From 9fc7c63a1d6e9920038ced782390a54888ed70a6 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: jbd2: fix the way the b_modified flag is cleared Currently at the start of a journal commit we loop through all of the buffers on the committing transaction and clear the b_modified flag (the flag that is set when a transaction modifies the buffer) under the j_list_lock. The problem is that everywhere else this flag is modified only under the jbd2 lock buffer flag, so it will race with a running transaction who could potentially set it, and have it unset by the committing transaction. This is also a big waste, you can have several thousands of buffers that you are clearing the modified flag on when you may not need to. This patch removes this code and instead clears the b_modified flag upon entering do_get_write_access/journal_get_create_access, so if that transaction does indeed use the buffer then it will be accounted for properly, and if it does not then we know we didn't use it. That will be important for the next patch in this series. Tested thoroughly by myself using postmark/iozone/bonnie++. Cc: Cc: Jan Kara Signed-off-by: Josef Bacik Signed-off-by: Andrew Morton Signed-off-by: "Theodore Ts'o" --- fs/jbd2/commit.c | 16 ---------------- fs/jbd2/transaction.c | 13 +++++++++++++ 2 files changed, 13 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index a8173081f831..988fbec1143c 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -519,22 +519,6 @@ void jbd2_journal_commit_transaction(journal_t *journal) jbd_debug (3, "JBD: commit phase 2\n"); - /* - * First, drop modified flag: all accesses to the buffers - * will be tracked for a new trasaction only -bzzz - */ - spin_lock(&journal->j_list_lock); - if (commit_transaction->t_buffers) { - new_jh = jh = commit_transaction->t_buffers->b_tnext; - do { - J_ASSERT_JH(new_jh, new_jh->b_modified == 1 || - new_jh->b_modified == 0); - new_jh->b_modified = 0; - new_jh = new_jh->b_tnext; - } while (new_jh != jh); - } - spin_unlock(&journal->j_list_lock); - /* * Now start flushing things to disk, in the order they appear * on the transaction lists. Data blocks go first. diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index b9b0b6f899b9..9dc71a6b62e6 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -617,6 +617,12 @@ repeat: jh->b_next_transaction == transaction) goto done; + /* + * this is the first time this transaction is touching this buffer, + * reset the modified flag + */ + jh->b_modified = 0; + /* * If there is already a copy-out version of this buffer, then we don't * need to make another one @@ -829,9 +835,16 @@ int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh) if (jh->b_transaction == NULL) { jh->b_transaction = transaction; + + /* first access by this transaction */ + jh->b_modified = 0; + JBUFFER_TRACE(jh, "file as BJ_Reserved"); __jbd2_journal_file_buffer(jh, transaction, BJ_Reserved); } else if (jh->b_transaction == journal->j_committing_transaction) { + /* first access by this transaction */ + jh->b_modified = 0; + JBUFFER_TRACE(jh, "set next transaction"); jh->b_next_transaction = transaction; } -- cgit v1.2.3 From 1dfc3220d963385a317264b11154c462a83596ed Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: jbd2: fix possible journal overflow issues There are several cases where the running transaction can get buffers added to its BJ_Metadata list which it never dirtied, which makes its t_nr_buffers counter end up larger than its t_outstanding_credits counter. This will cause issues when starting new transactions as while we are logging buffers we decrement t_outstanding_buffers, so when t_outstanding_buffers goes negative, we will report that we need less space in the journal than we actually need, so transactions will be started even though there may not be enough room for them. In the worst case scenario (which admittedly is almost impossible to reproduce) this will result in the journal running out of space. The fix is to only refile buffers from the committing transaction to the running transactions BJ_Modified list when b_modified is set on that journal, which is the only way to be sure if the running transaction has modified that buffer. This patch also fixes an accounting error in journal_forget, it is possible that we can call journal_forget on a buffer without having modified it, only gotten write access to it, so instead of freeing a credit, we only do so if the buffer was modified. The assert will help catch if this problem occurs. Without these two patches I could hit this assert within minutes of running postmark, with them this issue no longer arises. Cc: Cc: Jan Kara Signed-off-by: Josef Bacik Signed-off-by: Andrew Morton Signed-off-by: "Theodore Ts'o" --- fs/jbd2/commit.c | 3 +++ fs/jbd2/transaction.c | 21 ++++++++++++++++++--- 2 files changed, 21 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 988fbec1143c..e0139786f717 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -568,6 +568,9 @@ void jbd2_journal_commit_transaction(journal_t *journal) stats.u.run.rs_blocks = commit_transaction->t_outstanding_credits; stats.u.run.rs_blocks_logged = 0; + J_ASSERT(commit_transaction->t_nr_buffers <= + commit_transaction->t_outstanding_credits); + descriptor = NULL; bufs = 0; while (commit_transaction->t_buffers) { diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 9dc71a6b62e6..70245d6638b7 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -1243,6 +1243,7 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh) struct journal_head *jh; int drop_reserve = 0; int err = 0; + int was_modified = 0; BUFFER_TRACE(bh, "entry"); @@ -1261,6 +1262,9 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh) goto not_jbd; } + /* keep track of wether or not this transaction modified us */ + was_modified = jh->b_modified; + /* * The buffer's going from the transaction, we must drop * all references -bzzz @@ -1278,7 +1282,12 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh) JBUFFER_TRACE(jh, "belongs to current transaction: unfile"); - drop_reserve = 1; + /* + * we only want to drop a reference if this transaction + * modified the buffer + */ + if (was_modified) + drop_reserve = 1; /* * We are no longer going to journal this buffer. @@ -1318,7 +1327,13 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh) if (jh->b_next_transaction) { J_ASSERT(jh->b_next_transaction == transaction); jh->b_next_transaction = NULL; - drop_reserve = 1; + + /* + * only drop a reference if this transaction modified + * the buffer + */ + if (was_modified) + drop_reserve = 1; } } @@ -2090,7 +2105,7 @@ void __jbd2_journal_refile_buffer(struct journal_head *jh) jh->b_transaction = jh->b_next_transaction; jh->b_next_transaction = NULL; __jbd2_journal_file_buffer(jh, jh->b_transaction, - was_dirty ? BJ_Metadata : BJ_Reserved); + jh->b_modified ? BJ_Metadata : BJ_Reserved); J_ASSERT_JH(jh, jh->b_transaction->t_state == T_RUNNING); if (was_dirty) -- cgit v1.2.3 From 97bd42b9c8be748ad85b362ba3bd401f4d35be80 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Tue, 29 Apr 2008 22:04:56 -0400 Subject: ext4: check return of ext4_orphan_get properly This patch fix a panic while running fsfuzzer. We are improperly checking the return of ext4_orphan_get. Signed-off-by: Josef Bacik Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/super.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/ext4/super.c b/fs/ext4/super.c index c81a8e759bad..425f42778efa 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1594,8 +1594,8 @@ static void ext4_orphan_cleanup (struct super_block * sb, while (es->s_last_orphan) { struct inode *inode; - if (!(inode = - ext4_orphan_get(sb, le32_to_cpu(es->s_last_orphan)))) { + inode = ext4_orphan_get(sb, le32_to_cpu(es->s_last_orphan)); + if (IS_ERR(inode)) { es->s_last_orphan = 0; break; } -- cgit v1.2.3 From 53c550e9750434ddc4275fe0405170e0d1b46731 Mon Sep 17 00:00:00 2001 From: Hisashi Hifumi Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: ext4: fdatasync should skip metadata writeout when overwriting Currently fdatasync is identical to fsync in ext3. I think fdatasync should skip journal flush in data=ordered and data=writeback mode when it overwrites to already-instantiated blocks on HDD. When I_DIRTY_DATASYNC flag is not set, fdatasync should skip journal writeout because this indicates only atime or/and mtime updates. Following patch is the same approach of ext2's fsync code(ext2_sync_file). I did a performance test using the sysbench. #sysbench --num-threads=128 --max-requests=50000 --test=fileio --file-total-size=128G --file-test-mode=rndwr --file-fsync-mode=fdatasync run The result on ext3 was: -2.6.24 Operations performed: 0 Read, 50080 Write, 59600 Other = 109680 Total Read 0b Written 782.5Mb Total transferred 782.5Mb (12.116Mb/sec) 775.45 Requests/sec executed Test execution summary: total time: 64.5814s total number of events: 50080 total time taken by event execution: 3713.9836 per-request statistics: min: 0.0000s avg: 0.0742s max: 0.9375s approx. 95 percentile: 0.2901s Threads fairness: events (avg/stddev): 391.2500/23.26 execution time (avg/stddev): 29.0155/1.99 -2.6.24-patched Operations performed: 0 Read, 50009 Write, 61596 Other = 111605 Total Read 0b Written 781.39Mb Total transferred 781.39Mb (16.419Mb/sec) 1050.83 Requests/sec executed Test execution summary: total time: 47.5900s total number of events: 50009 total time taken by event execution: 2934.5768 per-request statistics: min: 0.0000s avg: 0.0587s max: 0.8938s approx. 95 percentile: 0.1993s Threads fairness: events (avg/stddev): 390.6953/22.64 execution time (avg/stddev): 22.9264/1.17 Filesystem I/O throughput was improved. Signed-off-by :Hisashi Hifumi Acked-by: Jan Kara Cc: Signed-off-by: Andrew Morton Signed-off-by: "Theodore Ts'o" --- fs/ext4/fsync.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index 8d50879d1c2c..a04a1ac4e0cf 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -72,6 +72,9 @@ int ext4_sync_file(struct file * file, struct dentry *dentry, int datasync) goto out; } + if (datasync && !(inode->i_state & I_DIRTY_DATASYNC)) + goto out; + /* * The VFS has written the file data. If the inode is unaltered * then we need not start a commit. -- cgit v1.2.3 From f3f12faa7414595f502721c90c34deccc1a03c71 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Tue, 29 Apr 2008 22:05:28 -0400 Subject: ext4: fix mount option parsing The "resize" option won't be noticed as it comes after the NULL option, so if you try to mount (or in this case remount) with that option it won't be recognized. Cc: Signed-off-by: Josef Bacik Signed-off-by: Andrew Morton Signed-off-by: "Theodore Ts'o" --- fs/ext4/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 425f42778efa..9d2d9a7fdea1 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -945,8 +945,8 @@ static match_table_t tokens = { {Opt_mballoc, "mballoc"}, {Opt_nomballoc, "nomballoc"}, {Opt_stripe, "stripe=%u"}, - {Opt_err, NULL}, {Opt_resize, "resize"}, + {Opt_err, NULL}, }; static ext4_fsblk_t get_sb_block(void **data) -- cgit v1.2.3 From 95c3889cb88ca4833096553c12cde9e7eb792f4c Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: ext4: Fix fallocate error path Put the old extent details back if we fail to split the uninitialized extent. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 26 ++++++++++++++++++++++++-- 1 file changed, 24 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 9ae6e67090cd..a38f32427826 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -2154,7 +2154,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, ext4_lblk_t iblock, unsigned long max_blocks) { - struct ext4_extent *ex, newex; + struct ext4_extent *ex, newex, orig_ex; struct ext4_extent *ex1 = NULL; struct ext4_extent *ex2 = NULL; struct ext4_extent *ex3 = NULL; @@ -2173,6 +2173,9 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, allocated = ee_len - (iblock - ee_block); newblock = iblock - ee_block + ext_pblock(ex); ex2 = ex; + orig_ex.ee_block = ex->ee_block; + orig_ex.ee_len = cpu_to_le16(ee_len); + ext4_ext_store_pblock(&orig_ex, ext_pblock(ex)); err = ext4_ext_get_access(handle, inode, path + depth); if (err) @@ -2201,13 +2204,25 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, ex3->ee_len = cpu_to_le16(allocated - max_blocks); ext4_ext_mark_uninitialized(ex3); err = ext4_ext_insert_extent(handle, inode, path, ex3); - if (err) + if (err) { + ex->ee_block = orig_ex.ee_block; + ex->ee_len = orig_ex.ee_len; + ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); + ext4_ext_mark_uninitialized(ex); + ext4_ext_dirty(handle, inode, path + depth); goto out; + } /* * The depth, and hence eh & ex might change * as part of the insert above. */ newdepth = ext_depth(inode); + /* + * update the extent length after successfull insert of the + * split extent + */ + orig_ex.ee_len = cpu_to_le16(ee_len - + ext4_ext_get_actual_len(ex3)); if (newdepth != depth) { depth = newdepth; ext4_ext_drop_refs(path); @@ -2282,6 +2297,13 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, goto out; insert: err = ext4_ext_insert_extent(handle, inode, path, &newex); + if (err) { + ex->ee_block = orig_ex.ee_block; + ex->ee_len = orig_ex.ee_len; + ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); + ext4_ext_mark_uninitialized(ex); + ext4_ext_dirty(handle, inode, path + depth); + } out: return err ? err : allocated; } -- cgit v1.2.3 From e65187e6d0d541f992e684f88a7e090dcff1aac8 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Tue, 29 Apr 2008 08:11:12 -0400 Subject: ext4: Enable extent format for symlinks. This patch enables extent-formatted normal symlinks. Using extents format allows a symlink to refer to a block number larger than 2^32 on large filesystems. We still don't enable extent format for fast symlinks, which are contained in the inode itself. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/ialloc.c | 4 ++-- fs/ext4/namei.c | 2 ++ 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index 486e46a3918d..cb14646117f0 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -750,8 +750,8 @@ got: goto fail_free_drop; } if (test_opt(sb, EXTENTS)) { - /* set extent flag only for directory and file */ - if (S_ISDIR(mode) || S_ISREG(mode)) { + /* set extent flag only for diretory, file and normal symlink*/ + if (S_ISDIR(mode) || S_ISREG(mode) || S_ISLNK(mode)) { EXT4_I(inode)->i_flags |= EXT4_EXTENTS_FL; ext4_ext_tree_init(handle, inode); err = ext4_update_incompat_feature(handle, sb, diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 28aa2ed4297e..68611945687d 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -2217,6 +2217,8 @@ retry: goto out_stop; } } else { + /* clear the extent format for fast symlink */ + EXT4_I(inode)->i_flags &= ~EXT4_EXTENTS_FL; inode->i_op = &ext4_fast_symlink_inode_operations; memcpy((char*)&EXT4_I(inode)->i_data,symname,l); inode->i_size = l-1; -- cgit v1.2.3 From 093a088b76352e0a6fdca84eb78b3aa65fbe6dd1 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Tue, 29 Apr 2008 08:11:12 -0400 Subject: ext4: ENOSPC error handling for writing to an uninitialized extent This patch handles possible ENOSPC errors when writing to an uninitialized extent in case the filesystem is full. A write to a prealloc area causes the split of an unititalized extent into initialized and uninitialized extents. If we don't have space to add new extent information, instead of returning error, convert the existing uninitialized extent to initialized one. We need to zero out the blocks corresponding to the entire extent to prevent uninitialized data reaching userspace. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 106 ++++++++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 99 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index a38f32427826..d279ea8c4661 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -2138,6 +2138,80 @@ void ext4_ext_release(struct super_block *sb) #endif } +static void bi_complete(struct bio *bio, int error) +{ + complete((struct completion *)bio->bi_private); +} + +/* FIXME!! we need to try to merge to left or right after zero-out */ +static int ext4_ext_zeroout(struct inode *inode, struct ext4_extent *ex) +{ + int ret = -EIO; + struct bio *bio; + int blkbits, blocksize; + sector_t ee_pblock; + struct completion event; + unsigned int ee_len, len, done, offset; + + + blkbits = inode->i_blkbits; + blocksize = inode->i_sb->s_blocksize; + ee_len = ext4_ext_get_actual_len(ex); + ee_pblock = ext_pblock(ex); + + /* convert ee_pblock to 512 byte sectors */ + ee_pblock = ee_pblock << (blkbits - 9); + + while (ee_len > 0) { + + if (ee_len > BIO_MAX_PAGES) + len = BIO_MAX_PAGES; + else + len = ee_len; + + bio = bio_alloc(GFP_NOIO, len); + if (!bio) + return -ENOMEM; + bio->bi_sector = ee_pblock; + bio->bi_bdev = inode->i_sb->s_bdev; + + done = 0; + offset = 0; + while (done < len) { + ret = bio_add_page(bio, ZERO_PAGE(0), + blocksize, offset); + if (ret != blocksize) { + /* + * We can't add any more pages because of + * hardware limitations. Start a new bio. + */ + break; + } + done++; + offset += blocksize; + if (offset >= PAGE_CACHE_SIZE) + offset = 0; + } + + init_completion(&event); + bio->bi_private = &event; + bio->bi_end_io = bi_complete; + submit_bio(WRITE, bio); + wait_for_completion(&event); + + if (test_bit(BIO_UPTODATE, &bio->bi_flags)) + ret = 0; + else { + ret = -EIO; + break; + } + bio_put(bio); + ee_len -= done; + ee_pblock += done << (blkbits - 9); + } + return ret; +} + /* * This function is called by ext4_ext_get_blocks() if someone tries to write * to an uninitialized extent. It may result in splitting the uninitialized @@ -2204,14 +2278,19 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, ex3->ee_len = cpu_to_le16(allocated - max_blocks); ext4_ext_mark_uninitialized(ex3); err = ext4_ext_insert_extent(handle, inode, path, ex3); - if (err) { + if (err == -ENOSPC) { + err = ext4_ext_zeroout(inode, &orig_ex); + if (err) + goto fix_extent_len; + /* update the extent length and mark as initialized */ ex->ee_block = orig_ex.ee_block; ex->ee_len = orig_ex.ee_len; ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); - ext4_ext_mark_uninitialized(ex); ext4_ext_dirty(handle, inode, path + depth); - goto out; - } + return le16_to_cpu(ex->ee_len); + + } else if (err) + goto fix_extent_len; /* * The depth, and hence eh & ex might change * as part of the insert above. @@ -2297,15 +2376,28 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, goto out; insert: err = ext4_ext_insert_extent(handle, inode, path, &newex); - if (err) { + if (err == -ENOSPC) { + err = ext4_ext_zeroout(inode, &orig_ex); + if (err) + goto fix_extent_len; + /* update the extent length and mark as initialized */ ex->ee_block = orig_ex.ee_block; ex->ee_len = orig_ex.ee_len; ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); - ext4_ext_mark_uninitialized(ex); ext4_ext_dirty(handle, inode, path + depth); - } + return le16_to_cpu(ex->ee_len); + } else if (err) + goto fix_extent_len; out: return err ? err : allocated; + +fix_extent_len: + ex->ee_block = orig_ex.ee_block; + ex->ee_len = orig_ex.ee_len; + ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); + ext4_ext_mark_uninitialized(ex); + ext4_ext_dirty(handle, inode, path + depth); + return err; } /* -- cgit v1.2.3 From 3977c965ec35ce1a7eac988ad313f0fc9aee9660 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: ext4: zero out small extents when writing to prealloc area. If the preallocated area is small zero out the full extent instead of splitting them. This should avoid the "write every alternate block" problem that could grow the number of extents dramatically. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 63 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 63 insertions(+) (limited to 'fs') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index d279ea8c4661..668d82e494dd 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -2212,6 +2212,8 @@ static int ext4_ext_zeroout(struct inode *inode, struct ext4_extent *ex) return ret; } +#define EXT4_EXT_ZERO_LEN 7 + /* * This function is called by ext4_ext_get_blocks() if someone tries to write * to an uninitialized extent. It may result in splitting the uninitialized @@ -2254,6 +2256,18 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, err = ext4_ext_get_access(handle, inode, path + depth); if (err) goto out; + /* If extent has less than 2*EXT4_EXT_ZERO_LEN zerout directly */ + if (ee_len <= 2*EXT4_EXT_ZERO_LEN) { + err = ext4_ext_zeroout(inode, &orig_ex); + if (err) + goto fix_extent_len; + /* update the extent length and mark as initialized */ + ex->ee_block = orig_ex.ee_block; + ex->ee_len = orig_ex.ee_len; + ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); + ext4_ext_dirty(handle, inode, path + depth); + return le16_to_cpu(ex->ee_len); + } /* ex1: ee_block to iblock - 1 : uninitialized */ if (iblock > ee_block) { @@ -2272,6 +2286,38 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, /* ex3: to ee_block + ee_len : uninitialised */ if (allocated > max_blocks) { unsigned int newdepth; + /* If extent has less than EXT4_EXT_ZERO_LEN zerout directly */ + if (allocated <= EXT4_EXT_ZERO_LEN) { + /* Mark first half uninitialized. + * Mark second half initialized and zero out the + * initialized extent + */ + ex->ee_block = orig_ex.ee_block; + ex->ee_len = cpu_to_le16(ee_len - allocated); + ext4_ext_mark_uninitialized(ex); + ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); + ext4_ext_dirty(handle, inode, path + depth); + + ex3 = &newex; + ex3->ee_block = cpu_to_le32(iblock); + ext4_ext_store_pblock(ex3, newblock); + ex3->ee_len = cpu_to_le16(allocated); + err = ext4_ext_insert_extent(handle, inode, path, ex3); + if (err == -ENOSPC) { + err = ext4_ext_zeroout(inode, &orig_ex); + if (err) + goto fix_extent_len; + ex->ee_block = orig_ex.ee_block; + ex->ee_len = orig_ex.ee_len; + ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); + ext4_ext_dirty(handle, inode, path + depth); + return le16_to_cpu(ex->ee_len); + + } else if (err) + goto fix_extent_len; + + return allocated; + } ex3 = &newex; ex3->ee_block = cpu_to_le32(iblock + max_blocks); ext4_ext_store_pblock(ex3, newblock + max_blocks); @@ -2320,6 +2366,23 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, goto out; } allocated = max_blocks; + + /* If extent has less than EXT4_EXT_ZERO_LEN and we are trying + * to insert a extent in the middle zerout directly + * otherwise give the extent a chance to merge to left + */ + if (le16_to_cpu(orig_ex.ee_len) <= EXT4_EXT_ZERO_LEN && + iblock != ee_block) { + err = ext4_ext_zeroout(inode, &orig_ex); + if (err) + goto fix_extent_len; + /* update the extent length and mark as initialized */ + ex->ee_block = orig_ex.ee_block; + ex->ee_len = orig_ex.ee_len; + ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); + ext4_ext_dirty(handle, inode, path + depth); + return le16_to_cpu(ex->ee_len); + } } /* * If there was a change of depth as part of the -- cgit v1.2.3 From 267e4db9ac28a09973476e7ec2cb6807e609d35a Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Tue, 29 Apr 2008 08:11:12 -0400 Subject: ext4: Fix race between migration and mmap write Fail migrate if we allocated new blocks via mmap write. If we write to holes in the file via mmap, we end up allocating new blocks. This block allocation happens without taking inode->i_mutex. Since migrate is protected by i_mutex and migrate expects that no new blocks get allocated during migrate, fail migrate if new blocks get allocated. We can't take inode->i_mutex in the mmap write path because that would result in a locking order violation between i_mutex and mmap_sem. Also adding a separate rw_sempahore for protection is really high overhead for a rare operation such as migrate. Signed-off-by: Aneesh Kumar K.V Acked-by: Jan Kara Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 13 ++++++++++++- fs/ext4/migrate.c | 39 ++++++++++++++++++++++++++++++++++----- include/linux/ext4_fs.h | 1 + 3 files changed, 47 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 8fab233cb05f..24a2604dde7b 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -985,6 +985,16 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, } else { retval = ext4_get_blocks_handle(handle, inode, block, max_blocks, bh, create, extend_disksize); + + if (retval > 0 && buffer_new(bh)) { + /* + * We allocated new blocks which will result in + * i_data's format changing. Force the migrate + * to fail by clearing migrate flags + */ + EXT4_I(inode)->i_flags = EXT4_I(inode)->i_flags & + ~EXT4_EXT_MIGRATE; + } } up_write((&EXT4_I(inode)->i_data_sem)); return retval; @@ -2976,7 +2986,8 @@ static int ext4_do_update_inode(handle_t *handle, if (ext4_inode_blocks_set(handle, raw_inode, ei)) goto out_brelse; raw_inode->i_dtime = cpu_to_le32(ei->i_dtime); - raw_inode->i_flags = cpu_to_le32(ei->i_flags); + /* clear the migrate flag in the raw_inode */ + raw_inode->i_flags = cpu_to_le32(ei->i_flags & ~EXT4_EXT_MIGRATE); if (EXT4_SB(inode->i_sb)->s_es->s_creator_os != cpu_to_le32(EXT4_OS_HURD)) raw_inode->i_file_acl_high = diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c index 5c1e27de7755..9b4fb07d192c 100644 --- a/fs/ext4/migrate.c +++ b/fs/ext4/migrate.c @@ -327,7 +327,7 @@ static int free_ind_block(handle_t *handle, struct inode *inode, __le32 *i_data) } static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode, - struct inode *tmp_inode) + struct inode *tmp_inode) { int retval; __le32 i_data[3]; @@ -339,7 +339,7 @@ static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode, * i_data field of the original inode */ retval = ext4_journal_extend(handle, 1); - if (retval != 0) { + if (retval) { retval = ext4_journal_restart(handle, 1); if (retval) goto err_out; @@ -350,6 +350,18 @@ static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode, i_data[2] = ei->i_data[EXT4_TIND_BLOCK]; down_write(&EXT4_I(inode)->i_data_sem); + /* + * if EXT4_EXT_MIGRATE is cleared a block allocation + * happened after we started the migrate. We need to + * fail the migrate + */ + if (!(EXT4_I(inode)->i_flags & EXT4_EXT_MIGRATE)) { + retval = -EAGAIN; + up_write(&EXT4_I(inode)->i_data_sem); + goto err_out; + } else + EXT4_I(inode)->i_flags = EXT4_I(inode)->i_flags & + ~EXT4_EXT_MIGRATE; /* * We have the extent map build with the tmp inode. * Now copy the i_data across @@ -508,6 +520,17 @@ int ext4_ext_migrate(struct inode *inode, struct file *filp, * switch the inode format to prevent read. */ mutex_lock(&(inode->i_mutex)); + /* + * Even though we take i_mutex we can still cause block allocation + * via mmap write to holes. If we have allocated new blocks we fail + * migrate. New block allocation will clear EXT4_EXT_MIGRATE flag. + * The flag is updated with i_data_sem held to prevent racing with + * block allocation. + */ + down_read((&EXT4_I(inode)->i_data_sem)); + EXT4_I(inode)->i_flags = EXT4_I(inode)->i_flags | EXT4_EXT_MIGRATE; + up_read((&EXT4_I(inode)->i_data_sem)); + handle = ext4_journal_start(inode, 1); ei = EXT4_I(inode); @@ -559,9 +582,15 @@ err_out: * tmp_inode */ free_ext_block(handle, tmp_inode); - else - retval = ext4_ext_swap_inode_data(handle, inode, - tmp_inode); + else { + retval = ext4_ext_swap_inode_data(handle, inode, tmp_inode); + if (retval) + /* + * if we fail to swap inode data free the extent + * details of the tmp inode + */ + free_ext_block(handle, tmp_inode); + } /* We mark the tmp_inode dirty via ext4_ext_tree_init. */ if (ext4_journal_extend(handle, 1) != 0) diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h index 250032548597..105337ca9ed0 100644 --- a/include/linux/ext4_fs.h +++ b/include/linux/ext4_fs.h @@ -231,6 +231,7 @@ struct ext4_group_desc #define EXT4_TOPDIR_FL 0x00020000 /* Top of directory hierarchies*/ #define EXT4_HUGE_FILE_FL 0x00040000 /* Set to each huge file */ #define EXT4_EXTENTS_FL 0x00080000 /* Inode uses extents */ +#define EXT4_EXT_MIGRATE 0x00100000 /* Inode is migrating */ #define EXT4_RESERVED_FL 0x80000000 /* reserved for ext4 lib */ #define EXT4_FL_USER_VISIBLE 0x000BDFFF /* User visible flags */ -- cgit v1.2.3 From fd28784adc079afa905df56204b1298ddb4d0bfe Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Tue, 29 Apr 2008 08:11:12 -0400 Subject: ext4: Fix fallocate to update the file size in each transaction ext4_fallocate needs to update file size in each transaction. Otherwise if we crash the file size won't be seen. We were also not marking the inode dirty after updating file size before. Also when we try to retry allocation due to ENOSPC, make sure we reset the variable ret so that we actually do a retry. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 89 ++++++++++++++++++++++++------------------------------- 1 file changed, 38 insertions(+), 51 deletions(-) (limited to 'fs') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 668d82e494dd..c2b004e29ad8 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -2785,6 +2785,28 @@ int ext4_ext_writepage_trans_blocks(struct inode *inode, int num) return needed; } +static void ext4_falloc_update_inode(struct inode *inode, + int mode, loff_t new_size, int update_ctime) +{ + struct timespec now; + + if (update_ctime) { + now = current_fs_time(inode->i_sb); + if (!timespec_equal(&inode->i_ctime, &now)) + inode->i_ctime = now; + } + /* + * Update only when preallocation was requested beyond + * the file size. + */ + if (!(mode & FALLOC_FL_KEEP_SIZE) && + new_size > i_size_read(inode)) { + i_size_write(inode, new_size); + EXT4_I(inode)->i_disksize = new_size; + } + +} + /* * preallocate space for a file. This implements ext4's fallocate inode * operation, which gets called from sys_fallocate system call. @@ -2796,8 +2818,8 @@ long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len) { handle_t *handle; ext4_lblk_t block; + loff_t new_size; unsigned long max_blocks; - ext4_fsblk_t nblocks = 0; int ret = 0; int ret2 = 0; int retries = 0; @@ -2816,9 +2838,12 @@ long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len) return -ENODEV; block = offset >> blkbits; + /* + * We can't just convert len to max_blocks because + * If blocksize = 4096 offset = 3072 and len = 2048 + */ max_blocks = (EXT4_BLOCK_ALIGN(len + offset, blkbits) >> blkbits) - - block; - + - block; /* * credits to insert 1 extent into extent tree + buffers to be able to * modify 1 super block, 1 block bitmap and 1 group descriptor. @@ -2834,7 +2859,6 @@ retry: ret = PTR_ERR(handle); break; } - ret = ext4_get_blocks_wrap(handle, inode, block, max_blocks, &map_bh, EXT4_CREATE_UNINITIALIZED_EXT, 0); @@ -2850,61 +2874,24 @@ retry: ret2 = ext4_journal_stop(handle); break; } - if (ret > 0) { - /* check wrap through sign-bit/zero here */ - if ((block + ret) < 0 || (block + ret) < block) { - ret = -EIO; - ext4_mark_inode_dirty(handle, inode); - ret2 = ext4_journal_stop(handle); - break; - } - if (buffer_new(&map_bh) && ((block + ret) > - (EXT4_BLOCK_ALIGN(i_size_read(inode), blkbits) - >> blkbits))) - nblocks = nblocks + ret; - } - - /* Update ctime if new blocks get allocated */ - if (nblocks) { - struct timespec now; - - now = current_fs_time(inode->i_sb); - if (!timespec_equal(&inode->i_ctime, &now)) - inode->i_ctime = now; - } + if ((block + ret) >= (EXT4_BLOCK_ALIGN(offset + len, + blkbits) >> blkbits)) + new_size = offset + len; + else + new_size = (block + ret) << blkbits; + ext4_falloc_update_inode(inode, mode, new_size, + buffer_new(&map_bh)); ext4_mark_inode_dirty(handle, inode); ret2 = ext4_journal_stop(handle); if (ret2) break; } - - if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) + if (ret == -ENOSPC && + ext4_should_retry_alloc(inode->i_sb, &retries)) { + ret = 0; goto retry; - - /* - * Time to update the file size. - * Update only when preallocation was requested beyond the file size. - */ - if (!(mode & FALLOC_FL_KEEP_SIZE) && - (offset + len) > i_size_read(inode)) { - if (ret > 0) { - /* - * if no error, we assume preallocation succeeded - * completely - */ - i_size_write(inode, offset + len); - EXT4_I(inode)->i_disksize = i_size_read(inode); - } else if (ret < 0 && nblocks) { - /* Handle partial allocation scenario */ - loff_t newsize; - - newsize = (nblocks << blkbits) + i_size_read(inode); - i_size_write(inode, EXT4_BLOCK_ALIGN(newsize, blkbits)); - EXT4_I(inode)->i_disksize = i_size_read(inode); - } } - mutex_unlock(&inode->i_mutex); return ret > 0 ? ret2 : ret; } -- cgit v1.2.3 From e067ba0078cd6f00eb6c4052fec630b78ebe59de Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Tue, 29 Apr 2008 08:11:12 -0400 Subject: ext4: make ext4_ext_get_blocks always return <= max_blocks ext4_ext_get_blocks() returns number of blocks allocated with buffer heads unmapped for a read from prealloc space. This is needed so that delayed allocation doesn't do block reservation for prealloc space since the blocks are already resevred on disk. Fix ext4_ext_get_blocks to not return greater than max_blocks, since some of the code paths cannot handle such a return value. Signed-off-by: Aneesh Kumar K.V Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index c2b004e29ad8..f4ef0b745a53 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -2570,8 +2570,18 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, } if (create == EXT4_CREATE_UNINITIALIZED_EXT) goto out; - if (!create) + if (!create) { + /* + * We have blocks reserved already. We + * return allocated blocks so that delalloc + * won't do block reservation for us. But + * the buffer head will be unmapped so that + * a read from the block returns 0s. + */ + if (allocated > max_blocks) + allocated = max_blocks; goto out2; + } ret = ext4_ext_convert_to_initialized(handle, inode, path, iblock, -- cgit v1.2.3 From 1a89734d4057066344356e9c7e13b6379497aebe Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Tue, 29 Apr 2008 08:11:12 -0400 Subject: ext4: Return unwritten buffer head when trying to read from prealloc space. ext4_ext_get_blocks() returns the number of blocks allocated with buffer head unmapped for a read from prealloc space. This is needed so that delayed allocation doesn't do block reservation for prealloc space since the blocks are already reserved on disk. Mark the buffer head unwritten. Some code paths try to read the block if the buffer_head is not new and no uptodate. Marking the buffer head unwritten avoids this reading. Signed-off-by: Aneesh Kumar K.V Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index f4ef0b745a53..7733e2943367 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -2580,6 +2580,8 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, */ if (allocated > max_blocks) allocated = max_blocks; + /* mark the buffer unwritten */ + __set_bit(BH_Unwritten, &bh_result->b_state); goto out2; } -- cgit v1.2.3 From 161e7b7c1d24112d188df9a7b30d468a8d135b96 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Tue, 29 Apr 2008 22:03:59 -0400 Subject: ext4: Cache the correct extent length for uninit extents When we convert an uninitialized extent to an initialized extent we need to make sure we return the number of blocks in the extent from the file system block corresponding to logical file block. Otherwise we cache wrong extent details and this results in file system corruption. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 48 +++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 43 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 7733e2943367..c7c42c9c7bf7 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -2266,7 +2266,8 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, ex->ee_len = orig_ex.ee_len; ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); ext4_ext_dirty(handle, inode, path + depth); - return le16_to_cpu(ex->ee_len); + /* zeroed the full extent */ + return allocated; } /* ex1: ee_block to iblock - 1 : uninitialized */ @@ -2311,11 +2312,45 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, ex->ee_len = orig_ex.ee_len; ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); ext4_ext_dirty(handle, inode, path + depth); - return le16_to_cpu(ex->ee_len); + /* zeroed the full extent */ + return allocated; } else if (err) goto fix_extent_len; + /* + * We need to zero out the second half because + * an fallocate request can update file size and + * converting the second half to initialized extent + * implies that we can leak some junk data to user + * space. + */ + err = ext4_ext_zeroout(inode, ex3); + if (err) { + /* + * We should actually mark the + * second half as uninit and return error + * Insert would have changed the extent + */ + depth = ext_depth(inode); + ext4_ext_drop_refs(path); + path = ext4_ext_find_extent(inode, + iblock, path); + if (IS_ERR(path)) { + err = PTR_ERR(path); + return err; + } + ex = path[depth].p_ext; + err = ext4_ext_get_access(handle, inode, + path + depth); + if (err) + return err; + ext4_ext_mark_uninitialized(ex); + ext4_ext_dirty(handle, inode, path + depth); + return err; + } + + /* zeroed the second half */ return allocated; } ex3 = &newex; @@ -2333,7 +2368,8 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, ex->ee_len = orig_ex.ee_len; ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); ext4_ext_dirty(handle, inode, path + depth); - return le16_to_cpu(ex->ee_len); + /* zeroed the full extent */ + return allocated; } else if (err) goto fix_extent_len; @@ -2381,7 +2417,8 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, ex->ee_len = orig_ex.ee_len; ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); ext4_ext_dirty(handle, inode, path + depth); - return le16_to_cpu(ex->ee_len); + /* zero out the first half */ + return allocated; } } /* @@ -2448,7 +2485,8 @@ insert: ex->ee_len = orig_ex.ee_len; ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); ext4_ext_dirty(handle, inode, path + depth); - return le16_to_cpu(ex->ee_len); + /* zero out the first half */ + return allocated; } else if (err) goto fix_extent_len; out: -- cgit v1.2.3 From 5cdd7b2d7716a7ed7d6dc7588e2d015f04d46640 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 29 Apr 2008 22:03:54 -0400 Subject: Convert ext4 to use unlocked_ioctl I checked ext4_ioctl and it looked largely safe to not be used without BKL. So convert it over to unlocked_ioctl. Signed-off-by: Andi Kleen Signed-off-by: Theodore Ts'o --- fs/ext4/dir.c | 2 +- fs/ext4/file.c | 2 +- fs/ext4/ioctl.c | 12 +++--------- include/linux/ext4_fs.h | 3 +-- 4 files changed, 6 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index 2c23bade9aa6..88c97f7312be 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -42,7 +42,7 @@ const struct file_operations ext4_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, .readdir = ext4_readdir, /* we take BKL. needed?*/ - .ioctl = ext4_ioctl, /* BKL held */ + .unlocked_ioctl = ext4_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ext4_compat_ioctl, #endif diff --git a/fs/ext4/file.c b/fs/ext4/file.c index ac35ec58db55..20507a24506a 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -129,7 +129,7 @@ const struct file_operations ext4_file_operations = { .write = do_sync_write, .aio_read = generic_file_aio_read, .aio_write = ext4_file_write, - .ioctl = ext4_ioctl, + .unlocked_ioctl = ext4_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ext4_compat_ioctl, #endif diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index 25b13ede8086..ce937fe432a0 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -18,9 +18,9 @@ #include #include -int ext4_ioctl (struct inode * inode, struct file * filp, unsigned int cmd, - unsigned long arg) +long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { + struct inode *inode = filp->f_dentry->d_inode; struct ext4_inode_info *ei = EXT4_I(inode); unsigned int flags; unsigned short rsv_window_size; @@ -277,9 +277,6 @@ setversion_out: #ifdef CONFIG_COMPAT long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { - struct inode *inode = file->f_path.dentry->d_inode; - int ret; - /* These are just misnamed, they actually get/put from/to user an int */ switch (cmd) { case EXT4_IOC32_GETFLAGS: @@ -319,9 +316,6 @@ long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) default: return -ENOIOCTLCMD; } - lock_kernel(); - ret = ext4_ioctl(inode, file, cmd, (unsigned long) compat_ptr(arg)); - unlock_kernel(); - return ret; + return ext4_ioctl(file, cmd, (unsigned long) compat_ptr(arg)); } #endif diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h index 105337ca9ed0..33bc88568c54 100644 --- a/include/linux/ext4_fs.h +++ b/include/linux/ext4_fs.h @@ -1050,8 +1050,7 @@ extern int ext4_block_truncate_page(handle_t *handle, struct page *page, struct address_space *mapping, loff_t from); /* ioctl.c */ -extern int ext4_ioctl (struct inode *, struct file *, unsigned int, - unsigned long); +extern long ext4_ioctl(struct file *, unsigned int, unsigned long); extern long ext4_compat_ioctl (struct file *, unsigned int, unsigned long); /* migrate.c */ -- cgit v1.2.3 From 4ddfef7b41aebbbede73f361cb938800ba3072dc Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Tue, 29 Apr 2008 08:11:12 -0400 Subject: ext4: reduce mballoc stack usage with noinline_for_stack mballoc.c is a whole lot of static functions, which gcc seems to really like to inline. With the changes below, on x86, I can at least get from: 432 ext4_mb_new_blocks 240 ext4_mb_free_blocks 208 ext4_mb_discard_group_preallocations 188 ext4_mb_seq_groups_show 164 ext4_mb_init_cache 152 ext4_mb_release_inode_pa 136 ext4_mb_seq_history_show ... to 220 ext4_mb_free_blocks 188 ext4_mb_seq_groups_show 176 ext4_mb_regular_allocator 164 ext4_mb_init_cache 156 ext4_mb_new_blocks 152 ext4_mb_release_inode_pa 136 ext4_mb_seq_history_show 124 ext4_mb_release_group_pa ... which still has some big functions in there, but not 432 bytes! Signed-off-by: Eric Sandeen Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/mballoc.c | 45 +++++++++++++++++++++++++++++---------------- 1 file changed, 29 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 9d57695de746..66e1451f64e7 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -1168,8 +1168,9 @@ out: return err; } -static int ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group, - struct ext4_buddy *e4b) +static noinline_for_stack int +ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group, + struct ext4_buddy *e4b) { struct ext4_sb_info *sbi = EXT4_SB(sb); struct inode *inode = sbi->s_buddy_cache; @@ -1965,7 +1966,8 @@ static int ext4_mb_good_group(struct ext4_allocation_context *ac, return 0; } -static int ext4_mb_regular_allocator(struct ext4_allocation_context *ac) +static noinline_for_stack int +ext4_mb_regular_allocator(struct ext4_allocation_context *ac) { ext4_group_t group; ext4_group_t i; @@ -2465,7 +2467,8 @@ static void ext4_mb_history_init(struct super_block *sb) /* if we can't allocate history, then we simple won't use it */ } -static void ext4_mb_store_history(struct ext4_allocation_context *ac) +static noinline_for_stack void +ext4_mb_store_history(struct ext4_allocation_context *ac) { struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); struct ext4_mb_history h; @@ -2801,7 +2804,8 @@ int ext4_mb_release(struct super_block *sb) return 0; } -static void ext4_mb_free_committed_blocks(struct super_block *sb) +static noinline_for_stack void +ext4_mb_free_committed_blocks(struct super_block *sb) { struct ext4_sb_info *sbi = EXT4_SB(sb); int err; @@ -3021,7 +3025,8 @@ void exit_ext4_mballoc(void) * Check quota and mark choosed space (ac->ac_b_ex) non-free in bitmaps * Returns 0 if success or error code */ -static int ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, +static noinline_for_stack int +ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, handle_t *handle) { struct buffer_head *bitmap_bh = NULL; @@ -3138,7 +3143,8 @@ static void ext4_mb_normalize_group_request(struct ext4_allocation_context *ac) * Normalization means making request better in terms of * size and alignment */ -static void ext4_mb_normalize_request(struct ext4_allocation_context *ac, +static noinline_for_stack void +ext4_mb_normalize_request(struct ext4_allocation_context *ac, struct ext4_allocation_request *ar) { int bsbits, max; @@ -3404,7 +3410,8 @@ static void ext4_mb_use_group_pa(struct ext4_allocation_context *ac, /* * search goal blocks in preallocated space */ -static int ext4_mb_use_preallocated(struct ext4_allocation_context *ac) +static noinline_for_stack int +ext4_mb_use_preallocated(struct ext4_allocation_context *ac) { struct ext4_inode_info *ei = EXT4_I(ac->ac_inode); struct ext4_locality_group *lg; @@ -3571,7 +3578,8 @@ static void ext4_mb_put_pa(struct ext4_allocation_context *ac, /* * creates new preallocated space for given inode */ -static int ext4_mb_new_inode_pa(struct ext4_allocation_context *ac) +static noinline_for_stack int +ext4_mb_new_inode_pa(struct ext4_allocation_context *ac) { struct super_block *sb = ac->ac_sb; struct ext4_prealloc_space *pa; @@ -3658,7 +3666,8 @@ static int ext4_mb_new_inode_pa(struct ext4_allocation_context *ac) /* * creates new preallocated space for locality group inodes belongs to */ -static int ext4_mb_new_group_pa(struct ext4_allocation_context *ac) +static noinline_for_stack int +ext4_mb_new_group_pa(struct ext4_allocation_context *ac) { struct super_block *sb = ac->ac_sb; struct ext4_locality_group *lg; @@ -3731,8 +3740,8 @@ static int ext4_mb_new_preallocation(struct ext4_allocation_context *ac) * the caller MUST hold group/inode locks. * TODO: optimize the case when there are no in-core structures yet */ -static int ext4_mb_release_inode_pa(struct ext4_buddy *e4b, - struct buffer_head *bitmap_bh, +static noinline_for_stack int +ext4_mb_release_inode_pa(struct ext4_buddy *e4b, struct buffer_head *bitmap_bh, struct ext4_prealloc_space *pa) { struct ext4_allocation_context *ac; @@ -3803,7 +3812,8 @@ static int ext4_mb_release_inode_pa(struct ext4_buddy *e4b, return err; } -static int ext4_mb_release_group_pa(struct ext4_buddy *e4b, +static noinline_for_stack int +ext4_mb_release_group_pa(struct ext4_buddy *e4b, struct ext4_prealloc_space *pa) { struct ext4_allocation_context *ac; @@ -3845,7 +3855,8 @@ static int ext4_mb_release_group_pa(struct ext4_buddy *e4b, * - how many do we discard * 1) how many requested */ -static int ext4_mb_discard_group_preallocations(struct super_block *sb, +static noinline_for_stack int +ext4_mb_discard_group_preallocations(struct super_block *sb, ext4_group_t group, int needed) { struct ext4_group_info *grp = ext4_get_group_info(sb, group); @@ -4167,7 +4178,8 @@ static void ext4_mb_group_or_file(struct ext4_allocation_context *ac) mutex_lock(&ac->ac_lg->lg_mutex); } -static int ext4_mb_initialize_context(struct ext4_allocation_context *ac, +static noinline_for_stack int +ext4_mb_initialize_context(struct ext4_allocation_context *ac, struct ext4_allocation_request *ar) { struct super_block *sb = ar->inode->i_sb; @@ -4398,7 +4410,8 @@ static void ext4_mb_poll_new_transaction(struct super_block *sb, ext4_mb_free_committed_blocks(sb); } -static int ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b, +static noinline_for_stack int +ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b, ext4_group_t group, ext4_grpblk_t block, int count) { struct ext4_group_info *db = e4b->bd_info; -- cgit v1.2.3 From 9a0762c5af40e4aa64fef999967459c98e6ae4c9 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: ext4: Convert list_for_each_rcu() to list_for_each_entry_rcu() The list_for_each_entry_rcu() primitive should be used instead of list_for_each_rcu(), as the former is easier to use and provides better type safety. http://groups.google.com/group/linux.kernel/browse_thread/thread/45749c83451cebeb/0633a65759ce7713?lnk=raot Signed-off-by: Aneesh Kumar K.V Signed-off-by: Roel Kluin <12o3l@tiscali.nl> Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/mballoc.c | 18 +++++------------- 1 file changed, 5 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 66e1451f64e7..aaaccf5986f9 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -3149,10 +3149,10 @@ ext4_mb_normalize_request(struct ext4_allocation_context *ac, { int bsbits, max; ext4_lblk_t end; - struct list_head *cur; loff_t size, orig_size, start_off; ext4_lblk_t start, orig_start; struct ext4_inode_info *ei = EXT4_I(ac->ac_inode); + struct ext4_prealloc_space *pa; /* do normalize only data requests, metadata requests do not need preallocation */ @@ -3238,12 +3238,9 @@ ext4_mb_normalize_request(struct ext4_allocation_context *ac, /* check we don't cross already preallocated blocks */ rcu_read_lock(); - list_for_each_rcu(cur, &ei->i_prealloc_list) { - struct ext4_prealloc_space *pa; + list_for_each_entry_rcu(pa, &ei->i_prealloc_list, pa_inode_list) { unsigned long pa_end; - pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list); - if (pa->pa_deleted) continue; spin_lock(&pa->pa_lock); @@ -3285,10 +3282,8 @@ ext4_mb_normalize_request(struct ext4_allocation_context *ac, /* XXX: extra loop to check we really don't overlap preallocations */ rcu_read_lock(); - list_for_each_rcu(cur, &ei->i_prealloc_list) { - struct ext4_prealloc_space *pa; + list_for_each_entry_rcu(pa, &ei->i_prealloc_list, pa_inode_list) { unsigned long pa_end; - pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list); spin_lock(&pa->pa_lock); if (pa->pa_deleted == 0) { pa_end = pa->pa_lstart + pa->pa_len; @@ -3416,7 +3411,6 @@ ext4_mb_use_preallocated(struct ext4_allocation_context *ac) struct ext4_inode_info *ei = EXT4_I(ac->ac_inode); struct ext4_locality_group *lg; struct ext4_prealloc_space *pa; - struct list_head *cur; /* only data can be preallocated */ if (!(ac->ac_flags & EXT4_MB_HINT_DATA)) @@ -3424,8 +3418,7 @@ ext4_mb_use_preallocated(struct ext4_allocation_context *ac) /* first, try per-file preallocation */ rcu_read_lock(); - list_for_each_rcu(cur, &ei->i_prealloc_list) { - pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list); + list_for_each_entry_rcu(pa, &ei->i_prealloc_list, pa_inode_list) { /* all fields in this condition don't change, * so we can skip locking for them */ @@ -3457,8 +3450,7 @@ ext4_mb_use_preallocated(struct ext4_allocation_context *ac) return 0; rcu_read_lock(); - list_for_each_rcu(cur, &lg->lg_prealloc_list) { - pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list); + list_for_each_entry_rcu(pa, &lg->lg_prealloc_list, pa_inode_list) { spin_lock(&pa->pa_lock); if (pa->pa_deleted == 0 && pa->pa_free >= ac->ac_o_ex.fe_len) { atomic_inc(&pa->pa_count); -- cgit v1.2.3 From e8546d0615542684ca02ba03edebec1a503beb6b Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: ext4: le*_add_cpu conversion replace all: little_endian_variable = cpu_to_leX(leX_to_cpu(little_endian_variable) + expression_in_cpu_byteorder); with: leX_add_cpu(&little_endian_variable, expression_in_cpu_byteorder); generated with semantic patch Signed-off-by: Marcin Slusarz Signed-off-by: "Theodore Ts'o" Cc: linux-ext4@vger.kernel.org Cc: sct@redhat.com Cc: Andrew Morton Cc: adilger@clusterfs.com Cc: Mingming Cao --- fs/ext4/balloc.c | 7 ++----- fs/ext4/extents.c | 20 +++++++++----------- fs/ext4/ialloc.c | 12 ++++-------- fs/ext4/mballoc.c | 7 ++----- fs/ext4/resize.c | 6 ++---- fs/ext4/super.c | 2 +- fs/ext4/xattr.c | 6 ++---- 7 files changed, 22 insertions(+), 38 deletions(-) (limited to 'fs') diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 0737e05ba3dd..5d348998ff38 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -752,9 +752,7 @@ do_more: jbd_unlock_bh_state(bitmap_bh); spin_lock(sb_bgl_lock(sbi, block_group)); - desc->bg_free_blocks_count = - cpu_to_le16(le16_to_cpu(desc->bg_free_blocks_count) + - group_freed); + le16_add_cpu(&desc->bg_free_blocks_count, group_freed); desc->bg_checksum = ext4_group_desc_csum(sbi, block_group, desc); spin_unlock(sb_bgl_lock(sbi, block_group)); percpu_counter_add(&sbi->s_freeblocks_counter, count); @@ -1823,8 +1821,7 @@ allocated: spin_lock(sb_bgl_lock(sbi, group_no)); if (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT); - gdp->bg_free_blocks_count = - cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count)-num); + le16_add_cpu(&gdp->bg_free_blocks_count, -num); gdp->bg_checksum = ext4_group_desc_csum(sbi, group_no, gdp); spin_unlock(sb_bgl_lock(sbi, group_no)); percpu_counter_sub(&sbi->s_freeblocks_counter, num); diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index c7c42c9c7bf7..3d5a35f373d4 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -614,7 +614,7 @@ static int ext4_ext_insert_index(handle_t *handle, struct inode *inode, ix->ei_block = cpu_to_le32(logical); ext4_idx_store_pblock(ix, ptr); - curp->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(curp->p_hdr->eh_entries)+1); + le16_add_cpu(&curp->p_hdr->eh_entries, 1); BUG_ON(le16_to_cpu(curp->p_hdr->eh_entries) > le16_to_cpu(curp->p_hdr->eh_max)); @@ -736,7 +736,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, } if (m) { memmove(ex, path[depth].p_ext-m, sizeof(struct ext4_extent)*m); - neh->eh_entries = cpu_to_le16(le16_to_cpu(neh->eh_entries)+m); + le16_add_cpu(&neh->eh_entries, m); } set_buffer_uptodate(bh); @@ -753,8 +753,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, err = ext4_ext_get_access(handle, inode, path + depth); if (err) goto cleanup; - path[depth].p_hdr->eh_entries = - cpu_to_le16(le16_to_cpu(path[depth].p_hdr->eh_entries)-m); + le16_add_cpu(&path[depth].p_hdr->eh_entries, -m); err = ext4_ext_dirty(handle, inode, path + depth); if (err) goto cleanup; @@ -817,8 +816,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, if (m) { memmove(++fidx, path[i].p_idx - m, sizeof(struct ext4_extent_idx) * m); - neh->eh_entries = - cpu_to_le16(le16_to_cpu(neh->eh_entries) + m); + le16_add_cpu(&neh->eh_entries, m); } set_buffer_uptodate(bh); unlock_buffer(bh); @@ -834,7 +832,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, err = ext4_ext_get_access(handle, inode, path + i); if (err) goto cleanup; - path[i].p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path[i].p_hdr->eh_entries)-m); + le16_add_cpu(&path[i].p_hdr->eh_entries, -m); err = ext4_ext_dirty(handle, inode, path + i); if (err) goto cleanup; @@ -1369,7 +1367,7 @@ int ext4_ext_try_to_merge(struct inode *inode, * sizeof(struct ext4_extent); memmove(ex + 1, ex + 2, len); } - eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries) - 1); + le16_add_cpu(&eh->eh_entries, -1); merge_done = 1; WARN_ON(eh->eh_entries == 0); if (!eh->eh_entries) @@ -1560,7 +1558,7 @@ has_space: path[depth].p_ext = nearex; } - eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)+1); + le16_add_cpu(&eh->eh_entries, 1); nearex = path[depth].p_ext; nearex->ee_block = newext->ee_block; ext4_ext_store_pblock(nearex, ext_pblock(newext)); @@ -1699,7 +1697,7 @@ static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, err = ext4_ext_get_access(handle, inode, path); if (err) return err; - path->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path->p_hdr->eh_entries)-1); + le16_add_cpu(&path->p_hdr->eh_entries, -1); err = ext4_ext_dirty(handle, inode, path); if (err) return err; @@ -1902,7 +1900,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, if (num == 0) { /* this extent is removed; mark slot entirely unused */ ext4_ext_store_pblock(ex, 0); - eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)-1); + le16_add_cpu(&eh->eh_entries, -1); } ex->ee_block = cpu_to_le32(block); diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index cb14646117f0..e8d24f24f282 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -223,11 +223,9 @@ void ext4_free_inode (handle_t *handle, struct inode * inode) if (gdp) { spin_lock(sb_bgl_lock(sbi, block_group)); - gdp->bg_free_inodes_count = cpu_to_le16( - le16_to_cpu(gdp->bg_free_inodes_count) + 1); + le16_add_cpu(&gdp->bg_free_inodes_count, 1); if (is_directory) - gdp->bg_used_dirs_count = cpu_to_le16( - le16_to_cpu(gdp->bg_used_dirs_count) - 1); + le16_add_cpu(&gdp->bg_used_dirs_count, -1); gdp->bg_checksum = ext4_group_desc_csum(sbi, block_group, gdp); spin_unlock(sb_bgl_lock(sbi, block_group)); @@ -664,11 +662,9 @@ got: cpu_to_le16(EXT4_INODES_PER_GROUP(sb) - ino); } - gdp->bg_free_inodes_count = - cpu_to_le16(le16_to_cpu(gdp->bg_free_inodes_count) - 1); + le16_add_cpu(&gdp->bg_free_inodes_count, -1); if (S_ISDIR(mode)) { - gdp->bg_used_dirs_count = - cpu_to_le16(le16_to_cpu(gdp->bg_used_dirs_count) + 1); + le16_add_cpu(&gdp->bg_used_dirs_count, 1); } gdp->bg_checksum = ext4_group_desc_csum(sbi, group, gdp); spin_unlock(sb_bgl_lock(sbi, group)); diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index aaaccf5986f9..4865d318d7e1 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -3099,9 +3099,7 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, ac->ac_b_ex.fe_group, gdp)); } - gdp->bg_free_blocks_count = - cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) - - ac->ac_b_ex.fe_len); + le16_add_cpu(&gdp->bg_free_blocks_count, -ac->ac_b_ex.fe_len); gdp->bg_checksum = ext4_group_desc_csum(sbi, ac->ac_b_ex.fe_group, gdp); spin_unlock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group)); percpu_counter_sub(&sbi->s_freeblocks_counter, ac->ac_b_ex.fe_len); @@ -4593,8 +4591,7 @@ do_more: } spin_lock(sb_bgl_lock(sbi, block_group)); - gdp->bg_free_blocks_count = - cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) + count); + le16_add_cpu(&gdp->bg_free_blocks_count, count); gdp->bg_checksum = ext4_group_desc_csum(sbi, block_group, gdp); spin_unlock(sb_bgl_lock(sbi, block_group)); percpu_counter_add(&sbi->s_freeblocks_counter, count); diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index e29efa0f9d62..728e3efa84b5 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -502,8 +502,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, EXT4_SB(sb)->s_gdb_count++; kfree(o_group_desc); - es->s_reserved_gdt_blocks = - cpu_to_le16(le16_to_cpu(es->s_reserved_gdt_blocks) - 1); + le16_add_cpu(&es->s_reserved_gdt_blocks, -1); ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); return 0; @@ -877,8 +876,7 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) */ ext4_blocks_count_set(es, ext4_blocks_count(es) + input->blocks_count); - es->s_inodes_count = cpu_to_le32(le32_to_cpu(es->s_inodes_count) + - EXT4_INODES_PER_GROUP(sb)); + le32_add_cpu(&es->s_inodes_count, EXT4_INODES_PER_GROUP(sb)); /* * We need to protect s_groups_count against other CPUs seeing diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 9d2d9a7fdea1..9c0a3448ec6b 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1392,7 +1392,7 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, #endif if (!(__s16) le16_to_cpu(es->s_max_mnt_count)) es->s_max_mnt_count = cpu_to_le16(EXT4_DFL_MAX_MNT_COUNT); - es->s_mnt_count=cpu_to_le16(le16_to_cpu(es->s_mnt_count) + 1); + le16_add_cpu(&es->s_mnt_count, 1); es->s_mtime = cpu_to_le32(get_seconds()); ext4_update_dynamic_rev(sb); EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index e9054c1c7d93..726716b618d8 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -484,8 +484,7 @@ ext4_xattr_release_block(handle_t *handle, struct inode *inode, get_bh(bh); ext4_forget(handle, 1, inode, bh, bh->b_blocknr); } else { - BHDR(bh)->h_refcount = cpu_to_le32( - le32_to_cpu(BHDR(bh)->h_refcount) - 1); + le32_add_cpu(&BHDR(bh)->h_refcount, -1); error = ext4_journal_dirty_metadata(handle, bh); if (IS_SYNC(inode)) handle->h_sync = 1; @@ -789,8 +788,7 @@ inserted: if (error) goto cleanup_dquot; lock_buffer(new_bh); - BHDR(new_bh)->h_refcount = cpu_to_le32(1 + - le32_to_cpu(BHDR(new_bh)->h_refcount)); + le32_add_cpu(&BHDR(new_bh)->h_refcount, 1); ea_bdebug(new_bh, "reusing; refcount now=%d", le32_to_cpu(BHDR(new_bh)->h_refcount)); unlock_buffer(new_bh); -- cgit v1.2.3 From 216c34b2b8a3687afed4d269acec140c8baf23fe Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: ext4: convert byte order of constant instead of variable Convert byte order of constant instead of variable which can be done at compile time (vs run time). Signed-off-by: Marcin Slusarz Cc: Signed-off-by: Andrew Morton Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 9c0a3448ec6b..01d163964e69 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1388,7 +1388,7 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, * a plain journaled filesystem we can keep it set as * valid forever! :) */ - es->s_state = cpu_to_le16(le16_to_cpu(es->s_state) & ~EXT4_VALID_FS); + es->s_state &= cpu_to_le16(~EXT4_VALID_FS); #endif if (!(__s16) le16_to_cpu(es->s_max_mnt_count)) es->s_max_mnt_count = cpu_to_le16(EXT4_DFL_MAX_MNT_COUNT); -- cgit v1.2.3 From d00a6d7b40b44ee6b03f492a6c58f5bc4649c784 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: ext4: use ext4_group_first_block_no() Use ext4_group_first_block_no() and assign the return values to ext4_fsblk_t variables. Signed-off-by: Akinobu Mita Signed-off-by: "Theodore Ts'o" Cc: Stephen Tweedie Cc: adilger@clusterfs.com Cc: Andrew Morton Cc: Mingming Cao --- fs/ext4/balloc.c | 6 +++--- fs/ext4/xattr.c | 6 ++---- 2 files changed, 5 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 5d348998ff38..a080d7f5face 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -48,7 +48,6 @@ void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr, unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, ext4_group_t block_group, struct ext4_group_desc *gdp) { - unsigned long start; int bit, bit_max; unsigned free_blocks, group_blocks; struct ext4_sb_info *sbi = EXT4_SB(sb); @@ -106,11 +105,12 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, free_blocks = group_blocks - bit_max; if (bh) { + ext4_fsblk_t start; + for (bit = 0; bit < bit_max; bit++) ext4_set_bit(bit, bh->b_data); - start = block_group * EXT4_BLOCKS_PER_GROUP(sb) + - le32_to_cpu(sbi->s_es->s_first_data_block); + start = ext4_group_first_block_no(sb, block_group); /* Set bits for block and inode bitmaps, and inode table */ ext4_set_bit(ext4_block_bitmap(sb, gdp) - start, bh->b_data); diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 726716b618d8..bc1d7daac8eb 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -806,10 +806,8 @@ inserted: get_bh(new_bh); } else { /* We need to allocate a new block */ - ext4_fsblk_t goal = le32_to_cpu( - EXT4_SB(sb)->s_es->s_first_data_block) + - (ext4_fsblk_t)EXT4_I(inode)->i_block_group * - EXT4_BLOCKS_PER_GROUP(sb); + ext4_fsblk_t goal = ext4_group_first_block_no(sb, + EXT4_I(inode)->i_block_group); ext4_fsblk_t block = ext4_new_block(handle, inode, goal, &error); if (error) -- cgit v1.2.3 From c0a4ef38ac90d9053fcf3e22f81520a507c1a7bd Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: ext4: use ext4_get_group_desc() Use ext4_get_group_desc() in ext4_get_inode_block() instead of open coding the functionality. Signed-off-by: Akinobu Mita Signed-off-by: "Theodore Ts'o" Cc: Stephen Tweedie Cc: adilger@clusterfs.com Cc: Andrew Morton Cc: Mingming Cao --- fs/ext4/inode.c | 20 +++----------------- 1 file changed, 3 insertions(+), 17 deletions(-) (limited to 'fs') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 24a2604dde7b..aa0fb985b15e 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2511,12 +2511,10 @@ out_stop: static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, unsigned long ino, struct ext4_iloc *iloc) { - unsigned long desc, group_desc; ext4_group_t block_group; unsigned long offset; ext4_fsblk_t block; - struct buffer_head *bh; - struct ext4_group_desc * gdp; + struct ext4_group_desc *gdp; if (!ext4_valid_inum(sb, ino)) { /* @@ -2528,22 +2526,10 @@ static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, } block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); - if (block_group >= EXT4_SB(sb)->s_groups_count) { - ext4_error(sb,"ext4_get_inode_block","group >= groups count"); + gdp = ext4_get_group_desc(sb, block_group, NULL); + if (!gdp) return 0; - } - smp_rmb(); - group_desc = block_group >> EXT4_DESC_PER_BLOCK_BITS(sb); - desc = block_group & (EXT4_DESC_PER_BLOCK(sb) - 1); - bh = EXT4_SB(sb)->s_group_desc[group_desc]; - if (!bh) { - ext4_error (sb, "ext4_get_inode_block", - "Descriptor not loaded"); - return 0; - } - gdp = (struct ext4_group_desc *)((__u8 *)bh->b_data + - desc * EXT4_DESC_SIZE(sb)); /* * Figure out the offset within the block group inode table */ -- cgit v1.2.3 From a871611b474bfcdee422c0cf5d16f509dce096f5 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: ext4: check ext4_journal_get_write_access() errors Check ext4_journal_get_write_access() errors. Signed-off-by: Akinobu Mita Signed-off-by: "Theodore Ts'o" Cc: Stephen Tweedie Cc: adilger@clusterfs.com Cc: Andrew Morton Cc: Mingming Cao --- fs/ext4/namei.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 68611945687d..63c33e053478 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -57,10 +57,15 @@ static struct buffer_head *ext4_append(handle_t *handle, *block = inode->i_size >> inode->i_sb->s_blocksize_bits; - if ((bh = ext4_bread(handle, inode, *block, 1, err))) { + bh = ext4_bread(handle, inode, *block, 1, err); + if (bh) { inode->i_size += inode->i_sb->s_blocksize; EXT4_I(inode)->i_disksize = inode->i_size; - ext4_journal_get_write_access(handle,bh); + *err = ext4_journal_get_write_access(handle, bh); + if (*err) { + brelse(bh); + bh = NULL; + } } return bh; } -- cgit v1.2.3 From 14499f3592f3f52ceb7a639466de9ca21e2c1914 Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: ext4: remove extra define of ext4_new_blocks_old from mballoc.c The function prototype of ext4_new_blocks_old() is defined in ext4_fs.h, so we don't need the extra function prototype in mballoc.c Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/mballoc.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 4865d318d7e1..d9136b539c1e 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -576,8 +576,6 @@ static void ext4_mb_store_history(struct ext4_allocation_context *ac); static struct proc_dir_entry *proc_root_ext4; struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t); -ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, unsigned long *count, int *errp); static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap, ext4_group_t group); -- cgit v1.2.3 From d3a95d477d4fcb2c276b8357087a6c862c9e1949 Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: ext4: make ext4_xattr_list() static This patch makes the needlessly global ext4_xattr_list() static. Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/xattr.c | 4 +++- fs/ext4/xattr.h | 7 ------- 2 files changed, 3 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index bc1d7daac8eb..739930427b8a 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -92,6 +92,8 @@ static struct buffer_head *ext4_xattr_cache_find(struct inode *, struct mb_cache_entry **); static void ext4_xattr_rehash(struct ext4_xattr_header *, struct ext4_xattr_entry *); +static int ext4_xattr_list(struct inode *inode, char *buffer, + size_t buffer_size); static struct mb_cache *ext4_xattr_cache; @@ -420,7 +422,7 @@ cleanup: * Returns a negative error number on failure, or the number of bytes * used / required on success. */ -int +static int ext4_xattr_list(struct inode *inode, char *buffer, size_t buffer_size) { int i_error, b_error; diff --git a/fs/ext4/xattr.h b/fs/ext4/xattr.h index d7f5d6a12651..5992fe979bb9 100644 --- a/fs/ext4/xattr.h +++ b/fs/ext4/xattr.h @@ -74,7 +74,6 @@ extern struct xattr_handler ext4_xattr_security_handler; extern ssize_t ext4_listxattr(struct dentry *, char *, size_t); extern int ext4_xattr_get(struct inode *, int, const char *, void *, size_t); -extern int ext4_xattr_list(struct inode *, char *, size_t); extern int ext4_xattr_set(struct inode *, int, const char *, const void *, size_t, int); extern int ext4_xattr_set_handle(handle_t *, struct inode *, int, const char *, const void *, size_t, int); @@ -98,12 +97,6 @@ ext4_xattr_get(struct inode *inode, int name_index, const char *name, return -EOPNOTSUPP; } -static inline int -ext4_xattr_list(struct inode *inode, void *buffer, size_t size) -{ - return -EOPNOTSUPP; -} - static inline int ext4_xattr_set(struct inode *inode, int name_index, const char *name, const void *value, size_t size, int flags) -- cgit v1.2.3 From 9fa27c85de57d38ca698f4e34fdd1ab06b6c8e49 Mon Sep 17 00:00:00 2001 From: Duane Griffin Date: Mon, 28 Apr 2008 09:40:00 -0400 Subject: jbd2: tidy up revoke cache initialisation and destruction Make revocation cache destruction safe to call if initialisation fails partially or entirely. This allows it to be used to cleanup in the case of initialisation failure, simplifying the code slightly. Signed-off-by: Duane Griffin Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" Signed-off-by: Aneesh Kumar K.V --- fs/jbd2/revoke.c | 36 ++++++++++++++++++++++-------------- 1 file changed, 22 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 2e1453a5e998..72b260896bbd 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -167,33 +167,41 @@ static struct jbd2_revoke_record_s *find_revoke_record(journal_t *journal, return NULL; } +void jbd2_journal_destroy_revoke_caches(void) +{ + if (jbd2_revoke_record_cache) { + kmem_cache_destroy(jbd2_revoke_record_cache); + jbd2_revoke_record_cache = NULL; + } + if (jbd2_revoke_table_cache) { + kmem_cache_destroy(jbd2_revoke_table_cache); + jbd2_revoke_table_cache = NULL; + } +} + int __init jbd2_journal_init_revoke_caches(void) { + J_ASSERT(!jbd2_revoke_record_cache); + J_ASSERT(!jbd2_revoke_table_cache); + jbd2_revoke_record_cache = kmem_cache_create("jbd2_revoke_record", sizeof(struct jbd2_revoke_record_s), 0, SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY, NULL); if (!jbd2_revoke_record_cache) - return -ENOMEM; + goto record_cache_failure; jbd2_revoke_table_cache = kmem_cache_create("jbd2_revoke_table", sizeof(struct jbd2_revoke_table_s), 0, SLAB_TEMPORARY, NULL); - if (!jbd2_revoke_table_cache) { - kmem_cache_destroy(jbd2_revoke_record_cache); - jbd2_revoke_record_cache = NULL; - return -ENOMEM; - } + if (!jbd2_revoke_table_cache) + goto table_cache_failure; return 0; -} - -void jbd2_journal_destroy_revoke_caches(void) -{ - kmem_cache_destroy(jbd2_revoke_record_cache); - jbd2_revoke_record_cache = NULL; - kmem_cache_destroy(jbd2_revoke_table_cache); - jbd2_revoke_table_cache = NULL; +table_cache_failure: + jbd2_journal_destroy_revoke_caches(); +record_cache_failure: + return -ENOMEM; } /* Initialise the revoke table for a given journal to a given size. */ -- cgit v1.2.3 From 83c49523c91fff10493f5b3c102063b02ab76907 Mon Sep 17 00:00:00 2001 From: Duane Griffin Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: jbd2: eliminate duplicated code in revocation table init/destroy functions The revocation table initialisation/destruction code is repeated for each of the two revocation tables stored in the journal. Refactoring the duplicated code into functions is tidier, simplifies the logic in initialisation in particular, and slightly reduces the code size. There should not be any functional change. Signed-off-by: Duane Griffin Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/jbd2/revoke.c | 127 ++++++++++++++++++++++--------------------------------- 1 file changed, 51 insertions(+), 76 deletions(-) (limited to 'fs') diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 72b260896bbd..144a2065f03e 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -204,109 +204,84 @@ record_cache_failure: return -ENOMEM; } -/* Initialise the revoke table for a given journal to a given size. */ - -int jbd2_journal_init_revoke(journal_t *journal, int hash_size) +static struct jbd2_revoke_table_s *jbd2_journal_init_revoke_table(int hash_size) { - int shift, tmp; + int shift = 0; + int tmp = hash_size; + struct jbd2_revoke_table_s *table; - J_ASSERT (journal->j_revoke_table[0] == NULL); + table = kmem_cache_alloc(jbd2_revoke_table_cache, GFP_KERNEL); + if (!table) + goto out; - shift = 0; - tmp = hash_size; while((tmp >>= 1UL) != 0UL) shift++; - journal->j_revoke_table[0] = kmem_cache_alloc(jbd2_revoke_table_cache, GFP_KERNEL); - if (!journal->j_revoke_table[0]) - return -ENOMEM; - journal->j_revoke = journal->j_revoke_table[0]; - - /* Check that the hash_size is a power of two */ - J_ASSERT(is_power_of_2(hash_size)); - - journal->j_revoke->hash_size = hash_size; - - journal->j_revoke->hash_shift = shift; - - journal->j_revoke->hash_table = + table->hash_size = hash_size; + table->hash_shift = shift; + table->hash_table = kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL); - if (!journal->j_revoke->hash_table) { - kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[0]); - journal->j_revoke = NULL; - return -ENOMEM; + if (!table->hash_table) { + kmem_cache_free(jbd2_revoke_table_cache, table); + table = NULL; + goto out; } for (tmp = 0; tmp < hash_size; tmp++) - INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]); + INIT_LIST_HEAD(&table->hash_table[tmp]); - journal->j_revoke_table[1] = kmem_cache_alloc(jbd2_revoke_table_cache, GFP_KERNEL); - if (!journal->j_revoke_table[1]) { - kfree(journal->j_revoke_table[0]->hash_table); - kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[0]); - return -ENOMEM; +out: + return table; +} + +static void jbd2_journal_destroy_revoke_table(struct jbd2_revoke_table_s *table) +{ + int i; + struct list_head *hash_list; + + for (i = 0; i < table->hash_size; i++) { + hash_list = &table->hash_table[i]; + J_ASSERT(list_empty(hash_list)); } - journal->j_revoke = journal->j_revoke_table[1]; + kfree(table->hash_table); + kmem_cache_free(jbd2_revoke_table_cache, table); +} - /* Check that the hash_size is a power of two */ +/* Initialise the revoke table for a given journal to a given size. */ +int jbd2_journal_init_revoke(journal_t *journal, int hash_size) +{ + J_ASSERT(journal->j_revoke_table[0] == NULL); J_ASSERT(is_power_of_2(hash_size)); - journal->j_revoke->hash_size = hash_size; + journal->j_revoke_table[0] = jbd2_journal_init_revoke_table(hash_size); + if (!journal->j_revoke_table[0]) + goto fail0; - journal->j_revoke->hash_shift = shift; + journal->j_revoke_table[1] = jbd2_journal_init_revoke_table(hash_size); + if (!journal->j_revoke_table[1]) + goto fail1; - journal->j_revoke->hash_table = - kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL); - if (!journal->j_revoke->hash_table) { - kfree(journal->j_revoke_table[0]->hash_table); - kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[0]); - kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[1]); - journal->j_revoke = NULL; - return -ENOMEM; - } - - for (tmp = 0; tmp < hash_size; tmp++) - INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]); + journal->j_revoke = journal->j_revoke_table[1]; spin_lock_init(&journal->j_revoke_lock); return 0; -} -/* Destoy a journal's revoke table. The table must already be empty! */ +fail1: + jbd2_journal_destroy_revoke_table(journal->j_revoke_table[0]); +fail0: + return -ENOMEM; +} +/* Destroy a journal's revoke table. The table must already be empty! */ void jbd2_journal_destroy_revoke(journal_t *journal) { - struct jbd2_revoke_table_s *table; - struct list_head *hash_list; - int i; - - table = journal->j_revoke_table[0]; - if (!table) - return; - - for (i=0; ihash_size; i++) { - hash_list = &table->hash_table[i]; - J_ASSERT (list_empty(hash_list)); - } - - kfree(table->hash_table); - kmem_cache_free(jbd2_revoke_table_cache, table); - journal->j_revoke = NULL; - - table = journal->j_revoke_table[1]; - if (!table) - return; - - for (i=0; ihash_size; i++) { - hash_list = &table->hash_table[i]; - J_ASSERT (list_empty(hash_list)); - } - - kfree(table->hash_table); - kmem_cache_free(jbd2_revoke_table_cache, table); journal->j_revoke = NULL; + if (journal->j_revoke_table[0]) + jbd2_journal_destroy_revoke_table(journal->j_revoke_table[0]); + if (journal->j_revoke_table[1]) + jbd2_journal_destroy_revoke_table(journal->j_revoke_table[1]); } -- cgit v1.2.3 From 8a9362eb405e380432e6883cb83830df3b6cdf78 Mon Sep 17 00:00:00 2001 From: Duane Griffin Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: jbd2: replace potentially false assertion with if block If an error occurs during jbd2 cache initialisation it is possible for the journal_head_cache to be NULL when jbd2_journal_destroy_journal_head_cache is called. Replace the J_ASSERT with an if block to handle the situation correctly. Note that even with this fix things will break badly if jbd2 is statically compiled in and cache initialisation fails. Signed-off-by: Duane Griffin Cc: Signed-off-by: Andrew Morton Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/jbd2/journal.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index eb7eb6c27bcb..2dbf23e82c6f 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1976,9 +1976,10 @@ static int journal_init_jbd2_journal_head_cache(void) static void jbd2_journal_destroy_jbd2_journal_head_cache(void) { - J_ASSERT(jbd2_journal_head_cache != NULL); - kmem_cache_destroy(jbd2_journal_head_cache); - jbd2_journal_head_cache = NULL; + if (jbd2_journal_head_cache) { + kmem_cache_destroy(jbd2_journal_head_cache); + jbd2_journal_head_cache = NULL; + } } /* -- cgit v1.2.3 From 5648ba5b2dc0d07a8108fabc7b9100962e9e1d88 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: jbd2: fix kernel-doc notation Fix kernel-doc notation in jbd2. Signed-off-by: Randy Dunlap Signed-off-by: Mingming Cao Signed-off-by: Andrew Morton Signed-off-by: "Theodore Ts'o" --- fs/jbd2/journal.c | 5 +++-- fs/jbd2/transaction.c | 3 ++- 2 files changed, 5 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 2dbf23e82c6f..d707a219e21b 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -997,13 +997,14 @@ fail: */ /** - * journal_t * jbd2_journal_init_dev() - creates an initialises a journal structure + * journal_t * jbd2_journal_init_dev() - creates and initialises a journal structure * @bdev: Block device on which to create the journal * @fs_dev: Device which hold journalled filesystem for this journal. * @start: Block nr Start of journal. * @len: Length of the journal in blocks. * @blocksize: blocksize of journalling device - * @returns: a newly created journal_t * + * + * Returns: a newly created journal_t * * * jbd2_journal_init_dev creates a journal which maps a fixed contiguous * range of blocks on an arbitrary block device. diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 70245d6638b7..401bf9d23656 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -1462,7 +1462,8 @@ int jbd2_journal_stop(handle_t *handle) return err; } -/**int jbd2_journal_force_commit() - force any uncommitted transactions +/** + * int jbd2_journal_force_commit() - force any uncommitted transactions * @journal: journal to force * * For synchronous operations: force any uncommitted transactions -- cgit v1.2.3 From 620de4e19890c623eb4ba293ec19b42e2e391b89 Mon Sep 17 00:00:00 2001 From: Duane Griffin Date: Tue, 29 Apr 2008 22:02:47 -0400 Subject: jbd2: only create debugfs and stats entries if init is successful jbd2 debugfs and stats entries should only be created if cache initialisation is successful. At the moment they are being created unconditionally which will leave them dangling if cache (and hence module) initialisation fails. Signed-off-by: Duane Griffin Cc: Signed-off-by: Andrew Morton Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/jbd2/journal.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index d707a219e21b..64356e85a10f 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -2307,10 +2307,12 @@ static int __init journal_init(void) BUILD_BUG_ON(sizeof(struct journal_superblock_s) != 1024); ret = journal_init_caches(); - if (ret != 0) + if (ret == 0) { + jbd2_create_debugfs_entry(); + jbd2_create_jbd_stats_proc_entry(); + } else { jbd2_journal_destroy_caches(); - jbd2_create_debugfs_entry(); - jbd2_create_jbd_stats_proc_entry(); + } return ret; } -- cgit v1.2.3 From 46e665e9d297525d286989640cf4247cbe941df6 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: ext4: replace remaining __FUNCTION__ occurrences __FUNCTION__ is gcc-specific, use __func__ Signed-off-by: Harvey Harrison Cc: Signed-off-by: Andrew Morton Signed-off-by: "Theodore Ts'o" --- fs/ext4/balloc.c | 14 +++++------ fs/ext4/ext4_jbd2.c | 12 ++++----- fs/ext4/extents.c | 2 +- fs/ext4/ialloc.c | 10 ++++---- fs/ext4/inode.c | 8 +++--- fs/ext4/mballoc.c | 20 +++++++-------- fs/ext4/namei.c | 26 ++++++++++---------- fs/ext4/resize.c | 70 ++++++++++++++++++++++++++--------------------------- fs/ext4/super.c | 16 ++++++------ fs/ext4/xattr.c | 16 ++++++------ 10 files changed, 97 insertions(+), 97 deletions(-) (limited to 'fs') diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index a080d7f5face..af5032b23c29 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -58,7 +58,7 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, /* If checksum is bad mark all blocks used to prevent allocation * essentially implementing a per-group read-only flag. */ if (!ext4_group_desc_csum_verify(sbi, block_group, gdp)) { - ext4_error(sb, __FUNCTION__, + ext4_error(sb, __func__, "Checksum bad for group %lu\n", block_group); gdp->bg_free_blocks_count = 0; gdp->bg_free_inodes_count = 0; @@ -235,7 +235,7 @@ static int ext4_valid_block_bitmap(struct super_block *sb, return 1; err_out: - ext4_error(sb, __FUNCTION__, + ext4_error(sb, __func__, "Invalid block bitmap - " "block_group = %d, block = %llu", block_group, bitmap_blk); @@ -264,7 +264,7 @@ read_block_bitmap(struct super_block *sb, ext4_group_t block_group) bitmap_blk = ext4_block_bitmap(sb, desc); bh = sb_getblk(sb, bitmap_blk); if (unlikely(!bh)) { - ext4_error(sb, __FUNCTION__, + ext4_error(sb, __func__, "Cannot read block bitmap - " "block_group = %d, block_bitmap = %llu", (int)block_group, (unsigned long long)bitmap_blk); @@ -281,7 +281,7 @@ read_block_bitmap(struct super_block *sb, ext4_group_t block_group) } if (bh_submit_read(bh) < 0) { put_bh(bh); - ext4_error(sb, __FUNCTION__, + ext4_error(sb, __func__, "Cannot read block bitmap - " "block_group = %d, block_bitmap = %llu", (int)block_group, (unsigned long long)bitmap_blk); @@ -360,7 +360,7 @@ restart: BUG(); } #define rsv_window_dump(root, verbose) \ - __rsv_window_dump((root), (verbose), __FUNCTION__) + __rsv_window_dump((root), (verbose), __func__) #else #define rsv_window_dump(root, verbose) do {} while (0) #endif @@ -740,7 +740,7 @@ do_more: if (!ext4_clear_bit_atomic(sb_bgl_lock(sbi, block_group), bit + i, bitmap_bh->b_data)) { jbd_unlock_bh_state(bitmap_bh); - ext4_error(sb, __FUNCTION__, + ext4_error(sb, __func__, "bit already cleared for block %llu", (ext4_fsblk_t)(block + i)); jbd_lock_bh_state(bitmap_bh); @@ -1796,7 +1796,7 @@ allocated: if (ext4_test_bit(grp_alloc_blk+i, bh2jh(bitmap_bh)->b_committed_data)) { printk("%s: block was unexpectedly set in " - "b_committed_data\n", __FUNCTION__); + "b_committed_data\n", __func__); } } } diff --git a/fs/ext4/ext4_jbd2.c b/fs/ext4/ext4_jbd2.c index d6afe4e27340..2e6007d418dc 100644 --- a/fs/ext4/ext4_jbd2.c +++ b/fs/ext4/ext4_jbd2.c @@ -9,7 +9,7 @@ int __ext4_journal_get_undo_access(const char *where, handle_t *handle, { int err = jbd2_journal_get_undo_access(handle, bh); if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext4_journal_abort_handle(where, __func__, bh, handle, err); return err; } @@ -18,7 +18,7 @@ int __ext4_journal_get_write_access(const char *where, handle_t *handle, { int err = jbd2_journal_get_write_access(handle, bh); if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext4_journal_abort_handle(where, __func__, bh, handle, err); return err; } @@ -27,7 +27,7 @@ int __ext4_journal_forget(const char *where, handle_t *handle, { int err = jbd2_journal_forget(handle, bh); if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext4_journal_abort_handle(where, __func__, bh, handle, err); return err; } @@ -36,7 +36,7 @@ int __ext4_journal_revoke(const char *where, handle_t *handle, { int err = jbd2_journal_revoke(handle, blocknr, bh); if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext4_journal_abort_handle(where, __func__, bh, handle, err); return err; } @@ -45,7 +45,7 @@ int __ext4_journal_get_create_access(const char *where, { int err = jbd2_journal_get_create_access(handle, bh); if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext4_journal_abort_handle(where, __func__, bh, handle, err); return err; } @@ -54,6 +54,6 @@ int __ext4_journal_dirty_metadata(const char *where, { int err = jbd2_journal_dirty_metadata(handle, bh); if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext4_journal_abort_handle(where, __func__, bh, handle, err); return err; } diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 3d5a35f373d4..1c8aab4dad23 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -308,7 +308,7 @@ corrupted: } #define ext4_ext_check_header(inode, eh, depth) \ - __ext4_ext_check_header(__FUNCTION__, inode, eh, depth) + __ext4_ext_check_header(__func__, inode, eh, depth) #ifdef EXT_DEBUG static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path) diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index e8d24f24f282..a86377401ff0 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -75,7 +75,7 @@ unsigned ext4_init_inode_bitmap(struct super_block *sb, struct buffer_head *bh, /* If checksum is bad mark all blocks and inodes use to prevent * allocation, essentially implementing a per-group read-only flag. */ if (!ext4_group_desc_csum_verify(sbi, block_group, gdp)) { - ext4_error(sb, __FUNCTION__, "Checksum bad for group %lu\n", + ext4_error(sb, __func__, "Checksum bad for group %lu\n", block_group); gdp->bg_free_blocks_count = 0; gdp->bg_free_inodes_count = 0; @@ -586,7 +586,7 @@ got: ino++; if ((group == 0 && ino < EXT4_FIRST_INO(sb)) || ino > EXT4_INODES_PER_GROUP(sb)) { - ext4_error(sb, __FUNCTION__, + ext4_error(sb, __func__, "reserved inode or inode > inodes count - " "block_group = %lu, inode=%lu", group, ino + group * EXT4_INODES_PER_GROUP(sb)); @@ -792,7 +792,7 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino) /* Error cases - e2fsck has already cleaned up for us */ if (ino > max_ino) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "bad orphan ino %lu! e2fsck was run?", ino); goto error; } @@ -801,7 +801,7 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino) bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb); bitmap_bh = read_inode_bitmap(sb, block_group); if (!bitmap_bh) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "inode bitmap error for orphan %lu", ino); goto error; } @@ -826,7 +826,7 @@ iget_failed: err = PTR_ERR(inode); inode = NULL; bad_orphan: - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "bad orphan inode %lu! e2fsck was run?", ino); printk(KERN_NOTICE "ext4_test_bit(bit=%d, block=%llu) = %d\n", bit, (unsigned long long)bitmap_bh->b_blocknr, diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index aa0fb985b15e..bd1a391725c0 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -93,7 +93,7 @@ int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode, BUFFER_TRACE(bh, "call ext4_journal_revoke"); err = ext4_journal_revoke(handle, blocknr, bh); if (err) - ext4_abort(inode->i_sb, __FUNCTION__, + ext4_abort(inode->i_sb, __func__, "error %d when attempting revoke", err); BUFFER_TRACE(bh, "exit"); return err; @@ -1240,7 +1240,7 @@ int ext4_journal_dirty_data(handle_t *handle, struct buffer_head *bh) { int err = jbd2_journal_dirty_data(handle, bh); if (err) - ext4_journal_abort_handle(__FUNCTION__, __FUNCTION__, + ext4_journal_abort_handle(__func__, __func__, bh, handle, err); return err; } @@ -3371,7 +3371,7 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode) EXT4_I(inode)->i_state |= EXT4_STATE_NO_EXPAND; if (mnt_count != le16_to_cpu(sbi->s_es->s_mnt_count)) { - ext4_warning(inode->i_sb, __FUNCTION__, + ext4_warning(inode->i_sb, __func__, "Unable to expand inode %lu. Delete" " some EAs or run e2fsck.", inode->i_ino); @@ -3412,7 +3412,7 @@ void ext4_dirty_inode(struct inode *inode) current_handle->h_transaction != handle->h_transaction) { /* This task has a transaction open against a different fs */ printk(KERN_EMERG "%s: transactions do not match!\n", - __FUNCTION__); + __func__); } else { jbd_debug(5, "marking dirty. outer handle=%p\n", current_handle); diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index d9136b539c1e..0b46fc0ca196 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -734,7 +734,7 @@ static void mb_free_blocks_double(struct inode *inode, struct ext4_buddy *e4b, blocknr += le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); - ext4_error(sb, __FUNCTION__, "double-free of inode" + ext4_error(sb, __func__, "double-free of inode" " %lu's block %llu(bit %u in group %lu)\n", inode ? inode->i_ino : 0, blocknr, first + i, e4b->bd_group); @@ -906,7 +906,7 @@ static int __mb_check_buddy(struct ext4_buddy *e4b, char *file, } #undef MB_CHECK_ASSERT #define mb_check_buddy(e4b) __mb_check_buddy(e4b, \ - __FILE__, __FUNCTION__, __LINE__) + __FILE__, __func__, __LINE__) #else #define mb_check_buddy(e4b) #endif @@ -980,7 +980,7 @@ static void ext4_mb_generate_buddy(struct super_block *sb, grp->bb_fragments = fragments; if (free != grp->bb_free) { - ext4_error(sb, __FUNCTION__, + ext4_error(sb, __func__, "EXT4-fs: group %lu: %u blocks in bitmap, %u in gd\n", group, free, grp->bb_free); /* @@ -1366,7 +1366,7 @@ static int mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b, blocknr += le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); - ext4_error(sb, __FUNCTION__, "double-free of inode" + ext4_error(sb, __func__, "double-free of inode" " %lu's block %llu(bit %u in group %lu)\n", inode ? inode->i_ino : 0, blocknr, block, e4b->bd_group); @@ -1847,7 +1847,7 @@ static void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac, * free blocks even though group info says we * we have free blocks */ - ext4_error(sb, __FUNCTION__, "%d free blocks as per " + ext4_error(sb, __func__, "%d free blocks as per " "group info. But bitmap says 0\n", free); break; @@ -1856,7 +1856,7 @@ static void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac, mb_find_extent(e4b, 0, i, ac->ac_g_ex.fe_len, &ex); BUG_ON(ex.fe_len <= 0); if (free < ex.fe_len) { - ext4_error(sb, __FUNCTION__, "%d free blocks as per " + ext4_error(sb, __func__, "%d free blocks as per " "group info. But got %d blocks\n", free, ex.fe_len); /* @@ -3073,7 +3073,7 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, in_range(block, ext4_inode_table(sb, gdp), EXT4_SB(sb)->s_itb_per_group)) { - ext4_error(sb, __FUNCTION__, + ext4_error(sb, __func__, "Allocating block in system zone - block = %llu", block); } @@ -3786,7 +3786,7 @@ ext4_mb_release_inode_pa(struct ext4_buddy *e4b, struct buffer_head *bitmap_bh, pa, (unsigned long) pa->pa_lstart, (unsigned long) pa->pa_pstart, (unsigned long) pa->pa_len); - ext4_error(sb, __FUNCTION__, "free %u, pa_free %u\n", + ext4_error(sb, __func__, "free %u, pa_free %u\n", free, pa->pa_free); /* * pa is already deleted so we use the value obtained @@ -4490,7 +4490,7 @@ void ext4_mb_free_blocks(handle_t *handle, struct inode *inode, if (block < le32_to_cpu(es->s_first_data_block) || block + count < block || block + count > ext4_blocks_count(es)) { - ext4_error(sb, __FUNCTION__, + ext4_error(sb, __func__, "Freeing blocks not in datazone - " "block = %lu, count = %lu", block, count); goto error_return; @@ -4531,7 +4531,7 @@ do_more: in_range(block + count - 1, ext4_inode_table(sb, gdp), EXT4_SB(sb)->s_itb_per_group)) { - ext4_error(sb, __FUNCTION__, + ext4_error(sb, __func__, "Freeing blocks in system zone - " "Block = %lu, count = %lu", block, count); } diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 63c33e053478..02cdaec39e21 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -353,7 +353,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, if (root->info.hash_version != DX_HASH_TEA && root->info.hash_version != DX_HASH_HALF_MD4 && root->info.hash_version != DX_HASH_LEGACY) { - ext4_warning(dir->i_sb, __FUNCTION__, + ext4_warning(dir->i_sb, __func__, "Unrecognised inode hash code %d", root->info.hash_version); brelse(bh); @@ -367,7 +367,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, hash = hinfo->hash; if (root->info.unused_flags & 1) { - ext4_warning(dir->i_sb, __FUNCTION__, + ext4_warning(dir->i_sb, __func__, "Unimplemented inode hash flags: %#06x", root->info.unused_flags); brelse(bh); @@ -376,7 +376,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, } if ((indirect = root->info.indirect_levels) > 1) { - ext4_warning(dir->i_sb, __FUNCTION__, + ext4_warning(dir->i_sb, __func__, "Unimplemented inode hash depth: %#06x", root->info.indirect_levels); brelse(bh); @@ -389,7 +389,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, if (dx_get_limit(entries) != dx_root_limit(dir, root->info.info_length)) { - ext4_warning(dir->i_sb, __FUNCTION__, + ext4_warning(dir->i_sb, __func__, "dx entry: limit != root limit"); brelse(bh); *err = ERR_BAD_DX_DIR; @@ -401,7 +401,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, { count = dx_get_count(entries); if (!count || count > dx_get_limit(entries)) { - ext4_warning(dir->i_sb, __FUNCTION__, + ext4_warning(dir->i_sb, __func__, "dx entry: no count or count > limit"); brelse(bh); *err = ERR_BAD_DX_DIR; @@ -446,7 +446,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, goto fail2; at = entries = ((struct dx_node *) bh->b_data)->entries; if (dx_get_limit(entries) != dx_node_limit (dir)) { - ext4_warning(dir->i_sb, __FUNCTION__, + ext4_warning(dir->i_sb, __func__, "dx entry: limit != node limit"); brelse(bh); *err = ERR_BAD_DX_DIR; @@ -462,7 +462,7 @@ fail2: } fail: if (*err == ERR_BAD_DX_DIR) - ext4_warning(dir->i_sb, __FUNCTION__, + ext4_warning(dir->i_sb, __func__, "Corrupt dir inode %ld, running e2fsck is " "recommended.", dir->i_ino); return NULL; @@ -919,7 +919,7 @@ restart: wait_on_buffer(bh); if (!buffer_uptodate(bh)) { /* read error, skip block & hope for the best */ - ext4_error(sb, __FUNCTION__, "reading directory #%lu " + ext4_error(sb, __func__, "reading directory #%lu " "offset %lu", dir->i_ino, (unsigned long)block); brelse(bh); @@ -1012,7 +1012,7 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, retval = ext4_htree_next_block(dir, hash, frame, frames, NULL); if (retval < 0) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "error reading index page in directory #%lu", dir->i_ino); *err = retval; @@ -1537,7 +1537,7 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, if (levels && (dx_get_count(frames->entries) == dx_get_limit(frames->entries))) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Directory index full!"); err = -ENOSPC; goto cleanup; @@ -1865,11 +1865,11 @@ static int empty_dir (struct inode * inode) if (inode->i_size < EXT4_DIR_REC_LEN(1) + EXT4_DIR_REC_LEN(2) || !(bh = ext4_bread (NULL, inode, 0, 0, &err))) { if (err) - ext4_error(inode->i_sb, __FUNCTION__, + ext4_error(inode->i_sb, __func__, "error %d reading directory #%lu offset 0", err, inode->i_ino); else - ext4_warning(inode->i_sb, __FUNCTION__, + ext4_warning(inode->i_sb, __func__, "bad directory (dir #%lu) - no data block", inode->i_ino); return 1; @@ -1898,7 +1898,7 @@ static int empty_dir (struct inode * inode) offset >> EXT4_BLOCK_SIZE_BITS(sb), 0, &err); if (!bh) { if (err) - ext4_error(sb, __FUNCTION__, + ext4_error(sb, __func__, "error %d reading directory" " #%lu offset %lu", err, inode->i_ino, offset); diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 728e3efa84b5..3e0f5d06f3ee 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -50,63 +50,63 @@ static int verify_group_input(struct super_block *sb, ext4_get_group_no_and_offset(sb, start, NULL, &offset); if (group != sbi->s_groups_count) - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Cannot add at group %u (only %lu groups)", input->group, sbi->s_groups_count); else if (offset != 0) - ext4_warning(sb, __FUNCTION__, "Last group not full"); + ext4_warning(sb, __func__, "Last group not full"); else if (input->reserved_blocks > input->blocks_count / 5) - ext4_warning(sb, __FUNCTION__, "Reserved blocks too high (%u)", + ext4_warning(sb, __func__, "Reserved blocks too high (%u)", input->reserved_blocks); else if (free_blocks_count < 0) - ext4_warning(sb, __FUNCTION__, "Bad blocks count %u", + ext4_warning(sb, __func__, "Bad blocks count %u", input->blocks_count); else if (!(bh = sb_bread(sb, end - 1))) - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Cannot read last block (%llu)", end - 1); else if (outside(input->block_bitmap, start, end)) - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Block bitmap not in group (block %llu)", (unsigned long long)input->block_bitmap); else if (outside(input->inode_bitmap, start, end)) - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Inode bitmap not in group (block %llu)", (unsigned long long)input->inode_bitmap); else if (outside(input->inode_table, start, end) || outside(itend - 1, start, end)) - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Inode table not in group (blocks %llu-%llu)", (unsigned long long)input->inode_table, itend - 1); else if (input->inode_bitmap == input->block_bitmap) - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Block bitmap same as inode bitmap (%llu)", (unsigned long long)input->block_bitmap); else if (inside(input->block_bitmap, input->inode_table, itend)) - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Block bitmap (%llu) in inode table (%llu-%llu)", (unsigned long long)input->block_bitmap, (unsigned long long)input->inode_table, itend - 1); else if (inside(input->inode_bitmap, input->inode_table, itend)) - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Inode bitmap (%llu) in inode table (%llu-%llu)", (unsigned long long)input->inode_bitmap, (unsigned long long)input->inode_table, itend - 1); else if (inside(input->block_bitmap, start, metaend)) - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Block bitmap (%llu) in GDT table" " (%llu-%llu)", (unsigned long long)input->block_bitmap, start, metaend - 1); else if (inside(input->inode_bitmap, start, metaend)) - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Inode bitmap (%llu) in GDT table" " (%llu-%llu)", (unsigned long long)input->inode_bitmap, start, metaend - 1); else if (inside(input->inode_table, start, metaend) || inside(itend - 1, start, metaend)) - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Inode table (%llu-%llu) overlaps" "GDT table (%llu-%llu)", (unsigned long long)input->inode_table, @@ -368,7 +368,7 @@ static int verify_reserved_gdb(struct super_block *sb, while ((grp = ext4_list_backups(sb, &three, &five, &seven)) < end) { if (le32_to_cpu(*p++) != grp * EXT4_BLOCKS_PER_GROUP(sb) + blk){ - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "reserved GDT %llu" " missing grp %d (%llu)", blk, grp, @@ -424,7 +424,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, */ if (EXT4_SB(sb)->s_sbh->b_blocknr != le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block)) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "won't resize using backup superblock at %llu", (unsigned long long)EXT4_SB(sb)->s_sbh->b_blocknr); return -EPERM; @@ -448,7 +448,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, data = (__le32 *)dind->b_data; if (le32_to_cpu(data[gdb_num % EXT4_ADDR_PER_BLOCK(sb)]) != gdblock) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "new group %u GDT block %llu not reserved", input->group, gdblock); err = -EINVAL; @@ -472,7 +472,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, GFP_KERNEL); if (!n_group_desc) { err = -ENOMEM; - ext4_warning (sb, __FUNCTION__, + ext4_warning(sb, __func__, "not enough memory for %lu groups", gdb_num + 1); goto exit_inode; } @@ -570,7 +570,7 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode, /* Get each reserved primary GDT block and verify it holds backups */ for (res = 0; res < reserved_gdb; res++, blk++) { if (le32_to_cpu(*data) != blk) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "reserved block %llu" " not at offset %ld", blk, @@ -714,7 +714,7 @@ static void update_backups(struct super_block *sb, */ exit_err: if (err) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "can't update backup for group %lu (err %d), " "forcing fsck on next reboot", group, err); sbi->s_mount_state &= ~EXT4_VALID_FS; @@ -754,33 +754,33 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) if (gdb_off == 0 && !EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER)) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Can't resize non-sparse filesystem further"); return -EPERM; } if (ext4_blocks_count(es) + input->blocks_count < ext4_blocks_count(es)) { - ext4_warning(sb, __FUNCTION__, "blocks_count overflow\n"); + ext4_warning(sb, __func__, "blocks_count overflow\n"); return -EINVAL; } if (le32_to_cpu(es->s_inodes_count) + EXT4_INODES_PER_GROUP(sb) < le32_to_cpu(es->s_inodes_count)) { - ext4_warning(sb, __FUNCTION__, "inodes_count overflow\n"); + ext4_warning(sb, __func__, "inodes_count overflow\n"); return -EINVAL; } if (reserved_gdb || gdb_off == 0) { if (!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_RESIZE_INODE)){ - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "No reserved GDT blocks, can't resize"); return -EPERM; } inode = ext4_iget(sb, EXT4_RESIZE_INO); if (IS_ERR(inode)) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "Error opening resize inode"); return PTR_ERR(inode); } @@ -809,7 +809,7 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) lock_super(sb); if (input->group != sbi->s_groups_count) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "multiple resizers run on filesystem!"); err = -EBUSY; goto exit_journal; @@ -975,13 +975,13 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, " too large to resize to %llu blocks safely\n", sb->s_id, n_blocks_count); if (sizeof(sector_t) < 8) - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "CONFIG_LBD not enabled\n"); return -EINVAL; } if (n_blocks_count < o_blocks_count) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "can't shrink FS - resize aborted"); return -EBUSY; } @@ -990,7 +990,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, ext4_get_group_no_and_offset(sb, o_blocks_count, NULL, &last); if (last == 0) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "need to use ext2online to resize further"); return -EPERM; } @@ -998,7 +998,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, add = EXT4_BLOCKS_PER_GROUP(sb) - last; if (o_blocks_count + add < o_blocks_count) { - ext4_warning(sb, __FUNCTION__, "blocks_count overflow"); + ext4_warning(sb, __func__, "blocks_count overflow"); return -EINVAL; } @@ -1006,7 +1006,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, add = n_blocks_count - o_blocks_count; if (o_blocks_count + add < n_blocks_count) - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "will only finish group (%llu" " blocks, %u new)", o_blocks_count + add, add); @@ -1014,7 +1014,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, /* See if the device is actually as big as what was requested */ bh = sb_bread(sb, o_blocks_count + add -1); if (!bh) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "can't read last block, resize aborted"); return -ENOSPC; } @@ -1026,13 +1026,13 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, handle = ext4_journal_start_sb(sb, 3); if (IS_ERR(handle)) { err = PTR_ERR(handle); - ext4_warning(sb, __FUNCTION__, "error %d on journal start",err); + ext4_warning(sb, __func__, "error %d on journal start", err); goto exit_put; } lock_super(sb); if (o_blocks_count != ext4_blocks_count(es)) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "multiple resizers run on filesystem!"); unlock_super(sb); ext4_journal_stop(handle); @@ -1042,7 +1042,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, if ((err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh))) { - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "error %d on journal write access", err); unlock_super(sb); ext4_journal_stop(handle); diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 01d163964e69..93a7e746f425 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -135,7 +135,7 @@ handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks) * take the FS itself readonly cleanly. */ journal = EXT4_SB(sb)->s_journal; if (is_journal_aborted(journal)) { - ext4_abort(sb, __FUNCTION__, + ext4_abort(sb, __func__, "Detected aborted journal"); return ERR_PTR(-EROFS); } @@ -355,7 +355,7 @@ void ext4_update_dynamic_rev(struct super_block *sb) if (le32_to_cpu(es->s_rev_level) > EXT4_GOOD_OLD_REV) return; - ext4_warning(sb, __FUNCTION__, + ext4_warning(sb, __func__, "updating to rev %d because of new feature flag, " "running e2fsck is recommended", EXT4_DYNAMIC_REV); @@ -1511,7 +1511,7 @@ static int ext4_check_descriptors(struct super_block *sb) return 0; } if (!ext4_group_desc_csum_verify(sbi, i, gdp)) { - ext4_error(sb, __FUNCTION__, + ext4_error(sb, __func__, "Checksum for group %lu failed (%u!=%u)\n", i, le16_to_cpu(ext4_group_desc_csum(sbi, i, gdp)), le16_to_cpu(gdp->bg_checksum)); @@ -1605,7 +1605,7 @@ static void ext4_orphan_cleanup (struct super_block * sb, if (inode->i_nlink) { printk(KERN_DEBUG "%s: truncating inode %lu to %Ld bytes\n", - __FUNCTION__, inode->i_ino, inode->i_size); + __func__, inode->i_ino, inode->i_size); jbd_debug(2, "truncating inode %lu to %Ld bytes\n", inode->i_ino, inode->i_size); ext4_truncate(inode); @@ -1613,7 +1613,7 @@ static void ext4_orphan_cleanup (struct super_block * sb, } else { printk(KERN_DEBUG "%s: deleting unreferenced inode %lu\n", - __FUNCTION__, inode->i_ino); + __func__, inode->i_ino); jbd_debug(2, "deleting unreferenced inode %lu\n", inode->i_ino); nr_orphans++; @@ -2699,9 +2699,9 @@ static void ext4_clear_journal_err(struct super_block * sb, char nbuf[16]; errstr = ext4_decode_error(sb, j_errno, nbuf); - ext4_warning(sb, __FUNCTION__, "Filesystem error recorded " + ext4_warning(sb, __func__, "Filesystem error recorded " "from previous mount: %s", errstr); - ext4_warning(sb, __FUNCTION__, "Marking fs in need of " + ext4_warning(sb, __func__, "Marking fs in need of " "filesystem check."); EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS; @@ -2828,7 +2828,7 @@ static int ext4_remount (struct super_block * sb, int * flags, char * data) } if (sbi->s_mount_opt & EXT4_MOUNT_ABORT) - ext4_abort(sb, __FUNCTION__, "Abort forced by user"); + ext4_abort(sb, __func__, "Abort forced by user"); sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | ((sbi->s_mount_opt & EXT4_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 739930427b8a..f56598df6880 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -227,7 +227,7 @@ ext4_xattr_block_get(struct inode *inode, int name_index, const char *name, ea_bdebug(bh, "b_count=%d, refcount=%d", atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount)); if (ext4_xattr_check_block(bh)) { -bad_block: ext4_error(inode->i_sb, __FUNCTION__, +bad_block: ext4_error(inode->i_sb, __func__, "inode %lu: bad block %llu", inode->i_ino, EXT4_I(inode)->i_file_acl); error = -EIO; @@ -369,7 +369,7 @@ ext4_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size) ea_bdebug(bh, "b_count=%d, refcount=%d", atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount)); if (ext4_xattr_check_block(bh)) { - ext4_error(inode->i_sb, __FUNCTION__, + ext4_error(inode->i_sb, __func__, "inode %lu: bad block %llu", inode->i_ino, EXT4_I(inode)->i_file_acl); error = -EIO; @@ -661,7 +661,7 @@ ext4_xattr_block_find(struct inode *inode, struct ext4_xattr_info *i, atomic_read(&(bs->bh->b_count)), le32_to_cpu(BHDR(bs->bh)->h_refcount)); if (ext4_xattr_check_block(bs->bh)) { - ext4_error(sb, __FUNCTION__, + ext4_error(sb, __func__, "inode %lu: bad block %llu", inode->i_ino, EXT4_I(inode)->i_file_acl); error = -EIO; @@ -861,7 +861,7 @@ cleanup_dquot: goto cleanup; bad_block: - ext4_error(inode->i_sb, __FUNCTION__, + ext4_error(inode->i_sb, __func__, "inode %lu: bad block %llu", inode->i_ino, EXT4_I(inode)->i_file_acl); goto cleanup; @@ -1164,7 +1164,7 @@ retry: if (!bh) goto cleanup; if (ext4_xattr_check_block(bh)) { - ext4_error(inode->i_sb, __FUNCTION__, + ext4_error(inode->i_sb, __func__, "inode %lu: bad block %llu", inode->i_ino, EXT4_I(inode)->i_file_acl); error = -EIO; @@ -1339,14 +1339,14 @@ ext4_xattr_delete_inode(handle_t *handle, struct inode *inode) goto cleanup; bh = sb_bread(inode->i_sb, EXT4_I(inode)->i_file_acl); if (!bh) { - ext4_error(inode->i_sb, __FUNCTION__, + ext4_error(inode->i_sb, __func__, "inode %lu: block %llu read error", inode->i_ino, EXT4_I(inode)->i_file_acl); goto cleanup; } if (BHDR(bh)->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC) || BHDR(bh)->h_blocks != cpu_to_le32(1)) { - ext4_error(inode->i_sb, __FUNCTION__, + ext4_error(inode->i_sb, __func__, "inode %lu: bad block %llu", inode->i_ino, EXT4_I(inode)->i_file_acl); goto cleanup; @@ -1473,7 +1473,7 @@ again: } bh = sb_bread(inode->i_sb, ce->e_block); if (!bh) { - ext4_error(inode->i_sb, __FUNCTION__, + ext4_error(inode->i_sb, __func__, "inode %lu: block %lu read error", inode->i_ino, (unsigned long) ce->e_block); } else if (le32_to_cpu(BHDR(bh)->h_refcount) >= -- cgit v1.2.3 From 329d291f50d53f77d15769051f3eb494a9fd54b7 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Thu, 17 Apr 2008 10:38:59 -0400 Subject: jdb2: replace remaining __FUNCTION__ occurrences __FUNCTION__ is gcc-specific, use __func__ Signed-off-by: Harvey Harrison Cc: Signed-off-by: Andrew Morton Signed-off-by: "Theodore Ts'o" --- fs/jbd2/journal.c | 18 +++++++++--------- fs/jbd2/revoke.c | 2 +- fs/jbd2/transaction.c | 4 ++-- 3 files changed, 12 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 64356e85a10f..53632e3e8457 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -534,7 +534,7 @@ int jbd2_log_wait_commit(journal_t *journal, tid_t tid) if (!tid_geq(journal->j_commit_request, tid)) { printk(KERN_EMERG "%s: error: j_commit_request=%d, tid=%d\n", - __FUNCTION__, journal->j_commit_request, tid); + __func__, journal->j_commit_request, tid); } spin_unlock(&journal->j_state_lock); #endif @@ -599,7 +599,7 @@ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr, printk(KERN_ALERT "%s: journal block not found " "at offset %lu on %s\n", - __FUNCTION__, + __func__, blocknr, bdevname(journal->j_dev, b)); err = -EIO; @@ -1028,7 +1028,7 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev, journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL); if (!journal->j_wbuf) { printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n", - __FUNCTION__); + __func__); kfree(journal); journal = NULL; goto out; @@ -1084,7 +1084,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode) journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL); if (!journal->j_wbuf) { printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n", - __FUNCTION__); + __func__); kfree(journal); return NULL; } @@ -1093,7 +1093,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode) /* If that failed, give up */ if (err) { printk(KERN_ERR "%s: Cannnot locate journal superblock\n", - __FUNCTION__); + __func__); kfree(journal); return NULL; } @@ -1179,7 +1179,7 @@ int jbd2_journal_create(journal_t *journal) */ printk(KERN_EMERG "%s: creation of journal on external device!\n", - __FUNCTION__); + __func__); BUG(); } @@ -1999,7 +1999,7 @@ static struct journal_head *journal_alloc_journal_head(void) jbd_debug(1, "out of memory for journal_head\n"); if (time_after(jiffies, last_warning + 5*HZ)) { printk(KERN_NOTICE "ENOMEM in %s, retrying.\n", - __FUNCTION__); + __func__); last_warning = jiffies; } while (!ret) { @@ -2136,13 +2136,13 @@ static void __journal_remove_journal_head(struct buffer_head *bh) if (jh->b_frozen_data) { printk(KERN_WARNING "%s: freeing " "b_frozen_data\n", - __FUNCTION__); + __func__); jbd2_free(jh->b_frozen_data, bh->b_size); } if (jh->b_committed_data) { printk(KERN_WARNING "%s: freeing " "b_committed_data\n", - __FUNCTION__); + __func__); jbd2_free(jh->b_committed_data, bh->b_size); } bh->b_private = NULL; diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 144a2065f03e..257ff2625765 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -139,7 +139,7 @@ repeat: oom: if (!journal_oom_retry) return -ENOMEM; - jbd_debug(1, "ENOMEM in %s, retrying\n", __FUNCTION__); + jbd_debug(1, "ENOMEM in %s, retrying\n", __func__); yield(); goto repeat; } diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 401bf9d23656..d6e006e67804 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -696,7 +696,7 @@ repeat: if (!frozen_buffer) { printk(KERN_EMERG "%s: OOM for frozen_buffer\n", - __FUNCTION__); + __func__); JBUFFER_TRACE(jh, "oom!"); error = -ENOMEM; jbd_lock_bh_state(bh); @@ -914,7 +914,7 @@ repeat: committed_data = jbd2_alloc(jh2bh(jh)->b_size, GFP_NOFS); if (!committed_data) { printk(KERN_EMERG "%s: No memory for committed data\n", - __FUNCTION__); + __func__); err = -ENOMEM; goto out; } -- cgit v1.2.3 From 42173f6860af7e016a950a9a19a66679cfc46d98 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 22 Apr 2008 17:33:25 +1000 Subject: [XFS] Remove VN_IS* macros and related cruft. We can just check i_mode / di_mode directly. SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30896a Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/linux-2.6/xfs_vnode.h | 24 ------------------------ fs/xfs/xfs_acl.c | 6 +++--- fs/xfs/xfs_vnodeops.c | 20 ++++---------------- 3 files changed, 7 insertions(+), 43 deletions(-) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/xfs_vnode.h b/fs/xfs/linux-2.6/xfs_vnode.h index 8b4d63ce8694..9d73cb5c0fc7 100644 --- a/fs/xfs/linux-2.6/xfs_vnode.h +++ b/fs/xfs/linux-2.6/xfs_vnode.h @@ -25,12 +25,6 @@ struct attrlist_cursor_kern; typedef struct inode bhv_vnode_t; -#define VN_ISLNK(vp) S_ISLNK((vp)->i_mode) -#define VN_ISREG(vp) S_ISREG((vp)->i_mode) -#define VN_ISDIR(vp) S_ISDIR((vp)->i_mode) -#define VN_ISCHR(vp) S_ISCHR((vp)->i_mode) -#define VN_ISBLK(vp) S_ISBLK((vp)->i_mode) - /* * Vnode to Linux inode mapping. */ @@ -151,24 +145,6 @@ typedef struct bhv_vattr { XFS_AT_TYPE|XFS_AT_BLKSIZE|XFS_AT_NBLOCKS|XFS_AT_VCODE|\ XFS_AT_NEXTENTS|XFS_AT_ANEXTENTS|XFS_AT_GENCOUNT) -/* - * Modes. - */ -#define VSUID S_ISUID /* set user id on execution */ -#define VSGID S_ISGID /* set group id on execution */ -#define VSVTX S_ISVTX /* save swapped text even after use */ -#define VREAD S_IRUSR /* read, write, execute permissions */ -#define VWRITE S_IWUSR -#define VEXEC S_IXUSR - -#define MODEMASK S_IALLUGO /* mode bits plus permission bits */ - -/* - * Check whether mandatory file locking is enabled. - */ -#define MANDLOCK(vp, mode) \ - (VN_ISREG(vp) && ((mode) & (VSGID|(VEXEC>>3))) == VSGID) - extern void vn_init(void); extern int vn_revalidate(bhv_vnode_t *); diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c index 8e130b9720ae..b1275cc45617 100644 --- a/fs/xfs/xfs_acl.c +++ b/fs/xfs/xfs_acl.c @@ -72,7 +72,7 @@ xfs_acl_vhasacl_default( { int error; - if (!VN_ISDIR(vp)) + if (!S_ISDIR(vp->i_mode)) return 0; xfs_acl_get_attr(vp, NULL, _ACL_TYPE_DEFAULT, ATTR_KERNOVAL, &error); return (error == 0); @@ -379,7 +379,7 @@ xfs_acl_allow_set( if (vp->i_flags & (S_IMMUTABLE|S_APPEND)) return EPERM; - if (kind == _ACL_TYPE_DEFAULT && !VN_ISDIR(vp)) + if (kind == _ACL_TYPE_DEFAULT && !S_ISDIR(vp->i_mode)) return ENOTDIR; if (vp->i_sb->s_flags & MS_RDONLY) return EROFS; @@ -719,7 +719,7 @@ xfs_acl_inherit( * If the new file is a directory, its default ACL is a copy of * the containing directory's default ACL. */ - if (VN_ISDIR(vp)) + if (S_ISDIR(vp->i_mode)) xfs_acl_set_attr(vp, pdaclp, _ACL_TYPE_DEFAULT, &error); if (!error && !basicperms) xfs_acl_set_attr(vp, cacl, _ACL_TYPE_ACCESS, &error); diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 6650601c64f7..3fef54b11582 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -211,7 +211,6 @@ xfs_setattr( int flags, cred_t *credp) { - bhv_vnode_t *vp = XFS_ITOV(ip); xfs_mount_t *mp = ip->i_mount; xfs_trans_t *tp; int mask; @@ -222,7 +221,6 @@ xfs_setattr( gid_t gid=0, igid=0; int timeflags = 0; xfs_prid_t projid=0, iprojid=0; - int mandlock_before, mandlock_after; struct xfs_dquot *udqp, *gdqp, *olddquot1, *olddquot2; int file_owner; int need_iolock = 1; @@ -383,7 +381,7 @@ xfs_setattr( m |= S_ISGID; #if 0 /* Linux allows this, Irix doesn't. */ - if ((vap->va_mode & S_ISVTX) && !VN_ISDIR(vp)) + if ((vap->va_mode & S_ISVTX) && !S_ISDIR(ip->i_d.di_mode)) m |= S_ISVTX; #endif if (m && !capable(CAP_FSETID)) @@ -461,10 +459,10 @@ xfs_setattr( goto error_return; } - if (VN_ISDIR(vp)) { + if (S_ISDIR(ip->i_d.di_mode)) { code = XFS_ERROR(EISDIR); goto error_return; - } else if (!VN_ISREG(vp)) { + } else if (!S_ISREG(ip->i_d.di_mode)) { code = XFS_ERROR(EINVAL); goto error_return; } @@ -626,9 +624,6 @@ xfs_setattr( xfs_trans_ihold(tp, ip); } - /* determine whether mandatory locking mode changes */ - mandlock_before = MANDLOCK(vp, ip->i_d.di_mode); - /* * Truncate file. Must have write permission and not be a directory. */ @@ -858,13 +853,6 @@ xfs_setattr( code = xfs_trans_commit(tp, commit_flags); } - /* - * If the (regular) file's mandatory locking mode changed, then - * notify the vnode. We do this under the inode lock to prevent - * racing calls to vop_vnode_change. - */ - mandlock_after = MANDLOCK(vp, ip->i_d.di_mode); - xfs_iunlock(ip, lock_flags); /* @@ -1491,7 +1479,7 @@ xfs_release( xfs_mount_t *mp = ip->i_mount; int error; - if (!VN_ISREG(vp) || (ip->i_d.di_mode == 0)) + if (!S_ISREG(ip->i_d.di_mode) || (ip->i_d.di_mode == 0)) return 0; /* If this is a read-only mount, don't do this (would generate I/O) */ -- cgit v1.2.3 From 4e5dbb3498e74514b9936d691413afc55fb84ea9 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 22 Apr 2008 17:33:33 +1000 Subject: [XFS] kill xfs_getattr It's currently used by the ACL code to read di_mode/di_uid, but these are simple 32bit scalar values we can just read directly without locking. SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30897a Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_acl.c | 40 +++------------- fs/xfs/xfs_vnodeops.c | 126 -------------------------------------------------- fs/xfs/xfs_vnodeops.h | 1 - 3 files changed, 7 insertions(+), 160 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c index b1275cc45617..796e76ef2713 100644 --- a/fs/xfs/xfs_acl.c +++ b/fs/xfs/xfs_acl.c @@ -238,15 +238,8 @@ xfs_acl_vget( error = EINVAL; goto out; } - if (kind == _ACL_TYPE_ACCESS) { - bhv_vattr_t va; - - va.va_mask = XFS_AT_MODE; - error = xfs_getattr(xfs_vtoi(vp), &va, 0); - if (error) - goto out; - xfs_acl_sync_mode(va.va_mode, xfs_acl); - } + if (kind == _ACL_TYPE_ACCESS) + xfs_acl_sync_mode(xfs_vtoi(vp)->i_d.di_mode, xfs_acl); error = -posix_acl_xfs_to_xattr(xfs_acl, ext_acl, size); } out: @@ -373,23 +366,15 @@ xfs_acl_allow_set( bhv_vnode_t *vp, int kind) { - xfs_inode_t *ip = xfs_vtoi(vp); - bhv_vattr_t va; - int error; - if (vp->i_flags & (S_IMMUTABLE|S_APPEND)) return EPERM; if (kind == _ACL_TYPE_DEFAULT && !S_ISDIR(vp->i_mode)) return ENOTDIR; if (vp->i_sb->s_flags & MS_RDONLY) return EROFS; - va.va_mask = XFS_AT_UID; - error = xfs_getattr(ip, &va, 0); - if (error) - return error; - if (va.va_uid != current->fsuid && !capable(CAP_FOWNER)) + if (xfs_vtoi(vp)->i_d.di_uid != current->fsuid && !capable(CAP_FOWNER)) return EPERM; - return error; + return 0; } /* @@ -643,7 +628,6 @@ xfs_acl_vtoacl( xfs_acl_t *access_acl, xfs_acl_t *default_acl) { - bhv_vattr_t va; int error = 0; if (access_acl) { @@ -652,16 +636,10 @@ xfs_acl_vtoacl( * be obtained for some reason, invalidate the access ACL. */ xfs_acl_get_attr(vp, access_acl, _ACL_TYPE_ACCESS, 0, &error); - if (!error) { - /* Got the ACL, need the mode... */ - va.va_mask = XFS_AT_MODE; - error = xfs_getattr(xfs_vtoi(vp), &va, 0); - } - if (error) access_acl->acl_cnt = XFS_ACL_NOT_PRESENT; else /* We have a good ACL and the file mode, synchronize. */ - xfs_acl_sync_mode(va.va_mode, access_acl); + xfs_acl_sync_mode(xfs_vtoi(vp)->i_d.di_mode, access_acl); } if (default_acl) { @@ -744,7 +722,7 @@ xfs_acl_setmode( bhv_vattr_t va; xfs_acl_entry_t *ap; xfs_acl_entry_t *gap = NULL; - int i, error, nomask = 1; + int i, nomask = 1; *basicperms = 1; @@ -756,11 +734,7 @@ xfs_acl_setmode( * mode. The m:: bits take precedence over the g:: bits. */ va.va_mask = XFS_AT_MODE; - error = xfs_getattr(xfs_vtoi(vp), &va, 0); - if (error) - return error; - - va.va_mask = XFS_AT_MODE; + va.va_mode = xfs_vtoi(vp)->i_d.di_mode; va.va_mode &= ~(S_IRWXU|S_IRWXG|S_IRWXO); ap = acl->acl_entry; for (i = 0; i < acl->acl_cnt; ++i) { diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 3fef54b11582..04f3e302feee 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -75,132 +75,6 @@ xfs_open( return 0; } -/* - * xfs_getattr - */ -int -xfs_getattr( - xfs_inode_t *ip, - bhv_vattr_t *vap, - int flags) -{ - bhv_vnode_t *vp = XFS_ITOV(ip); - xfs_mount_t *mp = ip->i_mount; - - xfs_itrace_entry(ip); - - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); - - if (!(flags & ATTR_LAZY)) - xfs_ilock(ip, XFS_ILOCK_SHARED); - - vap->va_size = XFS_ISIZE(ip); - if (vap->va_mask == XFS_AT_SIZE) - goto all_done; - - vap->va_nblocks = - XFS_FSB_TO_BB(mp, ip->i_d.di_nblocks + ip->i_delayed_blks); - vap->va_nodeid = ip->i_ino; -#if XFS_BIG_INUMS - vap->va_nodeid += mp->m_inoadd; -#endif - vap->va_nlink = ip->i_d.di_nlink; - - /* - * Quick exit for non-stat callers - */ - if ((vap->va_mask & - ~(XFS_AT_SIZE|XFS_AT_FSID|XFS_AT_NODEID| - XFS_AT_NLINK|XFS_AT_BLKSIZE)) == 0) - goto all_done; - - /* - * Copy from in-core inode. - */ - vap->va_mode = ip->i_d.di_mode; - vap->va_uid = ip->i_d.di_uid; - vap->va_gid = ip->i_d.di_gid; - vap->va_projid = ip->i_d.di_projid; - - /* - * Check vnode type block/char vs. everything else. - */ - switch (ip->i_d.di_mode & S_IFMT) { - case S_IFBLK: - case S_IFCHR: - vap->va_rdev = ip->i_df.if_u2.if_rdev; - vap->va_blocksize = BLKDEV_IOSIZE; - break; - default: - vap->va_rdev = 0; - - if (!(XFS_IS_REALTIME_INODE(ip))) { - vap->va_blocksize = xfs_preferred_iosize(mp); - } else { - - /* - * If the file blocks are being allocated from a - * realtime partition, then return the inode's - * realtime extent size or the realtime volume's - * extent size. - */ - vap->va_blocksize = - xfs_get_extsz_hint(ip) << mp->m_sb.sb_blocklog; - } - break; - } - - vn_atime_to_timespec(vp, &vap->va_atime); - vap->va_mtime.tv_sec = ip->i_d.di_mtime.t_sec; - vap->va_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec; - vap->va_ctime.tv_sec = ip->i_d.di_ctime.t_sec; - vap->va_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec; - - /* - * Exit for stat callers. See if any of the rest of the fields - * to be filled in are needed. - */ - if ((vap->va_mask & - (XFS_AT_XFLAGS|XFS_AT_EXTSIZE|XFS_AT_NEXTENTS|XFS_AT_ANEXTENTS| - XFS_AT_GENCOUNT|XFS_AT_VCODE)) == 0) - goto all_done; - - /* - * Convert di_flags to xflags. - */ - vap->va_xflags = xfs_ip2xflags(ip); - - /* - * Exit for inode revalidate. See if any of the rest of - * the fields to be filled in are needed. - */ - if ((vap->va_mask & - (XFS_AT_EXTSIZE|XFS_AT_NEXTENTS|XFS_AT_ANEXTENTS| - XFS_AT_GENCOUNT|XFS_AT_VCODE)) == 0) - goto all_done; - - vap->va_extsize = ip->i_d.di_extsize << mp->m_sb.sb_blocklog; - vap->va_nextents = - (ip->i_df.if_flags & XFS_IFEXTENTS) ? - ip->i_df.if_bytes / sizeof(xfs_bmbt_rec_t) : - ip->i_d.di_nextents; - if (ip->i_afp) - vap->va_anextents = - (ip->i_afp->if_flags & XFS_IFEXTENTS) ? - ip->i_afp->if_bytes / sizeof(xfs_bmbt_rec_t) : - ip->i_d.di_anextents; - else - vap->va_anextents = 0; - vap->va_gen = ip->i_d.di_gen; - - all_done: - if (!(flags & ATTR_LAZY)) - xfs_iunlock(ip, XFS_ILOCK_SHARED); - return 0; -} - - /* * xfs_setattr */ diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h index 24c53923dc2c..6b66904a3837 100644 --- a/fs/xfs/xfs_vnodeops.h +++ b/fs/xfs/xfs_vnodeops.h @@ -15,7 +15,6 @@ struct xfs_iomap; int xfs_open(struct xfs_inode *ip); -int xfs_getattr(struct xfs_inode *ip, struct bhv_vattr *vap, int flags); int xfs_setattr(struct xfs_inode *ip, struct bhv_vattr *vap, int flags, struct cred *credp); int xfs_readlink(struct xfs_inode *ip, char *link); -- cgit v1.2.3 From 6a7f422d47d4af461704ebb9d7a389d9e59766b2 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 22 Apr 2008 17:33:40 +1000 Subject: [XFS] kill di_mode checks after xfs_iget Unless XFS_IGET_CREATE is passed xfs_iget will return ENOENT if it encounters an inode with di_mode == 0. Remove the duplicated checks in the callers. (the log recovery case is not touched for now) SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30898a Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/linux-2.6/xfs_export.c | 2 +- fs/xfs/linux-2.6/xfs_ioctl.c | 2 +- fs/xfs/quota/xfs_qm.c | 6 ------ fs/xfs/quota/xfs_qm_syscalls.c | 6 ------ fs/xfs/xfs_itable.c | 6 ------ 5 files changed, 2 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c index 265f0168ab76..c672b3238b14 100644 --- a/fs/xfs/linux-2.6/xfs_export.c +++ b/fs/xfs/linux-2.6/xfs_export.c @@ -133,7 +133,7 @@ xfs_nfs_get_inode( if (!ip) return ERR_PTR(-EIO); - if (!ip->i_d.di_mode || ip->i_d.di_gen != generation) { + if (ip->i_d.di_gen != generation) { xfs_iput_new(ip, XFS_ILOCK_SHARED); return ERR_PTR(-ENOENT); } diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index 4ddb86b73c6b..98e87804991d 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -238,7 +238,7 @@ xfs_vget_fsop_handlereq( return error; if (ip == NULL) return XFS_ERROR(EIO); - if (ip->i_d.di_mode == 0 || ip->i_d.di_gen != igen) { + if (ip->i_d.di_gen != igen) { xfs_iput_new(ip, XFS_ILOCK_SHARED); return XFS_ERROR(ENOENT); } diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c index 40ea56409561..fb624a17f15b 100644 --- a/fs/xfs/quota/xfs_qm.c +++ b/fs/xfs/quota/xfs_qm.c @@ -1737,12 +1737,6 @@ xfs_qm_dqusage_adjust( return error; } - if (ip->i_d.di_mode == 0) { - xfs_iput_new(ip, XFS_ILOCK_EXCL); - *res = BULKSTAT_RV_NOTHING; - return XFS_ERROR(ENOENT); - } - /* * Obtain the locked dquots. In case of an error (eg. allocation * fails for ENOSPC), we return the negative of the error number diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c index 8342823dbdc3..768a3b27d2b6 100644 --- a/fs/xfs/quota/xfs_qm_syscalls.c +++ b/fs/xfs/quota/xfs_qm_syscalls.c @@ -1366,12 +1366,6 @@ xfs_qm_internalqcheck_adjust( return (error); } - if (ip->i_d.di_mode == 0) { - xfs_iput_new(ip, lock_flags); - *res = BULKSTAT_RV_NOTHING; - return XFS_ERROR(ENOENT); - } - /* * This inode can have blocks after eof which can get released * when we send it to inactive. Since we don't check the dquot diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index eb85bdedad0c..419de15aeb43 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -71,11 +71,6 @@ xfs_bulkstat_one_iget( ASSERT(ip != NULL); ASSERT(ip->i_blkno != (xfs_daddr_t)0); - if (ip->i_d.di_mode == 0) { - *stat = BULKSTAT_RV_NOTHING; - error = XFS_ERROR(ENOENT); - goto out_iput; - } vp = XFS_ITOV(ip); dic = &ip->i_d; @@ -124,7 +119,6 @@ xfs_bulkstat_one_iget( break; } - out_iput: xfs_iput(ip, XFS_ILOCK_SHARED); return error; } -- cgit v1.2.3 From d4377d84189349357e1812eaff6d0504766eea06 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 22 Apr 2008 17:33:46 +1000 Subject: [XFS] xfs_rename: pass resblks to xfs_dir_removename Similar to rmdir and remove - avoids a potential transaction reservation overrun. SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30900a Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_vnodeops.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 04f3e302feee..7093d749589b 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -2241,7 +2241,7 @@ xfs_remove( */ XFS_BMAP_INIT(&free_list, &first_block); error = xfs_dir_removename(tp, dp, name, ip->i_ino, - &first_block, &free_list, 0); + &first_block, &free_list, resblks); if (error) { ASSERT(error != ENOENT); REMOVE_DEBUG_TRACE(__LINE__); -- cgit v1.2.3 From eca450b7c23f804597b87085b2a05bfc5b3ccb8b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 22 Apr 2008 17:33:52 +1000 Subject: [XFS] simplify xfs_lookup Opencode xfs-kill-xfs_dir_lookup_int here, which gets rid of a lock roundtrip, and lots of stack space. Also kill the di_mode == 0 check that has been done in xfs_iget for a few years now. SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30901a Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_vnodeops.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 7093d749589b..637fc1a2bb44 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -1636,8 +1636,7 @@ xfs_lookup( struct xfs_name *name, xfs_inode_t **ipp) { - xfs_inode_t *ip; - xfs_ino_t e_inum; + xfs_ino_t inum; int error; uint lock_mode; @@ -1647,12 +1646,21 @@ xfs_lookup( return XFS_ERROR(EIO); lock_mode = xfs_ilock_map_shared(dp); - error = xfs_dir_lookup_int(dp, lock_mode, name, &e_inum, &ip); - if (!error) { - *ipp = ip; - xfs_itrace_ref(ip); - } + error = xfs_dir_lookup(NULL, dp, name, &inum); xfs_iunlock_map_shared(dp, lock_mode); + + if (error) + goto out; + + error = xfs_iget(dp->i_mount, NULL, inum, 0, 0, ipp, 0); + if (error) + goto out; + + xfs_itrace_ref(*ipp); + return 0; + + out: + *ipp = NULL; return error; } -- cgit v1.2.3 From 579aa9caf552c639fc78168db4cfe7ffcf00c3b3 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 22 Apr 2008 17:34:00 +1000 Subject: [XFS] shrink mrlock_t The writer field is not needed for non_DEBU builds so remove it. While we're at i also clean up the interface for is locked asserts to go through and xfs_iget.c helper with an interface like the xfs_ilock routines to isolated the XFS codebase from mrlock internals. That way we can kill mrlock_t entirely once rw_semaphores grow an islocked facility. Also remove unused flags to the ilock family of functions. SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30902a Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/linux-2.6/mrlock.h | 60 +++++++----------- fs/xfs/linux-2.6/xfs_lrw.c | 21 +++---- fs/xfs/quota/xfs_dquot.c | 4 +- fs/xfs/quota/xfs_qm.c | 21 ++++--- fs/xfs/quota/xfs_quota_priv.h | 5 -- fs/xfs/quota/xfs_trans_dquot.c | 2 +- fs/xfs/xfs_bmap.c | 1 - fs/xfs/xfs_iget.c | 140 ++++++++++++++++++++++------------------- fs/xfs/xfs_inode.c | 25 ++++---- fs/xfs/xfs_inode.h | 14 +---- fs/xfs/xfs_inode_item.c | 12 ++-- fs/xfs/xfs_iomap.c | 6 +- fs/xfs/xfs_trans_inode.c | 12 ++-- fs/xfs/xfs_utils.c | 2 +- fs/xfs/xfs_vnodeops.c | 4 +- 15 files changed, 154 insertions(+), 175 deletions(-) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/mrlock.h b/fs/xfs/linux-2.6/mrlock.h index c110bb002665..ff6a19873e5c 100644 --- a/fs/xfs/linux-2.6/mrlock.h +++ b/fs/xfs/linux-2.6/mrlock.h @@ -20,29 +20,24 @@ #include -enum { MR_NONE, MR_ACCESS, MR_UPDATE }; - typedef struct { struct rw_semaphore mr_lock; +#ifdef DEBUG int mr_writer; +#endif } mrlock_t; +#ifdef DEBUG #define mrinit(mrp, name) \ do { (mrp)->mr_writer = 0; init_rwsem(&(mrp)->mr_lock); } while (0) +#else +#define mrinit(mrp, name) \ + do { init_rwsem(&(mrp)->mr_lock); } while (0) +#endif + #define mrlock_init(mrp, t,n,s) mrinit(mrp, n) #define mrfree(mrp) do { } while (0) -static inline void mraccess(mrlock_t *mrp) -{ - down_read(&mrp->mr_lock); -} - -static inline void mrupdate(mrlock_t *mrp) -{ - down_write(&mrp->mr_lock); - mrp->mr_writer = 1; -} - static inline void mraccess_nested(mrlock_t *mrp, int subclass) { down_read_nested(&mrp->mr_lock, subclass); @@ -51,10 +46,11 @@ static inline void mraccess_nested(mrlock_t *mrp, int subclass) static inline void mrupdate_nested(mrlock_t *mrp, int subclass) { down_write_nested(&mrp->mr_lock, subclass); +#ifdef DEBUG mrp->mr_writer = 1; +#endif } - static inline int mrtryaccess(mrlock_t *mrp) { return down_read_trylock(&mrp->mr_lock); @@ -64,39 +60,31 @@ static inline int mrtryupdate(mrlock_t *mrp) { if (!down_write_trylock(&mrp->mr_lock)) return 0; +#ifdef DEBUG mrp->mr_writer = 1; +#endif return 1; } -static inline void mrunlock(mrlock_t *mrp) +static inline void mrunlock_excl(mrlock_t *mrp) { - if (mrp->mr_writer) { - mrp->mr_writer = 0; - up_write(&mrp->mr_lock); - } else { - up_read(&mrp->mr_lock); - } +#ifdef DEBUG + mrp->mr_writer = 0; +#endif + up_write(&mrp->mr_lock); } -static inline void mrdemote(mrlock_t *mrp) +static inline void mrunlock_shared(mrlock_t *mrp) { - mrp->mr_writer = 0; - downgrade_write(&mrp->mr_lock); + up_read(&mrp->mr_lock); } -#ifdef DEBUG -/* - * Debug-only routine, without some platform-specific asm code, we can - * now only answer requests regarding whether we hold the lock for write - * (reader state is outside our visibility, we only track writer state). - * Note: means !ismrlocked would give false positives, so don't do that. - */ -static inline int ismrlocked(mrlock_t *mrp, int type) +static inline void mrdemote(mrlock_t *mrp) { - if (mrp && type == MR_UPDATE) - return mrp->mr_writer; - return 1; -} +#ifdef DEBUG + mrp->mr_writer = 0; #endif + downgrade_write(&mrp->mr_lock); +} #endif /* __XFS_SUPPORT_MRLOCK_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c index 1ebd8004469c..5e3b57516ec7 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.c +++ b/fs/xfs/linux-2.6/xfs_lrw.c @@ -394,7 +394,7 @@ xfs_zero_last_block( int error = 0; xfs_bmbt_irec_t imap; - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE) != 0); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); zero_offset = XFS_B_FSB_OFFSET(mp, isize); if (zero_offset == 0) { @@ -425,14 +425,14 @@ xfs_zero_last_block( * out sync. We need to drop the ilock while we do this so we * don't deadlock when the buffer cache calls back to us. */ - xfs_iunlock(ip, XFS_ILOCK_EXCL| XFS_EXTSIZE_RD); + xfs_iunlock(ip, XFS_ILOCK_EXCL); zero_len = mp->m_sb.sb_blocksize - zero_offset; if (isize + zero_len > offset) zero_len = offset - isize; error = xfs_iozero(ip, isize, zero_len); - xfs_ilock(ip, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD); + xfs_ilock(ip, XFS_ILOCK_EXCL); ASSERT(error >= 0); return error; } @@ -465,8 +465,7 @@ xfs_zero_eof( int error = 0; xfs_bmbt_irec_t imap; - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); - ASSERT(ismrlocked(&ip->i_iolock, MR_UPDATE)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); ASSERT(offset > isize); /* @@ -475,8 +474,7 @@ xfs_zero_eof( */ error = xfs_zero_last_block(ip, offset, isize); if (error) { - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); - ASSERT(ismrlocked(&ip->i_iolock, MR_UPDATE)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); return error; } @@ -507,8 +505,7 @@ xfs_zero_eof( error = xfs_bmapi(NULL, ip, start_zero_fsb, zero_count_fsb, 0, NULL, 0, &imap, &nimaps, NULL, NULL); if (error) { - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); - ASSERT(ismrlocked(&ip->i_iolock, MR_UPDATE)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); return error; } ASSERT(nimaps > 0); @@ -532,7 +529,7 @@ xfs_zero_eof( * Drop the inode lock while we're doing the I/O. * We'll still have the iolock to protect us. */ - xfs_iunlock(ip, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD); + xfs_iunlock(ip, XFS_ILOCK_EXCL); zero_off = XFS_FSB_TO_B(mp, start_zero_fsb); zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount); @@ -548,13 +545,13 @@ xfs_zero_eof( start_zero_fsb = imap.br_startoff + imap.br_blockcount; ASSERT(start_zero_fsb <= (end_zero_fsb + 1)); - xfs_ilock(ip, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD); + xfs_ilock(ip, XFS_ILOCK_EXCL); } return 0; out_lock: - xfs_ilock(ip, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD); + xfs_ilock(ip, XFS_ILOCK_EXCL); ASSERT(error >= 0); return error; } diff --git a/fs/xfs/quota/xfs_dquot.c b/fs/xfs/quota/xfs_dquot.c index 631ebb31b295..85df3288efd5 100644 --- a/fs/xfs/quota/xfs_dquot.c +++ b/fs/xfs/quota/xfs_dquot.c @@ -933,7 +933,7 @@ xfs_qm_dqget( type == XFS_DQ_PROJ || type == XFS_DQ_GROUP); if (ip) { - ASSERT(XFS_ISLOCKED_INODE_EXCL(ip)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); if (type == XFS_DQ_USER) ASSERT(ip->i_udquot == NULL); else @@ -1088,7 +1088,7 @@ xfs_qm_dqget( xfs_qm_mplist_unlock(mp); XFS_DQ_HASH_UNLOCK(h); dqret: - ASSERT((ip == NULL) || XFS_ISLOCKED_INODE_EXCL(ip)); + ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL)); xfs_dqtrace_entry(dqp, "DQGET DONE"); *O_dqpp = dqp; return (0); diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c index fb624a17f15b..d31cce1165c5 100644 --- a/fs/xfs/quota/xfs_qm.c +++ b/fs/xfs/quota/xfs_qm.c @@ -670,7 +670,7 @@ xfs_qm_dqattach_one( xfs_dquot_t *dqp; int error; - ASSERT(XFS_ISLOCKED_INODE_EXCL(ip)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); error = 0; /* * See if we already have it in the inode itself. IO_idqpp is @@ -874,7 +874,7 @@ xfs_qm_dqattach( return 0; ASSERT((flags & XFS_QMOPT_ILOCKED) == 0 || - XFS_ISLOCKED_INODE_EXCL(ip)); + xfs_isilocked(ip, XFS_ILOCK_EXCL)); if (! (flags & XFS_QMOPT_ILOCKED)) xfs_ilock(ip, XFS_ILOCK_EXCL); @@ -888,7 +888,8 @@ xfs_qm_dqattach( goto done; nquotas++; } - ASSERT(XFS_ISLOCKED_INODE_EXCL(ip)); + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); if (XFS_IS_OQUOTA_ON(mp)) { error = XFS_IS_GQUOTA_ON(mp) ? xfs_qm_dqattach_one(ip, ip->i_d.di_gid, XFS_DQ_GROUP, @@ -913,7 +914,7 @@ xfs_qm_dqattach( * This WON'T, in general, result in a thrash. */ if (nquotas == 2) { - ASSERT(XFS_ISLOCKED_INODE_EXCL(ip)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); ASSERT(ip->i_udquot); ASSERT(ip->i_gdquot); @@ -956,7 +957,7 @@ xfs_qm_dqattach( #ifdef QUOTADEBUG else - ASSERT(XFS_ISLOCKED_INODE_EXCL(ip)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); #endif return error; } @@ -1291,7 +1292,7 @@ xfs_qm_dqget_noattach( xfs_mount_t *mp; xfs_dquot_t *udqp, *gdqp; - ASSERT(XFS_ISLOCKED_INODE_EXCL(ip)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); mp = ip->i_mount; udqp = NULL; gdqp = NULL; @@ -1392,7 +1393,7 @@ xfs_qm_qino_alloc( * Keep an extra reference to this quota inode. This inode is * locked exclusively and joined to the transaction already. */ - ASSERT(XFS_ISLOCKED_INODE_EXCL(*ip)); + ASSERT(xfs_isilocked(*ip, XFS_ILOCK_EXCL)); VN_HOLD(XFS_ITOV((*ip))); /* @@ -2557,7 +2558,7 @@ xfs_qm_vop_chown( uint bfield = XFS_IS_REALTIME_INODE(ip) ? XFS_TRANS_DQ_RTBCOUNT : XFS_TRANS_DQ_BCOUNT; - ASSERT(XFS_ISLOCKED_INODE_EXCL(ip)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); ASSERT(XFS_IS_QUOTA_RUNNING(ip->i_mount)); /* old dquot */ @@ -2601,7 +2602,7 @@ xfs_qm_vop_chown_reserve( uint delblks, blkflags, prjflags = 0; xfs_dquot_t *unresudq, *unresgdq, *delblksudq, *delblksgdq; - ASSERT(XFS_ISLOCKED_INODE(ip)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); mp = ip->i_mount; ASSERT(XFS_IS_QUOTA_RUNNING(mp)); @@ -2711,7 +2712,7 @@ xfs_qm_vop_dqattach_and_dqmod_newinode( if (!XFS_IS_QUOTA_ON(tp->t_mountp)) return; - ASSERT(XFS_ISLOCKED_INODE_EXCL(ip)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); ASSERT(XFS_IS_QUOTA_RUNNING(tp->t_mountp)); if (udqp) { diff --git a/fs/xfs/quota/xfs_quota_priv.h b/fs/xfs/quota/xfs_quota_priv.h index a8b85e2be9d5..5e4a40b1c565 100644 --- a/fs/xfs/quota/xfs_quota_priv.h +++ b/fs/xfs/quota/xfs_quota_priv.h @@ -27,11 +27,6 @@ /* Number of dquots that fit in to a dquot block */ #define XFS_QM_DQPERBLK(mp) ((mp)->m_quotainfo->qi_dqperchunk) -#define XFS_ISLOCKED_INODE(ip) (ismrlocked(&(ip)->i_lock, \ - MR_UPDATE | MR_ACCESS) != 0) -#define XFS_ISLOCKED_INODE_EXCL(ip) (ismrlocked(&(ip)->i_lock, \ - MR_UPDATE) != 0) - #define XFS_DQ_IS_ADDEDTO_TRX(t, d) ((d)->q_transp == (t)) #define XFS_QI_MPLRECLAIMS(mp) ((mp)->m_quotainfo->qi_dqreclaims) diff --git a/fs/xfs/quota/xfs_trans_dquot.c b/fs/xfs/quota/xfs_trans_dquot.c index f441f836ca8b..99611381e740 100644 --- a/fs/xfs/quota/xfs_trans_dquot.c +++ b/fs/xfs/quota/xfs_trans_dquot.c @@ -834,7 +834,7 @@ xfs_trans_reserve_quota_nblks( ASSERT(ip->i_ino != mp->m_sb.sb_uquotino); ASSERT(ip->i_ino != mp->m_sb.sb_gquotino); - ASSERT(XFS_ISLOCKED_INODE_EXCL(ip)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); ASSERT(XFS_IS_QUOTA_RUNNING(ip->i_mount)); ASSERT((flags & ~(XFS_QMOPT_FORCE_RES | XFS_QMOPT_ENOSPC)) == XFS_TRANS_DQ_RES_RTBLKS || diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index eb198c01c35d..53c259f5a5af 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -4074,7 +4074,6 @@ xfs_bmap_add_attrfork( error2: xfs_bmap_cancel(&flist); error1: - ASSERT(ismrlocked(&ip->i_lock,MR_UPDATE)); xfs_iunlock(ip, XFS_ILOCK_EXCL); error0: xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT); diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index e657c5128460..b07604b94d9f 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -593,8 +593,9 @@ xfs_iunlock_map_shared( * XFS_IOLOCK_EXCL | XFS_ILOCK_EXCL */ void -xfs_ilock(xfs_inode_t *ip, - uint lock_flags) +xfs_ilock( + xfs_inode_t *ip, + uint lock_flags) { /* * You can't set both SHARED and EXCL for the same lock, @@ -607,16 +608,16 @@ xfs_ilock(xfs_inode_t *ip, (XFS_ILOCK_SHARED | XFS_ILOCK_EXCL)); ASSERT((lock_flags & ~(XFS_LOCK_MASK | XFS_LOCK_DEP_MASK)) == 0); - if (lock_flags & XFS_IOLOCK_EXCL) { + if (lock_flags & XFS_IOLOCK_EXCL) mrupdate_nested(&ip->i_iolock, XFS_IOLOCK_DEP(lock_flags)); - } else if (lock_flags & XFS_IOLOCK_SHARED) { + else if (lock_flags & XFS_IOLOCK_SHARED) mraccess_nested(&ip->i_iolock, XFS_IOLOCK_DEP(lock_flags)); - } - if (lock_flags & XFS_ILOCK_EXCL) { + + if (lock_flags & XFS_ILOCK_EXCL) mrupdate_nested(&ip->i_lock, XFS_ILOCK_DEP(lock_flags)); - } else if (lock_flags & XFS_ILOCK_SHARED) { + else if (lock_flags & XFS_ILOCK_SHARED) mraccess_nested(&ip->i_lock, XFS_ILOCK_DEP(lock_flags)); - } + xfs_ilock_trace(ip, 1, lock_flags, (inst_t *)__return_address); } @@ -631,15 +632,12 @@ xfs_ilock(xfs_inode_t *ip, * lock_flags -- this parameter indicates the inode's locks to be * to be locked. See the comment for xfs_ilock() for a list * of valid values. - * */ int -xfs_ilock_nowait(xfs_inode_t *ip, - uint lock_flags) +xfs_ilock_nowait( + xfs_inode_t *ip, + uint lock_flags) { - int iolocked; - int ilocked; - /* * You can't set both SHARED and EXCL for the same lock, * and only XFS_IOLOCK_SHARED, XFS_IOLOCK_EXCL, XFS_ILOCK_SHARED, @@ -651,37 +649,30 @@ xfs_ilock_nowait(xfs_inode_t *ip, (XFS_ILOCK_SHARED | XFS_ILOCK_EXCL)); ASSERT((lock_flags & ~(XFS_LOCK_MASK | XFS_LOCK_DEP_MASK)) == 0); - iolocked = 0; if (lock_flags & XFS_IOLOCK_EXCL) { - iolocked = mrtryupdate(&ip->i_iolock); - if (!iolocked) { - return 0; - } + if (!mrtryupdate(&ip->i_iolock)) + goto out; } else if (lock_flags & XFS_IOLOCK_SHARED) { - iolocked = mrtryaccess(&ip->i_iolock); - if (!iolocked) { - return 0; - } + if (!mrtryaccess(&ip->i_iolock)) + goto out; } if (lock_flags & XFS_ILOCK_EXCL) { - ilocked = mrtryupdate(&ip->i_lock); - if (!ilocked) { - if (iolocked) { - mrunlock(&ip->i_iolock); - } - return 0; - } + if (!mrtryupdate(&ip->i_lock)) + goto out_undo_iolock; } else if (lock_flags & XFS_ILOCK_SHARED) { - ilocked = mrtryaccess(&ip->i_lock); - if (!ilocked) { - if (iolocked) { - mrunlock(&ip->i_iolock); - } - return 0; - } + if (!mrtryaccess(&ip->i_lock)) + goto out_undo_iolock; } xfs_ilock_trace(ip, 2, lock_flags, (inst_t *)__return_address); return 1; + + out_undo_iolock: + if (lock_flags & XFS_IOLOCK_EXCL) + mrunlock_excl(&ip->i_iolock); + else if (lock_flags & XFS_IOLOCK_SHARED) + mrunlock_shared(&ip->i_iolock); + out: + return 0; } /* @@ -697,8 +688,9 @@ xfs_ilock_nowait(xfs_inode_t *ip, * */ void -xfs_iunlock(xfs_inode_t *ip, - uint lock_flags) +xfs_iunlock( + xfs_inode_t *ip, + uint lock_flags) { /* * You can't set both SHARED and EXCL for the same lock, @@ -713,31 +705,25 @@ xfs_iunlock(xfs_inode_t *ip, XFS_LOCK_DEP_MASK)) == 0); ASSERT(lock_flags != 0); - if (lock_flags & (XFS_IOLOCK_SHARED | XFS_IOLOCK_EXCL)) { - ASSERT(!(lock_flags & XFS_IOLOCK_SHARED) || - (ismrlocked(&ip->i_iolock, MR_ACCESS))); - ASSERT(!(lock_flags & XFS_IOLOCK_EXCL) || - (ismrlocked(&ip->i_iolock, MR_UPDATE))); - mrunlock(&ip->i_iolock); - } + if (lock_flags & XFS_IOLOCK_EXCL) + mrunlock_excl(&ip->i_iolock); + else if (lock_flags & XFS_IOLOCK_SHARED) + mrunlock_shared(&ip->i_iolock); - if (lock_flags & (XFS_ILOCK_SHARED | XFS_ILOCK_EXCL)) { - ASSERT(!(lock_flags & XFS_ILOCK_SHARED) || - (ismrlocked(&ip->i_lock, MR_ACCESS))); - ASSERT(!(lock_flags & XFS_ILOCK_EXCL) || - (ismrlocked(&ip->i_lock, MR_UPDATE))); - mrunlock(&ip->i_lock); + if (lock_flags & XFS_ILOCK_EXCL) + mrunlock_excl(&ip->i_lock); + else if (lock_flags & XFS_ILOCK_SHARED) + mrunlock_shared(&ip->i_lock); + if ((lock_flags & (XFS_ILOCK_SHARED | XFS_ILOCK_EXCL)) && + !(lock_flags & XFS_IUNLOCK_NONOTIFY) && ip->i_itemp) { /* * Let the AIL know that this item has been unlocked in case * it is in the AIL and anyone is waiting on it. Don't do * this if the caller has asked us not to. */ - if (!(lock_flags & XFS_IUNLOCK_NONOTIFY) && - ip->i_itemp != NULL) { - xfs_trans_unlocked_item(ip->i_mount, - (xfs_log_item_t*)(ip->i_itemp)); - } + xfs_trans_unlocked_item(ip->i_mount, + (xfs_log_item_t*)(ip->i_itemp)); } xfs_ilock_trace(ip, 3, lock_flags, (inst_t *)__return_address); } @@ -747,21 +733,47 @@ xfs_iunlock(xfs_inode_t *ip, * if it is being demoted. */ void -xfs_ilock_demote(xfs_inode_t *ip, - uint lock_flags) +xfs_ilock_demote( + xfs_inode_t *ip, + uint lock_flags) { ASSERT(lock_flags & (XFS_IOLOCK_EXCL|XFS_ILOCK_EXCL)); ASSERT((lock_flags & ~(XFS_IOLOCK_EXCL|XFS_ILOCK_EXCL)) == 0); - if (lock_flags & XFS_ILOCK_EXCL) { - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); + if (lock_flags & XFS_ILOCK_EXCL) mrdemote(&ip->i_lock); - } - if (lock_flags & XFS_IOLOCK_EXCL) { - ASSERT(ismrlocked(&ip->i_iolock, MR_UPDATE)); + if (lock_flags & XFS_IOLOCK_EXCL) mrdemote(&ip->i_iolock); +} + +#ifdef DEBUG +/* + * Debug-only routine, without additional rw_semaphore APIs, we can + * now only answer requests regarding whether we hold the lock for write + * (reader state is outside our visibility, we only track writer state). + * + * Note: this means !xfs_isilocked would give false positives, so don't do that. + */ +int +xfs_isilocked( + xfs_inode_t *ip, + uint lock_flags) +{ + if ((lock_flags & (XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)) == + XFS_ILOCK_EXCL) { + if (!ip->i_lock.mr_writer) + return 0; } + + if ((lock_flags & (XFS_IOLOCK_EXCL|XFS_IOLOCK_SHARED)) == + XFS_IOLOCK_EXCL) { + if (!ip->i_iolock.mr_writer) + return 0; + } + + return 1; } +#endif /* * The following three routines simply manage the i_flock diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index ca12acb90394..cf0bb9c1d621 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1291,7 +1291,7 @@ xfs_file_last_byte( xfs_fileoff_t size_last_block; int error; - ASSERT(ismrlocked(&(ip->i_iolock), MR_UPDATE | MR_ACCESS)); + ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL|XFS_IOLOCK_SHARED)); mp = ip->i_mount; /* @@ -1402,7 +1402,7 @@ xfs_itruncate_start( bhv_vnode_t *vp; int error = 0; - ASSERT(ismrlocked(&ip->i_iolock, MR_UPDATE) != 0); + ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL)); ASSERT((new_size == 0) || (new_size <= ip->i_size)); ASSERT((flags == XFS_ITRUNC_DEFINITE) || (flags == XFS_ITRUNC_MAYBE)); @@ -1528,8 +1528,7 @@ xfs_itruncate_finish( xfs_bmap_free_t free_list; int error; - ASSERT(ismrlocked(&ip->i_iolock, MR_UPDATE) != 0); - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE) != 0); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); ASSERT((new_size == 0) || (new_size <= ip->i_size)); ASSERT(*tp != NULL); ASSERT((*tp)->t_flags & XFS_TRANS_PERM_LOG_RES); @@ -1780,8 +1779,7 @@ xfs_igrow_start( xfs_fsize_t new_size, cred_t *credp) { - ASSERT(ismrlocked(&(ip->i_lock), MR_UPDATE) != 0); - ASSERT(ismrlocked(&(ip->i_iolock), MR_UPDATE) != 0); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); ASSERT(new_size > ip->i_size); /* @@ -1809,8 +1807,7 @@ xfs_igrow_finish( xfs_fsize_t new_size, int change_flag) { - ASSERT(ismrlocked(&(ip->i_lock), MR_UPDATE) != 0); - ASSERT(ismrlocked(&(ip->i_iolock), MR_UPDATE) != 0); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); ASSERT(ip->i_transp == tp); ASSERT(new_size > ip->i_size); @@ -2287,7 +2284,7 @@ xfs_ifree( xfs_dinode_t *dip; xfs_buf_t *ibp; - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); ASSERT(ip->i_transp == tp); ASSERT(ip->i_d.di_nlink == 0); ASSERT(ip->i_d.di_nextents == 0); @@ -2746,7 +2743,7 @@ void xfs_ipin( xfs_inode_t *ip) { - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); atomic_inc(&ip->i_pincount); } @@ -2779,7 +2776,7 @@ __xfs_iunpin_wait( { xfs_inode_log_item_t *iip = ip->i_itemp; - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE | MR_ACCESS)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); if (atomic_read(&ip->i_pincount) == 0) return; @@ -2829,7 +2826,7 @@ xfs_iextents_copy( xfs_fsblock_t start_block; ifp = XFS_IFORK_PTR(ip, whichfork); - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE|MR_ACCESS)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); ASSERT(ifp->if_bytes > 0); nrecs = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); @@ -3132,7 +3129,7 @@ xfs_iflush( XFS_STATS_INC(xs_iflush_count); - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE|MR_ACCESS)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); ASSERT(issemalocked(&(ip->i_flock))); ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE || ip->i_d.di_nextents > ip->i_df.if_ext_max); @@ -3297,7 +3294,7 @@ xfs_iflush_int( int first; #endif - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE|MR_ACCESS)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); ASSERT(issemalocked(&(ip->i_flock))); ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE || ip->i_d.di_nextents > ip->i_df.if_ext_max); diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 93c37697a72c..877d71adbc1e 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -386,20 +386,9 @@ xfs_iflags_test_and_clear(xfs_inode_t *ip, unsigned short flags) #define XFS_ILOCK_EXCL (1<<2) #define XFS_ILOCK_SHARED (1<<3) #define XFS_IUNLOCK_NONOTIFY (1<<4) -/* #define XFS_IOLOCK_NESTED (1<<5) */ -#define XFS_EXTENT_TOKEN_RD (1<<6) -#define XFS_SIZE_TOKEN_RD (1<<7) -#define XFS_EXTSIZE_RD (XFS_EXTENT_TOKEN_RD|XFS_SIZE_TOKEN_RD) -#define XFS_WILLLEND (1<<8) /* Always acquire tokens for lending */ -#define XFS_EXTENT_TOKEN_WR (XFS_EXTENT_TOKEN_RD | XFS_WILLLEND) -#define XFS_SIZE_TOKEN_WR (XFS_SIZE_TOKEN_RD | XFS_WILLLEND) -#define XFS_EXTSIZE_WR (XFS_EXTSIZE_RD | XFS_WILLLEND) -/* TODO:XFS_SIZE_TOKEN_WANT (1<<9) */ #define XFS_LOCK_MASK (XFS_IOLOCK_EXCL | XFS_IOLOCK_SHARED \ - | XFS_ILOCK_EXCL | XFS_ILOCK_SHARED \ - | XFS_EXTENT_TOKEN_RD | XFS_SIZE_TOKEN_RD \ - | XFS_WILLLEND) + | XFS_ILOCK_EXCL | XFS_ILOCK_SHARED) /* * Flags for lockdep annotations. @@ -483,6 +472,7 @@ void xfs_ilock(xfs_inode_t *, uint); int xfs_ilock_nowait(xfs_inode_t *, uint); void xfs_iunlock(xfs_inode_t *, uint); void xfs_ilock_demote(xfs_inode_t *, uint); +int xfs_isilocked(xfs_inode_t *, uint); void xfs_iflock(xfs_inode_t *); int xfs_iflock_nowait(xfs_inode_t *); uint xfs_ilock_map_shared(xfs_inode_t *); diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index 93b5db453ea2..167b33f15772 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -547,7 +547,7 @@ STATIC void xfs_inode_item_pin( xfs_inode_log_item_t *iip) { - ASSERT(ismrlocked(&(iip->ili_inode->i_lock), MR_UPDATE)); + ASSERT(xfs_isilocked(iip->ili_inode, XFS_ILOCK_EXCL)); xfs_ipin(iip->ili_inode); } @@ -664,13 +664,13 @@ xfs_inode_item_unlock( ASSERT(iip != NULL); ASSERT(iip->ili_inode->i_itemp != NULL); - ASSERT(ismrlocked(&(iip->ili_inode->i_lock), MR_UPDATE)); + ASSERT(xfs_isilocked(iip->ili_inode, XFS_ILOCK_EXCL)); ASSERT((!(iip->ili_inode->i_itemp->ili_flags & XFS_ILI_IOLOCKED_EXCL)) || - ismrlocked(&(iip->ili_inode->i_iolock), MR_UPDATE)); + xfs_isilocked(iip->ili_inode, XFS_IOLOCK_EXCL)); ASSERT((!(iip->ili_inode->i_itemp->ili_flags & XFS_ILI_IOLOCKED_SHARED)) || - ismrlocked(&(iip->ili_inode->i_iolock), MR_ACCESS)); + xfs_isilocked(iip->ili_inode, XFS_IOLOCK_SHARED)); /* * Clear the transaction pointer in the inode. */ @@ -769,7 +769,7 @@ xfs_inode_item_pushbuf( ip = iip->ili_inode; - ASSERT(ismrlocked(&(ip->i_lock), MR_ACCESS)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED)); /* * The ili_pushbuf_flag keeps others from @@ -857,7 +857,7 @@ xfs_inode_item_push( ip = iip->ili_inode; - ASSERT(ismrlocked(&(ip->i_lock), MR_ACCESS)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED)); ASSERT(issemalocked(&(ip->i_flock))); /* * Since we were able to lock the inode's flush lock and diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index fb3cf1191419..a2c3200a099f 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -196,14 +196,14 @@ xfs_iomap( break; case BMAPI_WRITE: xfs_iomap_enter_trace(XFS_IOMAP_WRITE_ENTER, ip, offset, count); - lockmode = XFS_ILOCK_EXCL|XFS_EXTSIZE_WR; + lockmode = XFS_ILOCK_EXCL; if (flags & BMAPI_IGNSTATE) bmapi_flags |= XFS_BMAPI_IGSTATE|XFS_BMAPI_ENTIRE; xfs_ilock(ip, lockmode); break; case BMAPI_ALLOCATE: xfs_iomap_enter_trace(XFS_IOMAP_ALLOC_ENTER, ip, offset, count); - lockmode = XFS_ILOCK_SHARED|XFS_EXTSIZE_RD; + lockmode = XFS_ILOCK_SHARED; bmapi_flags = XFS_BMAPI_ENTIRE; /* Attempt non-blocking lock */ @@ -624,7 +624,7 @@ xfs_iomap_write_delay( int prealloc, fsynced = 0; int error; - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE) != 0); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); /* * Make sure that the dquots are there. This doesn't hold diff --git a/fs/xfs/xfs_trans_inode.c b/fs/xfs/xfs_trans_inode.c index b8db1d5cde5a..4c70bf5e9985 100644 --- a/fs/xfs/xfs_trans_inode.c +++ b/fs/xfs/xfs_trans_inode.c @@ -111,13 +111,13 @@ xfs_trans_iget( */ ASSERT(ip->i_itemp != NULL); ASSERT(lock_flags & XFS_ILOCK_EXCL); - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); ASSERT((!(lock_flags & XFS_IOLOCK_EXCL)) || - ismrlocked(&ip->i_iolock, MR_UPDATE)); + xfs_isilocked(ip, XFS_IOLOCK_EXCL)); ASSERT((!(lock_flags & XFS_IOLOCK_EXCL)) || (ip->i_itemp->ili_flags & XFS_ILI_IOLOCKED_EXCL)); ASSERT((!(lock_flags & XFS_IOLOCK_SHARED)) || - ismrlocked(&ip->i_iolock, (MR_UPDATE | MR_ACCESS))); + xfs_isilocked(ip, XFS_IOLOCK_EXCL|XFS_IOLOCK_SHARED)); ASSERT((!(lock_flags & XFS_IOLOCK_SHARED)) || (ip->i_itemp->ili_flags & XFS_ILI_IOLOCKED_ANY)); @@ -185,7 +185,7 @@ xfs_trans_ijoin( xfs_inode_log_item_t *iip; ASSERT(ip->i_transp == NULL); - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); ASSERT(lock_flags & XFS_ILOCK_EXCL); if (ip->i_itemp == NULL) xfs_inode_item_init(ip, ip->i_mount); @@ -232,7 +232,7 @@ xfs_trans_ihold( { ASSERT(ip->i_transp == tp); ASSERT(ip->i_itemp != NULL); - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); ip->i_itemp->ili_flags |= XFS_ILI_HOLD; } @@ -257,7 +257,7 @@ xfs_trans_log_inode( ASSERT(ip->i_transp == tp); ASSERT(ip->i_itemp != NULL); - ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)(ip->i_itemp)); ASSERT(lidp != NULL); diff --git a/fs/xfs/xfs_utils.c b/fs/xfs/xfs_utils.c index 2b8dc7e40772..27075c9060ef 100644 --- a/fs/xfs/xfs_utils.c +++ b/fs/xfs/xfs_utils.c @@ -310,7 +310,7 @@ xfs_bump_ino_vers2( { xfs_mount_t *mp; - ASSERT(ismrlocked (&ip->i_lock, MR_UPDATE)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); ASSERT(ip->i_d.di_version == XFS_DINODE_VERSION_1); ip->i_d.di_version = XFS_DINODE_VERSION_2; diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 637fc1a2bb44..322ba094dcc8 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -1305,7 +1305,7 @@ xfs_inactive_attrs( int error; xfs_mount_t *mp; - ASSERT(ismrlocked(&ip->i_iolock, MR_UPDATE)); + ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL)); tp = *tpp; mp = ip->i_mount; ASSERT(ip->i_d.di_forkoff != 0); @@ -1776,7 +1776,7 @@ xfs_create( * It is locked (and joined to the transaction). */ - ASSERT(ismrlocked (&ip->i_lock, MR_UPDATE)); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); /* * Now we join the directory inode to the transaction. We do not do it -- cgit v1.2.3 From cfa853e47df4fbee441ac0ac3fb592f076233145 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 22 Apr 2008 17:34:06 +1000 Subject: [XFS] remove manual lookup from xfs_rename and simplify locking ->rename already gets the target inode passed if it exits. Pass it down to xfs_rename so that we can avoid looking it up again. Also simplify locking as the first lock section in xfs_rename can go away now: the isdir is an invariant over the lifetime of the inode, and new_parent and the nlink check are namespace topology protected by i_mutex in the VFS. The projid check needs to move into the second lock section anyway to not be racy. Also kill the now unused xfs_dir_lookup_int and remove the now-unused first_locked argumet to xfs_lock_inodes. SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30903a Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/linux-2.6/xfs_iops.c | 3 +- fs/xfs/xfs_dfrag.c | 4 +- fs/xfs/xfs_inode.h | 2 +- fs/xfs/xfs_rename.c | 185 +++++++++++--------------------------------- fs/xfs/xfs_utils.c | 43 ---------- fs/xfs/xfs_utils.h | 2 - fs/xfs/xfs_vnodeops.c | 14 +--- fs/xfs/xfs_vnodeops.h | 2 +- 8 files changed, 55 insertions(+), 200 deletions(-) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index a1237dad6430..2bf287ef5489 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -511,7 +511,8 @@ xfs_vn_rename( xfs_dentry_to_name(&nname, ndentry); error = xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode), - XFS_I(ndir), &nname); + XFS_I(ndir), &nname, new_inode ? + XFS_I(new_inode) : NULL); if (likely(!error)) { if (new_inode) xfs_validate_fields(new_inode); diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c index 3f53fad356a3..5f3647cb9885 100644 --- a/fs/xfs/xfs_dfrag.c +++ b/fs/xfs/xfs_dfrag.c @@ -162,7 +162,7 @@ xfs_swap_extents( ips[1] = ip; } - xfs_lock_inodes(ips, 2, 0, lock_flags); + xfs_lock_inodes(ips, 2, lock_flags); locked = 1; /* Verify that both files have the same format */ @@ -265,7 +265,7 @@ xfs_swap_extents( locked = 0; goto error0; } - xfs_lock_inodes(ips, 2, 0, XFS_ILOCK_EXCL); + xfs_lock_inodes(ips, 2, XFS_ILOCK_EXCL); /* * Count the number of extended attribute blocks diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 877d71adbc1e..0a999fee4f03 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -524,7 +524,7 @@ int xfs_iflush(xfs_inode_t *, uint); void xfs_iflush_all(struct xfs_mount *); void xfs_ichgtime(xfs_inode_t *, int); xfs_fsize_t xfs_file_last_byte(xfs_inode_t *); -void xfs_lock_inodes(xfs_inode_t **, int, int, uint); +void xfs_lock_inodes(xfs_inode_t **, int, uint); void xfs_synchronize_atime(xfs_inode_t *); void xfs_mark_inode_dirty_sync(xfs_inode_t *); diff --git a/fs/xfs/xfs_rename.c b/fs/xfs/xfs_rename.c index ee371890d85d..6a141427f68a 100644 --- a/fs/xfs/xfs_rename.c +++ b/fs/xfs/xfs_rename.c @@ -55,85 +55,32 @@ xfs_rename_unlock4( xfs_iunlock(i_tab[0], lock_mode); for (i = 1; i < 4; i++) { - if (i_tab[i] == NULL) { + if (i_tab[i] == NULL) break; - } + /* * Watch out for duplicate entries in the table. */ - if (i_tab[i] != i_tab[i-1]) { + if (i_tab[i] != i_tab[i-1]) xfs_iunlock(i_tab[i], lock_mode); - } } } -#ifdef DEBUG -int xfs_rename_skip, xfs_rename_nskip; -#endif - /* - * The following routine will acquire the locks required for a rename - * operation. The code understands the semantics of renames and will - * validate that name1 exists under dp1 & that name2 may or may not - * exist under dp2. - * - * We are renaming dp1/name1 to dp2/name2. - * - * Return ENOENT if dp1 does not exist, other lookup errors, or 0 for success. + * Enter all inodes for a rename transaction into a sorted array. */ -STATIC int -xfs_lock_for_rename( +STATIC void +xfs_sort_for_rename( xfs_inode_t *dp1, /* in: old (source) directory inode */ xfs_inode_t *dp2, /* in: new (target) directory inode */ xfs_inode_t *ip1, /* in: inode of old entry */ - struct xfs_name *name2, /* in: new entry name */ - xfs_inode_t **ipp2, /* out: inode of new entry, if it + xfs_inode_t *ip2, /* in: inode of new entry, if it already exists, NULL otherwise. */ xfs_inode_t **i_tab,/* out: array of inode returned, sorted */ int *num_inodes) /* out: number of inodes in array */ { - xfs_inode_t *ip2 = NULL; xfs_inode_t *temp; - xfs_ino_t inum1, inum2; - int error; int i, j; - uint lock_mode; - int diff_dirs = (dp1 != dp2); - - /* - * First, find out the current inums of the entries so that we - * can determine the initial locking order. We'll have to - * sanity check stuff after all the locks have been acquired - * to see if we still have the right inodes, directories, etc. - */ - lock_mode = xfs_ilock_map_shared(dp1); - IHOLD(ip1); - xfs_itrace_ref(ip1); - - inum1 = ip1->i_ino; - - /* - * Unlock dp1 and lock dp2 if they are different. - */ - if (diff_dirs) { - xfs_iunlock_map_shared(dp1, lock_mode); - lock_mode = xfs_ilock_map_shared(dp2); - } - - error = xfs_dir_lookup_int(dp2, lock_mode, name2, &inum2, &ip2); - if (error == ENOENT) { /* target does not need to exist. */ - inum2 = 0; - } else if (error) { - /* - * If dp2 and dp1 are the same, the next line unlocks dp1. - * Got it? - */ - xfs_iunlock_map_shared(dp2, lock_mode); - IRELE (ip1); - return error; - } else { - xfs_itrace_ref(ip2); - } /* * i_tab contains a list of pointers to inodes. We initialize @@ -145,21 +92,20 @@ xfs_lock_for_rename( i_tab[0] = dp1; i_tab[1] = dp2; i_tab[2] = ip1; - if (inum2 == 0) { - *num_inodes = 3; - i_tab[3] = NULL; - } else { + if (ip2) { *num_inodes = 4; i_tab[3] = ip2; + } else { + *num_inodes = 3; + i_tab[3] = NULL; } - *ipp2 = i_tab[3]; /* * Sort the elements via bubble sort. (Remember, there are at * most 4 elements to sort, so this is adequate.) */ - for (i=0; i < *num_inodes; i++) { - for (j=1; j < *num_inodes; j++) { + for (i = 0; i < *num_inodes; i++) { + for (j = 1; j < *num_inodes; j++) { if (i_tab[j]->i_ino < i_tab[j-1]->i_ino) { temp = i_tab[j]; i_tab[j] = i_tab[j-1]; @@ -167,30 +113,6 @@ xfs_lock_for_rename( } } } - - /* - * We have dp2 locked. If it isn't first, unlock it. - * If it is first, tell xfs_lock_inodes so it can skip it - * when locking. if dp1 == dp2, xfs_lock_inodes will skip both - * since they are equal. xfs_lock_inodes needs all these inodes - * so that it can unlock and retry if there might be a dead-lock - * potential with the log. - */ - - if (i_tab[0] == dp2 && lock_mode == XFS_ILOCK_SHARED) { -#ifdef DEBUG - xfs_rename_skip++; -#endif - xfs_lock_inodes(i_tab, *num_inodes, 1, XFS_ILOCK_SHARED); - } else { -#ifdef DEBUG - xfs_rename_nskip++; -#endif - xfs_iunlock_map_shared(dp2, lock_mode); - xfs_lock_inodes(i_tab, *num_inodes, 0, XFS_ILOCK_SHARED); - } - - return 0; } /* @@ -202,10 +124,10 @@ xfs_rename( struct xfs_name *src_name, xfs_inode_t *src_ip, xfs_inode_t *target_dp, - struct xfs_name *target_name) + struct xfs_name *target_name, + xfs_inode_t *target_ip) { - xfs_trans_t *tp; - xfs_inode_t *target_ip; + xfs_trans_t *tp = NULL; xfs_mount_t *mp = src_dp->i_mount; int new_parent; /* moving to a new dir */ int src_is_directory; /* src_name is a directory */ @@ -230,64 +152,31 @@ xfs_rename( target_dp, DM_RIGHT_NULL, src_name->name, target_name->name, 0, 0, 0); - if (error) { + if (error) return error; - } } /* Return through std_return after this point. */ - /* - * Lock all the participating inodes. Depending upon whether - * the target_name exists in the target directory, and - * whether the target directory is the same as the source - * directory, we can lock from 2 to 4 inodes. - * xfs_lock_for_rename() will return ENOENT if src_name - * does not exist in the source directory. - */ - tp = NULL; - error = xfs_lock_for_rename(src_dp, target_dp, src_ip, target_name, - &target_ip, inodes, &num_inodes); - if (error) { - /* - * We have nothing locked, no inode references, and - * no transaction, so just get out. - */ - goto std_return; - } - - ASSERT(src_ip != NULL); + new_parent = (src_dp != target_dp); + src_is_directory = ((src_ip->i_d.di_mode & S_IFMT) == S_IFDIR); - if ((src_ip->i_d.di_mode & S_IFMT) == S_IFDIR) { + if (src_is_directory) { /* * Check for link count overflow on target_dp */ - if (target_ip == NULL && (src_dp != target_dp) && + if (target_ip == NULL && new_parent && target_dp->i_d.di_nlink >= XFS_MAXLINK) { error = XFS_ERROR(EMLINK); - xfs_rename_unlock4(inodes, XFS_ILOCK_SHARED); - goto rele_return; + goto std_return; } } - /* - * If we are using project inheritance, we only allow renames - * into our tree when the project IDs are the same; else the - * tree quota mechanism would be circumvented. - */ - if (unlikely((target_dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) && - (target_dp->i_d.di_projid != src_ip->i_d.di_projid))) { - error = XFS_ERROR(EXDEV); - xfs_rename_unlock4(inodes, XFS_ILOCK_SHARED); - goto rele_return; - } - - new_parent = (src_dp != target_dp); - src_is_directory = ((src_ip->i_d.di_mode & S_IFMT) == S_IFDIR); + xfs_sort_for_rename(src_dp, target_dp, src_ip, target_ip, + inodes, &num_inodes); - /* - * Drop the locks on our inodes so that we can start the transaction. - */ - xfs_rename_unlock4(inodes, XFS_ILOCK_SHARED); + IHOLD(src_ip); + if (target_ip) + IHOLD(target_ip); XFS_BMAP_INIT(&free_list, &first_block); tp = xfs_trans_alloc(mp, XFS_TRANS_RENAME); @@ -314,9 +203,25 @@ xfs_rename( } /* - * Reacquire the inode locks we dropped above. + * Lock all the participating inodes. Depending upon whether + * the target_name exists in the target directory, and + * whether the target directory is the same as the source + * directory, we can lock from 2 to 4 inodes. */ - xfs_lock_inodes(inodes, num_inodes, 0, XFS_ILOCK_EXCL); + xfs_lock_inodes(inodes, num_inodes, XFS_ILOCK_EXCL); + + /* + * If we are using project inheritance, we only allow renames + * into our tree when the project IDs are the same; else the + * tree quota mechanism would be circumvented. + */ + if (unlikely((target_dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) && + (target_dp->i_d.di_projid != src_ip->i_d.di_projid))) { + error = XFS_ERROR(EXDEV); + xfs_rename_unlock4(inodes, XFS_ILOCK_SHARED); + xfs_trans_cancel(tp, cancel_flags); + goto rele_return; + } /* * Join all the inodes to the transaction. From this point on, diff --git a/fs/xfs/xfs_utils.c b/fs/xfs/xfs_utils.c index 27075c9060ef..98e5f110ba5f 100644 --- a/fs/xfs/xfs_utils.c +++ b/fs/xfs/xfs_utils.c @@ -41,49 +41,6 @@ #include "xfs_utils.h" -int -xfs_dir_lookup_int( - xfs_inode_t *dp, - uint lock_mode, - struct xfs_name *name, - xfs_ino_t *inum, - xfs_inode_t **ipp) -{ - int error; - - xfs_itrace_entry(dp); - - error = xfs_dir_lookup(NULL, dp, name, inum); - if (!error) { - /* - * Unlock the directory. We do this because we can't - * hold the directory lock while doing the vn_get() - * in xfs_iget(). Doing so could cause us to hold - * a lock while waiting for the inode to finish - * being inactive while it's waiting for a log - * reservation in the inactive routine. - */ - xfs_iunlock(dp, lock_mode); - error = xfs_iget(dp->i_mount, NULL, *inum, 0, 0, ipp, 0); - xfs_ilock(dp, lock_mode); - - if (error) { - *ipp = NULL; - } else if ((*ipp)->i_d.di_mode == 0) { - /* - * The inode has been freed. Something is - * wrong so just get out of here. - */ - xfs_iunlock(dp, lock_mode); - xfs_iput_new(*ipp, 0); - *ipp = NULL; - xfs_ilock(dp, lock_mode); - error = XFS_ERROR(ENOENT); - } - } - return error; -} - /* * Allocates a new inode from disk and return a pointer to the * incore copy. This routine will internally commit the current diff --git a/fs/xfs/xfs_utils.h b/fs/xfs/xfs_utils.h index 175b126d2cab..f316cb85d8e2 100644 --- a/fs/xfs/xfs_utils.h +++ b/fs/xfs/xfs_utils.h @@ -21,8 +21,6 @@ #define IRELE(ip) VN_RELE(XFS_ITOV(ip)) #define IHOLD(ip) VN_HOLD(XFS_ITOV(ip)) -extern int xfs_dir_lookup_int(xfs_inode_t *, uint, struct xfs_name *, - xfs_ino_t *, xfs_inode_t **); extern int xfs_truncate_file(xfs_mount_t *, xfs_inode_t *); extern int xfs_dir_ialloc(xfs_trans_t **, xfs_inode_t *, mode_t, xfs_nlink_t, xfs_dev_t, cred_t *, prid_t, int, diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 322ba094dcc8..308dfff76ae2 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -1982,7 +1982,7 @@ again: ips[0] = ip; ips[1] = dp; - xfs_lock_inodes(ips, 2, 0, XFS_ILOCK_EXCL); + xfs_lock_inodes(ips, 2, XFS_ILOCK_EXCL); } /* else e_inum == dp->i_ino */ /* This can happen if we're asked to lock /x/.. @@ -2030,7 +2030,6 @@ void xfs_lock_inodes( xfs_inode_t **ips, int inodes, - int first_locked, uint lock_mode) { int attempts = 0, i, j, try_lock; @@ -2038,13 +2037,8 @@ xfs_lock_inodes( ASSERT(ips && (inodes >= 2)); /* we need at least two */ - if (first_locked) { - try_lock = 1; - i = 1; - } else { - try_lock = 0; - i = 0; - } + try_lock = 0; + i = 0; again: for (; i < inodes; i++) { @@ -2406,7 +2400,7 @@ xfs_link( ips[1] = sip; } - xfs_lock_inodes(ips, 2, 0, XFS_ILOCK_EXCL); + xfs_lock_inodes(ips, 2, XFS_ILOCK_EXCL); /* * Increment vnode ref counts since xfs_trans_commit & diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h index 6b66904a3837..ba798fccf72b 100644 --- a/fs/xfs/xfs_vnodeops.h +++ b/fs/xfs/xfs_vnodeops.h @@ -47,7 +47,7 @@ int xfs_change_file_space(struct xfs_inode *ip, int cmd, struct cred *credp, int attr_flags); int xfs_rename(struct xfs_inode *src_dp, struct xfs_name *src_name, struct xfs_inode *src_ip, struct xfs_inode *target_dp, - struct xfs_name *target_name); + struct xfs_name *target_name, struct xfs_inode *target_ip); int xfs_attr_get(struct xfs_inode *ip, const char *name, char *value, int *valuelenp, int flags, cred_t *cred); int xfs_attr_set(struct xfs_inode *dp, const char *name, char *value, -- cgit v1.2.3 From 1ac74e01df959e3e91baded7c83399372af945a2 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 22 Apr 2008 17:34:12 +1000 Subject: [XFS] kill usesless IHOLD calls in xfs_rename Similar to to the previous patch for remove and rmdir only grab a reference to inodes when we join them to transaction to balance the decrement on transaction completion. Everything else it taken care of by the VFS. Note that the old case had leaks of inode count when src == target or src or target == one of the parent inodes, but these cases are fortunately already rejected by the VFS. SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30904a Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_rename.c | 75 +++++++---------------------------------------------- 1 file changed, 10 insertions(+), 65 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_rename.c b/fs/xfs/xfs_rename.c index 6a141427f68a..d8063e1ad298 100644 --- a/fs/xfs/xfs_rename.c +++ b/fs/xfs/xfs_rename.c @@ -137,9 +137,7 @@ xfs_rename( int cancel_flags; int committed; xfs_inode_t *inodes[4]; - int target_ip_dropped = 0; /* dropped target_ip link? */ int spaceres; - int target_link_zero = 0; int num_inodes; xfs_itrace_entry(src_dp); @@ -174,10 +172,6 @@ xfs_rename( xfs_sort_for_rename(src_dp, target_dp, src_ip, target_ip, inodes, &num_inodes); - IHOLD(src_ip); - if (target_ip) - IHOLD(target_ip); - XFS_BMAP_INIT(&free_list, &first_block); tp = xfs_trans_alloc(mp, XFS_TRANS_RENAME); cancel_flags = XFS_TRANS_RELEASE_LOG_RES; @@ -191,7 +185,7 @@ xfs_rename( } if (error) { xfs_trans_cancel(tp, 0); - goto rele_return; + goto std_return; } /* @@ -199,7 +193,7 @@ xfs_rename( */ if ((error = XFS_QM_DQVOPRENAME(mp, inodes))) { xfs_trans_cancel(tp, cancel_flags); - goto rele_return; + goto std_return; } /* @@ -220,7 +214,7 @@ xfs_rename( error = XFS_ERROR(EXDEV); xfs_rename_unlock4(inodes, XFS_ILOCK_SHARED); xfs_trans_cancel(tp, cancel_flags); - goto rele_return; + goto std_return; } /* @@ -233,17 +227,17 @@ xfs_rename( */ IHOLD(src_dp); xfs_trans_ijoin(tp, src_dp, XFS_ILOCK_EXCL); + if (new_parent) { IHOLD(target_dp); xfs_trans_ijoin(tp, target_dp, XFS_ILOCK_EXCL); } - if ((src_ip != src_dp) && (src_ip != target_dp)) { - xfs_trans_ijoin(tp, src_ip, XFS_ILOCK_EXCL); - } - if ((target_ip != NULL) && - (target_ip != src_ip) && - (target_ip != src_dp) && - (target_ip != target_dp)) { + + IHOLD(src_ip); + xfs_trans_ijoin(tp, src_ip, XFS_ILOCK_EXCL); + + if (target_ip) { + IHOLD(target_ip); xfs_trans_ijoin(tp, target_ip, XFS_ILOCK_EXCL); } @@ -317,7 +311,6 @@ xfs_rename( error = xfs_droplink(tp, target_ip); if (error) goto abort_return; - target_ip_dropped = 1; if (src_is_directory) { /* @@ -327,10 +320,6 @@ xfs_rename( if (error) goto abort_return; } - - /* Do this test while we still hold the locks */ - target_link_zero = (target_ip)->i_d.di_nlink==0; - } /* target_ip != NULL */ /* @@ -396,15 +385,6 @@ xfs_rename( xfs_trans_log_inode(tp, target_dp, XFS_ILOG_CORE); } - /* - * If there was a target inode, take an extra reference on - * it here so that it doesn't go to xfs_inactive() from - * within the commit. - */ - if (target_ip != NULL) { - IHOLD(target_ip); - } - /* * If this is a synchronous mount, make sure that the * rename transaction goes to disk before returning to @@ -414,30 +394,11 @@ xfs_rename( xfs_trans_set_sync(tp); } - /* - * Take refs. for vop_link_removed calls below. No need to worry - * about directory refs. because the caller holds them. - * - * Do holds before the xfs_bmap_finish since it might rele them down - * to zero. - */ - - if (target_ip_dropped) - IHOLD(target_ip); - IHOLD(src_ip); - error = xfs_bmap_finish(&tp, &free_list, &committed); if (error) { xfs_bmap_cancel(&free_list); xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT)); - if (target_ip != NULL) { - IRELE(target_ip); - } - if (target_ip_dropped) { - IRELE(target_ip); - } - IRELE(src_ip); goto std_return; } @@ -446,15 +407,6 @@ xfs_rename( * the vnode references. */ error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); - if (target_ip != NULL) - IRELE(target_ip); - /* - * Let interposed file systems know about removed links. - */ - if (target_ip_dropped) - IRELE(target_ip); - - IRELE(src_ip); /* Fall through to std_return with error = 0 or errno from * xfs_trans_commit */ @@ -476,11 +428,4 @@ std_return: xfs_bmap_cancel(&free_list); xfs_trans_cancel(tp, cancel_flags); goto std_return; - - rele_return: - IRELE(src_ip); - if (target_ip != NULL) { - IRELE(target_ip); - } - goto std_return; } -- cgit v1.2.3 From 82dab941a192d081dd0b7cde3ed32603372d5acc Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 22 Apr 2008 17:34:18 +1000 Subject: [XFS] kill parent == child checks in xfs_remove and xfs_rmdir VFS guaranteed these can't happen. SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30911a Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_vnodeops.c | 25 ++++++------------------- 1 file changed, 6 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 308dfff76ae2..2ebfc60097d1 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -2180,7 +2180,7 @@ xfs_remove( xfs_itrace_ref(ip); error = XFS_QM_DQATTACH(mp, dp, 0); - if (!error && dp != ip) + if (!error) error = XFS_QM_DQATTACH(mp, ip, 0); if (error) { REMOVE_DEBUG_TRACE(__LINE__); @@ -2228,15 +2228,9 @@ xfs_remove( * inodes locked. */ xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL); - if (dp != ip) { - /* - * Increment vnode ref count only in this case since - * there's an extra vnode reference in the case where - * dp == ip. - */ - IHOLD(dp); - xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); - } + + IHOLD(dp); + xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); /* * Entry must exist since we did a lookup in xfs_lock_dir_and_entry. @@ -2747,7 +2741,7 @@ xfs_rmdir( * Get the dquots for the inodes. */ error = XFS_QM_DQATTACH(mp, dp, 0); - if (!error && dp != cdp) + if (!error) error = XFS_QM_DQATTACH(mp, cdp, 0); if (error) { IRELE(cdp); @@ -2796,14 +2790,7 @@ xfs_rmdir( } xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL); - if (dp != cdp) { - /* - * Only increment the parent directory vnode count if - * we didn't bump it in looking up cdp. The only time - * we don't bump it is when we're looking up ".". - */ - VN_HOLD(dir_vp); - } + VN_HOLD(dir_vp); xfs_itrace_ref(cdp); xfs_trans_ijoin(tp, cdp, XFS_ILOCK_EXCL); -- cgit v1.2.3 From 5df78e73d328e870a1cd8a9e0f39bf094e42ce9d Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 22 Apr 2008 17:34:24 +1000 Subject: [XFS] kill usesless IHOLD calls in xfs_remove and xfs_rmdir The VFS always has an inode reference when we call these functions. So we only need to grab a signle reference to each inode that's joined to a transaction - all the other bumping and dropping is as useless as the comments describing the IRIX semantics. SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30912a Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_vnodeops.c | 63 ++++----------------------------------------------- 1 file changed, 4 insertions(+), 59 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 2ebfc60097d1..70702a60b4bb 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -2162,20 +2162,6 @@ xfs_remove( return error; } - /* - * We need to get a reference to ip before we get our log - * reservation. The reason for this is that we cannot call - * xfs_iget for an inode for which we do not have a reference - * once we've acquired a log reservation. This is because the - * inode we are trying to get might be in xfs_inactive going - * for a log reservation. Since we'll have to wait for the - * inactive code to complete before returning from xfs_iget, - * we need to make sure that we don't have log space reserved - * when we call xfs_iget. Instead we get an unlocked reference - * to the inode before getting our log reservation. - */ - IHOLD(ip); - xfs_itrace_entry(ip); xfs_itrace_ref(ip); @@ -2184,7 +2170,6 @@ xfs_remove( error = XFS_QM_DQATTACH(mp, ip, 0); if (error) { REMOVE_DEBUG_TRACE(__LINE__); - IRELE(ip); goto std_return; } @@ -2211,7 +2196,6 @@ xfs_remove( ASSERT(error != ENOSPC); REMOVE_DEBUG_TRACE(__LINE__); xfs_trans_cancel(tp, 0); - IRELE(ip); return error; } @@ -2219,7 +2203,6 @@ xfs_remove( if (error) { REMOVE_DEBUG_TRACE(__LINE__); xfs_trans_cancel(tp, cancel_flags); - IRELE(ip); goto std_return; } @@ -2227,6 +2210,7 @@ xfs_remove( * At this point, we've gotten both the directory and the entry * inodes locked. */ + IHOLD(ip); xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL); IHOLD(dp); @@ -2259,12 +2243,6 @@ xfs_remove( */ link_zero = (ip)->i_d.di_nlink==0; - /* - * Take an extra ref on the inode so that it doesn't - * go to xfs_inactive() from within the commit. - */ - IHOLD(ip); - /* * If this is a synchronous mount, make sure that the * remove transaction goes to disk before returning to @@ -2281,10 +2259,8 @@ xfs_remove( } error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); - if (error) { - IRELE(ip); + if (error) goto std_return; - } /* * If we are using filestreams, kill the stream association. @@ -2296,7 +2272,6 @@ xfs_remove( xfs_filestream_deassociate(ip); xfs_itrace_exit(ip); - IRELE(ip); /* Fall through to std_return with error = 0 */ std_return: @@ -2325,8 +2300,6 @@ xfs_remove( cancel_flags |= XFS_TRANS_ABORT; xfs_trans_cancel(tp, cancel_flags); - IRELE(ip); - goto std_return; } @@ -2698,7 +2671,6 @@ xfs_rmdir( struct xfs_name *name, xfs_inode_t *cdp) { - bhv_vnode_t *dir_vp = XFS_ITOV(dp); xfs_mount_t *mp = dp->i_mount; xfs_trans_t *tp; int error; @@ -2723,20 +2695,6 @@ xfs_rmdir( return XFS_ERROR(error); } - /* - * We need to get a reference to cdp before we get our log - * reservation. The reason for this is that we cannot call - * xfs_iget for an inode for which we do not have a reference - * once we've acquired a log reservation. This is because the - * inode we are trying to get might be in xfs_inactive going - * for a log reservation. Since we'll have to wait for the - * inactive code to complete before returning from xfs_iget, - * we need to make sure that we don't have log space reserved - * when we call xfs_iget. Instead we get an unlocked reference - * to the inode before getting our log reservation. - */ - IHOLD(cdp); - /* * Get the dquots for the inodes. */ @@ -2744,7 +2702,6 @@ xfs_rmdir( if (!error) error = XFS_QM_DQATTACH(mp, cdp, 0); if (error) { - IRELE(cdp); REMOVE_DEBUG_TRACE(__LINE__); goto std_return; } @@ -2771,7 +2728,6 @@ xfs_rmdir( if (error) { ASSERT(error != ENOSPC); cancel_flags = 0; - IRELE(cdp); goto error_return; } XFS_BMAP_INIT(&free_list, &first_block); @@ -2785,14 +2741,13 @@ xfs_rmdir( error = xfs_lock_dir_and_entry(dp, cdp); if (error) { xfs_trans_cancel(tp, cancel_flags); - IRELE(cdp); goto std_return; } + IHOLD(dp); xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL); - VN_HOLD(dir_vp); - xfs_itrace_ref(cdp); + IHOLD(cdp); xfs_trans_ijoin(tp, cdp, XFS_ILOCK_EXCL); ASSERT(cdp->i_d.di_nlink >= 2); @@ -2845,12 +2800,6 @@ xfs_rmdir( /* Determine these before committing transaction */ last_cdp_link = (cdp)->i_d.di_nlink==0; - /* - * Take an extra ref on the child vnode so that it - * does not go to xfs_inactive() from within the commit. - */ - IHOLD(cdp); - /* * If this is a synchronous mount, make sure that the * rmdir transaction goes to disk before returning to @@ -2865,19 +2814,15 @@ xfs_rmdir( xfs_bmap_cancel(&free_list); xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT)); - IRELE(cdp); goto std_return; } error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); if (error) { - IRELE(cdp); goto std_return; } - IRELE(cdp); - /* Fall through to std_return with error = 0 or the errno * from xfs_trans_commit. */ std_return: -- cgit v1.2.3 From e8b0ebaa115ac46b21622b103c29927f5805aeaa Mon Sep 17 00:00:00 2001 From: Barry Naujok Date: Tue, 22 Apr 2008 17:34:31 +1000 Subject: [XFS] Cleanup xfs_attr a bit with xfs_name and remove cred SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30913a Signed-off-by: Barry Naujok Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/linux-2.6/xfs_ioctl.c | 6 +-- fs/xfs/xfs_acl.c | 7 ++-- fs/xfs/xfs_attr.c | 93 +++++++++++++++++++++++++------------------- fs/xfs/xfs_attr.h | 6 +-- fs/xfs/xfs_vnodeops.h | 2 +- 5 files changed, 62 insertions(+), 52 deletions(-) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index 98e87804991d..a42ba9d71156 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -505,14 +505,14 @@ xfs_attrmulti_attr_get( { char *kbuf; int error = EFAULT; - + if (*len > XATTR_SIZE_MAX) return EINVAL; kbuf = kmalloc(*len, GFP_KERNEL); if (!kbuf) return ENOMEM; - error = xfs_attr_get(XFS_I(inode), name, kbuf, (int *)len, flags, NULL); + error = xfs_attr_get(XFS_I(inode), name, kbuf, (int *)len, flags); if (error) goto out_kfree; @@ -546,7 +546,7 @@ xfs_attrmulti_attr_set( if (copy_from_user(kbuf, ubuf, len)) goto out_kfree; - + error = xfs_attr_set(XFS_I(inode), name, kbuf, len, flags); out_kfree: diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c index 796e76ef2713..ebee3a4f703a 100644 --- a/fs/xfs/xfs_acl.c +++ b/fs/xfs/xfs_acl.c @@ -334,14 +334,15 @@ xfs_acl_iaccess( { xfs_acl_t *acl; int rval; + struct xfs_name acl_name = {SGI_ACL_FILE, SGI_ACL_FILE_SIZE}; if (!(_ACL_ALLOC(acl))) return -1; /* If the file has no ACL return -1. */ rval = sizeof(xfs_acl_t); - if (xfs_attr_fetch(ip, SGI_ACL_FILE, SGI_ACL_FILE_SIZE, - (char *)acl, &rval, ATTR_ROOT | ATTR_KERNACCESS, cr)) { + if (xfs_attr_fetch(ip, &acl_name, (char *)acl, &rval, + ATTR_ROOT | ATTR_KERNACCESS)) { _ACL_FREE(acl); return -1; } @@ -579,7 +580,7 @@ xfs_acl_get_attr( *error = xfs_attr_get(xfs_vtoi(vp), kind == _ACL_TYPE_ACCESS ? SGI_ACL_FILE : SGI_ACL_DEFAULT, - (char *)aclp, &len, flags, sys_cred); + (char *)aclp, &len, flags); if (*error || (flags & ATTR_KERNOVAL)) return; xfs_acl_get_endian(aclp); diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c index 36d781ee5fcc..df151a859186 100644 --- a/fs/xfs/xfs_attr.c +++ b/fs/xfs/xfs_attr.c @@ -101,14 +101,28 @@ STATIC int xfs_attr_rmtval_remove(xfs_da_args_t *args); ktrace_t *xfs_attr_trace_buf; #endif +STATIC int +xfs_attr_name_to_xname( + struct xfs_name *xname, + const char *aname) +{ + if (!aname) + return EINVAL; + xname->name = aname; + xname->len = strlen(aname); + if (xname->len >= MAXNAMELEN) + return EFAULT; /* match IRIX behaviour */ + + return 0; +} /*======================================================================== * Overall external interface routines. *========================================================================*/ int -xfs_attr_fetch(xfs_inode_t *ip, const char *name, int namelen, - char *value, int *valuelenp, int flags, struct cred *cred) +xfs_attr_fetch(xfs_inode_t *ip, struct xfs_name *name, + char *value, int *valuelenp, int flags) { xfs_da_args_t args; int error; @@ -122,8 +136,8 @@ xfs_attr_fetch(xfs_inode_t *ip, const char *name, int namelen, * Fill in the arg structure for this request. */ memset((char *)&args, 0, sizeof(args)); - args.name = name; - args.namelen = namelen; + args.name = name->name; + args.namelen = name->len; args.value = value; args.valuelen = *valuelenp; args.flags = flags; @@ -162,31 +176,29 @@ xfs_attr_get( const char *name, char *value, int *valuelenp, - int flags, - cred_t *cred) + int flags) { - int error, namelen; + int error; + struct xfs_name xname; XFS_STATS_INC(xs_attr_get); - if (!name) - return(EINVAL); - namelen = strlen(name); - if (namelen >= MAXNAMELEN) - return(EFAULT); /* match IRIX behaviour */ - if (XFS_FORCED_SHUTDOWN(ip->i_mount)) return(EIO); + error = xfs_attr_name_to_xname(&xname, name); + if (error) + return error; + xfs_ilock(ip, XFS_ILOCK_SHARED); - error = xfs_attr_fetch(ip, name, namelen, value, valuelenp, flags, cred); + error = xfs_attr_fetch(ip, &xname, value, valuelenp, flags); xfs_iunlock(ip, XFS_ILOCK_SHARED); return(error); } -int -xfs_attr_set_int(xfs_inode_t *dp, const char *name, int namelen, - char *value, int valuelen, int flags) +STATIC int +xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name, + char *value, int valuelen, int flags) { xfs_da_args_t args; xfs_fsblock_t firstblock; @@ -209,7 +221,7 @@ xfs_attr_set_int(xfs_inode_t *dp, const char *name, int namelen, */ if (XFS_IFORK_Q(dp) == 0) { int sf_size = sizeof(xfs_attr_sf_hdr_t) + - XFS_ATTR_SF_ENTSIZE_BYNAME(namelen, valuelen); + XFS_ATTR_SF_ENTSIZE_BYNAME(name->len, valuelen); if ((error = xfs_bmap_add_attrfork(dp, sf_size, rsvd))) return(error); @@ -219,8 +231,8 @@ xfs_attr_set_int(xfs_inode_t *dp, const char *name, int namelen, * Fill in the arg structure for this request. */ memset((char *)&args, 0, sizeof(args)); - args.name = name; - args.namelen = namelen; + args.name = name->name; + args.namelen = name->len; args.value = value; args.valuelen = valuelen; args.flags = flags; @@ -236,7 +248,7 @@ xfs_attr_set_int(xfs_inode_t *dp, const char *name, int namelen, * Determine space new attribute will use, and if it would be * "local" or "remote" (note: local != inline). */ - size = xfs_attr_leaf_newentsize(namelen, valuelen, + size = xfs_attr_leaf_newentsize(name->len, valuelen, mp->m_sb.sb_blocksize, &local); nblks = XFS_DAENTER_SPACE_RES(mp, XFS_ATTR_FORK); @@ -429,26 +441,27 @@ xfs_attr_set( int valuelen, int flags) { - int namelen; - - namelen = strlen(name); - if (namelen >= MAXNAMELEN) - return EFAULT; /* match IRIX behaviour */ + int error; + struct xfs_name xname; XFS_STATS_INC(xs_attr_set); if (XFS_FORCED_SHUTDOWN(dp->i_mount)) return (EIO); - return xfs_attr_set_int(dp, name, namelen, value, valuelen, flags); + error = xfs_attr_name_to_xname(&xname, name); + if (error) + return error; + + return xfs_attr_set_int(dp, &xname, value, valuelen, flags); } /* * Generic handler routine to remove a name from an attribute list. * Transitions attribute list from Btree to shortform as necessary. */ -int -xfs_attr_remove_int(xfs_inode_t *dp, const char *name, int namelen, int flags) +STATIC int +xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags) { xfs_da_args_t args; xfs_fsblock_t firstblock; @@ -460,8 +473,8 @@ xfs_attr_remove_int(xfs_inode_t *dp, const char *name, int namelen, int flags) * Fill in the arg structure for this request. */ memset((char *)&args, 0, sizeof(args)); - args.name = name; - args.namelen = namelen; + args.name = name->name; + args.namelen = name->len; args.flags = flags; args.hashval = xfs_da_hashname(args.name, args.namelen); args.dp = dp; @@ -575,17 +588,18 @@ xfs_attr_remove( const char *name, int flags) { - int namelen; - - namelen = strlen(name); - if (namelen >= MAXNAMELEN) - return EFAULT; /* match IRIX behaviour */ + int error; + struct xfs_name xname; XFS_STATS_INC(xs_attr_remove); if (XFS_FORCED_SHUTDOWN(dp->i_mount)) return (EIO); + error = xfs_attr_name_to_xname(&xname, name); + if (error) + return error; + xfs_ilock(dp, XFS_ILOCK_SHARED); if (XFS_IFORK_Q(dp) == 0 || (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS && @@ -595,10 +609,10 @@ xfs_attr_remove( } xfs_iunlock(dp, XFS_ILOCK_SHARED); - return xfs_attr_remove_int(dp, name, namelen, flags); + return xfs_attr_remove_int(dp, &xname, flags); } -int /* error */ +STATIC int xfs_attr_list_int(xfs_attr_list_context_t *context) { int error; @@ -2522,8 +2536,7 @@ attr_generic_get( { int error, asize = size; - error = xfs_attr_get(xfs_vtoi(vp), name, data, - &asize, xflags, NULL); + error = xfs_attr_get(xfs_vtoi(vp), name, data, &asize, xflags); if (!error) return asize; return -error; diff --git a/fs/xfs/xfs_attr.h b/fs/xfs/xfs_attr.h index 786eba3121c4..6cfc9384fe35 100644 --- a/fs/xfs/xfs_attr.h +++ b/fs/xfs/xfs_attr.h @@ -158,14 +158,10 @@ struct xfs_da_args; /* * Overall external interface routines. */ -int xfs_attr_set_int(struct xfs_inode *, const char *, int, char *, int, int); -int xfs_attr_remove_int(struct xfs_inode *, const char *, int, int); -int xfs_attr_list_int(struct xfs_attr_list_context *); int xfs_attr_inactive(struct xfs_inode *dp); int xfs_attr_shortform_getvalue(struct xfs_da_args *); -int xfs_attr_fetch(struct xfs_inode *, const char *, int, - char *, int *, int, struct cred *); +int xfs_attr_fetch(struct xfs_inode *, struct xfs_name *, char *, int *, int); int xfs_attr_rmtval_get(struct xfs_da_args *args); #endif /* __XFS_ATTR_H__ */ diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h index ba798fccf72b..827afc997008 100644 --- a/fs/xfs/xfs_vnodeops.h +++ b/fs/xfs/xfs_vnodeops.h @@ -49,7 +49,7 @@ int xfs_rename(struct xfs_inode *src_dp, struct xfs_name *src_name, struct xfs_inode *src_ip, struct xfs_inode *target_dp, struct xfs_name *target_name, struct xfs_inode *target_ip); int xfs_attr_get(struct xfs_inode *ip, const char *name, char *value, - int *valuelenp, int flags, cred_t *cred); + int *valuelenp, int flags); int xfs_attr_set(struct xfs_inode *dp, const char *name, char *value, int valuelen, int flags); int xfs_attr_remove(struct xfs_inode *dp, const char *name, int flags); -- cgit v1.2.3 From d4d90b577ee5af5c1b29bd693aca026a77a1a2f1 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 22 Apr 2008 17:34:37 +1000 Subject: [XFS] Add xfs_icsb_sync_counters_locked for when m_sb_lock already held Add a new xfs_icsb_sync_counters_locked for the case where m_sb_lock is already taken and add a flags argument to xfs_icsb_sync_counters so that xfs_icsb_sync_counters_flags is not needed. SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30917a Signed-off-by: Christoph Hellwig Signed-off-by: David Chinner Signed-off-by: Lachlan McIlroy --- fs/xfs/linux-2.6/xfs_super.c | 2 +- fs/xfs/xfs_fsops.c | 4 ++-- fs/xfs/xfs_mount.c | 23 ++++++++--------------- fs/xfs/xfs_mount.h | 5 +++-- 4 files changed, 14 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 865eb708aa95..742b2c7852c1 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -1181,7 +1181,7 @@ xfs_fs_statfs( statp->f_fsid.val[0] = (u32)id; statp->f_fsid.val[1] = (u32)(id >> 32); - xfs_icsb_sync_counters_flags(mp, XFS_ICSB_LAZY_COUNT); + xfs_icsb_sync_counters(mp, XFS_ICSB_LAZY_COUNT); spin_lock(&mp->m_sb_lock); statp->f_bsize = sbp->sb_blocksize; diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index d3a0f538d6a6..5d5e9b34dd02 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -462,7 +462,7 @@ xfs_fs_counts( xfs_mount_t *mp, xfs_fsop_counts_t *cnt) { - xfs_icsb_sync_counters_flags(mp, XFS_ICSB_LAZY_COUNT); + xfs_icsb_sync_counters(mp, XFS_ICSB_LAZY_COUNT); spin_lock(&mp->m_sb_lock); cnt->freedata = mp->m_sb.sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); cnt->freertx = mp->m_sb.sb_frextents; @@ -524,7 +524,7 @@ xfs_reserve_blocks( */ retry: spin_lock(&mp->m_sb_lock); - xfs_icsb_sync_counters_flags(mp, XFS_ICSB_SB_LOCKED); + xfs_icsb_sync_counters_locked(mp, 0); /* * If our previous reservation was larger than the current value, diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 2fec452afbcc..a2fad07fd844 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -55,7 +55,6 @@ STATIC void xfs_unmountfs_wait(xfs_mount_t *); STATIC void xfs_icsb_destroy_counters(xfs_mount_t *); STATIC void xfs_icsb_balance_counter(xfs_mount_t *, xfs_sb_field_t, int, int); -STATIC void xfs_icsb_sync_counters(xfs_mount_t *); STATIC int xfs_icsb_modify_counters(xfs_mount_t *, xfs_sb_field_t, int64_t, int); STATIC void xfs_icsb_disable_counter(xfs_mount_t *, xfs_sb_field_t); @@ -64,7 +63,6 @@ STATIC void xfs_icsb_disable_counter(xfs_mount_t *, xfs_sb_field_t); #define xfs_icsb_destroy_counters(mp) do { } while (0) #define xfs_icsb_balance_counter(mp, a, b, c) do { } while (0) -#define xfs_icsb_sync_counters(mp) do { } while (0) #define xfs_icsb_modify_counters(mp, a, b, c) do { } while (0) #endif @@ -1400,7 +1398,7 @@ xfs_log_sbcount( if (!xfs_fs_writable(mp)) return 0; - xfs_icsb_sync_counters(mp); + xfs_icsb_sync_counters(mp, 0); /* * we don't need to do this if we are updating the superblock @@ -2278,38 +2276,33 @@ xfs_icsb_enable_counter( } void -xfs_icsb_sync_counters_flags( +xfs_icsb_sync_counters_locked( xfs_mount_t *mp, int flags) { xfs_icsb_cnts_t cnt; - /* Pass 1: lock all counters */ - if ((flags & XFS_ICSB_SB_LOCKED) == 0) - spin_lock(&mp->m_sb_lock); - xfs_icsb_count(mp, &cnt, flags); - /* Step 3: update mp->m_sb fields */ if (!xfs_icsb_counter_disabled(mp, XFS_SBS_ICOUNT)) mp->m_sb.sb_icount = cnt.icsb_icount; if (!xfs_icsb_counter_disabled(mp, XFS_SBS_IFREE)) mp->m_sb.sb_ifree = cnt.icsb_ifree; if (!xfs_icsb_counter_disabled(mp, XFS_SBS_FDBLOCKS)) mp->m_sb.sb_fdblocks = cnt.icsb_fdblocks; - - if ((flags & XFS_ICSB_SB_LOCKED) == 0) - spin_unlock(&mp->m_sb_lock); } /* * Accurate update of per-cpu counters to incore superblock */ -STATIC void +void xfs_icsb_sync_counters( - xfs_mount_t *mp) + xfs_mount_t *mp, + int flags) { - xfs_icsb_sync_counters_flags(mp, 0); + spin_lock(&mp->m_sb_lock); + xfs_icsb_sync_counters_locked(mp, flags); + spin_unlock(&mp->m_sb_lock); } /* diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 1ed575110ff0..06ecaeb338a5 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -211,12 +211,13 @@ typedef struct xfs_icsb_cnts { extern int xfs_icsb_init_counters(struct xfs_mount *); extern void xfs_icsb_reinit_counters(struct xfs_mount *); -extern void xfs_icsb_sync_counters_flags(struct xfs_mount *, int); +extern void xfs_icsb_sync_counters(struct xfs_mount *, int); +extern void xfs_icsb_sync_counters_locked(struct xfs_mount *, int); #else #define xfs_icsb_init_counters(mp) (0) #define xfs_icsb_reinit_counters(mp) do { } while (0) -#define xfs_icsb_sync_counters_flags(mp, flags) do { } while (0) +#define xfs_icsb_sync_counters(mp, flags) do { } while (0) #endif typedef struct xfs_ail { -- cgit v1.2.3 From 45af6c6de6453b385c80555c0ee40ab5fc4a033b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 22 Apr 2008 17:34:44 +1000 Subject: [XFS] split xfs_icsb_balance_counter Add an xfs_icsb_balance_counter_locked for the case where mp->m_sb_lock is already locked. SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30918a Signed-off-by: Christoph Hellwig Signed-off-by: David Chinner Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_mount.c | 58 +++++++++++++++++++++++++++++------------------------- 1 file changed, 31 insertions(+), 27 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index a2fad07fd844..8bdc16381bc5 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -54,7 +54,9 @@ STATIC void xfs_unmountfs_wait(xfs_mount_t *); #ifdef HAVE_PERCPU_SB STATIC void xfs_icsb_destroy_counters(xfs_mount_t *); STATIC void xfs_icsb_balance_counter(xfs_mount_t *, xfs_sb_field_t, - int, int); + int); +STATIC void xfs_icsb_balance_counter_locked(xfs_mount_t *, xfs_sb_field_t, + int); STATIC int xfs_icsb_modify_counters(xfs_mount_t *, xfs_sb_field_t, int64_t, int); STATIC void xfs_icsb_disable_counter(xfs_mount_t *, xfs_sb_field_t); @@ -62,7 +64,8 @@ STATIC void xfs_icsb_disable_counter(xfs_mount_t *, xfs_sb_field_t); #else #define xfs_icsb_destroy_counters(mp) do { } while (0) -#define xfs_icsb_balance_counter(mp, a, b, c) do { } while (0) +#define xfs_icsb_balance_counter(mp, a, b) do { } while (0) +#define xfs_icsb_balance_counter_locked(mp, a, b) do { } while (0) #define xfs_icsb_modify_counters(mp, a, b, c) do { } while (0) #endif @@ -2024,9 +2027,9 @@ xfs_icsb_cpu_notify( case CPU_ONLINE: case CPU_ONLINE_FROZEN: xfs_icsb_lock(mp); - xfs_icsb_balance_counter(mp, XFS_SBS_ICOUNT, 0, 0); - xfs_icsb_balance_counter(mp, XFS_SBS_IFREE, 0, 0); - xfs_icsb_balance_counter(mp, XFS_SBS_FDBLOCKS, 0, 0); + xfs_icsb_balance_counter(mp, XFS_SBS_ICOUNT, 0); + xfs_icsb_balance_counter(mp, XFS_SBS_IFREE, 0); + xfs_icsb_balance_counter(mp, XFS_SBS_FDBLOCKS, 0); xfs_icsb_unlock(mp); break; case CPU_DEAD: @@ -2046,12 +2049,9 @@ xfs_icsb_cpu_notify( memset(cntp, 0, sizeof(xfs_icsb_cnts_t)); - xfs_icsb_balance_counter(mp, XFS_SBS_ICOUNT, - XFS_ICSB_SB_LOCKED, 0); - xfs_icsb_balance_counter(mp, XFS_SBS_IFREE, - XFS_ICSB_SB_LOCKED, 0); - xfs_icsb_balance_counter(mp, XFS_SBS_FDBLOCKS, - XFS_ICSB_SB_LOCKED, 0); + xfs_icsb_balance_counter_locked(mp, XFS_SBS_ICOUNT, 0); + xfs_icsb_balance_counter_locked(mp, XFS_SBS_IFREE, 0); + xfs_icsb_balance_counter_locked(mp, XFS_SBS_FDBLOCKS, 0); spin_unlock(&mp->m_sb_lock); xfs_icsb_unlock(mp); break; @@ -2103,9 +2103,9 @@ xfs_icsb_reinit_counters( * initial balance kicks us off correctly */ mp->m_icsb_counters = -1; - xfs_icsb_balance_counter(mp, XFS_SBS_ICOUNT, 0, 0); - xfs_icsb_balance_counter(mp, XFS_SBS_IFREE, 0, 0); - xfs_icsb_balance_counter(mp, XFS_SBS_FDBLOCKS, 0, 0); + xfs_icsb_balance_counter(mp, XFS_SBS_ICOUNT, 0); + xfs_icsb_balance_counter(mp, XFS_SBS_IFREE, 0); + xfs_icsb_balance_counter(mp, XFS_SBS_FDBLOCKS, 0); xfs_icsb_unlock(mp); } @@ -2325,19 +2325,15 @@ xfs_icsb_sync_counters( #define XFS_ICSB_FDBLK_CNTR_REENABLE(mp) \ (uint64_t)(512 + XFS_ALLOC_SET_ASIDE(mp)) STATIC void -xfs_icsb_balance_counter( +xfs_icsb_balance_counter_locked( xfs_mount_t *mp, xfs_sb_field_t field, - int flags, int min_per_cpu) { uint64_t count, resid; int weight = num_online_cpus(); uint64_t min = (uint64_t)min_per_cpu; - if (!(flags & XFS_ICSB_SB_LOCKED)) - spin_lock(&mp->m_sb_lock); - /* disable counter and sync counter */ xfs_icsb_disable_counter(mp, field); @@ -2347,19 +2343,19 @@ xfs_icsb_balance_counter( count = mp->m_sb.sb_icount; resid = do_div(count, weight); if (count < max(min, XFS_ICSB_INO_CNTR_REENABLE)) - goto out; + return; break; case XFS_SBS_IFREE: count = mp->m_sb.sb_ifree; resid = do_div(count, weight); if (count < max(min, XFS_ICSB_INO_CNTR_REENABLE)) - goto out; + return; break; case XFS_SBS_FDBLOCKS: count = mp->m_sb.sb_fdblocks; resid = do_div(count, weight); if (count < max(min, XFS_ICSB_FDBLK_CNTR_REENABLE(mp))) - goto out; + return; break; default: BUG(); @@ -2368,9 +2364,17 @@ xfs_icsb_balance_counter( } xfs_icsb_enable_counter(mp, field, count, resid); -out: - if (!(flags & XFS_ICSB_SB_LOCKED)) - spin_unlock(&mp->m_sb_lock); +} + +STATIC void +xfs_icsb_balance_counter( + xfs_mount_t *mp, + xfs_sb_field_t fields, + int min_per_cpu) +{ + spin_lock(&mp->m_sb_lock); + xfs_icsb_balance_counter_locked(mp, fields, min_per_cpu); + spin_unlock(&mp->m_sb_lock); } STATIC int @@ -2477,7 +2481,7 @@ slow_path: * we are done. */ if (ret != ENOSPC) - xfs_icsb_balance_counter(mp, field, 0, 0); + xfs_icsb_balance_counter(mp, field, 0); xfs_icsb_unlock(mp); return ret; @@ -2501,7 +2505,7 @@ balance_counter: * will either succeed through the fast path or slow path without * another balance operation being required. */ - xfs_icsb_balance_counter(mp, field, 0, delta); + xfs_icsb_balance_counter(mp, field, delta); xfs_icsb_unlock(mp); goto again; } -- cgit v1.2.3 From ce46193bcaaf3c769718bcec6eae94719b8f53ed Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 22 Apr 2008 17:34:50 +1000 Subject: [XFS] kill XFS_ICSB_SB_LOCKED With the last two patches XFS_ICSB_SB_LOCKED is never checked and only superflously passed to xfs_icsb_count, so kill it. SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30920a Signed-off-by: Christoph Hellwig Signed-off-by: David Chinner Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_mount.c | 2 +- fs/xfs/xfs_mount.h | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 8bdc16381bc5..da3988453b71 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -2221,7 +2221,7 @@ xfs_icsb_disable_counter( if (!test_and_set_bit(field, &mp->m_icsb_counters)) { /* drain back to superblock */ - xfs_icsb_count(mp, &cnt, XFS_ICSB_SB_LOCKED|XFS_ICSB_LAZY_COUNT); + xfs_icsb_count(mp, &cnt, XFS_ICSB_LAZY_COUNT); switch(field) { case XFS_SBS_ICOUNT: mp->m_sb.sb_icount = cnt.icsb_icount; diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 06ecaeb338a5..27b558ee576f 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -206,7 +206,6 @@ typedef struct xfs_icsb_cnts { #define XFS_ICSB_FLAG_LOCK (1 << 0) /* counter lock bit */ -#define XFS_ICSB_SB_LOCKED (1 << 0) /* sb already locked */ #define XFS_ICSB_LAZY_COUNT (1 << 1) /* accuracy not needed */ extern int xfs_icsb_init_counters(struct xfs_mount *); -- cgit v1.2.3 From 18d18208daced52123de9ba0808447058d3442d8 Mon Sep 17 00:00:00 2001 From: Donald Douwsma Date: Tue, 22 Apr 2008 17:34:56 +1000 Subject: [XFS] Fix broken HAVE_SPLICE removal commit. Commit e687330b5ed1ea899fdaf0dea50aba196b6e019a was meant to remove the unused HAVE_SPLICE macro, instead an unrelated change was checked enabling QUOTADEBUG when building DEBUG XFS. Restore the intended changes. SGI-PV: 971046 SGI-Modid: xfs-linux-melb:xfs-kern:30924a Signed-off-by: Donald Douwsma Signed-off-by: Barry Naujok Signed-off-by: Lachlan McIlroy --- fs/xfs/linux-2.6/xfs_linux.h | 1 - fs/xfs/xfs.h | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h index e5143323e71f..1bc9f600365f 100644 --- a/fs/xfs/linux-2.6/xfs_linux.h +++ b/fs/xfs/linux-2.6/xfs_linux.h @@ -99,7 +99,6 @@ /* * Feature macros (disable/enable) */ -#define HAVE_SPLICE /* a splice(2) exists in 2.6, but not in 2.4 */ #ifdef CONFIG_SMP #define HAVE_PERCPU_SB /* per cpu superblock counters are a 2.6 feature */ #else diff --git a/fs/xfs/xfs.h b/fs/xfs/xfs.h index 765aaf65e2d3..540e4c989825 100644 --- a/fs/xfs/xfs.h +++ b/fs/xfs/xfs.h @@ -22,7 +22,7 @@ #define STATIC #define DEBUG 1 #define XFS_BUF_LOCK_TRACKING 1 -#define QUOTADEBUG 1 +/* #define QUOTADEBUG 1 */ #endif #ifdef CONFIG_XFS_TRACE -- cgit v1.2.3 From 7155054c9d8b5974f6e788b46939b419bd5fb020 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Tue, 29 Apr 2008 12:53:00 +1000 Subject: [XFS] fix non-smp xfs build xfs_reserve_blocks() calls xfs_icsb_sync_counters_locked(), which is not defined if !CONFIG_SMP/!HAVE_PERCPU_SB SGI-PV: 976035 SGI-Modid: xfs-linux-melb:xfs-kern:30991a Signed-off-by: Eric Sandeen Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_mount.h | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 27b558ee576f..63e0693a358a 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -217,6 +217,7 @@ extern void xfs_icsb_sync_counters_locked(struct xfs_mount *, int); #define xfs_icsb_init_counters(mp) (0) #define xfs_icsb_reinit_counters(mp) do { } while (0) #define xfs_icsb_sync_counters(mp, flags) do { } while (0) +#define xfs_icsb_sync_counters_locked(mp, flags) do { } while (0) #endif typedef struct xfs_ail { -- cgit v1.2.3 From fe0754f0e5c0f070bf82b6e7e5e8fa5a188163fc Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 29 Apr 2008 12:53:08 +1000 Subject: [XFS] remove xfs_log_ticket_zone on rmmod Fix bug introduced in commit eb01c9cd87c7a9998c2edf209721ea069e3e3652 aka "[XFS] Remove the xlog_ticket allocator" SGI-PV: 980887 SGI-Modid: xfs-linux-melb:xfs-kern:30995a Signed-off-by: Alexey Dobriyan Signed-off-by: David Chinner Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_vfsops.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/xfs/xfs_vfsops.c b/fs/xfs/xfs_vfsops.c index fc48158fe479..30bacd8bb0e5 100644 --- a/fs/xfs/xfs_vfsops.c +++ b/fs/xfs/xfs_vfsops.c @@ -186,6 +186,7 @@ xfs_cleanup(void) kmem_zone_destroy(xfs_efi_zone); kmem_zone_destroy(xfs_ifork_zone); kmem_zone_destroy(xfs_ili_zone); + kmem_zone_destroy(xfs_log_ticket_zone); } /* -- cgit v1.2.3 From d349404ff14758dc9a2d3df032073ed795085860 Mon Sep 17 00:00:00 2001 From: David Chinner Date: Tue, 29 Apr 2008 12:53:15 +1000 Subject: [XFS] Don't double count reserved block changes on UP. On uniprocessor machines, the incore superblock is used for all in memory accounting of free blocks. in this situation, changes to the reserved block count are accounted twice; once directly and once via xfs_mod_incore_sb(). Seeing as the modification on SMP is done via xfs_mod_incore_sb(), make this the only update mechanism that UP uses as well. SGI-PV: 980654 SGI-Modid: xfs-linux-melb:xfs-kern:30997a Signed-off-by: David Chinner Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_fsops.c | 4 ---- 1 file changed, 4 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index 5d5e9b34dd02..381ebda4f7bc 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -552,11 +552,8 @@ retry: mp->m_resblks += free; mp->m_resblks_avail += free; fdblks_delta = -free; - mp->m_sb.sb_fdblocks = XFS_ALLOC_SET_ASIDE(mp); } else { fdblks_delta = -delta; - mp->m_sb.sb_fdblocks = - lcounter + XFS_ALLOC_SET_ASIDE(mp); mp->m_resblks = request; mp->m_resblks_avail += delta; } @@ -587,7 +584,6 @@ out: if (error == ENOSPC) goto retry; } - return 0; } -- cgit v1.2.3 From 86c4d62305649848164ae311a0959fc569b0d964 Mon Sep 17 00:00:00 2001 From: David Chinner Date: Tue, 29 Apr 2008 12:53:21 +1000 Subject: [XFS] Fix check for block zero access in xfs_write_iomap_allocate() The check for block zero access should be done on non-realtime inodes. Fix the logic error in xfs_write_iomap_allocate(), and simplify the logic on all checks for block zero access in xfs_iomap.c SGI-PV: 980888 SGI-Modid: xfs-linux-melb:xfs-kern:30998a Signed-off-by: David Chinner Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_iomap.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index a2c3200a099f..7edcde691d1a 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -523,8 +523,7 @@ xfs_iomap_write_direct( goto error_out; } - if (unlikely(!imap.br_startblock && - !(XFS_IS_REALTIME_INODE(ip)))) { + if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip))) { error = xfs_cmn_err_fsblock_zero(ip, &imap); goto error_out; } @@ -686,8 +685,7 @@ retry: goto retry; } - if (unlikely(!imap[0].br_startblock && - !(XFS_IS_REALTIME_INODE(ip)))) + if (!(imap[0].br_startblock || XFS_IS_REALTIME_INODE(ip))) return xfs_cmn_err_fsblock_zero(ip, &imap[0]); *ret_imap = imap[0]; @@ -838,9 +836,9 @@ xfs_iomap_write_allocate( * See if we were able to allocate an extent that * covers at least part of the callers request */ - if (unlikely(!imap.br_startblock && - XFS_IS_REALTIME_INODE(ip))) + if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip))) return xfs_cmn_err_fsblock_zero(ip, &imap); + if ((offset_fsb >= imap.br_startoff) && (offset_fsb < (imap.br_startoff + imap.br_blockcount))) { @@ -934,8 +932,7 @@ xfs_iomap_write_unwritten( if (error) return XFS_ERROR(error); - if (unlikely(!imap.br_startblock && - !(XFS_IS_REALTIME_INODE(ip)))) + if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip))) return xfs_cmn_err_fsblock_zero(ip, &imap); if ((numblks_fsb = imap.br_blockcount) == 0) { -- cgit v1.2.3 From 359346a9655c8800408ed3ca44517ac7ea95c197 Mon Sep 17 00:00:00 2001 From: David Chinner Date: Tue, 29 Apr 2008 12:53:32 +1000 Subject: [XFS] Don't initialise new inode generation numbers to zero When we allocation new inode chunks, we initialise the generation numbers to zero. This works fine until we delete a chunk and then reallocate it, resulting in the same inode numbers but with a reset generation count. This can result in inode/generation pairs of different inodes occurring relatively close together. Given that the inode/gen pair makes up the "unique" portion of an NFS filehandle on XFS, this can result in file handles cached on clients being seen on the wire from the server but refer to a different file. This causes .... issues for NFS clients. Hence we need a unique generation number initialisation for each inode to prevent reuse of a small portion of the generation number space. Use a random number to initialise the generation number so we don't need to keep any new state on disk whilst making the new number difficult to guess from previous allocations. SGI-PV: 979416 SGI-Modid: xfs-linux-melb:xfs-kern:31001a Signed-off-by: David Chinner Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/xfs_ialloc.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'fs') diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c index a64dfbd565a5..aad8c5da38af 100644 --- a/fs/xfs/xfs_ialloc.c +++ b/fs/xfs/xfs_ialloc.c @@ -147,6 +147,7 @@ xfs_ialloc_ag_alloc( int version; /* inode version number to use */ int isaligned = 0; /* inode allocation at stripe unit */ /* boundary */ + unsigned int gen; args.tp = tp; args.mp = tp->t_mountp; @@ -290,6 +291,14 @@ xfs_ialloc_ag_alloc( else version = XFS_DINODE_VERSION_1; + /* + * Seed the new inode cluster with a random generation number. This + * prevents short-term reuse of generation numbers if a chunk is + * freed and then immediately reallocated. We use random numbers + * rather than a linear progression to prevent the next generation + * number from being easily guessable. + */ + gen = random32(); for (j = 0; j < nbufs; j++) { /* * Get the block. @@ -309,6 +318,7 @@ xfs_ialloc_ag_alloc( free = XFS_MAKE_IPTR(args.mp, fbuf, i); free->di_core.di_magic = cpu_to_be16(XFS_DINODE_MAGIC); free->di_core.di_version = version; + free->di_core.di_gen = cpu_to_be32(gen); free->di_next_unlinked = cpu_to_be32(NULLAGINO); xfs_ialloc_log_di(tp, fbuf, i, XFS_DI_CORE_BITS | XFS_DI_NEXT_UNLINKED); -- cgit v1.2.3 From 7788fae6cce616fe2c624273fcfe54cf50f5c38b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 21 Apr 2008 17:22:27 +1000 Subject: [XFS] allow enabling CONFIG_XFS_DEBUG Back when I first submitted XFS for mainline inclusion we made the decision that the debug code is far to extensive to be accidentally enabled by users in mainline. But then again it's often quite useful to track problems down and hacking the makefile all the time is rather annoying. Given all the debug options with even more overhead like lockdep or DEBUG_PAGE_ALLOC users (or rather developers) should know by now what they're doing. Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/Kconfig | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'fs') diff --git a/fs/xfs/Kconfig b/fs/xfs/Kconfig index 524021ff5436..3f53dd101f99 100644 --- a/fs/xfs/Kconfig +++ b/fs/xfs/Kconfig @@ -64,3 +64,16 @@ config XFS_RT See the xfs man page in section 5 for additional information. If unsure, say N. + +config XFS_DEBUG + bool "XFS Debugging support (EXPERIMENTAL)" + depends on XFS_FS && EXPERIMENTAL + help + Say Y here to get an XFS build with many debugging features, + including ASSERT checks, function wrappers around macros, + and extra sanity-checking functions in various code paths. + + Note that the resulting code will be HUGE and SLOW, and probably + not useful unless you are debugging a particular problem. + + Say N unless you are an XFS developer, or you play one on TV. -- cgit v1.2.3 From 3a738a5c73e0617d11b27ac46dd6a1a8f752017b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 21 Apr 2008 17:25:35 +1000 Subject: [XFS] remove sendfile leftovers Remove the last sendfile leftovers in mainline. This code is already gone in CVS. Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/linux-2.6/xfs_lrw.h | 1 - fs/xfs/xfs_vnodeops.h | 3 --- 2 files changed, 4 deletions(-) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/xfs_lrw.h b/fs/xfs/linux-2.6/xfs_lrw.h index e1d498b4ba7a..e6be37dbd0e9 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.h +++ b/fs/xfs/linux-2.6/xfs_lrw.h @@ -50,7 +50,6 @@ struct xfs_iomap; #define XFS_INVAL_CACHED 18 #define XFS_DIORD_ENTER 19 #define XFS_DIOWR_ENTER 20 -#define XFS_SENDFILE_ENTER 21 #define XFS_WRITEPAGE_ENTER 22 #define XFS_RELEASEPAGE_ENTER 23 #define XFS_INVALIDPAGE_ENTER 24 diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h index 827afc997008..8abe8f186e20 100644 --- a/fs/xfs/xfs_vnodeops.h +++ b/fs/xfs/xfs_vnodeops.h @@ -60,9 +60,6 @@ int xfs_ioctl(struct xfs_inode *ip, struct file *filp, ssize_t xfs_read(struct xfs_inode *ip, struct kiocb *iocb, const struct iovec *iovp, unsigned int segs, loff_t *offset, int ioflags); -ssize_t xfs_sendfile(struct xfs_inode *ip, struct file *filp, - loff_t *offset, int ioflags, size_t count, - read_actor_t actor, void *target); ssize_t xfs_splice_read(struct xfs_inode *ip, struct file *infilp, loff_t *ppos, struct pipe_inode_info *pipe, size_t count, int flags, int ioflags); -- cgit v1.2.3 From c5acbaf43da139fe014d78d1f0ca7754fa856ddb Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 21 Apr 2008 18:11:13 +1000 Subject: [XFS] remove dmapi cruft in xfs_file.c The dmapi cruft in xfs_file.c is totally out of date in mainline vs CVS, and at this point just removing this code which can't be used on mainline at all seems to be the best option to keep it maintainable. Signed-off-by: Christoph Hellwig Signed-off-by: Lachlan McIlroy --- fs/xfs/linux-2.6/xfs_file.c | 75 --------------------------------------------- 1 file changed, 75 deletions(-) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c index 05905246434d..65e78c13d4ae 100644 --- a/fs/xfs/linux-2.6/xfs_file.c +++ b/fs/xfs/linux-2.6/xfs_file.c @@ -43,9 +43,6 @@ #include static struct vm_operations_struct xfs_file_vm_ops; -#ifdef CONFIG_XFS_DMAPI -static struct vm_operations_struct xfs_dmapi_file_vm_ops; -#endif STATIC_INLINE ssize_t __xfs_file_read( @@ -202,22 +199,6 @@ xfs_file_fsync( (xfs_off_t)0, (xfs_off_t)-1); } -#ifdef CONFIG_XFS_DMAPI -STATIC int -xfs_vm_fault( - struct vm_area_struct *vma, - struct vm_fault *vmf) -{ - struct inode *inode = vma->vm_file->f_path.dentry->d_inode; - bhv_vnode_t *vp = vn_from_inode(inode); - - ASSERT_ALWAYS(vp->v_vfsp->vfs_flag & VFS_DMI); - if (XFS_SEND_MMAP(XFS_VFSTOM(vp->v_vfsp), vma, 0)) - return VM_FAULT_SIGBUS; - return filemap_fault(vma, vmf); -} -#endif /* CONFIG_XFS_DMAPI */ - /* * Unfortunately we can't just use the clean and simple readdir implementation * below, because nfs might call back into ->lookup from the filldir callback @@ -386,11 +367,6 @@ xfs_file_mmap( vma->vm_ops = &xfs_file_vm_ops; vma->vm_flags |= VM_CAN_NONLINEAR; -#ifdef CONFIG_XFS_DMAPI - if (XFS_M(filp->f_path.dentry->d_inode->i_sb)->m_flags & XFS_MOUNT_DMAPI) - vma->vm_ops = &xfs_dmapi_file_vm_ops; -#endif /* CONFIG_XFS_DMAPI */ - file_accessed(filp); return 0; } @@ -437,47 +413,6 @@ xfs_file_ioctl_invis( return error; } -#ifdef CONFIG_XFS_DMAPI -#ifdef HAVE_VMOP_MPROTECT -STATIC int -xfs_vm_mprotect( - struct vm_area_struct *vma, - unsigned int newflags) -{ - struct inode *inode = vma->vm_file->f_path.dentry->d_inode; - struct xfs_mount *mp = XFS_M(inode->i_sb); - int error = 0; - - if (mp->m_flags & XFS_MOUNT_DMAPI) { - if ((vma->vm_flags & VM_MAYSHARE) && - (newflags & VM_WRITE) && !(vma->vm_flags & VM_WRITE)) - error = XFS_SEND_MMAP(mp, vma, VM_WRITE); - } - return error; -} -#endif /* HAVE_VMOP_MPROTECT */ -#endif /* CONFIG_XFS_DMAPI */ - -#ifdef HAVE_FOP_OPEN_EXEC -/* If the user is attempting to execute a file that is offline then - * we have to trigger a DMAPI READ event before the file is marked as busy - * otherwise the invisible I/O will not be able to write to the file to bring - * it back online. - */ -STATIC int -xfs_file_open_exec( - struct inode *inode) -{ - struct xfs_mount *mp = XFS_M(inode->i_sb); - struct xfs_inode *ip = XFS_I(inode); - - if (unlikely(mp->m_flags & XFS_MOUNT_DMAPI) && - DM_EVENT_ENABLED(ip, DM_EVENT_READ)) - return -XFS_SEND_DATA(mp, DM_EVENT_READ, ip, 0, 0, 0, NULL); - return 0; -} -#endif /* HAVE_FOP_OPEN_EXEC */ - /* * mmap()d file has taken write protection fault and is being made * writable. We can set the page state up correctly for a writable @@ -546,13 +481,3 @@ static struct vm_operations_struct xfs_file_vm_ops = { .fault = filemap_fault, .page_mkwrite = xfs_vm_page_mkwrite, }; - -#ifdef CONFIG_XFS_DMAPI -static struct vm_operations_struct xfs_dmapi_file_vm_ops = { - .fault = xfs_vm_fault, - .page_mkwrite = xfs_vm_page_mkwrite, -#ifdef HAVE_VMOP_MPROTECT - .mprotect = xfs_vm_mprotect, -#endif -}; -#endif /* CONFIG_XFS_DMAPI */ -- cgit v1.2.3 From adaa693b845373296631766176ebf0f73a342e10 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Tue, 22 Apr 2008 15:26:13 +1000 Subject: [XFS] Fix build failure after enabling CONFIG_XFS_DEBUG Signed-off-by: Stephen Rothwell Signed-off-by: Lachlan McIlroy --- fs/xfs/linux-2.6/xfs_buf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 52f6846101d5..5105015a75ad 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -886,7 +886,7 @@ int xfs_buf_lock_value( xfs_buf_t *bp) { - return atomic_read(&bp->b_sema.count); + return bp->b_sema.count; } #endif -- cgit v1.2.3 From c3270e577c18b3d0e984c3371493205a4807db9d Mon Sep 17 00:00:00 2001 From: Tom Zanussi Date: Thu, 24 Apr 2008 12:52:20 +0200 Subject: relay: fix splice problem Splice isn't always incrementing the ppos correctly, which broke relay splice. Signed-off-by: Tom Zanussi Signed-off-by: Jens Axboe --- fs/splice.c | 2 +- kernel/relay.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/splice.c b/fs/splice.c index eeb1a86a7014..633f58ebfb72 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -1075,7 +1075,7 @@ long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, ret = splice_direct_to_actor(in, &sd, direct_splice_actor); if (ret > 0) - *ppos += ret; + *ppos = sd.pos; return ret; } diff --git a/kernel/relay.c b/kernel/relay.c index d6204a485818..dc873fba90d2 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -1162,7 +1162,7 @@ static ssize_t relay_file_splice_read(struct file *in, ret = 0; spliced = 0; - while (len) { + while (len && !spliced) { ret = subbuf_splice_actor(in, ppos, pipe, len, flags, &nonpad_ret); if (ret < 0) break; -- cgit v1.2.3 From 68154e90c9d1492d570671ae181d9a8f8530da55 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Fri, 25 Apr 2008 12:47:50 +0200 Subject: block: add dma alignment and padding support to blk_rq_map_kern This patch adds bio_copy_kern similar to bio_copy_user. blk_rq_map_kern uses bio_copy_kern instead of bio_map_kern if necessary. bio_copy_kern uses temporary pages and the bi_end_io callback frees these pages. bio_copy_kern saves the original kernel buffer at bio->bi_private it doesn't use something like struct bio_map_data to store the information about the caller. Signed-off-by: FUJITA Tomonori Cc: Tejun Heo Signed-off-by: Jens Axboe --- block/blk-map.c | 21 ++++++++++++- fs/bio.c | 90 +++++++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/bio.h | 2 ++ 3 files changed, 112 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/block/blk-map.c b/block/blk-map.c index 3c942bd6422a..0b1af5a3537c 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -255,10 +255,18 @@ EXPORT_SYMBOL(blk_rq_unmap_user); * @kbuf: the kernel buffer * @len: length of user data * @gfp_mask: memory allocation flags + * + * Description: + * Data will be mapped directly if possible. Otherwise a bounce + * buffer is used. */ int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf, unsigned int len, gfp_t gfp_mask) { + unsigned long kaddr; + unsigned int alignment; + int reading = rq_data_dir(rq) == READ; + int do_copy = 0; struct bio *bio; if (len > (q->max_hw_sectors << 9)) @@ -266,13 +274,24 @@ int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf, if (!len || !kbuf) return -EINVAL; - bio = bio_map_kern(q, kbuf, len, gfp_mask); + kaddr = (unsigned long)kbuf; + alignment = queue_dma_alignment(q) | q->dma_pad_mask; + do_copy = ((kaddr & alignment) || (len & alignment)); + + if (do_copy) + bio = bio_copy_kern(q, kbuf, len, gfp_mask, reading); + else + bio = bio_map_kern(q, kbuf, len, gfp_mask); + if (IS_ERR(bio)) return PTR_ERR(bio); if (rq_data_dir(rq) == WRITE) bio->bi_rw |= (1 << BIO_RW); + if (do_copy) + rq->cmd_flags |= REQ_COPY_USER; + blk_rq_bio_prep(q, rq, bio); blk_queue_bounce(q, &rq->bio); rq->buffer = rq->data = NULL; diff --git a/fs/bio.c b/fs/bio.c index 6e0b6f66df03..799f86deff24 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -937,6 +937,95 @@ struct bio *bio_map_kern(struct request_queue *q, void *data, unsigned int len, return ERR_PTR(-EINVAL); } +static void bio_copy_kern_endio(struct bio *bio, int err) +{ + struct bio_vec *bvec; + const int read = bio_data_dir(bio) == READ; + char *p = bio->bi_private; + int i; + + __bio_for_each_segment(bvec, bio, i, 0) { + char *addr = page_address(bvec->bv_page); + + if (read && !err) + memcpy(p, addr, bvec->bv_len); + + __free_page(bvec->bv_page); + p += bvec->bv_len; + } + + bio_put(bio); +} + +/** + * bio_copy_kern - copy kernel address into bio + * @q: the struct request_queue for the bio + * @data: pointer to buffer to copy + * @len: length in bytes + * @gfp_mask: allocation flags for bio and page allocation + * + * copy the kernel address into a bio suitable for io to a block + * device. Returns an error pointer in case of error. + */ +struct bio *bio_copy_kern(struct request_queue *q, void *data, unsigned int len, + gfp_t gfp_mask, int reading) +{ + unsigned long kaddr = (unsigned long)data; + unsigned long end = (kaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT; + unsigned long start = kaddr >> PAGE_SHIFT; + const int nr_pages = end - start; + struct bio *bio; + struct bio_vec *bvec; + int i, ret; + + bio = bio_alloc(gfp_mask, nr_pages); + if (!bio) + return ERR_PTR(-ENOMEM); + + while (len) { + struct page *page; + unsigned int bytes = PAGE_SIZE; + + if (bytes > len) + bytes = len; + + page = alloc_page(q->bounce_gfp | gfp_mask); + if (!page) { + ret = -ENOMEM; + goto cleanup; + } + + if (bio_add_pc_page(q, bio, page, bytes, 0) < bytes) { + ret = -EINVAL; + goto cleanup; + } + + len -= bytes; + } + + if (!reading) { + void *p = data; + + bio_for_each_segment(bvec, bio, i) { + char *addr = page_address(bvec->bv_page); + + memcpy(addr, p, bvec->bv_len); + p += bvec->bv_len; + } + } + + bio->bi_private = data; + bio->bi_end_io = bio_copy_kern_endio; + return bio; +cleanup: + bio_for_each_segment(bvec, bio, i) + __free_page(bvec->bv_page); + + bio_put(bio); + + return ERR_PTR(ret); +} + /* * bio_set_pages_dirty() and bio_check_pages_dirty() are support functions * for performing direct-IO in BIOs. @@ -1273,6 +1362,7 @@ EXPORT_SYMBOL(bio_get_nr_vecs); EXPORT_SYMBOL(bio_map_user); EXPORT_SYMBOL(bio_unmap_user); EXPORT_SYMBOL(bio_map_kern); +EXPORT_SYMBOL(bio_copy_kern); EXPORT_SYMBOL(bio_pair_release); EXPORT_SYMBOL(bio_split); EXPORT_SYMBOL(bio_split_pool); diff --git a/include/linux/bio.h b/include/linux/bio.h index d259690863fb..61c15eaf3fb3 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -324,6 +324,8 @@ extern struct bio *bio_map_user_iov(struct request_queue *, extern void bio_unmap_user(struct bio *); extern struct bio *bio_map_kern(struct request_queue *, void *, unsigned int, gfp_t); +extern struct bio *bio_copy_kern(struct request_queue *, void *, unsigned int, + gfp_t, int); extern void bio_set_pages_dirty(struct bio *bio); extern void bio_check_pages_dirty(struct bio *bio); extern struct bio *bio_copy_user(struct request_queue *, unsigned long, unsigned int, int); -- cgit v1.2.3 From cdac75e6f2fec9abc21d0abb4e5d80720eeebb10 Mon Sep 17 00:00:00 2001 From: Davide Libenzi Date: Tue, 29 Apr 2008 00:58:34 -0700 Subject: epoll: avoid kmemcheck warning Epoll calls rb_set_parent(n, n) to initialize the rb-tree node, but rb_set_parent() accesses node's pointer in its code. This creates a warning in kmemcheck (reported by Vegard Nossum) about an uninitialized memory access. The warning is harmless since the following rb-tree node insert is going to overwrite the node data. In any case I think it's better to not have that happening at all, and fix it by simplifying the code to get rid of a few lines that became superfluous after the previous epoll changes. Signed-off-by: Davide Libenzi Cc: Vegard Nossum Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/eventpoll.c | 27 +++------------------------ 1 file changed, 3 insertions(+), 24 deletions(-) (limited to 'fs') diff --git a/fs/eventpoll.c b/fs/eventpoll.c index a415f42d32cf..0d237182d721 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -257,25 +257,6 @@ static inline int ep_cmp_ffd(struct epoll_filefd *p1, (p1->file < p2->file ? -1 : p1->fd - p2->fd)); } -/* Special initialization for the RB tree node to detect linkage */ -static inline void ep_rb_initnode(struct rb_node *n) -{ - rb_set_parent(n, n); -} - -/* Removes a node from the RB tree and marks it for a fast is-linked check */ -static inline void ep_rb_erase(struct rb_node *n, struct rb_root *r) -{ - rb_erase(n, r); - rb_set_parent(n, n); -} - -/* Fast check to verify that the item is linked to the main RB tree */ -static inline int ep_rb_linked(struct rb_node *n) -{ - return rb_parent(n) != n; -} - /* Tells us if the item is currently linked */ static inline int ep_is_linked(struct list_head *p) { @@ -283,13 +264,13 @@ static inline int ep_is_linked(struct list_head *p) } /* Get the "struct epitem" from a wait queue pointer */ -static inline struct epitem * ep_item_from_wait(wait_queue_t *p) +static inline struct epitem *ep_item_from_wait(wait_queue_t *p) { return container_of(p, struct eppoll_entry, wait)->base; } /* Get the "struct epitem" from an epoll queue wrapper */ -static inline struct epitem * ep_item_from_epqueue(poll_table *p) +static inline struct epitem *ep_item_from_epqueue(poll_table *p) { return container_of(p, struct ep_pqueue, pt)->epi; } @@ -411,8 +392,7 @@ static int ep_remove(struct eventpoll *ep, struct epitem *epi) list_del_init(&epi->fllink); spin_unlock(&file->f_ep_lock); - if (ep_rb_linked(&epi->rbn)) - ep_rb_erase(&epi->rbn, &ep->rbr); + rb_erase(&epi->rbn, &ep->rbr); spin_lock_irqsave(&ep->lock, flags); if (ep_is_linked(&epi->rdllink)) @@ -728,7 +708,6 @@ static int ep_insert(struct eventpoll *ep, struct epoll_event *event, goto error_return; /* Item initialization follow here ... */ - ep_rb_initnode(&epi->rbn); INIT_LIST_HEAD(&epi->rdllink); INIT_LIST_HEAD(&epi->fllink); INIT_LIST_HEAD(&epi->pwqlist); -- cgit v1.2.3 From e5949050f2610fa526b154e0d8379218e54f49d1 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Tue, 29 Apr 2008 00:58:41 -0700 Subject: adfs: work around bogus sparse warning fs/adfs/dir_f.c:126:4: warning: do-while statement is not a compound statement Signed-off-by: Harvey Harrison Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/adfs/dir_f.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/adfs/dir_f.c b/fs/adfs/dir_f.c index b9b2b27b68c3..ea7df2146921 100644 --- a/fs/adfs/dir_f.c +++ b/fs/adfs/dir_f.c @@ -122,9 +122,9 @@ adfs_dir_checkbyte(const struct adfs_dir *dir) ptr.ptr8 = bufoff(bh, i); end.ptr8 = ptr.ptr8 + last - i; - do + do { dircheck = *ptr.ptr8++ ^ ror13(dircheck); - while (ptr.ptr8 < end.ptr8); + } while (ptr.ptr8 < end.ptr8); } /* -- cgit v1.2.3 From 9fe76c763f0e18582bcb670c386978e83a755d05 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Tue, 29 Apr 2008 00:58:44 -0700 Subject: coda: add static to functions in dir.c coda_unlink, coda_rmdir, coda_readdir can all be static, the forward declarations already were. Signed-off-by: Harvey Harrison Cc: Jan Harkes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/coda/dir.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/coda/dir.c b/fs/coda/dir.c index f89ff083079b..3d2580e00a3e 100644 --- a/fs/coda/dir.c +++ b/fs/coda/dir.c @@ -345,7 +345,7 @@ static int coda_symlink(struct inode *dir_inode, struct dentry *de, } /* destruction routines: unlink, rmdir */ -int coda_unlink(struct inode *dir, struct dentry *de) +static int coda_unlink(struct inode *dir, struct dentry *de) { int error; const char *name = de->d_name.name; @@ -365,7 +365,7 @@ int coda_unlink(struct inode *dir, struct dentry *de) return 0; } -int coda_rmdir(struct inode *dir, struct dentry *de) +static int coda_rmdir(struct inode *dir, struct dentry *de) { const char *name = de->d_name.name; int len = de->d_name.len; @@ -424,7 +424,7 @@ static int coda_rename(struct inode *old_dir, struct dentry *old_dentry, /* file operations for directories */ -int coda_readdir(struct file *coda_file, void *buf, filldir_t filldir) +static int coda_readdir(struct file *coda_file, void *buf, filldir_t filldir) { struct coda_file_info *cfi; struct file *host_file; -- cgit v1.2.3 From 63e3453e547b20321381b212cb1ee11537dc843d Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Tue, 29 Apr 2008 00:58:44 -0700 Subject: befs: fix sparse warning in linuxvfs.c Use link as the variable name to avoid shadowing the arg. fs/befs/linuxvfs.c:492:8: warning: symbol 'p' shadows an earlier one fs/befs/linuxvfs.c:488:77: originally declared here Signed-off-by: Harvey Harrison Cc: "Sergey S. Kostyliov" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/befs/linuxvfs.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index 82123ff3e1dd..e8717de3bab3 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -489,9 +489,9 @@ static void befs_put_link(struct dentry *dentry, struct nameidata *nd, void *p) { befs_inode_info *befs_ino = BEFS_I(dentry->d_inode); if (befs_ino->i_flags & BEFS_LONG_SYMLINK) { - char *p = nd_get_link(nd); - if (!IS_ERR(p)) - kfree(p); + char *link = nd_get_link(nd); + if (!IS_ERR(link)) + kfree(link); } } -- cgit v1.2.3 From 4488c59c942bd6004fc97f0c2a7603a2f5dd80e0 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 29 Apr 2008 00:58:51 -0700 Subject: fs/ramfs/ extern cleanup - internal.h shouldn't duplicate the extern declaration for ramfs_file_operations already in include/linux/ramfs.h - file-mmu.c needs two #include's for seeing the extern declarations of it's global struct's Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ramfs/file-mmu.c | 3 +++ fs/ramfs/internal.h | 1 - 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ramfs/file-mmu.c b/fs/ramfs/file-mmu.c index b41a514b0976..9590b9024300 100644 --- a/fs/ramfs/file-mmu.c +++ b/fs/ramfs/file-mmu.c @@ -26,6 +26,9 @@ #include #include +#include + +#include "internal.h" const struct address_space_operations ramfs_aops = { .readpage = simple_readpage, diff --git a/fs/ramfs/internal.h b/fs/ramfs/internal.h index af7cc074a476..6b330639b51d 100644 --- a/fs/ramfs/internal.h +++ b/fs/ramfs/internal.h @@ -11,5 +11,4 @@ extern const struct address_space_operations ramfs_aops; -extern const struct file_operations ramfs_file_operations; extern const struct inode_operations ramfs_file_inode_operations; -- cgit v1.2.3 From 4b0a8da7a7bbe7f84c7bd16a5e965a129f461881 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 29 Apr 2008 00:58:52 -0700 Subject: fs/hfsplus/: proper externs Add proper extern declarations for two structs in fs/hfsplus/hfsplus_fs.h Signed-off-by: Adrian Bunk Cc: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hfsplus/hfsplus_fs.h | 4 ++++ fs/hfsplus/inode.c | 3 --- 2 files changed, 4 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h index d72d0a8b25aa..9e59537b43d5 100644 --- a/fs/hfsplus/hfsplus_fs.h +++ b/fs/hfsplus/hfsplus_fs.h @@ -311,6 +311,10 @@ int hfsplus_delete_cat(u32, struct inode *, struct qstr *); int hfsplus_rename_cat(u32, struct inode *, struct qstr *, struct inode *, struct qstr *); +/* dir.c */ +extern const struct inode_operations hfsplus_dir_inode_operations; +extern const struct file_operations hfsplus_dir_operations; + /* extents.c */ int hfsplus_ext_cmp_key(const hfsplus_btree_key *, const hfsplus_btree_key *); void hfsplus_ext_write_extent(struct inode *); diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index 37744cf3706a..d53b2af91c25 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -278,9 +278,6 @@ static int hfsplus_file_release(struct inode *inode, struct file *file) return 0; } -extern const struct inode_operations hfsplus_dir_inode_operations; -extern struct file_operations hfsplus_dir_operations; - static const struct inode_operations hfsplus_file_inode_operations = { .lookup = hfsplus_file_lookup, .truncate = hfsplus_file_truncate, -- cgit v1.2.3 From 8b1919a1e8b8968e0ac9030a4f14f0d2cd69e7cf Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 29 Apr 2008 00:58:54 -0700 Subject: fs/freevxfs/: proper externs Move the extern declarations of several structs to vxfs_extern.h Signed-off-by: Adrian Bunk Acked-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/freevxfs/vxfs_extern.h | 5 +++++ fs/freevxfs/vxfs_immed.c | 1 + fs/freevxfs/vxfs_inode.c | 5 ----- 3 files changed, 6 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/freevxfs/vxfs_extern.h b/fs/freevxfs/vxfs_extern.h index 2b46064f66b2..50ab5eecb99b 100644 --- a/fs/freevxfs/vxfs_extern.h +++ b/fs/freevxfs/vxfs_extern.h @@ -50,7 +50,11 @@ extern daddr_t vxfs_bmap1(struct inode *, long); /* vxfs_fshead.c */ extern int vxfs_read_fshead(struct super_block *); +/* vxfs_immed.c */ +extern const struct inode_operations vxfs_immed_symlink_iops; + /* vxfs_inode.c */ +extern const struct address_space_operations vxfs_immed_aops; extern struct kmem_cache *vxfs_inode_cachep; extern void vxfs_dumpi(struct vxfs_inode_info *, ino_t); extern struct inode * vxfs_get_fake_inode(struct super_block *, @@ -69,6 +73,7 @@ extern const struct file_operations vxfs_dir_operations; extern int vxfs_read_olt(struct super_block *, u_long); /* vxfs_subr.c */ +extern const struct address_space_operations vxfs_aops; extern struct page * vxfs_get_page(struct address_space *, u_long); extern void vxfs_put_page(struct page *); extern struct buffer_head * vxfs_bread(struct inode *, int); diff --git a/fs/freevxfs/vxfs_immed.c b/fs/freevxfs/vxfs_immed.c index 8a5959a61ba9..c36aeaf92e41 100644 --- a/fs/freevxfs/vxfs_immed.c +++ b/fs/freevxfs/vxfs_immed.c @@ -35,6 +35,7 @@ #include #include "vxfs.h" +#include "vxfs_extern.h" #include "vxfs_inode.h" diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index ad88d2364bc2..9f3f2ceb73f0 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -41,11 +41,6 @@ #include "vxfs_extern.h" -extern const struct address_space_operations vxfs_aops; -extern const struct address_space_operations vxfs_immed_aops; - -extern const struct inode_operations vxfs_immed_symlink_iops; - struct kmem_cache *vxfs_inode_cachep; -- cgit v1.2.3 From 6b09ae66922ca198e5830c0a4d74400a507a9170 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 29 Apr 2008 00:58:54 -0700 Subject: make __put_super() static Make the needlessly global __put_super() static. Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/super.c | 2 +- include/linux/fs.h | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/super.c b/fs/super.c index a5a4aca7e22f..453877c5697b 100644 --- a/fs/super.c +++ b/fs/super.c @@ -117,7 +117,7 @@ static inline void destroy_super(struct super_block *s) * Drop a superblock's refcount. Returns non-zero if the superblock was * destroyed. The caller must hold sb_lock. */ -int __put_super(struct super_block *sb) +static int __put_super(struct super_block *sb) { int ret = 0; diff --git a/include/linux/fs.h b/include/linux/fs.h index 2c925747bc49..1de9d72178e1 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1521,7 +1521,6 @@ extern int get_sb_pseudo(struct file_system_type *, char *, const struct super_operations *ops, unsigned long, struct vfsmount *mnt); extern int simple_set_mnt(struct vfsmount *mnt, struct super_block *sb); -int __put_super(struct super_block *sb); int __put_super_and_need_restart(struct super_block *sb); void unnamed_dev_init(void); -- cgit v1.2.3 From 67cde595374dd0e4e4a537dbf9dff70fd3d7bd7b Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 29 Apr 2008 00:58:55 -0700 Subject: make vfs_ioctl() static Make the needlessly global vfs_ioctl() static. Signed-off-by: Adrian Bunk Acked-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ioctl.c | 4 ++-- include/linux/fs.h | 1 - 2 files changed, 2 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/ioctl.c b/fs/ioctl.c index f32fbde2175e..7db32b3382d3 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -28,8 +28,8 @@ * * Returns 0 on success, -errno on error. */ -long vfs_ioctl(struct file *filp, unsigned int cmd, - unsigned long arg) +static long vfs_ioctl(struct file *filp, unsigned int cmd, + unsigned long arg) { int error = -ENOTTY; diff --git a/include/linux/fs.h b/include/linux/fs.h index 1de9d72178e1..a1ba005d08e7 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1964,7 +1964,6 @@ extern int vfs_stat_fd(int dfd, char __user *, struct kstat *); extern int vfs_lstat_fd(int dfd, char __user *, struct kstat *); extern int vfs_fstat(unsigned int, struct kstat *); -extern long vfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); extern int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd, unsigned long arg); -- cgit v1.2.3 From f11b00f3bd89c91c684d56b2082d1b0241ff20ae Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 29 Apr 2008 00:58:56 -0700 Subject: fs/fs-writeback.c: make 2 functions static Make the following needlessly global functions static: - writeback_acquire() - writeback_release() Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fs-writeback.c | 78 ++++++++++++++++++++++----------------------- include/linux/backing-dev.h | 2 -- 2 files changed, 39 insertions(+), 41 deletions(-) (limited to 'fs') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 06557679ca41..ae45f77765c0 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -25,6 +25,45 @@ #include #include "internal.h" + +/** + * writeback_acquire - attempt to get exclusive writeback access to a device + * @bdi: the device's backing_dev_info structure + * + * It is a waste of resources to have more than one pdflush thread blocked on + * a single request queue. Exclusion at the request_queue level is obtained + * via a flag in the request_queue's backing_dev_info.state. + * + * Non-request_queue-backed address_spaces will share default_backing_dev_info, + * unless they implement their own. Which is somewhat inefficient, as this + * may prevent concurrent writeback against multiple devices. + */ +static int writeback_acquire(struct backing_dev_info *bdi) +{ + return !test_and_set_bit(BDI_pdflush, &bdi->state); +} + +/** + * writeback_in_progress - determine whether there is writeback in progress + * @bdi: the device's backing_dev_info structure. + * + * Determine whether there is writeback in progress against a backing device. + */ +int writeback_in_progress(struct backing_dev_info *bdi) +{ + return test_bit(BDI_pdflush, &bdi->state); +} + +/** + * writeback_release - relinquish exclusive writeback access against a device. + * @bdi: the device's backing_dev_info structure + */ +static void writeback_release(struct backing_dev_info *bdi) +{ + BUG_ON(!writeback_in_progress(bdi)); + clear_bit(BDI_pdflush, &bdi->state); +} + /** * __mark_inode_dirty - internal function * @inode: inode to mark @@ -747,43 +786,4 @@ int generic_osync_inode(struct inode *inode, struct address_space *mapping, int return err; } - EXPORT_SYMBOL(generic_osync_inode); - -/** - * writeback_acquire - attempt to get exclusive writeback access to a device - * @bdi: the device's backing_dev_info structure - * - * It is a waste of resources to have more than one pdflush thread blocked on - * a single request queue. Exclusion at the request_queue level is obtained - * via a flag in the request_queue's backing_dev_info.state. - * - * Non-request_queue-backed address_spaces will share default_backing_dev_info, - * unless they implement their own. Which is somewhat inefficient, as this - * may prevent concurrent writeback against multiple devices. - */ -int writeback_acquire(struct backing_dev_info *bdi) -{ - return !test_and_set_bit(BDI_pdflush, &bdi->state); -} - -/** - * writeback_in_progress - determine whether there is writeback in progress - * @bdi: the device's backing_dev_info structure. - * - * Determine whether there is writeback in progress against a backing device. - */ -int writeback_in_progress(struct backing_dev_info *bdi) -{ - return test_bit(BDI_pdflush, &bdi->state); -} - -/** - * writeback_release - relinquish exclusive writeback access against a device. - * @bdi: the device's backing_dev_info structure - */ -void writeback_release(struct backing_dev_info *bdi) -{ - BUG_ON(!writeback_in_progress(bdi)); - clear_bit(BDI_pdflush, &bdi->state); -} diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 48a62baace58..b66fa2bdfd9c 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -156,9 +156,7 @@ static inline unsigned long bdi_stat_error(struct backing_dev_info *bdi) extern struct backing_dev_info default_backing_dev_info; void default_unplug_io_fn(struct backing_dev_info *bdi, struct page *page); -int writeback_acquire(struct backing_dev_info *bdi); int writeback_in_progress(struct backing_dev_info *bdi); -void writeback_release(struct backing_dev_info *bdi); static inline int bdi_congested(struct backing_dev_info *bdi, int bdi_bits) { -- cgit v1.2.3 From 07d45da616f8514651360b502314fc9554223a03 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 29 Apr 2008 00:58:57 -0700 Subject: fs/drop_caches.c: make 2 functions static Make the following needlessly global functions static: - drop_pagecache() - drop_slab() Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/drop_caches.c | 4 ++-- include/linux/mm.h | 2 -- 2 files changed, 2 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/drop_caches.c b/fs/drop_caches.c index 59375efcf39d..e2c6b6500c78 100644 --- a/fs/drop_caches.c +++ b/fs/drop_caches.c @@ -25,7 +25,7 @@ static void drop_pagecache_sb(struct super_block *sb) spin_unlock(&inode_lock); } -void drop_pagecache(void) +static void drop_pagecache(void) { struct super_block *sb; @@ -45,7 +45,7 @@ restart: spin_unlock(&sb_lock); } -void drop_slab(void) +static void drop_slab(void) { int nr_objects; diff --git a/include/linux/mm.h b/include/linux/mm.h index 8b7f4a5d4f6a..fef602d82722 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1230,8 +1230,6 @@ int drop_caches_sysctl_handler(struct ctl_table *, int, struct file *, void __user *, size_t *, loff_t *); unsigned long shrink_slab(unsigned long scanned, gfp_t gfp_mask, unsigned long lru_pages); -void drop_pagecache(void); -void drop_slab(void); #ifndef CONFIG_MMU #define randomize_va_space 0 -- cgit v1.2.3 From d5470b596abdd566339b2417e807b1198be64b97 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 29 Apr 2008 00:58:57 -0700 Subject: fs/aio.c: make 3 functions static Make the following needlessly global functions static: - __put_ioctx() - lookup_ioctx() - io_submit_one() Signed-off-by: Adrian Bunk Cc: Zach Brown Cc: Benjamin LaHaise Cc: Badari Pulavarty Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/aio.c | 67 +++++++++++++++++++++++++++++++---------------------- include/linux/aio.h | 19 --------------- 2 files changed, 39 insertions(+), 47 deletions(-) (limited to 'fs') diff --git a/fs/aio.c b/fs/aio.c index ae94e1dea266..81c01290939b 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -191,6 +191,43 @@ static int aio_setup_ring(struct kioctx *ctx) kunmap_atomic((void *)((unsigned long)__event & PAGE_MASK), km); \ } while(0) + +/* __put_ioctx + * Called when the last user of an aio context has gone away, + * and the struct needs to be freed. + */ +static void __put_ioctx(struct kioctx *ctx) +{ + unsigned nr_events = ctx->max_reqs; + + BUG_ON(ctx->reqs_active); + + cancel_delayed_work(&ctx->wq); + cancel_work_sync(&ctx->wq.work); + aio_free_ring(ctx); + mmdrop(ctx->mm); + ctx->mm = NULL; + pr_debug("__put_ioctx: freeing %p\n", ctx); + kmem_cache_free(kioctx_cachep, ctx); + + if (nr_events) { + spin_lock(&aio_nr_lock); + BUG_ON(aio_nr - nr_events > aio_nr); + aio_nr -= nr_events; + spin_unlock(&aio_nr_lock); + } +} + +#define get_ioctx(kioctx) do { \ + BUG_ON(atomic_read(&(kioctx)->users) <= 0); \ + atomic_inc(&(kioctx)->users); \ +} while (0) +#define put_ioctx(kioctx) do { \ + BUG_ON(atomic_read(&(kioctx)->users) <= 0); \ + if (unlikely(atomic_dec_and_test(&(kioctx)->users))) \ + __put_ioctx(kioctx); \ +} while (0) + /* ioctx_alloc * Allocates and initializes an ioctx. Returns an ERR_PTR if it failed. */ @@ -361,32 +398,6 @@ void exit_aio(struct mm_struct *mm) } } -/* __put_ioctx - * Called when the last user of an aio context has gone away, - * and the struct needs to be freed. - */ -void __put_ioctx(struct kioctx *ctx) -{ - unsigned nr_events = ctx->max_reqs; - - BUG_ON(ctx->reqs_active); - - cancel_delayed_work(&ctx->wq); - cancel_work_sync(&ctx->wq.work); - aio_free_ring(ctx); - mmdrop(ctx->mm); - ctx->mm = NULL; - pr_debug("__put_ioctx: freeing %p\n", ctx); - kmem_cache_free(kioctx_cachep, ctx); - - if (nr_events) { - spin_lock(&aio_nr_lock); - BUG_ON(aio_nr - nr_events > aio_nr); - aio_nr -= nr_events; - spin_unlock(&aio_nr_lock); - } -} - /* aio_get_req * Allocate a slot for an aio request. Increments the users count * of the kioctx so that the kioctx stays around until all requests are @@ -545,7 +556,7 @@ int aio_put_req(struct kiocb *req) /* Lookup an ioctx id. ioctx_list is lockless for reads. * FIXME: this is O(n) and is only suitable for development. */ -struct kioctx *lookup_ioctx(unsigned long ctx_id) +static struct kioctx *lookup_ioctx(unsigned long ctx_id) { struct kioctx *ioctx; struct mm_struct *mm; @@ -1552,7 +1563,7 @@ static int aio_wake_function(wait_queue_t *wait, unsigned mode, return 1; } -int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, +static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, struct iocb *iocb) { struct kiocb *req; diff --git a/include/linux/aio.h b/include/linux/aio.h index 0d0b7f629bd3..b51ddd28444e 100644 --- a/include/linux/aio.h +++ b/include/linux/aio.h @@ -209,27 +209,8 @@ extern ssize_t wait_on_sync_kiocb(struct kiocb *iocb); extern int aio_put_req(struct kiocb *iocb); extern void kick_iocb(struct kiocb *iocb); extern int aio_complete(struct kiocb *iocb, long res, long res2); -extern void __put_ioctx(struct kioctx *ctx); struct mm_struct; extern void exit_aio(struct mm_struct *mm); -extern struct kioctx *lookup_ioctx(unsigned long ctx_id); -extern int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, - struct iocb *iocb); - -/* semi private, but used by the 32bit emulations: */ -struct kioctx *lookup_ioctx(unsigned long ctx_id); -int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, - struct iocb *iocb); - -#define get_ioctx(kioctx) do { \ - BUG_ON(atomic_read(&(kioctx)->users) <= 0); \ - atomic_inc(&(kioctx)->users); \ -} while (0) -#define put_ioctx(kioctx) do { \ - BUG_ON(atomic_read(&(kioctx)->users) <= 0); \ - if (unlikely(atomic_dec_and_test(&(kioctx)->users))) \ - __put_ioctx(kioctx); \ -} while (0) #define io_wait_to_kiocb(wait) container_of(wait, struct kiocb, ki_wait) -- cgit v1.2.3 From 45cc2b96f20fa27088a650587e5d9dc5fa5e32c0 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 29 Apr 2008 00:58:59 -0700 Subject: fs/timerfd.c should #include Every file should include the headers containing the prototypes for its global functions (in this case for sys_timerfd_*()). Signed-off-by: Adrian Bunk Cc: Davide Libenzi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/timerfd.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/timerfd.c b/fs/timerfd.c index 10c80b59ec4b..5400524e9cb1 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -20,6 +20,7 @@ #include #include #include +#include struct timerfd_ctx { struct hrtimer tmr; -- cgit v1.2.3 From 946a57b526a16e5662235cb8f573337bc8ecdc48 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 29 Apr 2008 00:59:00 -0700 Subject: remove generic_commit_write() Remove the obsolete and no longer used generic_commit_write(). Signed-off-by: Adrian Bunk Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 18 ------------------ include/linux/buffer_head.h | 1 - 2 files changed, 19 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 3db4a26adc44..22ed55198f3d 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2328,23 +2328,6 @@ int block_commit_write(struct page *page, unsigned from, unsigned to) return 0; } -int generic_commit_write(struct file *file, struct page *page, - unsigned from, unsigned to) -{ - struct inode *inode = page->mapping->host; - loff_t pos = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; - __block_commit_write(inode,page,from,to); - /* - * No need to use i_size_read() here, the i_size - * cannot change under us because we hold i_mutex. - */ - if (pos > inode->i_size) { - i_size_write(inode, pos); - mark_inode_dirty(inode); - } - return 0; -} - /* * block_page_mkwrite() is not allowed to change the file size as it gets * called from a page fault handler when a page is first dirtied. Hence we must @@ -3315,7 +3298,6 @@ EXPORT_SYMBOL(end_buffer_write_sync); EXPORT_SYMBOL(file_fsync); EXPORT_SYMBOL(fsync_bdev); EXPORT_SYMBOL(generic_block_bmap); -EXPORT_SYMBOL(generic_commit_write); EXPORT_SYMBOL(generic_cont_expand_simple); EXPORT_SYMBOL(init_buffer); EXPORT_SYMBOL(invalidate_bdev); diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index 932eb02a2753..82aa36c53ea7 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -225,7 +225,6 @@ int block_page_mkwrite(struct vm_area_struct *vma, struct page *page, get_block_t get_block); void block_sync_page(struct page *); sector_t generic_block_bmap(struct address_space *, sector_t, get_block_t *); -int generic_commit_write(struct file *, struct page *, unsigned, unsigned); int block_truncate_page(struct address_space *, loff_t, get_block_t *); int file_fsync(struct file *, struct dentry *, int); int nobh_write_begin(struct file *, struct address_space *, -- cgit v1.2.3 From f1e3af72c10ba74fb15864c354515ec1bd8bf2a5 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 29 Apr 2008 00:59:01 -0700 Subject: make fs/buffer.c:cont_expand_zero() static cont_expand_zero() can become static. Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 22ed55198f3d..189efa4efc6e 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2211,8 +2211,8 @@ out: return err; } -int cont_expand_zero(struct file *file, struct address_space *mapping, - loff_t pos, loff_t *bytes) +static int cont_expand_zero(struct file *file, struct address_space *mapping, + loff_t pos, loff_t *bytes) { struct inode *inode = mapping->host; unsigned blocksize = 1 << inode->i_blkbits; -- cgit v1.2.3 From 3202e1811fd312f3f32ddc8f526aa2691b64ec55 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 29 Apr 2008 00:59:02 -0700 Subject: make BINFMT_FLAT a bool I have not yet seen anyone saying he has a reasonable use case for using BINFMT_FLAT modular on his embedded device. Considering that fs/binfmt_flat.c even lacks a MODULE_LICENSE() I really doubt there is any, and this patch therefore makes BINFMT_FLAT a bool. Signed-off-by: Adrian Bunk Acked-by: Bryan Wu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/Kconfig.binfmt | 2 +- fs/binfmt_flat.c | 6 ------ 2 files changed, 1 insertion(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/Kconfig.binfmt b/fs/Kconfig.binfmt index 853845abcca6..55e8ee1900a5 100644 --- a/fs/Kconfig.binfmt +++ b/fs/Kconfig.binfmt @@ -41,7 +41,7 @@ config BINFMT_ELF_FDPIC It is also possible to run FDPIC ELF binaries on MMU linux also. config BINFMT_FLAT - tristate "Kernel support for flat binaries" + bool "Kernel support for flat binaries" depends on !MMU help Support uClinux FLAT format binaries. diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 0498b181dd52..c12cc362fd3b 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -932,14 +932,8 @@ static int __init init_flat_binfmt(void) return register_binfmt(&flat_format); } -static void __exit exit_flat_binfmt(void) -{ - unregister_binfmt(&flat_format); -} - /****************************************************************************/ core_initcall(init_flat_binfmt); -module_exit(exit_flat_binfmt); /****************************************************************************/ -- cgit v1.2.3 From f249fdd8c19ff65825c0be67212cdf22e556668e Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Tue, 29 Apr 2008 00:59:03 -0700 Subject: autofs4: fix sparse warning in root.c fs/autofs4/root.c:536:23: warning: symbol 'ino' shadows an earlier one fs/autofs4/root.c:510:22: originally declared here There is no need to redeclare, we are at the end of the loop and in the next iteration of the loop, ino will be reset. Signed-off-by: Harvey Harrison Acked-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/root.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index a54a946a50ae..aa4c5ff8a40d 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -533,9 +533,9 @@ static struct dentry *autofs4_lookup_unhashed(struct autofs_sb_info *sbi, struct goto next; if (d_unhashed(dentry)) { - struct autofs_info *ino = autofs4_dentry_ino(dentry); struct inode *inode = dentry->d_inode; + ino = autofs4_dentry_ino(dentry); list_del_init(&ino->rehash); dget(dentry); /* -- cgit v1.2.3 From 61d64576a21275114d6bffff3c1cac6c8e2f7cf2 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Tue, 29 Apr 2008 00:59:05 -0700 Subject: fs: remove unused fops from struct char_device_struct struct char_device_struct::fops is no longer used: remove it. Signed-off-by: Jiri Olsa Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/char_dev.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/char_dev.c b/fs/char_dev.c index 038674aa88a7..68e510b88457 100644 --- a/fs/char_dev.c +++ b/fs/char_dev.c @@ -55,7 +55,6 @@ static struct char_device_struct { unsigned int baseminor; int minorct; char name[64]; - struct file_operations *fops; struct cdev *cdev; /* will die */ } *chrdevs[CHRDEV_MAJOR_HASH_SIZE]; -- cgit v1.2.3 From 6db27dd9d26fb270adaa4c265df65ccb49638bd0 Mon Sep 17 00:00:00 2001 From: Jim Meyering Date: Tue, 29 Apr 2008 00:59:06 -0700 Subject: affs: handle match_strdup failure fs/affs/super.c (parse_options): Remove useless initialization. Handle match_strdup failure. Signed-off-by: Jim Meyering Cc: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/affs/super.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/affs/super.c b/fs/affs/super.c index d2dc047cb479..01d25d532541 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -199,7 +199,6 @@ parse_options(char *options, uid_t *uid, gid_t *gid, int *mode, int *reserved, s case Opt_prefix: /* Free any previous prefix */ kfree(*prefix); - *prefix = NULL; *prefix = match_strdup(&args[0]); if (!*prefix) return 0; @@ -233,6 +232,8 @@ parse_options(char *options, uid_t *uid, gid_t *gid, int *mode, int *reserved, s break; case Opt_volume: { char *vol = match_strdup(&args[0]); + if (!vol) + return 0; strlcpy(volume, vol, 32); kfree(vol); break; -- cgit v1.2.3 From 3fbe5c31009d26c7b6b73d5c69fe930a5e9d2e26 Mon Sep 17 00:00:00 2001 From: Jim Meyering Date: Tue, 29 Apr 2008 00:59:07 -0700 Subject: hfs: handle match_strdup failure fs/hfs/super.c (parse_options): Handle match_strdup failure, twice. Signed-off-by: Jim Meyering Cc: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hfs/super.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/hfs/super.c b/fs/hfs/super.c index 32de44ed0021..8cf67974adf6 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -297,7 +297,8 @@ static int parse_options(char *options, struct hfs_sb_info *hsb) return 0; } p = match_strdup(&args[0]); - hsb->nls_disk = load_nls(p); + if (p) + hsb->nls_disk = load_nls(p); if (!hsb->nls_disk) { printk(KERN_ERR "hfs: unable to load codepage \"%s\"\n", p); kfree(p); @@ -311,7 +312,8 @@ static int parse_options(char *options, struct hfs_sb_info *hsb) return 0; } p = match_strdup(&args[0]); - hsb->nls_io = load_nls(p); + if (p) + hsb->nls_io = load_nls(p); if (!hsb->nls_io) { printk(KERN_ERR "hfs: unable to load iocharset \"%s\"\n", p); kfree(p); -- cgit v1.2.3 From cd6fda36089cf3b450821228c2f575a3b5d0e7a7 Mon Sep 17 00:00:00 2001 From: Jim Meyering Date: Tue, 29 Apr 2008 00:59:08 -0700 Subject: hfsplus: handle match_strdup failure fs/hfsplus/options.c (hfsplus_parse_options): Handle match_strdup failure. Signed-off-by: Jim Meyering Cc: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hfsplus/options.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/hfsplus/options.c b/fs/hfsplus/options.c index dc64fac00831..9997cbf8beb5 100644 --- a/fs/hfsplus/options.c +++ b/fs/hfsplus/options.c @@ -132,7 +132,8 @@ int hfsplus_parse_options(char *input, struct hfsplus_sb_info *sbi) return 0; } p = match_strdup(&args[0]); - sbi->nls = load_nls(p); + if (p) + sbi->nls = load_nls(p); if (!sbi->nls) { printk(KERN_ERR "hfs: unable to load nls mapping \"%s\"\n", p); kfree(p); -- cgit v1.2.3 From 8d4b69002e56e93f1cfe8bb863846ecde3990032 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Tue, 29 Apr 2008 00:59:12 -0700 Subject: fs/affs/file.c: use BUG_ON if (...) BUG(); should be replaced with BUG_ON(...) when the test has no side-effects to allow a definition of BUG_ON that drops the code completely. The semantic patch that makes this change is as follows: (http://www.emn.fr/x-info/coccinelle/) // @ disable unlikely @ expression E,f; @@ ( if (<... f(...) ...>) { BUG(); } | - if (unlikely(E)) { BUG(); } + BUG_ON(E); ) @@ expression E,f; @@ ( if (<... f(...) ...>) { BUG(); } | - if (E) { BUG(); } + BUG_ON(E); ) // Signed-off-by: Julia Lawall Cc: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/affs/file.c | 21 +++++++-------------- 1 file changed, 7 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/affs/file.c b/fs/affs/file.c index 6e0c9399200e..e87ede608f77 100644 --- a/fs/affs/file.c +++ b/fs/affs/file.c @@ -325,8 +325,7 @@ affs_get_block(struct inode *inode, sector_t block, struct buffer_head *bh_resul pr_debug("AFFS: get_block(%u, %lu)\n", (u32)inode->i_ino, (unsigned long)block); - if (block > (sector_t)0x7fffffffUL) - BUG(); + BUG_ON(block > (sector_t)0x7fffffffUL); if (block >= AFFS_I(inode)->i_blkcnt) { if (block > AFFS_I(inode)->i_blkcnt || !create) @@ -493,8 +492,7 @@ affs_do_readpage_ofs(struct file *file, struct page *page, unsigned from, unsign u32 tmp; pr_debug("AFFS: read_page(%u, %ld, %d, %d)\n", (u32)inode->i_ino, page->index, from, to); - if (from > to || to > PAGE_CACHE_SIZE) - BUG(); + BUG_ON(from > to || to > PAGE_CACHE_SIZE); kmap(page); data = page_address(page); bsize = AFFS_SB(sb)->s_data_blksize; @@ -507,8 +505,7 @@ affs_do_readpage_ofs(struct file *file, struct page *page, unsigned from, unsign if (IS_ERR(bh)) return PTR_ERR(bh); tmp = min(bsize - boff, to - from); - if (from + tmp > to || tmp > bsize) - BUG(); + BUG_ON(from + tmp > to || tmp > bsize); memcpy(data + from, AFFS_DATA(bh) + boff, tmp); affs_brelse(bh); bidx++; @@ -540,8 +537,7 @@ affs_extent_file_ofs(struct inode *inode, u32 newsize) if (IS_ERR(bh)) return PTR_ERR(bh); tmp = min(bsize - boff, newsize - size); - if (boff + tmp > bsize || tmp > bsize) - BUG(); + BUG_ON(boff + tmp > bsize || tmp > bsize); memset(AFFS_DATA(bh) + boff, 0, tmp); AFFS_DATA_HEAD(bh)->size = cpu_to_be32(be32_to_cpu(AFFS_DATA_HEAD(bh)->size) + tmp); affs_fix_checksum(sb, bh); @@ -560,8 +556,7 @@ affs_extent_file_ofs(struct inode *inode, u32 newsize) if (IS_ERR(bh)) goto out; tmp = min(bsize, newsize - size); - if (tmp > bsize) - BUG(); + BUG_ON(tmp > bsize); AFFS_DATA_HEAD(bh)->ptype = cpu_to_be32(T_DATA); AFFS_DATA_HEAD(bh)->key = cpu_to_be32(inode->i_ino); AFFS_DATA_HEAD(bh)->sequence = cpu_to_be32(bidx); @@ -683,8 +678,7 @@ static int affs_write_end_ofs(struct file *file, struct address_space *mapping, if (IS_ERR(bh)) return PTR_ERR(bh); tmp = min(bsize - boff, to - from); - if (boff + tmp > bsize || tmp > bsize) - BUG(); + BUG_ON(boff + tmp > bsize || tmp > bsize); memcpy(AFFS_DATA(bh) + boff, data + from, tmp); AFFS_DATA_HEAD(bh)->size = cpu_to_be32(be32_to_cpu(AFFS_DATA_HEAD(bh)->size) + tmp); affs_fix_checksum(sb, bh); @@ -732,8 +726,7 @@ static int affs_write_end_ofs(struct file *file, struct address_space *mapping, if (IS_ERR(bh)) goto out; tmp = min(bsize, to - from); - if (tmp > bsize) - BUG(); + BUG_ON(tmp > bsize); memcpy(AFFS_DATA(bh), data + from, tmp); if (buffer_new(bh)) { AFFS_DATA_HEAD(bh)->ptype = cpu_to_be32(T_DATA); -- cgit v1.2.3 From 175a06ae300188af8a61db68a78e1af44dc7d44f Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Tue, 29 Apr 2008 00:59:17 -0700 Subject: exec: remove argv_len from struct linux_binprm I noticed that 2.6.24.2 calculates bprm->argv_len at do_execve(). But it doesn't update bprm->argv_len after "remove_arg_zero() + copy_strings_kernel()" at load_script() etc. audit_bprm() is called from search_binary_handler() and search_binary_handler() is called from load_script() etc. Thus, I think the condition check if (bprm->argv_len > (audit_argv_kb << 10)) return -E2BIG; in audit_bprm() might return wrong result when strlen(removed_arg) != strlen(spliced_args). Why not update bprm->argv_len at load_script() etc. ? By the way, 2.6.25-rc3 seems to not doing the condition check. Is the field bprm->argv_len no longer needed? Signed-off-by: Tetsuo Handa Cc: Ollie Wild Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 3 --- include/linux/binfmts.h | 1 - 2 files changed, 4 deletions(-) (limited to 'fs') diff --git a/fs/exec.c b/fs/exec.c index b152029f18f6..7768453dc986 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1268,7 +1268,6 @@ int do_execve(char * filename, { struct linux_binprm *bprm; struct file *file; - unsigned long env_p; struct files_struct *displaced; int retval; @@ -1321,11 +1320,9 @@ int do_execve(char * filename, if (retval < 0) goto out; - env_p = bprm->p; retval = copy_strings(bprm->argc, argv, bprm); if (retval < 0) goto out; - bprm->argv_len = env_p - bprm->p; retval = search_binary_handler(bprm,regs); if (retval >= 0) { diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h index b7fc55ec8d48..1dd756731c95 100644 --- a/include/linux/binfmts.h +++ b/include/linux/binfmts.h @@ -48,7 +48,6 @@ struct linux_binprm{ unsigned interp_flags; unsigned interp_data; unsigned long loader, exec; - unsigned long argv_len; }; #define BINPRM_FLAGS_ENFORCE_NONDUMP_BIT 0 -- cgit v1.2.3 From a2d416dcc92e576d0e339efd641bd3d8ee2bfb4d Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 29 Apr 2008 00:59:22 -0700 Subject: codafs: fix build warning powerpc: fs/coda/coda_linux.c: In function 'coda_iattr_to_vattr': fs/coda/coda_linux.c:137: warning: large integer implicitly truncated to unsigned type Cc: Jan Harkes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/coda/coda_linux.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/coda/coda_linux.c b/fs/coda/coda_linux.c index 95a54253c047..e1c854890f94 100644 --- a/fs/coda/coda_linux.c +++ b/fs/coda/coda_linux.c @@ -134,7 +134,7 @@ void coda_iattr_to_vattr(struct iattr *iattr, struct coda_vattr *vattr) unsigned int valid; /* clean out */ - vattr->va_mode = (umode_t) -1; + vattr->va_mode = -1; vattr->va_uid = (vuid_t) -1; vattr->va_gid = (vgid_t) -1; vattr->va_size = (off_t) -1; -- cgit v1.2.3 From 3a2e7f47d71e1df86acc1dda6826890b6546a4e1 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Tue, 29 Apr 2008 00:59:24 -0700 Subject: binfmt_misc.c: avoid potential kernel stack overflow This can be triggered with root help only, but... Register the ":text:E::txt::/root/cat.txt:' rule in binfmt_misc (by root) and try launching the cat.txt file (by anyone) :) The result is - the endless recursion in the load_misc_binary -> open_exec -> load_misc_binary chain and stack overflow. There's a similar problem with binfmt_script, and there's a sh_bang memner on linux_binprm structure to handle this, but simply raising this in binfmt_misc may break some setups when the interpreter of some misc binaries is a script. So the proposal is to turn sh_bang into a bit, add a new one (the misc_bang) and raise it in load_misc_binary. After this, even if we set up the misc -> script -> misc loop for binfmts one of them will step on its own bang and exit. Signed-off-by: Pavel Emelyanov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/binfmt_em86.c | 2 +- fs/binfmt_misc.c | 6 ++++++ fs/binfmt_script.c | 2 +- include/linux/binfmts.h | 3 ++- 4 files changed, 10 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/binfmt_em86.c b/fs/binfmt_em86.c index f95ae9789c91..f9c88d0c8ced 100644 --- a/fs/binfmt_em86.c +++ b/fs/binfmt_em86.c @@ -43,7 +43,7 @@ static int load_em86(struct linux_binprm *bprm,struct pt_regs *regs) return -ENOEXEC; } - bprm->sh_bang++; /* Well, the bang-shell is implicit... */ + bprm->sh_bang = 1; /* Well, the bang-shell is implicit... */ allow_write_access(bprm->file); fput(bprm->file); bprm->file = NULL; diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index dbf0ac0523de..7191306367c5 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -115,6 +115,12 @@ static int load_misc_binary(struct linux_binprm *bprm, struct pt_regs *regs) if (!enabled) goto _ret; + retval = -ENOEXEC; + if (bprm->misc_bang) + goto _ret; + + bprm->misc_bang = 1; + /* to keep locking time low, we copy the interpreter string */ read_lock(&entries_lock); fmt = check_file(bprm); diff --git a/fs/binfmt_script.c b/fs/binfmt_script.c index ab33939b12a7..9e3963f7ebf1 100644 --- a/fs/binfmt_script.c +++ b/fs/binfmt_script.c @@ -29,7 +29,7 @@ static int load_script(struct linux_binprm *bprm,struct pt_regs *regs) * Sorta complicated, but hopefully it will work. -TYT */ - bprm->sh_bang++; + bprm->sh_bang = 1; allow_write_access(bprm->file); fput(bprm->file); bprm->file = NULL; diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h index 1dd756731c95..b512e48f6d8e 100644 --- a/include/linux/binfmts.h +++ b/include/linux/binfmts.h @@ -34,7 +34,8 @@ struct linux_binprm{ #endif struct mm_struct *mm; unsigned long p; /* current top of mem */ - int sh_bang; + unsigned int sh_bang:1, + misc_bang:1; struct file * file; int e_uid, e_gid; kernel_cap_t cap_inheritable, cap_permitted; -- cgit v1.2.3 From 2e50b6ccdaaf0d933bb9d8409cac4b2f088f5a2f Mon Sep 17 00:00:00 2001 From: "S.Caglar Onur" Date: Tue, 29 Apr 2008 00:59:26 -0700 Subject: fs/binfmt_aout.c: use printk_ratelimit() Use printk_ratelimit() instead of jiffies based arithmetic, suggested by Geert Uytterhoeven Signed-off-by: S.Caglar Onur Cc: Geert Uytterhoeven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/binfmt_aout.c | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index a1bb2244cac7..ba4cddb92f1d 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -372,21 +372,17 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs) flush_icache_range(text_addr, text_addr+ex.a_text+ex.a_data); } else { - static unsigned long error_time, error_time2; if ((ex.a_text & 0xfff || ex.a_data & 0xfff) && - (N_MAGIC(ex) != NMAGIC) && (jiffies-error_time2) > 5*HZ) + (N_MAGIC(ex) != NMAGIC) && printk_ratelimit()) { printk(KERN_NOTICE "executable not page aligned\n"); - error_time2 = jiffies; } - if ((fd_offset & ~PAGE_MASK) != 0 && - (jiffies-error_time) > 5*HZ) + if ((fd_offset & ~PAGE_MASK) != 0 && printk_ratelimit()) { printk(KERN_WARNING "fd_offset is not page aligned. Please convert program: %s\n", bprm->file->f_path.dentry->d_name.name); - error_time = jiffies; } if (!bprm->file->f_op->mmap||((fd_offset & ~PAGE_MASK) != 0)) { @@ -495,15 +491,13 @@ static int load_aout_library(struct file *file) start_addr = ex.a_entry & 0xfffff000; if ((N_TXTOFF(ex) & ~PAGE_MASK) != 0) { - static unsigned long error_time; loff_t pos = N_TXTOFF(ex); - if ((jiffies-error_time) > 5*HZ) + if (printk_ratelimit()) { printk(KERN_WARNING "N_TXTOFF is not page aligned. Please convert library: %s\n", file->f_path.dentry->d_name.name); - error_time = jiffies; } down_write(¤t->mm->mmap_sem); do_brk(start_addr, ex.a_text + ex.a_data + ex.a_bss); -- cgit v1.2.3 From e1d2c8b69ad81ea103b1e87809eba51931e16874 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 29 Apr 2008 00:59:34 -0700 Subject: fdpic: check that the size returned by kernel_read() is what we asked for Check that the size of the read returned by kernel_read() is what we asked for. If it isn't, then reject the binary as being a badly formatted. Signed-off-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/binfmt_elf_fdpic.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 32649f2a1654..ddd35d873391 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -136,8 +136,8 @@ static int elf_fdpic_fetch_phdrs(struct elf_fdpic_params *params, retval = kernel_read(file, params->hdr.e_phoff, (char *) params->phdrs, size); - if (retval < 0) - return retval; + if (unlikely(retval != size)) + return retval < 0 ? retval : -ENOEXEC; /* determine stack size for this binary */ phdr = params->phdrs; @@ -218,8 +218,11 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, phdr->p_offset, interpreter_name, phdr->p_filesz); - if (retval < 0) + if (unlikely(retval != phdr->p_filesz)) { + if (retval >= 0) + retval = -ENOEXEC; goto error; + } retval = -ENOENT; if (interpreter_name[phdr->p_filesz - 1] != '\0') @@ -245,8 +248,11 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, retval = kernel_read(interpreter, 0, bprm->buf, BINPRM_BUF_SIZE); - if (retval < 0) + if (unlikely(retval != BINPRM_BUF_SIZE)) { + if (retval >= 0) + retval = -ENOEXEC; goto error; + } interp_params.hdr = *((struct elfhdr *) bprm->buf); break; -- cgit v1.2.3 From eccb95cee4f0d56faa46ef22fb94dd4a3578d3eb Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 29 Apr 2008 00:59:37 -0700 Subject: vfs: fix lock inversion in drop_pagecache_sb() Fix longstanding lock inversion in drop_pagecache_sb by dropping inode_lock before calling __invalidate_mapping_pages(). We just have to make sure inode won't go away from under us by keeping reference to it and putting the reference only after we have safely resumed the scan of the inode list. A bit tricky but not too bad... Signed-off-by: Jan Kara Cc: Fengguang Wu Cc: David Chinner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/drop_caches.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/drop_caches.c b/fs/drop_caches.c index e2c6b6500c78..50f9087635d8 100644 --- a/fs/drop_caches.c +++ b/fs/drop_caches.c @@ -14,15 +14,21 @@ int sysctl_drop_caches; static void drop_pagecache_sb(struct super_block *sb) { - struct inode *inode; + struct inode *inode, *toput_inode = NULL; spin_lock(&inode_lock); list_for_each_entry(inode, &sb->s_inodes, i_sb_list) { if (inode->i_state & (I_FREEING|I_WILL_FREE)) continue; + __iget(inode); + spin_unlock(&inode_lock); __invalidate_mapping_pages(inode->i_mapping, 0, -1, true); + iput(toput_inode); + toput_inode = inode; + spin_lock(&inode_lock); } spin_unlock(&inode_lock); + iput(toput_inode); } static void drop_pagecache(void) -- cgit v1.2.3 From af065b8a19041554196971d8b6ae1459798d3b14 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 29 Apr 2008 00:59:39 -0700 Subject: vfs: skip inodes without pages to free in drop_pagecache_sb() Many inodes have no pagecache, so we can avoid lots of lock-takings. Signed-off-by: Jan Kara Cc: Fengguang Wu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/drop_caches.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/drop_caches.c b/fs/drop_caches.c index 50f9087635d8..3e5637fc3779 100644 --- a/fs/drop_caches.c +++ b/fs/drop_caches.c @@ -20,6 +20,8 @@ static void drop_pagecache_sb(struct super_block *sb) list_for_each_entry(inode, &sb->s_inodes, i_sb_list) { if (inode->i_state & (I_FREEING|I_WILL_FREE)) continue; + if (inode->i_mapping->nrpages == 0) + continue; __iget(inode); spin_unlock(&inode_lock); __invalidate_mapping_pages(inode->i_mapping, 0, -1, true); -- cgit v1.2.3 From c5c8be3ce59dc59baf20b33dae3f8eb70af7b1f1 Mon Sep 17 00:00:00 2001 From: Matthias Kaehlcke Date: Tue, 29 Apr 2008 00:59:40 -0700 Subject: fs/inode.c: use hlist_for_each_entry() fs/inode.c: use hlist_for_each_entry() in find_inode() and find_inode_fast() [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Matthias Kaehlcke Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/inode.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/inode.c b/fs/inode.c index 27ee1af50d02..bf6478130424 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -495,8 +495,7 @@ static struct inode * find_inode(struct super_block * sb, struct hlist_head *hea struct inode * inode = NULL; repeat: - hlist_for_each (node, head) { - inode = hlist_entry(node, struct inode, i_hash); + hlist_for_each_entry(inode, node, head, i_hash) { if (inode->i_sb != sb) continue; if (!test(inode, data)) @@ -520,8 +519,7 @@ static struct inode * find_inode_fast(struct super_block * sb, struct hlist_head struct inode * inode = NULL; repeat: - hlist_for_each (node, head) { - inode = hlist_entry(node, struct inode, i_hash); + hlist_for_each_entry(inode, node, head, i_hash) { if (inode->i_ino != ino) continue; if (inode->i_sb != sb) -- cgit v1.2.3 From 7ec02ef1596bb3c829a7e8b65ebf13b87faf1819 Mon Sep 17 00:00:00 2001 From: Jan Blunck Date: Tue, 29 Apr 2008 00:59:40 -0700 Subject: vfs: remove lives_below_in_same_fs() Remove lives_below_in_same_fs() since is_subdir() from fs/dcache.c is providing the same functionality. Signed-off-by: Jan Blunck Acked-by: Miklos Szeredi Cc: Al Viro Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/namespace.c | 13 +------------ 1 file changed, 1 insertion(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/namespace.c b/fs/namespace.c index fe376805cf5f..061e5edb4d27 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1176,17 +1176,6 @@ static int mount_is_safe(struct nameidata *nd) #endif } -static int lives_below_in_same_fs(struct dentry *d, struct dentry *dentry) -{ - while (1) { - if (d == dentry) - return 1; - if (d == NULL || d == d->d_parent) - return 0; - d = d->d_parent; - } -} - struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, int flag) { @@ -1203,7 +1192,7 @@ struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, p = mnt; list_for_each_entry(r, &mnt->mnt_mounts, mnt_child) { - if (!lives_below_in_same_fs(r->mnt_mountpoint, dentry)) + if (!is_subdir(r->mnt_mountpoint, dentry)) continue; for (s = r; s; s = next_mnt(s, r)) { -- cgit v1.2.3 From 8f0cfa52a1d4ffacd8e7de906d19662f5da58d58 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 29 Apr 2008 00:59:41 -0700 Subject: xattr: add missing consts to function arguments Add missing consts to xattr function arguments. Signed-off-by: David Howells Cc: Andreas Gruenbacher Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/xattr.c | 41 ++++++++++++++++++----------------- include/linux/security.h | 43 ++++++++++++++++++++----------------- include/linux/syscalls.h | 30 ++++++++++++++------------ include/linux/xattr.h | 6 +++--- security/commoncap.c | 6 +++--- security/dummy.c | 13 +++++------ security/security.c | 12 +++++------ security/selinux/hooks.c | 14 ++++++------ security/selinux/include/security.h | 2 +- security/selinux/ss/services.c | 4 ++-- security/smack/smack_lsm.c | 12 +++++------ 11 files changed, 96 insertions(+), 87 deletions(-) (limited to 'fs') diff --git a/fs/xattr.c b/fs/xattr.c index 89a942f07e1b..4706a8b1f495 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -67,7 +67,7 @@ xattr_permission(struct inode *inode, const char *name, int mask) } int -vfs_setxattr(struct dentry *dentry, char *name, void *value, +vfs_setxattr(struct dentry *dentry, const char *name, const void *value, size_t size, int flags) { struct inode *inode = dentry->d_inode; @@ -131,7 +131,7 @@ out_noalloc: EXPORT_SYMBOL_GPL(xattr_getsecurity); ssize_t -vfs_getxattr(struct dentry *dentry, char *name, void *value, size_t size) +vfs_getxattr(struct dentry *dentry, const char *name, void *value, size_t size) { struct inode *inode = dentry->d_inode; int error; @@ -187,7 +187,7 @@ vfs_listxattr(struct dentry *d, char *list, size_t size) EXPORT_SYMBOL_GPL(vfs_listxattr); int -vfs_removexattr(struct dentry *dentry, char *name) +vfs_removexattr(struct dentry *dentry, const char *name) { struct inode *inode = dentry->d_inode; int error; @@ -218,7 +218,7 @@ EXPORT_SYMBOL_GPL(vfs_removexattr); * Extended attribute SET operations */ static long -setxattr(struct dentry *d, char __user *name, void __user *value, +setxattr(struct dentry *d, const char __user *name, const void __user *value, size_t size, int flags) { int error; @@ -252,8 +252,8 @@ setxattr(struct dentry *d, char __user *name, void __user *value, } asmlinkage long -sys_setxattr(char __user *path, char __user *name, void __user *value, - size_t size, int flags) +sys_setxattr(const char __user *path, const char __user *name, + const void __user *value, size_t size, int flags) { struct nameidata nd; int error; @@ -271,8 +271,8 @@ sys_setxattr(char __user *path, char __user *name, void __user *value, } asmlinkage long -sys_lsetxattr(char __user *path, char __user *name, void __user *value, - size_t size, int flags) +sys_lsetxattr(const char __user *path, const char __user *name, + const void __user *value, size_t size, int flags) { struct nameidata nd; int error; @@ -290,7 +290,7 @@ sys_lsetxattr(char __user *path, char __user *name, void __user *value, } asmlinkage long -sys_fsetxattr(int fd, char __user *name, void __user *value, +sys_fsetxattr(int fd, const char __user *name, const void __user *value, size_t size, int flags) { struct file *f; @@ -315,7 +315,8 @@ sys_fsetxattr(int fd, char __user *name, void __user *value, * Extended attribute GET operations */ static ssize_t -getxattr(struct dentry *d, char __user *name, void __user *value, size_t size) +getxattr(struct dentry *d, const char __user *name, void __user *value, + size_t size) { ssize_t error; void *kvalue = NULL; @@ -349,8 +350,8 @@ getxattr(struct dentry *d, char __user *name, void __user *value, size_t size) } asmlinkage ssize_t -sys_getxattr(char __user *path, char __user *name, void __user *value, - size_t size) +sys_getxattr(const char __user *path, const char __user *name, + void __user *value, size_t size) { struct nameidata nd; ssize_t error; @@ -364,7 +365,7 @@ sys_getxattr(char __user *path, char __user *name, void __user *value, } asmlinkage ssize_t -sys_lgetxattr(char __user *path, char __user *name, void __user *value, +sys_lgetxattr(const char __user *path, const char __user *name, void __user *value, size_t size) { struct nameidata nd; @@ -379,7 +380,7 @@ sys_lgetxattr(char __user *path, char __user *name, void __user *value, } asmlinkage ssize_t -sys_fgetxattr(int fd, char __user *name, void __user *value, size_t size) +sys_fgetxattr(int fd, const char __user *name, void __user *value, size_t size) { struct file *f; ssize_t error = -EBADF; @@ -424,7 +425,7 @@ listxattr(struct dentry *d, char __user *list, size_t size) } asmlinkage ssize_t -sys_listxattr(char __user *path, char __user *list, size_t size) +sys_listxattr(const char __user *path, char __user *list, size_t size) { struct nameidata nd; ssize_t error; @@ -438,7 +439,7 @@ sys_listxattr(char __user *path, char __user *list, size_t size) } asmlinkage ssize_t -sys_llistxattr(char __user *path, char __user *list, size_t size) +sys_llistxattr(const char __user *path, char __user *list, size_t size) { struct nameidata nd; ssize_t error; @@ -470,7 +471,7 @@ sys_flistxattr(int fd, char __user *list, size_t size) * Extended attribute REMOVE operations */ static long -removexattr(struct dentry *d, char __user *name) +removexattr(struct dentry *d, const char __user *name) { int error; char kname[XATTR_NAME_MAX + 1]; @@ -485,7 +486,7 @@ removexattr(struct dentry *d, char __user *name) } asmlinkage long -sys_removexattr(char __user *path, char __user *name) +sys_removexattr(const char __user *path, const char __user *name) { struct nameidata nd; int error; @@ -503,7 +504,7 @@ sys_removexattr(char __user *path, char __user *name) } asmlinkage long -sys_lremovexattr(char __user *path, char __user *name) +sys_lremovexattr(const char __user *path, const char __user *name) { struct nameidata nd; int error; @@ -521,7 +522,7 @@ sys_lremovexattr(char __user *path, char __user *name) } asmlinkage long -sys_fremovexattr(int fd, char __user *name) +sys_fremovexattr(int fd, const char __user *name) { struct file *f; struct dentry *dentry; diff --git a/include/linux/security.h b/include/linux/security.h index d0a28fd1747a..3ebcdd00b17d 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -53,8 +53,9 @@ extern void cap_capset_set(struct task_struct *target, kernel_cap_t *effective, extern int cap_bprm_set_security(struct linux_binprm *bprm); extern void cap_bprm_apply_creds(struct linux_binprm *bprm, int unsafe); extern int cap_bprm_secureexec(struct linux_binprm *bprm); -extern int cap_inode_setxattr(struct dentry *dentry, char *name, void *value, size_t size, int flags); -extern int cap_inode_removexattr(struct dentry *dentry, char *name); +extern int cap_inode_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags); +extern int cap_inode_removexattr(struct dentry *dentry, const char *name); extern int cap_inode_need_killpriv(struct dentry *dentry); extern int cap_inode_killpriv(struct dentry *dentry); extern int cap_task_post_setuid(uid_t old_ruid, uid_t old_euid, uid_t old_suid, int flags); @@ -1362,13 +1363,13 @@ struct security_operations { int (*inode_setattr) (struct dentry *dentry, struct iattr *attr); int (*inode_getattr) (struct vfsmount *mnt, struct dentry *dentry); void (*inode_delete) (struct inode *inode); - int (*inode_setxattr) (struct dentry *dentry, char *name, void *value, - size_t size, int flags); - void (*inode_post_setxattr) (struct dentry *dentry, char *name, void *value, - size_t size, int flags); - int (*inode_getxattr) (struct dentry *dentry, char *name); + int (*inode_setxattr) (struct dentry *dentry, const char *name, + const void *value, size_t size, int flags); + void (*inode_post_setxattr) (struct dentry *dentry, const char *name, + const void *value, size_t size, int flags); + int (*inode_getxattr) (struct dentry *dentry, const char *name); int (*inode_listxattr) (struct dentry *dentry); - int (*inode_removexattr) (struct dentry *dentry, char *name); + int (*inode_removexattr) (struct dentry *dentry, const char *name); int (*inode_need_killpriv) (struct dentry *dentry); int (*inode_killpriv) (struct dentry *dentry); int (*inode_getsecurity) (const struct inode *inode, const char *name, void **buffer, bool alloc); @@ -1633,13 +1634,13 @@ int security_inode_permission(struct inode *inode, int mask, struct nameidata *n int security_inode_setattr(struct dentry *dentry, struct iattr *attr); int security_inode_getattr(struct vfsmount *mnt, struct dentry *dentry); void security_inode_delete(struct inode *inode); -int security_inode_setxattr(struct dentry *dentry, char *name, - void *value, size_t size, int flags); -void security_inode_post_setxattr(struct dentry *dentry, char *name, - void *value, size_t size, int flags); -int security_inode_getxattr(struct dentry *dentry, char *name); +int security_inode_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags); +void security_inode_post_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags); +int security_inode_getxattr(struct dentry *dentry, const char *name); int security_inode_listxattr(struct dentry *dentry); -int security_inode_removexattr(struct dentry *dentry, char *name); +int security_inode_removexattr(struct dentry *dentry, const char *name); int security_inode_need_killpriv(struct dentry *dentry); int security_inode_killpriv(struct dentry *dentry); int security_inode_getsecurity(const struct inode *inode, const char *name, void **buffer, bool alloc); @@ -2041,17 +2042,18 @@ static inline int security_inode_getattr(struct vfsmount *mnt, static inline void security_inode_delete(struct inode *inode) { } -static inline int security_inode_setxattr(struct dentry *dentry, char *name, - void *value, size_t size, int flags) +static inline int security_inode_setxattr(struct dentry *dentry, + const char *name, const void *value, size_t size, int flags) { return cap_inode_setxattr(dentry, name, value, size, flags); } -static inline void security_inode_post_setxattr(struct dentry *dentry, char *name, - void *value, size_t size, int flags) +static inline void security_inode_post_setxattr(struct dentry *dentry, + const char *name, const void *value, size_t size, int flags) { } -static inline int security_inode_getxattr(struct dentry *dentry, char *name) +static inline int security_inode_getxattr(struct dentry *dentry, + const char *name) { return 0; } @@ -2061,7 +2063,8 @@ static inline int security_inode_listxattr(struct dentry *dentry) return 0; } -static inline int security_inode_removexattr(struct dentry *dentry, char *name) +static inline int security_inode_removexattr(struct dentry *dentry, + const char *name) { return cap_inode_removexattr(dentry, name); } diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 8df6d1382ac8..0522f368f9d7 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -240,26 +240,28 @@ asmlinkage long sys_truncate64(const char __user *path, loff_t length); asmlinkage long sys_ftruncate64(unsigned int fd, loff_t length); #endif -asmlinkage long sys_setxattr(char __user *path, char __user *name, - void __user *value, size_t size, int flags); -asmlinkage long sys_lsetxattr(char __user *path, char __user *name, - void __user *value, size_t size, int flags); -asmlinkage long sys_fsetxattr(int fd, char __user *name, void __user *value, - size_t size, int flags); -asmlinkage ssize_t sys_getxattr(char __user *path, char __user *name, +asmlinkage long sys_setxattr(const char __user *path, const char __user *name, + const void __user *value, size_t size, int flags); +asmlinkage long sys_lsetxattr(const char __user *path, const char __user *name, + const void __user *value, size_t size, int flags); +asmlinkage long sys_fsetxattr(int fd, const char __user *name, + const void __user *value, size_t size, int flags); +asmlinkage ssize_t sys_getxattr(const char __user *path, const char __user *name, void __user *value, size_t size); -asmlinkage ssize_t sys_lgetxattr(char __user *path, char __user *name, +asmlinkage ssize_t sys_lgetxattr(const char __user *path, const char __user *name, void __user *value, size_t size); -asmlinkage ssize_t sys_fgetxattr(int fd, char __user *name, +asmlinkage ssize_t sys_fgetxattr(int fd, const char __user *name, void __user *value, size_t size); -asmlinkage ssize_t sys_listxattr(char __user *path, char __user *list, +asmlinkage ssize_t sys_listxattr(const char __user *path, char __user *list, size_t size); -asmlinkage ssize_t sys_llistxattr(char __user *path, char __user *list, +asmlinkage ssize_t sys_llistxattr(const char __user *path, char __user *list, size_t size); asmlinkage ssize_t sys_flistxattr(int fd, char __user *list, size_t size); -asmlinkage long sys_removexattr(char __user *path, char __user *name); -asmlinkage long sys_lremovexattr(char __user *path, char __user *name); -asmlinkage long sys_fremovexattr(int fd, char __user *name); +asmlinkage long sys_removexattr(const char __user *path, + const char __user *name); +asmlinkage long sys_lremovexattr(const char __user *path, + const char __user *name); +asmlinkage long sys_fremovexattr(int fd, const char __user *name); asmlinkage unsigned long sys_brk(unsigned long brk); asmlinkage long sys_mprotect(unsigned long start, size_t len, diff --git a/include/linux/xattr.h b/include/linux/xattr.h index df6b95d2218e..d131e352cfe1 100644 --- a/include/linux/xattr.h +++ b/include/linux/xattr.h @@ -47,10 +47,10 @@ struct xattr_handler { }; ssize_t xattr_getsecurity(struct inode *, const char *, void *, size_t); -ssize_t vfs_getxattr(struct dentry *, char *, void *, size_t); +ssize_t vfs_getxattr(struct dentry *, const char *, void *, size_t); ssize_t vfs_listxattr(struct dentry *d, char *list, size_t size); -int vfs_setxattr(struct dentry *, char *, void *, size_t, int); -int vfs_removexattr(struct dentry *, char *); +int vfs_setxattr(struct dentry *, const char *, const void *, size_t, int); +int vfs_removexattr(struct dentry *, const char *); ssize_t generic_getxattr(struct dentry *dentry, const char *name, void *buffer, size_t size); ssize_t generic_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size); diff --git a/security/commoncap.c b/security/commoncap.c index e8c3f5e46705..5edabc7542ae 100644 --- a/security/commoncap.c +++ b/security/commoncap.c @@ -383,8 +383,8 @@ int cap_bprm_secureexec (struct linux_binprm *bprm) current->egid != current->gid); } -int cap_inode_setxattr(struct dentry *dentry, char *name, void *value, - size_t size, int flags) +int cap_inode_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags) { if (!strcmp(name, XATTR_NAME_CAPS)) { if (!capable(CAP_SETFCAP)) @@ -397,7 +397,7 @@ int cap_inode_setxattr(struct dentry *dentry, char *name, void *value, return 0; } -int cap_inode_removexattr(struct dentry *dentry, char *name) +int cap_inode_removexattr(struct dentry *dentry, const char *name) { if (!strcmp(name, XATTR_NAME_CAPS)) { if (!capable(CAP_SETFCAP)) diff --git a/security/dummy.c b/security/dummy.c index 58d4dd1af5c7..26ee06ef0e93 100644 --- a/security/dummy.c +++ b/security/dummy.c @@ -365,8 +365,8 @@ static void dummy_inode_delete (struct inode *ino) return; } -static int dummy_inode_setxattr (struct dentry *dentry, char *name, void *value, - size_t size, int flags) +static int dummy_inode_setxattr (struct dentry *dentry, const char *name, + const void *value, size_t size, int flags) { if (!strncmp(name, XATTR_SECURITY_PREFIX, sizeof(XATTR_SECURITY_PREFIX) - 1) && @@ -375,12 +375,13 @@ static int dummy_inode_setxattr (struct dentry *dentry, char *name, void *value, return 0; } -static void dummy_inode_post_setxattr (struct dentry *dentry, char *name, void *value, - size_t size, int flags) +static void dummy_inode_post_setxattr (struct dentry *dentry, const char *name, + const void *value, size_t size, + int flags) { } -static int dummy_inode_getxattr (struct dentry *dentry, char *name) +static int dummy_inode_getxattr (struct dentry *dentry, const char *name) { return 0; } @@ -390,7 +391,7 @@ static int dummy_inode_listxattr (struct dentry *dentry) return 0; } -static int dummy_inode_removexattr (struct dentry *dentry, char *name) +static int dummy_inode_removexattr (struct dentry *dentry, const char *name) { if (!strncmp(name, XATTR_SECURITY_PREFIX, sizeof(XATTR_SECURITY_PREFIX) - 1) && diff --git a/security/security.c b/security/security.c index d5cb5898d967..a809035441ab 100644 --- a/security/security.c +++ b/security/security.c @@ -491,23 +491,23 @@ void security_inode_delete(struct inode *inode) security_ops->inode_delete(inode); } -int security_inode_setxattr(struct dentry *dentry, char *name, - void *value, size_t size, int flags) +int security_inode_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags) { if (unlikely(IS_PRIVATE(dentry->d_inode))) return 0; return security_ops->inode_setxattr(dentry, name, value, size, flags); } -void security_inode_post_setxattr(struct dentry *dentry, char *name, - void *value, size_t size, int flags) +void security_inode_post_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags) { if (unlikely(IS_PRIVATE(dentry->d_inode))) return; security_ops->inode_post_setxattr(dentry, name, value, size, flags); } -int security_inode_getxattr(struct dentry *dentry, char *name) +int security_inode_getxattr(struct dentry *dentry, const char *name) { if (unlikely(IS_PRIVATE(dentry->d_inode))) return 0; @@ -521,7 +521,7 @@ int security_inode_listxattr(struct dentry *dentry) return security_ops->inode_listxattr(dentry); } -int security_inode_removexattr(struct dentry *dentry, char *name) +int security_inode_removexattr(struct dentry *dentry, const char *name) { if (unlikely(IS_PRIVATE(dentry->d_inode))) return 0; diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 04acb5af8317..047365ac9faa 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -2619,7 +2619,7 @@ static int selinux_inode_getattr(struct vfsmount *mnt, struct dentry *dentry) return dentry_has_perm(current, mnt, dentry, FILE__GETATTR); } -static int selinux_inode_setotherxattr(struct dentry *dentry, char *name) +static int selinux_inode_setotherxattr(struct dentry *dentry, const char *name) { if (!strncmp(name, XATTR_SECURITY_PREFIX, sizeof XATTR_SECURITY_PREFIX - 1)) { @@ -2638,7 +2638,8 @@ static int selinux_inode_setotherxattr(struct dentry *dentry, char *name) return dentry_has_perm(current, NULL, dentry, FILE__SETATTR); } -static int selinux_inode_setxattr(struct dentry *dentry, char *name, void *value, size_t size, int flags) +static int selinux_inode_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags) { struct task_security_struct *tsec = current->security; struct inode *inode = dentry->d_inode; @@ -2687,8 +2688,9 @@ static int selinux_inode_setxattr(struct dentry *dentry, char *name, void *value &ad); } -static void selinux_inode_post_setxattr(struct dentry *dentry, char *name, - void *value, size_t size, int flags) +static void selinux_inode_post_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, + int flags) { struct inode *inode = dentry->d_inode; struct inode_security_struct *isec = inode->i_security; @@ -2711,7 +2713,7 @@ static void selinux_inode_post_setxattr(struct dentry *dentry, char *name, return; } -static int selinux_inode_getxattr(struct dentry *dentry, char *name) +static int selinux_inode_getxattr(struct dentry *dentry, const char *name) { return dentry_has_perm(current, NULL, dentry, FILE__GETATTR); } @@ -2721,7 +2723,7 @@ static int selinux_inode_listxattr(struct dentry *dentry) return dentry_has_perm(current, NULL, dentry, FILE__GETATTR); } -static int selinux_inode_removexattr(struct dentry *dentry, char *name) +static int selinux_inode_removexattr(struct dentry *dentry, const char *name) { if (strcmp(name, XATTR_NAME_SELINUX)) return selinux_inode_setotherxattr(dentry, name); diff --git a/security/selinux/include/security.h b/security/selinux/include/security.h index 6445b6440648..cdb14add27d2 100644 --- a/security/selinux/include/security.h +++ b/security/selinux/include/security.h @@ -93,7 +93,7 @@ int security_change_sid(u32 ssid, u32 tsid, int security_sid_to_context(u32 sid, char **scontext, u32 *scontext_len); -int security_context_to_sid(char *scontext, u32 scontext_len, +int security_context_to_sid(const char *scontext, u32 scontext_len, u32 *out_sid); int security_context_to_sid_default(char *scontext, u32 scontext_len, diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index 2daaddbb301d..25cac5a2aa8e 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -708,7 +708,7 @@ out: } -static int security_context_to_sid_core(char *scontext, u32 scontext_len, +static int security_context_to_sid_core(const char *scontext, u32 scontext_len, u32 *sid, u32 def_sid, gfp_t gfp_flags) { char *scontext2; @@ -835,7 +835,7 @@ out: * Returns -%EINVAL if the context is invalid, -%ENOMEM if insufficient * memory is available, or 0 on success. */ -int security_context_to_sid(char *scontext, u32 scontext_len, u32 *sid) +int security_context_to_sid(const char *scontext, u32 scontext_len, u32 *sid) { return security_context_to_sid_core(scontext, scontext_len, sid, SECSID_NULL, GFP_KERNEL); diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c index 77ec16a3b68b..5d2ec5650e61 100644 --- a/security/smack/smack_lsm.c +++ b/security/smack/smack_lsm.c @@ -574,8 +574,8 @@ static int smack_inode_getattr(struct vfsmount *mnt, struct dentry *dentry) * * Returns 0 if access is permitted, an error code otherwise */ -static int smack_inode_setxattr(struct dentry *dentry, char *name, - void *value, size_t size, int flags) +static int smack_inode_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags) { int rc = 0; @@ -604,8 +604,8 @@ static int smack_inode_setxattr(struct dentry *dentry, char *name, * Set the pointer in the inode blob to the entry found * in the master label list. */ -static void smack_inode_post_setxattr(struct dentry *dentry, char *name, - void *value, size_t size, int flags) +static void smack_inode_post_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags) { struct inode_smack *isp; char *nsp; @@ -641,7 +641,7 @@ static void smack_inode_post_setxattr(struct dentry *dentry, char *name, * * Returns 0 if access is permitted, an error code otherwise */ -static int smack_inode_getxattr(struct dentry *dentry, char *name) +static int smack_inode_getxattr(struct dentry *dentry, const char *name) { return smk_curacc(smk_of_inode(dentry->d_inode), MAY_READ); } @@ -655,7 +655,7 @@ static int smack_inode_getxattr(struct dentry *dentry, char *name) * * Returns 0 if access is permitted, an error code otherwise */ -static int smack_inode_removexattr(struct dentry *dentry, char *name) +static int smack_inode_removexattr(struct dentry *dentry, const char *name) { int rc = 0; -- cgit v1.2.3 From 762873c251b056c6c1b29e83a4dabafb064e5421 Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Tue, 29 Apr 2008 00:59:42 -0700 Subject: vfs: fix unconditional write_super() call in file_fsync() We need to check ->s_dirt before calling write_super(). It became the cause of an unneeded write. This bug was noticed by Sudhanshu Saxena. Signed-off-by: OGAWA Hirofumi Cc: Al Viro Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/sync.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/sync.c b/fs/sync.c index 7cd005ea7639..228e17b5e9ee 100644 --- a/fs/sync.c +++ b/fs/sync.c @@ -64,7 +64,7 @@ int file_fsync(struct file *filp, struct dentry *dentry, int datasync) /* sync the superblock to buffers */ sb = inode->i_sb; lock_super(sb); - if (sb->s_op->write_super) + if (sb->s_dirt && sb->s_op->write_super) sb->s_op->write_super(sb); unlock_super(sb); -- cgit v1.2.3 From 05db67a4f2c14dab5bcaa46c7d4e9237bd11b37c Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 29 Apr 2008 00:59:47 -0700 Subject: remove ecryptfs_header_cache_0 Remove the no longer used ecryptfs_header_cache_0. Signed-off-by: Adrian Bunk Cc: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/crypto.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index a066e109ad9c..dcbbb2b4455d 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -1246,7 +1246,6 @@ ecryptfs_write_header_metadata(char *virt, (*written) = 6; } -struct kmem_cache *ecryptfs_header_cache_0; struct kmem_cache *ecryptfs_header_cache_1; struct kmem_cache *ecryptfs_header_cache_2; -- cgit v1.2.3 From 18d1dbf1d401e8f9d74cf1cf799fdb19cff150c6 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Tue, 29 Apr 2008 00:59:48 -0700 Subject: ecryptfs: replace remaining __FUNCTION__ occurrences __FUNCTION__ is gcc-specific, use __func__ Signed-off-by: Harvey Harrison Cc: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/crypto.c | 32 ++++++++++++++++---------------- fs/ecryptfs/ecryptfs_kernel.h | 2 +- fs/ecryptfs/inode.c | 2 +- fs/ecryptfs/main.c | 2 +- fs/ecryptfs/mmap.c | 18 +++++++++--------- fs/ecryptfs/read_write.c | 16 ++++++++-------- 6 files changed, 36 insertions(+), 36 deletions(-) (limited to 'fs') diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index dcbbb2b4455d..cd62d75b2cc0 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -119,21 +119,21 @@ static int ecryptfs_calculate_md5(char *dst, if (rc) { printk(KERN_ERR "%s: Error initializing crypto hash; rc = [%d]\n", - __FUNCTION__, rc); + __func__, rc); goto out; } rc = crypto_hash_update(&desc, &sg, len); if (rc) { printk(KERN_ERR "%s: Error updating crypto hash; rc = [%d]\n", - __FUNCTION__, rc); + __func__, rc); goto out; } rc = crypto_hash_final(&desc, dst); if (rc) { printk(KERN_ERR "%s: Error finalizing crypto hash; rc = [%d]\n", - __FUNCTION__, rc); + __func__, rc); goto out; } out: @@ -437,7 +437,7 @@ static int ecryptfs_encrypt_extent(struct page *enc_extent_page, if (rc < 0) { printk(KERN_ERR "%s: Error attempting to encrypt page with " "page->index = [%ld], extent_offset = [%ld]; " - "rc = [%d]\n", __FUNCTION__, page->index, extent_offset, + "rc = [%d]\n", __func__, page->index, extent_offset, rc); goto out; } @@ -487,7 +487,7 @@ int ecryptfs_encrypt_page(struct page *page) 0, PAGE_CACHE_SIZE); if (rc) printk(KERN_ERR "%s: Error attempting to copy " - "page at index [%ld]\n", __FUNCTION__, + "page at index [%ld]\n", __func__, page->index); goto out; } @@ -508,7 +508,7 @@ int ecryptfs_encrypt_page(struct page *page) extent_offset); if (rc) { printk(KERN_ERR "%s: Error encrypting extent; " - "rc = [%d]\n", __FUNCTION__, rc); + "rc = [%d]\n", __func__, rc); goto out; } ecryptfs_lower_offset_for_extent( @@ -569,7 +569,7 @@ static int ecryptfs_decrypt_extent(struct page *page, if (rc < 0) { printk(KERN_ERR "%s: Error attempting to decrypt to page with " "page->index = [%ld], extent_offset = [%ld]; " - "rc = [%d]\n", __FUNCTION__, page->index, extent_offset, + "rc = [%d]\n", __func__, page->index, extent_offset, rc); goto out; } @@ -622,7 +622,7 @@ int ecryptfs_decrypt_page(struct page *page) ecryptfs_inode); if (rc) printk(KERN_ERR "%s: Error attempting to copy " - "page at index [%ld]\n", __FUNCTION__, + "page at index [%ld]\n", __func__, page->index); goto out; } @@ -656,7 +656,7 @@ int ecryptfs_decrypt_page(struct page *page) extent_offset); if (rc) { printk(KERN_ERR "%s: Error encrypting extent; " - "rc = [%d]\n", __FUNCTION__, rc); + "rc = [%d]\n", __func__, rc); goto out; } } @@ -1215,7 +1215,7 @@ int ecryptfs_read_and_validate_header_region(char *data, ecryptfs_inode); if (rc) { printk(KERN_ERR "%s: Error reading header region; rc = [%d]\n", - __FUNCTION__, rc); + __func__, rc); goto out; } if (!contains_ecryptfs_marker(data + ECRYPTFS_FILE_SIZE_BYTES)) { @@ -1319,7 +1319,7 @@ ecryptfs_write_metadata_to_contents(struct ecryptfs_crypt_stat *crypt_stat, 0, crypt_stat->num_header_bytes_at_front); if (rc) printk(KERN_ERR "%s: Error attempting to write header " - "information to lower file; rc = [%d]\n", __FUNCTION__, + "information to lower file; rc = [%d]\n", __func__, rc); return rc; } @@ -1364,14 +1364,14 @@ int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry) } } else { printk(KERN_WARNING "%s: Encrypted flag not set\n", - __FUNCTION__); + __func__); rc = -EINVAL; goto out; } /* Released in this function */ virt = kzalloc(crypt_stat->num_header_bytes_at_front, GFP_KERNEL); if (!virt) { - printk(KERN_ERR "%s: Out of memory\n", __FUNCTION__); + printk(KERN_ERR "%s: Out of memory\n", __func__); rc = -ENOMEM; goto out; } @@ -1379,7 +1379,7 @@ int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry) ecryptfs_dentry); if (unlikely(rc)) { printk(KERN_ERR "%s: Error whilst writing headers; rc = [%d]\n", - __FUNCTION__, rc); + __func__, rc); goto out_free; } if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR) @@ -1390,7 +1390,7 @@ int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry) ecryptfs_dentry, virt); if (rc) { printk(KERN_ERR "%s: Error writing metadata out to lower file; " - "rc = [%d]\n", __FUNCTION__, rc); + "rc = [%d]\n", __func__, rc); goto out_free; } out_free: @@ -1584,7 +1584,7 @@ int ecryptfs_read_metadata(struct dentry *ecryptfs_dentry) if (!page_virt) { rc = -ENOMEM; printk(KERN_ERR "%s: Unable to allocate page_virt\n", - __FUNCTION__); + __func__); goto out; } rc = ecryptfs_read_lower(page_virt, 0, crypt_stat->extent_size, diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index 5007f788da01..342e8d37b421 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -500,7 +500,7 @@ ecryptfs_set_dentry_lower_mnt(struct dentry *dentry, struct vfsmount *lower_mnt) } #define ecryptfs_printk(type, fmt, arg...) \ - __ecryptfs_printk(type "%s: " fmt, __FUNCTION__, ## arg); + __ecryptfs_printk(type "%s: " fmt, __func__, ## arg); void __ecryptfs_printk(const char *fmt, ...); extern const struct file_operations ecryptfs_main_fops; diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index e23861152101..a7d5d7d2b2a7 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -121,7 +121,7 @@ ecryptfs_do_create(struct inode *directory_inode, ecryptfs_dentry, mode, nd); if (rc) { printk(KERN_ERR "%s: Failure to create dentry in lower fs; " - "rc = [%d]\n", __FUNCTION__, rc); + "rc = [%d]\n", __func__, rc); goto out_lock; } rc = ecryptfs_interpose(lower_dentry, ecryptfs_dentry, diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index d25ac9500a92..d603631601eb 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -219,7 +219,7 @@ int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry, if (rc) { printk(KERN_ERR "%s: Error attempting to initialize the " "persistent file for the dentry with name [%s]; " - "rc = [%d]\n", __FUNCTION__, dentry->d_name.name, rc); + "rc = [%d]\n", __func__, dentry->d_name.name, rc); goto out; } out: diff --git a/fs/ecryptfs/mmap.c b/fs/ecryptfs/mmap.c index 6df1debdccce..2b6fe1e6e8ba 100644 --- a/fs/ecryptfs/mmap.c +++ b/fs/ecryptfs/mmap.c @@ -153,7 +153,7 @@ ecryptfs_copy_up_encrypted_with_header(struct page *page, flush_dcache_page(page); if (rc) { printk(KERN_ERR "%s: Error reading xattr " - "region; rc = [%d]\n", __FUNCTION__, rc); + "region; rc = [%d]\n", __func__, rc); goto out; } } else { @@ -169,7 +169,7 @@ ecryptfs_copy_up_encrypted_with_header(struct page *page, if (rc) { printk(KERN_ERR "%s: Error attempting to read " "extent at offset [%lld] in the lower " - "file; rc = [%d]\n", __FUNCTION__, + "file; rc = [%d]\n", __func__, lower_offset, rc); goto out; } @@ -212,7 +212,7 @@ static int ecryptfs_readpage(struct file *file, struct page *page) "the encrypted content from the lower " "file whilst inserting the metadata " "from the xattr into the header; rc = " - "[%d]\n", __FUNCTION__, rc); + "[%d]\n", __func__, rc); goto out; } @@ -293,7 +293,7 @@ static int ecryptfs_prepare_write(struct file *file, struct page *page, if (rc) { printk(KERN_ERR "%s: Error attemping to read " "lower page segment; rc = [%d]\n", - __FUNCTION__, rc); + __func__, rc); ClearPageUptodate(page); goto out; } else @@ -308,7 +308,7 @@ static int ecryptfs_prepare_write(struct file *file, struct page *page, "from the lower file whilst " "inserting the metadata from " "the xattr into the header; rc " - "= [%d]\n", __FUNCTION__, rc); + "= [%d]\n", __func__, rc); ClearPageUptodate(page); goto out; } @@ -320,7 +320,7 @@ static int ecryptfs_prepare_write(struct file *file, struct page *page, if (rc) { printk(KERN_ERR "%s: Error reading " "page; rc = [%d]\n", - __FUNCTION__, rc); + __func__, rc); ClearPageUptodate(page); goto out; } @@ -331,7 +331,7 @@ static int ecryptfs_prepare_write(struct file *file, struct page *page, if (rc) { printk(KERN_ERR "%s: Error decrypting page " "at index [%ld]; rc = [%d]\n", - __FUNCTION__, page->index, rc); + __func__, page->index, rc); ClearPageUptodate(page); goto out; } @@ -348,7 +348,7 @@ static int ecryptfs_prepare_write(struct file *file, struct page *page, if (rc) { printk(KERN_ERR "%s: Error on attempt to " "truncate to (higher) offset [%lld];" - " rc = [%d]\n", __FUNCTION__, + " rc = [%d]\n", __func__, prev_page_end_size, rc); goto out; } @@ -389,7 +389,7 @@ static int ecryptfs_write_inode_size_to_header(struct inode *ecryptfs_inode) kfree(file_size_virt); if (rc) printk(KERN_ERR "%s: Error writing file size to header; " - "rc = [%d]\n", __FUNCTION__, rc); + "rc = [%d]\n", __func__, rc); out: return rc; } diff --git a/fs/ecryptfs/read_write.c b/fs/ecryptfs/read_write.c index 0c4928623bbc..ebf55150be56 100644 --- a/fs/ecryptfs/read_write.c +++ b/fs/ecryptfs/read_write.c @@ -55,7 +55,7 @@ int ecryptfs_write_lower(struct inode *ecryptfs_inode, char *data, set_fs(fs_save); if (octets_written < 0) { printk(KERN_ERR "%s: octets_written = [%td]; " - "expected [%td]\n", __FUNCTION__, octets_written, size); + "expected [%td]\n", __func__, octets_written, size); rc = -EINVAL; } mutex_unlock(&inode_info->lower_file_mutex); @@ -153,7 +153,7 @@ int ecryptfs_write(struct file *ecryptfs_file, char *data, loff_t offset, rc = PTR_ERR(ecryptfs_page); printk(KERN_ERR "%s: Error getting page at " "index [%ld] from eCryptfs inode " - "mapping; rc = [%d]\n", __FUNCTION__, + "mapping; rc = [%d]\n", __func__, ecryptfs_page_idx, rc); goto out; } @@ -165,7 +165,7 @@ int ecryptfs_write(struct file *ecryptfs_file, char *data, loff_t offset, if (rc) { printk(KERN_ERR "%s: Error decrypting " "page; rc = [%d]\n", - __FUNCTION__, rc); + __func__, rc); ClearPageUptodate(ecryptfs_page); page_cache_release(ecryptfs_page); goto out; @@ -202,7 +202,7 @@ int ecryptfs_write(struct file *ecryptfs_file, char *data, loff_t offset, page_cache_release(ecryptfs_page); if (rc) { printk(KERN_ERR "%s: Error encrypting " - "page; rc = [%d]\n", __FUNCTION__, rc); + "page; rc = [%d]\n", __func__, rc); goto out; } pos += num_bytes; @@ -254,7 +254,7 @@ int ecryptfs_read_lower(char *data, loff_t offset, size_t size, set_fs(fs_save); if (octets_read < 0) { printk(KERN_ERR "%s: octets_read = [%td]; " - "expected [%td]\n", __FUNCTION__, octets_read, size); + "expected [%td]\n", __func__, octets_read, size); rc = -EINVAL; } mutex_unlock(&inode_info->lower_file_mutex); @@ -327,7 +327,7 @@ int ecryptfs_read(char *data, loff_t offset, size_t size, printk(KERN_ERR "%s: Attempt to read data past the end of the " "file; offset = [%lld]; size = [%td]; " "ecryptfs_file_size = [%lld]\n", - __FUNCTION__, offset, size, ecryptfs_file_size); + __func__, offset, size, ecryptfs_file_size); goto out; } pos = offset; @@ -345,14 +345,14 @@ int ecryptfs_read(char *data, loff_t offset, size_t size, rc = PTR_ERR(ecryptfs_page); printk(KERN_ERR "%s: Error getting page at " "index [%ld] from eCryptfs inode " - "mapping; rc = [%d]\n", __FUNCTION__, + "mapping; rc = [%d]\n", __func__, ecryptfs_page_idx, rc); goto out; } rc = ecryptfs_decrypt_page(ecryptfs_page); if (rc) { printk(KERN_ERR "%s: Error decrypting " - "page; rc = [%d]\n", __FUNCTION__, rc); + "page; rc = [%d]\n", __func__, rc); ClearPageUptodate(ecryptfs_page); page_cache_release(ecryptfs_page); goto out; -- cgit v1.2.3 From 9c3580aa52195699065bc2d7242b1c7e3e6903fa Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 29 Apr 2008 00:59:48 -0700 Subject: ecryptfs: add missing lock around notify_change Callers of notify_change() need to hold i_mutex. Signed-off-by: Miklos Szeredi Cc: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/inode.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index a7d5d7d2b2a7..1623ebf25e51 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -908,7 +908,9 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia) if (ia->ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID)) ia->ia_valid &= ~ATTR_MODE; + mutex_lock(&lower_dentry->d_inode->i_mutex); rc = notify_change(lower_dentry, ia); + mutex_unlock(&lower_dentry->d_inode->i_mutex); out: fsstack_copy_attr_all(inode, lower_inode, NULL); return rc; -- cgit v1.2.3 From 8bf2debd5f7bf12d122124e34fec14af5b1e8ecf Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Tue, 29 Apr 2008 00:59:50 -0700 Subject: eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/miscdev.c | 580 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 580 insertions(+) create mode 100644 fs/ecryptfs/miscdev.c (limited to 'fs') diff --git a/fs/ecryptfs/miscdev.c b/fs/ecryptfs/miscdev.c new file mode 100644 index 000000000000..72dfec48ea22 --- /dev/null +++ b/fs/ecryptfs/miscdev.c @@ -0,0 +1,580 @@ +/** + * eCryptfs: Linux filesystem encryption layer + * + * Copyright (C) 2008 International Business Machines Corp. + * Author(s): Michael A. Halcrow + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version + * 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + */ + +#include +#include +#include +#include +#include +#include +#include +#include "ecryptfs_kernel.h" + +static atomic_t ecryptfs_num_miscdev_opens; + +/** + * ecryptfs_miscdev_poll + * @file: dev file (ignored) + * @pt: dev poll table (ignored) + * + * Returns the poll mask + */ +static unsigned int +ecryptfs_miscdev_poll(struct file *file, poll_table *pt) +{ + struct ecryptfs_daemon *daemon; + unsigned int mask = 0; + int rc; + + mutex_lock(&ecryptfs_daemon_hash_mux); + /* TODO: Just use file->private_data? */ + rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid); + BUG_ON(rc || !daemon); + mutex_lock(&daemon->mux); + mutex_unlock(&ecryptfs_daemon_hash_mux); + if (daemon->flags & ECRYPTFS_DAEMON_ZOMBIE) { + printk(KERN_WARNING "%s: Attempt to poll on zombified " + "daemon\n", __func__); + goto out_unlock_daemon; + } + if (daemon->flags & ECRYPTFS_DAEMON_IN_READ) + goto out_unlock_daemon; + if (daemon->flags & ECRYPTFS_DAEMON_IN_POLL) + goto out_unlock_daemon; + daemon->flags |= ECRYPTFS_DAEMON_IN_POLL; + mutex_unlock(&daemon->mux); + poll_wait(file, &daemon->wait, pt); + mutex_lock(&daemon->mux); + if (!list_empty(&daemon->msg_ctx_out_queue)) + mask |= POLLIN | POLLRDNORM; +out_unlock_daemon: + daemon->flags &= ~ECRYPTFS_DAEMON_IN_POLL; + mutex_unlock(&daemon->mux); + return mask; +} + +/** + * ecryptfs_miscdev_open + * @inode: inode of miscdev handle (ignored) + * @file: file for miscdev handle (ignored) + * + * Returns zero on success; non-zero otherwise + */ +static int +ecryptfs_miscdev_open(struct inode *inode, struct file *file) +{ + struct ecryptfs_daemon *daemon = NULL; + int rc; + + mutex_lock(&ecryptfs_daemon_hash_mux); + rc = try_module_get(THIS_MODULE); + if (rc == 0) { + rc = -EIO; + printk(KERN_ERR "%s: Error attempting to increment module use " + "count; rc = [%d]\n", __func__, rc); + goto out_unlock_daemon_list; + } + rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid); + if (rc || !daemon) { + rc = ecryptfs_spawn_daemon(&daemon, current->euid, + current->pid); + if (rc) { + printk(KERN_ERR "%s: Error attempting to spawn daemon; " + "rc = [%d]\n", __func__, rc); + goto out_module_put_unlock_daemon_list; + } + } + mutex_lock(&daemon->mux); + if (daemon->pid != current->pid) { + rc = -EINVAL; + printk(KERN_ERR "%s: pid [%d] has registered with euid [%d], " + "but pid [%d] has attempted to open the handle " + "instead\n", __func__, daemon->pid, daemon->euid, + current->pid); + goto out_unlock_daemon; + } + if (daemon->flags & ECRYPTFS_DAEMON_MISCDEV_OPEN) { + rc = -EBUSY; + printk(KERN_ERR "%s: Miscellaneous device handle may only be " + "opened once per daemon; pid [%d] already has this " + "handle open\n", __func__, daemon->pid); + goto out_unlock_daemon; + } + daemon->flags |= ECRYPTFS_DAEMON_MISCDEV_OPEN; + atomic_inc(&ecryptfs_num_miscdev_opens); +out_unlock_daemon: + mutex_unlock(&daemon->mux); +out_module_put_unlock_daemon_list: + if (rc) + module_put(THIS_MODULE); +out_unlock_daemon_list: + mutex_unlock(&ecryptfs_daemon_hash_mux); + return rc; +} + +/** + * ecryptfs_miscdev_release + * @inode: inode of fs/ecryptfs/euid handle (ignored) + * @file: file for fs/ecryptfs/euid handle (ignored) + * + * This keeps the daemon registered until the daemon sends another + * ioctl to fs/ecryptfs/ctl or until the kernel module unregisters. + * + * Returns zero on success; non-zero otherwise + */ +static int +ecryptfs_miscdev_release(struct inode *inode, struct file *file) +{ + struct ecryptfs_daemon *daemon = NULL; + int rc; + + mutex_lock(&ecryptfs_daemon_hash_mux); + rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid); + BUG_ON(rc || !daemon); + mutex_lock(&daemon->mux); + BUG_ON(daemon->pid != current->pid); + BUG_ON(!(daemon->flags & ECRYPTFS_DAEMON_MISCDEV_OPEN)); + daemon->flags &= ~ECRYPTFS_DAEMON_MISCDEV_OPEN; + atomic_dec(&ecryptfs_num_miscdev_opens); + mutex_unlock(&daemon->mux); + rc = ecryptfs_exorcise_daemon(daemon); + if (rc) { + printk(KERN_CRIT "%s: Fatal error whilst attempting to " + "shut down daemon; rc = [%d]. Please report this " + "bug.\n", __func__, rc); + BUG(); + } + module_put(THIS_MODULE); + mutex_unlock(&ecryptfs_daemon_hash_mux); + return rc; +} + +/** + * ecryptfs_send_miscdev + * @data: Data to send to daemon; may be NULL + * @data_size: Amount of data to send to daemon + * @msg_ctx: Message context, which is used to handle the reply. If + * this is NULL, then we do not expect a reply. + * @msg_type: Type of message + * @msg_flags: Flags for message + * @daemon: eCryptfs daemon object + * + * Add msg_ctx to queue and then, if it exists, notify the blocked + * miscdevess about the data being available. Must be called with + * ecryptfs_daemon_hash_mux held. + * + * Returns zero on success; non-zero otherwise + */ +int ecryptfs_send_miscdev(char *data, size_t data_size, + struct ecryptfs_msg_ctx *msg_ctx, u8 msg_type, + u16 msg_flags, struct ecryptfs_daemon *daemon) +{ + int rc = 0; + + mutex_lock(&msg_ctx->mux); + if (data) { + msg_ctx->msg = kmalloc((sizeof(*msg_ctx->msg) + data_size), + GFP_KERNEL); + if (!msg_ctx->msg) { + rc = -ENOMEM; + printk(KERN_ERR "%s: Out of memory whilst attempting " + "to kmalloc(%d, GFP_KERNEL)\n", __func__, + (sizeof(*msg_ctx->msg) + data_size)); + goto out_unlock; + } + } else + msg_ctx->msg = NULL; + msg_ctx->msg->index = msg_ctx->index; + msg_ctx->msg->data_len = data_size; + msg_ctx->type = msg_type; + if (data) { + memcpy(msg_ctx->msg->data, data, data_size); + msg_ctx->msg_size = (sizeof(*msg_ctx->msg) + data_size); + } else + msg_ctx->msg_size = 0; + mutex_lock(&daemon->mux); + list_add_tail(&msg_ctx->daemon_out_list, &daemon->msg_ctx_out_queue); + daemon->num_queued_msg_ctx++; + wake_up_interruptible(&daemon->wait); + mutex_unlock(&daemon->mux); +out_unlock: + mutex_unlock(&msg_ctx->mux); + return rc; +} + +/** + * ecryptfs_miscdev_read - format and send message from queue + * @file: fs/ecryptfs/euid miscdevfs handle (ignored) + * @buf: User buffer into which to copy the next message on the daemon queue + * @count: Amount of space available in @buf + * @ppos: Offset in file (ignored) + * + * Pulls the most recent message from the daemon queue, formats it for + * being sent via a miscdevfs handle, and copies it into @buf + * + * Returns the number of bytes copied into the user buffer + */ +static int +ecryptfs_miscdev_read(struct file *file, char __user *buf, size_t count, + loff_t *ppos) +{ + struct ecryptfs_daemon *daemon; + struct ecryptfs_msg_ctx *msg_ctx; + size_t packet_length_size; + u32 counter_nbo; + char packet_length[3]; + size_t i; + size_t total_length; + int rc; + + mutex_lock(&ecryptfs_daemon_hash_mux); + /* TODO: Just use file->private_data? */ + rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid); + BUG_ON(rc || !daemon); + mutex_lock(&daemon->mux); + if (daemon->flags & ECRYPTFS_DAEMON_ZOMBIE) { + rc = 0; + printk(KERN_WARNING "%s: Attempt to read from zombified " + "daemon\n", __func__); + goto out_unlock_daemon; + } + if (daemon->flags & ECRYPTFS_DAEMON_IN_READ) { + rc = 0; + goto out_unlock_daemon; + } + /* This daemon will not go away so long as this flag is set */ + daemon->flags |= ECRYPTFS_DAEMON_IN_READ; + mutex_unlock(&ecryptfs_daemon_hash_mux); +check_list: + if (list_empty(&daemon->msg_ctx_out_queue)) { + mutex_unlock(&daemon->mux); + rc = wait_event_interruptible( + daemon->wait, !list_empty(&daemon->msg_ctx_out_queue)); + mutex_lock(&daemon->mux); + if (rc < 0) { + rc = 0; + goto out_unlock_daemon; + } + } + if (daemon->flags & ECRYPTFS_DAEMON_ZOMBIE) { + rc = 0; + goto out_unlock_daemon; + } + if (list_empty(&daemon->msg_ctx_out_queue)) { + /* Something else jumped in since the + * wait_event_interruptable() and removed the + * message from the queue; try again */ + goto check_list; + } + BUG_ON(current->euid != daemon->euid); + BUG_ON(current->pid != daemon->pid); + msg_ctx = list_first_entry(&daemon->msg_ctx_out_queue, + struct ecryptfs_msg_ctx, daemon_out_list); + BUG_ON(!msg_ctx); + mutex_lock(&msg_ctx->mux); + if (msg_ctx->msg) { + rc = ecryptfs_write_packet_length(packet_length, + msg_ctx->msg_size, + &packet_length_size); + if (rc) { + rc = 0; + printk(KERN_WARNING "%s: Error writing packet length; " + "rc = [%d]\n", __func__, rc); + goto out_unlock_msg_ctx; + } + } else { + packet_length_size = 0; + msg_ctx->msg_size = 0; + } + /* miscdevfs packet format: + * Octet 0: Type + * Octets 1-4: network byte order msg_ctx->counter + * Octets 5-N0: Size of struct ecryptfs_message to follow + * Octets N0-N1: struct ecryptfs_message (including data) + * + * Octets 5-N1 not written if the packet type does not + * include a message */ + total_length = (1 + 4 + packet_length_size + msg_ctx->msg_size); + if (count < total_length) { + rc = 0; + printk(KERN_WARNING "%s: Only given user buffer of " + "size [%Zd], but we need [%Zd] to read the " + "pending message\n", __func__, count, total_length); + goto out_unlock_msg_ctx; + } + i = 0; + buf[i++] = msg_ctx->type; + counter_nbo = cpu_to_be32(msg_ctx->counter); + memcpy(&buf[i], (char *)&counter_nbo, 4); + i += 4; + if (msg_ctx->msg) { + memcpy(&buf[i], packet_length, packet_length_size); + i += packet_length_size; + rc = copy_to_user(&buf[i], msg_ctx->msg, msg_ctx->msg_size); + if (rc) { + printk(KERN_ERR "%s: copy_to_user returned error " + "[%d]\n", __func__, rc); + goto out_unlock_msg_ctx; + } + i += msg_ctx->msg_size; + } + rc = i; + list_del(&msg_ctx->daemon_out_list); + kfree(msg_ctx->msg); + msg_ctx->msg = NULL; + /* We do not expect a reply from the userspace daemon for any + * message type other than ECRYPTFS_MSG_REQUEST */ + if (msg_ctx->type != ECRYPTFS_MSG_REQUEST) + ecryptfs_msg_ctx_alloc_to_free(msg_ctx); +out_unlock_msg_ctx: + mutex_unlock(&msg_ctx->mux); +out_unlock_daemon: + daemon->flags &= ~ECRYPTFS_DAEMON_IN_READ; + mutex_unlock(&daemon->mux); + return rc; +} + +/** + * ecryptfs_miscdev_helo + * @euid: effective user id of miscdevess sending helo packet + * @pid: miscdevess id of miscdevess sending helo packet + * + * Returns zero on success; non-zero otherwise + */ +static int ecryptfs_miscdev_helo(uid_t uid, pid_t pid) +{ + int rc; + + rc = ecryptfs_process_helo(ECRYPTFS_TRANSPORT_MISCDEV, uid, pid); + if (rc) + printk(KERN_WARNING "Error processing HELO; rc = [%d]\n", rc); + return rc; +} + +/** + * ecryptfs_miscdev_quit + * @euid: effective user id of miscdevess sending quit packet + * @pid: miscdevess id of miscdevess sending quit packet + * + * Returns zero on success; non-zero otherwise + */ +static int ecryptfs_miscdev_quit(uid_t euid, pid_t pid) +{ + int rc; + + rc = ecryptfs_process_quit(euid, pid); + if (rc) + printk(KERN_WARNING + "Error processing QUIT message; rc = [%d]\n", rc); + return rc; +} + +/** + * ecryptfs_miscdev_response - miscdevess response to message previously sent to daemon + * @data: Bytes comprising struct ecryptfs_message + * @data_size: sizeof(struct ecryptfs_message) + data len + * @euid: Effective user id of miscdevess sending the miscdev response + * @pid: Miscdevess id of miscdevess sending the miscdev response + * @seq: Sequence number for miscdev response packet + * + * Returns zero on success; non-zero otherwise + */ +static int ecryptfs_miscdev_response(char *data, size_t data_size, + uid_t euid, pid_t pid, u32 seq) +{ + struct ecryptfs_message *msg = (struct ecryptfs_message *)data; + int rc; + + if ((sizeof(*msg) + msg->data_len) != data_size) { + printk(KERN_WARNING "%s: (sizeof(*msg) + msg->data_len) = " + "[%Zd]; data_size = [%Zd]. Invalid packet.\n", __func__, + (sizeof(*msg) + msg->data_len), data_size); + rc = -EINVAL; + goto out; + } + rc = ecryptfs_process_response(msg, euid, pid, seq); + if (rc) + printk(KERN_ERR + "Error processing response message; rc = [%d]\n", rc); +out: + return rc; +} + +/** + * ecryptfs_miscdev_write - handle write to daemon miscdev handle + * @file: File for misc dev handle (ignored) + * @buf: Buffer containing user data + * @count: Amount of data in @buf + * @ppos: Pointer to offset in file (ignored) + * + * miscdevfs packet format: + * Octet 0: Type + * Octets 1-4: network byte order msg_ctx->counter (0's for non-response) + * Octets 5-N0: Size of struct ecryptfs_message to follow + * Octets N0-N1: struct ecryptfs_message (including data) + * + * Returns the number of bytes read from @buf + */ +static ssize_t +ecryptfs_miscdev_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + u32 counter_nbo, seq; + size_t packet_size, packet_size_length, i; + ssize_t sz = 0; + char *data; + int rc; + + if (count == 0) + goto out; + data = kmalloc(count, GFP_KERNEL); + if (!data) { + printk(KERN_ERR "%s: Out of memory whilst attempting to " + "kmalloc([%Zd], GFP_KERNEL)\n", __func__, count); + goto out; + } + rc = copy_from_user(data, buf, count); + if (rc) { + printk(KERN_ERR "%s: copy_from_user returned error [%d]\n", + __func__, rc); + goto out_free; + } + sz = count; + i = 0; + switch (data[i++]) { + case ECRYPTFS_MSG_RESPONSE: + if (count < (1 + 4 + 1 + sizeof(struct ecryptfs_message))) { + printk(KERN_WARNING "%s: Minimum acceptable packet " + "size is [%Zd], but amount of data written is " + "only [%Zd]. Discarding response packet.\n", + __func__, + (1 + 4 + 1 + sizeof(struct ecryptfs_message)), + count); + goto out_free; + } + memcpy((char *)&counter_nbo, &data[i], 4); + seq = be32_to_cpu(counter_nbo); + i += 4; + rc = ecryptfs_parse_packet_length(&data[i], &packet_size, + &packet_size_length); + if (rc) { + printk(KERN_WARNING "%s: Error parsing packet length; " + "rc = [%d]\n", __func__, rc); + goto out_free; + } + i += packet_size_length; + if ((1 + 4 + packet_size_length + packet_size) != count) { + printk(KERN_WARNING "%s: (1 + packet_size_length([%Zd])" + " + packet_size([%Zd]))([%Zd]) != " + "count([%Zd]). Invalid packet format.\n", + __func__, packet_size_length, packet_size, + (1 + packet_size_length + packet_size), count); + goto out_free; + } + rc = ecryptfs_miscdev_response(&data[i], packet_size, + current->euid, + current->pid, seq); + if (rc) + printk(KERN_WARNING "%s: Failed to deliver miscdev " + "response to requesting operation; rc = [%d]\n", + __func__, rc); + break; + case ECRYPTFS_MSG_HELO: + rc = ecryptfs_miscdev_helo(current->euid, current->pid); + if (rc) { + printk(KERN_ERR "%s: Error attempting to process " + "helo from pid [%d]; rc = [%d]\n", __func__, + current->pid, rc); + goto out_free; + } + break; + case ECRYPTFS_MSG_QUIT: + rc = ecryptfs_miscdev_quit(current->euid, current->pid); + if (rc) { + printk(KERN_ERR "%s: Error attempting to process " + "quit from pid [%d]; rc = [%d]\n", __func__, + current->pid, rc); + goto out_free; + } + break; + default: + ecryptfs_printk(KERN_WARNING, "Dropping miscdev " + "message of unrecognized type [%d]\n", + data[0]); + break; + } +out_free: + kfree(data); +out: + return sz; +} + + +static const struct file_operations ecryptfs_miscdev_fops = { + .open = ecryptfs_miscdev_open, + .poll = ecryptfs_miscdev_poll, + .read = ecryptfs_miscdev_read, + .write = ecryptfs_miscdev_write, + .release = ecryptfs_miscdev_release, +}; + +static struct miscdevice ecryptfs_miscdev = { + .minor = MISC_DYNAMIC_MINOR, + .name = "ecryptfs", + .fops = &ecryptfs_miscdev_fops +}; + +/** + * ecryptfs_init_ecryptfs_miscdev + * + * Messages sent to the userspace daemon from the kernel are placed on + * a queue associated with the daemon. The next read against the + * miscdev handle by that daemon will return the oldest message placed + * on the message queue for the daemon. + * + * Returns zero on success; non-zero otherwise + */ +int ecryptfs_init_ecryptfs_miscdev(void) +{ + int rc; + + atomic_set(&ecryptfs_num_miscdev_opens, 0); + mutex_lock(&ecryptfs_daemon_hash_mux); + rc = misc_register(&ecryptfs_miscdev); + if (rc) + printk(KERN_ERR "%s: Failed to register miscellaneous device " + "for communications with userspace daemons; rc = [%d]\n", + __func__, rc); + mutex_unlock(&ecryptfs_daemon_hash_mux); + return rc; +} + +/** + * ecryptfs_destroy_ecryptfs_miscdev + * + * All of the daemons must be exorcised prior to calling this + * function. + */ +void ecryptfs_destroy_ecryptfs_miscdev(void) +{ + BUG_ON(atomic_read(&ecryptfs_num_miscdev_opens) != 0); + misc_deregister(&ecryptfs_miscdev); +} -- cgit v1.2.3 From f66e883eb6186bc43a79581b67aff7d1a69d0ff1 Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Tue, 29 Apr 2008 00:59:51 -0700 Subject: eCryptfs: integrate eCryptfs device handle into the module. Update the versioning information. Make the message types generic. Add an outgoing message queue to the daemon struct. Make the functions to parse and write the packet lengths available to the rest of the module. Add functions to create and destroy the daemon structs. Clean up some of the comments and make the code a little more consistent with itself. [akpm@linux-foundation.org: printk fixes] Signed-off-by: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/Makefile | 2 +- fs/ecryptfs/ecryptfs_kernel.h | 79 +++++-- fs/ecryptfs/keystore.c | 89 ++++---- fs/ecryptfs/messaging.c | 479 ++++++++++++++++++++++++++++-------------- fs/ecryptfs/miscdev.c | 4 +- fs/ecryptfs/netlink.c | 8 +- 6 files changed, 435 insertions(+), 226 deletions(-) (limited to 'fs') diff --git a/fs/ecryptfs/Makefile b/fs/ecryptfs/Makefile index 768857015516..1e34a7fd4884 100644 --- a/fs/ecryptfs/Makefile +++ b/fs/ecryptfs/Makefile @@ -4,4 +4,4 @@ obj-$(CONFIG_ECRYPT_FS) += ecryptfs.o -ecryptfs-objs := dentry.o file.o inode.o main.o super.o mmap.o read_write.o crypto.o keystore.o messaging.o netlink.o debug.o +ecryptfs-objs := dentry.o file.o inode.o main.o super.o mmap.o read_write.o crypto.o keystore.o messaging.o netlink.o miscdev.o debug.o diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index 342e8d37b421..72e117706a68 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -4,7 +4,7 @@ * * Copyright (C) 1997-2003 Erez Zadok * Copyright (C) 2001-2003 Stony Brook University - * Copyright (C) 2004-2007 International Business Machines Corp. + * Copyright (C) 2004-2008 International Business Machines Corp. * Author(s): Michael A. Halcrow * Trevor S. Highland * Tyler Hicks @@ -49,11 +49,13 @@ #define ECRYPTFS_VERSIONING_POLICY 0x00000008 #define ECRYPTFS_VERSIONING_XATTR 0x00000010 #define ECRYPTFS_VERSIONING_MULTKEY 0x00000020 +#define ECRYPTFS_VERSIONING_DEVMISC 0x00000040 #define ECRYPTFS_VERSIONING_MASK (ECRYPTFS_VERSIONING_PASSPHRASE \ | ECRYPTFS_VERSIONING_PLAINTEXT_PASSTHROUGH \ | ECRYPTFS_VERSIONING_PUBKEY \ | ECRYPTFS_VERSIONING_XATTR \ - | ECRYPTFS_VERSIONING_MULTKEY) + | ECRYPTFS_VERSIONING_MULTKEY \ + | ECRYPTFS_VERSIONING_DEVMISC) #define ECRYPTFS_MAX_PASSWORD_LENGTH 64 #define ECRYPTFS_MAX_PASSPHRASE_BYTES ECRYPTFS_MAX_PASSWORD_LENGTH #define ECRYPTFS_SALT_SIZE 8 @@ -73,17 +75,14 @@ #define ECRYPTFS_DEFAULT_MSG_CTX_ELEMS 32 #define ECRYPTFS_DEFAULT_SEND_TIMEOUT HZ #define ECRYPTFS_MAX_MSG_CTX_TTL (HZ*3) -#define ECRYPTFS_NLMSG_HELO 100 -#define ECRYPTFS_NLMSG_QUIT 101 -#define ECRYPTFS_NLMSG_REQUEST 102 -#define ECRYPTFS_NLMSG_RESPONSE 103 #define ECRYPTFS_MAX_PKI_NAME_BYTES 16 #define ECRYPTFS_DEFAULT_NUM_USERS 4 #define ECRYPTFS_MAX_NUM_USERS 32768 #define ECRYPTFS_TRANSPORT_NETLINK 0 #define ECRYPTFS_TRANSPORT_CONNECTOR 1 #define ECRYPTFS_TRANSPORT_RELAYFS 2 -#define ECRYPTFS_DEFAULT_TRANSPORT ECRYPTFS_TRANSPORT_NETLINK +#define ECRYPTFS_TRANSPORT_MISCDEV 3 +#define ECRYPTFS_DEFAULT_TRANSPORT ECRYPTFS_TRANSPORT_MISCDEV #define ECRYPTFS_XATTR_NAME "user.ecryptfs" #define RFC2440_CIPHER_DES3_EDE 0x02 @@ -366,32 +365,62 @@ struct ecryptfs_auth_tok_list_item { }; struct ecryptfs_message { + /* Can never be greater than ecryptfs_message_buf_len */ + /* Used to find the parent msg_ctx */ + /* Inherits from msg_ctx->index */ u32 index; u32 data_len; u8 data[]; }; struct ecryptfs_msg_ctx { -#define ECRYPTFS_MSG_CTX_STATE_FREE 0x0001 -#define ECRYPTFS_MSG_CTX_STATE_PENDING 0x0002 -#define ECRYPTFS_MSG_CTX_STATE_DONE 0x0003 - u32 state; - unsigned int index; - unsigned int counter; +#define ECRYPTFS_MSG_CTX_STATE_FREE 0x01 +#define ECRYPTFS_MSG_CTX_STATE_PENDING 0x02 +#define ECRYPTFS_MSG_CTX_STATE_DONE 0x03 +#define ECRYPTFS_MSG_CTX_STATE_NO_REPLY 0x04 + u8 state; +#define ECRYPTFS_MSG_HELO 100 +#define ECRYPTFS_MSG_QUIT 101 +#define ECRYPTFS_MSG_REQUEST 102 +#define ECRYPTFS_MSG_RESPONSE 103 + u8 type; + u32 index; + /* Counter converts to a sequence number. Each message sent + * out for which we expect a response has an associated + * sequence number. The response must have the same sequence + * number as the counter for the msg_stc for the message to be + * valid. */ + u32 counter; + size_t msg_size; struct ecryptfs_message *msg; struct task_struct *task; struct list_head node; + struct list_head daemon_out_list; struct mutex mux; }; extern unsigned int ecryptfs_transport; -struct ecryptfs_daemon_id { +struct ecryptfs_daemon; + +struct ecryptfs_daemon { +#define ECRYPTFS_DAEMON_IN_READ 0x00000001 +#define ECRYPTFS_DAEMON_IN_POLL 0x00000002 +#define ECRYPTFS_DAEMON_ZOMBIE 0x00000004 +#define ECRYPTFS_DAEMON_MISCDEV_OPEN 0x00000008 + u32 flags; + u32 num_queued_msg_ctx; pid_t pid; - uid_t uid; - struct hlist_node id_chain; + uid_t euid; + struct task_struct *task; + struct mutex mux; + struct list_head msg_ctx_out_queue; + wait_queue_head_t wait; + struct hlist_node euid_chain; }; +extern struct mutex ecryptfs_daemon_hash_mux; + static inline struct ecryptfs_file_info * ecryptfs_file_to_private(struct file *file) { @@ -593,13 +622,13 @@ int ecryptfs_init_messaging(unsigned int transport); void ecryptfs_release_messaging(unsigned int transport); int ecryptfs_send_netlink(char *data, int data_len, - struct ecryptfs_msg_ctx *msg_ctx, u16 msg_type, + struct ecryptfs_msg_ctx *msg_ctx, u8 msg_type, u16 msg_flags, pid_t daemon_pid); int ecryptfs_init_netlink(void); void ecryptfs_release_netlink(void); int ecryptfs_send_connector(char *data, int data_len, - struct ecryptfs_msg_ctx *msg_ctx, u16 msg_type, + struct ecryptfs_msg_ctx *msg_ctx, u8 msg_type, u16 msg_flags, pid_t daemon_pid); int ecryptfs_init_connector(void); void ecryptfs_release_connector(void); @@ -642,5 +671,19 @@ int ecryptfs_read_lower_page_segment(struct page *page_for_ecryptfs, size_t offset_in_page, size_t size, struct inode *ecryptfs_inode); struct page *ecryptfs_get_locked_page(struct file *file, loff_t index); +int ecryptfs_exorcise_daemon(struct ecryptfs_daemon *daemon); +int ecryptfs_find_daemon_by_euid(struct ecryptfs_daemon **daemon, uid_t euid); +int ecryptfs_parse_packet_length(unsigned char *data, size_t *size, + size_t *length_size); +int ecryptfs_write_packet_length(char *dest, size_t size, + size_t *packet_size_length); +int ecryptfs_init_ecryptfs_miscdev(void); +void ecryptfs_destroy_ecryptfs_miscdev(void); +int ecryptfs_send_miscdev(char *data, size_t data_size, + struct ecryptfs_msg_ctx *msg_ctx, u8 msg_type, + u16 msg_flags, struct ecryptfs_daemon *daemon); +void ecryptfs_msg_ctx_alloc_to_free(struct ecryptfs_msg_ctx *msg_ctx); +int +ecryptfs_spawn_daemon(struct ecryptfs_daemon **daemon, uid_t euid, pid_t pid); #endif /* #ifndef ECRYPTFS_KERNEL_H */ diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index 682b1b2482c2..e82b457180be 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -65,7 +65,7 @@ static int process_request_key_err(long err_code) } /** - * parse_packet_length + * ecryptfs_parse_packet_length * @data: Pointer to memory containing length at offset * @size: This function writes the decoded size to this memory * address; zero on error @@ -73,8 +73,8 @@ static int process_request_key_err(long err_code) * * Returns zero on success; non-zero on error */ -static int parse_packet_length(unsigned char *data, size_t *size, - size_t *length_size) +int ecryptfs_parse_packet_length(unsigned char *data, size_t *size, + size_t *length_size) { int rc = 0; @@ -105,7 +105,7 @@ out: } /** - * write_packet_length + * ecryptfs_write_packet_length * @dest: The byte array target into which to write the length. Must * have at least 5 bytes allocated. * @size: The length to write. @@ -114,8 +114,8 @@ out: * * Returns zero on success; non-zero on error. */ -static int write_packet_length(char *dest, size_t size, - size_t *packet_size_length) +int ecryptfs_write_packet_length(char *dest, size_t size, + size_t *packet_size_length) { int rc = 0; @@ -162,8 +162,8 @@ write_tag_64_packet(char *signature, struct ecryptfs_session_key *session_key, goto out; } message[i++] = ECRYPTFS_TAG_64_PACKET_TYPE; - rc = write_packet_length(&message[i], ECRYPTFS_SIG_SIZE_HEX, - &packet_size_len); + rc = ecryptfs_write_packet_length(&message[i], ECRYPTFS_SIG_SIZE_HEX, + &packet_size_len); if (rc) { ecryptfs_printk(KERN_ERR, "Error generating tag 64 packet " "header; cannot generate packet length\n"); @@ -172,8 +172,9 @@ write_tag_64_packet(char *signature, struct ecryptfs_session_key *session_key, i += packet_size_len; memcpy(&message[i], signature, ECRYPTFS_SIG_SIZE_HEX); i += ECRYPTFS_SIG_SIZE_HEX; - rc = write_packet_length(&message[i], session_key->encrypted_key_size, - &packet_size_len); + rc = ecryptfs_write_packet_length(&message[i], + session_key->encrypted_key_size, + &packet_size_len); if (rc) { ecryptfs_printk(KERN_ERR, "Error generating tag 64 packet " "header; cannot generate packet length\n"); @@ -225,7 +226,7 @@ parse_tag_65_packet(struct ecryptfs_session_key *session_key, u8 *cipher_code, rc = -EIO; goto out; } - rc = parse_packet_length(&data[i], &m_size, &data_len); + rc = ecryptfs_parse_packet_length(&data[i], &m_size, &data_len); if (rc) { ecryptfs_printk(KERN_WARNING, "Error parsing packet length; " "rc = [%d]\n", rc); @@ -304,8 +305,8 @@ write_tag_66_packet(char *signature, u8 cipher_code, goto out; } message[i++] = ECRYPTFS_TAG_66_PACKET_TYPE; - rc = write_packet_length(&message[i], ECRYPTFS_SIG_SIZE_HEX, - &packet_size_len); + rc = ecryptfs_write_packet_length(&message[i], ECRYPTFS_SIG_SIZE_HEX, + &packet_size_len); if (rc) { ecryptfs_printk(KERN_ERR, "Error generating tag 66 packet " "header; cannot generate packet length\n"); @@ -315,8 +316,8 @@ write_tag_66_packet(char *signature, u8 cipher_code, memcpy(&message[i], signature, ECRYPTFS_SIG_SIZE_HEX); i += ECRYPTFS_SIG_SIZE_HEX; /* The encrypted key includes 1 byte cipher code and 2 byte checksum */ - rc = write_packet_length(&message[i], crypt_stat->key_size + 3, - &packet_size_len); + rc = ecryptfs_write_packet_length(&message[i], crypt_stat->key_size + 3, + &packet_size_len); if (rc) { ecryptfs_printk(KERN_ERR, "Error generating tag 66 packet " "header; cannot generate packet length\n"); @@ -357,20 +358,25 @@ parse_tag_67_packet(struct ecryptfs_key_record *key_rec, /* verify that everything through the encrypted FEK size is present */ if (message_len < 4) { rc = -EIO; + printk(KERN_ERR "%s: message_len is [%Zd]; minimum acceptable " + "message length is [%d]\n", __func__, message_len, 4); goto out; } if (data[i++] != ECRYPTFS_TAG_67_PACKET_TYPE) { - ecryptfs_printk(KERN_ERR, "Type should be ECRYPTFS_TAG_67\n"); rc = -EIO; + printk(KERN_ERR "%s: Type should be ECRYPTFS_TAG_67\n", + __func__); goto out; } if (data[i++]) { - ecryptfs_printk(KERN_ERR, "Status indicator has non zero value" - " [%d]\n", data[i-1]); rc = -EIO; + printk(KERN_ERR "%s: Status indicator has non zero " + "value [%d]\n", __func__, data[i-1]); + goto out; } - rc = parse_packet_length(&data[i], &key_rec->enc_key_size, &data_len); + rc = ecryptfs_parse_packet_length(&data[i], &key_rec->enc_key_size, + &data_len); if (rc) { ecryptfs_printk(KERN_WARNING, "Error parsing packet length; " "rc = [%d]\n", rc); @@ -378,17 +384,17 @@ parse_tag_67_packet(struct ecryptfs_key_record *key_rec, } i += data_len; if (message_len < (i + key_rec->enc_key_size)) { - ecryptfs_printk(KERN_ERR, "message_len [%d]; max len is [%d]\n", - message_len, (i + key_rec->enc_key_size)); rc = -EIO; + printk(KERN_ERR "%s: message_len [%Zd]; max len is [%Zd]\n", + __func__, message_len, (i + key_rec->enc_key_size)); goto out; } if (key_rec->enc_key_size > ECRYPTFS_MAX_ENCRYPTED_KEY_BYTES) { - ecryptfs_printk(KERN_ERR, "Encrypted key_size [%d] larger than " - "the maximum key size [%d]\n", - key_rec->enc_key_size, - ECRYPTFS_MAX_ENCRYPTED_KEY_BYTES); rc = -EIO; + printk(KERN_ERR "%s: Encrypted key_size [%Zd] larger than " + "the maximum key size [%d]\n", __func__, + key_rec->enc_key_size, + ECRYPTFS_MAX_ENCRYPTED_KEY_BYTES); goto out; } memcpy(key_rec->enc_key, &data[i], key_rec->enc_key_size); @@ -445,7 +451,7 @@ decrypt_pki_encrypted_session_key(struct ecryptfs_auth_tok *auth_tok, rc = write_tag_64_packet(auth_tok_sig, &(auth_tok->session_key), &netlink_message, &netlink_message_length); if (rc) { - ecryptfs_printk(KERN_ERR, "Failed to write tag 64 packet"); + ecryptfs_printk(KERN_ERR, "Failed to write tag 64 packet\n"); goto out; } rc = ecryptfs_send_message(ecryptfs_transport, netlink_message, @@ -570,8 +576,8 @@ parse_tag_1_packet(struct ecryptfs_crypt_stat *crypt_stat, goto out; } (*new_auth_tok) = &auth_tok_list_item->auth_tok; - rc = parse_packet_length(&data[(*packet_size)], &body_size, - &length_size); + rc = ecryptfs_parse_packet_length(&data[(*packet_size)], &body_size, + &length_size); if (rc) { printk(KERN_WARNING "Error parsing packet length; " "rc = [%d]\n", rc); @@ -704,8 +710,8 @@ parse_tag_3_packet(struct ecryptfs_crypt_stat *crypt_stat, goto out; } (*new_auth_tok) = &auth_tok_list_item->auth_tok; - rc = parse_packet_length(&data[(*packet_size)], &body_size, - &length_size); + rc = ecryptfs_parse_packet_length(&data[(*packet_size)], &body_size, + &length_size); if (rc) { printk(KERN_WARNING "Error parsing packet length; rc = [%d]\n", rc); @@ -852,8 +858,8 @@ parse_tag_11_packet(unsigned char *data, unsigned char *contents, rc = -EINVAL; goto out; } - rc = parse_packet_length(&data[(*packet_size)], &body_size, - &length_size); + rc = ecryptfs_parse_packet_length(&data[(*packet_size)], &body_size, + &length_size); if (rc) { printk(KERN_WARNING "Invalid tag 11 packet format\n"); goto out; @@ -1405,8 +1411,8 @@ write_tag_1_packet(char *dest, size_t *remaining_bytes, auth_tok->token.private_key.key_size; rc = pki_encrypt_session_key(auth_tok, crypt_stat, key_rec); if (rc) { - ecryptfs_printk(KERN_ERR, "Failed to encrypt session key " - "via a pki"); + printk(KERN_ERR "Failed to encrypt session key via a key " + "module; rc = [%d]\n", rc); goto out; } if (ecryptfs_verbosity > 0) { @@ -1430,8 +1436,9 @@ encrypted_session_key_set: goto out; } dest[(*packet_size)++] = ECRYPTFS_TAG_1_PACKET_TYPE; - rc = write_packet_length(&dest[(*packet_size)], (max_packet_size - 4), - &packet_size_length); + rc = ecryptfs_write_packet_length(&dest[(*packet_size)], + (max_packet_size - 4), + &packet_size_length); if (rc) { ecryptfs_printk(KERN_ERR, "Error generating tag 1 packet " "header; cannot generate packet length\n"); @@ -1489,8 +1496,9 @@ write_tag_11_packet(char *dest, size_t *remaining_bytes, char *contents, goto out; } dest[(*packet_length)++] = ECRYPTFS_TAG_11_PACKET_TYPE; - rc = write_packet_length(&dest[(*packet_length)], - (max_packet_size - 4), &packet_size_length); + rc = ecryptfs_write_packet_length(&dest[(*packet_length)], + (max_packet_size - 4), + &packet_size_length); if (rc) { printk(KERN_ERR "Error generating tag 11 packet header; cannot " "generate packet length. rc = [%d]\n", rc); @@ -1682,8 +1690,9 @@ encrypted_session_key_set: dest[(*packet_size)++] = ECRYPTFS_TAG_3_PACKET_TYPE; /* Chop off the Tag 3 identifier(1) and Tag 3 packet size(3) * to get the number of octets in the actual Tag 3 packet */ - rc = write_packet_length(&dest[(*packet_size)], (max_packet_size - 4), - &packet_size_length); + rc = ecryptfs_write_packet_length(&dest[(*packet_size)], + (max_packet_size - 4), + &packet_size_length); if (rc) { printk(KERN_ERR "Error generating tag 3 packet header; cannot " "generate packet length. rc = [%d]\n", rc); diff --git a/fs/ecryptfs/messaging.c b/fs/ecryptfs/messaging.c index 9cc2aec27b0d..c6038bd60897 100644 --- a/fs/ecryptfs/messaging.c +++ b/fs/ecryptfs/messaging.c @@ -1,7 +1,7 @@ /** * eCryptfs: Linux filesystem encryption layer * - * Copyright (C) 2004-2006 International Business Machines Corp. + * Copyright (C) 2004-2008 International Business Machines Corp. * Author(s): Michael A. Halcrow * Tyler Hicks * @@ -26,13 +26,13 @@ static LIST_HEAD(ecryptfs_msg_ctx_free_list); static LIST_HEAD(ecryptfs_msg_ctx_alloc_list); static struct mutex ecryptfs_msg_ctx_lists_mux; -static struct hlist_head *ecryptfs_daemon_id_hash; -static struct mutex ecryptfs_daemon_id_hash_mux; +static struct hlist_head *ecryptfs_daemon_hash; +struct mutex ecryptfs_daemon_hash_mux; static int ecryptfs_hash_buckets; #define ecryptfs_uid_hash(uid) \ hash_long((unsigned long)uid, ecryptfs_hash_buckets) -static unsigned int ecryptfs_msg_counter; +static u32 ecryptfs_msg_counter; static struct ecryptfs_msg_ctx *ecryptfs_msg_ctx_arr; /** @@ -40,9 +40,10 @@ static struct ecryptfs_msg_ctx *ecryptfs_msg_ctx_arr; * @msg_ctx: The context that was acquired from the free list * * Acquires a context element from the free list and locks the mutex - * on the context. Returns zero on success; non-zero on error or upon - * failure to acquire a free context element. Be sure to lock the - * list mutex before calling. + * on the context. Sets the msg_ctx task to current. Returns zero on + * success; non-zero on error or upon failure to acquire a free + * context element. Must be called with ecryptfs_msg_ctx_lists_mux + * held. */ static int ecryptfs_acquire_free_msg_ctx(struct ecryptfs_msg_ctx **msg_ctx) { @@ -50,11 +51,11 @@ static int ecryptfs_acquire_free_msg_ctx(struct ecryptfs_msg_ctx **msg_ctx) int rc; if (list_empty(&ecryptfs_msg_ctx_free_list)) { - ecryptfs_printk(KERN_WARNING, "The eCryptfs free " - "context list is empty. It may be helpful to " - "specify the ecryptfs_message_buf_len " - "parameter to be greater than the current " - "value of [%d]\n", ecryptfs_message_buf_len); + printk(KERN_WARNING "%s: The eCryptfs free " + "context list is empty. It may be helpful to " + "specify the ecryptfs_message_buf_len " + "parameter to be greater than the current " + "value of [%d]\n", __func__, ecryptfs_message_buf_len); rc = -ENOMEM; goto out; } @@ -75,8 +76,7 @@ out: * ecryptfs_msg_ctx_free_to_alloc * @msg_ctx: The context to move from the free list to the alloc list * - * Be sure to lock the list mutex and the context mutex before - * calling. + * Must be called with ecryptfs_msg_ctx_lists_mux held. */ static void ecryptfs_msg_ctx_free_to_alloc(struct ecryptfs_msg_ctx *msg_ctx) { @@ -89,36 +89,37 @@ static void ecryptfs_msg_ctx_free_to_alloc(struct ecryptfs_msg_ctx *msg_ctx) * ecryptfs_msg_ctx_alloc_to_free * @msg_ctx: The context to move from the alloc list to the free list * - * Be sure to lock the list mutex and the context mutex before - * calling. + * Must be called with ecryptfs_msg_ctx_lists_mux held. */ -static void ecryptfs_msg_ctx_alloc_to_free(struct ecryptfs_msg_ctx *msg_ctx) +void ecryptfs_msg_ctx_alloc_to_free(struct ecryptfs_msg_ctx *msg_ctx) { list_move(&(msg_ctx->node), &ecryptfs_msg_ctx_free_list); if (msg_ctx->msg) kfree(msg_ctx->msg); + msg_ctx->msg = NULL; msg_ctx->state = ECRYPTFS_MSG_CTX_STATE_FREE; } /** - * ecryptfs_find_daemon_id - * @uid: The user id which maps to the desired daemon id - * @id: If return value is zero, points to the desired daemon id - * pointer + * ecryptfs_find_daemon_by_euid + * @euid: The effective user id which maps to the desired daemon id + * @daemon: If return value is zero, points to the desired daemon pointer * - * Search the hash list for the given user id. Returns zero if the - * user id exists in the list; non-zero otherwise. The daemon id hash - * mutex should be held before calling this function. + * Must be called with ecryptfs_daemon_hash_mux held. + * + * Search the hash list for the given user id. + * + * Returns zero if the user id exists in the list; non-zero otherwise. */ -static int ecryptfs_find_daemon_id(uid_t uid, struct ecryptfs_daemon_id **id) +int ecryptfs_find_daemon_by_euid(struct ecryptfs_daemon **daemon, uid_t euid) { struct hlist_node *elem; int rc; - hlist_for_each_entry(*id, elem, - &ecryptfs_daemon_id_hash[ecryptfs_uid_hash(uid)], - id_chain) { - if ((*id)->uid == uid) { + hlist_for_each_entry(*daemon, elem, + &ecryptfs_daemon_hash[ecryptfs_uid_hash(euid)], + euid_chain) { + if ((*daemon)->euid == euid) { rc = 0; goto out; } @@ -128,181 +129,291 @@ out: return rc; } -static int ecryptfs_send_raw_message(unsigned int transport, u16 msg_type, - pid_t pid) +static int +ecryptfs_send_message_locked(unsigned int transport, char *data, int data_len, + u8 msg_type, struct ecryptfs_msg_ctx **msg_ctx); + +/** + * ecryptfs_send_raw_message + * @transport: Transport type + * @msg_type: Message type + * @daemon: Daemon struct for recipient of message + * + * A raw message is one that does not include an ecryptfs_message + * struct. It simply has a type. + * + * Must be called with ecryptfs_daemon_hash_mux held. + * + * Returns zero on success; non-zero otherwise + */ +static int ecryptfs_send_raw_message(unsigned int transport, u8 msg_type, + struct ecryptfs_daemon *daemon) { + struct ecryptfs_msg_ctx *msg_ctx; int rc; switch(transport) { case ECRYPTFS_TRANSPORT_NETLINK: - rc = ecryptfs_send_netlink(NULL, 0, NULL, msg_type, 0, pid); + rc = ecryptfs_send_netlink(NULL, 0, NULL, msg_type, 0, + daemon->pid); + break; + case ECRYPTFS_TRANSPORT_MISCDEV: + rc = ecryptfs_send_message_locked(transport, NULL, 0, msg_type, + &msg_ctx); + if (rc) { + printk(KERN_ERR "%s: Error whilst attempting to send " + "message via procfs; rc = [%d]\n", __func__, rc); + goto out; + } + /* Raw messages are logically context-free (e.g., no + * reply is expected), so we set the state of the + * ecryptfs_msg_ctx object to indicate that it should + * be freed as soon as the transport sends out the message. */ + mutex_lock(&msg_ctx->mux); + msg_ctx->state = ECRYPTFS_MSG_CTX_STATE_NO_REPLY; + mutex_unlock(&msg_ctx->mux); break; case ECRYPTFS_TRANSPORT_CONNECTOR: case ECRYPTFS_TRANSPORT_RELAYFS: default: rc = -ENOSYS; } +out: + return rc; +} + +/** + * ecryptfs_spawn_daemon - Create and initialize a new daemon struct + * @daemon: Pointer to set to newly allocated daemon struct + * @euid: Effective user id for the daemon + * @pid: Process id for the daemon + * + * Must be called ceremoniously while in possession of + * ecryptfs_sacred_daemon_hash_mux + * + * Returns zero on success; non-zero otherwise + */ +int +ecryptfs_spawn_daemon(struct ecryptfs_daemon **daemon, uid_t euid, pid_t pid) +{ + int rc = 0; + + (*daemon) = kzalloc(sizeof(**daemon), GFP_KERNEL); + if (!(*daemon)) { + rc = -ENOMEM; + printk(KERN_ERR "%s: Failed to allocate [%Zd] bytes of " + "GFP_KERNEL memory\n", __func__, sizeof(**daemon)); + goto out; + } + (*daemon)->euid = euid; + (*daemon)->pid = pid; + (*daemon)->task = current; + mutex_init(&(*daemon)->mux); + INIT_LIST_HEAD(&(*daemon)->msg_ctx_out_queue); + init_waitqueue_head(&(*daemon)->wait); + (*daemon)->num_queued_msg_ctx = 0; + hlist_add_head(&(*daemon)->euid_chain, + &ecryptfs_daemon_hash[ecryptfs_uid_hash(euid)]); +out: return rc; } /** * ecryptfs_process_helo * @transport: The underlying transport (netlink, etc.) - * @uid: The user ID owner of the message + * @euid: The user ID owner of the message * @pid: The process ID for the userspace program that sent the * message * - * Adds the uid and pid values to the daemon id hash. If a uid + * Adds the euid and pid values to the daemon euid hash. If an euid * already has a daemon pid registered, the daemon will be - * unregistered before the new daemon id is put into the hash list. - * Returns zero after adding a new daemon id to the hash list; + * unregistered before the new daemon is put into the hash list. + * Returns zero after adding a new daemon to the hash list; * non-zero otherwise. */ -int ecryptfs_process_helo(unsigned int transport, uid_t uid, pid_t pid) +int ecryptfs_process_helo(unsigned int transport, uid_t euid, pid_t pid) { - struct ecryptfs_daemon_id *new_id; - struct ecryptfs_daemon_id *old_id; + struct ecryptfs_daemon *new_daemon; + struct ecryptfs_daemon *old_daemon; int rc; - mutex_lock(&ecryptfs_daemon_id_hash_mux); - new_id = kmalloc(sizeof(*new_id), GFP_KERNEL); - if (!new_id) { - rc = -ENOMEM; - ecryptfs_printk(KERN_ERR, "Failed to allocate memory; unable " - "to register daemon [%d] for user [%d]\n", - pid, uid); - goto unlock; - } - if (!ecryptfs_find_daemon_id(uid, &old_id)) { + mutex_lock(&ecryptfs_daemon_hash_mux); + rc = ecryptfs_find_daemon_by_euid(&old_daemon, euid); + if (rc != 0) { printk(KERN_WARNING "Received request from user [%d] " "to register daemon [%d]; unregistering daemon " - "[%d]\n", uid, pid, old_id->pid); - hlist_del(&old_id->id_chain); - rc = ecryptfs_send_raw_message(transport, ECRYPTFS_NLMSG_QUIT, - old_id->pid); + "[%d]\n", euid, pid, old_daemon->pid); + rc = ecryptfs_send_raw_message(transport, ECRYPTFS_MSG_QUIT, + old_daemon); if (rc) printk(KERN_WARNING "Failed to send QUIT " "message to daemon [%d]; rc = [%d]\n", - old_id->pid, rc); - kfree(old_id); + old_daemon->pid, rc); + hlist_del(&old_daemon->euid_chain); + kfree(old_daemon); } - new_id->uid = uid; - new_id->pid = pid; - hlist_add_head(&new_id->id_chain, - &ecryptfs_daemon_id_hash[ecryptfs_uid_hash(uid)]); - rc = 0; -unlock: - mutex_unlock(&ecryptfs_daemon_id_hash_mux); + rc = ecryptfs_spawn_daemon(&new_daemon, euid, pid); + if (rc) + printk(KERN_ERR "%s: The gods are displeased with this attempt " + "to create a new daemon object for euid [%d]; pid [%d]; " + "rc = [%d]\n", __func__, euid, pid, rc); + mutex_unlock(&ecryptfs_daemon_hash_mux); + return rc; +} + +/** + * ecryptfs_exorcise_daemon - Destroy the daemon struct + * + * Must be called ceremoniously while in possession of + * ecryptfs_daemon_hash_mux and the daemon's own mux. + */ +int ecryptfs_exorcise_daemon(struct ecryptfs_daemon *daemon) +{ + struct ecryptfs_msg_ctx *msg_ctx, *msg_ctx_tmp; + int rc = 0; + + mutex_lock(&daemon->mux); + if ((daemon->flags & ECRYPTFS_DAEMON_IN_READ) + || (daemon->flags & ECRYPTFS_DAEMON_IN_POLL)) { + rc = -EBUSY; + printk(KERN_WARNING "%s: Attempt to destroy daemon with pid " + "[%d], but it is in the midst of a read or a poll\n", + __func__, daemon->pid); + mutex_unlock(&daemon->mux); + goto out; + } + list_for_each_entry_safe(msg_ctx, msg_ctx_tmp, + &daemon->msg_ctx_out_queue, daemon_out_list) { + list_del(&msg_ctx->daemon_out_list); + daemon->num_queued_msg_ctx--; + printk(KERN_WARNING "%s: Warning: dropping message that is in " + "the out queue of a dying daemon\n", __func__); + ecryptfs_msg_ctx_alloc_to_free(msg_ctx); + } + hlist_del(&daemon->euid_chain); + if (daemon->task) + wake_up_process(daemon->task); + mutex_unlock(&daemon->mux); + memset(daemon, 0, sizeof(*daemon)); + kfree(daemon); +out: return rc; } /** * ecryptfs_process_quit - * @uid: The user ID owner of the message + * @euid: The user ID owner of the message * @pid: The process ID for the userspace program that sent the * message * - * Deletes the corresponding daemon id for the given uid and pid, if + * Deletes the corresponding daemon for the given euid and pid, if * it is the registered that is requesting the deletion. Returns zero - * after deleting the desired daemon id; non-zero otherwise. + * after deleting the desired daemon; non-zero otherwise. */ -int ecryptfs_process_quit(uid_t uid, pid_t pid) +int ecryptfs_process_quit(uid_t euid, pid_t pid) { - struct ecryptfs_daemon_id *id; + struct ecryptfs_daemon *daemon; int rc; - mutex_lock(&ecryptfs_daemon_id_hash_mux); - if (ecryptfs_find_daemon_id(uid, &id)) { - rc = -EINVAL; - ecryptfs_printk(KERN_ERR, "Received request from user [%d] to " - "unregister unrecognized daemon [%d]\n", uid, - pid); - goto unlock; - } - if (id->pid != pid) { + mutex_lock(&ecryptfs_daemon_hash_mux); + rc = ecryptfs_find_daemon_by_euid(&daemon, euid); + if (rc || !daemon) { rc = -EINVAL; - ecryptfs_printk(KERN_WARNING, "Received request from user [%d] " - "with pid [%d] to unregister daemon [%d]\n", - uid, pid, id->pid); - goto unlock; + printk(KERN_ERR "Received request from user [%d] to " + "unregister unrecognized daemon [%d]\n", euid, pid); + goto out_unlock; } - hlist_del(&id->id_chain); - kfree(id); - rc = 0; -unlock: - mutex_unlock(&ecryptfs_daemon_id_hash_mux); + rc = ecryptfs_exorcise_daemon(daemon); +out_unlock: + mutex_unlock(&ecryptfs_daemon_hash_mux); return rc; } /** * ecryptfs_process_reponse * @msg: The ecryptfs message received; the caller should sanity check - * msg->data_len + * msg->data_len and free the memory * @pid: The process ID of the userspace application that sent the * message - * @seq: The sequence number of the message + * @seq: The sequence number of the message; must match the sequence + * number for the existing message context waiting for this + * response + * + * Processes a response message after sending an operation request to + * userspace. Some other process is awaiting this response. Before + * sending out its first communications, the other process allocated a + * msg_ctx from the ecryptfs_msg_ctx_arr at a particular index. The + * response message contains this index so that we can copy over the + * response message into the msg_ctx that the process holds a + * reference to. The other process is going to wake up, check to see + * that msg_ctx->state == ECRYPTFS_MSG_CTX_STATE_DONE, and then + * proceed to read off and process the response message. Returns zero + * upon delivery to desired context element; non-zero upon delivery + * failure or error. * - * Processes a response message after sending a operation request to - * userspace. Returns zero upon delivery to desired context element; - * non-zero upon delivery failure or error. + * Returns zero on success; non-zero otherwise */ -int ecryptfs_process_response(struct ecryptfs_message *msg, uid_t uid, +int ecryptfs_process_response(struct ecryptfs_message *msg, uid_t euid, pid_t pid, u32 seq) { - struct ecryptfs_daemon_id *id; + struct ecryptfs_daemon *daemon; struct ecryptfs_msg_ctx *msg_ctx; - int msg_size; + size_t msg_size; int rc; if (msg->index >= ecryptfs_message_buf_len) { rc = -EINVAL; - ecryptfs_printk(KERN_ERR, "Attempt to reference " - "context buffer at index [%d]; maximum " - "allowable is [%d]\n", msg->index, - (ecryptfs_message_buf_len - 1)); + printk(KERN_ERR "%s: Attempt to reference " + "context buffer at index [%d]; maximum " + "allowable is [%d]\n", __func__, msg->index, + (ecryptfs_message_buf_len - 1)); goto out; } msg_ctx = &ecryptfs_msg_ctx_arr[msg->index]; mutex_lock(&msg_ctx->mux); - if (ecryptfs_find_daemon_id(msg_ctx->task->euid, &id)) { + mutex_lock(&ecryptfs_daemon_hash_mux); + rc = ecryptfs_find_daemon_by_euid(&daemon, msg_ctx->task->euid); + mutex_unlock(&ecryptfs_daemon_hash_mux); + if (rc) { rc = -EBADMSG; - ecryptfs_printk(KERN_WARNING, "User [%d] received a " - "message response from process [%d] but does " - "not have a registered daemon\n", - msg_ctx->task->euid, pid); + printk(KERN_WARNING "%s: User [%d] received a " + "message response from process [%d] but does " + "not have a registered daemon\n", __func__, + msg_ctx->task->euid, pid); goto wake_up; } - if (msg_ctx->task->euid != uid) { + if (msg_ctx->task->euid != euid) { rc = -EBADMSG; - ecryptfs_printk(KERN_WARNING, "Received message from user " - "[%d]; expected message from user [%d]\n", - uid, msg_ctx->task->euid); + printk(KERN_WARNING "%s: Received message from user " + "[%d]; expected message from user [%d]\n", __func__, + euid, msg_ctx->task->euid); goto unlock; } - if (id->pid != pid) { + if (daemon->pid != pid) { rc = -EBADMSG; - ecryptfs_printk(KERN_ERR, "User [%d] received a " - "message response from an unrecognized " - "process [%d]\n", msg_ctx->task->euid, pid); + printk(KERN_ERR "%s: User [%d] sent a message response " + "from an unrecognized process [%d]\n", + __func__, msg_ctx->task->euid, pid); goto unlock; } if (msg_ctx->state != ECRYPTFS_MSG_CTX_STATE_PENDING) { rc = -EINVAL; - ecryptfs_printk(KERN_WARNING, "Desired context element is not " - "pending a response\n"); + printk(KERN_WARNING "%s: Desired context element is not " + "pending a response\n", __func__); goto unlock; } else if (msg_ctx->counter != seq) { rc = -EINVAL; - ecryptfs_printk(KERN_WARNING, "Invalid message sequence; " - "expected [%d]; received [%d]\n", - msg_ctx->counter, seq); + printk(KERN_WARNING "%s: Invalid message sequence; " + "expected [%d]; received [%d]\n", __func__, + msg_ctx->counter, seq); goto unlock; } - msg_size = sizeof(*msg) + msg->data_len; + msg_size = (sizeof(*msg) + msg->data_len); msg_ctx->msg = kmalloc(msg_size, GFP_KERNEL); if (!msg_ctx->msg) { rc = -ENOMEM; - ecryptfs_printk(KERN_ERR, "Failed to allocate memory\n"); + printk(KERN_ERR "%s: Failed to allocate [%Zd] bytes of " + "GFP_KERNEL memory\n", __func__, msg_size); goto unlock; } memcpy(msg_ctx->msg, msg, msg_size); @@ -317,34 +428,37 @@ out: } /** - * ecryptfs_send_message + * ecryptfs_send_message_locked * @transport: The transport over which to send the message (i.e., * netlink) * @data: The data to send * @data_len: The length of data * @msg_ctx: The message context allocated for the send + * + * Must be called with ecryptfs_daemon_hash_mux held. + * + * Returns zero on success; non-zero otherwise */ -int ecryptfs_send_message(unsigned int transport, char *data, int data_len, - struct ecryptfs_msg_ctx **msg_ctx) +static int +ecryptfs_send_message_locked(unsigned int transport, char *data, int data_len, + u8 msg_type, struct ecryptfs_msg_ctx **msg_ctx) { - struct ecryptfs_daemon_id *id; + struct ecryptfs_daemon *daemon; int rc; - mutex_lock(&ecryptfs_daemon_id_hash_mux); - if (ecryptfs_find_daemon_id(current->euid, &id)) { - mutex_unlock(&ecryptfs_daemon_id_hash_mux); + rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid); + if (rc || !daemon) { rc = -ENOTCONN; - ecryptfs_printk(KERN_ERR, "User [%d] does not have a daemon " - "registered\n", current->euid); + printk(KERN_ERR "%s: User [%d] does not have a daemon " + "registered\n", __func__, current->euid); goto out; } - mutex_unlock(&ecryptfs_daemon_id_hash_mux); mutex_lock(&ecryptfs_msg_ctx_lists_mux); rc = ecryptfs_acquire_free_msg_ctx(msg_ctx); if (rc) { mutex_unlock(&ecryptfs_msg_ctx_lists_mux); - ecryptfs_printk(KERN_WARNING, "Could not claim a free " - "context element\n"); + printk(KERN_WARNING "%s: Could not claim a free " + "context element\n", __func__); goto out; } ecryptfs_msg_ctx_free_to_alloc(*msg_ctx); @@ -352,22 +466,49 @@ int ecryptfs_send_message(unsigned int transport, char *data, int data_len, mutex_unlock(&ecryptfs_msg_ctx_lists_mux); switch (transport) { case ECRYPTFS_TRANSPORT_NETLINK: - rc = ecryptfs_send_netlink(data, data_len, *msg_ctx, - ECRYPTFS_NLMSG_REQUEST, 0, id->pid); + rc = ecryptfs_send_netlink(data, data_len, *msg_ctx, msg_type, + 0, daemon->pid); + break; + case ECRYPTFS_TRANSPORT_MISCDEV: + rc = ecryptfs_send_miscdev(data, data_len, *msg_ctx, msg_type, + 0, daemon); break; case ECRYPTFS_TRANSPORT_CONNECTOR: case ECRYPTFS_TRANSPORT_RELAYFS: default: rc = -ENOSYS; } - if (rc) { - printk(KERN_ERR "Error attempting to send message to userspace " - "daemon; rc = [%d]\n", rc); - } + if (rc) + printk(KERN_ERR "%s: Error attempting to send message to " + "userspace daemon; rc = [%d]\n", __func__, rc); out: return rc; } +/** + * ecryptfs_send_message + * @transport: The transport over which to send the message (i.e., + * netlink) + * @data: The data to send + * @data_len: The length of data + * @msg_ctx: The message context allocated for the send + * + * Grabs ecryptfs_daemon_hash_mux. + * + * Returns zero on success; non-zero otherwise + */ +int ecryptfs_send_message(unsigned int transport, char *data, int data_len, + struct ecryptfs_msg_ctx **msg_ctx) +{ + int rc; + + mutex_lock(&ecryptfs_daemon_hash_mux); + rc = ecryptfs_send_message_locked(transport, data, data_len, + ECRYPTFS_MSG_REQUEST, msg_ctx); + mutex_unlock(&ecryptfs_daemon_hash_mux); + return rc; +} + /** * ecryptfs_wait_for_response * @msg_ctx: The context that was assigned when sending a message @@ -377,7 +518,7 @@ out: * of time exceeds ecryptfs_message_wait_timeout. If zero is * returned, msg will point to a valid message from userspace; a * non-zero value is returned upon failure to receive a message or an - * error occurs. + * error occurs. Callee must free @msg on success. */ int ecryptfs_wait_for_response(struct ecryptfs_msg_ctx *msg_ctx, struct ecryptfs_message **msg) @@ -413,32 +554,32 @@ int ecryptfs_init_messaging(unsigned int transport) if (ecryptfs_number_of_users > ECRYPTFS_MAX_NUM_USERS) { ecryptfs_number_of_users = ECRYPTFS_MAX_NUM_USERS; - ecryptfs_printk(KERN_WARNING, "Specified number of users is " - "too large, defaulting to [%d] users\n", - ecryptfs_number_of_users); + printk(KERN_WARNING "%s: Specified number of users is " + "too large, defaulting to [%d] users\n", __func__, + ecryptfs_number_of_users); } - mutex_init(&ecryptfs_daemon_id_hash_mux); - mutex_lock(&ecryptfs_daemon_id_hash_mux); + mutex_init(&ecryptfs_daemon_hash_mux); + mutex_lock(&ecryptfs_daemon_hash_mux); ecryptfs_hash_buckets = 1; while (ecryptfs_number_of_users >> ecryptfs_hash_buckets) ecryptfs_hash_buckets++; - ecryptfs_daemon_id_hash = kmalloc(sizeof(struct hlist_head) - * ecryptfs_hash_buckets, GFP_KERNEL); - if (!ecryptfs_daemon_id_hash) { + ecryptfs_daemon_hash = kmalloc((sizeof(struct hlist_head) + * ecryptfs_hash_buckets), GFP_KERNEL); + if (!ecryptfs_daemon_hash) { rc = -ENOMEM; - ecryptfs_printk(KERN_ERR, "Failed to allocate memory\n"); - mutex_unlock(&ecryptfs_daemon_id_hash_mux); + printk(KERN_ERR "%s: Failed to allocate memory\n", __func__); + mutex_unlock(&ecryptfs_daemon_hash_mux); goto out; } for (i = 0; i < ecryptfs_hash_buckets; i++) - INIT_HLIST_HEAD(&ecryptfs_daemon_id_hash[i]); - mutex_unlock(&ecryptfs_daemon_id_hash_mux); - + INIT_HLIST_HEAD(&ecryptfs_daemon_hash[i]); + mutex_unlock(&ecryptfs_daemon_hash_mux); ecryptfs_msg_ctx_arr = kmalloc((sizeof(struct ecryptfs_msg_ctx) - * ecryptfs_message_buf_len), GFP_KERNEL); + * ecryptfs_message_buf_len), + GFP_KERNEL); if (!ecryptfs_msg_ctx_arr) { rc = -ENOMEM; - ecryptfs_printk(KERN_ERR, "Failed to allocate memory\n"); + printk(KERN_ERR "%s: Failed to allocate memory\n", __func__); goto out; } mutex_init(&ecryptfs_msg_ctx_lists_mux); @@ -446,6 +587,7 @@ int ecryptfs_init_messaging(unsigned int transport) ecryptfs_msg_counter = 0; for (i = 0; i < ecryptfs_message_buf_len; i++) { INIT_LIST_HEAD(&ecryptfs_msg_ctx_arr[i].node); + INIT_LIST_HEAD(&ecryptfs_msg_ctx_arr[i].daemon_out_list); mutex_init(&ecryptfs_msg_ctx_arr[i].mux); mutex_lock(&ecryptfs_msg_ctx_arr[i].mux); ecryptfs_msg_ctx_arr[i].index = i; @@ -464,6 +606,11 @@ int ecryptfs_init_messaging(unsigned int transport) if (rc) ecryptfs_release_messaging(transport); break; + case ECRYPTFS_TRANSPORT_MISCDEV: + rc = ecryptfs_init_ecryptfs_miscdev(); + if (rc) + ecryptfs_release_messaging(transport); + break; case ECRYPTFS_TRANSPORT_CONNECTOR: case ECRYPTFS_TRANSPORT_RELAYFS: default: @@ -488,27 +635,37 @@ void ecryptfs_release_messaging(unsigned int transport) kfree(ecryptfs_msg_ctx_arr); mutex_unlock(&ecryptfs_msg_ctx_lists_mux); } - if (ecryptfs_daemon_id_hash) { + if (ecryptfs_daemon_hash) { struct hlist_node *elem; - struct ecryptfs_daemon_id *id; + struct ecryptfs_daemon *daemon; int i; - mutex_lock(&ecryptfs_daemon_id_hash_mux); + mutex_lock(&ecryptfs_daemon_hash_mux); for (i = 0; i < ecryptfs_hash_buckets; i++) { - hlist_for_each_entry(id, elem, - &ecryptfs_daemon_id_hash[i], - id_chain) { - hlist_del(elem); - kfree(id); + int rc; + + hlist_for_each_entry(daemon, elem, + &ecryptfs_daemon_hash[i], + euid_chain) { + rc = ecryptfs_exorcise_daemon(daemon); + if (rc) + printk(KERN_ERR "%s: Error whilst " + "attempting to destroy daemon; " + "rc = [%d]. Dazed and confused, " + "but trying to continue.\n", + __func__, rc); } } - kfree(ecryptfs_daemon_id_hash); - mutex_unlock(&ecryptfs_daemon_id_hash_mux); + kfree(ecryptfs_daemon_hash); + mutex_unlock(&ecryptfs_daemon_hash_mux); } switch(transport) { case ECRYPTFS_TRANSPORT_NETLINK: ecryptfs_release_netlink(); break; + case ECRYPTFS_TRANSPORT_MISCDEV: + ecryptfs_destroy_ecryptfs_miscdev(); + break; case ECRYPTFS_TRANSPORT_CONNECTOR: case ECRYPTFS_TRANSPORT_RELAYFS: default: diff --git a/fs/ecryptfs/miscdev.c b/fs/ecryptfs/miscdev.c index 72dfec48ea22..0c559731ae34 100644 --- a/fs/ecryptfs/miscdev.c +++ b/fs/ecryptfs/miscdev.c @@ -196,7 +196,7 @@ int ecryptfs_send_miscdev(char *data, size_t data_size, if (!msg_ctx->msg) { rc = -ENOMEM; printk(KERN_ERR "%s: Out of memory whilst attempting " - "to kmalloc(%d, GFP_KERNEL)\n", __func__, + "to kmalloc(%Zd, GFP_KERNEL)\n", __func__, (sizeof(*msg_ctx->msg) + data_size)); goto out_unlock; } @@ -232,7 +232,7 @@ out_unlock: * * Returns the number of bytes copied into the user buffer */ -static int +static ssize_t ecryptfs_miscdev_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { diff --git a/fs/ecryptfs/netlink.c b/fs/ecryptfs/netlink.c index f638a698dc52..eb70f69d705d 100644 --- a/fs/ecryptfs/netlink.c +++ b/fs/ecryptfs/netlink.c @@ -44,7 +44,7 @@ static struct sock *ecryptfs_nl_sock; * upon sending the message; non-zero upon error. */ int ecryptfs_send_netlink(char *data, int data_len, - struct ecryptfs_msg_ctx *msg_ctx, u16 msg_type, + struct ecryptfs_msg_ctx *msg_ctx, u8 msg_type, u16 msg_flags, pid_t daemon_pid) { struct sk_buff *skb; @@ -176,20 +176,20 @@ static void ecryptfs_receive_nl_message(struct sk_buff *skb) goto free; } switch (nlh->nlmsg_type) { - case ECRYPTFS_NLMSG_RESPONSE: + case ECRYPTFS_MSG_RESPONSE: if (ecryptfs_process_nl_response(skb)) { ecryptfs_printk(KERN_WARNING, "Failed to " "deliver netlink response to " "requesting operation\n"); } break; - case ECRYPTFS_NLMSG_HELO: + case ECRYPTFS_MSG_HELO: if (ecryptfs_process_nl_helo(skb)) { ecryptfs_printk(KERN_WARNING, "Failed to " "fulfill HELO request\n"); } break; - case ECRYPTFS_NLMSG_QUIT: + case ECRYPTFS_MSG_QUIT: if (ecryptfs_process_nl_quit(skb)) { ecryptfs_printk(KERN_WARNING, "Failed to " "fulfill QUIT request\n"); -- cgit v1.2.3 From 6a3fd92e73fffd9e583650c56ad9558afe51dc5c Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Tue, 29 Apr 2008 00:59:52 -0700 Subject: eCryptfs: make key module subsystem respect namespaces Make eCryptfs key module subsystem respect namespaces. Since I will be removing the netlink interface in a future patch, I just made changes to the netlink.c code so that it will not break the build. With my recent patches, the kernel module currently defaults to the device handle interface rather than the netlink interface. [akpm@linux-foundation.org: export free_user_ns()] Signed-off-by: Michael Halcrow Acked-by: Serge Hallyn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/ecryptfs_kernel.h | 25 ++++++++----- fs/ecryptfs/messaging.c | 81 ++++++++++++++++++++++++++++++++----------- fs/ecryptfs/miscdev.c | 68 +++++++++++++++++++++++------------- fs/ecryptfs/netlink.c | 25 ++++++++----- kernel/user_namespace.c | 1 + 5 files changed, 136 insertions(+), 64 deletions(-) (limited to 'fs') diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index 72e117706a68..951ee33a022d 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -34,6 +34,7 @@ #include #include #include +#include /* Version verification for shared data structures w/ userspace */ #define ECRYPTFS_VERSION_MAJOR 0x00 @@ -410,8 +411,9 @@ struct ecryptfs_daemon { #define ECRYPTFS_DAEMON_MISCDEV_OPEN 0x00000008 u32 flags; u32 num_queued_msg_ctx; - pid_t pid; + struct pid *pid; uid_t euid; + struct user_namespace *user_ns; struct task_struct *task; struct mutex mux; struct list_head msg_ctx_out_queue; @@ -610,10 +612,13 @@ int ecryptfs_setxattr(struct dentry *dentry, const char *name, const void *value, size_t size, int flags); int ecryptfs_read_xattr_region(char *page_virt, struct inode *ecryptfs_inode); -int ecryptfs_process_helo(unsigned int transport, uid_t uid, pid_t pid); -int ecryptfs_process_quit(uid_t uid, pid_t pid); -int ecryptfs_process_response(struct ecryptfs_message *msg, uid_t uid, - pid_t pid, u32 seq); +int ecryptfs_process_helo(unsigned int transport, uid_t euid, + struct user_namespace *user_ns, struct pid *pid); +int ecryptfs_process_quit(uid_t euid, struct user_namespace *user_ns, + struct pid *pid); +int ecryptfs_process_response(struct ecryptfs_message *msg, uid_t euid, + struct user_namespace *user_ns, struct pid *pid, + u32 seq); int ecryptfs_send_message(unsigned int transport, char *data, int data_len, struct ecryptfs_msg_ctx **msg_ctx); int ecryptfs_wait_for_response(struct ecryptfs_msg_ctx *msg_ctx, @@ -623,13 +628,13 @@ void ecryptfs_release_messaging(unsigned int transport); int ecryptfs_send_netlink(char *data, int data_len, struct ecryptfs_msg_ctx *msg_ctx, u8 msg_type, - u16 msg_flags, pid_t daemon_pid); + u16 msg_flags, struct pid *daemon_pid); int ecryptfs_init_netlink(void); void ecryptfs_release_netlink(void); int ecryptfs_send_connector(char *data, int data_len, struct ecryptfs_msg_ctx *msg_ctx, u8 msg_type, - u16 msg_flags, pid_t daemon_pid); + u16 msg_flags, struct pid *daemon_pid); int ecryptfs_init_connector(void); void ecryptfs_release_connector(void); void @@ -672,7 +677,8 @@ int ecryptfs_read_lower_page_segment(struct page *page_for_ecryptfs, struct inode *ecryptfs_inode); struct page *ecryptfs_get_locked_page(struct file *file, loff_t index); int ecryptfs_exorcise_daemon(struct ecryptfs_daemon *daemon); -int ecryptfs_find_daemon_by_euid(struct ecryptfs_daemon **daemon, uid_t euid); +int ecryptfs_find_daemon_by_euid(struct ecryptfs_daemon **daemon, uid_t euid, + struct user_namespace *user_ns); int ecryptfs_parse_packet_length(unsigned char *data, size_t *size, size_t *length_size); int ecryptfs_write_packet_length(char *dest, size_t size, @@ -684,6 +690,7 @@ int ecryptfs_send_miscdev(char *data, size_t data_size, u16 msg_flags, struct ecryptfs_daemon *daemon); void ecryptfs_msg_ctx_alloc_to_free(struct ecryptfs_msg_ctx *msg_ctx); int -ecryptfs_spawn_daemon(struct ecryptfs_daemon **daemon, uid_t euid, pid_t pid); +ecryptfs_spawn_daemon(struct ecryptfs_daemon **daemon, uid_t euid, + struct user_namespace *user_ns, struct pid *pid); #endif /* #ifndef ECRYPTFS_KERNEL_H */ diff --git a/fs/ecryptfs/messaging.c b/fs/ecryptfs/messaging.c index c6038bd60897..1b5c20058acb 100644 --- a/fs/ecryptfs/messaging.c +++ b/fs/ecryptfs/messaging.c @@ -20,6 +20,8 @@ * 02111-1307, USA. */ #include +#include +#include #include "ecryptfs_kernel.h" static LIST_HEAD(ecryptfs_msg_ctx_free_list); @@ -103,6 +105,7 @@ void ecryptfs_msg_ctx_alloc_to_free(struct ecryptfs_msg_ctx *msg_ctx) /** * ecryptfs_find_daemon_by_euid * @euid: The effective user id which maps to the desired daemon id + * @user_ns: The namespace in which @euid applies * @daemon: If return value is zero, points to the desired daemon pointer * * Must be called with ecryptfs_daemon_hash_mux held. @@ -111,7 +114,8 @@ void ecryptfs_msg_ctx_alloc_to_free(struct ecryptfs_msg_ctx *msg_ctx) * * Returns zero if the user id exists in the list; non-zero otherwise. */ -int ecryptfs_find_daemon_by_euid(struct ecryptfs_daemon **daemon, uid_t euid) +int ecryptfs_find_daemon_by_euid(struct ecryptfs_daemon **daemon, uid_t euid, + struct user_namespace *user_ns) { struct hlist_node *elem; int rc; @@ -119,7 +123,7 @@ int ecryptfs_find_daemon_by_euid(struct ecryptfs_daemon **daemon, uid_t euid) hlist_for_each_entry(*daemon, elem, &ecryptfs_daemon_hash[ecryptfs_uid_hash(euid)], euid_chain) { - if ((*daemon)->euid == euid) { + if ((*daemon)->euid == euid && (*daemon)->user_ns == user_ns) { rc = 0; goto out; } @@ -186,6 +190,7 @@ out: * ecryptfs_spawn_daemon - Create and initialize a new daemon struct * @daemon: Pointer to set to newly allocated daemon struct * @euid: Effective user id for the daemon + * @user_ns: The namespace in which @euid applies * @pid: Process id for the daemon * * Must be called ceremoniously while in possession of @@ -194,7 +199,8 @@ out: * Returns zero on success; non-zero otherwise */ int -ecryptfs_spawn_daemon(struct ecryptfs_daemon **daemon, uid_t euid, pid_t pid) +ecryptfs_spawn_daemon(struct ecryptfs_daemon **daemon, uid_t euid, + struct user_namespace *user_ns, struct pid *pid) { int rc = 0; @@ -206,7 +212,8 @@ ecryptfs_spawn_daemon(struct ecryptfs_daemon **daemon, uid_t euid, pid_t pid) goto out; } (*daemon)->euid = euid; - (*daemon)->pid = pid; + (*daemon)->user_ns = get_user_ns(user_ns); + (*daemon)->pid = get_pid(pid); (*daemon)->task = current; mutex_init(&(*daemon)->mux); INIT_LIST_HEAD(&(*daemon)->msg_ctx_out_queue); @@ -222,6 +229,7 @@ out: * ecryptfs_process_helo * @transport: The underlying transport (netlink, etc.) * @euid: The user ID owner of the message + * @user_ns: The namespace in which @euid applies * @pid: The process ID for the userspace program that sent the * message * @@ -231,32 +239,33 @@ out: * Returns zero after adding a new daemon to the hash list; * non-zero otherwise. */ -int ecryptfs_process_helo(unsigned int transport, uid_t euid, pid_t pid) +int ecryptfs_process_helo(unsigned int transport, uid_t euid, + struct user_namespace *user_ns, struct pid *pid) { struct ecryptfs_daemon *new_daemon; struct ecryptfs_daemon *old_daemon; int rc; mutex_lock(&ecryptfs_daemon_hash_mux); - rc = ecryptfs_find_daemon_by_euid(&old_daemon, euid); + rc = ecryptfs_find_daemon_by_euid(&old_daemon, euid, user_ns); if (rc != 0) { printk(KERN_WARNING "Received request from user [%d] " - "to register daemon [%d]; unregistering daemon " - "[%d]\n", euid, pid, old_daemon->pid); + "to register daemon [0x%p]; unregistering daemon " + "[0x%p]\n", euid, pid, old_daemon->pid); rc = ecryptfs_send_raw_message(transport, ECRYPTFS_MSG_QUIT, old_daemon); if (rc) printk(KERN_WARNING "Failed to send QUIT " - "message to daemon [%d]; rc = [%d]\n", + "message to daemon [0x%p]; rc = [%d]\n", old_daemon->pid, rc); hlist_del(&old_daemon->euid_chain); kfree(old_daemon); } - rc = ecryptfs_spawn_daemon(&new_daemon, euid, pid); + rc = ecryptfs_spawn_daemon(&new_daemon, euid, user_ns, pid); if (rc) printk(KERN_ERR "%s: The gods are displeased with this attempt " - "to create a new daemon object for euid [%d]; pid [%d]; " - "rc = [%d]\n", __func__, euid, pid, rc); + "to create a new daemon object for euid [%d]; pid " + "[0x%p]; rc = [%d]\n", __func__, euid, pid, rc); mutex_unlock(&ecryptfs_daemon_hash_mux); return rc; } @@ -277,7 +286,7 @@ int ecryptfs_exorcise_daemon(struct ecryptfs_daemon *daemon) || (daemon->flags & ECRYPTFS_DAEMON_IN_POLL)) { rc = -EBUSY; printk(KERN_WARNING "%s: Attempt to destroy daemon with pid " - "[%d], but it is in the midst of a read or a poll\n", + "[0x%p], but it is in the midst of a read or a poll\n", __func__, daemon->pid); mutex_unlock(&daemon->mux); goto out; @@ -293,6 +302,10 @@ int ecryptfs_exorcise_daemon(struct ecryptfs_daemon *daemon) hlist_del(&daemon->euid_chain); if (daemon->task) wake_up_process(daemon->task); + if (daemon->pid) + put_pid(daemon->pid); + if (daemon->user_ns) + put_user_ns(daemon->user_ns); mutex_unlock(&daemon->mux); memset(daemon, 0, sizeof(*daemon)); kfree(daemon); @@ -303,6 +316,7 @@ out: /** * ecryptfs_process_quit * @euid: The user ID owner of the message + * @user_ns: The namespace in which @euid applies * @pid: The process ID for the userspace program that sent the * message * @@ -310,17 +324,18 @@ out: * it is the registered that is requesting the deletion. Returns zero * after deleting the desired daemon; non-zero otherwise. */ -int ecryptfs_process_quit(uid_t euid, pid_t pid) +int ecryptfs_process_quit(uid_t euid, struct user_namespace *user_ns, + struct pid *pid) { struct ecryptfs_daemon *daemon; int rc; mutex_lock(&ecryptfs_daemon_hash_mux); - rc = ecryptfs_find_daemon_by_euid(&daemon, euid); + rc = ecryptfs_find_daemon_by_euid(&daemon, euid, user_ns); if (rc || !daemon) { rc = -EINVAL; printk(KERN_ERR "Received request from user [%d] to " - "unregister unrecognized daemon [%d]\n", euid, pid); + "unregister unrecognized daemon [0x%p]\n", euid, pid); goto out_unlock; } rc = ecryptfs_exorcise_daemon(daemon); @@ -354,11 +369,14 @@ out_unlock: * Returns zero on success; non-zero otherwise */ int ecryptfs_process_response(struct ecryptfs_message *msg, uid_t euid, - pid_t pid, u32 seq) + struct user_namespace *user_ns, struct pid *pid, + u32 seq) { struct ecryptfs_daemon *daemon; struct ecryptfs_msg_ctx *msg_ctx; size_t msg_size; + struct nsproxy *nsproxy; + struct user_namespace *current_user_ns; int rc; if (msg->index >= ecryptfs_message_buf_len) { @@ -372,12 +390,25 @@ int ecryptfs_process_response(struct ecryptfs_message *msg, uid_t euid, msg_ctx = &ecryptfs_msg_ctx_arr[msg->index]; mutex_lock(&msg_ctx->mux); mutex_lock(&ecryptfs_daemon_hash_mux); - rc = ecryptfs_find_daemon_by_euid(&daemon, msg_ctx->task->euid); + rcu_read_lock(); + nsproxy = task_nsproxy(msg_ctx->task); + if (nsproxy == NULL) { + rc = -EBADMSG; + printk(KERN_ERR "%s: Receiving process is a zombie. Dropping " + "message.\n", __func__); + rcu_read_unlock(); + mutex_unlock(&ecryptfs_daemon_hash_mux); + goto wake_up; + } + current_user_ns = nsproxy->user_ns; + rc = ecryptfs_find_daemon_by_euid(&daemon, msg_ctx->task->euid, + current_user_ns); + rcu_read_unlock(); mutex_unlock(&ecryptfs_daemon_hash_mux); if (rc) { rc = -EBADMSG; printk(KERN_WARNING "%s: User [%d] received a " - "message response from process [%d] but does " + "message response from process [0x%p] but does " "not have a registered daemon\n", __func__, msg_ctx->task->euid, pid); goto wake_up; @@ -389,10 +420,17 @@ int ecryptfs_process_response(struct ecryptfs_message *msg, uid_t euid, euid, msg_ctx->task->euid); goto unlock; } + if (current_user_ns != user_ns) { + rc = -EBADMSG; + printk(KERN_WARNING "%s: Received message from user_ns " + "[0x%p]; expected message from user_ns [0x%p]\n", + __func__, user_ns, nsproxy->user_ns); + goto unlock; + } if (daemon->pid != pid) { rc = -EBADMSG; printk(KERN_ERR "%s: User [%d] sent a message response " - "from an unrecognized process [%d]\n", + "from an unrecognized process [0x%p]\n", __func__, msg_ctx->task->euid, pid); goto unlock; } @@ -446,7 +484,8 @@ ecryptfs_send_message_locked(unsigned int transport, char *data, int data_len, struct ecryptfs_daemon *daemon; int rc; - rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid); + rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid, + current->nsproxy->user_ns); if (rc || !daemon) { rc = -ENOTCONN; printk(KERN_ERR "%s: User [%d] does not have a daemon " diff --git a/fs/ecryptfs/miscdev.c b/fs/ecryptfs/miscdev.c index 0c559731ae34..788995efd1d3 100644 --- a/fs/ecryptfs/miscdev.c +++ b/fs/ecryptfs/miscdev.c @@ -46,7 +46,8 @@ ecryptfs_miscdev_poll(struct file *file, poll_table *pt) mutex_lock(&ecryptfs_daemon_hash_mux); /* TODO: Just use file->private_data? */ - rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid); + rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid, + current->nsproxy->user_ns); BUG_ON(rc || !daemon); mutex_lock(&daemon->mux); mutex_unlock(&ecryptfs_daemon_hash_mux); @@ -92,10 +93,12 @@ ecryptfs_miscdev_open(struct inode *inode, struct file *file) "count; rc = [%d]\n", __func__, rc); goto out_unlock_daemon_list; } - rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid); + rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid, + current->nsproxy->user_ns); if (rc || !daemon) { rc = ecryptfs_spawn_daemon(&daemon, current->euid, - current->pid); + current->nsproxy->user_ns, + task_pid(current)); if (rc) { printk(KERN_ERR "%s: Error attempting to spawn daemon; " "rc = [%d]\n", __func__, rc); @@ -103,18 +106,18 @@ ecryptfs_miscdev_open(struct inode *inode, struct file *file) } } mutex_lock(&daemon->mux); - if (daemon->pid != current->pid) { + if (daemon->pid != task_pid(current)) { rc = -EINVAL; - printk(KERN_ERR "%s: pid [%d] has registered with euid [%d], " - "but pid [%d] has attempted to open the handle " + printk(KERN_ERR "%s: pid [0x%p] has registered with euid [%d], " + "but pid [0x%p] has attempted to open the handle " "instead\n", __func__, daemon->pid, daemon->euid, - current->pid); + task_pid(current)); goto out_unlock_daemon; } if (daemon->flags & ECRYPTFS_DAEMON_MISCDEV_OPEN) { rc = -EBUSY; printk(KERN_ERR "%s: Miscellaneous device handle may only be " - "opened once per daemon; pid [%d] already has this " + "opened once per daemon; pid [0x%p] already has this " "handle open\n", __func__, daemon->pid); goto out_unlock_daemon; } @@ -147,10 +150,11 @@ ecryptfs_miscdev_release(struct inode *inode, struct file *file) int rc; mutex_lock(&ecryptfs_daemon_hash_mux); - rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid); + rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid, + current->nsproxy->user_ns); BUG_ON(rc || !daemon); mutex_lock(&daemon->mux); - BUG_ON(daemon->pid != current->pid); + BUG_ON(daemon->pid != task_pid(current)); BUG_ON(!(daemon->flags & ECRYPTFS_DAEMON_MISCDEV_OPEN)); daemon->flags &= ~ECRYPTFS_DAEMON_MISCDEV_OPEN; atomic_dec(&ecryptfs_num_miscdev_opens); @@ -247,7 +251,8 @@ ecryptfs_miscdev_read(struct file *file, char __user *buf, size_t count, mutex_lock(&ecryptfs_daemon_hash_mux); /* TODO: Just use file->private_data? */ - rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid); + rc = ecryptfs_find_daemon_by_euid(&daemon, current->euid, + current->nsproxy->user_ns); BUG_ON(rc || !daemon); mutex_lock(&daemon->mux); if (daemon->flags & ECRYPTFS_DAEMON_ZOMBIE) { @@ -285,7 +290,8 @@ check_list: goto check_list; } BUG_ON(current->euid != daemon->euid); - BUG_ON(current->pid != daemon->pid); + BUG_ON(current->nsproxy->user_ns != daemon->user_ns); + BUG_ON(task_pid(current) != daemon->pid); msg_ctx = list_first_entry(&daemon->msg_ctx_out_queue, struct ecryptfs_msg_ctx, daemon_out_list); BUG_ON(!msg_ctx); @@ -355,15 +361,18 @@ out_unlock_daemon: /** * ecryptfs_miscdev_helo * @euid: effective user id of miscdevess sending helo packet + * @user_ns: The namespace in which @euid applies * @pid: miscdevess id of miscdevess sending helo packet * * Returns zero on success; non-zero otherwise */ -static int ecryptfs_miscdev_helo(uid_t uid, pid_t pid) +static int ecryptfs_miscdev_helo(uid_t euid, struct user_namespace *user_ns, + struct pid *pid) { int rc; - rc = ecryptfs_process_helo(ECRYPTFS_TRANSPORT_MISCDEV, uid, pid); + rc = ecryptfs_process_helo(ECRYPTFS_TRANSPORT_MISCDEV, euid, user_ns, + pid); if (rc) printk(KERN_WARNING "Error processing HELO; rc = [%d]\n", rc); return rc; @@ -372,15 +381,17 @@ static int ecryptfs_miscdev_helo(uid_t uid, pid_t pid) /** * ecryptfs_miscdev_quit * @euid: effective user id of miscdevess sending quit packet + * @user_ns: The namespace in which @euid applies * @pid: miscdevess id of miscdevess sending quit packet * * Returns zero on success; non-zero otherwise */ -static int ecryptfs_miscdev_quit(uid_t euid, pid_t pid) +static int ecryptfs_miscdev_quit(uid_t euid, struct user_namespace *user_ns, + struct pid *pid) { int rc; - rc = ecryptfs_process_quit(euid, pid); + rc = ecryptfs_process_quit(euid, user_ns, pid); if (rc) printk(KERN_WARNING "Error processing QUIT message; rc = [%d]\n", rc); @@ -392,13 +403,15 @@ static int ecryptfs_miscdev_quit(uid_t euid, pid_t pid) * @data: Bytes comprising struct ecryptfs_message * @data_size: sizeof(struct ecryptfs_message) + data len * @euid: Effective user id of miscdevess sending the miscdev response + * @user_ns: The namespace in which @euid applies * @pid: Miscdevess id of miscdevess sending the miscdev response * @seq: Sequence number for miscdev response packet * * Returns zero on success; non-zero otherwise */ static int ecryptfs_miscdev_response(char *data, size_t data_size, - uid_t euid, pid_t pid, u32 seq) + uid_t euid, struct user_namespace *user_ns, + struct pid *pid, u32 seq) { struct ecryptfs_message *msg = (struct ecryptfs_message *)data; int rc; @@ -410,7 +423,7 @@ static int ecryptfs_miscdev_response(char *data, size_t data_size, rc = -EINVAL; goto out; } - rc = ecryptfs_process_response(msg, euid, pid, seq); + rc = ecryptfs_process_response(msg, euid, user_ns, pid, seq); if (rc) printk(KERN_ERR "Error processing response message; rc = [%d]\n", rc); @@ -491,27 +504,32 @@ ecryptfs_miscdev_write(struct file *file, const char __user *buf, } rc = ecryptfs_miscdev_response(&data[i], packet_size, current->euid, - current->pid, seq); + current->nsproxy->user_ns, + task_pid(current), seq); if (rc) printk(KERN_WARNING "%s: Failed to deliver miscdev " "response to requesting operation; rc = [%d]\n", __func__, rc); break; case ECRYPTFS_MSG_HELO: - rc = ecryptfs_miscdev_helo(current->euid, current->pid); + rc = ecryptfs_miscdev_helo(current->euid, + current->nsproxy->user_ns, + task_pid(current)); if (rc) { printk(KERN_ERR "%s: Error attempting to process " - "helo from pid [%d]; rc = [%d]\n", __func__, - current->pid, rc); + "helo from pid [0x%p]; rc = [%d]\n", __func__, + task_pid(current), rc); goto out_free; } break; case ECRYPTFS_MSG_QUIT: - rc = ecryptfs_miscdev_quit(current->euid, current->pid); + rc = ecryptfs_miscdev_quit(current->euid, + current->nsproxy->user_ns, + task_pid(current)); if (rc) { printk(KERN_ERR "%s: Error attempting to process " - "quit from pid [%d]; rc = [%d]\n", __func__, - current->pid, rc); + "quit from pid [0x%p]; rc = [%d]\n", __func__, + task_pid(current), rc); goto out_free; } break; diff --git a/fs/ecryptfs/netlink.c b/fs/ecryptfs/netlink.c index eb70f69d705d..e0abad62b395 100644 --- a/fs/ecryptfs/netlink.c +++ b/fs/ecryptfs/netlink.c @@ -45,7 +45,7 @@ static struct sock *ecryptfs_nl_sock; */ int ecryptfs_send_netlink(char *data, int data_len, struct ecryptfs_msg_ctx *msg_ctx, u8 msg_type, - u16 msg_flags, pid_t daemon_pid) + u16 msg_flags, struct pid *daemon_pid) { struct sk_buff *skb; struct nlmsghdr *nlh; @@ -60,7 +60,7 @@ int ecryptfs_send_netlink(char *data, int data_len, ecryptfs_printk(KERN_ERR, "Failed to allocate socket buffer\n"); goto out; } - nlh = NLMSG_PUT(skb, daemon_pid, msg_ctx ? msg_ctx->counter : 0, + nlh = NLMSG_PUT(skb, pid_nr(daemon_pid), msg_ctx ? msg_ctx->counter : 0, msg_type, payload_len); nlh->nlmsg_flags = msg_flags; if (msg_ctx && payload_len) { @@ -69,7 +69,7 @@ int ecryptfs_send_netlink(char *data, int data_len, msg->data_len = data_len; memcpy(msg->data, data, data_len); } - rc = netlink_unicast(ecryptfs_nl_sock, skb, daemon_pid, 0); + rc = netlink_unicast(ecryptfs_nl_sock, skb, pid_nr(daemon_pid), 0); if (rc < 0) { ecryptfs_printk(KERN_ERR, "Failed to send eCryptfs netlink " "message; rc = [%d]\n", rc); @@ -99,6 +99,7 @@ static int ecryptfs_process_nl_response(struct sk_buff *skb) { struct nlmsghdr *nlh = nlmsg_hdr(skb); struct ecryptfs_message *msg = NLMSG_DATA(nlh); + struct pid *pid; int rc; if (skb->len - NLMSG_HDRLEN - sizeof(*msg) != msg->data_len) { @@ -107,8 +108,10 @@ static int ecryptfs_process_nl_response(struct sk_buff *skb) "incorrectly specified data length\n"); goto out; } - rc = ecryptfs_process_response(msg, NETLINK_CREDS(skb)->uid, - NETLINK_CREDS(skb)->pid, nlh->nlmsg_seq); + pid = find_get_pid(NETLINK_CREDS(skb)->pid); + rc = ecryptfs_process_response(msg, NETLINK_CREDS(skb)->uid, NULL, + pid, nlh->nlmsg_seq); + put_pid(pid); if (rc) printk(KERN_ERR "Error processing response message; rc = [%d]\n", rc); @@ -126,11 +129,13 @@ out: */ static int ecryptfs_process_nl_helo(struct sk_buff *skb) { + struct pid *pid; int rc; + pid = find_get_pid(NETLINK_CREDS(skb)->pid); rc = ecryptfs_process_helo(ECRYPTFS_TRANSPORT_NETLINK, - NETLINK_CREDS(skb)->uid, - NETLINK_CREDS(skb)->pid); + NETLINK_CREDS(skb)->uid, NULL, pid); + put_pid(pid); if (rc) printk(KERN_WARNING "Error processing HELO; rc = [%d]\n", rc); return rc; @@ -147,10 +152,12 @@ static int ecryptfs_process_nl_helo(struct sk_buff *skb) */ static int ecryptfs_process_nl_quit(struct sk_buff *skb) { + struct pid *pid; int rc; - rc = ecryptfs_process_quit(NETLINK_CREDS(skb)->uid, - NETLINK_CREDS(skb)->pid); + pid = find_get_pid(NETLINK_CREDS(skb)->pid); + rc = ecryptfs_process_quit(NETLINK_CREDS(skb)->uid, NULL, pid); + put_pid(pid); if (rc) printk(KERN_WARNING "Error processing QUIT message; rc = [%d]\n", rc); diff --git a/kernel/user_namespace.c b/kernel/user_namespace.c index 2731ba80e30b..a9ab0596de44 100644 --- a/kernel/user_namespace.c +++ b/kernel/user_namespace.c @@ -74,3 +74,4 @@ void free_user_ns(struct kref *kref) release_uids(ns); kfree(ns); } +EXPORT_SYMBOL(free_user_ns); -- cgit v1.2.3 From 2f9b12a31fcb738ea8c9eb0d4ddf906c6f1d696c Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Tue, 29 Apr 2008 00:59:52 -0700 Subject: eCryptfs: protect crypt_stat->flags in ecryptfs_open() Make sure crypt_stat->flags is protected with a lock in ecryptfs_open(). Signed-off-by: Michael Halcrow Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/file.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index 2b8f5ed4adea..2258b8f654a6 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -195,7 +195,9 @@ static int ecryptfs_open(struct inode *inode, struct file *file) file, ecryptfs_inode_to_private(inode)->lower_file); if (S_ISDIR(ecryptfs_dentry->d_inode->i_mode)) { ecryptfs_printk(KERN_DEBUG, "This is a directory\n"); + mutex_lock(&crypt_stat->cs_mutex); crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED); + mutex_unlock(&crypt_stat->cs_mutex); rc = 0; goto out; } -- cgit v1.2.3 From 08ce5f16ee466ffc5bf243800deeecd77d9eaf50 Mon Sep 17 00:00:00 2001 From: "Serge E. Hallyn" Date: Tue, 29 Apr 2008 01:00:10 -0700 Subject: cgroups: implement device whitelist Implement a cgroup to track and enforce open and mknod restrictions on device files. A device cgroup associates a device access whitelist with each cgroup. A whitelist entry has 4 fields. 'type' is a (all), c (char), or b (block). 'all' means it applies to all types and all major and minor numbers. Major and minor are either an integer or * for all. Access is a composition of r (read), w (write), and m (mknod). The root device cgroup starts with rwm to 'all'. A child devcg gets a copy of the parent. Admins can then remove devices from the whitelist or add new entries. A child cgroup can never receive a device access which is denied its parent. However when a device access is removed from a parent it will not also be removed from the child(ren). An entry is added using devices.allow, and removed using devices.deny. For instance echo 'c 1:3 mr' > /cgroups/1/devices.allow allows cgroup 1 to read and mknod the device usually known as /dev/null. Doing echo a > /cgroups/1/devices.deny will remove the default 'a *:* mrw' entry. CAP_SYS_ADMIN is needed to change permissions or move another task to a new cgroup. A cgroup may not be granted more permissions than the cgroup's parent has. Any task can move itself between cgroups. This won't be sufficient, but we can decide the best way to adequately restrict movement later. [akpm@linux-foundation.org: coding-style fixes] [akpm@linux-foundation.org: fix may-be-used-uninitialized warning] Signed-off-by: Serge E. Hallyn Acked-by: James Morris Looks-good-to: Pavel Emelyanov Cc: Daniel Hokka Zakrisson Cc: Li Zefan Cc: Paul Menage Cc: Balbir Singh Cc: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/controllers/devices.txt | 48 +++ fs/namei.c | 9 + include/linux/cgroup_subsys.h | 6 + include/linux/device_cgroup.h | 12 + init/Kconfig | 7 + security/Makefile | 1 + security/device_cgroup.c | 603 ++++++++++++++++++++++++++++++++++ 7 files changed, 686 insertions(+) create mode 100644 Documentation/controllers/devices.txt create mode 100644 include/linux/device_cgroup.h create mode 100644 security/device_cgroup.c (limited to 'fs') diff --git a/Documentation/controllers/devices.txt b/Documentation/controllers/devices.txt new file mode 100644 index 000000000000..4dcea42432c2 --- /dev/null +++ b/Documentation/controllers/devices.txt @@ -0,0 +1,48 @@ +Device Whitelist Controller + +1. Description: + +Implement a cgroup to track and enforce open and mknod restrictions +on device files. A device cgroup associates a device access +whitelist with each cgroup. A whitelist entry has 4 fields. +'type' is a (all), c (char), or b (block). 'all' means it applies +to all types and all major and minor numbers. Major and minor are +either an integer or * for all. Access is a composition of r +(read), w (write), and m (mknod). + +The root device cgroup starts with rwm to 'all'. A child device +cgroup gets a copy of the parent. Administrators can then remove +devices from the whitelist or add new entries. A child cgroup can +never receive a device access which is denied its parent. However +when a device access is removed from a parent it will not also be +removed from the child(ren). + +2. User Interface + +An entry is added using devices.allow, and removed using +devices.deny. For instance + + echo 'c 1:3 mr' > /cgroups/1/devices.allow + +allows cgroup 1 to read and mknod the device usually known as +/dev/null. Doing + + echo a > /cgroups/1/devices.deny + +will remove the default 'a *:* mrw' entry. + +3. Security + +Any task can move itself between cgroups. This clearly won't +suffice, but we can decide the best way to adequately restrict +movement as people get some experience with this. We may just want +to require CAP_SYS_ADMIN, which at least is a separate bit from +CAP_MKNOD. We may want to just refuse moving to a cgroup which +isn't a descendent of the current one. Or we may want to use +CAP_MAC_ADMIN, since we really are trying to lock down root. + +CAP_SYS_ADMIN is needed to modify the whitelist or move another +task to a new cgroup. (Again we'll probably want to change that). + +A cgroup may not be granted more permissions than the cgroup's +parent has. diff --git a/fs/namei.c b/fs/namei.c index e179f71bfcb0..32fd9655485b 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include @@ -281,6 +282,10 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) if (retval) return retval; + retval = devcgroup_inode_permission(inode, mask); + if (retval) + return retval; + return security_inode_permission(inode, mask, nd); } @@ -2028,6 +2033,10 @@ int vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) if (!dir->i_op || !dir->i_op->mknod) return -EPERM; + error = devcgroup_inode_mknod(mode, dev); + if (error) + return error; + error = security_inode_mknod(dir, dentry, mode, dev); if (error) return error; diff --git a/include/linux/cgroup_subsys.h b/include/linux/cgroup_subsys.h index 1ddebfc52565..e2877454ec82 100644 --- a/include/linux/cgroup_subsys.h +++ b/include/linux/cgroup_subsys.h @@ -42,3 +42,9 @@ SUBSYS(mem_cgroup) #endif /* */ + +#ifdef CONFIG_CGROUP_DEVICE +SUBSYS(devices) +#endif + +/* */ diff --git a/include/linux/device_cgroup.h b/include/linux/device_cgroup.h new file mode 100644 index 000000000000..0b0d9c39ed67 --- /dev/null +++ b/include/linux/device_cgroup.h @@ -0,0 +1,12 @@ +#include +#include + +#ifdef CONFIG_CGROUP_DEVICE +extern int devcgroup_inode_permission(struct inode *inode, int mask); +extern int devcgroup_inode_mknod(int mode, dev_t dev); +#else +static inline int devcgroup_inode_permission(struct inode *inode, int mask) +{ return 0; } +static inline int devcgroup_inode_mknod(int mode, dev_t dev) +{ return 0; } +#endif diff --git a/init/Kconfig b/init/Kconfig index 6ce16bdbec76..a3457926342a 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -298,6 +298,13 @@ config CGROUP_NS for instance virtual servers and checkpoint/restart jobs. +config CGROUP_DEVICE + bool "Device controller for cgroups" + depends on CGROUPS && EXPERIMENTAL + help + Provides a cgroup implementing whitelists for devices which + a process in the cgroup can mknod or open. + config CPUSETS bool "Cpuset support" depends on SMP && CGROUPS diff --git a/security/Makefile b/security/Makefile index 9e8b02525014..7ef1107a7287 100644 --- a/security/Makefile +++ b/security/Makefile @@ -18,3 +18,4 @@ obj-$(CONFIG_SECURITY_SELINUX) += selinux/built-in.o obj-$(CONFIG_SECURITY_SMACK) += commoncap.o smack/built-in.o obj-$(CONFIG_SECURITY_CAPABILITIES) += commoncap.o capability.o obj-$(CONFIG_SECURITY_ROOTPLUG) += commoncap.o root_plug.o +obj-$(CONFIG_CGROUP_DEVICE) += device_cgroup.o diff --git a/security/device_cgroup.c b/security/device_cgroup.c new file mode 100644 index 000000000000..4237b19e8fb3 --- /dev/null +++ b/security/device_cgroup.c @@ -0,0 +1,603 @@ +/* + * dev_cgroup.c - device cgroup subsystem + * + * Copyright 2007 IBM Corp + */ + +#include +#include +#include +#include +#include + +#define ACC_MKNOD 1 +#define ACC_READ 2 +#define ACC_WRITE 4 +#define ACC_MASK (ACC_MKNOD | ACC_READ | ACC_WRITE) + +#define DEV_BLOCK 1 +#define DEV_CHAR 2 +#define DEV_ALL 4 /* this represents all devices */ + +/* + * whitelist locking rules: + * cgroup_lock() cannot be taken under dev_cgroup->lock. + * dev_cgroup->lock can be taken with or without cgroup_lock(). + * + * modifications always require cgroup_lock + * modifications to a list which is visible require the + * dev_cgroup->lock *and* cgroup_lock() + * walking the list requires dev_cgroup->lock or cgroup_lock(). + * + * reasoning: dev_whitelist_copy() needs to kmalloc, so needs + * a mutex, which the cgroup_lock() is. Since modifying + * a visible list requires both locks, either lock can be + * taken for walking the list. + */ + +struct dev_whitelist_item { + u32 major, minor; + short type; + short access; + struct list_head list; +}; + +struct dev_cgroup { + struct cgroup_subsys_state css; + struct list_head whitelist; + spinlock_t lock; +}; + +static inline struct dev_cgroup *cgroup_to_devcgroup(struct cgroup *cgroup) +{ + return container_of(cgroup_subsys_state(cgroup, devices_subsys_id), + struct dev_cgroup, css); +} + +struct cgroup_subsys devices_subsys; + +static int devcgroup_can_attach(struct cgroup_subsys *ss, + struct cgroup *new_cgroup, struct task_struct *task) +{ + if (current != task && !capable(CAP_SYS_ADMIN)) + return -EPERM; + + return 0; +} + +/* + * called under cgroup_lock() + */ +static int dev_whitelist_copy(struct list_head *dest, struct list_head *orig) +{ + struct dev_whitelist_item *wh, *tmp, *new; + + list_for_each_entry(wh, orig, list) { + new = kmalloc(sizeof(*wh), GFP_KERNEL); + if (!new) + goto free_and_exit; + new->major = wh->major; + new->minor = wh->minor; + new->type = wh->type; + new->access = wh->access; + list_add_tail(&new->list, dest); + } + + return 0; + +free_and_exit: + list_for_each_entry_safe(wh, tmp, dest, list) { + list_del(&wh->list); + kfree(wh); + } + return -ENOMEM; +} + +/* Stupid prototype - don't bother combining existing entries */ +/* + * called under cgroup_lock() + * since the list is visible to other tasks, we need the spinlock also + */ +static int dev_whitelist_add(struct dev_cgroup *dev_cgroup, + struct dev_whitelist_item *wh) +{ + struct dev_whitelist_item *whcopy; + + whcopy = kmalloc(sizeof(*whcopy), GFP_KERNEL); + if (!whcopy) + return -ENOMEM; + + memcpy(whcopy, wh, sizeof(*whcopy)); + spin_lock(&dev_cgroup->lock); + list_add_tail(&whcopy->list, &dev_cgroup->whitelist); + spin_unlock(&dev_cgroup->lock); + return 0; +} + +/* + * called under cgroup_lock() + * since the list is visible to other tasks, we need the spinlock also + */ +static void dev_whitelist_rm(struct dev_cgroup *dev_cgroup, + struct dev_whitelist_item *wh) +{ + struct dev_whitelist_item *walk, *tmp; + + spin_lock(&dev_cgroup->lock); + list_for_each_entry_safe(walk, tmp, &dev_cgroup->whitelist, list) { + if (walk->type == DEV_ALL) + goto remove; + if (walk->type != wh->type) + continue; + if (walk->major != ~0 && walk->major != wh->major) + continue; + if (walk->minor != ~0 && walk->minor != wh->minor) + continue; + +remove: + walk->access &= ~wh->access; + if (!walk->access) { + list_del(&walk->list); + kfree(walk); + } + } + spin_unlock(&dev_cgroup->lock); +} + +/* + * called from kernel/cgroup.c with cgroup_lock() held. + */ +static struct cgroup_subsys_state *devcgroup_create(struct cgroup_subsys *ss, + struct cgroup *cgroup) +{ + struct dev_cgroup *dev_cgroup, *parent_dev_cgroup; + struct cgroup *parent_cgroup; + int ret; + + dev_cgroup = kzalloc(sizeof(*dev_cgroup), GFP_KERNEL); + if (!dev_cgroup) + return ERR_PTR(-ENOMEM); + INIT_LIST_HEAD(&dev_cgroup->whitelist); + parent_cgroup = cgroup->parent; + + if (parent_cgroup == NULL) { + struct dev_whitelist_item *wh; + wh = kmalloc(sizeof(*wh), GFP_KERNEL); + if (!wh) { + kfree(dev_cgroup); + return ERR_PTR(-ENOMEM); + } + wh->minor = wh->major = ~0; + wh->type = DEV_ALL; + wh->access = ACC_MKNOD | ACC_READ | ACC_WRITE; + list_add(&wh->list, &dev_cgroup->whitelist); + } else { + parent_dev_cgroup = cgroup_to_devcgroup(parent_cgroup); + ret = dev_whitelist_copy(&dev_cgroup->whitelist, + &parent_dev_cgroup->whitelist); + if (ret) { + kfree(dev_cgroup); + return ERR_PTR(ret); + } + } + + spin_lock_init(&dev_cgroup->lock); + return &dev_cgroup->css; +} + +static void devcgroup_destroy(struct cgroup_subsys *ss, + struct cgroup *cgroup) +{ + struct dev_cgroup *dev_cgroup; + struct dev_whitelist_item *wh, *tmp; + + dev_cgroup = cgroup_to_devcgroup(cgroup); + list_for_each_entry_safe(wh, tmp, &dev_cgroup->whitelist, list) { + list_del(&wh->list); + kfree(wh); + } + kfree(dev_cgroup); +} + +#define DEVCG_ALLOW 1 +#define DEVCG_DENY 2 + +static void set_access(char *acc, short access) +{ + int idx = 0; + memset(acc, 0, 4); + if (access & ACC_READ) + acc[idx++] = 'r'; + if (access & ACC_WRITE) + acc[idx++] = 'w'; + if (access & ACC_MKNOD) + acc[idx++] = 'm'; +} + +static char type_to_char(short type) +{ + if (type == DEV_ALL) + return 'a'; + if (type == DEV_CHAR) + return 'c'; + if (type == DEV_BLOCK) + return 'b'; + return 'X'; +} + +static void set_majmin(char *str, int len, unsigned m) +{ + memset(str, 0, len); + if (m == ~0) + sprintf(str, "*"); + else + snprintf(str, len, "%d", m); +} + +static char *print_whitelist(struct dev_cgroup *devcgroup, int *len) +{ + char *buf, *s, acc[4]; + struct dev_whitelist_item *wh; + int ret; + int count = 0; + char maj[10], min[10]; + + buf = kmalloc(4096, GFP_KERNEL); + if (!buf) + return ERR_PTR(-ENOMEM); + s = buf; + *s = '\0'; + *len = 0; + + spin_lock(&devcgroup->lock); + list_for_each_entry(wh, &devcgroup->whitelist, list) { + set_access(acc, wh->access); + set_majmin(maj, 10, wh->major); + set_majmin(min, 10, wh->minor); + ret = snprintf(s, 4095-(s-buf), "%c %s:%s %s\n", + type_to_char(wh->type), maj, min, acc); + if (s+ret >= buf+4095) { + kfree(buf); + buf = ERR_PTR(-ENOMEM); + break; + } + s += ret; + *len += ret; + count++; + } + spin_unlock(&devcgroup->lock); + + return buf; +} + +static ssize_t devcgroup_access_read(struct cgroup *cgroup, + struct cftype *cft, struct file *file, + char __user *userbuf, size_t nbytes, loff_t *ppos) +{ + struct dev_cgroup *devcgroup = cgroup_to_devcgroup(cgroup); + int filetype = cft->private; + char *buffer; + int uninitialized_var(len); + int retval; + + if (filetype != DEVCG_ALLOW) + return -EINVAL; + buffer = print_whitelist(devcgroup, &len); + if (IS_ERR(buffer)) + return PTR_ERR(buffer); + + retval = simple_read_from_buffer(userbuf, nbytes, ppos, buffer, len); + kfree(buffer); + return retval; +} + +/* + * may_access_whitelist: + * does the access granted to dev_cgroup c contain the access + * requested in whitelist item refwh. + * return 1 if yes, 0 if no. + * call with c->lock held + */ +static int may_access_whitelist(struct dev_cgroup *c, + struct dev_whitelist_item *refwh) +{ + struct dev_whitelist_item *whitem; + + list_for_each_entry(whitem, &c->whitelist, list) { + if (whitem->type & DEV_ALL) + return 1; + if ((refwh->type & DEV_BLOCK) && !(whitem->type & DEV_BLOCK)) + continue; + if ((refwh->type & DEV_CHAR) && !(whitem->type & DEV_CHAR)) + continue; + if (whitem->major != ~0 && whitem->major != refwh->major) + continue; + if (whitem->minor != ~0 && whitem->minor != refwh->minor) + continue; + if (refwh->access & (~(whitem->access | ACC_MASK))) + continue; + return 1; + } + return 0; +} + +/* + * parent_has_perm: + * when adding a new allow rule to a device whitelist, the rule + * must be allowed in the parent device + */ +static int parent_has_perm(struct cgroup *childcg, + struct dev_whitelist_item *wh) +{ + struct cgroup *pcg = childcg->parent; + struct dev_cgroup *parent; + int ret; + + if (!pcg) + return 1; + parent = cgroup_to_devcgroup(pcg); + spin_lock(&parent->lock); + ret = may_access_whitelist(parent, wh); + spin_unlock(&parent->lock); + return ret; +} + +/* + * Modify the whitelist using allow/deny rules. + * CAP_SYS_ADMIN is needed for this. It's at least separate from CAP_MKNOD + * so we can give a container CAP_MKNOD to let it create devices but not + * modify the whitelist. + * It seems likely we'll want to add a CAP_CONTAINER capability to allow + * us to also grant CAP_SYS_ADMIN to containers without giving away the + * device whitelist controls, but for now we'll stick with CAP_SYS_ADMIN + * + * Taking rules away is always allowed (given CAP_SYS_ADMIN). Granting + * new access is only allowed if you're in the top-level cgroup, or your + * parent cgroup has the access you're asking for. + */ +static ssize_t devcgroup_access_write(struct cgroup *cgroup, struct cftype *cft, + struct file *file, const char __user *userbuf, + size_t nbytes, loff_t *ppos) +{ + struct cgroup *cur_cgroup; + struct dev_cgroup *devcgroup, *cur_devcgroup; + int filetype = cft->private; + char *buffer, *b; + int retval = 0, count; + struct dev_whitelist_item wh; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + devcgroup = cgroup_to_devcgroup(cgroup); + cur_cgroup = task_cgroup(current, devices_subsys.subsys_id); + cur_devcgroup = cgroup_to_devcgroup(cur_cgroup); + + buffer = kmalloc(nbytes+1, GFP_KERNEL); + if (!buffer) + return -ENOMEM; + + if (copy_from_user(buffer, userbuf, nbytes)) { + retval = -EFAULT; + goto out1; + } + buffer[nbytes] = 0; /* nul-terminate */ + + cgroup_lock(); + if (cgroup_is_removed(cgroup)) { + retval = -ENODEV; + goto out2; + } + + memset(&wh, 0, sizeof(wh)); + b = buffer; + + switch (*b) { + case 'a': + wh.type = DEV_ALL; + wh.access = ACC_MASK; + goto handle; + case 'b': + wh.type = DEV_BLOCK; + break; + case 'c': + wh.type = DEV_CHAR; + break; + default: + retval = -EINVAL; + goto out2; + } + b++; + if (!isspace(*b)) { + retval = -EINVAL; + goto out2; + } + b++; + if (*b == '*') { + wh.major = ~0; + b++; + } else if (isdigit(*b)) { + wh.major = 0; + while (isdigit(*b)) { + wh.major = wh.major*10+(*b-'0'); + b++; + } + } else { + retval = -EINVAL; + goto out2; + } + if (*b != ':') { + retval = -EINVAL; + goto out2; + } + b++; + + /* read minor */ + if (*b == '*') { + wh.minor = ~0; + b++; + } else if (isdigit(*b)) { + wh.minor = 0; + while (isdigit(*b)) { + wh.minor = wh.minor*10+(*b-'0'); + b++; + } + } else { + retval = -EINVAL; + goto out2; + } + if (!isspace(*b)) { + retval = -EINVAL; + goto out2; + } + for (b++, count = 0; count < 3; count++, b++) { + switch (*b) { + case 'r': + wh.access |= ACC_READ; + break; + case 'w': + wh.access |= ACC_WRITE; + break; + case 'm': + wh.access |= ACC_MKNOD; + break; + case '\n': + case '\0': + count = 3; + break; + default: + retval = -EINVAL; + goto out2; + } + } + +handle: + retval = 0; + switch (filetype) { + case DEVCG_ALLOW: + if (!parent_has_perm(cgroup, &wh)) + retval = -EPERM; + else + retval = dev_whitelist_add(devcgroup, &wh); + break; + case DEVCG_DENY: + dev_whitelist_rm(devcgroup, &wh); + break; + default: + retval = -EINVAL; + goto out2; + } + + if (retval == 0) + retval = nbytes; + +out2: + cgroup_unlock(); +out1: + kfree(buffer); + return retval; +} + +static struct cftype dev_cgroup_files[] = { + { + .name = "allow", + .read = devcgroup_access_read, + .write = devcgroup_access_write, + .private = DEVCG_ALLOW, + }, + { + .name = "deny", + .write = devcgroup_access_write, + .private = DEVCG_DENY, + }, +}; + +static int devcgroup_populate(struct cgroup_subsys *ss, + struct cgroup *cgroup) +{ + return cgroup_add_files(cgroup, ss, dev_cgroup_files, + ARRAY_SIZE(dev_cgroup_files)); +} + +struct cgroup_subsys devices_subsys = { + .name = "devices", + .can_attach = devcgroup_can_attach, + .create = devcgroup_create, + .destroy = devcgroup_destroy, + .populate = devcgroup_populate, + .subsys_id = devices_subsys_id, +}; + +int devcgroup_inode_permission(struct inode *inode, int mask) +{ + struct cgroup *cgroup; + struct dev_cgroup *dev_cgroup; + struct dev_whitelist_item *wh; + + dev_t device = inode->i_rdev; + if (!device) + return 0; + if (!S_ISBLK(inode->i_mode) && !S_ISCHR(inode->i_mode)) + return 0; + cgroup = task_cgroup(current, devices_subsys.subsys_id); + dev_cgroup = cgroup_to_devcgroup(cgroup); + if (!dev_cgroup) + return 0; + + spin_lock(&dev_cgroup->lock); + list_for_each_entry(wh, &dev_cgroup->whitelist, list) { + if (wh->type & DEV_ALL) + goto acc_check; + if ((wh->type & DEV_BLOCK) && !S_ISBLK(inode->i_mode)) + continue; + if ((wh->type & DEV_CHAR) && !S_ISCHR(inode->i_mode)) + continue; + if (wh->major != ~0 && wh->major != imajor(inode)) + continue; + if (wh->minor != ~0 && wh->minor != iminor(inode)) + continue; +acc_check: + if ((mask & MAY_WRITE) && !(wh->access & ACC_WRITE)) + continue; + if ((mask & MAY_READ) && !(wh->access & ACC_READ)) + continue; + spin_unlock(&dev_cgroup->lock); + return 0; + } + spin_unlock(&dev_cgroup->lock); + + return -EPERM; +} + +int devcgroup_inode_mknod(int mode, dev_t dev) +{ + struct cgroup *cgroup; + struct dev_cgroup *dev_cgroup; + struct dev_whitelist_item *wh; + + cgroup = task_cgroup(current, devices_subsys.subsys_id); + dev_cgroup = cgroup_to_devcgroup(cgroup); + if (!dev_cgroup) + return 0; + + spin_lock(&dev_cgroup->lock); + list_for_each_entry(wh, &dev_cgroup->whitelist, list) { + if (wh->type & DEV_ALL) + goto acc_check; + if ((wh->type & DEV_BLOCK) && !S_ISBLK(mode)) + continue; + if ((wh->type & DEV_CHAR) && !S_ISCHR(mode)) + continue; + if (wh->major != ~0 && wh->major != MAJOR(dev)) + continue; + if (wh->minor != ~0 && wh->minor != MINOR(dev)) + continue; +acc_check: + if (!(wh->access & ACC_MKNOD)) + continue; + spin_unlock(&dev_cgroup->lock); + return 0; + } + spin_unlock(&dev_cgroup->lock); + return -EPERM; +} -- cgit v1.2.3 From cf475ad28ac35cc9ba612d67158f29b73b38b05d Mon Sep 17 00:00:00 2001 From: Balbir Singh Date: Tue, 29 Apr 2008 01:00:16 -0700 Subject: cgroups: add an owner to the mm_struct Remove the mem_cgroup member from mm_struct and instead adds an owner. This approach was suggested by Paul Menage. The advantage of this approach is that, once the mm->owner is known, using the subsystem id, the cgroup can be determined. It also allows several control groups that are virtually grouped by mm_struct, to exist independent of the memory controller i.e., without adding mem_cgroup's for each controller, to mm_struct. A new config option CONFIG_MM_OWNER is added and the memory resource controller selects this config option. This patch also adds cgroup callbacks to notify subsystems when mm->owner changes. The mm_cgroup_changed callback is called with the task_lock() of the new task held and is called just prior to changing the mm->owner. I am indebted to Paul Menage for the several reviews of this patchset and helping me make it lighter and simpler. This patch was tested on a powerpc box, it was compiled with both the MM_OWNER config turned on and off. After the thread group leader exits, it's moved to init_css_state by cgroup_exit(), thus all future charges from runnings threads would be redirected to the init_css_set's subsystem. Signed-off-by: Balbir Singh Cc: Pavel Emelianov Cc: Hugh Dickins Cc: Sudhir Kumar Cc: YAMAMOTO Takashi Cc: Hirokazu Takahashi Cc: David Rientjes , Cc: Balbir Singh Acked-by: KAMEZAWA Hiroyuki Acked-by: Pekka Enberg Reviewed-by: Paul Menage Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 1 + include/linux/cgroup.h | 15 +++++++++ include/linux/memcontrol.h | 16 ++------- include/linux/mm_types.h | 5 +-- include/linux/sched.h | 13 ++++++++ init/Kconfig | 7 ++++ init/main.c | 1 + kernel/cgroup.c | 30 +++++++++++++++++ kernel/exit.c | 83 ++++++++++++++++++++++++++++++++++++++++++++++ kernel/fork.c | 11 ++++-- mm/memcontrol.c | 28 +++------------- 11 files changed, 169 insertions(+), 41 deletions(-) (limited to 'fs') diff --git a/fs/exec.c b/fs/exec.c index 7768453dc986..711bc45d789c 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -735,6 +735,7 @@ static int exec_mmap(struct mm_struct *mm) tsk->active_mm = mm; activate_mm(active_mm, mm); task_unlock(tsk); + mm_update_next_owner(mm); arch_pick_mmap_layout(mm); if (old_mm) { up_read(&old_mm->mmap_sem); diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 095248082b7e..e155aa78d859 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -305,6 +305,12 @@ struct cgroup_subsys { struct cgroup *cgrp); void (*post_clone)(struct cgroup_subsys *ss, struct cgroup *cgrp); void (*bind)(struct cgroup_subsys *ss, struct cgroup *root); + /* + * This routine is called with the task_lock of mm->owner held + */ + void (*mm_owner_changed)(struct cgroup_subsys *ss, + struct cgroup *old, + struct cgroup *new); int subsys_id; int active; int disabled; @@ -390,4 +396,13 @@ static inline int cgroupstats_build(struct cgroupstats *stats, #endif /* !CONFIG_CGROUPS */ +#ifdef CONFIG_MM_OWNER +extern void +cgroup_mm_owner_callbacks(struct task_struct *old, struct task_struct *new); +#else /* !CONFIG_MM_OWNER */ +static inline void +cgroup_mm_owner_callbacks(struct task_struct *old, struct task_struct *new) +{ +} +#endif /* CONFIG_MM_OWNER */ #endif /* _LINUX_CGROUP_H */ diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 8b1c4295848b..e6608776bc96 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -27,9 +27,6 @@ struct mm_struct; #ifdef CONFIG_CGROUP_MEM_RES_CTLR -extern void mm_init_cgroup(struct mm_struct *mm, struct task_struct *p); -extern void mm_free_cgroup(struct mm_struct *mm); - #define page_reset_bad_cgroup(page) ((page)->page_cgroup = 0) extern struct page_cgroup *page_get_page_cgroup(struct page *page); @@ -48,8 +45,10 @@ extern unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan, extern void mem_cgroup_out_of_memory(struct mem_cgroup *mem, gfp_t gfp_mask); int task_in_mem_cgroup(struct task_struct *task, const struct mem_cgroup *mem); +extern struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p); + #define mm_match_cgroup(mm, cgroup) \ - ((cgroup) == rcu_dereference((mm)->mem_cgroup)) + ((cgroup) == mem_cgroup_from_task((mm)->owner)) extern int mem_cgroup_prepare_migration(struct page *page); extern void mem_cgroup_end_migration(struct page *page); @@ -73,15 +72,6 @@ extern long mem_cgroup_calc_reclaim_inactive(struct mem_cgroup *mem, struct zone *zone, int priority); #else /* CONFIG_CGROUP_MEM_RES_CTLR */ -static inline void mm_init_cgroup(struct mm_struct *mm, - struct task_struct *p) -{ -} - -static inline void mm_free_cgroup(struct mm_struct *mm) -{ -} - static inline void page_reset_bad_cgroup(struct page *page) { } diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index e2bae8dde35a..bc97bd54f606 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -225,8 +225,9 @@ struct mm_struct { /* aio bits */ rwlock_t ioctx_list_lock; /* aio lock */ struct kioctx *ioctx_list; -#ifdef CONFIG_CGROUP_MEM_RES_CTLR - struct mem_cgroup *mem_cgroup; +#ifdef CONFIG_MM_OWNER + struct task_struct *owner; /* The thread group leader that */ + /* owns the mm_struct. */ #endif }; diff --git a/include/linux/sched.h b/include/linux/sched.h index 024d72b47a0c..1d02babdb2c7 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2148,6 +2148,19 @@ static inline void migration_init(void) #define TASK_SIZE_OF(tsk) TASK_SIZE #endif +#ifdef CONFIG_MM_OWNER +extern void mm_update_next_owner(struct mm_struct *mm); +extern void mm_init_owner(struct mm_struct *mm, struct task_struct *p); +#else +static inline void mm_update_next_owner(struct mm_struct *mm) +{ +} + +static inline void mm_init_owner(struct mm_struct *mm, struct task_struct *p) +{ +} +#endif /* CONFIG_MM_OWNER */ + #endif /* __KERNEL__ */ #endif diff --git a/init/Kconfig b/init/Kconfig index a3457926342a..98fa96eac415 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -378,9 +378,13 @@ config RESOURCE_COUNTERS infrastructure that works with cgroups depends on CGROUPS +config MM_OWNER + bool + config CGROUP_MEM_RES_CTLR bool "Memory Resource Controller for Control Groups" depends on CGROUPS && RESOURCE_COUNTERS + select MM_OWNER help Provides a memory resource controller that manages both page cache and RSS memory. @@ -393,6 +397,9 @@ config CGROUP_MEM_RES_CTLR Only enable when you're ok with these trade offs and really sure you need the memory resource controller. + This config option also selects MM_OWNER config option, which + could in turn add some fork/exit overhead. + config SYSFS_DEPRECATED bool diff --git a/init/main.c b/init/main.c index 1116d2f40cc1..c62c98f381f2 100644 --- a/init/main.c +++ b/init/main.c @@ -559,6 +559,7 @@ asmlinkage void __init start_kernel(void) printk(KERN_NOTICE); printk(linux_banner); setup_arch(&command_line); + mm_init_owner(&init_mm, &init_task); setup_command_line(command_line); unwind_setup(); setup_per_cpu_areas(); diff --git a/kernel/cgroup.c b/kernel/cgroup.c index abc433772e5a..b9d467d83fc1 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -119,6 +119,7 @@ static int root_count; * be called. */ static int need_forkexit_callback; +static int need_mm_owner_callback __read_mostly; /* convenient tests for these bits */ inline int cgroup_is_removed(const struct cgroup *cgrp) @@ -2498,6 +2499,7 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss) init_css_set.subsys[ss->subsys_id] = dummytop->subsys[ss->subsys_id]; need_forkexit_callback |= ss->fork || ss->exit; + need_mm_owner_callback |= !!ss->mm_owner_changed; /* At system boot, before all subsystems have been * registered, no tasks have been forked, so we don't @@ -2748,6 +2750,34 @@ void cgroup_fork_callbacks(struct task_struct *child) } } +#ifdef CONFIG_MM_OWNER +/** + * cgroup_mm_owner_callbacks - run callbacks when the mm->owner changes + * @p: the new owner + * + * Called on every change to mm->owner. mm_init_owner() does not + * invoke this routine, since it assigns the mm->owner the first time + * and does not change it. + */ +void cgroup_mm_owner_callbacks(struct task_struct *old, struct task_struct *new) +{ + struct cgroup *oldcgrp, *newcgrp; + + if (need_mm_owner_callback) { + int i; + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { + struct cgroup_subsys *ss = subsys[i]; + oldcgrp = task_cgroup(old, ss->subsys_id); + newcgrp = task_cgroup(new, ss->subsys_id); + if (oldcgrp == newcgrp) + continue; + if (ss->mm_owner_changed) + ss->mm_owner_changed(ss, oldcgrp, newcgrp); + } + } +} +#endif /* CONFIG_MM_OWNER */ + /** * cgroup_post_fork - called on a new task after adding it to the task list * @child: the task in question diff --git a/kernel/exit.c b/kernel/exit.c index 2a9d98c641ac..ae0f2c4e452b 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -557,6 +557,88 @@ void exit_fs(struct task_struct *tsk) EXPORT_SYMBOL_GPL(exit_fs); +#ifdef CONFIG_MM_OWNER +/* + * Task p is exiting and it owned mm, lets find a new owner for it + */ +static inline int +mm_need_new_owner(struct mm_struct *mm, struct task_struct *p) +{ + /* + * If there are other users of the mm and the owner (us) is exiting + * we need to find a new owner to take on the responsibility. + */ + if (!mm) + return 0; + if (atomic_read(&mm->mm_users) <= 1) + return 0; + if (mm->owner != p) + return 0; + return 1; +} + +void mm_update_next_owner(struct mm_struct *mm) +{ + struct task_struct *c, *g, *p = current; + +retry: + if (!mm_need_new_owner(mm, p)) + return; + + read_lock(&tasklist_lock); + /* + * Search in the children + */ + list_for_each_entry(c, &p->children, sibling) { + if (c->mm == mm) + goto assign_new_owner; + } + + /* + * Search in the siblings + */ + list_for_each_entry(c, &p->parent->children, sibling) { + if (c->mm == mm) + goto assign_new_owner; + } + + /* + * Search through everything else. We should not get + * here often + */ + do_each_thread(g, c) { + if (c->mm == mm) + goto assign_new_owner; + } while_each_thread(g, c); + + read_unlock(&tasklist_lock); + return; + +assign_new_owner: + BUG_ON(c == p); + get_task_struct(c); + /* + * The task_lock protects c->mm from changing. + * We always want mm->owner->mm == mm + */ + task_lock(c); + /* + * Delay read_unlock() till we have the task_lock() + * to ensure that c does not slip away underneath us + */ + read_unlock(&tasklist_lock); + if (c->mm != mm) { + task_unlock(c); + put_task_struct(c); + goto retry; + } + cgroup_mm_owner_callbacks(mm->owner, c); + mm->owner = c; + task_unlock(c); + put_task_struct(c); +} +#endif /* CONFIG_MM_OWNER */ + /* * Turn us into a lazy TLB process if we * aren't already.. @@ -596,6 +678,7 @@ static void exit_mm(struct task_struct * tsk) /* We don't want this task to be frozen prematurely */ clear_freeze_flag(tsk); task_unlock(tsk); + mm_update_next_owner(mm); mmput(mm); } diff --git a/kernel/fork.c b/kernel/fork.c index 6067e429f281..156db96ff754 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -381,14 +381,13 @@ static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p) mm->ioctx_list = NULL; mm->free_area_cache = TASK_UNMAPPED_BASE; mm->cached_hole_size = ~0UL; - mm_init_cgroup(mm, p); + mm_init_owner(mm, p); if (likely(!mm_alloc_pgd(mm))) { mm->def_flags = 0; return mm; } - mm_free_cgroup(mm); free_mm(mm); return NULL; } @@ -438,7 +437,6 @@ void mmput(struct mm_struct *mm) spin_unlock(&mmlist_lock); } put_swap_token(mm); - mm_free_cgroup(mm); mmdrop(mm); } } @@ -982,6 +980,13 @@ static void rt_mutex_init_task(struct task_struct *p) #endif } +#ifdef CONFIG_MM_OWNER +void mm_init_owner(struct mm_struct *mm, struct task_struct *p) +{ + mm->owner = p; +} +#endif /* CONFIG_MM_OWNER */ + /* * This creates a new process as a copy of the old one, * but does not actually start it yet. diff --git a/mm/memcontrol.c b/mm/memcontrol.c index d12795cc7622..49d80814798b 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -236,26 +236,12 @@ static struct mem_cgroup *mem_cgroup_from_cont(struct cgroup *cont) css); } -static struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p) +struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p) { return container_of(task_subsys_state(p, mem_cgroup_subsys_id), struct mem_cgroup, css); } -void mm_init_cgroup(struct mm_struct *mm, struct task_struct *p) -{ - struct mem_cgroup *mem; - - mem = mem_cgroup_from_task(p); - css_get(&mem->css); - mm->mem_cgroup = mem; -} - -void mm_free_cgroup(struct mm_struct *mm) -{ - css_put(&mm->mem_cgroup->css); -} - static inline int page_cgroup_locked(struct page *page) { return bit_spin_is_locked(PAGE_CGROUP_LOCK_BIT, &page->page_cgroup); @@ -476,6 +462,7 @@ unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan, int zid = zone_idx(z); struct mem_cgroup_per_zone *mz; + BUG_ON(!mem_cont); mz = mem_cgroup_zoneinfo(mem_cont, nid, zid); if (active) src = &mz->active_list; @@ -574,7 +561,7 @@ retry: mm = &init_mm; rcu_read_lock(); - mem = rcu_dereference(mm->mem_cgroup); + mem = mem_cgroup_from_task(rcu_dereference(mm->owner)); /* * For every charge from the cgroup, increment reference count */ @@ -985,10 +972,9 @@ mem_cgroup_create(struct cgroup_subsys *ss, struct cgroup *cont) struct mem_cgroup *mem; int node; - if (unlikely((cont->parent) == NULL)) { + if (unlikely((cont->parent) == NULL)) mem = &init_mem_cgroup; - init_mm.mem_cgroup = mem; - } else + else mem = kzalloc(sizeof(struct mem_cgroup), GFP_KERNEL); if (mem == NULL) @@ -1067,10 +1053,6 @@ static void mem_cgroup_move_task(struct cgroup_subsys *ss, if (!thread_group_leader(p)) goto out; - css_get(&mem->css); - rcu_assign_pointer(mm->mem_cgroup, mem); - css_put(&old_mem->css); - out: mmput(mm); } -- cgit v1.2.3 From 6970c8eff85dd450e7eff69dad710dcf594b1bb8 Mon Sep 17 00:00:00 2001 From: Cyrill Gorcunov Date: Tue, 29 Apr 2008 01:01:18 -0700 Subject: BINFMT: fill_elf_header cleanup - use straight memset first This patch does simplify fill_elf_header function by setting to zero the whole elf header first. So we fillup the fields we really need only. before: text data bss dec hex filename 11735 80 0 11815 2e27 fs/binfmt_elf.o after: text data bss dec hex filename 11710 80 0 11790 2e0e fs/binfmt_elf.o viola, 25 bytes of text is freed Signed-off-by: Cyrill Gorcunov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/binfmt_elf.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 9924581df6f6..6cb4efbae885 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1255,26 +1255,23 @@ static int writenote(struct memelfnote *men, struct file *file, static void fill_elf_header(struct elfhdr *elf, int segs, u16 machine, u32 flags, u8 osabi) { + memset(elf, 0, sizeof(*elf)); + memcpy(elf->e_ident, ELFMAG, SELFMAG); elf->e_ident[EI_CLASS] = ELF_CLASS; elf->e_ident[EI_DATA] = ELF_DATA; elf->e_ident[EI_VERSION] = EV_CURRENT; elf->e_ident[EI_OSABI] = ELF_OSABI; - memset(elf->e_ident+EI_PAD, 0, EI_NIDENT-EI_PAD); elf->e_type = ET_CORE; elf->e_machine = machine; elf->e_version = EV_CURRENT; - elf->e_entry = 0; elf->e_phoff = sizeof(struct elfhdr); - elf->e_shoff = 0; elf->e_flags = flags; elf->e_ehsize = sizeof(struct elfhdr); elf->e_phentsize = sizeof(struct elf_phdr); elf->e_phnum = segs; - elf->e_shentsize = 0; - elf->e_shnum = 0; - elf->e_shstrndx = 0; + return; } -- cgit v1.2.3 From 4220b7fe89f8c0623e09168ab81dd0da2fdadd72 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Tue, 29 Apr 2008 01:01:18 -0700 Subject: elf: fix shadowed variables in fs/binfmt_elf.c Fix these sparse warings: fs/binfmt_elf.c:1749:29: warning: symbol 'tmp' shadows an earlier one fs/binfmt_elf.c:1734:28: originally declared here fs/binfmt_elf.c:2009:26: warning: symbol 'vma' shadows an earlier one fs/binfmt_elf.c:1892:24: originally declared here [akpm@linux-foundation.org: chose better variable name] Signed-off-by: WANG Cong Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/binfmt_elf.c | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 6cb4efbae885..b25707fee2cc 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1722,26 +1722,25 @@ static int fill_note_info(struct elfhdr *elf, int phdrs, info->thread_status_size = 0; if (signr) { - struct elf_thread_status *tmp; + struct elf_thread_status *ets; rcu_read_lock(); do_each_thread(g, p) if (current->mm == p->mm && current != p) { - tmp = kzalloc(sizeof(*tmp), GFP_ATOMIC); - if (!tmp) { + ets = kzalloc(sizeof(*ets), GFP_ATOMIC); + if (!ets) { rcu_read_unlock(); return 0; } - tmp->thread = p; - list_add(&tmp->list, &info->thread_list); + ets->thread = p; + list_add(&ets->list, &info->thread_list); } while_each_thread(g, p); rcu_read_unlock(); list_for_each(t, &info->thread_list) { - struct elf_thread_status *tmp; int sz; - tmp = list_entry(t, struct elf_thread_status, list); - sz = elf_dump_thread_status(signr, tmp); + ets = list_entry(t, struct elf_thread_status, list); + sz = elf_dump_thread_status(signr, ets); info->thread_status_size += sz; } } @@ -1997,10 +1996,10 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, un for (addr = vma->vm_start; addr < end; addr += PAGE_SIZE) { struct page *page; - struct vm_area_struct *vma; + struct vm_area_struct *tmp_vma; if (get_user_pages(current, current->mm, addr, 1, 0, 1, - &page, &vma) <= 0) { + &page, &tmp_vma) <= 0) { DUMP_SEEK(PAGE_SIZE); } else { if (page == ZERO_PAGE(0)) { @@ -2010,7 +2009,7 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, un } } else { void *kaddr; - flush_cache_page(vma, addr, + flush_cache_page(tmp_vma, addr, page_to_pfn(page)); kaddr = kmap(page); if ((size += PAGE_SIZE) > limit || -- cgit v1.2.3 From e93b4ea20adb20f1f1f07f10ba5d7dd739d2843e Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 29 Apr 2008 01:01:35 -0700 Subject: proc: print more information when removing non-empty directories This usually saves one recompile to insert similar printk like below. :) Sample nastygram: remove_proc_entry: removing non-empty directory '/proc/foo', leaking at least 'bar' ------------[ cut here ]------------ WARNING: at fs/proc/generic.c:776 remove_proc_entry+0x18a/0x200() Modules linked in: foo(-) container fan battery dock sbs ac sbshc backlight ipv6 loop af_packet amd_rng sr_mod i2c_amd8111 i2c_amd756 cdrom i2c_core button thermal processor Pid: 3034, comm: rmmod Tainted: G M 2.6.25-rc1 #5 Call Trace: [] warn_on_slowpath+0x64/0x90 [] printk+0x4e/0x60 [] remove_proc_entry+0x18a/0x200 [] mutex_lock_nested+0x1c8/0x2d0 [] __try_stop_module+0x0/0x40 [] sys_delete_module+0x14d/0x200 [] lockdep_sys_exit_thunk+0x35/0x67 [] __up_read+0x27/0xa0 [] trace_hardirqs_on_thunk+0x35/0x3a [] system_call_after_swapgs+0x7b/0x80 ---[ end trace 10ef850597e89c54 ]--- Signed-off-by: Alexey Dobriyan Cc: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/generic.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/proc/generic.c b/fs/proc/generic.c index a36ad3c75cf4..f501f3211abc 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -777,7 +777,12 @@ continue_removing: if (S_ISDIR(de->mode)) parent->nlink--; de->nlink = 0; - WARN_ON(de->subdir); + if (de->subdir) { + printk(KERN_WARNING "%s: removing non-empty directory " + "'%s/%s', leaking at least '%s'\n", __func__, + de->parent->name, de->name, de->subdir->name); + WARN_ON(1); + } if (atomic_dec_and_test(&de->count)) free_proc_entry(de); break; -- cgit v1.2.3 From 925d1c401fa6cfd0df5d2e37da8981494ccdec07 Mon Sep 17 00:00:00 2001 From: Matt Helsley Date: Tue, 29 Apr 2008 01:01:36 -0700 Subject: procfs task exe symlink The kernel implements readlink of /proc/pid/exe by getting the file from the first executable VMA. Then the path to the file is reconstructed and reported as the result. Because of the VMA walk the code is slightly different on nommu systems. This patch avoids separate /proc/pid/exe code on nommu systems. Instead of walking the VMAs to find the first executable file-backed VMA we store a reference to the exec'd file in the mm_struct. That reference would prevent the filesystem holding the executable file from being unmounted even after unmapping the VMAs. So we track the number of VM_EXECUTABLE VMAs and drop the new reference when the last one is unmapped. This avoids pinning the mounted filesystem. [akpm@linux-foundation.org: improve comments] [yamamoto@valinux.co.jp: fix dup_mmap] Signed-off-by: Matt Helsley Cc: Oleg Nesterov Cc: David Howells Cc:"Eric W. Biederman" Cc: Christoph Hellwig Cc: Al Viro Cc: Hugh Dickins Signed-off-by: YAMAMOTO Takashi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/binfmt_flat.c | 3 +- fs/exec.c | 2 ++ fs/proc/base.c | 75 ++++++++++++++++++++++++++++++++++++++++++++++++ fs/proc/internal.h | 1 - fs/proc/task_mmu.c | 34 ---------------------- fs/proc/task_nommu.c | 34 ---------------------- include/linux/mm.h | 13 +++++++++ include/linux/mm_types.h | 6 ++++ include/linux/proc_fs.h | 20 ++++++++++++- kernel/fork.c | 3 ++ mm/mmap.c | 24 +++++++++++++--- mm/nommu.c | 23 +++++++++++---- 12 files changed, 157 insertions(+), 81 deletions(-) (limited to 'fs') diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index c12cc362fd3b..3b40d45a3a16 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -531,7 +531,8 @@ static int load_flat_file(struct linux_binprm * bprm, DBG_FLT("BINFMT_FLAT: ROM mapping of file (we hope)\n"); down_write(¤t->mm->mmap_sem); - textpos = do_mmap(bprm->file, 0, text_len, PROT_READ|PROT_EXEC, MAP_PRIVATE, 0); + textpos = do_mmap(bprm->file, 0, text_len, PROT_READ|PROT_EXEC, + MAP_PRIVATE|MAP_EXECUTABLE, 0); up_write(¤t->mm->mmap_sem); if (!textpos || textpos >= (unsigned long) -4096) { if (!textpos) diff --git a/fs/exec.c b/fs/exec.c index 711bc45d789c..a13883903ee9 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -964,6 +964,8 @@ int flush_old_exec(struct linux_binprm * bprm) if (retval) goto out; + set_mm_exe_file(bprm->mm, bprm->file); + /* * Release all of the old mmap stuff */ diff --git a/fs/proc/base.c b/fs/proc/base.c index c5e412a00b17..b48ddb119945 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1181,6 +1181,81 @@ static const struct file_operations proc_pid_sched_operations = { #endif +/* + * We added or removed a vma mapping the executable. The vmas are only mapped + * during exec and are not mapped with the mmap system call. + * Callers must hold down_write() on the mm's mmap_sem for these + */ +void added_exe_file_vma(struct mm_struct *mm) +{ + mm->num_exe_file_vmas++; +} + +void removed_exe_file_vma(struct mm_struct *mm) +{ + mm->num_exe_file_vmas--; + if ((mm->num_exe_file_vmas == 0) && mm->exe_file){ + fput(mm->exe_file); + mm->exe_file = NULL; + } + +} + +void set_mm_exe_file(struct mm_struct *mm, struct file *new_exe_file) +{ + if (new_exe_file) + get_file(new_exe_file); + if (mm->exe_file) + fput(mm->exe_file); + mm->exe_file = new_exe_file; + mm->num_exe_file_vmas = 0; +} + +struct file *get_mm_exe_file(struct mm_struct *mm) +{ + struct file *exe_file; + + /* We need mmap_sem to protect against races with removal of + * VM_EXECUTABLE vmas */ + down_read(&mm->mmap_sem); + exe_file = mm->exe_file; + if (exe_file) + get_file(exe_file); + up_read(&mm->mmap_sem); + return exe_file; +} + +void dup_mm_exe_file(struct mm_struct *oldmm, struct mm_struct *newmm) +{ + /* It's safe to write the exe_file pointer without exe_file_lock because + * this is called during fork when the task is not yet in /proc */ + newmm->exe_file = get_mm_exe_file(oldmm); +} + +static int proc_exe_link(struct inode *inode, struct path *exe_path) +{ + struct task_struct *task; + struct mm_struct *mm; + struct file *exe_file; + + task = get_proc_task(inode); + if (!task) + return -ENOENT; + mm = get_task_mm(task); + put_task_struct(task); + if (!mm) + return -ENOENT; + exe_file = get_mm_exe_file(mm); + mmput(mm); + if (exe_file) { + *exe_path = exe_file->f_path; + path_get(&exe_file->f_path); + fput(exe_file); + return 0; + } else + return -ENOENT; +} + static void *proc_pid_follow_link(struct dentry *dentry, struct nameidata *nd) { struct inode *inode = dentry->d_inode; diff --git a/fs/proc/internal.h b/fs/proc/internal.h index bc72f5c8c47d..45abb9803988 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -48,7 +48,6 @@ extern int maps_protect; extern void create_seq_entry(char *name, mode_t mode, const struct file_operations *f); -extern int proc_exe_link(struct inode *, struct path *); extern int proc_tid_stat(struct seq_file *m, struct pid_namespace *ns, struct pid *pid, struct task_struct *task); extern int proc_tgid_stat(struct seq_file *m, struct pid_namespace *ns, diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 7415eeb7cc3a..e2b8e769f510 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -75,40 +75,6 @@ int task_statm(struct mm_struct *mm, int *shared, int *text, return mm->total_vm; } -int proc_exe_link(struct inode *inode, struct path *path) -{ - struct vm_area_struct * vma; - int result = -ENOENT; - struct task_struct *task = get_proc_task(inode); - struct mm_struct * mm = NULL; - - if (task) { - mm = get_task_mm(task); - put_task_struct(task); - } - if (!mm) - goto out; - down_read(&mm->mmap_sem); - - vma = mm->mmap; - while (vma) { - if ((vma->vm_flags & VM_EXECUTABLE) && vma->vm_file) - break; - vma = vma->vm_next; - } - - if (vma) { - *path = vma->vm_file->f_path; - path_get(&vma->vm_file->f_path); - result = 0; - } - - up_read(&mm->mmap_sem); - mmput(mm); -out: - return result; -} - static void pad_len_spaces(struct seq_file *m, int len) { len = 25 + sizeof(void*) * 6 - len; diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c index 8011528518bd..4b733f108455 100644 --- a/fs/proc/task_nommu.c +++ b/fs/proc/task_nommu.c @@ -103,40 +103,6 @@ int task_statm(struct mm_struct *mm, int *shared, int *text, return size; } -int proc_exe_link(struct inode *inode, struct path *path) -{ - struct vm_list_struct *vml; - struct vm_area_struct *vma; - struct task_struct *task = get_proc_task(inode); - struct mm_struct *mm = get_task_mm(task); - int result = -ENOENT; - - if (!mm) - goto out; - down_read(&mm->mmap_sem); - - vml = mm->context.vmlist; - vma = NULL; - while (vml) { - if ((vml->vma->vm_flags & VM_EXECUTABLE) && vml->vma->vm_file) { - vma = vml->vma; - break; - } - vml = vml->next; - } - - if (vma) { - *path = vma->vm_file->f_path; - path_get(&vma->vm_file->f_path); - result = 0; - } - - up_read(&mm->mmap_sem); - mmput(mm); -out: - return result; -} - /* * display mapping lines for a particular process's /proc/pid/maps */ diff --git a/include/linux/mm.h b/include/linux/mm.h index fef602d82722..c31a9cd2a30e 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1066,6 +1066,19 @@ extern void unlink_file_vma(struct vm_area_struct *); extern struct vm_area_struct *copy_vma(struct vm_area_struct **, unsigned long addr, unsigned long len, pgoff_t pgoff); extern void exit_mmap(struct mm_struct *); + +#ifdef CONFIG_PROC_FS +/* From fs/proc/base.c. callers must _not_ hold the mm's exe_file_lock */ +extern void added_exe_file_vma(struct mm_struct *mm); +extern void removed_exe_file_vma(struct mm_struct *mm); +#else +static inline void added_exe_file_vma(struct mm_struct *mm) +{} + +static inline void removed_exe_file_vma(struct mm_struct *mm) +{} +#endif /* CONFIG_PROC_FS */ + extern int may_expand_vm(struct mm_struct *mm, unsigned long npages); extern int install_special_mapping(struct mm_struct *mm, unsigned long addr, unsigned long len, diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index bc97bd54f606..eb7c16cc9559 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -229,6 +229,12 @@ struct mm_struct { struct task_struct *owner; /* The thread group leader that */ /* owns the mm_struct. */ #endif + +#ifdef CONFIG_PROC_FS + /* store ref to file /proc//exe symlink points to */ + struct file *exe_file; + unsigned long num_exe_file_vmas; +#endif }; #endif /* _LINUX_MM_TYPES_H */ diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index 9b6c935f69cf..65f2299b772b 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -9,7 +9,6 @@ struct net; struct completion; - /* * The proc filesystem constants/structures */ @@ -206,6 +205,12 @@ extern void proc_net_remove(struct net *net, const char *name); extern struct proc_dir_entry *proc_net_mkdir(struct net *net, const char *name, struct proc_dir_entry *parent); +/* While the {get|set|dup}_mm_exe_file functions are for mm_structs, they are + * only needed to implement /proc/|self/exe so we define them here. */ +extern void set_mm_exe_file(struct mm_struct *mm, struct file *new_exe_file); +extern struct file *get_mm_exe_file(struct mm_struct *mm); +extern void dup_mm_exe_file(struct mm_struct *oldmm, struct mm_struct *newmm); + #else #define proc_root_driver NULL @@ -255,6 +260,19 @@ static inline void pid_ns_release_proc(struct pid_namespace *ns) { } +static inline void set_mm_exe_file(struct mm_struct *mm, + struct file *new_exe_file) +{} + +static inline struct file *get_mm_exe_file(struct mm_struct *mm) +{ + return NULL; +} + +static inline void dup_mm_exe_file(struct mm_struct *oldmm, + struct mm_struct *newmm) +{} + #endif /* CONFIG_PROC_FS */ #if !defined(CONFIG_PROC_KCORE) diff --git a/kernel/fork.c b/kernel/fork.c index de5c16c6b6ec..068ffe007529 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -431,6 +431,7 @@ void mmput(struct mm_struct *mm) if (atomic_dec_and_test(&mm->mm_users)) { exit_aio(mm); exit_mmap(mm); + set_mm_exe_file(mm, NULL); if (!list_empty(&mm->mmlist)) { spin_lock(&mmlist_lock); list_del(&mm->mmlist); @@ -543,6 +544,8 @@ struct mm_struct *dup_mm(struct task_struct *tsk) if (init_new_context(tsk, mm)) goto fail_nocontext; + dup_mm_exe_file(oldmm, mm); + err = dup_mmap(mm, oldmm); if (err) goto free_pt; diff --git a/mm/mmap.c b/mm/mmap.c index 677d184b0d42..fac66337da2a 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -230,8 +230,11 @@ static struct vm_area_struct *remove_vma(struct vm_area_struct *vma) might_sleep(); if (vma->vm_ops && vma->vm_ops->close) vma->vm_ops->close(vma); - if (vma->vm_file) + if (vma->vm_file) { fput(vma->vm_file); + if (vma->vm_flags & VM_EXECUTABLE) + removed_exe_file_vma(vma->vm_mm); + } mpol_put(vma_policy(vma)); kmem_cache_free(vm_area_cachep, vma); return next; @@ -623,8 +626,11 @@ again: remove_next = 1 + (end > next->vm_end); spin_unlock(&mapping->i_mmap_lock); if (remove_next) { - if (file) + if (file) { fput(file); + if (next->vm_flags & VM_EXECUTABLE) + removed_exe_file_vma(mm); + } mm->map_count--; mpol_put(vma_policy(next)); kmem_cache_free(vm_area_cachep, next); @@ -1154,6 +1160,8 @@ munmap_back: error = file->f_op->mmap(file, vma); if (error) goto unmap_and_free_vma; + if (vm_flags & VM_EXECUTABLE) + added_exe_file_vma(mm); } else if (vm_flags & VM_SHARED) { error = shmem_zero_setup(vma); if (error) @@ -1185,6 +1193,8 @@ munmap_back: mpol_put(vma_policy(vma)); kmem_cache_free(vm_area_cachep, vma); fput(file); + if (vm_flags & VM_EXECUTABLE) + removed_exe_file_vma(mm); } else { vma_link(mm, vma, prev, rb_link, rb_parent); file = vma->vm_file; @@ -1817,8 +1827,11 @@ int split_vma(struct mm_struct * mm, struct vm_area_struct * vma, } vma_set_policy(new, pol); - if (new->vm_file) + if (new->vm_file) { get_file(new->vm_file); + if (vma->vm_flags & VM_EXECUTABLE) + added_exe_file_vma(mm); + } if (new->vm_ops && new->vm_ops->open) new->vm_ops->open(new); @@ -2135,8 +2148,11 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap, new_vma->vm_start = addr; new_vma->vm_end = addr + len; new_vma->vm_pgoff = pgoff; - if (new_vma->vm_file) + if (new_vma->vm_file) { get_file(new_vma->vm_file); + if (vma->vm_flags & VM_EXECUTABLE) + added_exe_file_vma(mm); + } if (new_vma->vm_ops && new_vma->vm_ops->open) new_vma->vm_ops->open(new_vma); vma_link(mm, new_vma, prev, rb_link, rb_parent); diff --git a/mm/nommu.c b/mm/nommu.c index 1d32fe89d57b..ef8c62cec697 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -966,8 +966,13 @@ unsigned long do_mmap_pgoff(struct file *file, INIT_LIST_HEAD(&vma->anon_vma_node); atomic_set(&vma->vm_usage, 1); - if (file) + if (file) { get_file(file); + if (vm_flags & VM_EXECUTABLE) { + added_exe_file_vma(current->mm); + vma->vm_mm = current->mm; + } + } vma->vm_file = file; vma->vm_flags = vm_flags; vma->vm_start = addr; @@ -1022,8 +1027,11 @@ unsigned long do_mmap_pgoff(struct file *file, up_write(&nommu_vma_sem); kfree(vml); if (vma) { - if (vma->vm_file) + if (vma->vm_file) { fput(vma->vm_file); + if (vma->vm_flags & VM_EXECUTABLE) + removed_exe_file_vma(vma->vm_mm); + } kfree(vma); } return ret; @@ -1053,7 +1061,7 @@ EXPORT_SYMBOL(do_mmap_pgoff); /* * handle mapping disposal for uClinux */ -static void put_vma(struct vm_area_struct *vma) +static void put_vma(struct mm_struct *mm, struct vm_area_struct *vma) { if (vma) { down_write(&nommu_vma_sem); @@ -1075,8 +1083,11 @@ static void put_vma(struct vm_area_struct *vma) realalloc -= kobjsize(vma); askedalloc -= sizeof(*vma); - if (vma->vm_file) + if (vma->vm_file) { fput(vma->vm_file); + if (vma->vm_flags & VM_EXECUTABLE) + removed_exe_file_vma(mm); + } kfree(vma); } @@ -1113,7 +1124,7 @@ int do_munmap(struct mm_struct *mm, unsigned long addr, size_t len) found: vml = *parent; - put_vma(vml->vma); + put_vma(mm, vml->vma); *parent = vml->next; realalloc -= kobjsize(vml); @@ -1158,7 +1169,7 @@ void exit_mmap(struct mm_struct * mm) while ((tmp = mm->context.vmlist)) { mm->context.vmlist = tmp->next; - put_vma(tmp->vma); + put_vma(mm, tmp->vma); realalloc -= kobjsize(tmp); askedalloc -= sizeof(*tmp); -- cgit v1.2.3 From 0d5c9f5f59a61cf8e98e2925cb5d81cbe7694305 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 29 Apr 2008 01:01:37 -0700 Subject: proc: switch to proc_create() Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/internal.h | 2 -- fs/proc/nommu.c | 2 +- fs/proc/proc_misc.c | 66 ++++++++++++++++++----------------------------------- fs/proc/proc_tty.c | 5 +--- 4 files changed, 24 insertions(+), 51 deletions(-) (limited to 'fs') diff --git a/fs/proc/internal.h b/fs/proc/internal.h index 45abb9803988..b1d6df671edf 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -46,8 +46,6 @@ extern int nommu_vma_show(struct seq_file *, struct vm_area_struct *); extern int maps_protect; -extern void create_seq_entry(char *name, mode_t mode, - const struct file_operations *f); extern int proc_tid_stat(struct seq_file *m, struct pid_namespace *ns, struct pid *pid, struct task_struct *task); extern int proc_tgid_stat(struct seq_file *m, struct pid_namespace *ns, diff --git a/fs/proc/nommu.c b/fs/proc/nommu.c index 941e95114b5a..79ecd281d2cb 100644 --- a/fs/proc/nommu.c +++ b/fs/proc/nommu.c @@ -137,7 +137,7 @@ static const struct file_operations proc_nommu_vma_list_operations = { static int __init proc_nommu_init(void) { - create_seq_entry("maps", S_IRUGO, &proc_nommu_vma_list_operations); + proc_create("maps", S_IRUGO, NULL, &proc_nommu_vma_list_operations); return 0; } diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index 441a32f0e5f2..c4137bb94a9e 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -826,14 +826,6 @@ static struct file_operations proc_kpageflags_operations = { struct proc_dir_entry *proc_root_kcore; -void create_seq_entry(char *name, mode_t mode, const struct file_operations *f) -{ - struct proc_dir_entry *entry; - entry = create_proc_entry(name, mode, NULL); - if (entry) - entry->proc_fops = f; -} - void __init proc_misc_init(void) { static struct { @@ -862,66 +854,52 @@ void __init proc_misc_init(void) /* And now for trickier ones */ #ifdef CONFIG_PRINTK - { - struct proc_dir_entry *entry; - entry = create_proc_entry("kmsg", S_IRUSR, &proc_root); - if (entry) - entry->proc_fops = &proc_kmsg_operations; - } + proc_create("kmsg", S_IRUSR, &proc_root, &proc_kmsg_operations); #endif - create_seq_entry("locks", 0, &proc_locks_operations); - create_seq_entry("devices", 0, &proc_devinfo_operations); - create_seq_entry("cpuinfo", 0, &proc_cpuinfo_operations); + proc_create("locks", 0, NULL, &proc_locks_operations); + proc_create("devices", 0, NULL, &proc_devinfo_operations); + proc_create("cpuinfo", 0, NULL, &proc_cpuinfo_operations); #ifdef CONFIG_BLOCK - create_seq_entry("partitions", 0, &proc_partitions_operations); + proc_create("partitions", 0, NULL, &proc_partitions_operations); #endif - create_seq_entry("stat", 0, &proc_stat_operations); - create_seq_entry("interrupts", 0, &proc_interrupts_operations); + proc_create("stat", 0, NULL, &proc_stat_operations); + proc_create("interrupts", 0, NULL, &proc_interrupts_operations); #ifdef CONFIG_SLABINFO - create_seq_entry("slabinfo",S_IWUSR|S_IRUGO,&proc_slabinfo_operations); + proc_create("slabinfo",S_IWUSR|S_IRUGO,NULL,&proc_slabinfo_operations); #ifdef CONFIG_DEBUG_SLAB_LEAK - create_seq_entry("slab_allocators", 0 ,&proc_slabstats_operations); + proc_create("slab_allocators", 0, NULL, &proc_slabstats_operations); #endif #endif #ifdef CONFIG_MMU proc_create("vmallocinfo", S_IRUSR, NULL, &proc_vmalloc_operations); #endif - create_seq_entry("buddyinfo",S_IRUGO, &fragmentation_file_operations); - create_seq_entry("pagetypeinfo", S_IRUGO, &pagetypeinfo_file_ops); - create_seq_entry("vmstat",S_IRUGO, &proc_vmstat_file_operations); - create_seq_entry("zoneinfo",S_IRUGO, &proc_zoneinfo_file_operations); + proc_create("buddyinfo", S_IRUGO, NULL, &fragmentation_file_operations); + proc_create("pagetypeinfo", S_IRUGO, NULL, &pagetypeinfo_file_ops); + proc_create("vmstat", S_IRUGO, NULL, &proc_vmstat_file_operations); + proc_create("zoneinfo", S_IRUGO, NULL, &proc_zoneinfo_file_operations); #ifdef CONFIG_BLOCK - create_seq_entry("diskstats", 0, &proc_diskstats_operations); + proc_create("diskstats", 0, NULL, &proc_diskstats_operations); #endif #ifdef CONFIG_MODULES - create_seq_entry("modules", 0, &proc_modules_operations); + proc_create("modules", 0, NULL, &proc_modules_operations); #endif #ifdef CONFIG_SCHEDSTATS - create_seq_entry("schedstat", 0, &proc_schedstat_operations); + proc_create("schedstat", 0, NULL, &proc_schedstat_operations); #endif #ifdef CONFIG_PROC_KCORE - proc_root_kcore = create_proc_entry("kcore", S_IRUSR, NULL); - if (proc_root_kcore) { - proc_root_kcore->proc_fops = &proc_kcore_operations; + proc_root_kcore = proc_create("kcore", S_IRUSR, NULL, &proc_kcore_operations); + if (proc_root_kcore) proc_root_kcore->size = (size_t)high_memory - PAGE_OFFSET + PAGE_SIZE; - } #endif #ifdef CONFIG_PROC_PAGE_MONITOR - create_seq_entry("kpagecount", S_IRUSR, &proc_kpagecount_operations); - create_seq_entry("kpageflags", S_IRUSR, &proc_kpageflags_operations); + proc_create("kpagecount", S_IRUSR, NULL, &proc_kpagecount_operations); + proc_create("kpageflags", S_IRUSR, NULL, &proc_kpageflags_operations); #endif #ifdef CONFIG_PROC_VMCORE - proc_vmcore = create_proc_entry("vmcore", S_IRUSR, NULL); - if (proc_vmcore) - proc_vmcore->proc_fops = &proc_vmcore_operations; + proc_vmcore = proc_create("vmcore", S_IRUSR, NULL, &proc_vmcore_operations); #endif #ifdef CONFIG_MAGIC_SYSRQ - { - struct proc_dir_entry *entry; - entry = create_proc_entry("sysrq-trigger", S_IWUSR, NULL); - if (entry) - entry->proc_fops = &proc_sysrq_trigger_operations; - } + proc_create("sysrq-trigger", S_IWUSR, NULL, &proc_sysrq_trigger_operations); #endif } diff --git a/fs/proc/proc_tty.c b/fs/proc/proc_tty.c index 49816e00b51a..63f45383035d 100644 --- a/fs/proc/proc_tty.c +++ b/fs/proc/proc_tty.c @@ -214,7 +214,6 @@ void proc_tty_unregister_driver(struct tty_driver *driver) */ void __init proc_tty_init(void) { - struct proc_dir_entry *entry; if (!proc_mkdir("tty", NULL)) return; proc_tty_ldisc = proc_mkdir("tty/ldisc", NULL); @@ -227,7 +226,5 @@ void __init proc_tty_init(void) proc_tty_driver = proc_mkdir_mode("tty/driver", S_IRUSR | S_IXUSR, NULL); create_proc_read_entry("tty/ldiscs", 0, NULL, tty_ldiscs_read_proc, NULL); - entry = create_proc_entry("tty/drivers", 0, NULL); - if (entry) - entry->proc_fops = &proc_tty_drivers_operations; + proc_create("tty/drivers", 0, NULL, &proc_tty_drivers_operations); } -- cgit v1.2.3 From 638fa202cdb207083a12d6f73e313605a8fc1037 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Tue, 29 Apr 2008 01:01:38 -0700 Subject: procfs: mem permission cleanup This cleans up the permission checks done for /proc/PID/mem i/o calls. It puts all the logic in a new function, check_mem_permission(). The old code repeated the (!MAY_PTRACE(task) || !ptrace_may_attach(task)) magical expression multiple times. The new function does all that work in one place, with clear comments. The old code called security_ptrace() twice on successful checks, once in MAY_PTRACE() and once in __ptrace_may_attach(). Now it's only called once, and only if all other checks have succeeded. Signed-off-by: Roland McGrath Cc: Alexey Dobriyan Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/base.c | 38 +++++++++++++++++++++++++++++--------- 1 file changed, 29 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/proc/base.c b/fs/proc/base.c index b48ddb119945..fcf02f2deeba 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -195,12 +195,32 @@ static int proc_root_link(struct inode *inode, struct path *path) return result; } -#define MAY_PTRACE(task) \ - (task == current || \ - (task->parent == current && \ - (task->ptrace & PT_PTRACED) && \ - (task_is_stopped_or_traced(task)) && \ - security_ptrace(current,task) == 0)) +/* + * Return zero if current may access user memory in @task, -error if not. + */ +static int check_mem_permission(struct task_struct *task) +{ + /* + * A task can always look at itself, in case it chooses + * to use system calls instead of load instructions. + */ + if (task == current) + return 0; + + /* + * If current is actively ptrace'ing, and would also be + * permitted to freshly attach with ptrace now, permit it. + */ + if (task->parent == current && (task->ptrace & PT_PTRACED) && + task_is_stopped_or_traced(task) && + ptrace_may_attach(task)) + return 0; + + /* + * Noone else is allowed. + */ + return -EPERM; +} struct mm_struct *mm_for_maps(struct task_struct *task) { @@ -722,7 +742,7 @@ static ssize_t mem_read(struct file * file, char __user * buf, if (!task) goto out_no_task; - if (!MAY_PTRACE(task) || !ptrace_may_attach(task)) + if (check_mem_permission(task)) goto out; ret = -ENOMEM; @@ -748,7 +768,7 @@ static ssize_t mem_read(struct file * file, char __user * buf, this_len = (count > PAGE_SIZE) ? PAGE_SIZE : count; retval = access_process_vm(task, src, page, this_len, 0); - if (!retval || !MAY_PTRACE(task) || !ptrace_may_attach(task)) { + if (!retval || check_mem_permission(task)) { if (!ret) ret = -EIO; break; @@ -792,7 +812,7 @@ static ssize_t mem_write(struct file * file, const char __user *buf, if (!task) goto out_no_task; - if (!MAY_PTRACE(task) || !ptrace_may_attach(task)) + if (check_mem_permission(task)) goto out; copied = -ENOMEM; -- cgit v1.2.3 From f649d6d32605c7573884613289fb3b9fbd4f99a1 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 29 Apr 2008 01:01:39 -0700 Subject: proc: simplify locking in remove_proc_entry() proc_subdir_lock protects only modifying and walking through PDE lists, so after we've found PDE to remove and actually removed it from lists, there is no need to hold proc_subdir_lock for the rest of operation. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/generic.c | 82 +++++++++++++++++++++++++++---------------------------- 1 file changed, 40 insertions(+), 42 deletions(-) (limited to 'fs') diff --git a/fs/proc/generic.c b/fs/proc/generic.c index f501f3211abc..45d0076bc08e 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -734,60 +734,58 @@ void free_proc_entry(struct proc_dir_entry *de) void remove_proc_entry(const char *name, struct proc_dir_entry *parent) { struct proc_dir_entry **p; - struct proc_dir_entry *de; + struct proc_dir_entry *de = NULL; const char *fn = name; int len; if (!parent && xlate_proc_name(name, &parent, &fn) != 0) - goto out; + return; len = strlen(fn); spin_lock(&proc_subdir_lock); for (p = &parent->subdir; *p; p=&(*p)->next ) { - if (!proc_match(len, fn, *p)) - continue; - de = *p; - *p = de->next; - de->next = NULL; - - spin_lock(&de->pde_unload_lock); - /* - * Stop accepting new callers into module. If you're - * dynamically allocating ->proc_fops, save a pointer somewhere. - */ - de->proc_fops = NULL; - /* Wait until all existing callers into module are done. */ - if (de->pde_users > 0) { - DECLARE_COMPLETION_ONSTACK(c); - - if (!de->pde_unload_completion) - de->pde_unload_completion = &c; - - spin_unlock(&de->pde_unload_lock); - spin_unlock(&proc_subdir_lock); + if (proc_match(len, fn, *p)) { + de = *p; + *p = de->next; + de->next = NULL; + break; + } + } + spin_unlock(&proc_subdir_lock); + if (!de) + return; - wait_for_completion(de->pde_unload_completion); + spin_lock(&de->pde_unload_lock); + /* + * Stop accepting new callers into module. If you're + * dynamically allocating ->proc_fops, save a pointer somewhere. + */ + de->proc_fops = NULL; + /* Wait until all existing callers into module are done. */ + if (de->pde_users > 0) { + DECLARE_COMPLETION_ONSTACK(c); + + if (!de->pde_unload_completion) + de->pde_unload_completion = &c; - spin_lock(&proc_subdir_lock); - goto continue_removing; - } spin_unlock(&de->pde_unload_lock); + wait_for_completion(de->pde_unload_completion); + + goto continue_removing; + } + spin_unlock(&de->pde_unload_lock); + continue_removing: - if (S_ISDIR(de->mode)) - parent->nlink--; - de->nlink = 0; - if (de->subdir) { - printk(KERN_WARNING "%s: removing non-empty directory " - "'%s/%s', leaking at least '%s'\n", __func__, - de->parent->name, de->name, de->subdir->name); - WARN_ON(1); - } - if (atomic_dec_and_test(&de->count)) - free_proc_entry(de); - break; + if (S_ISDIR(de->mode)) + parent->nlink--; + de->nlink = 0; + if (de->subdir) { + printk(KERN_WARNING "%s: removing non-empty directory " + "'%s/%s', leaking at least '%s'\n", __func__, + de->parent->name, de->name, de->subdir->name); + WARN_ON(1); } - spin_unlock(&proc_subdir_lock); -out: - return; + if (atomic_dec_and_test(&de->count)) + free_proc_entry(de); } -- cgit v1.2.3 From 7cee4e00e0f8aa7290266382ea903a5a1b92c9a1 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 29 Apr 2008 01:01:40 -0700 Subject: proc: less special case in xlate code If valid "parent" is passed to proc_create/remove_proc_entry(), then name of PDE should consist of only one path component, otherwise creation or or removal will fail. However, if NULL is passed as parent then create/remove accept full path as a argument. This is arbitrary restriction -- all infrastructure is in place. So, patch allows the following to succeed: create_proc_entry("foo/bar", 0, pde_baz); remove_proc_entry("baz/foo/bar", &proc_root); Also makes the following to behave identically: create_proc_entry("foo/bar", 0, NULL); create_proc_entry("foo/bar", 0, &proc_root); Discrepancy noticed by Den Lunev (IIRC). Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/generic.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 45d0076bc08e..3c6f5669523a 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -277,8 +277,11 @@ static int xlate_proc_name(const char *name, int len; int rtn = 0; + de = *ret; + if (!de) + de = &proc_root; + spin_lock(&proc_subdir_lock); - de = &proc_root; while (1) { next = strchr(cp, '/'); if (!next) @@ -582,7 +585,7 @@ static struct proc_dir_entry *__proc_create(struct proc_dir_entry **parent, /* make sure name is valid */ if (!name || !strlen(name)) goto out; - if (!(*parent) && xlate_proc_name(name, parent, &fn) != 0) + if (xlate_proc_name(name, parent, &fn) != 0) goto out; /* At this point there must not be any '/' characters beyond *fn */ @@ -738,7 +741,7 @@ void remove_proc_entry(const char *name, struct proc_dir_entry *parent) const char *fn = name; int len; - if (!parent && xlate_proc_name(name, &parent, &fn) != 0) + if (xlate_proc_name(name, &parent, &fn) != 0) return; len = strlen(fn); -- cgit v1.2.3 From 5e971dce0b2f6896e02372512df0d1fb0bfe2d55 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 29 Apr 2008 01:01:41 -0700 Subject: proc: drop several "PDE valid/invalid" checks proc-misc code is noticeably full of "if (de)" checks when PDE passed is always valid. Remove them. Addition of such check in proc_lookup_de() is for failed lookup case. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/generic.c | 33 +++++++++++--------------- fs/proc/inode.c | 69 ++++++++++++++++++++++++++----------------------------- 2 files changed, 46 insertions(+), 56 deletions(-) (limited to 'fs') diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 3c6f5669523a..8b406e21a258 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -388,20 +388,18 @@ struct dentry *proc_lookup_de(struct proc_dir_entry *de, struct inode *dir, lock_kernel(); spin_lock(&proc_subdir_lock); - if (de) { - for (de = de->subdir; de ; de = de->next) { - if (de->namelen != dentry->d_name.len) - continue; - if (!memcmp(dentry->d_name.name, de->name, de->namelen)) { - unsigned int ino; - - ino = de->low_ino; - de_get(de); - spin_unlock(&proc_subdir_lock); - error = -EINVAL; - inode = proc_get_inode(dir->i_sb, ino, de); - goto out_unlock; - } + for (de = de->subdir; de ; de = de->next) { + if (de->namelen != dentry->d_name.len) + continue; + if (!memcmp(dentry->d_name.name, de->name, de->namelen)) { + unsigned int ino; + + ino = de->low_ino; + de_get(de); + spin_unlock(&proc_subdir_lock); + error = -EINVAL; + inode = proc_get_inode(dir->i_sb, ino, de); + goto out_unlock; } } spin_unlock(&proc_subdir_lock); @@ -413,7 +411,8 @@ out_unlock: d_add(dentry, inode); return NULL; } - de_put(de); + if (de) + de_put(de); return ERR_PTR(error); } @@ -443,10 +442,6 @@ int proc_readdir_de(struct proc_dir_entry *de, struct file *filp, void *dirent, lock_kernel(); ino = inode->i_ino; - if (!de) { - ret = -EINVAL; - goto out; - } i = filp->f_pos; switch (i) { case 0: diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 82b3a1b5a70b..6f4e8dc97da1 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -25,8 +25,7 @@ struct proc_dir_entry *de_get(struct proc_dir_entry *de) { - if (de) - atomic_inc(&de->count); + atomic_inc(&de->count); return de; } @@ -35,18 +34,16 @@ struct proc_dir_entry *de_get(struct proc_dir_entry *de) */ void de_put(struct proc_dir_entry *de) { - if (de) { - lock_kernel(); - if (!atomic_read(&de->count)) { - printk("de_put: entry %s already free!\n", de->name); - unlock_kernel(); - return; - } - - if (atomic_dec_and_test(&de->count)) - free_proc_entry(de); + lock_kernel(); + if (!atomic_read(&de->count)) { + printk("de_put: entry %s already free!\n", de->name); unlock_kernel(); + return; } + + if (atomic_dec_and_test(&de->count)) + free_proc_entry(de); + unlock_kernel(); } /* @@ -392,7 +389,7 @@ struct inode *proc_get_inode(struct super_block *sb, unsigned int ino, { struct inode * inode; - if (de != NULL && !try_module_get(de->owner)) + if (!try_module_get(de->owner)) goto out_mod; inode = iget_locked(sb, ino); @@ -402,30 +399,29 @@ struct inode *proc_get_inode(struct super_block *sb, unsigned int ino, inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; PROC_I(inode)->fd = 0; PROC_I(inode)->pde = de; - if (de) { - if (de->mode) { - inode->i_mode = de->mode; - inode->i_uid = de->uid; - inode->i_gid = de->gid; - } - if (de->size) - inode->i_size = de->size; - if (de->nlink) - inode->i_nlink = de->nlink; - if (de->proc_iops) - inode->i_op = de->proc_iops; - if (de->proc_fops) { - if (S_ISREG(inode->i_mode)) { + + if (de->mode) { + inode->i_mode = de->mode; + inode->i_uid = de->uid; + inode->i_gid = de->gid; + } + if (de->size) + inode->i_size = de->size; + if (de->nlink) + inode->i_nlink = de->nlink; + if (de->proc_iops) + inode->i_op = de->proc_iops; + if (de->proc_fops) { + if (S_ISREG(inode->i_mode)) { #ifdef CONFIG_COMPAT - if (!de->proc_fops->compat_ioctl) - inode->i_fop = - &proc_reg_file_ops_no_compat; - else + if (!de->proc_fops->compat_ioctl) + inode->i_fop = + &proc_reg_file_ops_no_compat; + else #endif - inode->i_fop = &proc_reg_file_ops; - } else { - inode->i_fop = de->proc_fops; - } + inode->i_fop = &proc_reg_file_ops; + } else { + inode->i_fop = de->proc_fops; } } unlock_new_inode(inode); @@ -433,8 +429,7 @@ struct inode *proc_get_inode(struct super_block *sb, unsigned int ino, return inode; out_ino: - if (de != NULL) - module_put(de->owner); + module_put(de->owner); out_mod: return NULL; } -- cgit v1.2.3 From 9c37066d888bf6e1b96ad12304971b3ddeabbad0 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 29 Apr 2008 01:01:41 -0700 Subject: proc: remove proc_bus Remove proc_bus export and variable itself. Using pathnames works fine and is slightly more understandable and greppable. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/ecard.c | 2 +- drivers/input/input.c | 6 +++--- drivers/nubus/proc.c | 2 +- drivers/pci/proc.c | 2 +- drivers/pnp/isapnp/proc.c | 2 +- drivers/pnp/pnpbios/proc.c | 4 ++-- drivers/usb/core/inode.c | 4 ++-- drivers/zorro/proc.c | 2 +- fs/proc/root.c | 5 ++--- include/linux/proc_fs.h | 2 -- 10 files changed, 14 insertions(+), 17 deletions(-) (limited to 'fs') diff --git a/arch/arm/kernel/ecard.c b/arch/arm/kernel/ecard.c index f56d48c451ea..2d8ab6b3a586 100644 --- a/arch/arm/kernel/ecard.c +++ b/arch/arm/kernel/ecard.c @@ -778,7 +778,7 @@ static struct proc_dir_entry *proc_bus_ecard_dir = NULL; static void ecard_proc_init(void) { - proc_bus_ecard_dir = proc_mkdir("ecard", proc_bus); + proc_bus_ecard_dir = proc_mkdir("bus/ecard", NULL); create_proc_info_entry("devices", 0, proc_bus_ecard_dir, get_ecard_dev_info); } diff --git a/drivers/input/input.c b/drivers/input/input.c index f02c242c3114..11426604d8a2 100644 --- a/drivers/input/input.c +++ b/drivers/input/input.c @@ -898,7 +898,7 @@ static int __init input_proc_init(void) { struct proc_dir_entry *entry; - proc_bus_input_dir = proc_mkdir("input", proc_bus); + proc_bus_input_dir = proc_mkdir("bus/input", NULL); if (!proc_bus_input_dir) return -ENOMEM; @@ -921,7 +921,7 @@ static int __init input_proc_init(void) return 0; fail2: remove_proc_entry("devices", proc_bus_input_dir); - fail1: remove_proc_entry("input", proc_bus); + fail1: remove_proc_entry("bus/input", NULL); return -ENOMEM; } @@ -929,7 +929,7 @@ static void input_proc_exit(void) { remove_proc_entry("devices", proc_bus_input_dir); remove_proc_entry("handlers", proc_bus_input_dir); - remove_proc_entry("input", proc_bus); + remove_proc_entry("bus/input", NULL); } #else /* !CONFIG_PROC_FS */ diff --git a/drivers/nubus/proc.c b/drivers/nubus/proc.c index e07492be1f4a..cb83acef9479 100644 --- a/drivers/nubus/proc.c +++ b/drivers/nubus/proc.c @@ -171,7 +171,7 @@ void __init nubus_proc_init(void) { if (!MACH_IS_MAC) return; - proc_bus_nubus_dir = proc_mkdir("nubus", proc_bus); + proc_bus_nubus_dir = proc_mkdir("bus/nubus", NULL); create_proc_info_entry("devices", 0, proc_bus_nubus_dir, get_nubus_dev_info); proc_bus_nubus_add_devices(); diff --git a/drivers/pci/proc.c b/drivers/pci/proc.c index ef18fcd641e2..7b5e45b2fd16 100644 --- a/drivers/pci/proc.c +++ b/drivers/pci/proc.c @@ -472,7 +472,7 @@ static int __init pci_proc_init(void) { struct proc_dir_entry *entry; struct pci_dev *dev = NULL; - proc_bus_pci_dir = proc_mkdir("pci", proc_bus); + proc_bus_pci_dir = proc_mkdir("bus/pci", NULL); entry = create_proc_entry("devices", 0, proc_bus_pci_dir); if (entry) entry->proc_fops = &proc_bus_pci_dev_operations; diff --git a/drivers/pnp/isapnp/proc.c b/drivers/pnp/isapnp/proc.c index 2b8266c3d40f..e1d1f2a10947 100644 --- a/drivers/pnp/isapnp/proc.c +++ b/drivers/pnp/isapnp/proc.c @@ -116,7 +116,7 @@ int __init isapnp_proc_init(void) { struct pnp_dev *dev; - isapnp_proc_bus_dir = proc_mkdir("isapnp", proc_bus); + isapnp_proc_bus_dir = proc_mkdir("bus/isapnp", NULL); protocol_for_each_dev(&isapnp_protocol, dev) { isapnp_proc_attach_device(dev); } diff --git a/drivers/pnp/pnpbios/proc.c b/drivers/pnp/pnpbios/proc.c index bb19bc957bad..46d506f66259 100644 --- a/drivers/pnp/pnpbios/proc.c +++ b/drivers/pnp/pnpbios/proc.c @@ -256,7 +256,7 @@ int pnpbios_interface_attach_device(struct pnp_bios_node *node) */ int __init pnpbios_proc_init(void) { - proc_pnp = proc_mkdir("pnp", proc_bus); + proc_pnp = proc_mkdir("bus/pnp", NULL); if (!proc_pnp) return -EIO; proc_pnp_boot = proc_mkdir("boot", proc_pnp); @@ -294,5 +294,5 @@ void __exit pnpbios_proc_exit(void) remove_proc_entry("configuration_info", proc_pnp); remove_proc_entry("devices", proc_pnp); remove_proc_entry("boot", proc_pnp); - remove_proc_entry("pnp", proc_bus); + remove_proc_entry("bus/pnp", NULL); } diff --git a/drivers/usb/core/inode.c b/drivers/usb/core/inode.c index 8607846e3c3f..1d253dd4ea81 100644 --- a/drivers/usb/core/inode.c +++ b/drivers/usb/core/inode.c @@ -773,7 +773,7 @@ int __init usbfs_init(void) usb_register_notify(&usbfs_nb); /* create mount point for usbfs */ - usbdir = proc_mkdir("usb", proc_bus); + usbdir = proc_mkdir("bus/usb", NULL); return 0; } @@ -783,6 +783,6 @@ void usbfs_cleanup(void) usb_unregister_notify(&usbfs_nb); unregister_filesystem(&usb_fs_type); if (usbdir) - remove_proc_entry("usb", proc_bus); + remove_proc_entry("bus/usb", NULL); } diff --git a/drivers/zorro/proc.c b/drivers/zorro/proc.c index 2ce4cebc31d9..b7a8c7b7f66e 100644 --- a/drivers/zorro/proc.c +++ b/drivers/zorro/proc.c @@ -128,7 +128,7 @@ static int __init zorro_proc_init(void) u_int slot; if (MACH_IS_AMIGA && AMIGAHW_PRESENT(ZORRO)) { - proc_bus_zorro_dir = proc_mkdir("zorro", proc_bus); + proc_bus_zorro_dir = proc_mkdir("bus/zorro", NULL); create_proc_info_entry("devices", 0, proc_bus_zorro_dir, get_zorro_dev_info); for (slot = 0; slot < zorro_num_autocon; slot++) diff --git a/fs/proc/root.c b/fs/proc/root.c index ef0fb57fc9ef..cc46fcba8029 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -22,7 +22,7 @@ #include "internal.h" -struct proc_dir_entry *proc_bus, *proc_root_fs, *proc_root_driver; +struct proc_dir_entry *proc_root_fs, *proc_root_driver; static int proc_test_super(struct super_block *sb, void *data) { @@ -137,7 +137,7 @@ void __init proc_root_init(void) #ifdef CONFIG_PROC_DEVICETREE proc_device_tree_init(); #endif - proc_bus = proc_mkdir("bus", NULL); + proc_mkdir("bus", NULL); proc_sys_init(); } @@ -236,5 +236,4 @@ EXPORT_SYMBOL(proc_create); EXPORT_SYMBOL(remove_proc_entry); EXPORT_SYMBOL(proc_root); EXPORT_SYMBOL(proc_root_fs); -EXPORT_SYMBOL(proc_bus); EXPORT_SYMBOL(proc_root_driver); diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index 65f2299b772b..78c8c41ea9cc 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -98,7 +98,6 @@ struct vmcore { extern struct proc_dir_entry proc_root; extern struct proc_dir_entry *proc_root_fs; -extern struct proc_dir_entry *proc_bus; extern struct proc_dir_entry *proc_root_driver; extern struct proc_dir_entry *proc_root_kcore; @@ -214,7 +213,6 @@ extern void dup_mm_exe_file(struct mm_struct *oldmm, struct mm_struct *newmm); #else #define proc_root_driver NULL -#define proc_bus NULL #define proc_net_fops_create(net, name, mode, fops) ({ (void)(mode), NULL; }) static inline void proc_net_remove(struct net *net, const char *name) {} -- cgit v1.2.3 From 36a5aeb8787fbf92510ed20d806e229c55726f93 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 29 Apr 2008 01:01:42 -0700 Subject: proc: remove proc_root_fs Use creation by full path instead: "fs/foo". Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/cifs/cifs_debug.c | 4 ++-- fs/ext4/mballoc.c | 7 +++---- fs/jfs/jfs_debug.c | 4 ++-- fs/nfs/client.c | 6 +++--- fs/proc/root.c | 5 ++--- include/linux/proc_fs.h | 1 - 6 files changed, 12 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index 0228ed06069e..cc950f69e51e 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -468,7 +468,7 @@ cifs_proc_init(void) { struct proc_dir_entry *pde; - proc_fs_cifs = proc_mkdir("cifs", proc_root_fs); + proc_fs_cifs = proc_mkdir("fs/cifs", NULL); if (proc_fs_cifs == NULL) return; @@ -559,7 +559,7 @@ cifs_proc_clean(void) remove_proc_entry("LinuxExtensionsEnabled", proc_fs_cifs); remove_proc_entry("Experimental", proc_fs_cifs); remove_proc_entry("LookupCacheEnabled", proc_fs_cifs); - remove_proc_entry("cifs", proc_root_fs); + remove_proc_entry("fs/cifs", NULL); } static int diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index ef97f19c2f9d..1efcb934c2d6 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2867,7 +2867,6 @@ static void ext4_mb_free_committed_blocks(struct super_block *sb) mb_debug("freed %u blocks in %u structures\n", count, count2); } -#define EXT4_ROOT "ext4" #define EXT4_MB_STATS_NAME "stats" #define EXT4_MB_MAX_TO_SCAN_NAME "max_to_scan" #define EXT4_MB_MIN_TO_SCAN_NAME "min_to_scan" @@ -3007,9 +3006,9 @@ int __init init_ext4_mballoc(void) return -ENOMEM; } #ifdef CONFIG_PROC_FS - proc_root_ext4 = proc_mkdir(EXT4_ROOT, proc_root_fs); + proc_root_ext4 = proc_mkdir("fs/ext4", NULL); if (proc_root_ext4 == NULL) - printk(KERN_ERR "EXT4-fs: Unable to create %s\n", EXT4_ROOT); + printk(KERN_ERR "EXT4-fs: Unable to create fs/ext4\n"); #endif return 0; } @@ -3020,7 +3019,7 @@ void exit_ext4_mballoc(void) kmem_cache_destroy(ext4_pspace_cachep); kmem_cache_destroy(ext4_ac_cachep); #ifdef CONFIG_PROC_FS - remove_proc_entry(EXT4_ROOT, proc_root_fs); + remove_proc_entry("fs/ext4", NULL); #endif } diff --git a/fs/jfs/jfs_debug.c b/fs/jfs/jfs_debug.c index 887f5759e536..bf6ab19b86ee 100644 --- a/fs/jfs/jfs_debug.c +++ b/fs/jfs/jfs_debug.c @@ -89,7 +89,7 @@ void jfs_proc_init(void) { int i; - if (!(base = proc_mkdir("jfs", proc_root_fs))) + if (!(base = proc_mkdir("fs/jfs", NULL))) return; base->owner = THIS_MODULE; @@ -109,7 +109,7 @@ void jfs_proc_clean(void) if (base) { for (i = 0; i < NPROCENT; i++) remove_proc_entry(Entries[i].name, base); - remove_proc_entry("jfs", proc_root_fs); + remove_proc_entry("fs/jfs", NULL); } } diff --git a/fs/nfs/client.c b/fs/nfs/client.c index f2f3b284e6dd..0e066dcd4700 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -1500,7 +1500,7 @@ int __init nfs_fs_proc_init(void) { struct proc_dir_entry *p; - proc_fs_nfs = proc_mkdir("nfsfs", proc_root_fs); + proc_fs_nfs = proc_mkdir("fs/nfsfs", NULL); if (!proc_fs_nfs) goto error_0; @@ -1526,7 +1526,7 @@ int __init nfs_fs_proc_init(void) error_2: remove_proc_entry("servers", proc_fs_nfs); error_1: - remove_proc_entry("nfsfs", proc_root_fs); + remove_proc_entry("fs/nfsfs", NULL); error_0: return -ENOMEM; } @@ -1538,7 +1538,7 @@ void nfs_fs_proc_exit(void) { remove_proc_entry("volumes", proc_fs_nfs); remove_proc_entry("servers", proc_fs_nfs); - remove_proc_entry("nfsfs", proc_root_fs); + remove_proc_entry("fs/nfsfs", NULL); } #endif /* CONFIG_PROC_FS */ diff --git a/fs/proc/root.c b/fs/proc/root.c index cc46fcba8029..596abb690afa 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -22,7 +22,7 @@ #include "internal.h" -struct proc_dir_entry *proc_root_fs, *proc_root_driver; +struct proc_dir_entry *proc_root_driver; static int proc_test_super(struct super_block *sb, void *data) { @@ -126,7 +126,7 @@ void __init proc_root_init(void) #ifdef CONFIG_SYSVIPC proc_mkdir("sysvipc", NULL); #endif - proc_root_fs = proc_mkdir("fs", NULL); + proc_mkdir("fs", NULL); proc_root_driver = proc_mkdir("driver", NULL); proc_mkdir("fs/nfsd", NULL); /* somewhere for the nfsd filesystem to be mounted */ #if defined(CONFIG_SUN_OPENPROMFS) || defined(CONFIG_SUN_OPENPROMFS_MODULE) @@ -235,5 +235,4 @@ EXPORT_SYMBOL(create_proc_entry); EXPORT_SYMBOL(proc_create); EXPORT_SYMBOL(remove_proc_entry); EXPORT_SYMBOL(proc_root); -EXPORT_SYMBOL(proc_root_fs); EXPORT_SYMBOL(proc_root_driver); diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index 78c8c41ea9cc..65582f0384e0 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -97,7 +97,6 @@ struct vmcore { #ifdef CONFIG_PROC_FS extern struct proc_dir_entry proc_root; -extern struct proc_dir_entry *proc_root_fs; extern struct proc_dir_entry *proc_root_driver; extern struct proc_dir_entry *proc_root_kcore; -- cgit v1.2.3 From 928b4d8c8963e75bdb133f562b03b07f9aa4844a Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 29 Apr 2008 01:01:44 -0700 Subject: proc: remove proc_root_driver Use creation by full path: "driver/foo". Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/cciss.c | 4 ++-- drivers/block/cpqarray.c | 4 ++-- drivers/block/pktcdvd.c | 4 ++-- drivers/net/wireless/airo.c | 8 ++++---- fs/proc/root.c | 5 +---- include/linux/proc_fs.h | 3 --- 6 files changed, 11 insertions(+), 17 deletions(-) (limited to 'fs') diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index cf6083a1f928..e539be5750dc 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -425,7 +425,7 @@ static void __devinit cciss_procinit(int i) struct proc_dir_entry *pde; if (proc_cciss == NULL) - proc_cciss = proc_mkdir("cciss", proc_root_driver); + proc_cciss = proc_mkdir("driver/cciss", NULL); if (!proc_cciss) return; pde = proc_create(hba[i]->devname, S_IWUSR | S_IRUSR | S_IRGRP | @@ -3700,7 +3700,7 @@ static void __exit cciss_cleanup(void) cciss_remove_one(hba[i]->pdev); } } - remove_proc_entry("cciss", proc_root_driver); + remove_proc_entry("driver/cciss", NULL); } static void fail_all_cmds(unsigned long ctlr) diff --git a/drivers/block/cpqarray.c b/drivers/block/cpqarray.c index 69199185ff4b..09c14341e6e3 100644 --- a/drivers/block/cpqarray.c +++ b/drivers/block/cpqarray.c @@ -214,7 +214,7 @@ static struct proc_dir_entry *proc_array; static void __init ida_procinit(int i) { if (proc_array == NULL) { - proc_array = proc_mkdir("cpqarray", proc_root_driver); + proc_array = proc_mkdir("driver/cpqarray", NULL); if (!proc_array) return; } @@ -1796,7 +1796,7 @@ static void __exit cpqarray_exit(void) } } - remove_proc_entry("cpqarray", proc_root_driver); + remove_proc_entry("driver/cpqarray", NULL); } module_init(cpqarray_init) diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index 18feb1c7c33b..0431e5977bcb 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -3101,7 +3101,7 @@ static int __init pkt_init(void) goto out_misc; } - pkt_proc = proc_mkdir(DRIVER_NAME, proc_root_driver); + pkt_proc = proc_mkdir("driver/"DRIVER_NAME, NULL); return 0; @@ -3117,7 +3117,7 @@ out2: static void __exit pkt_exit(void) { - remove_proc_entry(DRIVER_NAME, proc_root_driver); + remove_proc_entry("driver/"DRIVER_NAME, NULL); misc_deregister(&pkt_misc); pkt_debugfs_cleanup(); diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c index 932d6b1c9d0b..6c395fcece58 100644 --- a/drivers/net/wireless/airo.c +++ b/drivers/net/wireless/airo.c @@ -5625,9 +5625,9 @@ static int __init airo_init_module( void ) int have_isa_dev = 0; #endif - airo_entry = create_proc_entry("aironet", + airo_entry = create_proc_entry("driver/aironet", S_IFDIR | airo_perm, - proc_root_driver); + NULL); if (airo_entry) { airo_entry->uid = proc_uid; @@ -5651,7 +5651,7 @@ static int __init airo_init_module( void ) airo_print_info("", "Finished probing for PCI adapters"); if (i) { - remove_proc_entry("aironet", proc_root_driver); + remove_proc_entry("driver/aironet", NULL); return i; } #endif @@ -5673,7 +5673,7 @@ static void __exit airo_cleanup_module( void ) #ifdef CONFIG_PCI pci_unregister_driver(&airo_driver); #endif - remove_proc_entry("aironet", proc_root_driver); + remove_proc_entry("driver/aironet", NULL); } /* diff --git a/fs/proc/root.c b/fs/proc/root.c index 596abb690afa..5e93e9b0124e 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -22,8 +22,6 @@ #include "internal.h" -struct proc_dir_entry *proc_root_driver; - static int proc_test_super(struct super_block *sb, void *data) { return sb->s_fs_info == data; @@ -127,7 +125,7 @@ void __init proc_root_init(void) proc_mkdir("sysvipc", NULL); #endif proc_mkdir("fs", NULL); - proc_root_driver = proc_mkdir("driver", NULL); + proc_mkdir("driver", NULL); proc_mkdir("fs/nfsd", NULL); /* somewhere for the nfsd filesystem to be mounted */ #if defined(CONFIG_SUN_OPENPROMFS) || defined(CONFIG_SUN_OPENPROMFS_MODULE) /* just give it a mountpoint */ @@ -235,4 +233,3 @@ EXPORT_SYMBOL(create_proc_entry); EXPORT_SYMBOL(proc_create); EXPORT_SYMBOL(remove_proc_entry); EXPORT_SYMBOL(proc_root); -EXPORT_SYMBOL(proc_root_driver); diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index 65582f0384e0..f56205cbebc0 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -97,7 +97,6 @@ struct vmcore { #ifdef CONFIG_PROC_FS extern struct proc_dir_entry proc_root; -extern struct proc_dir_entry *proc_root_driver; extern struct proc_dir_entry *proc_root_kcore; extern spinlock_t proc_subdir_lock; @@ -211,8 +210,6 @@ extern void dup_mm_exe_file(struct mm_struct *oldmm, struct mm_struct *newmm); #else -#define proc_root_driver NULL - #define proc_net_fops_create(net, name, mode, fops) ({ (void)(mode), NULL; }) static inline void proc_net_remove(struct net *net, const char *name) {} -- cgit v1.2.3 From c74c120a21d87b0b6925ada5830d8cac21e852d9 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 29 Apr 2008 01:01:44 -0700 Subject: proc: remove proc_root from drivers Remove proc_root export. Creation and removal works well if parent PDE is supplied as NULL -- it worked always that way. So, one useless export removed and consistency added, some drivers created PDEs with &proc_root as parent but removed them as NULL and so on. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/atags.c | 2 +- arch/m68k/mac/iop.c | 2 +- arch/mips/basler/excite/excite_procfs.c | 2 +- arch/um/kernel/exitcode.c | 2 +- arch/um/kernel/process.c | 2 +- arch/x86/kernel/cpu/mtrr/if.c | 2 +- drivers/char/ip2/ip2main.c | 4 ++-- drivers/mca/mca-proc.c | 2 +- drivers/misc/hdpuftrs/hdpu_cpustate.c | 2 +- drivers/misc/hdpuftrs/hdpu_nexus.c | 12 ++++++------ drivers/s390/block/dasd_proc.c | 6 +++--- drivers/s390/char/tape_proc.c | 4 ++-- drivers/s390/cio/blacklist.c | 2 +- drivers/s390/cio/qdio.c | 4 ++-- drivers/scsi/megaraid.c | 6 +++--- drivers/video/clps711xfb.c | 2 +- fs/proc/internal.h | 1 + fs/proc/proc_misc.c | 2 +- fs/proc/root.c | 1 - include/linux/proc_fs.h | 3 --- kernel/configs.c | 5 ++--- sound/core/info.c | 4 ++-- 22 files changed, 34 insertions(+), 38 deletions(-) (limited to 'fs') diff --git a/arch/arm/kernel/atags.c b/arch/arm/kernel/atags.c index e2e934c38080..64c420805e6f 100644 --- a/arch/arm/kernel/atags.c +++ b/arch/arm/kernel/atags.c @@ -35,7 +35,7 @@ create_proc_entries(void) { struct proc_dir_entry* tags_entry; - tags_entry = create_proc_read_entry("atags", 0400, &proc_root, read_buffer, &tags_buffer); + tags_entry = create_proc_read_entry("atags", 0400, NULL, read_buffer, &tags_buffer); if (!tags_entry) return -ENOMEM; diff --git a/arch/m68k/mac/iop.c b/arch/m68k/mac/iop.c index 5b2799eb96a6..092d4b3d8f76 100644 --- a/arch/m68k/mac/iop.c +++ b/arch/m68k/mac/iop.c @@ -302,7 +302,7 @@ void __init iop_init(void) #if 0 /* Crashing in 2.4 now, not yet sure why. --jmt */ #ifdef CONFIG_PROC_FS - create_proc_info_entry("mac_iop", 0, &proc_root, iop_get_proc_info); + create_proc_info_entry("mac_iop", 0, NULL, iop_get_proc_info); #endif #endif } diff --git a/arch/mips/basler/excite/excite_procfs.c b/arch/mips/basler/excite/excite_procfs.c index 9ee67a95f6b9..6c08b386fdad 100644 --- a/arch/mips/basler/excite/excite_procfs.c +++ b/arch/mips/basler/excite/excite_procfs.c @@ -65,7 +65,7 @@ excite_bootrom_read(char *page, char **start, off_t off, int count, void excite_procfs_init(void) { /* Create & populate /proc/excite */ - struct proc_dir_entry * const pdir = proc_mkdir("excite", &proc_root); + struct proc_dir_entry * const pdir = proc_mkdir("excite", NULL); if (pdir) { struct proc_dir_entry * e; diff --git a/arch/um/kernel/exitcode.c b/arch/um/kernel/exitcode.c index 984f80e668ca..6540d2c9fbb7 100644 --- a/arch/um/kernel/exitcode.c +++ b/arch/um/kernel/exitcode.c @@ -59,7 +59,7 @@ static int make_proc_exitcode(void) { struct proc_dir_entry *ent; - ent = create_proc_entry("exitcode", 0600, &proc_root); + ent = create_proc_entry("exitcode", 0600, NULL); if (ent == NULL) { printk(KERN_WARNING "make_proc_exitcode : Failed to register " "/proc/exitcode\n"); diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c index e8cb9ff183e9..83603cfbde81 100644 --- a/arch/um/kernel/process.c +++ b/arch/um/kernel/process.c @@ -364,7 +364,7 @@ int __init make_proc_sysemu(void) if (!sysemu_supported) return 0; - ent = create_proc_entry("sysemu", 0600, &proc_root); + ent = create_proc_entry("sysemu", 0600, NULL); if (ent == NULL) { diff --git a/arch/x86/kernel/cpu/mtrr/if.c b/arch/x86/kernel/cpu/mtrr/if.c index 1960f1985e5e..84c480bb3715 100644 --- a/arch/x86/kernel/cpu/mtrr/if.c +++ b/arch/x86/kernel/cpu/mtrr/if.c @@ -424,7 +424,7 @@ static int __init mtrr_if_init(void) return -ENODEV; proc_root_mtrr = - proc_create("mtrr", S_IWUSR | S_IRUGO, &proc_root, &mtrr_fops); + proc_create("mtrr", S_IWUSR | S_IRUGO, NULL, &mtrr_fops); if (proc_root_mtrr) proc_root_mtrr->owner = THIS_MODULE; diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c index b1d6cad84282..a784f5e22ee9 100644 --- a/drivers/char/ip2/ip2main.c +++ b/drivers/char/ip2/ip2main.c @@ -423,7 +423,7 @@ cleanup_module(void) } put_tty_driver(ip2_tty_driver); unregister_chrdev(IP2_IPL_MAJOR, pcIpl); - remove_proc_entry("ip2mem", &proc_root); + remove_proc_entry("ip2mem", NULL); // free memory for (i = 0; i < IP2_MAX_BOARDS; i++) { @@ -695,7 +695,7 @@ ip2_loadmain(int *iop, int *irqp, unsigned char *firmware, int firmsize) } } /* Register the read_procmem thing */ - if (!create_proc_info_entry("ip2mem",0,&proc_root,ip2_read_procmem)) { + if (!create_proc_info_entry("ip2mem",0,NULL,ip2_read_procmem)) { printk(KERN_ERR "IP2: failed to register read_procmem\n"); } else { diff --git a/drivers/mca/mca-proc.c b/drivers/mca/mca-proc.c index 33d5e0820cc5..81ea0d377bf4 100644 --- a/drivers/mca/mca-proc.c +++ b/drivers/mca/mca-proc.c @@ -183,7 +183,7 @@ void __init mca_do_proc_init(void) struct proc_dir_entry* node = NULL; struct mca_device *mca_dev; - proc_mca = proc_mkdir("mca", &proc_root); + proc_mca = proc_mkdir("mca", NULL); create_proc_read_entry("pos",0,proc_mca,get_mca_info,NULL); create_proc_read_entry("machine",0,proc_mca,get_mca_machine_info,NULL); diff --git a/drivers/misc/hdpuftrs/hdpu_cpustate.c b/drivers/misc/hdpuftrs/hdpu_cpustate.c index 302e92418bbe..154155c9b638 100644 --- a/drivers/misc/hdpuftrs/hdpu_cpustate.c +++ b/drivers/misc/hdpuftrs/hdpu_cpustate.c @@ -210,7 +210,7 @@ static int hdpu_cpustate_probe(struct platform_device *pdev) return ret; } - proc_de = create_proc_entry("sky_cpustate", 0666, &proc_root); + proc_de = create_proc_entry("sky_cpustate", 0666, NULL); if (!proc_de) { printk(KERN_WARNING "sky_cpustate: " "Unable to create proc entry\n"); diff --git a/drivers/misc/hdpuftrs/hdpu_nexus.c b/drivers/misc/hdpuftrs/hdpu_nexus.c index 2fa36f7a6eb3..e92b7efccc72 100644 --- a/drivers/misc/hdpuftrs/hdpu_nexus.c +++ b/drivers/misc/hdpuftrs/hdpu_nexus.c @@ -102,8 +102,8 @@ static int hdpu_nexus_probe(struct platform_device *pdev) printk(KERN_ERR "sky_nexus: Could not map slot id\n"); } - hdpu_slot_id = create_proc_entry("sky_slot_id", 0666, &proc_root); - if (!hdpu_slot_id) { + hdpu_slot_id = create_proc_entry("sky_slot_id", 0666, NULL); + if (!hdpu_slot_id) printk(KERN_WARNING "sky_nexus: " "Unable to create proc dir entry: sky_slot_id\n"); } else { @@ -111,8 +111,8 @@ static int hdpu_nexus_probe(struct platform_device *pdev) hdpu_slot_id->owner = THIS_MODULE; } - hdpu_chassis_id = create_proc_entry("sky_chassis_id", 0666, &proc_root); - if (!hdpu_chassis_id) { + hdpu_chassis_id = create_proc_entry("sky_chassis_id", 0666, NULL); + if (!hdpu_chassis_id) printk(KERN_WARNING "sky_nexus: " "Unable to create proc dir entry: sky_chassis_id\n"); } else { @@ -128,8 +128,8 @@ static int hdpu_nexus_remove(struct platform_device *pdev) slot_id = -1; chassis_id = -1; - remove_proc_entry("sky_slot_id", &proc_root); - remove_proc_entry("sky_chassis_id", &proc_root); + remove_proc_entry("sky_slot_id", NULL); + remove_proc_entry("sky_chassis_id", NULL); hdpu_slot_id = 0; hdpu_chassis_id = 0; diff --git a/drivers/s390/block/dasd_proc.c b/drivers/s390/block/dasd_proc.c index 556063e8f7a9..8ae9406b10ad 100644 --- a/drivers/s390/block/dasd_proc.c +++ b/drivers/s390/block/dasd_proc.c @@ -311,7 +311,7 @@ out_error: int dasd_proc_init(void) { - dasd_proc_root_entry = proc_mkdir("dasd", &proc_root); + dasd_proc_root_entry = proc_mkdir("dasd", NULL); if (!dasd_proc_root_entry) goto out_nodasd; dasd_proc_root_entry->owner = THIS_MODULE; @@ -335,7 +335,7 @@ dasd_proc_init(void) out_nostatistics: remove_proc_entry("devices", dasd_proc_root_entry); out_nodevices: - remove_proc_entry("dasd", &proc_root); + remove_proc_entry("dasd", NULL); out_nodasd: return -ENOENT; } @@ -345,5 +345,5 @@ dasd_proc_exit(void) { remove_proc_entry("devices", dasd_proc_root_entry); remove_proc_entry("statistics", dasd_proc_root_entry); - remove_proc_entry("dasd", &proc_root); + remove_proc_entry("dasd", NULL); } diff --git a/drivers/s390/char/tape_proc.c b/drivers/s390/char/tape_proc.c index c9b96d51b28f..0c39636b2174 100644 --- a/drivers/s390/char/tape_proc.c +++ b/drivers/s390/char/tape_proc.c @@ -125,7 +125,7 @@ tape_proc_init(void) { tape_proc_devices = create_proc_entry ("tapedevices", S_IFREG | S_IRUGO | S_IWUSR, - &proc_root); + NULL); if (tape_proc_devices == NULL) { PRINT_WARN("tape: Cannot register procfs entry tapedevices\n"); return; @@ -141,5 +141,5 @@ void tape_proc_cleanup(void) { if (tape_proc_devices != NULL) - remove_proc_entry ("tapedevices", &proc_root); + remove_proc_entry ("tapedevices", NULL); } diff --git a/drivers/s390/cio/blacklist.c b/drivers/s390/cio/blacklist.c index e8597ec92247..ef33d5df2229 100644 --- a/drivers/s390/cio/blacklist.c +++ b/drivers/s390/cio/blacklist.c @@ -375,7 +375,7 @@ cio_ignore_proc_init (void) struct proc_dir_entry *entry; entry = create_proc_entry ("cio_ignore", S_IFREG | S_IRUGO | S_IWUSR, - &proc_root); + NULL); if (!entry) return -ENOENT; diff --git a/drivers/s390/cio/qdio.c b/drivers/s390/cio/qdio.c index 10aa1e780801..43876e287370 100644 --- a/drivers/s390/cio/qdio.c +++ b/drivers/s390/cio/qdio.c @@ -3632,7 +3632,7 @@ qdio_add_procfs_entry(void) { proc_perf_file_registration=0; qdio_perf_proc_file=create_proc_entry(QDIO_PERF, - S_IFREG|0444,&proc_root); + S_IFREG|0444,NULL); if (qdio_perf_proc_file) { qdio_perf_proc_file->read_proc=&qdio_perf_procfile_read; } else proc_perf_file_registration=-1; @@ -3647,7 +3647,7 @@ static void qdio_remove_procfs_entry(void) { if (!proc_perf_file_registration) /* means if it went ok earlier */ - remove_proc_entry(QDIO_PERF,&proc_root); + remove_proc_entry(QDIO_PERF,NULL); } /** diff --git a/drivers/scsi/megaraid.c b/drivers/scsi/megaraid.c index b135a1ed4b2c..18551aaf5e09 100644 --- a/drivers/scsi/megaraid.c +++ b/drivers/scsi/megaraid.c @@ -4996,7 +4996,7 @@ static int __init megaraid_init(void) max_mbox_busy_wait = MBOX_BUSY_WAIT; #ifdef CONFIG_PROC_FS - mega_proc_dir_entry = proc_mkdir("megaraid", &proc_root); + mega_proc_dir_entry = proc_mkdir("megaraid", NULL); if (!mega_proc_dir_entry) { printk(KERN_WARNING "megaraid: failed to create megaraid root\n"); @@ -5005,7 +5005,7 @@ static int __init megaraid_init(void) error = pci_register_driver(&megaraid_pci_driver); if (error) { #ifdef CONFIG_PROC_FS - remove_proc_entry("megaraid", &proc_root); + remove_proc_entry("megaraid", NULL); #endif return error; } @@ -5035,7 +5035,7 @@ static void __exit megaraid_exit(void) pci_unregister_driver(&megaraid_pci_driver); #ifdef CONFIG_PROC_FS - remove_proc_entry("megaraid", &proc_root); + remove_proc_entry("megaraid", NULL); #endif } diff --git a/drivers/video/clps711xfb.c b/drivers/video/clps711xfb.c index 17b5267f44d7..9f8a389dc7ae 100644 --- a/drivers/video/clps711xfb.c +++ b/drivers/video/clps711xfb.c @@ -381,7 +381,7 @@ int __init clps711xfb_init(void) /* Register the /proc entries. */ clps7111fb_backlight_proc_entry = create_proc_entry("backlight", 0444, - &proc_root); + NULL); if (clps7111fb_backlight_proc_entry == NULL) { printk("Couldn't create the /proc entry for the backlight.\n"); return -EINVAL; diff --git a/fs/proc/internal.h b/fs/proc/internal.h index b1d6df671edf..28cbca805905 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -11,6 +11,7 @@ #include +extern struct proc_dir_entry proc_root; #ifdef CONFIG_PROC_SYSCTL extern int proc_sys_init(void); #else diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index c4137bb94a9e..48bcf20cec2f 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -854,7 +854,7 @@ void __init proc_misc_init(void) /* And now for trickier ones */ #ifdef CONFIG_PRINTK - proc_create("kmsg", S_IRUSR, &proc_root, &proc_kmsg_operations); + proc_create("kmsg", S_IRUSR, NULL, &proc_kmsg_operations); #endif proc_create("locks", 0, NULL, &proc_locks_operations); proc_create("devices", 0, NULL, &proc_devinfo_operations); diff --git a/fs/proc/root.c b/fs/proc/root.c index 5e93e9b0124e..c741b45a5503 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -232,4 +232,3 @@ EXPORT_SYMBOL(proc_mkdir); EXPORT_SYMBOL(create_proc_entry); EXPORT_SYMBOL(proc_create); EXPORT_SYMBOL(remove_proc_entry); -EXPORT_SYMBOL(proc_root); diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index f56205cbebc0..2183ffdc5489 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -96,7 +96,6 @@ struct vmcore { #ifdef CONFIG_PROC_FS -extern struct proc_dir_entry proc_root; extern struct proc_dir_entry *proc_root_kcore; extern spinlock_t proc_subdir_lock; @@ -243,8 +242,6 @@ struct tty_driver; static inline void proc_tty_register_driver(struct tty_driver *driver) {}; static inline void proc_tty_unregister_driver(struct tty_driver *driver) {}; -extern struct proc_dir_entry proc_root; - static inline int pid_ns_prepare_proc(struct pid_namespace *ns) { return 0; diff --git a/kernel/configs.c b/kernel/configs.c index e84d3f9c6c7b..d3a4b82a8a96 100644 --- a/kernel/configs.c +++ b/kernel/configs.c @@ -79,8 +79,7 @@ static int __init ikconfig_init(void) struct proc_dir_entry *entry; /* create the current config file */ - entry = create_proc_entry("config.gz", S_IFREG | S_IRUGO, - &proc_root); + entry = create_proc_entry("config.gz", S_IFREG | S_IRUGO, NULL); if (!entry) return -ENOMEM; @@ -95,7 +94,7 @@ static int __init ikconfig_init(void) static void __exit ikconfig_cleanup(void) { - remove_proc_entry("config.gz", &proc_root); + remove_proc_entry("config.gz", NULL); } module_init(ikconfig_init); diff --git a/sound/core/info.c b/sound/core/info.c index 9977ec2eace3..cb5ead3e202d 100644 --- a/sound/core/info.c +++ b/sound/core/info.c @@ -544,7 +544,7 @@ int __init snd_info_init(void) { struct proc_dir_entry *p; - p = snd_create_proc_entry("asound", S_IFDIR | S_IRUGO | S_IXUGO, &proc_root); + p = snd_create_proc_entry("asound", S_IFDIR | S_IRUGO | S_IXUGO, NULL); if (p == NULL) return -ENOMEM; snd_proc_root = p; @@ -594,7 +594,7 @@ int __exit snd_info_done(void) #ifdef CONFIG_SND_OSSEMUL snd_info_free_entry(snd_oss_root); #endif - snd_remove_proc_entry(&proc_root, snd_proc_root); + snd_remove_proc_entry(NULL, snd_proc_root); } return 0; } -- cgit v1.2.3 From 8731f14d37825b54ad0c4c309cba2bc8fdf13a86 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 29 Apr 2008 01:01:58 -0700 Subject: proc: remove ->get_info infrastructure Now that last dozen or so users of ->get_info were removed, ditch it too. Everyone sane shouldd have switched to seq_file interface long ago. P.S.: Co-existing 3 interfaces (->get_info/->read_proc/->proc_fops) for proc is long-standing crap, BTW, thus a) put ->read_proc/->write_proc/read_proc_entry() users on death row, b) new such users should be rejected, c) everyone is encouraged to convert his favourite ->read_proc user or I'll do it, lazy bastards. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/generic.c | 7 +------ include/linux/proc_fs.h | 15 +-------------- 2 files changed, 2 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 8b406e21a258..0f3d97d41b0f 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -69,12 +69,7 @@ proc_file_read(struct file *file, char __user *buf, size_t nbytes, count = min_t(size_t, PROC_BLOCK_SIZE, nbytes); start = NULL; - if (dp->get_info) { - /* Handle old net routines */ - n = dp->get_info(page, &start, *ppos, count); - if (n < count) - eof = 1; - } else if (dp->read_proc) { + if (dp->read_proc) { /* * How to be a proc read function * ------------------------------ diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index 2183ffdc5489..29abcb805754 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -40,7 +40,7 @@ enum { * /proc file has a parent, but "subdir" is NULL for all * non-directory entries). * - * "get_info" is called at "read", while "owner" is used to protect module + * "owner" is used to protect module * from unloading while proc_dir_entry is in use */ @@ -48,7 +48,6 @@ typedef int (read_proc_t)(char *page, char **start, off_t off, int count, int *eof, void *data); typedef int (write_proc_t)(struct file *file, const char __user *buffer, unsigned long count, void *data); -typedef int (get_info_t)(char *, char **, off_t, int); struct proc_dir_entry { unsigned int low_ino; @@ -69,7 +68,6 @@ struct proc_dir_entry { * somewhere. */ const struct file_operations *proc_fops; - get_info_t *get_info; struct module *owner; struct proc_dir_entry *next, *parent, *subdir; void *data; @@ -187,14 +185,6 @@ static inline struct proc_dir_entry *create_proc_read_entry(const char *name, return res; } -static inline struct proc_dir_entry *create_proc_info_entry(const char *name, - mode_t mode, struct proc_dir_entry *base, get_info_t *get_info) -{ - struct proc_dir_entry *res=create_proc_entry(name,mode,base); - if (res) res->get_info=get_info; - return res; -} - extern struct proc_dir_entry *proc_net_fops_create(struct net *net, const char *name, mode_t mode, const struct file_operations *fops); extern void proc_net_remove(struct net *net, const char *name); @@ -234,9 +224,6 @@ static inline struct proc_dir_entry *proc_mkdir(const char *name, static inline struct proc_dir_entry *create_proc_read_entry(const char *name, mode_t mode, struct proc_dir_entry *base, read_proc_t *read_proc, void * data) { return NULL; } -static inline struct proc_dir_entry *create_proc_info_entry(const char *name, - mode_t mode, struct proc_dir_entry *base, get_info_t *get_info) - { return NULL; } struct tty_driver; static inline void proc_tty_register_driver(struct tty_driver *driver) {}; -- cgit v1.2.3 From b640a89ddd742782bd2d83873da30d4776d1b9c6 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 29 Apr 2008 01:01:58 -0700 Subject: proc: convert /proc/tty/ldiscs to seq_file interface Note: THIS_MODULE and header addition aren't technically needed because this code is not modular, but let's keep it anyway because people can copy this code into modular code. Signed-off-by: Alexey Dobriyan Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/proc_tty.c | 76 ++++++++++++++++++++++++++++++++---------------------- 1 file changed, 45 insertions(+), 31 deletions(-) (limited to 'fs') diff --git a/fs/proc/proc_tty.c b/fs/proc/proc_tty.c index 63f45383035d..ac26ccc25f42 100644 --- a/fs/proc/proc_tty.c +++ b/fs/proc/proc_tty.c @@ -5,7 +5,7 @@ */ #include - +#include #include #include #include @@ -136,39 +136,54 @@ static const struct file_operations proc_tty_drivers_operations = { .release = seq_release, }; -/* - * This is the handler for /proc/tty/ldiscs - */ -static int tty_ldiscs_read_proc(char *page, char **start, off_t off, - int count, int *eof, void *data) +static void * tty_ldiscs_seq_start(struct seq_file *m, loff_t *pos) { - int i; - int len = 0; - off_t begin = 0; + return (*pos < NR_LDISCS) ? pos : NULL; +} + +static void * tty_ldiscs_seq_next(struct seq_file *m, void *v, loff_t *pos) +{ + (*pos)++; + return (*pos < NR_LDISCS) ? pos : NULL; +} + +static void tty_ldiscs_seq_stop(struct seq_file *m, void *v) +{ +} + +static int tty_ldiscs_seq_show(struct seq_file *m, void *v) +{ + int i = *(loff_t *)v; struct tty_ldisc *ld; - for (i=0; i < NR_LDISCS; i++) { - ld = tty_ldisc_get(i); - if (ld == NULL) - continue; - len += sprintf(page+len, "%-10s %2d\n", - ld->name ? ld->name : "???", i); - tty_ldisc_put(i); - if (len+begin > off+count) - break; - if (len+begin < off) { - begin += len; - len = 0; - } - } - if (i >= NR_LDISCS) - *eof = 1; - if (off >= len+begin) + ld = tty_ldisc_get(i); + if (ld == NULL) return 0; - *start = page + (off-begin); - return ((count < begin+len-off) ? count : begin+len-off); + seq_printf(m, "%-10s %2d\n", ld->name ? ld->name : "???", i); + tty_ldisc_put(i); + return 0; +} + +static const struct seq_operations tty_ldiscs_seq_ops = { + .start = tty_ldiscs_seq_start, + .next = tty_ldiscs_seq_next, + .stop = tty_ldiscs_seq_stop, + .show = tty_ldiscs_seq_show, +}; + +static int proc_tty_ldiscs_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &tty_ldiscs_seq_ops); } +static const struct file_operations tty_ldiscs_proc_fops = { + .owner = THIS_MODULE, + .open = proc_tty_ldiscs_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + /* * This function is called by tty_register_driver() to handle * registering the driver's /proc handler into /proc/tty/driver/ @@ -223,8 +238,7 @@ void __init proc_tty_init(void) * password lengths and inter-keystroke timings during password * entry. */ - proc_tty_driver = proc_mkdir_mode("tty/driver", S_IRUSR | S_IXUSR, NULL); - - create_proc_read_entry("tty/ldiscs", 0, NULL, tty_ldiscs_read_proc, NULL); + proc_tty_driver = proc_mkdir_mode("tty/driver", S_IRUSR|S_IXUSR, NULL); + proc_create("tty/ldiscs", 0, NULL, &tty_ldiscs_proc_fops); proc_create("tty/drivers", 0, NULL, &proc_tty_drivers_operations); } -- cgit v1.2.3 From 59b7435149eab2dd06dd678742faff6049cb655f Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Tue, 29 Apr 2008 01:02:00 -0700 Subject: proc: introduce proc_create_data to setup de->data This set of patches fixes an proc ->open'less usage due to ->proc_fops flip in the most part of the kernel code. The original OOPS is described in the commit 2d3a4e3666325a9709cc8ea2e88151394e8f20fc: Typical PDE creation code looks like: pde = create_proc_entry("foo", 0, NULL); if (pde) pde->proc_fops = &foo_proc_fops; Notice that PDE is first created, only then ->proc_fops is set up to final value. This is a problem because right after creation a) PDE is fully visible in /proc , and b) ->proc_fops are proc_file_operations which do not have ->open callback. So, it's possible to ->read without ->open (see one class of oopses below). The fix is new API called proc_create() which makes sure ->proc_fops are set up before gluing PDE to main tree. Typical new code looks like: pde = proc_create("foo", 0, NULL, &foo_proc_fops); if (!pde) return -ENOMEM; Fix most networking users for a start. In the long run, create_proc_entry() for regular files will go. In addition to this, proc_create_data is introduced to fix reading from proc without PDE->data. The race is basically the same as above. create_proc_entries is replaced in the entire kernel code as new method is also simply better. This patch: The problem is the same as for de->proc_fops. Right now PDE becomes visible without data set. So, the entry could be looked up without data. This, in most cases, will simply OOPS. proc_create_data call is created to address this issue. proc_create now becomes a wrapper around it. Signed-off-by: Denis V. Lunev Cc: "Eric W. Biederman" Cc: "J. Bruce Fields" Cc: Alessandro Zummo Cc: Alexey Dobriyan Cc: Bartlomiej Zolnierkiewicz Cc: Benjamin Herrenschmidt Cc: Bjorn Helgaas Cc: Chris Mason Acked-by: David Howells Cc: Dmitry Torokhov Cc: Geert Uytterhoeven Cc: Grant Grundler Cc: Greg Kroah-Hartman Cc: Haavard Skinnemoen Cc: Heiko Carstens Cc: Ingo Molnar Cc: James Bottomley Cc: Jaroslav Kysela Cc: Jeff Garzik Cc: Jeff Mahoney Cc: Jesper Nilsson Cc: Karsten Keil Cc: Kyle McMartin Cc: Len Brown Cc: Martin Schwidefsky Cc: Mathieu Desnoyers Cc: Matthew Wilcox Cc: Mauro Carvalho Chehab Cc: Mikael Starvik Cc: Nadia Derbey Cc: Neil Brown Cc: Paul Mackerras Cc: Peter Osterlund Cc: Pierre Peiffer Cc: Russell King Cc: Takashi Iwai Cc: Tony Luck Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/generic.c | 8 +++++--- fs/proc/root.c | 2 +- include/linux/proc_fs.h | 17 +++++++++++++++-- 3 files changed, 21 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 0f3d97d41b0f..9d53b39a9cf8 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -675,9 +675,10 @@ struct proc_dir_entry *create_proc_entry(const char *name, mode_t mode, return ent; } -struct proc_dir_entry *proc_create(const char *name, mode_t mode, - struct proc_dir_entry *parent, - const struct file_operations *proc_fops) +struct proc_dir_entry *proc_create_data(const char *name, mode_t mode, + struct proc_dir_entry *parent, + const struct file_operations *proc_fops, + void *data) { struct proc_dir_entry *pde; nlink_t nlink; @@ -698,6 +699,7 @@ struct proc_dir_entry *proc_create(const char *name, mode_t mode, if (!pde) goto out; pde->proc_fops = proc_fops; + pde->data = data; if (proc_register(parent, pde) < 0) goto out_free; return pde; diff --git a/fs/proc/root.c b/fs/proc/root.c index c741b45a5503..95117538a4f6 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -230,5 +230,5 @@ void pid_ns_release_proc(struct pid_namespace *ns) EXPORT_SYMBOL(proc_symlink); EXPORT_SYMBOL(proc_mkdir); EXPORT_SYMBOL(create_proc_entry); -EXPORT_SYMBOL(proc_create); +EXPORT_SYMBOL(proc_create_data); EXPORT_SYMBOL(remove_proc_entry); diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index 29abcb805754..9883bc942262 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -116,9 +116,10 @@ void de_put(struct proc_dir_entry *de); extern struct proc_dir_entry *create_proc_entry(const char *name, mode_t mode, struct proc_dir_entry *parent); -struct proc_dir_entry *proc_create(const char *name, mode_t mode, +struct proc_dir_entry *proc_create_data(const char *name, mode_t mode, struct proc_dir_entry *parent, - const struct file_operations *proc_fops); + const struct file_operations *proc_fops, + void *data); extern void remove_proc_entry(const char *name, struct proc_dir_entry *parent); extern struct vfsmount *proc_mnt; @@ -173,6 +174,12 @@ extern struct proc_dir_entry *proc_mkdir(const char *,struct proc_dir_entry *); extern struct proc_dir_entry *proc_mkdir_mode(const char *name, mode_t mode, struct proc_dir_entry *parent); +static inline struct proc_dir_entry *proc_create(const char *name, mode_t mode, + struct proc_dir_entry *parent, const struct file_operations *proc_fops) +{ + return proc_create_data(name, mode, parent, proc_fops, NULL); +} + static inline struct proc_dir_entry *create_proc_read_entry(const char *name, mode_t mode, struct proc_dir_entry *base, read_proc_t *read_proc, void * data) @@ -214,6 +221,12 @@ static inline struct proc_dir_entry *proc_create(const char *name, { return NULL; } +static inline struct proc_dir_entry *proc_create_data(const char *name, + mode_t mode, struct proc_dir_entry *parent, + const struct file_operations *proc_fops, void *data) +{ + return NULL; +} #define remove_proc_entry(name, parent) do {} while (0) static inline struct proc_dir_entry *proc_symlink(const char *name, -- cgit v1.2.3 From 9ef2db2630652d68dfd336088648adae7ef0bcd4 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Tue, 29 Apr 2008 01:02:04 -0700 Subject: nfsd: use proc_create to setup de->proc_fops Use proc_create() to make sure that ->proc_fops be setup before gluing PDE to main tree. Signed-off-by: Denis V. Lunev Cc: Neil Brown Cc: "J. Bruce Fields" Cc: Alexey Dobriyan Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfsctl.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 42f3820ee8f5..5ac00c4fee91 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -169,6 +169,7 @@ static const struct file_operations exports_operations = { .read = seq_read, .llseek = seq_lseek, .release = seq_release, + .owner = THIS_MODULE, }; /*----------------------------------------------------------------------------*/ @@ -801,10 +802,9 @@ static int create_proc_exports_entry(void) entry = proc_mkdir("fs/nfs", NULL); if (!entry) return -ENOMEM; - entry = create_proc_entry("fs/nfs/exports", 0, NULL); + entry = proc_create("exports", 0, entry, &exports_operations); if (!entry) return -ENOMEM; - entry->proc_fops = &exports_operations; return 0; } #else /* CONFIG_PROC_FS */ -- cgit v1.2.3 From 34b37235c60fd23e4075da475c7bb22e6c7a466e Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Tue, 29 Apr 2008 01:02:07 -0700 Subject: nfs: use proc_create to setup de->proc_fops Use proc_create() to make sure that ->proc_fops be setup before gluing PDE to main tree. Signed-off-by: Denis V. Lunev Cc: "J. Bruce Fields" Cc: Trond Myklebust Cc: Alexey Dobriyan Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfs/client.c | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 0e066dcd4700..89ac5bb0401c 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -1321,6 +1321,7 @@ static const struct file_operations nfs_server_list_fops = { .read = seq_read, .llseek = seq_lseek, .release = seq_release, + .owner = THIS_MODULE, }; static int nfs_volume_list_open(struct inode *inode, struct file *file); @@ -1341,6 +1342,7 @@ static const struct file_operations nfs_volume_list_fops = { .read = seq_read, .llseek = seq_lseek, .release = seq_release, + .owner = THIS_MODULE, }; /* @@ -1507,20 +1509,16 @@ int __init nfs_fs_proc_init(void) proc_fs_nfs->owner = THIS_MODULE; /* a file of servers with which we're dealing */ - p = create_proc_entry("servers", S_IFREG|S_IRUGO, proc_fs_nfs); + p = proc_create("servers", S_IFREG|S_IRUGO, + proc_fs_nfs, &nfs_server_list_fops); if (!p) goto error_1; - p->proc_fops = &nfs_server_list_fops; - p->owner = THIS_MODULE; - /* a file of volumes that we have mounted */ - p = create_proc_entry("volumes", S_IFREG|S_IRUGO, proc_fs_nfs); + p = proc_create("volumes", S_IFREG|S_IRUGO, + proc_fs_nfs, &nfs_volume_list_fops); if (!p) goto error_2; - - p->proc_fops = &nfs_volume_list_fops; - p->owner = THIS_MODULE; return 0; error_2: -- cgit v1.2.3 From 21ac295b42b8bdc3d677aba6bd7308a38de28a9b Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Tue, 29 Apr 2008 01:02:07 -0700 Subject: afs: use non-racy method for proc entries creation Use proc_create()/proc_create_data() to make sure that ->proc_fops and ->data be setup before gluing PDE to main tree. Signed-off-by: Denis V. Lunev Acked-by: David Howells Cc: Alexey Dobriyan Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/afs/proc.c | 33 ++++++++++++++------------------- 1 file changed, 14 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/afs/proc.c b/fs/afs/proc.c index 846c7615ac9e..9f7d1ae70269 100644 --- a/fs/afs/proc.c +++ b/fs/afs/proc.c @@ -41,6 +41,7 @@ static const struct file_operations afs_proc_cells_fops = { .write = afs_proc_cells_write, .llseek = seq_lseek, .release = seq_release, + .owner = THIS_MODULE, }; static int afs_proc_rootcell_open(struct inode *inode, struct file *file); @@ -56,7 +57,8 @@ static const struct file_operations afs_proc_rootcell_fops = { .read = afs_proc_rootcell_read, .write = afs_proc_rootcell_write, .llseek = no_llseek, - .release = afs_proc_rootcell_release + .release = afs_proc_rootcell_release, + .owner = THIS_MODULE, }; static int afs_proc_cell_volumes_open(struct inode *inode, struct file *file); @@ -80,6 +82,7 @@ static const struct file_operations afs_proc_cell_volumes_fops = { .read = seq_read, .llseek = seq_lseek, .release = afs_proc_cell_volumes_release, + .owner = THIS_MODULE, }; static int afs_proc_cell_vlservers_open(struct inode *inode, @@ -104,6 +107,7 @@ static const struct file_operations afs_proc_cell_vlservers_fops = { .read = seq_read, .llseek = seq_lseek, .release = afs_proc_cell_vlservers_release, + .owner = THIS_MODULE, }; static int afs_proc_cell_servers_open(struct inode *inode, struct file *file); @@ -127,6 +131,7 @@ static const struct file_operations afs_proc_cell_servers_fops = { .read = seq_read, .llseek = seq_lseek, .release = afs_proc_cell_servers_release, + .owner = THIS_MODULE, }; /* @@ -143,17 +148,13 @@ int afs_proc_init(void) goto error_dir; proc_afs->owner = THIS_MODULE; - p = create_proc_entry("cells", 0, proc_afs); + p = proc_create("cells", 0, proc_afs, &afs_proc_cells_fops); if (!p) goto error_cells; - p->proc_fops = &afs_proc_cells_fops; - p->owner = THIS_MODULE; - p = create_proc_entry("rootcell", 0, proc_afs); + p = proc_create("rootcell", 0, proc_afs, &afs_proc_rootcell_fops); if (!p) goto error_rootcell; - p->proc_fops = &afs_proc_rootcell_fops; - p->owner = THIS_MODULE; _leave(" = 0"); return 0; @@ -395,26 +396,20 @@ int afs_proc_cell_setup(struct afs_cell *cell) if (!cell->proc_dir) goto error_dir; - p = create_proc_entry("servers", 0, cell->proc_dir); + p = proc_create_data("servers", 0, cell->proc_dir, + &afs_proc_cell_servers_fops, cell); if (!p) goto error_servers; - p->proc_fops = &afs_proc_cell_servers_fops; - p->owner = THIS_MODULE; - p->data = cell; - p = create_proc_entry("vlservers", 0, cell->proc_dir); + p = proc_create_data("vlservers", 0, cell->proc_dir, + &afs_proc_cell_vlservers_fops, cell); if (!p) goto error_vlservers; - p->proc_fops = &afs_proc_cell_vlservers_fops; - p->owner = THIS_MODULE; - p->data = cell; - p = create_proc_entry("volumes", 0, cell->proc_dir); + p = proc_create_data("volumes", 0, cell->proc_dir, + &afs_proc_cell_volumes_fops, cell); if (!p) goto error_volumes; - p->proc_fops = &afs_proc_cell_volumes_fops; - p->owner = THIS_MODULE; - p->data = cell; _leave(" = 0"); return 0; -- cgit v1.2.3 From 46fe74f2aed615c8c88164f4346b79c30cfd7c3d Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Tue, 29 Apr 2008 01:02:08 -0700 Subject: ext4: use non-racy method for proc entries creation Use proc_create()/proc_create_data() to make sure that ->proc_fops and ->data be setup before gluing PDE to main tree. Signed-off-by: Denis V. Lunev Cc: Cc: Alexey Dobriyan Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext4/mballoc.c | 15 ++++----------- 1 file changed, 4 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 1efcb934c2d6..9d57695de746 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2449,17 +2449,10 @@ static void ext4_mb_history_init(struct super_block *sb) int i; if (sbi->s_mb_proc != NULL) { - struct proc_dir_entry *p; - p = create_proc_entry("mb_history", S_IRUGO, sbi->s_mb_proc); - if (p) { - p->proc_fops = &ext4_mb_seq_history_fops; - p->data = sb; - } - p = create_proc_entry("mb_groups", S_IRUGO, sbi->s_mb_proc); - if (p) { - p->proc_fops = &ext4_mb_seq_groups_fops; - p->data = sb; - } + proc_create_data("mb_history", S_IRUGO, sbi->s_mb_proc, + &ext4_mb_seq_history_fops, sb); + proc_create_data("mb_groups", S_IRUGO, sbi->s_mb_proc, + &ext4_mb_seq_groups_fops, sb); } sbi->s_mb_history_max = 1000; -- cgit v1.2.3 From 19b4fc52d63b77adf700a215bfbabd680a8f1718 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Tue, 29 Apr 2008 01:02:09 -0700 Subject: reiserfs: use non-racy method for proc entries creation Use proc_create()/proc_create_data() to make sure that ->proc_fops and ->data be setup before gluing PDE to main tree. /proc entry owner is also added. Signed-off-by: Denis V. Lunev Cc: Jeff Mahoney Cc: Chris Mason Cc: Alexey Dobriyan Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/procfs.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c index 8f86c52b30d8..b9dbeeca7049 100644 --- a/fs/reiserfs/procfs.c +++ b/fs/reiserfs/procfs.c @@ -467,6 +467,7 @@ static const struct file_operations r_file_operations = { .read = seq_read, .llseek = seq_lseek, .release = seq_release, + .owner = THIS_MODULE, }; static struct proc_dir_entry *proc_info_root = NULL; @@ -475,12 +476,8 @@ static const char proc_info_root_name[] = "fs/reiserfs"; static void add_file(struct super_block *sb, char *name, int (*func) (struct seq_file *, struct super_block *)) { - struct proc_dir_entry *de; - de = create_proc_entry(name, 0, REISERFS_SB(sb)->procdir); - if (de) { - de->data = func; - de->proc_fops = &r_file_operations; - } + proc_create_data(name, 0, REISERFS_SB(sb)->procdir, + &r_file_operations, func); } int reiserfs_proc_info_init(struct super_block *sb) -- cgit v1.2.3 From 79da3664f61640057041bf172b1457e2d1969330 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Tue, 29 Apr 2008 01:02:11 -0700 Subject: jbd2: use non-racy method for proc entries creation Use proc_create()/proc_create_data() to make sure that ->proc_fops and ->data be setup before gluing PDE to main tree. Signed-off-by: Denis V. Lunev Cc: Cc: Alexey Dobriyan Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/jbd2/journal.c | 17 ++++------------- 1 file changed, 4 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 954cff001df6..eb7eb6c27bcb 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -904,19 +904,10 @@ static void jbd2_stats_proc_init(journal_t *journal) snprintf(name, sizeof(name) - 1, "%s", bdevname(journal->j_dev, name)); journal->j_proc_entry = proc_mkdir(name, proc_jbd2_stats); if (journal->j_proc_entry) { - struct proc_dir_entry *p; - p = create_proc_entry("history", S_IRUGO, - journal->j_proc_entry); - if (p) { - p->proc_fops = &jbd2_seq_history_fops; - p->data = journal; - p = create_proc_entry("info", S_IRUGO, - journal->j_proc_entry); - if (p) { - p->proc_fops = &jbd2_seq_info_fops; - p->data = journal; - } - } + proc_create_data("history", S_IRUGO, journal->j_proc_entry, + &jbd2_seq_history_fops, journal); + proc_create_data("info", S_IRUGO, journal->j_proc_entry, + &jbd2_seq_info_fops, journal); } } -- cgit v1.2.3 From 7708bfb1c855f2a076ef71cc21647deea022ebe7 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Tue, 29 Apr 2008 01:02:40 -0700 Subject: sysctl: merge equal proc_sys_read and proc_sys_write Many (most of) sysctls do not have a per-container sense. E.g. kernel.print_fatal_signals, vm.panic_on_oom, net.core.netdev_budget and so on and so forth. Besides, tuning then from inside a container is not even secure. On the other hand, hiding them completely from the container's tasks sometimes causes user-space to stop working. When developing net sysctl, the common practice was to duplicate a table and drop the write bits in table->mode, but this approach was not very elegant, lead to excessive memory consumption and was not suitable in general. Here's the alternative solution. To facilitate the per-container sysctls ctl_table_root-s were introduced. Each root contains a list of ctl_table_header-s that are visible to different namespaces. The idea of this set is to add the permissions() callback on the ctl_table_root to allow ctl root limit permissions to the same ctl_table-s. The main user of this functionality is the net-namespaces code, but later this will (should) be used by more and more namespaces, containers and control groups. Actually, this idea's core is in a single hunk in the third patch. First two patches are cleanups for sysctl code, while the third one mostly extends the arguments set of some sysctl functions. This patch: These ->read and ->write callbacks act in a very similar way, so merge these paths to reduce the number of places to patch later and shrink the .text size (a bit). Signed-off-by: Pavel Emelyanov Acked-by: "David S. Miller" Cc: "Eric W. Biederman" Cc: Alexey Dobriyan Cc: Denis V. Lunev Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/proc_sysctl.c | 50 +++++++++++--------------------------------------- 1 file changed, 11 insertions(+), 39 deletions(-) (limited to 'fs') diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 614c34b6d1c2..5e31585292a0 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -165,8 +165,8 @@ out: return err; } -static ssize_t proc_sys_read(struct file *filp, char __user *buf, - size_t count, loff_t *ppos) +static ssize_t proc_sys_call_handler(struct file *filp, void __user *buf, + size_t count, loff_t *ppos, int write) { struct dentry *dentry = filp->f_dentry; struct ctl_table_header *head; @@ -190,12 +190,12 @@ static ssize_t proc_sys_read(struct file *filp, char __user *buf, * and won't be until we finish. */ error = -EPERM; - if (sysctl_perm(table, MAY_READ)) + if (sysctl_perm(table, write ? MAY_WRITE : MAY_READ)) goto out; /* careful: calling conventions are nasty here */ res = count; - error = table->proc_handler(table, 0, filp, buf, &res, ppos); + error = table->proc_handler(table, write, filp, buf, &res, ppos); if (!error) error = res; out: @@ -204,44 +204,16 @@ out: return error; } -static ssize_t proc_sys_write(struct file *filp, const char __user *buf, +static ssize_t proc_sys_read(struct file *filp, char __user *buf, size_t count, loff_t *ppos) { - struct dentry *dentry = filp->f_dentry; - struct ctl_table_header *head; - struct ctl_table *table; - ssize_t error; - size_t res; - - table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head); - /* Has the sysctl entry disappeared on us? */ - error = -ENOENT; - if (!table) - goto out; - - /* Has the sysctl entry been replaced by a directory? */ - error = -EISDIR; - if (!table->proc_handler) - goto out; - - /* - * At this point we know that the sysctl was not unregistered - * and won't be until we finish. - */ - error = -EPERM; - if (sysctl_perm(table, MAY_WRITE)) - goto out; - - /* careful: calling conventions are nasty here */ - res = count; - error = table->proc_handler(table, 1, filp, (char __user *)buf, - &res, ppos); - if (!error) - error = res; -out: - sysctl_head_finish(head); + return proc_sys_call_handler(filp, (void __user *)buf, count, ppos, 0); +} - return error; +static ssize_t proc_sys_write(struct file *filp, const char __user *buf, + size_t count, loff_t *ppos) +{ + return proc_sys_call_handler(filp, (void __user *)buf, count, ppos, 1); } -- cgit v1.2.3 From d7321cd62470b70d2717dae5a963e7a8fabff4d5 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Tue, 29 Apr 2008 01:02:44 -0700 Subject: sysctl: add the ->permissions callback on the ctl_table_root When reading from/writing to some table, a root, which this table came from, may affect this table's permissions, depending on who is working with the table. The core hunk is at the bottom of this patch. All the rest is just pushing the ctl_table_root argument up to the sysctl_perm() function. This will be mostly (only?) used in the net sysctls. Signed-off-by: Pavel Emelyanov Acked-by: David S. Miller Cc: "Eric W. Biederman" Cc: Alexey Dobriyan Cc: Denis V. Lunev Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/proc_sysctl.c | 4 ++-- include/linux/sysctl.h | 7 ++++++- kernel/sysctl.c | 25 ++++++++++++++++++------- 3 files changed, 26 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 5e31585292a0..5acc001d49f6 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -190,7 +190,7 @@ static ssize_t proc_sys_call_handler(struct file *filp, void __user *buf, * and won't be until we finish. */ error = -EPERM; - if (sysctl_perm(table, write ? MAY_WRITE : MAY_READ)) + if (sysctl_perm(head->root, table, write ? MAY_WRITE : MAY_READ)) goto out; /* careful: calling conventions are nasty here */ @@ -388,7 +388,7 @@ static int proc_sys_permission(struct inode *inode, int mask, struct nameidata * goto out; /* Use the permissions on the sysctl table entry */ - error = sysctl_perm(table, mask); + error = sysctl_perm(head->root, table, mask); out: sysctl_head_finish(head); return error; diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 39eafd8f97a3..24141b4d1a11 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -945,11 +945,14 @@ enum /* For the /proc/sys support */ struct ctl_table; struct nsproxy; +struct ctl_table_root; + extern struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev); extern struct ctl_table_header *__sysctl_head_next(struct nsproxy *namespaces, struct ctl_table_header *prev); extern void sysctl_head_finish(struct ctl_table_header *prev); -extern int sysctl_perm(struct ctl_table *table, int op); +extern int sysctl_perm(struct ctl_table_root *root, + struct ctl_table *table, int op); typedef struct ctl_table ctl_table; @@ -1049,6 +1052,8 @@ struct ctl_table_root { struct list_head header_list; struct list_head *(*lookup)(struct ctl_table_root *root, struct nsproxy *namespaces); + int (*permissions)(struct ctl_table_root *root, + struct nsproxy *namespaces, struct ctl_table *table); }; /* struct ctl_table_header is used to maintain dynamic lists of diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 874e813e40c8..d7ffdc59816a 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1434,7 +1434,8 @@ void register_sysctl_root(struct ctl_table_root *root) #ifdef CONFIG_SYSCTL_SYSCALL /* Perform the actual read/write of a sysctl table entry. */ -static int do_sysctl_strategy(struct ctl_table *table, +static int do_sysctl_strategy(struct ctl_table_root *root, + struct ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen) @@ -1445,7 +1446,7 @@ static int do_sysctl_strategy(struct ctl_table *table, op |= 004; if (newval) op |= 002; - if (sysctl_perm(table, op)) + if (sysctl_perm(root, table, op)) return -EPERM; if (table->strategy) { @@ -1471,6 +1472,7 @@ static int do_sysctl_strategy(struct ctl_table *table, static int parse_table(int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen, + struct ctl_table_root *root, struct ctl_table *table) { int n; @@ -1485,14 +1487,14 @@ repeat: if (n == table->ctl_name) { int error; if (table->child) { - if (sysctl_perm(table, 001)) + if (sysctl_perm(root, table, 001)) return -EPERM; name++; nlen--; table = table->child; goto repeat; } - error = do_sysctl_strategy(table, name, nlen, + error = do_sysctl_strategy(root, table, name, nlen, oldval, oldlenp, newval, newlen); return error; @@ -1518,7 +1520,8 @@ int do_sysctl(int __user *name, int nlen, void __user *oldval, size_t __user *ol for (head = sysctl_head_next(NULL); head; head = sysctl_head_next(head)) { error = parse_table(name, nlen, oldval, oldlenp, - newval, newlen, head->ctl_table); + newval, newlen, + head->root, head->ctl_table); if (error != -ENOTDIR) { sysctl_head_finish(head); break; @@ -1564,13 +1567,21 @@ static int test_perm(int mode, int op) return -EACCES; } -int sysctl_perm(struct ctl_table *table, int op) +int sysctl_perm(struct ctl_table_root *root, struct ctl_table *table, int op) { int error; + int mode; + error = security_sysctl(table, op); if (error) return error; - return test_perm(table->mode, op); + + if (root->permissions) + mode = root->permissions(root, current->nsproxy, table); + else + mode = table->mode; + + return test_perm(mode, op); } static void sysctl_set_parent(struct ctl_table *parent, struct ctl_table *table) -- cgit v1.2.3 From 801678c5a3b4c79236970bcca27c733f5559e0d1 Mon Sep 17 00:00:00 2001 From: Hirofumi Nakagawa Date: Tue, 29 Apr 2008 01:03:09 -0700 Subject: Remove duplicated unlikely() in IS_ERR() Some drivers have duplicated unlikely() macros. IS_ERR() already has unlikely() in itself. This patch cleans up such pointless code. Signed-off-by: Hirofumi Nakagawa Acked-by: David S. Miller Acked-by: Jeff Garzik Cc: Paul Clements Cc: Richard Purdie Cc: Alessandro Zummo Cc: David Brownell Cc: James Bottomley Cc: Michael Halcrow Cc: Anton Altaparmakov Cc: Al Viro Cc: Carsten Otte Cc: Patrick McHardy Cc: Paul Mundt Cc: Jaroslav Kysela Cc: Takashi Iwai Acked-by: Mike Frysinger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/nbd.c | 2 +- drivers/leds/led-class.c | 2 +- drivers/message/i2o/i2o_block.c | 2 +- drivers/net/myri10ge/myri10ge.c | 2 +- drivers/net/tg3.c | 2 +- drivers/rtc/rtc-bfin.c | 2 +- drivers/scsi/scsi_scan.c | 2 +- fs/aio.c | 2 +- fs/ecryptfs/inode.c | 2 +- fs/inotify_user.c | 2 +- fs/ntfs/mft.c | 6 +++--- kernel/auditfilter.c | 10 +++++----- net/core/dev.c | 2 +- net/ipv4/af_inet.c | 2 +- net/netfilter/nf_queue.c | 2 +- net/xfrm/xfrm_output.c | 2 +- sound/sh/aica.c | 2 +- 17 files changed, 23 insertions(+), 23 deletions(-) (limited to 'fs') diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index a3da198ca429..bdba282f15e4 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -339,7 +339,7 @@ static struct request *nbd_read_stat(struct nbd_device *lo) } req = nbd_find_request(lo, *(struct request **)reply.handle); - if (unlikely(IS_ERR(req))) { + if (IS_ERR(req)) { result = PTR_ERR(req); if (result != -ENOENT) goto harderror; diff --git a/drivers/leds/led-class.c b/drivers/leds/led-class.c index ac05a928f764..b3c54be74556 100644 --- a/drivers/leds/led-class.c +++ b/drivers/leds/led-class.c @@ -105,7 +105,7 @@ int led_classdev_register(struct device *parent, struct led_classdev *led_cdev) led_cdev->dev = device_create(leds_class, parent, 0, "%s", led_cdev->name); - if (unlikely(IS_ERR(led_cdev->dev))) + if (IS_ERR(led_cdev->dev)) return PTR_ERR(led_cdev->dev); dev_set_drvdata(led_cdev->dev, led_cdev); diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c index a95314897402..81483de8c0fd 100644 --- a/drivers/message/i2o/i2o_block.c +++ b/drivers/message/i2o/i2o_block.c @@ -371,7 +371,7 @@ static int i2o_block_prep_req_fn(struct request_queue *q, struct request *req) /* connect the i2o_block_request to the request */ if (!req->special) { ireq = i2o_block_request_alloc(); - if (unlikely(IS_ERR(ireq))) { + if (IS_ERR(ireq)) { osm_debug("unable to allocate i2o_block_request!\n"); return BLKPREP_DEFER; } diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c index cead81e80f0c..ef63c8d2bd7e 100644 --- a/drivers/net/myri10ge/myri10ge.c +++ b/drivers/net/myri10ge/myri10ge.c @@ -2437,7 +2437,7 @@ static int myri10ge_sw_tso(struct sk_buff *skb, struct net_device *dev) int status; segs = skb_gso_segment(skb, dev->features & ~NETIF_F_TSO6); - if (unlikely(IS_ERR(segs))) + if (IS_ERR(segs)) goto drop; while (segs) { diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index e3f74c9f78bd..b66c75e3b8a1 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -4361,7 +4361,7 @@ static int tg3_tso_bug(struct tg3 *tp, struct sk_buff *skb) } segs = skb_gso_segment(skb, tp->dev->features & ~NETIF_F_TSO); - if (unlikely(IS_ERR(segs))) + if (IS_ERR(segs)) goto tg3_tso_bug_end; do { diff --git a/drivers/rtc/rtc-bfin.c b/drivers/rtc/rtc-bfin.c index 4f28045d9ef2..8624f55d0560 100644 --- a/drivers/rtc/rtc-bfin.c +++ b/drivers/rtc/rtc-bfin.c @@ -419,7 +419,7 @@ static int __devinit bfin_rtc_probe(struct platform_device *pdev) return -ENOMEM; rtc->rtc_dev = rtc_device_register(pdev->name, &pdev->dev, &bfin_rtc_ops, THIS_MODULE); - if (unlikely(IS_ERR(rtc))) { + if (IS_ERR(rtc)) { ret = PTR_ERR(rtc->rtc_dev); goto err; } diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index fcd7455ffc39..a00eee6f7be9 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -1828,7 +1828,7 @@ void scsi_scan_host(struct Scsi_Host *shost) } p = kthread_run(do_scan_async, data, "scsi_scan_%d", shost->host_no); - if (unlikely(IS_ERR(p))) + if (IS_ERR(p)) do_scan_async(data); } EXPORT_SYMBOL(scsi_scan_host); diff --git a/fs/aio.c b/fs/aio.c index 81c01290939b..a175ad650b66 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -1604,7 +1604,7 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, * event using the eventfd_signal() function. */ req->ki_eventfd = eventfd_fget((int) iocb->aio_resfd); - if (unlikely(IS_ERR(req->ki_eventfd))) { + if (IS_ERR(req->ki_eventfd)) { ret = PTR_ERR(req->ki_eventfd); goto out_put_req; } diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 1623ebf25e51..0a1397335a8e 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -111,7 +111,7 @@ ecryptfs_do_create(struct inode *directory_inode, lower_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry); lower_dir_dentry = lock_parent(lower_dentry); - if (unlikely(IS_ERR(lower_dir_dentry))) { + if (IS_ERR(lower_dir_dentry)) { ecryptfs_printk(KERN_ERR, "Error locking directory of " "dentry\n"); rc = PTR_ERR(lower_dir_dentry); diff --git a/fs/inotify_user.c b/fs/inotify_user.c index 7b94a1e3c015..6676c06bb7c1 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -598,7 +598,7 @@ asmlinkage long sys_inotify_init(void) } ih = inotify_init(&inotify_user_ops); - if (unlikely(IS_ERR(ih))) { + if (IS_ERR(ih)) { ret = PTR_ERR(ih); goto out_free_dev; } diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c index 2ad5c8b104b9..790defb847e7 100644 --- a/fs/ntfs/mft.c +++ b/fs/ntfs/mft.c @@ -1191,7 +1191,7 @@ static int ntfs_mft_bitmap_find_and_alloc_free_rec_nolock(ntfs_volume *vol, if (size) { page = ntfs_map_page(mftbmp_mapping, ofs >> PAGE_CACHE_SHIFT); - if (unlikely(IS_ERR(page))) { + if (IS_ERR(page)) { ntfs_error(vol->sb, "Failed to read mft " "bitmap, aborting."); return PTR_ERR(page); @@ -2118,7 +2118,7 @@ static int ntfs_mft_record_format(const ntfs_volume *vol, const s64 mft_no) } /* Read, map, and pin the page containing the mft record. */ page = ntfs_map_page(mft_vi->i_mapping, index); - if (unlikely(IS_ERR(page))) { + if (IS_ERR(page)) { ntfs_error(vol->sb, "Failed to map page containing mft record " "to format 0x%llx.", (long long)mft_no); return PTR_ERR(page); @@ -2519,7 +2519,7 @@ mft_rec_already_initialized: ofs = (bit << vol->mft_record_size_bits) & ~PAGE_CACHE_MASK; /* Read, map, and pin the page containing the mft record. */ page = ntfs_map_page(vol->mft_ino->i_mapping, index); - if (unlikely(IS_ERR(page))) { + if (IS_ERR(page)) { ntfs_error(vol->sb, "Failed to map page containing allocated " "mft record 0x%llx.", (long long)bit); err = PTR_ERR(page); diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c index 28fef6bf8534..13430176b3c9 100644 --- a/kernel/auditfilter.c +++ b/kernel/auditfilter.c @@ -272,7 +272,7 @@ static int audit_to_watch(struct audit_krule *krule, char *path, int len, return -EINVAL; watch = audit_init_watch(path); - if (unlikely(IS_ERR(watch))) + if (IS_ERR(watch)) return PTR_ERR(watch); audit_get_watch(watch); @@ -848,7 +848,7 @@ static struct audit_watch *audit_dupe_watch(struct audit_watch *old) return ERR_PTR(-ENOMEM); new = audit_init_watch(path); - if (unlikely(IS_ERR(new))) { + if (IS_ERR(new)) { kfree(path); goto out; } @@ -989,7 +989,7 @@ static void audit_update_watch(struct audit_parent *parent, audit_set_auditable(current->audit_context); nwatch = audit_dupe_watch(owatch); - if (unlikely(IS_ERR(nwatch))) { + if (IS_ERR(nwatch)) { mutex_unlock(&audit_filter_mutex); audit_panic("error updating watch, skipping"); return; @@ -1004,7 +1004,7 @@ static void audit_update_watch(struct audit_parent *parent, list_del_rcu(&oentry->list); nentry = audit_dupe_rule(&oentry->rule, nwatch); - if (unlikely(IS_ERR(nentry))) + if (IS_ERR(nentry)) audit_panic("error updating watch, removing"); else { int h = audit_hash_ino((u32)ino); @@ -1785,7 +1785,7 @@ int audit_update_lsm_rules(void) watch = entry->rule.watch; tree = entry->rule.tree; nentry = audit_dupe_rule(&entry->rule, watch); - if (unlikely(IS_ERR(nentry))) { + if (IS_ERR(nentry)) { /* save the first error encountered for the * return value */ if (!err) diff --git a/net/core/dev.c b/net/core/dev.c index e1df1ab3e04a..ed49da592051 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1524,7 +1524,7 @@ static int dev_gso_segment(struct sk_buff *skb) if (!segs) return 0; - if (unlikely(IS_ERR(segs))) + if (IS_ERR(segs)) return PTR_ERR(segs); skb->next = segs; diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index f2b5270efdaa..24eca23c2db3 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1234,7 +1234,7 @@ static struct sk_buff *inet_gso_segment(struct sk_buff *skb, int features) segs = ops->gso_segment(skb, features); rcu_read_unlock(); - if (!segs || unlikely(IS_ERR(segs))) + if (!segs || IS_ERR(segs)) goto out; skb = segs; diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index bbd26893c0c4..582ec3efc8a5 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -214,7 +214,7 @@ int nf_queue(struct sk_buff *skb, segs = skb_gso_segment(skb, 0); kfree_skb(skb); - if (unlikely(IS_ERR(segs))) + if (IS_ERR(segs)) return 1; do { diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c index 2519129c6d21..09cd9c0c2d80 100644 --- a/net/xfrm/xfrm_output.c +++ b/net/xfrm/xfrm_output.c @@ -150,7 +150,7 @@ static int xfrm_output_gso(struct sk_buff *skb) segs = skb_gso_segment(skb, 0); kfree_skb(skb); - if (unlikely(IS_ERR(segs))) + if (IS_ERR(segs)) return PTR_ERR(segs); do { diff --git a/sound/sh/aica.c b/sound/sh/aica.c index d49417bf78c6..9ca113326143 100644 --- a/sound/sh/aica.c +++ b/sound/sh/aica.c @@ -663,7 +663,7 @@ static int __init aica_init(void) return err; pd = platform_device_register_simple(SND_AICA_DRIVER, -1, aica_memory_space, 2); - if (unlikely(IS_ERR(pd))) { + if (IS_ERR(pd)) { platform_driver_unregister(&snd_aica_driver); return PTR_ERR(pd); } -- cgit v1.2.3 From 0ae52d6fbaf7ffe4d00876d25ea000e94f85819c Mon Sep 17 00:00:00 2001 From: "Robert P. J. Day" Date: Tue, 29 Apr 2008 01:03:20 -0700 Subject: afs: use the shorter LIST_HEAD for brevity Signed-off-by: Robert P. J. Day Acked-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/afs/cell.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/afs/cell.c b/fs/afs/cell.c index 584bb0f9c36a..5e1df14e16b1 100644 --- a/fs/afs/cell.c +++ b/fs/afs/cell.c @@ -20,7 +20,7 @@ DECLARE_RWSEM(afs_proc_cells_sem); LIST_HEAD(afs_proc_cells); -static struct list_head afs_cells = LIST_HEAD_INIT(afs_cells); +static LIST_HEAD(afs_cells); static DEFINE_RWLOCK(afs_cells_lock); static DECLARE_RWSEM(afs_cells_sem); /* add/remove serialisation */ static DECLARE_WAIT_QUEUE_HEAD(afs_cells_freeable_wq); -- cgit v1.2.3 From 7c80bcce34a355c0920f8cab250d766d7827341d Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 29 Apr 2008 01:03:21 -0700 Subject: afs: the AFS RPC op CBGetCapabilities is actually CBTellMeAboutYourself The AFS RxRPC op CBGetCapabilities is actually CBTellMeAboutYourself. Signed-off-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/afs/afs_cm.h | 2 +- fs/afs/cmservice.c | 26 +++++++++++++------------- 2 files changed, 14 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/afs/afs_cm.h b/fs/afs/afs_cm.h index 7b4d4fab4c80..5f62f3ea563e 100644 --- a/fs/afs/afs_cm.h +++ b/fs/afs/afs_cm.h @@ -24,7 +24,7 @@ enum AFS_CM_Operations { CBGetXStatsVersion = 209, /* get version of extended statistics */ CBGetXStats = 210, /* get contents of extended statistics data */ CBInitCallBackState3 = 213, /* initialise callback state, version 3 */ - CBGetCapabilities = 65538, /* get client capabilities */ + CBTellMeAboutYourself = 65538, /* get client capabilities */ }; #define AFS_CAP_ERROR_TRANSLATION 0x1 diff --git a/fs/afs/cmservice.c b/fs/afs/cmservice.c index 47b71c8947f9..75da3d04612c 100644 --- a/fs/afs/cmservice.c +++ b/fs/afs/cmservice.c @@ -26,8 +26,8 @@ static int afs_deliver_cb_init_call_back_state3(struct afs_call *, struct sk_buff *, bool); static int afs_deliver_cb_probe(struct afs_call *, struct sk_buff *, bool); static int afs_deliver_cb_callback(struct afs_call *, struct sk_buff *, bool); -static int afs_deliver_cb_get_capabilities(struct afs_call *, struct sk_buff *, - bool); +static int afs_deliver_cb_tell_me_about_yourself(struct afs_call *, + struct sk_buff *, bool); static void afs_cm_destructor(struct afs_call *); /* @@ -71,11 +71,11 @@ static const struct afs_call_type afs_SRXCBProbe = { }; /* - * CB.GetCapabilities operation type + * CB.TellMeAboutYourself operation type */ -static const struct afs_call_type afs_SRXCBGetCapabilites = { - .name = "CB.GetCapabilities", - .deliver = afs_deliver_cb_get_capabilities, +static const struct afs_call_type afs_SRXCBTellMeAboutYourself = { + .name = "CB.TellMeAboutYourself", + .deliver = afs_deliver_cb_tell_me_about_yourself, .abort_to_error = afs_abort_to_error, .destructor = afs_cm_destructor, }; @@ -103,8 +103,8 @@ bool afs_cm_incoming_call(struct afs_call *call) case CBProbe: call->type = &afs_SRXCBProbe; return true; - case CBGetCapabilities: - call->type = &afs_SRXCBGetCapabilites; + case CBTellMeAboutYourself: + call->type = &afs_SRXCBTellMeAboutYourself; return true; default: return false; @@ -395,7 +395,7 @@ static int afs_deliver_cb_probe(struct afs_call *call, struct sk_buff *skb, /* * allow the fileserver to ask about the cache manager's capabilities */ -static void SRXAFSCB_GetCapabilities(struct work_struct *work) +static void SRXAFSCB_TellMeAboutYourself(struct work_struct *work) { struct afs_interface *ifs; struct afs_call *call = container_of(work, struct afs_call, work); @@ -456,10 +456,10 @@ static void SRXAFSCB_GetCapabilities(struct work_struct *work) } /* - * deliver request data to a CB.GetCapabilities call + * deliver request data to a CB.TellMeAboutYourself call */ -static int afs_deliver_cb_get_capabilities(struct afs_call *call, - struct sk_buff *skb, bool last) +static int afs_deliver_cb_tell_me_about_yourself(struct afs_call *call, + struct sk_buff *skb, bool last) { _enter(",{%u},%d", skb->len, last); @@ -471,7 +471,7 @@ static int afs_deliver_cb_get_capabilities(struct afs_call *call, /* no unmarshalling required */ call->state = AFS_CALL_REPLYING; - INIT_WORK(&call->work, SRXAFSCB_GetCapabilities); + INIT_WORK(&call->work, SRXAFSCB_TellMeAboutYourself); schedule_work(&call->work); return 0; } -- cgit v1.2.3 From 9396d496d74587d46a74b93a8b6b41659d2daf2e Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 29 Apr 2008 01:03:22 -0700 Subject: afs: support the CB.ProbeUuid RPC op Add support for the CB.ProbeUuid cache manager RPC op. This allows a modern OpenAFS server to quickly ask if the client has been rebooted. Signed-off-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/afs/afs_cm.h | 1 + fs/afs/cmservice.c | 107 +++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 108 insertions(+) (limited to 'fs') diff --git a/fs/afs/afs_cm.h b/fs/afs/afs_cm.h index 5f62f3ea563e..255f5dd6040c 100644 --- a/fs/afs/afs_cm.h +++ b/fs/afs/afs_cm.h @@ -24,6 +24,7 @@ enum AFS_CM_Operations { CBGetXStatsVersion = 209, /* get version of extended statistics */ CBGetXStats = 210, /* get contents of extended statistics data */ CBInitCallBackState3 = 213, /* initialise callback state, version 3 */ + CBProbeUuid = 214, /* check the client hasn't rebooted */ CBTellMeAboutYourself = 65538, /* get client capabilities */ }; diff --git a/fs/afs/cmservice.c b/fs/afs/cmservice.c index 75da3d04612c..eb765489164f 100644 --- a/fs/afs/cmservice.c +++ b/fs/afs/cmservice.c @@ -26,6 +26,7 @@ static int afs_deliver_cb_init_call_back_state3(struct afs_call *, struct sk_buff *, bool); static int afs_deliver_cb_probe(struct afs_call *, struct sk_buff *, bool); static int afs_deliver_cb_callback(struct afs_call *, struct sk_buff *, bool); +static int afs_deliver_cb_probe_uuid(struct afs_call *, struct sk_buff *, bool); static int afs_deliver_cb_tell_me_about_yourself(struct afs_call *, struct sk_buff *, bool); static void afs_cm_destructor(struct afs_call *); @@ -70,6 +71,16 @@ static const struct afs_call_type afs_SRXCBProbe = { .destructor = afs_cm_destructor, }; +/* + * CB.ProbeUuid operation type + */ +static const struct afs_call_type afs_SRXCBProbeUuid = { + .name = "CB.ProbeUuid", + .deliver = afs_deliver_cb_probe_uuid, + .abort_to_error = afs_abort_to_error, + .destructor = afs_cm_destructor, +}; + /* * CB.TellMeAboutYourself operation type */ @@ -392,6 +403,102 @@ static int afs_deliver_cb_probe(struct afs_call *call, struct sk_buff *skb, return 0; } +/* + * allow the fileserver to quickly find out if the fileserver has been rebooted + */ +static void SRXAFSCB_ProbeUuid(struct work_struct *work) +{ + struct afs_call *call = container_of(work, struct afs_call, work); + struct afs_uuid *r = call->request; + + struct { + __be32 match; + } reply; + + _enter(""); + + + if (memcmp(r, &afs_uuid, sizeof(afs_uuid)) == 0) + reply.match = htonl(0); + else + reply.match = htonl(1); + + afs_send_simple_reply(call, &reply, sizeof(reply)); + _leave(""); +} + +/* + * deliver request data to a CB.ProbeUuid call + */ +static int afs_deliver_cb_probe_uuid(struct afs_call *call, struct sk_buff *skb, + bool last) +{ + struct afs_uuid *r; + unsigned loop; + __be32 *b; + int ret; + + _enter("{%u},{%u},%d", call->unmarshall, skb->len, last); + + if (skb->len > 0) + return -EBADMSG; + if (!last) + return 0; + + switch (call->unmarshall) { + case 0: + call->offset = 0; + call->buffer = kmalloc(11 * sizeof(__be32), GFP_KERNEL); + if (!call->buffer) + return -ENOMEM; + call->unmarshall++; + + case 1: + _debug("extract UUID"); + ret = afs_extract_data(call, skb, last, call->buffer, + 11 * sizeof(__be32)); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } + + _debug("unmarshall UUID"); + call->request = kmalloc(sizeof(struct afs_uuid), GFP_KERNEL); + if (!call->request) + return -ENOMEM; + + b = call->buffer; + r = call->request; + r->time_low = ntohl(b[0]); + r->time_mid = ntohl(b[1]); + r->time_hi_and_version = ntohl(b[2]); + r->clock_seq_hi_and_reserved = ntohl(b[3]); + r->clock_seq_low = ntohl(b[4]); + + for (loop = 0; loop < 6; loop++) + r->node[loop] = ntohl(b[loop + 5]); + + call->offset = 0; + call->unmarshall++; + + case 2: + _debug("trailer"); + if (skb->len != 0) + return -EBADMSG; + break; + } + + if (!last) + return 0; + + call->state = AFS_CALL_REPLYING; + + INIT_WORK(&call->work, SRXAFSCB_ProbeUuid); + schedule_work(&call->work); + return 0; +} + /* * allow the fileserver to ask about the cache manager's capabilities */ -- cgit v1.2.3 From 803f445f17aa1b71235ad6febae734dd7ad23ddd Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Tue, 29 Apr 2008 01:03:43 -0700 Subject: fat: use get/put_unaligned_* helpers Signed-off-by: Harvey Harrison Acked-by: OGAWA Hirofumi Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/inode.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 5f522a55b596..4e0a3dd9d677 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -1222,8 +1222,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, brelse(bh); goto out_invalid; } - logical_sector_size = - le16_to_cpu(get_unaligned((__le16 *)&b->sector_size)); + logical_sector_size = get_unaligned_le16(&b->sector_size); if (!is_power_of_2(logical_sector_size) || (logical_sector_size < 512) || (logical_sector_size > 4096)) { @@ -1322,8 +1321,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, sbi->dir_per_block_bits = ffs(sbi->dir_per_block) - 1; sbi->dir_start = sbi->fat_start + sbi->fats * sbi->fat_length; - sbi->dir_entries = - le16_to_cpu(get_unaligned((__le16 *)&b->dir_entries)); + sbi->dir_entries = get_unaligned_le16(&b->dir_entries); if (sbi->dir_entries & (sbi->dir_per_block - 1)) { if (!silent) printk(KERN_ERR "FAT: bogus directroy-entries per block" @@ -1335,7 +1333,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, rootdir_sectors = sbi->dir_entries * sizeof(struct msdos_dir_entry) / sb->s_blocksize; sbi->data_start = sbi->dir_start + rootdir_sectors; - total_sectors = le16_to_cpu(get_unaligned((__le16 *)&b->sectors)); + total_sectors = get_unaligned_le16(&b->sectors); if (total_sectors == 0) total_sectors = le32_to_cpu(b->total_sect); -- cgit v1.2.3 From 8b3789e5d552b8ba4841926066ef0ccd664e209c Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Tue, 29 Apr 2008 01:03:44 -0700 Subject: hfsplus: use get/put_unaligned_* helpers Signed-off-by: Harvey Harrison Cc: Roman Zippel Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hfsplus/wrapper.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/hfsplus/wrapper.c b/fs/hfsplus/wrapper.c index 72cab78f0509..175d08eacc86 100644 --- a/fs/hfsplus/wrapper.c +++ b/fs/hfsplus/wrapper.c @@ -47,7 +47,7 @@ static int hfsplus_read_mdb(void *bufptr, struct hfsplus_wd *wd) return 0; wd->ablk_start = be16_to_cpu(*(__be16 *)(bufptr + HFSP_WRAPOFF_ABLKSTART)); - extent = be32_to_cpu(get_unaligned((__be32 *)(bufptr + HFSP_WRAPOFF_EMBEDEXT))); + extent = get_unaligned_be32(bufptr + HFSP_WRAPOFF_EMBEDEXT); wd->embed_start = (extent >> 16) & 0xFFFF; wd->embed_count = extent & 0xFFFF; -- cgit v1.2.3 From 58d485d481013b47f50b7cd2cf9eab7795a0fcbd Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Tue, 29 Apr 2008 01:03:44 -0700 Subject: isofs: use get/put_unaligned_* helpers Signed-off-by: Harvey Harrison Cc: Jan Kara Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/isofs/isofs.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/isofs/isofs.h b/fs/isofs/isofs.h index d1bdf8adb351..ccbf72faf27a 100644 --- a/fs/isofs/isofs.h +++ b/fs/isofs/isofs.h @@ -78,29 +78,29 @@ static inline int isonum_712(char *p) } static inline unsigned int isonum_721(char *p) { - return le16_to_cpu(get_unaligned((__le16 *)p)); + return get_unaligned_le16(p); } static inline unsigned int isonum_722(char *p) { - return be16_to_cpu(get_unaligned((__le16 *)p)); + return get_unaligned_be16(p); } static inline unsigned int isonum_723(char *p) { /* Ignore bigendian datum due to broken mastering programs */ - return le16_to_cpu(get_unaligned((__le16 *)p)); + return get_unaligned_le16(p); } static inline unsigned int isonum_731(char *p) { - return le32_to_cpu(get_unaligned((__le32 *)p)); + return get_unaligned_le32(p); } static inline unsigned int isonum_732(char *p) { - return be32_to_cpu(get_unaligned((__le32 *)p)); + return get_unaligned_be32(p); } static inline unsigned int isonum_733(char *p) { /* Ignore bigendian datum due to broken mastering programs */ - return le32_to_cpu(get_unaligned((__le32 *)p)); + return get_unaligned_le32(p); } extern int iso_date(char *, int); -- cgit v1.2.3 From 97a4feb4a78ae5cd130be7d546471a0779f1aa14 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Tue, 29 Apr 2008 01:03:45 -0700 Subject: ncpfs: use get/put_unaligned_* helpers [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Harvey Harrison Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ncpfs/ncplib_kernel.c | 39 +++++++++++++++++++-------------------- 1 file changed, 19 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/ncpfs/ncplib_kernel.c b/fs/ncpfs/ncplib_kernel.c index df6d60bdfcd3..97645f112114 100644 --- a/fs/ncpfs/ncplib_kernel.c +++ b/fs/ncpfs/ncplib_kernel.c @@ -102,48 +102,47 @@ static inline void ncp_init_request_s(struct ncp_server *server, int subfunction } static inline char * - ncp_reply_data(struct ncp_server *server, int offset) +ncp_reply_data(struct ncp_server *server, int offset) { return &(server->packet[sizeof(struct ncp_reply_header) + offset]); } -static inline __u8 BVAL(void* data) +static inline u8 BVAL(void *data) { - return get_unaligned((__u8*)data); + return *(u8 *)data; } -static __u8 - ncp_reply_byte(struct ncp_server *server, int offset) +static u8 ncp_reply_byte(struct ncp_server *server, int offset) { - return get_unaligned((__u8 *) ncp_reply_data(server, offset)); + return *(u8 *)ncp_reply_data(server, offset); } -static inline __u16 WVAL_LH(void* data) +static inline u16 WVAL_LH(void *data) { - return le16_to_cpu(get_unaligned((__le16*)data)); + return get_unaligned_le16(data); } -static __u16 - ncp_reply_le16(struct ncp_server *server, int offset) +static u16 +ncp_reply_le16(struct ncp_server *server, int offset) { - return le16_to_cpu(get_unaligned((__le16 *) ncp_reply_data(server, offset))); + return get_unaligned_le16(ncp_reply_data(server, offset)); } -static __u16 - ncp_reply_be16(struct ncp_server *server, int offset) +static u16 +ncp_reply_be16(struct ncp_server *server, int offset) { - return be16_to_cpu(get_unaligned((__be16 *) ncp_reply_data(server, offset))); + return get_unaligned_be16(ncp_reply_data(server, offset)); } -static inline __u32 DVAL_LH(void* data) +static inline u32 DVAL_LH(void *data) { - return le32_to_cpu(get_unaligned((__le32*)data)); + return get_unaligned_le32(data); } static __le32 - ncp_reply_dword(struct ncp_server *server, int offset) +ncp_reply_dword(struct ncp_server *server, int offset) { - return get_unaligned((__le32 *) ncp_reply_data(server, offset)); + return get_unaligned((__le32 *)ncp_reply_data(server, offset)); } static inline __u32 ncp_reply_dword_lh(struct ncp_server* server, int offset) { @@ -1006,8 +1005,8 @@ ncp_read_bounce(struct ncp_server *server, const char *file_id, result = ncp_request2(server, 72, bounce, bufsize); ncp_unlock_server(server); if (!result) { - int len = be16_to_cpu(get_unaligned((__be16*)((char*)bounce + - sizeof(struct ncp_reply_header)))); + int len = get_unaligned_be16((char *)bounce + + sizeof(struct ncp_reply_header)); result = -EIO; if (len <= to_read) { char* source; -- cgit v1.2.3 From 39fa00311f21318cc498b139c2cc2830dcad98ff Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Tue, 29 Apr 2008 01:03:48 -0700 Subject: aio: fix misleading comments The FIXME comments are inaccurate. The locking comment over lookup_ioctx() is wrong. Signed-off-by: Jeff Moyer Signed-off-by: Zach Brown Signed-off-by: Shen Feng Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/aio.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/aio.c b/fs/aio.c index a175ad650b66..99c2352906a0 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -277,7 +277,7 @@ static struct kioctx *ioctx_alloc(unsigned nr_events) if (ctx->max_reqs == 0) goto out_cleanup; - /* now link into global list. kludge. FIXME */ + /* now link into global list. */ write_lock(&mm->ioctx_list_lock); ctx->next = mm->ioctx_list; mm->ioctx_list = ctx; @@ -553,9 +553,6 @@ int aio_put_req(struct kiocb *req) return ret; } -/* Lookup an ioctx id. ioctx_list is lockless for reads. - * FIXME: this is O(n) and is only suitable for development. - */ static struct kioctx *lookup_ioctx(unsigned long ctx_id) { struct kioctx *ioctx; -- cgit v1.2.3 From 355a46961b58012de239cafccbfce4c9321d4395 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Tue, 29 Apr 2008 16:01:22 -0400 Subject: trivial: fix user-visible typo in hfsplus Signed-off-by: Dave Jones Signed-off-by: Linus Torvalds --- fs/hfsplus/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index b0f9ad362d1d..946466cd9f25 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -357,7 +357,7 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) printk(KERN_WARNING "hfs: Filesystem is marked locked, mounting read-only.\n"); sb->s_flags |= MS_RDONLY; } else if (vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_JOURNALED)) { - printk(KERN_WARNING "hfs: write access to a jounaled filesystem is not supported, " + printk(KERN_WARNING "hfs: write access to a journaled filesystem is not supported, " "use the force option at your own risk, mounting read-only.\n"); sb->s_flags |= MS_RDONLY; } -- cgit v1.2.3 From 53b7e9f6807c1274eee19201396b4c2b5f721553 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 29 Apr 2008 22:02:11 -0400 Subject: ext4: Fix update of mtime and ctime on rename The patch below makes ext4 update mtime and ctime of the directory into which we move file even if the directory entry already exists. Signed-off-by: Jan Kara Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/namei.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 02cdaec39e21..7fc1bc1c16d1 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -2354,6 +2354,9 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, EXT4_FEATURE_INCOMPAT_FILETYPE)) new_de->file_type = old_de->file_type; new_dir->i_version++; + new_dir->i_ctime = new_dir->i_mtime = + ext4_current_time(new_dir); + ext4_mark_inode_dirty(handle, new_dir); BUFFER_TRACE(new_bh, "call ext4_journal_dirty_metadata"); ext4_journal_dirty_metadata(handle, new_bh); brelse(new_bh); -- cgit v1.2.3 From 2887df139c40512cdc147d1a84d95d4f3d261bd1 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 29 Apr 2008 22:02:07 -0400 Subject: ext4: Fix hang on umount with quotas when journal is aborted Call dquot_drop() from ext4_dquot_drop() even if we fail to start a transaction. Otherwise we never get to dropping references to quota structures from the inode and umount will hang indefinitely. Thanks to Payphone LIOU for spotting the problem. Signed-off-by: Jan Kara Signed-off-by: Mingming Cao CC: Payphone LIOU Signed-off-by: "Theodore Ts'o" --- fs/ext4/super.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 93a7e746f425..e3b3483b600d 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -3040,8 +3040,14 @@ static int ext4_dquot_drop(struct inode *inode) /* We may delete quota structure so we need to reserve enough blocks */ handle = ext4_journal_start(inode, 2*EXT4_QUOTA_DEL_BLOCKS(inode->i_sb)); - if (IS_ERR(handle)) + if (IS_ERR(handle)) { + /* + * We call dquot_drop() anyway to at least release references + * to quota structures so that umount does not hang. + */ + dquot_drop(inode); return PTR_ERR(handle); + } ret = dquot_drop(inode); err = ext4_journal_stop(handle); if (!ret) -- cgit v1.2.3 From 216553c4b7f3e3e2beb4981cddca9b2027523928 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Tue, 29 Apr 2008 22:02:02 -0400 Subject: ext4: fix wrong gfp type under transaction This fixes the allocations with GFP_KERNEL while under a transaction problems in ext4. This patch is the same as its ext3 counterpart, just switches these to GFP_NOFS. Signed-off-by: Josef Bacik Cc: Signed-off-by: Andrew Morton Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/acl.c | 8 ++++---- fs/ext4/extents.c | 2 +- fs/ext4/resize.c | 4 ++-- fs/ext4/xattr.c | 4 ++-- 4 files changed, 9 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c index a8bae8cd1d5d..cc92624a46a2 100644 --- a/fs/ext4/acl.c +++ b/fs/ext4/acl.c @@ -37,7 +37,7 @@ ext4_acl_from_disk(const void *value, size_t size) return ERR_PTR(-EINVAL); if (count == 0) return NULL; - acl = posix_acl_alloc(count, GFP_KERNEL); + acl = posix_acl_alloc(count, GFP_NOFS); if (!acl) return ERR_PTR(-ENOMEM); for (n=0; n < count; n++) { @@ -91,7 +91,7 @@ ext4_acl_to_disk(const struct posix_acl *acl, size_t *size) *size = ext4_acl_size(acl->a_count); ext_acl = kmalloc(sizeof(ext4_acl_header) + acl->a_count * - sizeof(ext4_acl_entry), GFP_KERNEL); + sizeof(ext4_acl_entry), GFP_NOFS); if (!ext_acl) return ERR_PTR(-ENOMEM); ext_acl->a_version = cpu_to_le32(EXT4_ACL_VERSION); @@ -187,7 +187,7 @@ ext4_get_acl(struct inode *inode, int type) } retval = ext4_xattr_get(inode, name_index, "", NULL, 0); if (retval > 0) { - value = kmalloc(retval, GFP_KERNEL); + value = kmalloc(retval, GFP_NOFS); if (!value) return ERR_PTR(-ENOMEM); retval = ext4_xattr_get(inode, name_index, "", value, retval); @@ -335,7 +335,7 @@ ext4_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) if (error) goto cleanup; } - clone = posix_acl_clone(acl, GFP_KERNEL); + clone = posix_acl_clone(acl, GFP_NOFS); error = -ENOMEM; if (!clone) goto cleanup; diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 1c8aab4dad23..4e6afc812fda 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -1977,7 +1977,7 @@ static int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start) * We start scanning from right side, freeing all the blocks * after i_size and walking into the tree depth-wise. */ - path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 1), GFP_KERNEL); + path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 1), GFP_NOFS); if (path == NULL) { ext4_journal_stop(handle); return -ENOMEM; diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 3e0f5d06f3ee..0ca63dcbdf88 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -469,7 +469,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, goto exit_dindj; n_group_desc = kmalloc((gdb_num + 1) * sizeof(struct buffer_head *), - GFP_KERNEL); + GFP_NOFS); if (!n_group_desc) { err = -ENOMEM; ext4_warning(sb, __func__, @@ -552,7 +552,7 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode, int res, i; int err; - primary = kmalloc(reserved_gdb * sizeof(*primary), GFP_KERNEL); + primary = kmalloc(reserved_gdb * sizeof(*primary), GFP_NOFS); if (!primary) return -ENOMEM; diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index f56598df6880..df4810d5a387 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -739,7 +739,7 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, ce = NULL; } ea_bdebug(bs->bh, "cloning"); - s->base = kmalloc(bs->bh->b_size, GFP_KERNEL); + s->base = kmalloc(bs->bh->b_size, GFP_NOFS); error = -ENOMEM; if (s->base == NULL) goto cleanup; @@ -751,7 +751,7 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, } } else { /* Allocate a buffer where we construct the new block. */ - s->base = kzalloc(sb->s_blocksize, GFP_KERNEL); + s->base = kzalloc(sb->s_blocksize, GFP_NOFS); /* assert(header == s->base) */ error = -ENOMEM; if (s->base == NULL) -- cgit v1.2.3 From 3dcf54515aa4981a647ad74859199032965193a5 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 29 Apr 2008 18:13:32 -0400 Subject: ext4: move headers out of include/linux Move ext4 headers out of include/linux. This is just the trivial move, there's some more thing that could be done later. Signed-off-by: Christoph Hellwig Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/acl.c | 4 +- fs/ext4/balloc.c | 6 +- fs/ext4/bitmap.c | 2 +- fs/ext4/dir.c | 2 +- fs/ext4/ext4.h | 1205 +++++++++++++++++++++++++++++++++++++++ fs/ext4/ext4_extents.h | 232 ++++++++ fs/ext4/ext4_i.h | 167 ++++++ fs/ext4/ext4_jbd2.c | 2 +- fs/ext4/ext4_jbd2.h | 231 ++++++++ fs/ext4/ext4_sb.h | 148 +++++ fs/ext4/extents.c | 4 +- fs/ext4/file.c | 4 +- fs/ext4/fsync.c | 4 +- fs/ext4/hash.c | 2 +- fs/ext4/ialloc.c | 5 +- fs/ext4/inode.c | 2 +- fs/ext4/ioctl.c | 4 +- fs/ext4/mballoc.c | 4 +- fs/ext4/migrate.c | 4 +- fs/ext4/namei.c | 4 +- fs/ext4/resize.c | 3 +- fs/ext4/super.c | 5 +- fs/ext4/symlink.c | 2 +- fs/ext4/xattr.c | 4 +- fs/ext4/xattr_security.c | 4 +- fs/ext4/xattr_trusted.c | 4 +- fs/ext4/xattr_user.c | 4 +- include/linux/ext4_fs.h | 1205 --------------------------------------- include/linux/ext4_fs_extents.h | 232 -------- include/linux/ext4_fs_i.h | 167 ------ include/linux/ext4_fs_sb.h | 148 ----- include/linux/ext4_jbd2.h | 231 -------- 32 files changed, 2021 insertions(+), 2024 deletions(-) create mode 100644 fs/ext4/ext4.h create mode 100644 fs/ext4/ext4_extents.h create mode 100644 fs/ext4/ext4_i.h create mode 100644 fs/ext4/ext4_jbd2.h create mode 100644 fs/ext4/ext4_sb.h delete mode 100644 include/linux/ext4_fs.h delete mode 100644 include/linux/ext4_fs_extents.h delete mode 100644 include/linux/ext4_fs_i.h delete mode 100644 include/linux/ext4_fs_sb.h delete mode 100644 include/linux/ext4_jbd2.h (limited to 'fs') diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c index cc92624a46a2..3c8dab880d91 100644 --- a/fs/ext4/acl.c +++ b/fs/ext4/acl.c @@ -9,8 +9,8 @@ #include #include #include -#include -#include +#include "ext4_jbd2.h" +#include "ext4.h" #include "xattr.h" #include "acl.h" diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index af5032b23c29..da994374ec3b 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -15,12 +15,12 @@ #include #include #include -#include -#include #include #include - +#include "ext4.h" +#include "ext4_jbd2.h" #include "group.h" + /* * balloc.c contains the blocks allocation and deallocation routines */ diff --git a/fs/ext4/bitmap.c b/fs/ext4/bitmap.c index 420554f8f79d..d37ea6750454 100644 --- a/fs/ext4/bitmap.c +++ b/fs/ext4/bitmap.c @@ -9,7 +9,7 @@ #include #include -#include +#include "ext4.h" #ifdef EXT4FS_DEBUG diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index 88c97f7312be..2bf0331ea194 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -23,10 +23,10 @@ #include #include -#include #include #include #include +#include "ext4.h" static unsigned char ext4_filetype_table[] = { DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, DT_FIFO, DT_SOCK, DT_LNK diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h new file mode 100644 index 000000000000..8158083f7ac0 --- /dev/null +++ b/fs/ext4/ext4.h @@ -0,0 +1,1205 @@ +/* + * ext4.h + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/include/linux/minix_fs.h + * + * Copyright (C) 1991, 1992 Linus Torvalds + */ + +#ifndef _EXT4_H +#define _EXT4_H + +#include +#include +#include +#include "ext4_i.h" + +/* + * The second extended filesystem constants/structures + */ + +/* + * Define EXT4FS_DEBUG to produce debug messages + */ +#undef EXT4FS_DEBUG + +/* + * Define EXT4_RESERVATION to reserve data blocks for expanding files + */ +#define EXT4_DEFAULT_RESERVE_BLOCKS 8 +/*max window size: 1024(direct blocks) + 3([t,d]indirect blocks) */ +#define EXT4_MAX_RESERVE_BLOCKS 1027 +#define EXT4_RESERVE_WINDOW_NOT_ALLOCATED 0 + +/* + * Debug code + */ +#ifdef EXT4FS_DEBUG +#define ext4_debug(f, a...) \ + do { \ + printk (KERN_DEBUG "EXT4-fs DEBUG (%s, %d): %s:", \ + __FILE__, __LINE__, __FUNCTION__); \ + printk (KERN_DEBUG f, ## a); \ + } while (0) +#else +#define ext4_debug(f, a...) do {} while (0) +#endif + +#define EXT4_MULTIBLOCK_ALLOCATOR 1 + +/* prefer goal again. length */ +#define EXT4_MB_HINT_MERGE 1 +/* blocks already reserved */ +#define EXT4_MB_HINT_RESERVED 2 +/* metadata is being allocated */ +#define EXT4_MB_HINT_METADATA 4 +/* first blocks in the file */ +#define EXT4_MB_HINT_FIRST 8 +/* search for the best chunk */ +#define EXT4_MB_HINT_BEST 16 +/* data is being allocated */ +#define EXT4_MB_HINT_DATA 32 +/* don't preallocate (for tails) */ +#define EXT4_MB_HINT_NOPREALLOC 64 +/* allocate for locality group */ +#define EXT4_MB_HINT_GROUP_ALLOC 128 +/* allocate goal blocks or none */ +#define EXT4_MB_HINT_GOAL_ONLY 256 +/* goal is meaningful */ +#define EXT4_MB_HINT_TRY_GOAL 512 + +struct ext4_allocation_request { + /* target inode for block we're allocating */ + struct inode *inode; + /* logical block in target inode */ + ext4_lblk_t logical; + /* phys. target (a hint) */ + ext4_fsblk_t goal; + /* the closest logical allocated block to the left */ + ext4_lblk_t lleft; + /* phys. block for ^^^ */ + ext4_fsblk_t pleft; + /* the closest logical allocated block to the right */ + ext4_lblk_t lright; + /* phys. block for ^^^ */ + ext4_fsblk_t pright; + /* how many blocks we want to allocate */ + unsigned long len; + /* flags. see above EXT4_MB_HINT_* */ + unsigned long flags; +}; + +/* + * Special inodes numbers + */ +#define EXT4_BAD_INO 1 /* Bad blocks inode */ +#define EXT4_ROOT_INO 2 /* Root inode */ +#define EXT4_BOOT_LOADER_INO 5 /* Boot loader inode */ +#define EXT4_UNDEL_DIR_INO 6 /* Undelete directory inode */ +#define EXT4_RESIZE_INO 7 /* Reserved group descriptors inode */ +#define EXT4_JOURNAL_INO 8 /* Journal inode */ + +/* First non-reserved inode for old ext4 filesystems */ +#define EXT4_GOOD_OLD_FIRST_INO 11 + +/* + * Maximal count of links to a file + */ +#define EXT4_LINK_MAX 65000 + +/* + * Macro-instructions used to manage several block sizes + */ +#define EXT4_MIN_BLOCK_SIZE 1024 +#define EXT4_MAX_BLOCK_SIZE 65536 +#define EXT4_MIN_BLOCK_LOG_SIZE 10 +#ifdef __KERNEL__ +# define EXT4_BLOCK_SIZE(s) ((s)->s_blocksize) +#else +# define EXT4_BLOCK_SIZE(s) (EXT4_MIN_BLOCK_SIZE << (s)->s_log_block_size) +#endif +#define EXT4_ADDR_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / sizeof (__u32)) +#ifdef __KERNEL__ +# define EXT4_BLOCK_SIZE_BITS(s) ((s)->s_blocksize_bits) +#else +# define EXT4_BLOCK_SIZE_BITS(s) ((s)->s_log_block_size + 10) +#endif +#ifdef __KERNEL__ +#define EXT4_ADDR_PER_BLOCK_BITS(s) (EXT4_SB(s)->s_addr_per_block_bits) +#define EXT4_INODE_SIZE(s) (EXT4_SB(s)->s_inode_size) +#define EXT4_FIRST_INO(s) (EXT4_SB(s)->s_first_ino) +#else +#define EXT4_INODE_SIZE(s) (((s)->s_rev_level == EXT4_GOOD_OLD_REV) ? \ + EXT4_GOOD_OLD_INODE_SIZE : \ + (s)->s_inode_size) +#define EXT4_FIRST_INO(s) (((s)->s_rev_level == EXT4_GOOD_OLD_REV) ? \ + EXT4_GOOD_OLD_FIRST_INO : \ + (s)->s_first_ino) +#endif +#define EXT4_BLOCK_ALIGN(size, blkbits) ALIGN((size), (1 << (blkbits))) + +/* + * Structure of a blocks group descriptor + */ +struct ext4_group_desc +{ + __le32 bg_block_bitmap_lo; /* Blocks bitmap block */ + __le32 bg_inode_bitmap_lo; /* Inodes bitmap block */ + __le32 bg_inode_table_lo; /* Inodes table block */ + __le16 bg_free_blocks_count; /* Free blocks count */ + __le16 bg_free_inodes_count; /* Free inodes count */ + __le16 bg_used_dirs_count; /* Directories count */ + __le16 bg_flags; /* EXT4_BG_flags (INODE_UNINIT, etc) */ + __u32 bg_reserved[2]; /* Likely block/inode bitmap checksum */ + __le16 bg_itable_unused; /* Unused inodes count */ + __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ + __le32 bg_block_bitmap_hi; /* Blocks bitmap block MSB */ + __le32 bg_inode_bitmap_hi; /* Inodes bitmap block MSB */ + __le32 bg_inode_table_hi; /* Inodes table block MSB */ + __le16 bg_free_blocks_count_hi;/* Free blocks count MSB */ + __le16 bg_free_inodes_count_hi;/* Free inodes count MSB */ + __le16 bg_used_dirs_count_hi; /* Directories count MSB */ + __le16 bg_itable_unused_hi; /* Unused inodes count MSB */ + __u32 bg_reserved2[3]; +}; + +#define EXT4_BG_INODE_UNINIT 0x0001 /* Inode table/bitmap not in use */ +#define EXT4_BG_BLOCK_UNINIT 0x0002 /* Block bitmap not in use */ +#define EXT4_BG_INODE_ZEROED 0x0004 /* On-disk itable initialized to zero */ + +#ifdef __KERNEL__ +#include "ext4_sb.h" +#endif +/* + * Macro-instructions used to manage group descriptors + */ +#define EXT4_MIN_DESC_SIZE 32 +#define EXT4_MIN_DESC_SIZE_64BIT 64 +#define EXT4_MAX_DESC_SIZE EXT4_MIN_BLOCK_SIZE +#define EXT4_DESC_SIZE(s) (EXT4_SB(s)->s_desc_size) +#ifdef __KERNEL__ +# define EXT4_BLOCKS_PER_GROUP(s) (EXT4_SB(s)->s_blocks_per_group) +# define EXT4_DESC_PER_BLOCK(s) (EXT4_SB(s)->s_desc_per_block) +# define EXT4_INODES_PER_GROUP(s) (EXT4_SB(s)->s_inodes_per_group) +# define EXT4_DESC_PER_BLOCK_BITS(s) (EXT4_SB(s)->s_desc_per_block_bits) +#else +# define EXT4_BLOCKS_PER_GROUP(s) ((s)->s_blocks_per_group) +# define EXT4_DESC_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / EXT4_DESC_SIZE(s)) +# define EXT4_INODES_PER_GROUP(s) ((s)->s_inodes_per_group) +#endif + +/* + * Constants relative to the data blocks + */ +#define EXT4_NDIR_BLOCKS 12 +#define EXT4_IND_BLOCK EXT4_NDIR_BLOCKS +#define EXT4_DIND_BLOCK (EXT4_IND_BLOCK + 1) +#define EXT4_TIND_BLOCK (EXT4_DIND_BLOCK + 1) +#define EXT4_N_BLOCKS (EXT4_TIND_BLOCK + 1) + +/* + * Inode flags + */ +#define EXT4_SECRM_FL 0x00000001 /* Secure deletion */ +#define EXT4_UNRM_FL 0x00000002 /* Undelete */ +#define EXT4_COMPR_FL 0x00000004 /* Compress file */ +#define EXT4_SYNC_FL 0x00000008 /* Synchronous updates */ +#define EXT4_IMMUTABLE_FL 0x00000010 /* Immutable file */ +#define EXT4_APPEND_FL 0x00000020 /* writes to file may only append */ +#define EXT4_NODUMP_FL 0x00000040 /* do not dump file */ +#define EXT4_NOATIME_FL 0x00000080 /* do not update atime */ +/* Reserved for compression usage... */ +#define EXT4_DIRTY_FL 0x00000100 +#define EXT4_COMPRBLK_FL 0x00000200 /* One or more compressed clusters */ +#define EXT4_NOCOMPR_FL 0x00000400 /* Don't compress */ +#define EXT4_ECOMPR_FL 0x00000800 /* Compression error */ +/* End compression flags --- maybe not all used */ +#define EXT4_INDEX_FL 0x00001000 /* hash-indexed directory */ +#define EXT4_IMAGIC_FL 0x00002000 /* AFS directory */ +#define EXT4_JOURNAL_DATA_FL 0x00004000 /* file data should be journaled */ +#define EXT4_NOTAIL_FL 0x00008000 /* file tail should not be merged */ +#define EXT4_DIRSYNC_FL 0x00010000 /* dirsync behaviour (directories only) */ +#define EXT4_TOPDIR_FL 0x00020000 /* Top of directory hierarchies*/ +#define EXT4_HUGE_FILE_FL 0x00040000 /* Set to each huge file */ +#define EXT4_EXTENTS_FL 0x00080000 /* Inode uses extents */ +#define EXT4_EXT_MIGRATE 0x00100000 /* Inode is migrating */ +#define EXT4_RESERVED_FL 0x80000000 /* reserved for ext4 lib */ + +#define EXT4_FL_USER_VISIBLE 0x000BDFFF /* User visible flags */ +#define EXT4_FL_USER_MODIFIABLE 0x000380FF /* User modifiable flags */ + +/* + * Inode dynamic state flags + */ +#define EXT4_STATE_JDATA 0x00000001 /* journaled data exists */ +#define EXT4_STATE_NEW 0x00000002 /* inode is newly created */ +#define EXT4_STATE_XATTR 0x00000004 /* has in-inode xattrs */ +#define EXT4_STATE_NO_EXPAND 0x00000008 /* No space for expansion */ + +/* Used to pass group descriptor data when online resize is done */ +struct ext4_new_group_input { + __u32 group; /* Group number for this data */ + __u64 block_bitmap; /* Absolute block number of block bitmap */ + __u64 inode_bitmap; /* Absolute block number of inode bitmap */ + __u64 inode_table; /* Absolute block number of inode table start */ + __u32 blocks_count; /* Total number of blocks in this group */ + __u16 reserved_blocks; /* Number of reserved blocks in this group */ + __u16 unused; +}; + +/* The struct ext4_new_group_input in kernel space, with free_blocks_count */ +struct ext4_new_group_data { + __u32 group; + __u64 block_bitmap; + __u64 inode_bitmap; + __u64 inode_table; + __u32 blocks_count; + __u16 reserved_blocks; + __u16 unused; + __u32 free_blocks_count; +}; + +/* + * Following is used by preallocation code to tell get_blocks() that we + * want uninitialzed extents. + */ +#define EXT4_CREATE_UNINITIALIZED_EXT 2 + +/* + * ioctl commands + */ +#define EXT4_IOC_GETFLAGS FS_IOC_GETFLAGS +#define EXT4_IOC_SETFLAGS FS_IOC_SETFLAGS +#define EXT4_IOC_GETVERSION _IOR('f', 3, long) +#define EXT4_IOC_SETVERSION _IOW('f', 4, long) +#define EXT4_IOC_GROUP_EXTEND _IOW('f', 7, unsigned long) +#define EXT4_IOC_GROUP_ADD _IOW('f', 8,struct ext4_new_group_input) +#define EXT4_IOC_GETVERSION_OLD FS_IOC_GETVERSION +#define EXT4_IOC_SETVERSION_OLD FS_IOC_SETVERSION +#ifdef CONFIG_JBD2_DEBUG +#define EXT4_IOC_WAIT_FOR_READONLY _IOR('f', 99, long) +#endif +#define EXT4_IOC_GETRSVSZ _IOR('f', 5, long) +#define EXT4_IOC_SETRSVSZ _IOW('f', 6, long) +#define EXT4_IOC_MIGRATE _IO('f', 7) + +/* + * ioctl commands in 32 bit emulation + */ +#define EXT4_IOC32_GETFLAGS FS_IOC32_GETFLAGS +#define EXT4_IOC32_SETFLAGS FS_IOC32_SETFLAGS +#define EXT4_IOC32_GETVERSION _IOR('f', 3, int) +#define EXT4_IOC32_SETVERSION _IOW('f', 4, int) +#define EXT4_IOC32_GETRSVSZ _IOR('f', 5, int) +#define EXT4_IOC32_SETRSVSZ _IOW('f', 6, int) +#define EXT4_IOC32_GROUP_EXTEND _IOW('f', 7, unsigned int) +#ifdef CONFIG_JBD2_DEBUG +#define EXT4_IOC32_WAIT_FOR_READONLY _IOR('f', 99, int) +#endif +#define EXT4_IOC32_GETVERSION_OLD FS_IOC32_GETVERSION +#define EXT4_IOC32_SETVERSION_OLD FS_IOC32_SETVERSION + + +/* + * Mount options + */ +struct ext4_mount_options { + unsigned long s_mount_opt; + uid_t s_resuid; + gid_t s_resgid; + unsigned long s_commit_interval; +#ifdef CONFIG_QUOTA + int s_jquota_fmt; + char *s_qf_names[MAXQUOTAS]; +#endif +}; + +/* + * Structure of an inode on the disk + */ +struct ext4_inode { + __le16 i_mode; /* File mode */ + __le16 i_uid; /* Low 16 bits of Owner Uid */ + __le32 i_size_lo; /* Size in bytes */ + __le32 i_atime; /* Access time */ + __le32 i_ctime; /* Inode Change time */ + __le32 i_mtime; /* Modification time */ + __le32 i_dtime; /* Deletion Time */ + __le16 i_gid; /* Low 16 bits of Group Id */ + __le16 i_links_count; /* Links count */ + __le32 i_blocks_lo; /* Blocks count */ + __le32 i_flags; /* File flags */ + union { + struct { + __le32 l_i_version; + } linux1; + struct { + __u32 h_i_translator; + } hurd1; + struct { + __u32 m_i_reserved1; + } masix1; + } osd1; /* OS dependent 1 */ + __le32 i_block[EXT4_N_BLOCKS];/* Pointers to blocks */ + __le32 i_generation; /* File version (for NFS) */ + __le32 i_file_acl_lo; /* File ACL */ + __le32 i_size_high; + __le32 i_obso_faddr; /* Obsoleted fragment address */ + union { + struct { + __le16 l_i_blocks_high; /* were l_i_reserved1 */ + __le16 l_i_file_acl_high; + __le16 l_i_uid_high; /* these 2 fields */ + __le16 l_i_gid_high; /* were reserved2[0] */ + __u32 l_i_reserved2; + } linux2; + struct { + __le16 h_i_reserved1; /* Obsoleted fragment number/size which are removed in ext4 */ + __u16 h_i_mode_high; + __u16 h_i_uid_high; + __u16 h_i_gid_high; + __u32 h_i_author; + } hurd2; + struct { + __le16 h_i_reserved1; /* Obsoleted fragment number/size which are removed in ext4 */ + __le16 m_i_file_acl_high; + __u32 m_i_reserved2[2]; + } masix2; + } osd2; /* OS dependent 2 */ + __le16 i_extra_isize; + __le16 i_pad1; + __le32 i_ctime_extra; /* extra Change time (nsec << 2 | epoch) */ + __le32 i_mtime_extra; /* extra Modification time(nsec << 2 | epoch) */ + __le32 i_atime_extra; /* extra Access time (nsec << 2 | epoch) */ + __le32 i_crtime; /* File Creation time */ + __le32 i_crtime_extra; /* extra FileCreationtime (nsec << 2 | epoch) */ + __le32 i_version_hi; /* high 32 bits for 64-bit version */ +}; + + +#define EXT4_EPOCH_BITS 2 +#define EXT4_EPOCH_MASK ((1 << EXT4_EPOCH_BITS) - 1) +#define EXT4_NSEC_MASK (~0UL << EXT4_EPOCH_BITS) + +/* + * Extended fields will fit into an inode if the filesystem was formatted + * with large inodes (-I 256 or larger) and there are not currently any EAs + * consuming all of the available space. For new inodes we always reserve + * enough space for the kernel's known extended fields, but for inodes + * created with an old kernel this might not have been the case. None of + * the extended inode fields is critical for correct filesystem operation. + * This macro checks if a certain field fits in the inode. Note that + * inode-size = GOOD_OLD_INODE_SIZE + i_extra_isize + */ +#define EXT4_FITS_IN_INODE(ext4_inode, einode, field) \ + ((offsetof(typeof(*ext4_inode), field) + \ + sizeof((ext4_inode)->field)) \ + <= (EXT4_GOOD_OLD_INODE_SIZE + \ + (einode)->i_extra_isize)) \ + +static inline __le32 ext4_encode_extra_time(struct timespec *time) +{ + return cpu_to_le32((sizeof(time->tv_sec) > 4 ? + time->tv_sec >> 32 : 0) | + ((time->tv_nsec << 2) & EXT4_NSEC_MASK)); +} + +static inline void ext4_decode_extra_time(struct timespec *time, __le32 extra) +{ + if (sizeof(time->tv_sec) > 4) + time->tv_sec |= (__u64)(le32_to_cpu(extra) & EXT4_EPOCH_MASK) + << 32; + time->tv_nsec = (le32_to_cpu(extra) & EXT4_NSEC_MASK) >> 2; +} + +#define EXT4_INODE_SET_XTIME(xtime, inode, raw_inode) \ +do { \ + (raw_inode)->xtime = cpu_to_le32((inode)->xtime.tv_sec); \ + if (EXT4_FITS_IN_INODE(raw_inode, EXT4_I(inode), xtime ## _extra)) \ + (raw_inode)->xtime ## _extra = \ + ext4_encode_extra_time(&(inode)->xtime); \ +} while (0) + +#define EXT4_EINODE_SET_XTIME(xtime, einode, raw_inode) \ +do { \ + if (EXT4_FITS_IN_INODE(raw_inode, einode, xtime)) \ + (raw_inode)->xtime = cpu_to_le32((einode)->xtime.tv_sec); \ + if (EXT4_FITS_IN_INODE(raw_inode, einode, xtime ## _extra)) \ + (raw_inode)->xtime ## _extra = \ + ext4_encode_extra_time(&(einode)->xtime); \ +} while (0) + +#define EXT4_INODE_GET_XTIME(xtime, inode, raw_inode) \ +do { \ + (inode)->xtime.tv_sec = (signed)le32_to_cpu((raw_inode)->xtime); \ + if (EXT4_FITS_IN_INODE(raw_inode, EXT4_I(inode), xtime ## _extra)) \ + ext4_decode_extra_time(&(inode)->xtime, \ + raw_inode->xtime ## _extra); \ +} while (0) + +#define EXT4_EINODE_GET_XTIME(xtime, einode, raw_inode) \ +do { \ + if (EXT4_FITS_IN_INODE(raw_inode, einode, xtime)) \ + (einode)->xtime.tv_sec = \ + (signed)le32_to_cpu((raw_inode)->xtime); \ + if (EXT4_FITS_IN_INODE(raw_inode, einode, xtime ## _extra)) \ + ext4_decode_extra_time(&(einode)->xtime, \ + raw_inode->xtime ## _extra); \ +} while (0) + +#define i_disk_version osd1.linux1.l_i_version + +#if defined(__KERNEL__) || defined(__linux__) +#define i_reserved1 osd1.linux1.l_i_reserved1 +#define i_file_acl_high osd2.linux2.l_i_file_acl_high +#define i_blocks_high osd2.linux2.l_i_blocks_high +#define i_uid_low i_uid +#define i_gid_low i_gid +#define i_uid_high osd2.linux2.l_i_uid_high +#define i_gid_high osd2.linux2.l_i_gid_high +#define i_reserved2 osd2.linux2.l_i_reserved2 + +#elif defined(__GNU__) + +#define i_translator osd1.hurd1.h_i_translator +#define i_uid_high osd2.hurd2.h_i_uid_high +#define i_gid_high osd2.hurd2.h_i_gid_high +#define i_author osd2.hurd2.h_i_author + +#elif defined(__masix__) + +#define i_reserved1 osd1.masix1.m_i_reserved1 +#define i_file_acl_high osd2.masix2.m_i_file_acl_high +#define i_reserved2 osd2.masix2.m_i_reserved2 + +#endif /* defined(__KERNEL__) || defined(__linux__) */ + +/* + * File system states + */ +#define EXT4_VALID_FS 0x0001 /* Unmounted cleanly */ +#define EXT4_ERROR_FS 0x0002 /* Errors detected */ +#define EXT4_ORPHAN_FS 0x0004 /* Orphans being recovered */ + +/* + * Misc. filesystem flags + */ +#define EXT2_FLAGS_SIGNED_HASH 0x0001 /* Signed dirhash in use */ +#define EXT2_FLAGS_UNSIGNED_HASH 0x0002 /* Unsigned dirhash in use */ +#define EXT2_FLAGS_TEST_FILESYS 0x0004 /* to test development code */ + +/* + * Mount flags + */ +#define EXT4_MOUNT_CHECK 0x00001 /* Do mount-time checks */ +#define EXT4_MOUNT_OLDALLOC 0x00002 /* Don't use the new Orlov allocator */ +#define EXT4_MOUNT_GRPID 0x00004 /* Create files with directory's group */ +#define EXT4_MOUNT_DEBUG 0x00008 /* Some debugging messages */ +#define EXT4_MOUNT_ERRORS_CONT 0x00010 /* Continue on errors */ +#define EXT4_MOUNT_ERRORS_RO 0x00020 /* Remount fs ro on errors */ +#define EXT4_MOUNT_ERRORS_PANIC 0x00040 /* Panic on errors */ +#define EXT4_MOUNT_MINIX_DF 0x00080 /* Mimics the Minix statfs */ +#define EXT4_MOUNT_NOLOAD 0x00100 /* Don't use existing journal*/ +#define EXT4_MOUNT_ABORT 0x00200 /* Fatal error detected */ +#define EXT4_MOUNT_DATA_FLAGS 0x00C00 /* Mode for data writes: */ +#define EXT4_MOUNT_JOURNAL_DATA 0x00400 /* Write data to journal */ +#define EXT4_MOUNT_ORDERED_DATA 0x00800 /* Flush data before commit */ +#define EXT4_MOUNT_WRITEBACK_DATA 0x00C00 /* No data ordering */ +#define EXT4_MOUNT_UPDATE_JOURNAL 0x01000 /* Update the journal format */ +#define EXT4_MOUNT_NO_UID32 0x02000 /* Disable 32-bit UIDs */ +#define EXT4_MOUNT_XATTR_USER 0x04000 /* Extended user attributes */ +#define EXT4_MOUNT_POSIX_ACL 0x08000 /* POSIX Access Control Lists */ +#define EXT4_MOUNT_RESERVATION 0x10000 /* Preallocation */ +#define EXT4_MOUNT_BARRIER 0x20000 /* Use block barriers */ +#define EXT4_MOUNT_NOBH 0x40000 /* No bufferheads */ +#define EXT4_MOUNT_QUOTA 0x80000 /* Some quota option set */ +#define EXT4_MOUNT_USRQUOTA 0x100000 /* "old" user quota */ +#define EXT4_MOUNT_GRPQUOTA 0x200000 /* "old" group quota */ +#define EXT4_MOUNT_EXTENTS 0x400000 /* Extents support */ +#define EXT4_MOUNT_JOURNAL_CHECKSUM 0x800000 /* Journal checksums */ +#define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT 0x1000000 /* Journal Async Commit */ +#define EXT4_MOUNT_I_VERSION 0x2000000 /* i_version support */ +#define EXT4_MOUNT_MBALLOC 0x4000000 /* Buddy allocation support */ +/* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */ +#ifndef _LINUX_EXT2_FS_H +#define clear_opt(o, opt) o &= ~EXT4_MOUNT_##opt +#define set_opt(o, opt) o |= EXT4_MOUNT_##opt +#define test_opt(sb, opt) (EXT4_SB(sb)->s_mount_opt & \ + EXT4_MOUNT_##opt) +#else +#define EXT2_MOUNT_NOLOAD EXT4_MOUNT_NOLOAD +#define EXT2_MOUNT_ABORT EXT4_MOUNT_ABORT +#define EXT2_MOUNT_DATA_FLAGS EXT4_MOUNT_DATA_FLAGS +#endif + +#define ext4_set_bit ext2_set_bit +#define ext4_set_bit_atomic ext2_set_bit_atomic +#define ext4_clear_bit ext2_clear_bit +#define ext4_clear_bit_atomic ext2_clear_bit_atomic +#define ext4_test_bit ext2_test_bit +#define ext4_find_first_zero_bit ext2_find_first_zero_bit +#define ext4_find_next_zero_bit ext2_find_next_zero_bit +#define ext4_find_next_bit ext2_find_next_bit + +/* + * Maximal mount counts between two filesystem checks + */ +#define EXT4_DFL_MAX_MNT_COUNT 20 /* Allow 20 mounts */ +#define EXT4_DFL_CHECKINTERVAL 0 /* Don't use interval check */ + +/* + * Behaviour when detecting errors + */ +#define EXT4_ERRORS_CONTINUE 1 /* Continue execution */ +#define EXT4_ERRORS_RO 2 /* Remount fs read-only */ +#define EXT4_ERRORS_PANIC 3 /* Panic */ +#define EXT4_ERRORS_DEFAULT EXT4_ERRORS_CONTINUE + +/* + * Structure of the super block + */ +struct ext4_super_block { +/*00*/ __le32 s_inodes_count; /* Inodes count */ + __le32 s_blocks_count_lo; /* Blocks count */ + __le32 s_r_blocks_count_lo; /* Reserved blocks count */ + __le32 s_free_blocks_count_lo; /* Free blocks count */ +/*10*/ __le32 s_free_inodes_count; /* Free inodes count */ + __le32 s_first_data_block; /* First Data Block */ + __le32 s_log_block_size; /* Block size */ + __le32 s_obso_log_frag_size; /* Obsoleted fragment size */ +/*20*/ __le32 s_blocks_per_group; /* # Blocks per group */ + __le32 s_obso_frags_per_group; /* Obsoleted fragments per group */ + __le32 s_inodes_per_group; /* # Inodes per group */ + __le32 s_mtime; /* Mount time */ +/*30*/ __le32 s_wtime; /* Write time */ + __le16 s_mnt_count; /* Mount count */ + __le16 s_max_mnt_count; /* Maximal mount count */ + __le16 s_magic; /* Magic signature */ + __le16 s_state; /* File system state */ + __le16 s_errors; /* Behaviour when detecting errors */ + __le16 s_minor_rev_level; /* minor revision level */ +/*40*/ __le32 s_lastcheck; /* time of last check */ + __le32 s_checkinterval; /* max. time between checks */ + __le32 s_creator_os; /* OS */ + __le32 s_rev_level; /* Revision level */ +/*50*/ __le16 s_def_resuid; /* Default uid for reserved blocks */ + __le16 s_def_resgid; /* Default gid for reserved blocks */ + /* + * These fields are for EXT4_DYNAMIC_REV superblocks only. + * + * Note: the difference between the compatible feature set and + * the incompatible feature set is that if there is a bit set + * in the incompatible feature set that the kernel doesn't + * know about, it should refuse to mount the filesystem. + * + * e2fsck's requirements are more strict; if it doesn't know + * about a feature in either the compatible or incompatible + * feature set, it must abort and not try to meddle with + * things it doesn't understand... + */ + __le32 s_first_ino; /* First non-reserved inode */ + __le16 s_inode_size; /* size of inode structure */ + __le16 s_block_group_nr; /* block group # of this superblock */ + __le32 s_feature_compat; /* compatible feature set */ +/*60*/ __le32 s_feature_incompat; /* incompatible feature set */ + __le32 s_feature_ro_compat; /* readonly-compatible feature set */ +/*68*/ __u8 s_uuid[16]; /* 128-bit uuid for volume */ +/*78*/ char s_volume_name[16]; /* volume name */ +/*88*/ char s_last_mounted[64]; /* directory where last mounted */ +/*C8*/ __le32 s_algorithm_usage_bitmap; /* For compression */ + /* + * Performance hints. Directory preallocation should only + * happen if the EXT4_FEATURE_COMPAT_DIR_PREALLOC flag is on. + */ + __u8 s_prealloc_blocks; /* Nr of blocks to try to preallocate*/ + __u8 s_prealloc_dir_blocks; /* Nr to preallocate for dirs */ + __le16 s_reserved_gdt_blocks; /* Per group desc for online growth */ + /* + * Journaling support valid if EXT4_FEATURE_COMPAT_HAS_JOURNAL set. + */ +/*D0*/ __u8 s_journal_uuid[16]; /* uuid of journal superblock */ +/*E0*/ __le32 s_journal_inum; /* inode number of journal file */ + __le32 s_journal_dev; /* device number of journal file */ + __le32 s_last_orphan; /* start of list of inodes to delete */ + __le32 s_hash_seed[4]; /* HTREE hash seed */ + __u8 s_def_hash_version; /* Default hash version to use */ + __u8 s_reserved_char_pad; + __le16 s_desc_size; /* size of group descriptor */ +/*100*/ __le32 s_default_mount_opts; + __le32 s_first_meta_bg; /* First metablock block group */ + __le32 s_mkfs_time; /* When the filesystem was created */ + __le32 s_jnl_blocks[17]; /* Backup of the journal inode */ + /* 64bit support valid if EXT4_FEATURE_COMPAT_64BIT */ +/*150*/ __le32 s_blocks_count_hi; /* Blocks count */ + __le32 s_r_blocks_count_hi; /* Reserved blocks count */ + __le32 s_free_blocks_count_hi; /* Free blocks count */ + __le16 s_min_extra_isize; /* All inodes have at least # bytes */ + __le16 s_want_extra_isize; /* New inodes should reserve # bytes */ + __le32 s_flags; /* Miscellaneous flags */ + __le16 s_raid_stride; /* RAID stride */ + __le16 s_mmp_interval; /* # seconds to wait in MMP checking */ + __le64 s_mmp_block; /* Block for multi-mount protection */ + __le32 s_raid_stripe_width; /* blocks on all data disks (N*stride)*/ + __u32 s_reserved[163]; /* Padding to the end of the block */ +}; + +#ifdef __KERNEL__ +static inline struct ext4_sb_info * EXT4_SB(struct super_block *sb) +{ + return sb->s_fs_info; +} +static inline struct ext4_inode_info *EXT4_I(struct inode *inode) +{ + return container_of(inode, struct ext4_inode_info, vfs_inode); +} + +static inline struct timespec ext4_current_time(struct inode *inode) +{ + return (inode->i_sb->s_time_gran < NSEC_PER_SEC) ? + current_fs_time(inode->i_sb) : CURRENT_TIME_SEC; +} + + +static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino) +{ + return ino == EXT4_ROOT_INO || + ino == EXT4_JOURNAL_INO || + ino == EXT4_RESIZE_INO || + (ino >= EXT4_FIRST_INO(sb) && + ino <= le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count)); +} +#else +/* Assume that user mode programs are passing in an ext4fs superblock, not + * a kernel struct super_block. This will allow us to call the feature-test + * macros from user land. */ +#define EXT4_SB(sb) (sb) +#endif + +#define NEXT_ORPHAN(inode) EXT4_I(inode)->i_dtime + +/* + * Codes for operating systems + */ +#define EXT4_OS_LINUX 0 +#define EXT4_OS_HURD 1 +#define EXT4_OS_MASIX 2 +#define EXT4_OS_FREEBSD 3 +#define EXT4_OS_LITES 4 + +/* + * Revision levels + */ +#define EXT4_GOOD_OLD_REV 0 /* The good old (original) format */ +#define EXT4_DYNAMIC_REV 1 /* V2 format w/ dynamic inode sizes */ + +#define EXT4_CURRENT_REV EXT4_GOOD_OLD_REV +#define EXT4_MAX_SUPP_REV EXT4_DYNAMIC_REV + +#define EXT4_GOOD_OLD_INODE_SIZE 128 + +/* + * Feature set definitions + */ + +#define EXT4_HAS_COMPAT_FEATURE(sb,mask) \ + ( EXT4_SB(sb)->s_es->s_feature_compat & cpu_to_le32(mask) ) +#define EXT4_HAS_RO_COMPAT_FEATURE(sb,mask) \ + ( EXT4_SB(sb)->s_es->s_feature_ro_compat & cpu_to_le32(mask) ) +#define EXT4_HAS_INCOMPAT_FEATURE(sb,mask) \ + ( EXT4_SB(sb)->s_es->s_feature_incompat & cpu_to_le32(mask) ) +#define EXT4_SET_COMPAT_FEATURE(sb,mask) \ + EXT4_SB(sb)->s_es->s_feature_compat |= cpu_to_le32(mask) +#define EXT4_SET_RO_COMPAT_FEATURE(sb,mask) \ + EXT4_SB(sb)->s_es->s_feature_ro_compat |= cpu_to_le32(mask) +#define EXT4_SET_INCOMPAT_FEATURE(sb,mask) \ + EXT4_SB(sb)->s_es->s_feature_incompat |= cpu_to_le32(mask) +#define EXT4_CLEAR_COMPAT_FEATURE(sb,mask) \ + EXT4_SB(sb)->s_es->s_feature_compat &= ~cpu_to_le32(mask) +#define EXT4_CLEAR_RO_COMPAT_FEATURE(sb,mask) \ + EXT4_SB(sb)->s_es->s_feature_ro_compat &= ~cpu_to_le32(mask) +#define EXT4_CLEAR_INCOMPAT_FEATURE(sb,mask) \ + EXT4_SB(sb)->s_es->s_feature_incompat &= ~cpu_to_le32(mask) + +#define EXT4_FEATURE_COMPAT_DIR_PREALLOC 0x0001 +#define EXT4_FEATURE_COMPAT_IMAGIC_INODES 0x0002 +#define EXT4_FEATURE_COMPAT_HAS_JOURNAL 0x0004 +#define EXT4_FEATURE_COMPAT_EXT_ATTR 0x0008 +#define EXT4_FEATURE_COMPAT_RESIZE_INODE 0x0010 +#define EXT4_FEATURE_COMPAT_DIR_INDEX 0x0020 + +#define EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER 0x0001 +#define EXT4_FEATURE_RO_COMPAT_LARGE_FILE 0x0002 +#define EXT4_FEATURE_RO_COMPAT_BTREE_DIR 0x0004 +#define EXT4_FEATURE_RO_COMPAT_HUGE_FILE 0x0008 +#define EXT4_FEATURE_RO_COMPAT_GDT_CSUM 0x0010 +#define EXT4_FEATURE_RO_COMPAT_DIR_NLINK 0x0020 +#define EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE 0x0040 + +#define EXT4_FEATURE_INCOMPAT_COMPRESSION 0x0001 +#define EXT4_FEATURE_INCOMPAT_FILETYPE 0x0002 +#define EXT4_FEATURE_INCOMPAT_RECOVER 0x0004 /* Needs recovery */ +#define EXT4_FEATURE_INCOMPAT_JOURNAL_DEV 0x0008 /* Journal device */ +#define EXT4_FEATURE_INCOMPAT_META_BG 0x0010 +#define EXT4_FEATURE_INCOMPAT_EXTENTS 0x0040 /* extents support */ +#define EXT4_FEATURE_INCOMPAT_64BIT 0x0080 +#define EXT4_FEATURE_INCOMPAT_MMP 0x0100 +#define EXT4_FEATURE_INCOMPAT_FLEX_BG 0x0200 + +#define EXT4_FEATURE_COMPAT_SUPP EXT2_FEATURE_COMPAT_EXT_ATTR +#define EXT4_FEATURE_INCOMPAT_SUPP (EXT4_FEATURE_INCOMPAT_FILETYPE| \ + EXT4_FEATURE_INCOMPAT_RECOVER| \ + EXT4_FEATURE_INCOMPAT_META_BG| \ + EXT4_FEATURE_INCOMPAT_EXTENTS| \ + EXT4_FEATURE_INCOMPAT_64BIT| \ + EXT4_FEATURE_INCOMPAT_FLEX_BG) +#define EXT4_FEATURE_RO_COMPAT_SUPP (EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER| \ + EXT4_FEATURE_RO_COMPAT_LARGE_FILE| \ + EXT4_FEATURE_RO_COMPAT_GDT_CSUM| \ + EXT4_FEATURE_RO_COMPAT_DIR_NLINK | \ + EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE | \ + EXT4_FEATURE_RO_COMPAT_BTREE_DIR |\ + EXT4_FEATURE_RO_COMPAT_HUGE_FILE) + +/* + * Default values for user and/or group using reserved blocks + */ +#define EXT4_DEF_RESUID 0 +#define EXT4_DEF_RESGID 0 + +/* + * Default mount options + */ +#define EXT4_DEFM_DEBUG 0x0001 +#define EXT4_DEFM_BSDGROUPS 0x0002 +#define EXT4_DEFM_XATTR_USER 0x0004 +#define EXT4_DEFM_ACL 0x0008 +#define EXT4_DEFM_UID16 0x0010 +#define EXT4_DEFM_JMODE 0x0060 +#define EXT4_DEFM_JMODE_DATA 0x0020 +#define EXT4_DEFM_JMODE_ORDERED 0x0040 +#define EXT4_DEFM_JMODE_WBACK 0x0060 + +/* + * Structure of a directory entry + */ +#define EXT4_NAME_LEN 255 + +struct ext4_dir_entry { + __le32 inode; /* Inode number */ + __le16 rec_len; /* Directory entry length */ + __le16 name_len; /* Name length */ + char name[EXT4_NAME_LEN]; /* File name */ +}; + +/* + * The new version of the directory entry. Since EXT4 structures are + * stored in intel byte order, and the name_len field could never be + * bigger than 255 chars, it's safe to reclaim the extra byte for the + * file_type field. + */ +struct ext4_dir_entry_2 { + __le32 inode; /* Inode number */ + __le16 rec_len; /* Directory entry length */ + __u8 name_len; /* Name length */ + __u8 file_type; + char name[EXT4_NAME_LEN]; /* File name */ +}; + +/* + * Ext4 directory file types. Only the low 3 bits are used. The + * other bits are reserved for now. + */ +#define EXT4_FT_UNKNOWN 0 +#define EXT4_FT_REG_FILE 1 +#define EXT4_FT_DIR 2 +#define EXT4_FT_CHRDEV 3 +#define EXT4_FT_BLKDEV 4 +#define EXT4_FT_FIFO 5 +#define EXT4_FT_SOCK 6 +#define EXT4_FT_SYMLINK 7 + +#define EXT4_FT_MAX 8 + +/* + * EXT4_DIR_PAD defines the directory entries boundaries + * + * NOTE: It must be a multiple of 4 + */ +#define EXT4_DIR_PAD 4 +#define EXT4_DIR_ROUND (EXT4_DIR_PAD - 1) +#define EXT4_DIR_REC_LEN(name_len) (((name_len) + 8 + EXT4_DIR_ROUND) & \ + ~EXT4_DIR_ROUND) +#define EXT4_MAX_REC_LEN ((1<<16)-1) + +static inline unsigned ext4_rec_len_from_disk(__le16 dlen) +{ + unsigned len = le16_to_cpu(dlen); + + if (len == EXT4_MAX_REC_LEN) + return 1 << 16; + return len; +} + +static inline __le16 ext4_rec_len_to_disk(unsigned len) +{ + if (len == (1 << 16)) + return cpu_to_le16(EXT4_MAX_REC_LEN); + else if (len > (1 << 16)) + BUG(); + return cpu_to_le16(len); +} + +/* + * Hash Tree Directory indexing + * (c) Daniel Phillips, 2001 + */ + +#define is_dx(dir) (EXT4_HAS_COMPAT_FEATURE(dir->i_sb, \ + EXT4_FEATURE_COMPAT_DIR_INDEX) && \ + (EXT4_I(dir)->i_flags & EXT4_INDEX_FL)) +#define EXT4_DIR_LINK_MAX(dir) (!is_dx(dir) && (dir)->i_nlink >= EXT4_LINK_MAX) +#define EXT4_DIR_LINK_EMPTY(dir) ((dir)->i_nlink == 2 || (dir)->i_nlink == 1) + +/* Legal values for the dx_root hash_version field: */ + +#define DX_HASH_LEGACY 0 +#define DX_HASH_HALF_MD4 1 +#define DX_HASH_TEA 2 + +#ifdef __KERNEL__ + +/* hash info structure used by the directory hash */ +struct dx_hash_info +{ + u32 hash; + u32 minor_hash; + int hash_version; + u32 *seed; +}; + +#define EXT4_HTREE_EOF 0x7fffffff + +/* + * Control parameters used by ext4_htree_next_block + */ +#define HASH_NB_ALWAYS 1 + + +/* + * Describe an inode's exact location on disk and in memory + */ +struct ext4_iloc +{ + struct buffer_head *bh; + unsigned long offset; + ext4_group_t block_group; +}; + +static inline struct ext4_inode *ext4_raw_inode(struct ext4_iloc *iloc) +{ + return (struct ext4_inode *) (iloc->bh->b_data + iloc->offset); +} + +/* + * This structure is stuffed into the struct file's private_data field + * for directories. It is where we put information so that we can do + * readdir operations in hash tree order. + */ +struct dir_private_info { + struct rb_root root; + struct rb_node *curr_node; + struct fname *extra_fname; + loff_t last_pos; + __u32 curr_hash; + __u32 curr_minor_hash; + __u32 next_hash; +}; + +/* calculate the first block number of the group */ +static inline ext4_fsblk_t +ext4_group_first_block_no(struct super_block *sb, ext4_group_t group_no) +{ + return group_no * (ext4_fsblk_t)EXT4_BLOCKS_PER_GROUP(sb) + + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); +} + +/* + * Special error return code only used by dx_probe() and its callers. + */ +#define ERR_BAD_DX_DIR -75000 + +void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr, + unsigned long *blockgrpp, ext4_grpblk_t *offsetp); + +/* + * Function prototypes + */ + +/* + * Ok, these declarations are also in but none of the + * ext4 source programs needs to include it so they are duplicated here. + */ +# define NORET_TYPE /**/ +# define ATTRIB_NORET __attribute__((noreturn)) +# define NORET_AND noreturn, + +/* balloc.c */ +extern unsigned int ext4_block_group(struct super_block *sb, + ext4_fsblk_t blocknr); +extern ext4_grpblk_t ext4_block_group_offset(struct super_block *sb, + ext4_fsblk_t blocknr); +extern int ext4_bg_has_super(struct super_block *sb, ext4_group_t group); +extern unsigned long ext4_bg_num_gdb(struct super_block *sb, + ext4_group_t group); +extern ext4_fsblk_t ext4_new_block (handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, int *errp); +extern ext4_fsblk_t ext4_new_blocks (handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, unsigned long *count, int *errp); +extern ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, unsigned long *count, int *errp); +extern void ext4_free_blocks (handle_t *handle, struct inode *inode, + ext4_fsblk_t block, unsigned long count, int metadata); +extern void ext4_free_blocks_sb (handle_t *handle, struct super_block *sb, + ext4_fsblk_t block, unsigned long count, + unsigned long *pdquot_freed_blocks); +extern ext4_fsblk_t ext4_count_free_blocks (struct super_block *); +extern void ext4_check_blocks_bitmap (struct super_block *); +extern struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, + ext4_group_t block_group, + struct buffer_head ** bh); +extern int ext4_should_retry_alloc(struct super_block *sb, int *retries); +extern void ext4_init_block_alloc_info(struct inode *); +extern void ext4_rsv_window_add(struct super_block *sb, struct ext4_reserve_window_node *rsv); + +/* dir.c */ +extern int ext4_check_dir_entry(const char *, struct inode *, + struct ext4_dir_entry_2 *, + struct buffer_head *, unsigned long); +extern int ext4_htree_store_dirent(struct file *dir_file, __u32 hash, + __u32 minor_hash, + struct ext4_dir_entry_2 *dirent); +extern void ext4_htree_free_dir_info(struct dir_private_info *p); + +/* fsync.c */ +extern int ext4_sync_file (struct file *, struct dentry *, int); + +/* hash.c */ +extern int ext4fs_dirhash(const char *name, int len, struct + dx_hash_info *hinfo); + +/* ialloc.c */ +extern struct inode * ext4_new_inode (handle_t *, struct inode *, int); +extern void ext4_free_inode (handle_t *, struct inode *); +extern struct inode * ext4_orphan_get (struct super_block *, unsigned long); +extern unsigned long ext4_count_free_inodes (struct super_block *); +extern unsigned long ext4_count_dirs (struct super_block *); +extern void ext4_check_inodes_bitmap (struct super_block *); +extern unsigned long ext4_count_free (struct buffer_head *, unsigned); + +/* mballoc.c */ +extern long ext4_mb_stats; +extern long ext4_mb_max_to_scan; +extern int ext4_mb_init(struct super_block *, int); +extern int ext4_mb_release(struct super_block *); +extern ext4_fsblk_t ext4_mb_new_blocks(handle_t *, + struct ext4_allocation_request *, int *); +extern int ext4_mb_reserve_blocks(struct super_block *, int); +extern void ext4_mb_discard_inode_preallocations(struct inode *); +extern int __init init_ext4_mballoc(void); +extern void exit_ext4_mballoc(void); +extern void ext4_mb_free_blocks(handle_t *, struct inode *, + unsigned long, unsigned long, int, unsigned long *); + + +/* inode.c */ +int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode, + struct buffer_head *bh, ext4_fsblk_t blocknr); +struct buffer_head *ext4_getblk(handle_t *, struct inode *, + ext4_lblk_t, int, int *); +struct buffer_head *ext4_bread(handle_t *, struct inode *, + ext4_lblk_t, int, int *); +int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, + ext4_lblk_t iblock, unsigned long maxblocks, + struct buffer_head *bh_result, + int create, int extend_disksize); + +extern struct inode *ext4_iget(struct super_block *, unsigned long); +extern int ext4_write_inode (struct inode *, int); +extern int ext4_setattr (struct dentry *, struct iattr *); +extern void ext4_delete_inode (struct inode *); +extern int ext4_sync_inode (handle_t *, struct inode *); +extern void ext4_discard_reservation (struct inode *); +extern void ext4_dirty_inode(struct inode *); +extern int ext4_change_inode_journal_flag(struct inode *, int); +extern int ext4_get_inode_loc(struct inode *, struct ext4_iloc *); +extern void ext4_truncate (struct inode *); +extern void ext4_set_inode_flags(struct inode *); +extern void ext4_get_inode_flags(struct ext4_inode_info *); +extern void ext4_set_aops(struct inode *inode); +extern int ext4_writepage_trans_blocks(struct inode *); +extern int ext4_block_truncate_page(handle_t *handle, struct page *page, + struct address_space *mapping, loff_t from); + +/* ioctl.c */ +extern long ext4_ioctl(struct file *, unsigned int, unsigned long); +extern long ext4_compat_ioctl (struct file *, unsigned int, unsigned long); + +/* migrate.c */ +extern int ext4_ext_migrate(struct inode *, struct file *, unsigned int, + unsigned long); +/* namei.c */ +extern int ext4_orphan_add(handle_t *, struct inode *); +extern int ext4_orphan_del(handle_t *, struct inode *); +extern int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash, + __u32 start_minor_hash, __u32 *next_hash); + +/* resize.c */ +extern int ext4_group_add(struct super_block *sb, + struct ext4_new_group_data *input); +extern int ext4_group_extend(struct super_block *sb, + struct ext4_super_block *es, + ext4_fsblk_t n_blocks_count); + +/* super.c */ +extern void ext4_error (struct super_block *, const char *, const char *, ...) + __attribute__ ((format (printf, 3, 4))); +extern void __ext4_std_error (struct super_block *, const char *, int); +extern void ext4_abort (struct super_block *, const char *, const char *, ...) + __attribute__ ((format (printf, 3, 4))); +extern void ext4_warning (struct super_block *, const char *, const char *, ...) + __attribute__ ((format (printf, 3, 4))); +extern void ext4_update_dynamic_rev (struct super_block *sb); +extern int ext4_update_compat_feature(handle_t *handle, struct super_block *sb, + __u32 compat); +extern int ext4_update_rocompat_feature(handle_t *handle, + struct super_block *sb, __u32 rocompat); +extern int ext4_update_incompat_feature(handle_t *handle, + struct super_block *sb, __u32 incompat); +extern ext4_fsblk_t ext4_block_bitmap(struct super_block *sb, + struct ext4_group_desc *bg); +extern ext4_fsblk_t ext4_inode_bitmap(struct super_block *sb, + struct ext4_group_desc *bg); +extern ext4_fsblk_t ext4_inode_table(struct super_block *sb, + struct ext4_group_desc *bg); +extern void ext4_block_bitmap_set(struct super_block *sb, + struct ext4_group_desc *bg, ext4_fsblk_t blk); +extern void ext4_inode_bitmap_set(struct super_block *sb, + struct ext4_group_desc *bg, ext4_fsblk_t blk); +extern void ext4_inode_table_set(struct super_block *sb, + struct ext4_group_desc *bg, ext4_fsblk_t blk); + +static inline ext4_fsblk_t ext4_blocks_count(struct ext4_super_block *es) +{ + return ((ext4_fsblk_t)le32_to_cpu(es->s_blocks_count_hi) << 32) | + le32_to_cpu(es->s_blocks_count_lo); +} + +static inline ext4_fsblk_t ext4_r_blocks_count(struct ext4_super_block *es) +{ + return ((ext4_fsblk_t)le32_to_cpu(es->s_r_blocks_count_hi) << 32) | + le32_to_cpu(es->s_r_blocks_count_lo); +} + +static inline ext4_fsblk_t ext4_free_blocks_count(struct ext4_super_block *es) +{ + return ((ext4_fsblk_t)le32_to_cpu(es->s_free_blocks_count_hi) << 32) | + le32_to_cpu(es->s_free_blocks_count_lo); +} + +static inline void ext4_blocks_count_set(struct ext4_super_block *es, + ext4_fsblk_t blk) +{ + es->s_blocks_count_lo = cpu_to_le32((u32)blk); + es->s_blocks_count_hi = cpu_to_le32(blk >> 32); +} + +static inline void ext4_free_blocks_count_set(struct ext4_super_block *es, + ext4_fsblk_t blk) +{ + es->s_free_blocks_count_lo = cpu_to_le32((u32)blk); + es->s_free_blocks_count_hi = cpu_to_le32(blk >> 32); +} + +static inline void ext4_r_blocks_count_set(struct ext4_super_block *es, + ext4_fsblk_t blk) +{ + es->s_r_blocks_count_lo = cpu_to_le32((u32)blk); + es->s_r_blocks_count_hi = cpu_to_le32(blk >> 32); +} + +static inline loff_t ext4_isize(struct ext4_inode *raw_inode) +{ + return ((loff_t)le32_to_cpu(raw_inode->i_size_high) << 32) | + le32_to_cpu(raw_inode->i_size_lo); +} + +static inline void ext4_isize_set(struct ext4_inode *raw_inode, loff_t i_size) +{ + raw_inode->i_size_lo = cpu_to_le32(i_size); + raw_inode->i_size_high = cpu_to_le32(i_size >> 32); +} + +static inline +struct ext4_group_info *ext4_get_group_info(struct super_block *sb, + ext4_group_t group) +{ + struct ext4_group_info ***grp_info; + long indexv, indexh; + grp_info = EXT4_SB(sb)->s_group_info; + indexv = group >> (EXT4_DESC_PER_BLOCK_BITS(sb)); + indexh = group & ((EXT4_DESC_PER_BLOCK(sb)) - 1); + return grp_info[indexv][indexh]; +} + + +#define ext4_std_error(sb, errno) \ +do { \ + if ((errno)) \ + __ext4_std_error((sb), __FUNCTION__, (errno)); \ +} while (0) + +/* + * Inodes and files operations + */ + +/* dir.c */ +extern const struct file_operations ext4_dir_operations; + +/* file.c */ +extern const struct inode_operations ext4_file_inode_operations; +extern const struct file_operations ext4_file_operations; + +/* namei.c */ +extern const struct inode_operations ext4_dir_inode_operations; +extern const struct inode_operations ext4_special_inode_operations; + +/* symlink.c */ +extern const struct inode_operations ext4_symlink_inode_operations; +extern const struct inode_operations ext4_fast_symlink_inode_operations; + +/* extents.c */ +extern int ext4_ext_tree_init(handle_t *handle, struct inode *); +extern int ext4_ext_writepage_trans_blocks(struct inode *, int); +extern int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, + ext4_lblk_t iblock, + unsigned long max_blocks, struct buffer_head *bh_result, + int create, int extend_disksize); +extern void ext4_ext_truncate(struct inode *, struct page *); +extern void ext4_ext_init(struct super_block *); +extern void ext4_ext_release(struct super_block *); +extern long ext4_fallocate(struct inode *inode, int mode, loff_t offset, + loff_t len); +extern int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, + sector_t block, unsigned long max_blocks, + struct buffer_head *bh, int create, + int extend_disksize); +#endif /* __KERNEL__ */ + +#endif /* _EXT4_H */ diff --git a/fs/ext4/ext4_extents.h b/fs/ext4/ext4_extents.h new file mode 100644 index 000000000000..75333b595fab --- /dev/null +++ b/fs/ext4/ext4_extents.h @@ -0,0 +1,232 @@ +/* + * Copyright (c) 2003-2006, Cluster File Systems, Inc, info@clusterfs.com + * Written by Alex Tomas + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public Licens + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111- + */ + +#ifndef _EXT4_EXTENTS +#define _EXT4_EXTENTS + +#include "ext4.h" + +/* + * With AGGRESSIVE_TEST defined, the capacity of index/leaf blocks + * becomes very small, so index split, in-depth growing and + * other hard changes happen much more often. + * This is for debug purposes only. + */ +#define AGGRESSIVE_TEST_ + +/* + * With EXTENTS_STATS defined, the number of blocks and extents + * are collected in the truncate path. They'll be shown at + * umount time. + */ +#define EXTENTS_STATS__ + +/* + * If CHECK_BINSEARCH is defined, then the results of the binary search + * will also be checked by linear search. + */ +#define CHECK_BINSEARCH__ + +/* + * If EXT_DEBUG is defined you can use the 'extdebug' mount option + * to get lots of info about what's going on. + */ +#define EXT_DEBUG__ +#ifdef EXT_DEBUG +#define ext_debug(a...) printk(a) +#else +#define ext_debug(a...) +#endif + +/* + * If EXT_STATS is defined then stats numbers are collected. + * These number will be displayed at umount time. + */ +#define EXT_STATS_ + + +/* + * ext4_inode has i_block array (60 bytes total). + * The first 12 bytes store ext4_extent_header; + * the remainder stores an array of ext4_extent. + */ + +/* + * This is the extent on-disk structure. + * It's used at the bottom of the tree. + */ +struct ext4_extent { + __le32 ee_block; /* first logical block extent covers */ + __le16 ee_len; /* number of blocks covered by extent */ + __le16 ee_start_hi; /* high 16 bits of physical block */ + __le32 ee_start_lo; /* low 32 bits of physical block */ +}; + +/* + * This is index on-disk structure. + * It's used at all the levels except the bottom. + */ +struct ext4_extent_idx { + __le32 ei_block; /* index covers logical blocks from 'block' */ + __le32 ei_leaf_lo; /* pointer to the physical block of the next * + * level. leaf or next index could be there */ + __le16 ei_leaf_hi; /* high 16 bits of physical block */ + __u16 ei_unused; +}; + +/* + * Each block (leaves and indexes), even inode-stored has header. + */ +struct ext4_extent_header { + __le16 eh_magic; /* probably will support different formats */ + __le16 eh_entries; /* number of valid entries */ + __le16 eh_max; /* capacity of store in entries */ + __le16 eh_depth; /* has tree real underlying blocks? */ + __le32 eh_generation; /* generation of the tree */ +}; + +#define EXT4_EXT_MAGIC cpu_to_le16(0xf30a) + +/* + * Array of ext4_ext_path contains path to some extent. + * Creation/lookup routines use it for traversal/splitting/etc. + * Truncate uses it to simulate recursive walking. + */ +struct ext4_ext_path { + ext4_fsblk_t p_block; + __u16 p_depth; + struct ext4_extent *p_ext; + struct ext4_extent_idx *p_idx; + struct ext4_extent_header *p_hdr; + struct buffer_head *p_bh; +}; + +/* + * structure for external API + */ + +#define EXT4_EXT_CACHE_NO 0 +#define EXT4_EXT_CACHE_GAP 1 +#define EXT4_EXT_CACHE_EXTENT 2 + + +#define EXT_MAX_BLOCK 0xffffffff + +/* + * EXT_INIT_MAX_LEN is the maximum number of blocks we can have in an + * initialized extent. This is 2^15 and not (2^16 - 1), since we use the + * MSB of ee_len field in the extent datastructure to signify if this + * particular extent is an initialized extent or an uninitialized (i.e. + * preallocated). + * EXT_UNINIT_MAX_LEN is the maximum number of blocks we can have in an + * uninitialized extent. + * If ee_len is <= 0x8000, it is an initialized extent. Otherwise, it is an + * uninitialized one. In other words, if MSB of ee_len is set, it is an + * uninitialized extent with only one special scenario when ee_len = 0x8000. + * In this case we can not have an uninitialized extent of zero length and + * thus we make it as a special case of initialized extent with 0x8000 length. + * This way we get better extent-to-group alignment for initialized extents. + * Hence, the maximum number of blocks we can have in an *initialized* + * extent is 2^15 (32768) and in an *uninitialized* extent is 2^15-1 (32767). + */ +#define EXT_INIT_MAX_LEN (1UL << 15) +#define EXT_UNINIT_MAX_LEN (EXT_INIT_MAX_LEN - 1) + + +#define EXT_FIRST_EXTENT(__hdr__) \ + ((struct ext4_extent *) (((char *) (__hdr__)) + \ + sizeof(struct ext4_extent_header))) +#define EXT_FIRST_INDEX(__hdr__) \ + ((struct ext4_extent_idx *) (((char *) (__hdr__)) + \ + sizeof(struct ext4_extent_header))) +#define EXT_HAS_FREE_INDEX(__path__) \ + (le16_to_cpu((__path__)->p_hdr->eh_entries) \ + < le16_to_cpu((__path__)->p_hdr->eh_max)) +#define EXT_LAST_EXTENT(__hdr__) \ + (EXT_FIRST_EXTENT((__hdr__)) + le16_to_cpu((__hdr__)->eh_entries) - 1) +#define EXT_LAST_INDEX(__hdr__) \ + (EXT_FIRST_INDEX((__hdr__)) + le16_to_cpu((__hdr__)->eh_entries) - 1) +#define EXT_MAX_EXTENT(__hdr__) \ + (EXT_FIRST_EXTENT((__hdr__)) + le16_to_cpu((__hdr__)->eh_max) - 1) +#define EXT_MAX_INDEX(__hdr__) \ + (EXT_FIRST_INDEX((__hdr__)) + le16_to_cpu((__hdr__)->eh_max) - 1) + +static inline struct ext4_extent_header *ext_inode_hdr(struct inode *inode) +{ + return (struct ext4_extent_header *) EXT4_I(inode)->i_data; +} + +static inline struct ext4_extent_header *ext_block_hdr(struct buffer_head *bh) +{ + return (struct ext4_extent_header *) bh->b_data; +} + +static inline unsigned short ext_depth(struct inode *inode) +{ + return le16_to_cpu(ext_inode_hdr(inode)->eh_depth); +} + +static inline void ext4_ext_tree_changed(struct inode *inode) +{ + EXT4_I(inode)->i_ext_generation++; +} + +static inline void +ext4_ext_invalidate_cache(struct inode *inode) +{ + EXT4_I(inode)->i_cached_extent.ec_type = EXT4_EXT_CACHE_NO; +} + +static inline void ext4_ext_mark_uninitialized(struct ext4_extent *ext) +{ + /* We can not have an uninitialized extent of zero length! */ + BUG_ON((le16_to_cpu(ext->ee_len) & ~EXT_INIT_MAX_LEN) == 0); + ext->ee_len |= cpu_to_le16(EXT_INIT_MAX_LEN); +} + +static inline int ext4_ext_is_uninitialized(struct ext4_extent *ext) +{ + /* Extent with ee_len of 0x8000 is treated as an initialized extent */ + return (le16_to_cpu(ext->ee_len) > EXT_INIT_MAX_LEN); +} + +static inline int ext4_ext_get_actual_len(struct ext4_extent *ext) +{ + return (le16_to_cpu(ext->ee_len) <= EXT_INIT_MAX_LEN ? + le16_to_cpu(ext->ee_len) : + (le16_to_cpu(ext->ee_len) - EXT_INIT_MAX_LEN)); +} + +extern ext4_fsblk_t idx_pblock(struct ext4_extent_idx *); +extern void ext4_ext_store_pblock(struct ext4_extent *, ext4_fsblk_t); +extern int ext4_extent_tree_init(handle_t *, struct inode *); +extern int ext4_ext_calc_credits_for_insert(struct inode *, struct ext4_ext_path *); +extern int ext4_ext_try_to_merge(struct inode *inode, + struct ext4_ext_path *path, + struct ext4_extent *); +extern unsigned int ext4_ext_check_overlap(struct inode *, struct ext4_extent *, struct ext4_ext_path *); +extern int ext4_ext_insert_extent(handle_t *, struct inode *, struct ext4_ext_path *, struct ext4_extent *); +extern struct ext4_ext_path *ext4_ext_find_extent(struct inode *, ext4_lblk_t, + struct ext4_ext_path *); +extern int ext4_ext_search_left(struct inode *, struct ext4_ext_path *, + ext4_lblk_t *, ext4_fsblk_t *); +extern int ext4_ext_search_right(struct inode *, struct ext4_ext_path *, + ext4_lblk_t *, ext4_fsblk_t *); +extern void ext4_ext_drop_refs(struct ext4_ext_path *); +#endif /* _EXT4_EXTENTS */ + diff --git a/fs/ext4/ext4_i.h b/fs/ext4/ext4_i.h new file mode 100644 index 000000000000..26a4ae255d79 --- /dev/null +++ b/fs/ext4/ext4_i.h @@ -0,0 +1,167 @@ +/* + * ext4_i.h + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/include/linux/minix_fs_i.h + * + * Copyright (C) 1991, 1992 Linus Torvalds + */ + +#ifndef _EXT4_I +#define _EXT4_I + +#include +#include +#include +#include + +/* data type for block offset of block group */ +typedef int ext4_grpblk_t; + +/* data type for filesystem-wide blocks number */ +typedef unsigned long long ext4_fsblk_t; + +/* data type for file logical block number */ +typedef __u32 ext4_lblk_t; + +/* data type for block group number */ +typedef unsigned long ext4_group_t; + +struct ext4_reserve_window { + ext4_fsblk_t _rsv_start; /* First byte reserved */ + ext4_fsblk_t _rsv_end; /* Last byte reserved or 0 */ +}; + +struct ext4_reserve_window_node { + struct rb_node rsv_node; + __u32 rsv_goal_size; + __u32 rsv_alloc_hit; + struct ext4_reserve_window rsv_window; +}; + +struct ext4_block_alloc_info { + /* information about reservation window */ + struct ext4_reserve_window_node rsv_window_node; + /* + * was i_next_alloc_block in ext4_inode_info + * is the logical (file-relative) number of the + * most-recently-allocated block in this file. + * We use this for detecting linearly ascending allocation requests. + */ + ext4_lblk_t last_alloc_logical_block; + /* + * Was i_next_alloc_goal in ext4_inode_info + * is the *physical* companion to i_next_alloc_block. + * it the physical block number of the block which was most-recentl + * allocated to this file. This give us the goal (target) for the next + * allocation when we detect linearly ascending requests. + */ + ext4_fsblk_t last_alloc_physical_block; +}; + +#define rsv_start rsv_window._rsv_start +#define rsv_end rsv_window._rsv_end + +/* + * storage for cached extent + */ +struct ext4_ext_cache { + ext4_fsblk_t ec_start; + ext4_lblk_t ec_block; + __u32 ec_len; /* must be 32bit to return holes */ + __u32 ec_type; +}; + +/* + * third extended file system inode data in memory + */ +struct ext4_inode_info { + __le32 i_data[15]; /* unconverted */ + __u32 i_flags; + ext4_fsblk_t i_file_acl; + __u32 i_dtime; + + /* + * i_block_group is the number of the block group which contains + * this file's inode. Constant across the lifetime of the inode, + * it is ued for making block allocation decisions - we try to + * place a file's data blocks near its inode block, and new inodes + * near to their parent directory's inode. + */ + ext4_group_t i_block_group; + __u32 i_state; /* Dynamic state flags for ext4 */ + + /* block reservation info */ + struct ext4_block_alloc_info *i_block_alloc_info; + + ext4_lblk_t i_dir_start_lookup; +#ifdef CONFIG_EXT4DEV_FS_XATTR + /* + * Extended attributes can be read independently of the main file + * data. Taking i_mutex even when reading would cause contention + * between readers of EAs and writers of regular file data, so + * instead we synchronize on xattr_sem when reading or changing + * EAs. + */ + struct rw_semaphore xattr_sem; +#endif +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + struct posix_acl *i_acl; + struct posix_acl *i_default_acl; +#endif + + struct list_head i_orphan; /* unlinked but open inodes */ + + /* + * i_disksize keeps track of what the inode size is ON DISK, not + * in memory. During truncate, i_size is set to the new size by + * the VFS prior to calling ext4_truncate(), but the filesystem won't + * set i_disksize to 0 until the truncate is actually under way. + * + * The intent is that i_disksize always represents the blocks which + * are used by this file. This allows recovery to restart truncate + * on orphans if we crash during truncate. We actually write i_disksize + * into the on-disk inode when writing inodes out, instead of i_size. + * + * The only time when i_disksize and i_size may be different is when + * a truncate is in progress. The only things which change i_disksize + * are ext4_get_block (growth) and ext4_truncate (shrinkth). + */ + loff_t i_disksize; + + /* on-disk additional length */ + __u16 i_extra_isize; + + /* + * i_data_sem is for serialising ext4_truncate() against + * ext4_getblock(). In the 2.4 ext2 design, great chunks of inode's + * data tree are chopped off during truncate. We can't do that in + * ext4 because whenever we perform intermediate commits during + * truncate, the inode and all the metadata blocks *must* be in a + * consistent state which allows truncation of the orphans to restart + * during recovery. Hence we must fix the get_block-vs-truncate race + * by other means, so we have i_data_sem. + */ + struct rw_semaphore i_data_sem; + struct inode vfs_inode; + + unsigned long i_ext_generation; + struct ext4_ext_cache i_cached_extent; + /* + * File creation time. Its function is same as that of + * struct timespec i_{a,c,m}time in the generic inode. + */ + struct timespec i_crtime; + + /* mballoc */ + struct list_head i_prealloc_list; + spinlock_t i_prealloc_lock; +}; + +#endif /* _EXT4_I */ diff --git a/fs/ext4/ext4_jbd2.c b/fs/ext4/ext4_jbd2.c index 2e6007d418dc..c75384b34f2c 100644 --- a/fs/ext4/ext4_jbd2.c +++ b/fs/ext4/ext4_jbd2.c @@ -2,7 +2,7 @@ * Interface between ext4 and JBD */ -#include +#include "ext4_jbd2.h" int __ext4_journal_get_undo_access(const char *where, handle_t *handle, struct buffer_head *bh) diff --git a/fs/ext4/ext4_jbd2.h b/fs/ext4/ext4_jbd2.h new file mode 100644 index 000000000000..9255a7d28b24 --- /dev/null +++ b/fs/ext4/ext4_jbd2.h @@ -0,0 +1,231 @@ +/* + * ext4_jbd2.h + * + * Written by Stephen C. Tweedie , 1999 + * + * Copyright 1998--1999 Red Hat corp --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Ext4-specific journaling extensions. + */ + +#ifndef _EXT4_JBD2_H +#define _EXT4_JBD2_H + +#include +#include +#include "ext4.h" + +#define EXT4_JOURNAL(inode) (EXT4_SB((inode)->i_sb)->s_journal) + +/* Define the number of blocks we need to account to a transaction to + * modify one block of data. + * + * We may have to touch one inode, one bitmap buffer, up to three + * indirection blocks, the group and superblock summaries, and the data + * block to complete the transaction. + * + * For extents-enabled fs we may have to allocate and modify up to + * 5 levels of tree + root which are stored in the inode. */ + +#define EXT4_SINGLEDATA_TRANS_BLOCKS(sb) \ + (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS) \ + || test_opt(sb, EXTENTS) ? 27U : 8U) + +/* Extended attribute operations touch at most two data buffers, + * two bitmap buffers, and two group summaries, in addition to the inode + * and the superblock, which are already accounted for. */ + +#define EXT4_XATTR_TRANS_BLOCKS 6U + +/* Define the minimum size for a transaction which modifies data. This + * needs to take into account the fact that we may end up modifying two + * quota files too (one for the group, one for the user quota). The + * superblock only gets updated once, of course, so don't bother + * counting that again for the quota updates. */ + +#define EXT4_DATA_TRANS_BLOCKS(sb) (EXT4_SINGLEDATA_TRANS_BLOCKS(sb) + \ + EXT4_XATTR_TRANS_BLOCKS - 2 + \ + 2*EXT4_QUOTA_TRANS_BLOCKS(sb)) + +/* Delete operations potentially hit one directory's namespace plus an + * entire inode, plus arbitrary amounts of bitmap/indirection data. Be + * generous. We can grow the delete transaction later if necessary. */ + +#define EXT4_DELETE_TRANS_BLOCKS(sb) (2 * EXT4_DATA_TRANS_BLOCKS(sb) + 64) + +/* Define an arbitrary limit for the amount of data we will anticipate + * writing to any given transaction. For unbounded transactions such as + * write(2) and truncate(2) we can write more than this, but we always + * start off at the maximum transaction size and grow the transaction + * optimistically as we go. */ + +#define EXT4_MAX_TRANS_DATA 64U + +/* We break up a large truncate or write transaction once the handle's + * buffer credits gets this low, we need either to extend the + * transaction or to start a new one. Reserve enough space here for + * inode, bitmap, superblock, group and indirection updates for at least + * one block, plus two quota updates. Quota allocations are not + * needed. */ + +#define EXT4_RESERVE_TRANS_BLOCKS 12U + +#define EXT4_INDEX_EXTRA_TRANS_BLOCKS 8 + +#ifdef CONFIG_QUOTA +/* Amount of blocks needed for quota update - we know that the structure was + * allocated so we need to update only inode+data */ +#define EXT4_QUOTA_TRANS_BLOCKS(sb) (test_opt(sb, QUOTA) ? 2 : 0) +/* Amount of blocks needed for quota insert/delete - we do some block writes + * but inode, sb and group updates are done only once */ +#define EXT4_QUOTA_INIT_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_INIT_ALLOC*\ + (EXT4_SINGLEDATA_TRANS_BLOCKS(sb)-3)+3+DQUOT_INIT_REWRITE) : 0) +#define EXT4_QUOTA_DEL_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_DEL_ALLOC*\ + (EXT4_SINGLEDATA_TRANS_BLOCKS(sb)-3)+3+DQUOT_DEL_REWRITE) : 0) +#else +#define EXT4_QUOTA_TRANS_BLOCKS(sb) 0 +#define EXT4_QUOTA_INIT_BLOCKS(sb) 0 +#define EXT4_QUOTA_DEL_BLOCKS(sb) 0 +#endif + +int +ext4_mark_iloc_dirty(handle_t *handle, + struct inode *inode, + struct ext4_iloc *iloc); + +/* + * On success, We end up with an outstanding reference count against + * iloc->bh. This _must_ be cleaned up later. + */ + +int ext4_reserve_inode_write(handle_t *handle, struct inode *inode, + struct ext4_iloc *iloc); + +int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode); + +/* + * Wrapper functions with which ext4 calls into JBD. The intent here is + * to allow these to be turned into appropriate stubs so ext4 can control + * ext2 filesystems, so ext2+ext4 systems only nee one fs. This work hasn't + * been done yet. + */ + +static inline void ext4_journal_release_buffer(handle_t *handle, + struct buffer_head *bh) +{ + jbd2_journal_release_buffer(handle, bh); +} + +void ext4_journal_abort_handle(const char *caller, const char *err_fn, + struct buffer_head *bh, handle_t *handle, int err); + +int __ext4_journal_get_undo_access(const char *where, handle_t *handle, + struct buffer_head *bh); + +int __ext4_journal_get_write_access(const char *where, handle_t *handle, + struct buffer_head *bh); + +int __ext4_journal_forget(const char *where, handle_t *handle, + struct buffer_head *bh); + +int __ext4_journal_revoke(const char *where, handle_t *handle, + ext4_fsblk_t blocknr, struct buffer_head *bh); + +int __ext4_journal_get_create_access(const char *where, + handle_t *handle, struct buffer_head *bh); + +int __ext4_journal_dirty_metadata(const char *where, + handle_t *handle, struct buffer_head *bh); + +#define ext4_journal_get_undo_access(handle, bh) \ + __ext4_journal_get_undo_access(__FUNCTION__, (handle), (bh)) +#define ext4_journal_get_write_access(handle, bh) \ + __ext4_journal_get_write_access(__FUNCTION__, (handle), (bh)) +#define ext4_journal_revoke(handle, blocknr, bh) \ + __ext4_journal_revoke(__FUNCTION__, (handle), (blocknr), (bh)) +#define ext4_journal_get_create_access(handle, bh) \ + __ext4_journal_get_create_access(__FUNCTION__, (handle), (bh)) +#define ext4_journal_dirty_metadata(handle, bh) \ + __ext4_journal_dirty_metadata(__FUNCTION__, (handle), (bh)) +#define ext4_journal_forget(handle, bh) \ + __ext4_journal_forget(__FUNCTION__, (handle), (bh)) + +int ext4_journal_dirty_data(handle_t *handle, struct buffer_head *bh); + +handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks); +int __ext4_journal_stop(const char *where, handle_t *handle); + +static inline handle_t *ext4_journal_start(struct inode *inode, int nblocks) +{ + return ext4_journal_start_sb(inode->i_sb, nblocks); +} + +#define ext4_journal_stop(handle) \ + __ext4_journal_stop(__FUNCTION__, (handle)) + +static inline handle_t *ext4_journal_current_handle(void) +{ + return journal_current_handle(); +} + +static inline int ext4_journal_extend(handle_t *handle, int nblocks) +{ + return jbd2_journal_extend(handle, nblocks); +} + +static inline int ext4_journal_restart(handle_t *handle, int nblocks) +{ + return jbd2_journal_restart(handle, nblocks); +} + +static inline int ext4_journal_blocks_per_page(struct inode *inode) +{ + return jbd2_journal_blocks_per_page(inode); +} + +static inline int ext4_journal_force_commit(journal_t *journal) +{ + return jbd2_journal_force_commit(journal); +} + +/* super.c */ +int ext4_force_commit(struct super_block *sb); + +static inline int ext4_should_journal_data(struct inode *inode) +{ + if (!S_ISREG(inode->i_mode)) + return 1; + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) + return 1; + if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) + return 1; + return 0; +} + +static inline int ext4_should_order_data(struct inode *inode) +{ + if (!S_ISREG(inode->i_mode)) + return 0; + if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) + return 0; + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA) + return 1; + return 0; +} + +static inline int ext4_should_writeback_data(struct inode *inode) +{ + if (!S_ISREG(inode->i_mode)) + return 0; + if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) + return 0; + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA) + return 1; + return 0; +} + +#endif /* _EXT4_JBD2_H */ diff --git a/fs/ext4/ext4_sb.h b/fs/ext4/ext4_sb.h new file mode 100644 index 000000000000..5802e69f2191 --- /dev/null +++ b/fs/ext4/ext4_sb.h @@ -0,0 +1,148 @@ +/* + * ext4_sb.h + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/include/linux/minix_fs_sb.h + * + * Copyright (C) 1991, 1992 Linus Torvalds + */ + +#ifndef _EXT4_SB +#define _EXT4_SB + +#ifdef __KERNEL__ +#include +#include +#include +#include +#endif +#include + +/* + * third extended-fs super-block data in memory + */ +struct ext4_sb_info { + unsigned long s_desc_size; /* Size of a group descriptor in bytes */ + unsigned long s_inodes_per_block;/* Number of inodes per block */ + unsigned long s_blocks_per_group;/* Number of blocks in a group */ + unsigned long s_inodes_per_group;/* Number of inodes in a group */ + unsigned long s_itb_per_group; /* Number of inode table blocks per group */ + unsigned long s_gdb_count; /* Number of group descriptor blocks */ + unsigned long s_desc_per_block; /* Number of group descriptors per block */ + ext4_group_t s_groups_count; /* Number of groups in the fs */ + unsigned long s_overhead_last; /* Last calculated overhead */ + unsigned long s_blocks_last; /* Last seen block count */ + loff_t s_bitmap_maxbytes; /* max bytes for bitmap files */ + struct buffer_head * s_sbh; /* Buffer containing the super block */ + struct ext4_super_block * s_es; /* Pointer to the super block in the buffer */ + struct buffer_head ** s_group_desc; + unsigned long s_mount_opt; + ext4_fsblk_t s_sb_block; + uid_t s_resuid; + gid_t s_resgid; + unsigned short s_mount_state; + unsigned short s_pad; + int s_addr_per_block_bits; + int s_desc_per_block_bits; + int s_inode_size; + int s_first_ino; + spinlock_t s_next_gen_lock; + u32 s_next_generation; + u32 s_hash_seed[4]; + int s_def_hash_version; + struct percpu_counter s_freeblocks_counter; + struct percpu_counter s_freeinodes_counter; + struct percpu_counter s_dirs_counter; + struct blockgroup_lock s_blockgroup_lock; + + /* root of the per fs reservation window tree */ + spinlock_t s_rsv_window_lock; + struct rb_root s_rsv_window_root; + struct ext4_reserve_window_node s_rsv_window_head; + + /* Journaling */ + struct inode * s_journal_inode; + struct journal_s * s_journal; + struct list_head s_orphan; + unsigned long s_commit_interval; + struct block_device *journal_bdev; +#ifdef CONFIG_JBD2_DEBUG + struct timer_list turn_ro_timer; /* For turning read-only (crash simulation) */ + wait_queue_head_t ro_wait_queue; /* For people waiting for the fs to go read-only */ +#endif +#ifdef CONFIG_QUOTA + char *s_qf_names[MAXQUOTAS]; /* Names of quota files with journalled quota */ + int s_jquota_fmt; /* Format of quota to use */ +#endif + unsigned int s_want_extra_isize; /* New inodes should reserve # bytes */ + +#ifdef EXTENTS_STATS + /* ext4 extents stats */ + unsigned long s_ext_min; + unsigned long s_ext_max; + unsigned long s_depth_max; + spinlock_t s_ext_stats_lock; + unsigned long s_ext_blocks; + unsigned long s_ext_extents; +#endif + + /* for buddy allocator */ + struct ext4_group_info ***s_group_info; + struct inode *s_buddy_cache; + long s_blocks_reserved; + spinlock_t s_reserve_lock; + struct list_head s_active_transaction; + struct list_head s_closed_transaction; + struct list_head s_committed_transaction; + spinlock_t s_md_lock; + tid_t s_last_transaction; + unsigned short *s_mb_offsets, *s_mb_maxs; + + /* tunables */ + unsigned long s_stripe; + unsigned long s_mb_stream_request; + unsigned long s_mb_max_to_scan; + unsigned long s_mb_min_to_scan; + unsigned long s_mb_stats; + unsigned long s_mb_order2_reqs; + unsigned long s_mb_group_prealloc; + /* where last allocation was done - for stream allocation */ + unsigned long s_mb_last_group; + unsigned long s_mb_last_start; + + /* history to debug policy */ + struct ext4_mb_history *s_mb_history; + int s_mb_history_cur; + int s_mb_history_max; + int s_mb_history_num; + struct proc_dir_entry *s_mb_proc; + spinlock_t s_mb_history_lock; + int s_mb_history_filter; + + /* stats for buddy allocator */ + spinlock_t s_mb_pa_lock; + atomic_t s_bal_reqs; /* number of reqs with len > 1 */ + atomic_t s_bal_success; /* we found long enough chunks */ + atomic_t s_bal_allocated; /* in blocks */ + atomic_t s_bal_ex_scanned; /* total extents scanned */ + atomic_t s_bal_goals; /* goal hits */ + atomic_t s_bal_breaks; /* too long searches */ + atomic_t s_bal_2orders; /* 2^order hits */ + spinlock_t s_bal_lock; + unsigned long s_mb_buddies_generated; + unsigned long long s_mb_generation_time; + atomic_t s_mb_lost_chunks; + atomic_t s_mb_preallocated; + atomic_t s_mb_discarded; + + /* locality groups */ + struct ext4_locality_group *s_locality_groups; +}; + +#endif /* _EXT4_SB */ diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 4e6afc812fda..a472bc046363 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -32,7 +32,6 @@ #include #include #include -#include #include #include #include @@ -40,8 +39,9 @@ #include #include #include -#include #include +#include "ext4_jbd2.h" +#include "ext4_extents.h" /* diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 20507a24506a..4159be6366ab 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -21,8 +21,8 @@ #include #include #include -#include -#include +#include "ext4.h" +#include "ext4_jbd2.h" #include "xattr.h" #include "acl.h" diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index a04a1ac4e0cf..1c8ba48d4f8d 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -27,8 +27,8 @@ #include #include #include -#include -#include +#include "ext4.h" +#include "ext4_jbd2.h" /* * akpm: A new design for ext4_sync_file(). diff --git a/fs/ext4/hash.c b/fs/ext4/hash.c index 1555024e3b36..1d6329dbe390 100644 --- a/fs/ext4/hash.c +++ b/fs/ext4/hash.c @@ -11,8 +11,8 @@ #include #include -#include #include +#include "ext4.h" #define DELTA 0x9E3779B9 diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index a86377401ff0..d59bdf7233b5 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -15,8 +15,6 @@ #include #include #include -#include -#include #include #include #include @@ -25,7 +23,8 @@ #include #include #include - +#include "ext4.h" +#include "ext4_jbd2.h" #include "xattr.h" #include "acl.h" #include "group.h" diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index bd1a391725c0..0c94db462c2f 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -25,7 +25,6 @@ #include #include #include -#include #include #include #include @@ -36,6 +35,7 @@ #include #include #include +#include "ext4_jbd2.h" #include "xattr.h" #include "acl.h" diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index ce937fe432a0..7a6c2f1faba6 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -10,13 +10,13 @@ #include #include #include -#include -#include #include #include #include #include #include +#include "ext4_jbd2.h" +#include "ext4.h" long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 0b46fc0ca196..f87471de3af7 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -24,8 +24,6 @@ #include #include #include -#include -#include #include #include #include @@ -34,6 +32,8 @@ #include #include #include +#include "ext4_jbd2.h" +#include "ext4.h" #include "group.h" /* diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c index 9b4fb07d192c..b9e077ba07e9 100644 --- a/fs/ext4/migrate.c +++ b/fs/ext4/migrate.c @@ -13,8 +13,8 @@ */ #include -#include -#include +#include "ext4_jbd2.h" +#include "ext4_extents.h" /* * The contiguous blocks details which can be diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 7fc1bc1c16d1..ab16beaa830d 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -28,14 +28,14 @@ #include #include #include -#include -#include #include #include #include #include #include #include +#include "ext4.h" +#include "ext4_jbd2.h" #include "namei.h" #include "xattr.h" diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 0ca63dcbdf88..9f086a6a472b 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -11,11 +11,10 @@ #define EXT4FS_DEBUG -#include - #include #include +#include "ext4_jbd2.h" #include "group.h" #define outside(b, first, last) ((b) < (first) || (b) >= (last)) diff --git a/fs/ext4/super.c b/fs/ext4/super.c index e3b3483b600d..3435184114c4 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -21,8 +21,6 @@ #include #include #include -#include -#include #include #include #include @@ -38,9 +36,10 @@ #include #include #include - #include +#include "ext4.h" +#include "ext4_jbd2.h" #include "xattr.h" #include "acl.h" #include "namei.h" diff --git a/fs/ext4/symlink.c b/fs/ext4/symlink.c index e6f9da4287c4..e9178643dc01 100644 --- a/fs/ext4/symlink.c +++ b/fs/ext4/symlink.c @@ -19,8 +19,8 @@ #include #include -#include #include +#include "ext4.h" #include "xattr.h" static void * ext4_follow_link(struct dentry *dentry, struct nameidata *nd) diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index df4810d5a387..3fbc2c6c3d0e 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -53,11 +53,11 @@ #include #include #include -#include -#include #include #include #include +#include "ext4_jbd2.h" +#include "ext4.h" #include "xattr.h" #include "acl.h" diff --git a/fs/ext4/xattr_security.c b/fs/ext4/xattr_security.c index f17eaf2321b9..ca5f89fc6cae 100644 --- a/fs/ext4/xattr_security.c +++ b/fs/ext4/xattr_security.c @@ -6,9 +6,9 @@ #include #include #include -#include -#include #include +#include "ext4_jbd2.h" +#include "ext4.h" #include "xattr.h" static size_t diff --git a/fs/ext4/xattr_trusted.c b/fs/ext4/xattr_trusted.c index e0f05acdafec..fff33382cadc 100644 --- a/fs/ext4/xattr_trusted.c +++ b/fs/ext4/xattr_trusted.c @@ -9,8 +9,8 @@ #include #include #include -#include -#include +#include "ext4_jbd2.h" +#include "ext4.h" #include "xattr.h" #define XATTR_TRUSTED_PREFIX "trusted." diff --git a/fs/ext4/xattr_user.c b/fs/ext4/xattr_user.c index 7ed3d8ebf096..67be723fcc4e 100644 --- a/fs/ext4/xattr_user.c +++ b/fs/ext4/xattr_user.c @@ -8,8 +8,8 @@ #include #include #include -#include -#include +#include "ext4_jbd2.h" +#include "ext4.h" #include "xattr.h" #define XATTR_USER_PREFIX "user." diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h deleted file mode 100644 index 1ae0f965f386..000000000000 --- a/include/linux/ext4_fs.h +++ /dev/null @@ -1,1205 +0,0 @@ -/* - * linux/include/linux/ext4_fs.h - * - * Copyright (C) 1992, 1993, 1994, 1995 - * Remy Card (card@masi.ibp.fr) - * Laboratoire MASI - Institut Blaise Pascal - * Universite Pierre et Marie Curie (Paris VI) - * - * from - * - * linux/include/linux/minix_fs.h - * - * Copyright (C) 1991, 1992 Linus Torvalds - */ - -#ifndef _LINUX_EXT4_FS_H -#define _LINUX_EXT4_FS_H - -#include -#include -#include -#include - -/* - * The second extended filesystem constants/structures - */ - -/* - * Define EXT4FS_DEBUG to produce debug messages - */ -#undef EXT4FS_DEBUG - -/* - * Define EXT4_RESERVATION to reserve data blocks for expanding files - */ -#define EXT4_DEFAULT_RESERVE_BLOCKS 8 -/*max window size: 1024(direct blocks) + 3([t,d]indirect blocks) */ -#define EXT4_MAX_RESERVE_BLOCKS 1027 -#define EXT4_RESERVE_WINDOW_NOT_ALLOCATED 0 - -/* - * Debug code - */ -#ifdef EXT4FS_DEBUG -#define ext4_debug(f, a...) \ - do { \ - printk (KERN_DEBUG "EXT4-fs DEBUG (%s, %d): %s:", \ - __FILE__, __LINE__, __FUNCTION__); \ - printk (KERN_DEBUG f, ## a); \ - } while (0) -#else -#define ext4_debug(f, a...) do {} while (0) -#endif - -#define EXT4_MULTIBLOCK_ALLOCATOR 1 - -/* prefer goal again. length */ -#define EXT4_MB_HINT_MERGE 1 -/* blocks already reserved */ -#define EXT4_MB_HINT_RESERVED 2 -/* metadata is being allocated */ -#define EXT4_MB_HINT_METADATA 4 -/* first blocks in the file */ -#define EXT4_MB_HINT_FIRST 8 -/* search for the best chunk */ -#define EXT4_MB_HINT_BEST 16 -/* data is being allocated */ -#define EXT4_MB_HINT_DATA 32 -/* don't preallocate (for tails) */ -#define EXT4_MB_HINT_NOPREALLOC 64 -/* allocate for locality group */ -#define EXT4_MB_HINT_GROUP_ALLOC 128 -/* allocate goal blocks or none */ -#define EXT4_MB_HINT_GOAL_ONLY 256 -/* goal is meaningful */ -#define EXT4_MB_HINT_TRY_GOAL 512 - -struct ext4_allocation_request { - /* target inode for block we're allocating */ - struct inode *inode; - /* logical block in target inode */ - ext4_lblk_t logical; - /* phys. target (a hint) */ - ext4_fsblk_t goal; - /* the closest logical allocated block to the left */ - ext4_lblk_t lleft; - /* phys. block for ^^^ */ - ext4_fsblk_t pleft; - /* the closest logical allocated block to the right */ - ext4_lblk_t lright; - /* phys. block for ^^^ */ - ext4_fsblk_t pright; - /* how many blocks we want to allocate */ - unsigned long len; - /* flags. see above EXT4_MB_HINT_* */ - unsigned long flags; -}; - -/* - * Special inodes numbers - */ -#define EXT4_BAD_INO 1 /* Bad blocks inode */ -#define EXT4_ROOT_INO 2 /* Root inode */ -#define EXT4_BOOT_LOADER_INO 5 /* Boot loader inode */ -#define EXT4_UNDEL_DIR_INO 6 /* Undelete directory inode */ -#define EXT4_RESIZE_INO 7 /* Reserved group descriptors inode */ -#define EXT4_JOURNAL_INO 8 /* Journal inode */ - -/* First non-reserved inode for old ext4 filesystems */ -#define EXT4_GOOD_OLD_FIRST_INO 11 - -/* - * Maximal count of links to a file - */ -#define EXT4_LINK_MAX 65000 - -/* - * Macro-instructions used to manage several block sizes - */ -#define EXT4_MIN_BLOCK_SIZE 1024 -#define EXT4_MAX_BLOCK_SIZE 65536 -#define EXT4_MIN_BLOCK_LOG_SIZE 10 -#ifdef __KERNEL__ -# define EXT4_BLOCK_SIZE(s) ((s)->s_blocksize) -#else -# define EXT4_BLOCK_SIZE(s) (EXT4_MIN_BLOCK_SIZE << (s)->s_log_block_size) -#endif -#define EXT4_ADDR_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / sizeof (__u32)) -#ifdef __KERNEL__ -# define EXT4_BLOCK_SIZE_BITS(s) ((s)->s_blocksize_bits) -#else -# define EXT4_BLOCK_SIZE_BITS(s) ((s)->s_log_block_size + 10) -#endif -#ifdef __KERNEL__ -#define EXT4_ADDR_PER_BLOCK_BITS(s) (EXT4_SB(s)->s_addr_per_block_bits) -#define EXT4_INODE_SIZE(s) (EXT4_SB(s)->s_inode_size) -#define EXT4_FIRST_INO(s) (EXT4_SB(s)->s_first_ino) -#else -#define EXT4_INODE_SIZE(s) (((s)->s_rev_level == EXT4_GOOD_OLD_REV) ? \ - EXT4_GOOD_OLD_INODE_SIZE : \ - (s)->s_inode_size) -#define EXT4_FIRST_INO(s) (((s)->s_rev_level == EXT4_GOOD_OLD_REV) ? \ - EXT4_GOOD_OLD_FIRST_INO : \ - (s)->s_first_ino) -#endif -#define EXT4_BLOCK_ALIGN(size, blkbits) ALIGN((size), (1 << (blkbits))) - -/* - * Structure of a blocks group descriptor - */ -struct ext4_group_desc -{ - __le32 bg_block_bitmap_lo; /* Blocks bitmap block */ - __le32 bg_inode_bitmap_lo; /* Inodes bitmap block */ - __le32 bg_inode_table_lo; /* Inodes table block */ - __le16 bg_free_blocks_count; /* Free blocks count */ - __le16 bg_free_inodes_count; /* Free inodes count */ - __le16 bg_used_dirs_count; /* Directories count */ - __le16 bg_flags; /* EXT4_BG_flags (INODE_UNINIT, etc) */ - __u32 bg_reserved[2]; /* Likely block/inode bitmap checksum */ - __le16 bg_itable_unused; /* Unused inodes count */ - __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ - __le32 bg_block_bitmap_hi; /* Blocks bitmap block MSB */ - __le32 bg_inode_bitmap_hi; /* Inodes bitmap block MSB */ - __le32 bg_inode_table_hi; /* Inodes table block MSB */ - __le16 bg_free_blocks_count_hi;/* Free blocks count MSB */ - __le16 bg_free_inodes_count_hi;/* Free inodes count MSB */ - __le16 bg_used_dirs_count_hi; /* Directories count MSB */ - __le16 bg_itable_unused_hi; /* Unused inodes count MSB */ - __u32 bg_reserved2[3]; -}; - -#define EXT4_BG_INODE_UNINIT 0x0001 /* Inode table/bitmap not in use */ -#define EXT4_BG_BLOCK_UNINIT 0x0002 /* Block bitmap not in use */ -#define EXT4_BG_INODE_ZEROED 0x0004 /* On-disk itable initialized to zero */ - -#ifdef __KERNEL__ -#include -#endif -/* - * Macro-instructions used to manage group descriptors - */ -#define EXT4_MIN_DESC_SIZE 32 -#define EXT4_MIN_DESC_SIZE_64BIT 64 -#define EXT4_MAX_DESC_SIZE EXT4_MIN_BLOCK_SIZE -#define EXT4_DESC_SIZE(s) (EXT4_SB(s)->s_desc_size) -#ifdef __KERNEL__ -# define EXT4_BLOCKS_PER_GROUP(s) (EXT4_SB(s)->s_blocks_per_group) -# define EXT4_DESC_PER_BLOCK(s) (EXT4_SB(s)->s_desc_per_block) -# define EXT4_INODES_PER_GROUP(s) (EXT4_SB(s)->s_inodes_per_group) -# define EXT4_DESC_PER_BLOCK_BITS(s) (EXT4_SB(s)->s_desc_per_block_bits) -#else -# define EXT4_BLOCKS_PER_GROUP(s) ((s)->s_blocks_per_group) -# define EXT4_DESC_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / EXT4_DESC_SIZE(s)) -# define EXT4_INODES_PER_GROUP(s) ((s)->s_inodes_per_group) -#endif - -/* - * Constants relative to the data blocks - */ -#define EXT4_NDIR_BLOCKS 12 -#define EXT4_IND_BLOCK EXT4_NDIR_BLOCKS -#define EXT4_DIND_BLOCK (EXT4_IND_BLOCK + 1) -#define EXT4_TIND_BLOCK (EXT4_DIND_BLOCK + 1) -#define EXT4_N_BLOCKS (EXT4_TIND_BLOCK + 1) - -/* - * Inode flags - */ -#define EXT4_SECRM_FL 0x00000001 /* Secure deletion */ -#define EXT4_UNRM_FL 0x00000002 /* Undelete */ -#define EXT4_COMPR_FL 0x00000004 /* Compress file */ -#define EXT4_SYNC_FL 0x00000008 /* Synchronous updates */ -#define EXT4_IMMUTABLE_FL 0x00000010 /* Immutable file */ -#define EXT4_APPEND_FL 0x00000020 /* writes to file may only append */ -#define EXT4_NODUMP_FL 0x00000040 /* do not dump file */ -#define EXT4_NOATIME_FL 0x00000080 /* do not update atime */ -/* Reserved for compression usage... */ -#define EXT4_DIRTY_FL 0x00000100 -#define EXT4_COMPRBLK_FL 0x00000200 /* One or more compressed clusters */ -#define EXT4_NOCOMPR_FL 0x00000400 /* Don't compress */ -#define EXT4_ECOMPR_FL 0x00000800 /* Compression error */ -/* End compression flags --- maybe not all used */ -#define EXT4_INDEX_FL 0x00001000 /* hash-indexed directory */ -#define EXT4_IMAGIC_FL 0x00002000 /* AFS directory */ -#define EXT4_JOURNAL_DATA_FL 0x00004000 /* file data should be journaled */ -#define EXT4_NOTAIL_FL 0x00008000 /* file tail should not be merged */ -#define EXT4_DIRSYNC_FL 0x00010000 /* dirsync behaviour (directories only) */ -#define EXT4_TOPDIR_FL 0x00020000 /* Top of directory hierarchies*/ -#define EXT4_HUGE_FILE_FL 0x00040000 /* Set to each huge file */ -#define EXT4_EXTENTS_FL 0x00080000 /* Inode uses extents */ -#define EXT4_EXT_MIGRATE 0x00100000 /* Inode is migrating */ -#define EXT4_RESERVED_FL 0x80000000 /* reserved for ext4 lib */ - -#define EXT4_FL_USER_VISIBLE 0x000BDFFF /* User visible flags */ -#define EXT4_FL_USER_MODIFIABLE 0x000380FF /* User modifiable flags */ - -/* - * Inode dynamic state flags - */ -#define EXT4_STATE_JDATA 0x00000001 /* journaled data exists */ -#define EXT4_STATE_NEW 0x00000002 /* inode is newly created */ -#define EXT4_STATE_XATTR 0x00000004 /* has in-inode xattrs */ -#define EXT4_STATE_NO_EXPAND 0x00000008 /* No space for expansion */ - -/* Used to pass group descriptor data when online resize is done */ -struct ext4_new_group_input { - __u32 group; /* Group number for this data */ - __u64 block_bitmap; /* Absolute block number of block bitmap */ - __u64 inode_bitmap; /* Absolute block number of inode bitmap */ - __u64 inode_table; /* Absolute block number of inode table start */ - __u32 blocks_count; /* Total number of blocks in this group */ - __u16 reserved_blocks; /* Number of reserved blocks in this group */ - __u16 unused; -}; - -/* The struct ext4_new_group_input in kernel space, with free_blocks_count */ -struct ext4_new_group_data { - __u32 group; - __u64 block_bitmap; - __u64 inode_bitmap; - __u64 inode_table; - __u32 blocks_count; - __u16 reserved_blocks; - __u16 unused; - __u32 free_blocks_count; -}; - -/* - * Following is used by preallocation code to tell get_blocks() that we - * want uninitialzed extents. - */ -#define EXT4_CREATE_UNINITIALIZED_EXT 2 - -/* - * ioctl commands - */ -#define EXT4_IOC_GETFLAGS FS_IOC_GETFLAGS -#define EXT4_IOC_SETFLAGS FS_IOC_SETFLAGS -#define EXT4_IOC_GETVERSION _IOR('f', 3, long) -#define EXT4_IOC_SETVERSION _IOW('f', 4, long) -#define EXT4_IOC_GROUP_EXTEND _IOW('f', 7, unsigned long) -#define EXT4_IOC_GROUP_ADD _IOW('f', 8,struct ext4_new_group_input) -#define EXT4_IOC_GETVERSION_OLD FS_IOC_GETVERSION -#define EXT4_IOC_SETVERSION_OLD FS_IOC_SETVERSION -#ifdef CONFIG_JBD2_DEBUG -#define EXT4_IOC_WAIT_FOR_READONLY _IOR('f', 99, long) -#endif -#define EXT4_IOC_GETRSVSZ _IOR('f', 5, long) -#define EXT4_IOC_SETRSVSZ _IOW('f', 6, long) -#define EXT4_IOC_MIGRATE _IO('f', 7) - -/* - * ioctl commands in 32 bit emulation - */ -#define EXT4_IOC32_GETFLAGS FS_IOC32_GETFLAGS -#define EXT4_IOC32_SETFLAGS FS_IOC32_SETFLAGS -#define EXT4_IOC32_GETVERSION _IOR('f', 3, int) -#define EXT4_IOC32_SETVERSION _IOW('f', 4, int) -#define EXT4_IOC32_GETRSVSZ _IOR('f', 5, int) -#define EXT4_IOC32_SETRSVSZ _IOW('f', 6, int) -#define EXT4_IOC32_GROUP_EXTEND _IOW('f', 7, unsigned int) -#ifdef CONFIG_JBD2_DEBUG -#define EXT4_IOC32_WAIT_FOR_READONLY _IOR('f', 99, int) -#endif -#define EXT4_IOC32_GETVERSION_OLD FS_IOC32_GETVERSION -#define EXT4_IOC32_SETVERSION_OLD FS_IOC32_SETVERSION - - -/* - * Mount options - */ -struct ext4_mount_options { - unsigned long s_mount_opt; - uid_t s_resuid; - gid_t s_resgid; - unsigned long s_commit_interval; -#ifdef CONFIG_QUOTA - int s_jquota_fmt; - char *s_qf_names[MAXQUOTAS]; -#endif -}; - -/* - * Structure of an inode on the disk - */ -struct ext4_inode { - __le16 i_mode; /* File mode */ - __le16 i_uid; /* Low 16 bits of Owner Uid */ - __le32 i_size_lo; /* Size in bytes */ - __le32 i_atime; /* Access time */ - __le32 i_ctime; /* Inode Change time */ - __le32 i_mtime; /* Modification time */ - __le32 i_dtime; /* Deletion Time */ - __le16 i_gid; /* Low 16 bits of Group Id */ - __le16 i_links_count; /* Links count */ - __le32 i_blocks_lo; /* Blocks count */ - __le32 i_flags; /* File flags */ - union { - struct { - __le32 l_i_version; - } linux1; - struct { - __u32 h_i_translator; - } hurd1; - struct { - __u32 m_i_reserved1; - } masix1; - } osd1; /* OS dependent 1 */ - __le32 i_block[EXT4_N_BLOCKS];/* Pointers to blocks */ - __le32 i_generation; /* File version (for NFS) */ - __le32 i_file_acl_lo; /* File ACL */ - __le32 i_size_high; - __le32 i_obso_faddr; /* Obsoleted fragment address */ - union { - struct { - __le16 l_i_blocks_high; /* were l_i_reserved1 */ - __le16 l_i_file_acl_high; - __le16 l_i_uid_high; /* these 2 fields */ - __le16 l_i_gid_high; /* were reserved2[0] */ - __u32 l_i_reserved2; - } linux2; - struct { - __le16 h_i_reserved1; /* Obsoleted fragment number/size which are removed in ext4 */ - __u16 h_i_mode_high; - __u16 h_i_uid_high; - __u16 h_i_gid_high; - __u32 h_i_author; - } hurd2; - struct { - __le16 h_i_reserved1; /* Obsoleted fragment number/size which are removed in ext4 */ - __le16 m_i_file_acl_high; - __u32 m_i_reserved2[2]; - } masix2; - } osd2; /* OS dependent 2 */ - __le16 i_extra_isize; - __le16 i_pad1; - __le32 i_ctime_extra; /* extra Change time (nsec << 2 | epoch) */ - __le32 i_mtime_extra; /* extra Modification time(nsec << 2 | epoch) */ - __le32 i_atime_extra; /* extra Access time (nsec << 2 | epoch) */ - __le32 i_crtime; /* File Creation time */ - __le32 i_crtime_extra; /* extra FileCreationtime (nsec << 2 | epoch) */ - __le32 i_version_hi; /* high 32 bits for 64-bit version */ -}; - - -#define EXT4_EPOCH_BITS 2 -#define EXT4_EPOCH_MASK ((1 << EXT4_EPOCH_BITS) - 1) -#define EXT4_NSEC_MASK (~0UL << EXT4_EPOCH_BITS) - -/* - * Extended fields will fit into an inode if the filesystem was formatted - * with large inodes (-I 256 or larger) and there are not currently any EAs - * consuming all of the available space. For new inodes we always reserve - * enough space for the kernel's known extended fields, but for inodes - * created with an old kernel this might not have been the case. None of - * the extended inode fields is critical for correct filesystem operation. - * This macro checks if a certain field fits in the inode. Note that - * inode-size = GOOD_OLD_INODE_SIZE + i_extra_isize - */ -#define EXT4_FITS_IN_INODE(ext4_inode, einode, field) \ - ((offsetof(typeof(*ext4_inode), field) + \ - sizeof((ext4_inode)->field)) \ - <= (EXT4_GOOD_OLD_INODE_SIZE + \ - (einode)->i_extra_isize)) \ - -static inline __le32 ext4_encode_extra_time(struct timespec *time) -{ - return cpu_to_le32((sizeof(time->tv_sec) > 4 ? - time->tv_sec >> 32 : 0) | - ((time->tv_nsec << 2) & EXT4_NSEC_MASK)); -} - -static inline void ext4_decode_extra_time(struct timespec *time, __le32 extra) -{ - if (sizeof(time->tv_sec) > 4) - time->tv_sec |= (__u64)(le32_to_cpu(extra) & EXT4_EPOCH_MASK) - << 32; - time->tv_nsec = (le32_to_cpu(extra) & EXT4_NSEC_MASK) >> 2; -} - -#define EXT4_INODE_SET_XTIME(xtime, inode, raw_inode) \ -do { \ - (raw_inode)->xtime = cpu_to_le32((inode)->xtime.tv_sec); \ - if (EXT4_FITS_IN_INODE(raw_inode, EXT4_I(inode), xtime ## _extra)) \ - (raw_inode)->xtime ## _extra = \ - ext4_encode_extra_time(&(inode)->xtime); \ -} while (0) - -#define EXT4_EINODE_SET_XTIME(xtime, einode, raw_inode) \ -do { \ - if (EXT4_FITS_IN_INODE(raw_inode, einode, xtime)) \ - (raw_inode)->xtime = cpu_to_le32((einode)->xtime.tv_sec); \ - if (EXT4_FITS_IN_INODE(raw_inode, einode, xtime ## _extra)) \ - (raw_inode)->xtime ## _extra = \ - ext4_encode_extra_time(&(einode)->xtime); \ -} while (0) - -#define EXT4_INODE_GET_XTIME(xtime, inode, raw_inode) \ -do { \ - (inode)->xtime.tv_sec = (signed)le32_to_cpu((raw_inode)->xtime); \ - if (EXT4_FITS_IN_INODE(raw_inode, EXT4_I(inode), xtime ## _extra)) \ - ext4_decode_extra_time(&(inode)->xtime, \ - raw_inode->xtime ## _extra); \ -} while (0) - -#define EXT4_EINODE_GET_XTIME(xtime, einode, raw_inode) \ -do { \ - if (EXT4_FITS_IN_INODE(raw_inode, einode, xtime)) \ - (einode)->xtime.tv_sec = \ - (signed)le32_to_cpu((raw_inode)->xtime); \ - if (EXT4_FITS_IN_INODE(raw_inode, einode, xtime ## _extra)) \ - ext4_decode_extra_time(&(einode)->xtime, \ - raw_inode->xtime ## _extra); \ -} while (0) - -#define i_disk_version osd1.linux1.l_i_version - -#if defined(__KERNEL__) || defined(__linux__) -#define i_reserved1 osd1.linux1.l_i_reserved1 -#define i_file_acl_high osd2.linux2.l_i_file_acl_high -#define i_blocks_high osd2.linux2.l_i_blocks_high -#define i_uid_low i_uid -#define i_gid_low i_gid -#define i_uid_high osd2.linux2.l_i_uid_high -#define i_gid_high osd2.linux2.l_i_gid_high -#define i_reserved2 osd2.linux2.l_i_reserved2 - -#elif defined(__GNU__) - -#define i_translator osd1.hurd1.h_i_translator -#define i_uid_high osd2.hurd2.h_i_uid_high -#define i_gid_high osd2.hurd2.h_i_gid_high -#define i_author osd2.hurd2.h_i_author - -#elif defined(__masix__) - -#define i_reserved1 osd1.masix1.m_i_reserved1 -#define i_file_acl_high osd2.masix2.m_i_file_acl_high -#define i_reserved2 osd2.masix2.m_i_reserved2 - -#endif /* defined(__KERNEL__) || defined(__linux__) */ - -/* - * File system states - */ -#define EXT4_VALID_FS 0x0001 /* Unmounted cleanly */ -#define EXT4_ERROR_FS 0x0002 /* Errors detected */ -#define EXT4_ORPHAN_FS 0x0004 /* Orphans being recovered */ - -/* - * Misc. filesystem flags - */ -#define EXT2_FLAGS_SIGNED_HASH 0x0001 /* Signed dirhash in use */ -#define EXT2_FLAGS_UNSIGNED_HASH 0x0002 /* Unsigned dirhash in use */ -#define EXT2_FLAGS_TEST_FILESYS 0x0004 /* to test development code */ - -/* - * Mount flags - */ -#define EXT4_MOUNT_CHECK 0x00001 /* Do mount-time checks */ -#define EXT4_MOUNT_OLDALLOC 0x00002 /* Don't use the new Orlov allocator */ -#define EXT4_MOUNT_GRPID 0x00004 /* Create files with directory's group */ -#define EXT4_MOUNT_DEBUG 0x00008 /* Some debugging messages */ -#define EXT4_MOUNT_ERRORS_CONT 0x00010 /* Continue on errors */ -#define EXT4_MOUNT_ERRORS_RO 0x00020 /* Remount fs ro on errors */ -#define EXT4_MOUNT_ERRORS_PANIC 0x00040 /* Panic on errors */ -#define EXT4_MOUNT_MINIX_DF 0x00080 /* Mimics the Minix statfs */ -#define EXT4_MOUNT_NOLOAD 0x00100 /* Don't use existing journal*/ -#define EXT4_MOUNT_ABORT 0x00200 /* Fatal error detected */ -#define EXT4_MOUNT_DATA_FLAGS 0x00C00 /* Mode for data writes: */ -#define EXT4_MOUNT_JOURNAL_DATA 0x00400 /* Write data to journal */ -#define EXT4_MOUNT_ORDERED_DATA 0x00800 /* Flush data before commit */ -#define EXT4_MOUNT_WRITEBACK_DATA 0x00C00 /* No data ordering */ -#define EXT4_MOUNT_UPDATE_JOURNAL 0x01000 /* Update the journal format */ -#define EXT4_MOUNT_NO_UID32 0x02000 /* Disable 32-bit UIDs */ -#define EXT4_MOUNT_XATTR_USER 0x04000 /* Extended user attributes */ -#define EXT4_MOUNT_POSIX_ACL 0x08000 /* POSIX Access Control Lists */ -#define EXT4_MOUNT_RESERVATION 0x10000 /* Preallocation */ -#define EXT4_MOUNT_BARRIER 0x20000 /* Use block barriers */ -#define EXT4_MOUNT_NOBH 0x40000 /* No bufferheads */ -#define EXT4_MOUNT_QUOTA 0x80000 /* Some quota option set */ -#define EXT4_MOUNT_USRQUOTA 0x100000 /* "old" user quota */ -#define EXT4_MOUNT_GRPQUOTA 0x200000 /* "old" group quota */ -#define EXT4_MOUNT_EXTENTS 0x400000 /* Extents support */ -#define EXT4_MOUNT_JOURNAL_CHECKSUM 0x800000 /* Journal checksums */ -#define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT 0x1000000 /* Journal Async Commit */ -#define EXT4_MOUNT_I_VERSION 0x2000000 /* i_version support */ -#define EXT4_MOUNT_MBALLOC 0x4000000 /* Buddy allocation support */ -/* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */ -#ifndef _LINUX_EXT2_FS_H -#define clear_opt(o, opt) o &= ~EXT4_MOUNT_##opt -#define set_opt(o, opt) o |= EXT4_MOUNT_##opt -#define test_opt(sb, opt) (EXT4_SB(sb)->s_mount_opt & \ - EXT4_MOUNT_##opt) -#else -#define EXT2_MOUNT_NOLOAD EXT4_MOUNT_NOLOAD -#define EXT2_MOUNT_ABORT EXT4_MOUNT_ABORT -#define EXT2_MOUNT_DATA_FLAGS EXT4_MOUNT_DATA_FLAGS -#endif - -#define ext4_set_bit ext2_set_bit -#define ext4_set_bit_atomic ext2_set_bit_atomic -#define ext4_clear_bit ext2_clear_bit -#define ext4_clear_bit_atomic ext2_clear_bit_atomic -#define ext4_test_bit ext2_test_bit -#define ext4_find_first_zero_bit ext2_find_first_zero_bit -#define ext4_find_next_zero_bit ext2_find_next_zero_bit -#define ext4_find_next_bit ext2_find_next_bit - -/* - * Maximal mount counts between two filesystem checks - */ -#define EXT4_DFL_MAX_MNT_COUNT 20 /* Allow 20 mounts */ -#define EXT4_DFL_CHECKINTERVAL 0 /* Don't use interval check */ - -/* - * Behaviour when detecting errors - */ -#define EXT4_ERRORS_CONTINUE 1 /* Continue execution */ -#define EXT4_ERRORS_RO 2 /* Remount fs read-only */ -#define EXT4_ERRORS_PANIC 3 /* Panic */ -#define EXT4_ERRORS_DEFAULT EXT4_ERRORS_CONTINUE - -/* - * Structure of the super block - */ -struct ext4_super_block { -/*00*/ __le32 s_inodes_count; /* Inodes count */ - __le32 s_blocks_count_lo; /* Blocks count */ - __le32 s_r_blocks_count_lo; /* Reserved blocks count */ - __le32 s_free_blocks_count_lo; /* Free blocks count */ -/*10*/ __le32 s_free_inodes_count; /* Free inodes count */ - __le32 s_first_data_block; /* First Data Block */ - __le32 s_log_block_size; /* Block size */ - __le32 s_obso_log_frag_size; /* Obsoleted fragment size */ -/*20*/ __le32 s_blocks_per_group; /* # Blocks per group */ - __le32 s_obso_frags_per_group; /* Obsoleted fragments per group */ - __le32 s_inodes_per_group; /* # Inodes per group */ - __le32 s_mtime; /* Mount time */ -/*30*/ __le32 s_wtime; /* Write time */ - __le16 s_mnt_count; /* Mount count */ - __le16 s_max_mnt_count; /* Maximal mount count */ - __le16 s_magic; /* Magic signature */ - __le16 s_state; /* File system state */ - __le16 s_errors; /* Behaviour when detecting errors */ - __le16 s_minor_rev_level; /* minor revision level */ -/*40*/ __le32 s_lastcheck; /* time of last check */ - __le32 s_checkinterval; /* max. time between checks */ - __le32 s_creator_os; /* OS */ - __le32 s_rev_level; /* Revision level */ -/*50*/ __le16 s_def_resuid; /* Default uid for reserved blocks */ - __le16 s_def_resgid; /* Default gid for reserved blocks */ - /* - * These fields are for EXT4_DYNAMIC_REV superblocks only. - * - * Note: the difference between the compatible feature set and - * the incompatible feature set is that if there is a bit set - * in the incompatible feature set that the kernel doesn't - * know about, it should refuse to mount the filesystem. - * - * e2fsck's requirements are more strict; if it doesn't know - * about a feature in either the compatible or incompatible - * feature set, it must abort and not try to meddle with - * things it doesn't understand... - */ - __le32 s_first_ino; /* First non-reserved inode */ - __le16 s_inode_size; /* size of inode structure */ - __le16 s_block_group_nr; /* block group # of this superblock */ - __le32 s_feature_compat; /* compatible feature set */ -/*60*/ __le32 s_feature_incompat; /* incompatible feature set */ - __le32 s_feature_ro_compat; /* readonly-compatible feature set */ -/*68*/ __u8 s_uuid[16]; /* 128-bit uuid for volume */ -/*78*/ char s_volume_name[16]; /* volume name */ -/*88*/ char s_last_mounted[64]; /* directory where last mounted */ -/*C8*/ __le32 s_algorithm_usage_bitmap; /* For compression */ - /* - * Performance hints. Directory preallocation should only - * happen if the EXT4_FEATURE_COMPAT_DIR_PREALLOC flag is on. - */ - __u8 s_prealloc_blocks; /* Nr of blocks to try to preallocate*/ - __u8 s_prealloc_dir_blocks; /* Nr to preallocate for dirs */ - __le16 s_reserved_gdt_blocks; /* Per group desc for online growth */ - /* - * Journaling support valid if EXT4_FEATURE_COMPAT_HAS_JOURNAL set. - */ -/*D0*/ __u8 s_journal_uuid[16]; /* uuid of journal superblock */ -/*E0*/ __le32 s_journal_inum; /* inode number of journal file */ - __le32 s_journal_dev; /* device number of journal file */ - __le32 s_last_orphan; /* start of list of inodes to delete */ - __le32 s_hash_seed[4]; /* HTREE hash seed */ - __u8 s_def_hash_version; /* Default hash version to use */ - __u8 s_reserved_char_pad; - __le16 s_desc_size; /* size of group descriptor */ -/*100*/ __le32 s_default_mount_opts; - __le32 s_first_meta_bg; /* First metablock block group */ - __le32 s_mkfs_time; /* When the filesystem was created */ - __le32 s_jnl_blocks[17]; /* Backup of the journal inode */ - /* 64bit support valid if EXT4_FEATURE_COMPAT_64BIT */ -/*150*/ __le32 s_blocks_count_hi; /* Blocks count */ - __le32 s_r_blocks_count_hi; /* Reserved blocks count */ - __le32 s_free_blocks_count_hi; /* Free blocks count */ - __le16 s_min_extra_isize; /* All inodes have at least # bytes */ - __le16 s_want_extra_isize; /* New inodes should reserve # bytes */ - __le32 s_flags; /* Miscellaneous flags */ - __le16 s_raid_stride; /* RAID stride */ - __le16 s_mmp_interval; /* # seconds to wait in MMP checking */ - __le64 s_mmp_block; /* Block for multi-mount protection */ - __le32 s_raid_stripe_width; /* blocks on all data disks (N*stride)*/ - __u32 s_reserved[163]; /* Padding to the end of the block */ -}; - -#ifdef __KERNEL__ -static inline struct ext4_sb_info * EXT4_SB(struct super_block *sb) -{ - return sb->s_fs_info; -} -static inline struct ext4_inode_info *EXT4_I(struct inode *inode) -{ - return container_of(inode, struct ext4_inode_info, vfs_inode); -} - -static inline struct timespec ext4_current_time(struct inode *inode) -{ - return (inode->i_sb->s_time_gran < NSEC_PER_SEC) ? - current_fs_time(inode->i_sb) : CURRENT_TIME_SEC; -} - - -static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino) -{ - return ino == EXT4_ROOT_INO || - ino == EXT4_JOURNAL_INO || - ino == EXT4_RESIZE_INO || - (ino >= EXT4_FIRST_INO(sb) && - ino <= le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count)); -} -#else -/* Assume that user mode programs are passing in an ext4fs superblock, not - * a kernel struct super_block. This will allow us to call the feature-test - * macros from user land. */ -#define EXT4_SB(sb) (sb) -#endif - -#define NEXT_ORPHAN(inode) EXT4_I(inode)->i_dtime - -/* - * Codes for operating systems - */ -#define EXT4_OS_LINUX 0 -#define EXT4_OS_HURD 1 -#define EXT4_OS_MASIX 2 -#define EXT4_OS_FREEBSD 3 -#define EXT4_OS_LITES 4 - -/* - * Revision levels - */ -#define EXT4_GOOD_OLD_REV 0 /* The good old (original) format */ -#define EXT4_DYNAMIC_REV 1 /* V2 format w/ dynamic inode sizes */ - -#define EXT4_CURRENT_REV EXT4_GOOD_OLD_REV -#define EXT4_MAX_SUPP_REV EXT4_DYNAMIC_REV - -#define EXT4_GOOD_OLD_INODE_SIZE 128 - -/* - * Feature set definitions - */ - -#define EXT4_HAS_COMPAT_FEATURE(sb,mask) \ - ( EXT4_SB(sb)->s_es->s_feature_compat & cpu_to_le32(mask) ) -#define EXT4_HAS_RO_COMPAT_FEATURE(sb,mask) \ - ( EXT4_SB(sb)->s_es->s_feature_ro_compat & cpu_to_le32(mask) ) -#define EXT4_HAS_INCOMPAT_FEATURE(sb,mask) \ - ( EXT4_SB(sb)->s_es->s_feature_incompat & cpu_to_le32(mask) ) -#define EXT4_SET_COMPAT_FEATURE(sb,mask) \ - EXT4_SB(sb)->s_es->s_feature_compat |= cpu_to_le32(mask) -#define EXT4_SET_RO_COMPAT_FEATURE(sb,mask) \ - EXT4_SB(sb)->s_es->s_feature_ro_compat |= cpu_to_le32(mask) -#define EXT4_SET_INCOMPAT_FEATURE(sb,mask) \ - EXT4_SB(sb)->s_es->s_feature_incompat |= cpu_to_le32(mask) -#define EXT4_CLEAR_COMPAT_FEATURE(sb,mask) \ - EXT4_SB(sb)->s_es->s_feature_compat &= ~cpu_to_le32(mask) -#define EXT4_CLEAR_RO_COMPAT_FEATURE(sb,mask) \ - EXT4_SB(sb)->s_es->s_feature_ro_compat &= ~cpu_to_le32(mask) -#define EXT4_CLEAR_INCOMPAT_FEATURE(sb,mask) \ - EXT4_SB(sb)->s_es->s_feature_incompat &= ~cpu_to_le32(mask) - -#define EXT4_FEATURE_COMPAT_DIR_PREALLOC 0x0001 -#define EXT4_FEATURE_COMPAT_IMAGIC_INODES 0x0002 -#define EXT4_FEATURE_COMPAT_HAS_JOURNAL 0x0004 -#define EXT4_FEATURE_COMPAT_EXT_ATTR 0x0008 -#define EXT4_FEATURE_COMPAT_RESIZE_INODE 0x0010 -#define EXT4_FEATURE_COMPAT_DIR_INDEX 0x0020 - -#define EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER 0x0001 -#define EXT4_FEATURE_RO_COMPAT_LARGE_FILE 0x0002 -#define EXT4_FEATURE_RO_COMPAT_BTREE_DIR 0x0004 -#define EXT4_FEATURE_RO_COMPAT_HUGE_FILE 0x0008 -#define EXT4_FEATURE_RO_COMPAT_GDT_CSUM 0x0010 -#define EXT4_FEATURE_RO_COMPAT_DIR_NLINK 0x0020 -#define EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE 0x0040 - -#define EXT4_FEATURE_INCOMPAT_COMPRESSION 0x0001 -#define EXT4_FEATURE_INCOMPAT_FILETYPE 0x0002 -#define EXT4_FEATURE_INCOMPAT_RECOVER 0x0004 /* Needs recovery */ -#define EXT4_FEATURE_INCOMPAT_JOURNAL_DEV 0x0008 /* Journal device */ -#define EXT4_FEATURE_INCOMPAT_META_BG 0x0010 -#define EXT4_FEATURE_INCOMPAT_EXTENTS 0x0040 /* extents support */ -#define EXT4_FEATURE_INCOMPAT_64BIT 0x0080 -#define EXT4_FEATURE_INCOMPAT_MMP 0x0100 -#define EXT4_FEATURE_INCOMPAT_FLEX_BG 0x0200 - -#define EXT4_FEATURE_COMPAT_SUPP EXT2_FEATURE_COMPAT_EXT_ATTR -#define EXT4_FEATURE_INCOMPAT_SUPP (EXT4_FEATURE_INCOMPAT_FILETYPE| \ - EXT4_FEATURE_INCOMPAT_RECOVER| \ - EXT4_FEATURE_INCOMPAT_META_BG| \ - EXT4_FEATURE_INCOMPAT_EXTENTS| \ - EXT4_FEATURE_INCOMPAT_64BIT| \ - EXT4_FEATURE_INCOMPAT_FLEX_BG) -#define EXT4_FEATURE_RO_COMPAT_SUPP (EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER| \ - EXT4_FEATURE_RO_COMPAT_LARGE_FILE| \ - EXT4_FEATURE_RO_COMPAT_GDT_CSUM| \ - EXT4_FEATURE_RO_COMPAT_DIR_NLINK | \ - EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE | \ - EXT4_FEATURE_RO_COMPAT_BTREE_DIR |\ - EXT4_FEATURE_RO_COMPAT_HUGE_FILE) - -/* - * Default values for user and/or group using reserved blocks - */ -#define EXT4_DEF_RESUID 0 -#define EXT4_DEF_RESGID 0 - -/* - * Default mount options - */ -#define EXT4_DEFM_DEBUG 0x0001 -#define EXT4_DEFM_BSDGROUPS 0x0002 -#define EXT4_DEFM_XATTR_USER 0x0004 -#define EXT4_DEFM_ACL 0x0008 -#define EXT4_DEFM_UID16 0x0010 -#define EXT4_DEFM_JMODE 0x0060 -#define EXT4_DEFM_JMODE_DATA 0x0020 -#define EXT4_DEFM_JMODE_ORDERED 0x0040 -#define EXT4_DEFM_JMODE_WBACK 0x0060 - -/* - * Structure of a directory entry - */ -#define EXT4_NAME_LEN 255 - -struct ext4_dir_entry { - __le32 inode; /* Inode number */ - __le16 rec_len; /* Directory entry length */ - __le16 name_len; /* Name length */ - char name[EXT4_NAME_LEN]; /* File name */ -}; - -/* - * The new version of the directory entry. Since EXT4 structures are - * stored in intel byte order, and the name_len field could never be - * bigger than 255 chars, it's safe to reclaim the extra byte for the - * file_type field. - */ -struct ext4_dir_entry_2 { - __le32 inode; /* Inode number */ - __le16 rec_len; /* Directory entry length */ - __u8 name_len; /* Name length */ - __u8 file_type; - char name[EXT4_NAME_LEN]; /* File name */ -}; - -/* - * Ext4 directory file types. Only the low 3 bits are used. The - * other bits are reserved for now. - */ -#define EXT4_FT_UNKNOWN 0 -#define EXT4_FT_REG_FILE 1 -#define EXT4_FT_DIR 2 -#define EXT4_FT_CHRDEV 3 -#define EXT4_FT_BLKDEV 4 -#define EXT4_FT_FIFO 5 -#define EXT4_FT_SOCK 6 -#define EXT4_FT_SYMLINK 7 - -#define EXT4_FT_MAX 8 - -/* - * EXT4_DIR_PAD defines the directory entries boundaries - * - * NOTE: It must be a multiple of 4 - */ -#define EXT4_DIR_PAD 4 -#define EXT4_DIR_ROUND (EXT4_DIR_PAD - 1) -#define EXT4_DIR_REC_LEN(name_len) (((name_len) + 8 + EXT4_DIR_ROUND) & \ - ~EXT4_DIR_ROUND) -#define EXT4_MAX_REC_LEN ((1<<16)-1) - -static inline unsigned ext4_rec_len_from_disk(__le16 dlen) -{ - unsigned len = le16_to_cpu(dlen); - - if (len == EXT4_MAX_REC_LEN) - return 1 << 16; - return len; -} - -static inline __le16 ext4_rec_len_to_disk(unsigned len) -{ - if (len == (1 << 16)) - return cpu_to_le16(EXT4_MAX_REC_LEN); - else if (len > (1 << 16)) - BUG(); - return cpu_to_le16(len); -} - -/* - * Hash Tree Directory indexing - * (c) Daniel Phillips, 2001 - */ - -#define is_dx(dir) (EXT4_HAS_COMPAT_FEATURE(dir->i_sb, \ - EXT4_FEATURE_COMPAT_DIR_INDEX) && \ - (EXT4_I(dir)->i_flags & EXT4_INDEX_FL)) -#define EXT4_DIR_LINK_MAX(dir) (!is_dx(dir) && (dir)->i_nlink >= EXT4_LINK_MAX) -#define EXT4_DIR_LINK_EMPTY(dir) ((dir)->i_nlink == 2 || (dir)->i_nlink == 1) - -/* Legal values for the dx_root hash_version field: */ - -#define DX_HASH_LEGACY 0 -#define DX_HASH_HALF_MD4 1 -#define DX_HASH_TEA 2 - -#ifdef __KERNEL__ - -/* hash info structure used by the directory hash */ -struct dx_hash_info -{ - u32 hash; - u32 minor_hash; - int hash_version; - u32 *seed; -}; - -#define EXT4_HTREE_EOF 0x7fffffff - -/* - * Control parameters used by ext4_htree_next_block - */ -#define HASH_NB_ALWAYS 1 - - -/* - * Describe an inode's exact location on disk and in memory - */ -struct ext4_iloc -{ - struct buffer_head *bh; - unsigned long offset; - ext4_group_t block_group; -}; - -static inline struct ext4_inode *ext4_raw_inode(struct ext4_iloc *iloc) -{ - return (struct ext4_inode *) (iloc->bh->b_data + iloc->offset); -} - -/* - * This structure is stuffed into the struct file's private_data field - * for directories. It is where we put information so that we can do - * readdir operations in hash tree order. - */ -struct dir_private_info { - struct rb_root root; - struct rb_node *curr_node; - struct fname *extra_fname; - loff_t last_pos; - __u32 curr_hash; - __u32 curr_minor_hash; - __u32 next_hash; -}; - -/* calculate the first block number of the group */ -static inline ext4_fsblk_t -ext4_group_first_block_no(struct super_block *sb, ext4_group_t group_no) -{ - return group_no * (ext4_fsblk_t)EXT4_BLOCKS_PER_GROUP(sb) + - le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); -} - -/* - * Special error return code only used by dx_probe() and its callers. - */ -#define ERR_BAD_DX_DIR -75000 - -void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr, - unsigned long *blockgrpp, ext4_grpblk_t *offsetp); - -/* - * Function prototypes - */ - -/* - * Ok, these declarations are also in but none of the - * ext4 source programs needs to include it so they are duplicated here. - */ -# define NORET_TYPE /**/ -# define ATTRIB_NORET __attribute__((noreturn)) -# define NORET_AND noreturn, - -/* balloc.c */ -extern unsigned int ext4_block_group(struct super_block *sb, - ext4_fsblk_t blocknr); -extern ext4_grpblk_t ext4_block_group_offset(struct super_block *sb, - ext4_fsblk_t blocknr); -extern int ext4_bg_has_super(struct super_block *sb, ext4_group_t group); -extern unsigned long ext4_bg_num_gdb(struct super_block *sb, - ext4_group_t group); -extern ext4_fsblk_t ext4_new_block (handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, int *errp); -extern ext4_fsblk_t ext4_new_blocks (handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, unsigned long *count, int *errp); -extern ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, unsigned long *count, int *errp); -extern void ext4_free_blocks (handle_t *handle, struct inode *inode, - ext4_fsblk_t block, unsigned long count, int metadata); -extern void ext4_free_blocks_sb (handle_t *handle, struct super_block *sb, - ext4_fsblk_t block, unsigned long count, - unsigned long *pdquot_freed_blocks); -extern ext4_fsblk_t ext4_count_free_blocks (struct super_block *); -extern void ext4_check_blocks_bitmap (struct super_block *); -extern struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, - ext4_group_t block_group, - struct buffer_head ** bh); -extern int ext4_should_retry_alloc(struct super_block *sb, int *retries); -extern void ext4_init_block_alloc_info(struct inode *); -extern void ext4_rsv_window_add(struct super_block *sb, struct ext4_reserve_window_node *rsv); - -/* dir.c */ -extern int ext4_check_dir_entry(const char *, struct inode *, - struct ext4_dir_entry_2 *, - struct buffer_head *, unsigned long); -extern int ext4_htree_store_dirent(struct file *dir_file, __u32 hash, - __u32 minor_hash, - struct ext4_dir_entry_2 *dirent); -extern void ext4_htree_free_dir_info(struct dir_private_info *p); - -/* fsync.c */ -extern int ext4_sync_file (struct file *, struct dentry *, int); - -/* hash.c */ -extern int ext4fs_dirhash(const char *name, int len, struct - dx_hash_info *hinfo); - -/* ialloc.c */ -extern struct inode * ext4_new_inode (handle_t *, struct inode *, int); -extern void ext4_free_inode (handle_t *, struct inode *); -extern struct inode * ext4_orphan_get (struct super_block *, unsigned long); -extern unsigned long ext4_count_free_inodes (struct super_block *); -extern unsigned long ext4_count_dirs (struct super_block *); -extern void ext4_check_inodes_bitmap (struct super_block *); -extern unsigned long ext4_count_free (struct buffer_head *, unsigned); - -/* mballoc.c */ -extern long ext4_mb_stats; -extern long ext4_mb_max_to_scan; -extern int ext4_mb_init(struct super_block *, int); -extern int ext4_mb_release(struct super_block *); -extern ext4_fsblk_t ext4_mb_new_blocks(handle_t *, - struct ext4_allocation_request *, int *); -extern int ext4_mb_reserve_blocks(struct super_block *, int); -extern void ext4_mb_discard_inode_preallocations(struct inode *); -extern int __init init_ext4_mballoc(void); -extern void exit_ext4_mballoc(void); -extern void ext4_mb_free_blocks(handle_t *, struct inode *, - unsigned long, unsigned long, int, unsigned long *); - - -/* inode.c */ -int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode, - struct buffer_head *bh, ext4_fsblk_t blocknr); -struct buffer_head *ext4_getblk(handle_t *, struct inode *, - ext4_lblk_t, int, int *); -struct buffer_head *ext4_bread(handle_t *, struct inode *, - ext4_lblk_t, int, int *); -int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, - ext4_lblk_t iblock, unsigned long maxblocks, - struct buffer_head *bh_result, - int create, int extend_disksize); - -extern struct inode *ext4_iget(struct super_block *, unsigned long); -extern int ext4_write_inode (struct inode *, int); -extern int ext4_setattr (struct dentry *, struct iattr *); -extern void ext4_delete_inode (struct inode *); -extern int ext4_sync_inode (handle_t *, struct inode *); -extern void ext4_discard_reservation (struct inode *); -extern void ext4_dirty_inode(struct inode *); -extern int ext4_change_inode_journal_flag(struct inode *, int); -extern int ext4_get_inode_loc(struct inode *, struct ext4_iloc *); -extern void ext4_truncate (struct inode *); -extern void ext4_set_inode_flags(struct inode *); -extern void ext4_get_inode_flags(struct ext4_inode_info *); -extern void ext4_set_aops(struct inode *inode); -extern int ext4_writepage_trans_blocks(struct inode *); -extern int ext4_block_truncate_page(handle_t *handle, struct page *page, - struct address_space *mapping, loff_t from); - -/* ioctl.c */ -extern long ext4_ioctl(struct file *, unsigned int, unsigned long); -extern long ext4_compat_ioctl (struct file *, unsigned int, unsigned long); - -/* migrate.c */ -extern int ext4_ext_migrate(struct inode *, struct file *, unsigned int, - unsigned long); -/* namei.c */ -extern int ext4_orphan_add(handle_t *, struct inode *); -extern int ext4_orphan_del(handle_t *, struct inode *); -extern int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash, - __u32 start_minor_hash, __u32 *next_hash); - -/* resize.c */ -extern int ext4_group_add(struct super_block *sb, - struct ext4_new_group_data *input); -extern int ext4_group_extend(struct super_block *sb, - struct ext4_super_block *es, - ext4_fsblk_t n_blocks_count); - -/* super.c */ -extern void ext4_error (struct super_block *, const char *, const char *, ...) - __attribute__ ((format (printf, 3, 4))); -extern void __ext4_std_error (struct super_block *, const char *, int); -extern void ext4_abort (struct super_block *, const char *, const char *, ...) - __attribute__ ((format (printf, 3, 4))); -extern void ext4_warning (struct super_block *, const char *, const char *, ...) - __attribute__ ((format (printf, 3, 4))); -extern void ext4_update_dynamic_rev (struct super_block *sb); -extern int ext4_update_compat_feature(handle_t *handle, struct super_block *sb, - __u32 compat); -extern int ext4_update_rocompat_feature(handle_t *handle, - struct super_block *sb, __u32 rocompat); -extern int ext4_update_incompat_feature(handle_t *handle, - struct super_block *sb, __u32 incompat); -extern ext4_fsblk_t ext4_block_bitmap(struct super_block *sb, - struct ext4_group_desc *bg); -extern ext4_fsblk_t ext4_inode_bitmap(struct super_block *sb, - struct ext4_group_desc *bg); -extern ext4_fsblk_t ext4_inode_table(struct super_block *sb, - struct ext4_group_desc *bg); -extern void ext4_block_bitmap_set(struct super_block *sb, - struct ext4_group_desc *bg, ext4_fsblk_t blk); -extern void ext4_inode_bitmap_set(struct super_block *sb, - struct ext4_group_desc *bg, ext4_fsblk_t blk); -extern void ext4_inode_table_set(struct super_block *sb, - struct ext4_group_desc *bg, ext4_fsblk_t blk); - -static inline ext4_fsblk_t ext4_blocks_count(struct ext4_super_block *es) -{ - return ((ext4_fsblk_t)le32_to_cpu(es->s_blocks_count_hi) << 32) | - le32_to_cpu(es->s_blocks_count_lo); -} - -static inline ext4_fsblk_t ext4_r_blocks_count(struct ext4_super_block *es) -{ - return ((ext4_fsblk_t)le32_to_cpu(es->s_r_blocks_count_hi) << 32) | - le32_to_cpu(es->s_r_blocks_count_lo); -} - -static inline ext4_fsblk_t ext4_free_blocks_count(struct ext4_super_block *es) -{ - return ((ext4_fsblk_t)le32_to_cpu(es->s_free_blocks_count_hi) << 32) | - le32_to_cpu(es->s_free_blocks_count_lo); -} - -static inline void ext4_blocks_count_set(struct ext4_super_block *es, - ext4_fsblk_t blk) -{ - es->s_blocks_count_lo = cpu_to_le32((u32)blk); - es->s_blocks_count_hi = cpu_to_le32(blk >> 32); -} - -static inline void ext4_free_blocks_count_set(struct ext4_super_block *es, - ext4_fsblk_t blk) -{ - es->s_free_blocks_count_lo = cpu_to_le32((u32)blk); - es->s_free_blocks_count_hi = cpu_to_le32(blk >> 32); -} - -static inline void ext4_r_blocks_count_set(struct ext4_super_block *es, - ext4_fsblk_t blk) -{ - es->s_r_blocks_count_lo = cpu_to_le32((u32)blk); - es->s_r_blocks_count_hi = cpu_to_le32(blk >> 32); -} - -static inline loff_t ext4_isize(struct ext4_inode *raw_inode) -{ - return ((loff_t)le32_to_cpu(raw_inode->i_size_high) << 32) | - le32_to_cpu(raw_inode->i_size_lo); -} - -static inline void ext4_isize_set(struct ext4_inode *raw_inode, loff_t i_size) -{ - raw_inode->i_size_lo = cpu_to_le32(i_size); - raw_inode->i_size_high = cpu_to_le32(i_size >> 32); -} - -static inline -struct ext4_group_info *ext4_get_group_info(struct super_block *sb, - ext4_group_t group) -{ - struct ext4_group_info ***grp_info; - long indexv, indexh; - grp_info = EXT4_SB(sb)->s_group_info; - indexv = group >> (EXT4_DESC_PER_BLOCK_BITS(sb)); - indexh = group & ((EXT4_DESC_PER_BLOCK(sb)) - 1); - return grp_info[indexv][indexh]; -} - - -#define ext4_std_error(sb, errno) \ -do { \ - if ((errno)) \ - __ext4_std_error((sb), __FUNCTION__, (errno)); \ -} while (0) - -/* - * Inodes and files operations - */ - -/* dir.c */ -extern const struct file_operations ext4_dir_operations; - -/* file.c */ -extern const struct inode_operations ext4_file_inode_operations; -extern const struct file_operations ext4_file_operations; - -/* namei.c */ -extern const struct inode_operations ext4_dir_inode_operations; -extern const struct inode_operations ext4_special_inode_operations; - -/* symlink.c */ -extern const struct inode_operations ext4_symlink_inode_operations; -extern const struct inode_operations ext4_fast_symlink_inode_operations; - -/* extents.c */ -extern int ext4_ext_tree_init(handle_t *handle, struct inode *); -extern int ext4_ext_writepage_trans_blocks(struct inode *, int); -extern int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, - ext4_lblk_t iblock, - unsigned long max_blocks, struct buffer_head *bh_result, - int create, int extend_disksize); -extern void ext4_ext_truncate(struct inode *, struct page *); -extern void ext4_ext_init(struct super_block *); -extern void ext4_ext_release(struct super_block *); -extern long ext4_fallocate(struct inode *inode, int mode, loff_t offset, - loff_t len); -extern int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, - sector_t block, unsigned long max_blocks, - struct buffer_head *bh, int create, - int extend_disksize); -#endif /* __KERNEL__ */ - -#endif /* _LINUX_EXT4_FS_H */ diff --git a/include/linux/ext4_fs_extents.h b/include/linux/ext4_fs_extents.h deleted file mode 100644 index 1285c583b2d8..000000000000 --- a/include/linux/ext4_fs_extents.h +++ /dev/null @@ -1,232 +0,0 @@ -/* - * Copyright (c) 2003-2006, Cluster File Systems, Inc, info@clusterfs.com - * Written by Alex Tomas - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public Licens - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111- - */ - -#ifndef _LINUX_EXT4_EXTENTS -#define _LINUX_EXT4_EXTENTS - -#include - -/* - * With AGGRESSIVE_TEST defined, the capacity of index/leaf blocks - * becomes very small, so index split, in-depth growing and - * other hard changes happen much more often. - * This is for debug purposes only. - */ -#define AGGRESSIVE_TEST_ - -/* - * With EXTENTS_STATS defined, the number of blocks and extents - * are collected in the truncate path. They'll be shown at - * umount time. - */ -#define EXTENTS_STATS__ - -/* - * If CHECK_BINSEARCH is defined, then the results of the binary search - * will also be checked by linear search. - */ -#define CHECK_BINSEARCH__ - -/* - * If EXT_DEBUG is defined you can use the 'extdebug' mount option - * to get lots of info about what's going on. - */ -#define EXT_DEBUG__ -#ifdef EXT_DEBUG -#define ext_debug(a...) printk(a) -#else -#define ext_debug(a...) -#endif - -/* - * If EXT_STATS is defined then stats numbers are collected. - * These number will be displayed at umount time. - */ -#define EXT_STATS_ - - -/* - * ext4_inode has i_block array (60 bytes total). - * The first 12 bytes store ext4_extent_header; - * the remainder stores an array of ext4_extent. - */ - -/* - * This is the extent on-disk structure. - * It's used at the bottom of the tree. - */ -struct ext4_extent { - __le32 ee_block; /* first logical block extent covers */ - __le16 ee_len; /* number of blocks covered by extent */ - __le16 ee_start_hi; /* high 16 bits of physical block */ - __le32 ee_start_lo; /* low 32 bits of physical block */ -}; - -/* - * This is index on-disk structure. - * It's used at all the levels except the bottom. - */ -struct ext4_extent_idx { - __le32 ei_block; /* index covers logical blocks from 'block' */ - __le32 ei_leaf_lo; /* pointer to the physical block of the next * - * level. leaf or next index could be there */ - __le16 ei_leaf_hi; /* high 16 bits of physical block */ - __u16 ei_unused; -}; - -/* - * Each block (leaves and indexes), even inode-stored has header. - */ -struct ext4_extent_header { - __le16 eh_magic; /* probably will support different formats */ - __le16 eh_entries; /* number of valid entries */ - __le16 eh_max; /* capacity of store in entries */ - __le16 eh_depth; /* has tree real underlying blocks? */ - __le32 eh_generation; /* generation of the tree */ -}; - -#define EXT4_EXT_MAGIC cpu_to_le16(0xf30a) - -/* - * Array of ext4_ext_path contains path to some extent. - * Creation/lookup routines use it for traversal/splitting/etc. - * Truncate uses it to simulate recursive walking. - */ -struct ext4_ext_path { - ext4_fsblk_t p_block; - __u16 p_depth; - struct ext4_extent *p_ext; - struct ext4_extent_idx *p_idx; - struct ext4_extent_header *p_hdr; - struct buffer_head *p_bh; -}; - -/* - * structure for external API - */ - -#define EXT4_EXT_CACHE_NO 0 -#define EXT4_EXT_CACHE_GAP 1 -#define EXT4_EXT_CACHE_EXTENT 2 - - -#define EXT_MAX_BLOCK 0xffffffff - -/* - * EXT_INIT_MAX_LEN is the maximum number of blocks we can have in an - * initialized extent. This is 2^15 and not (2^16 - 1), since we use the - * MSB of ee_len field in the extent datastructure to signify if this - * particular extent is an initialized extent or an uninitialized (i.e. - * preallocated). - * EXT_UNINIT_MAX_LEN is the maximum number of blocks we can have in an - * uninitialized extent. - * If ee_len is <= 0x8000, it is an initialized extent. Otherwise, it is an - * uninitialized one. In other words, if MSB of ee_len is set, it is an - * uninitialized extent with only one special scenario when ee_len = 0x8000. - * In this case we can not have an uninitialized extent of zero length and - * thus we make it as a special case of initialized extent with 0x8000 length. - * This way we get better extent-to-group alignment for initialized extents. - * Hence, the maximum number of blocks we can have in an *initialized* - * extent is 2^15 (32768) and in an *uninitialized* extent is 2^15-1 (32767). - */ -#define EXT_INIT_MAX_LEN (1UL << 15) -#define EXT_UNINIT_MAX_LEN (EXT_INIT_MAX_LEN - 1) - - -#define EXT_FIRST_EXTENT(__hdr__) \ - ((struct ext4_extent *) (((char *) (__hdr__)) + \ - sizeof(struct ext4_extent_header))) -#define EXT_FIRST_INDEX(__hdr__) \ - ((struct ext4_extent_idx *) (((char *) (__hdr__)) + \ - sizeof(struct ext4_extent_header))) -#define EXT_HAS_FREE_INDEX(__path__) \ - (le16_to_cpu((__path__)->p_hdr->eh_entries) \ - < le16_to_cpu((__path__)->p_hdr->eh_max)) -#define EXT_LAST_EXTENT(__hdr__) \ - (EXT_FIRST_EXTENT((__hdr__)) + le16_to_cpu((__hdr__)->eh_entries) - 1) -#define EXT_LAST_INDEX(__hdr__) \ - (EXT_FIRST_INDEX((__hdr__)) + le16_to_cpu((__hdr__)->eh_entries) - 1) -#define EXT_MAX_EXTENT(__hdr__) \ - (EXT_FIRST_EXTENT((__hdr__)) + le16_to_cpu((__hdr__)->eh_max) - 1) -#define EXT_MAX_INDEX(__hdr__) \ - (EXT_FIRST_INDEX((__hdr__)) + le16_to_cpu((__hdr__)->eh_max) - 1) - -static inline struct ext4_extent_header *ext_inode_hdr(struct inode *inode) -{ - return (struct ext4_extent_header *) EXT4_I(inode)->i_data; -} - -static inline struct ext4_extent_header *ext_block_hdr(struct buffer_head *bh) -{ - return (struct ext4_extent_header *) bh->b_data; -} - -static inline unsigned short ext_depth(struct inode *inode) -{ - return le16_to_cpu(ext_inode_hdr(inode)->eh_depth); -} - -static inline void ext4_ext_tree_changed(struct inode *inode) -{ - EXT4_I(inode)->i_ext_generation++; -} - -static inline void -ext4_ext_invalidate_cache(struct inode *inode) -{ - EXT4_I(inode)->i_cached_extent.ec_type = EXT4_EXT_CACHE_NO; -} - -static inline void ext4_ext_mark_uninitialized(struct ext4_extent *ext) -{ - /* We can not have an uninitialized extent of zero length! */ - BUG_ON((le16_to_cpu(ext->ee_len) & ~EXT_INIT_MAX_LEN) == 0); - ext->ee_len |= cpu_to_le16(EXT_INIT_MAX_LEN); -} - -static inline int ext4_ext_is_uninitialized(struct ext4_extent *ext) -{ - /* Extent with ee_len of 0x8000 is treated as an initialized extent */ - return (le16_to_cpu(ext->ee_len) > EXT_INIT_MAX_LEN); -} - -static inline int ext4_ext_get_actual_len(struct ext4_extent *ext) -{ - return (le16_to_cpu(ext->ee_len) <= EXT_INIT_MAX_LEN ? - le16_to_cpu(ext->ee_len) : - (le16_to_cpu(ext->ee_len) - EXT_INIT_MAX_LEN)); -} - -extern ext4_fsblk_t idx_pblock(struct ext4_extent_idx *); -extern void ext4_ext_store_pblock(struct ext4_extent *, ext4_fsblk_t); -extern int ext4_extent_tree_init(handle_t *, struct inode *); -extern int ext4_ext_calc_credits_for_insert(struct inode *, struct ext4_ext_path *); -extern int ext4_ext_try_to_merge(struct inode *inode, - struct ext4_ext_path *path, - struct ext4_extent *); -extern unsigned int ext4_ext_check_overlap(struct inode *, struct ext4_extent *, struct ext4_ext_path *); -extern int ext4_ext_insert_extent(handle_t *, struct inode *, struct ext4_ext_path *, struct ext4_extent *); -extern struct ext4_ext_path *ext4_ext_find_extent(struct inode *, ext4_lblk_t, - struct ext4_ext_path *); -extern int ext4_ext_search_left(struct inode *, struct ext4_ext_path *, - ext4_lblk_t *, ext4_fsblk_t *); -extern int ext4_ext_search_right(struct inode *, struct ext4_ext_path *, - ext4_lblk_t *, ext4_fsblk_t *); -extern void ext4_ext_drop_refs(struct ext4_ext_path *); -#endif /* _LINUX_EXT4_EXTENTS */ - diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h deleted file mode 100644 index d5508d3cf290..000000000000 --- a/include/linux/ext4_fs_i.h +++ /dev/null @@ -1,167 +0,0 @@ -/* - * linux/include/linux/ext4_fs_i.h - * - * Copyright (C) 1992, 1993, 1994, 1995 - * Remy Card (card@masi.ibp.fr) - * Laboratoire MASI - Institut Blaise Pascal - * Universite Pierre et Marie Curie (Paris VI) - * - * from - * - * linux/include/linux/minix_fs_i.h - * - * Copyright (C) 1991, 1992 Linus Torvalds - */ - -#ifndef _LINUX_EXT4_FS_I -#define _LINUX_EXT4_FS_I - -#include -#include -#include -#include - -/* data type for block offset of block group */ -typedef int ext4_grpblk_t; - -/* data type for filesystem-wide blocks number */ -typedef unsigned long long ext4_fsblk_t; - -/* data type for file logical block number */ -typedef __u32 ext4_lblk_t; - -/* data type for block group number */ -typedef unsigned long ext4_group_t; - -struct ext4_reserve_window { - ext4_fsblk_t _rsv_start; /* First byte reserved */ - ext4_fsblk_t _rsv_end; /* Last byte reserved or 0 */ -}; - -struct ext4_reserve_window_node { - struct rb_node rsv_node; - __u32 rsv_goal_size; - __u32 rsv_alloc_hit; - struct ext4_reserve_window rsv_window; -}; - -struct ext4_block_alloc_info { - /* information about reservation window */ - struct ext4_reserve_window_node rsv_window_node; - /* - * was i_next_alloc_block in ext4_inode_info - * is the logical (file-relative) number of the - * most-recently-allocated block in this file. - * We use this for detecting linearly ascending allocation requests. - */ - ext4_lblk_t last_alloc_logical_block; - /* - * Was i_next_alloc_goal in ext4_inode_info - * is the *physical* companion to i_next_alloc_block. - * it the physical block number of the block which was most-recentl - * allocated to this file. This give us the goal (target) for the next - * allocation when we detect linearly ascending requests. - */ - ext4_fsblk_t last_alloc_physical_block; -}; - -#define rsv_start rsv_window._rsv_start -#define rsv_end rsv_window._rsv_end - -/* - * storage for cached extent - */ -struct ext4_ext_cache { - ext4_fsblk_t ec_start; - ext4_lblk_t ec_block; - __u32 ec_len; /* must be 32bit to return holes */ - __u32 ec_type; -}; - -/* - * third extended file system inode data in memory - */ -struct ext4_inode_info { - __le32 i_data[15]; /* unconverted */ - __u32 i_flags; - ext4_fsblk_t i_file_acl; - __u32 i_dtime; - - /* - * i_block_group is the number of the block group which contains - * this file's inode. Constant across the lifetime of the inode, - * it is ued for making block allocation decisions - we try to - * place a file's data blocks near its inode block, and new inodes - * near to their parent directory's inode. - */ - ext4_group_t i_block_group; - __u32 i_state; /* Dynamic state flags for ext4 */ - - /* block reservation info */ - struct ext4_block_alloc_info *i_block_alloc_info; - - ext4_lblk_t i_dir_start_lookup; -#ifdef CONFIG_EXT4DEV_FS_XATTR - /* - * Extended attributes can be read independently of the main file - * data. Taking i_mutex even when reading would cause contention - * between readers of EAs and writers of regular file data, so - * instead we synchronize on xattr_sem when reading or changing - * EAs. - */ - struct rw_semaphore xattr_sem; -#endif -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL - struct posix_acl *i_acl; - struct posix_acl *i_default_acl; -#endif - - struct list_head i_orphan; /* unlinked but open inodes */ - - /* - * i_disksize keeps track of what the inode size is ON DISK, not - * in memory. During truncate, i_size is set to the new size by - * the VFS prior to calling ext4_truncate(), but the filesystem won't - * set i_disksize to 0 until the truncate is actually under way. - * - * The intent is that i_disksize always represents the blocks which - * are used by this file. This allows recovery to restart truncate - * on orphans if we crash during truncate. We actually write i_disksize - * into the on-disk inode when writing inodes out, instead of i_size. - * - * The only time when i_disksize and i_size may be different is when - * a truncate is in progress. The only things which change i_disksize - * are ext4_get_block (growth) and ext4_truncate (shrinkth). - */ - loff_t i_disksize; - - /* on-disk additional length */ - __u16 i_extra_isize; - - /* - * i_data_sem is for serialising ext4_truncate() against - * ext4_getblock(). In the 2.4 ext2 design, great chunks of inode's - * data tree are chopped off during truncate. We can't do that in - * ext4 because whenever we perform intermediate commits during - * truncate, the inode and all the metadata blocks *must* be in a - * consistent state which allows truncation of the orphans to restart - * during recovery. Hence we must fix the get_block-vs-truncate race - * by other means, so we have i_data_sem. - */ - struct rw_semaphore i_data_sem; - struct inode vfs_inode; - - unsigned long i_ext_generation; - struct ext4_ext_cache i_cached_extent; - /* - * File creation time. Its function is same as that of - * struct timespec i_{a,c,m}time in the generic inode. - */ - struct timespec i_crtime; - - /* mballoc */ - struct list_head i_prealloc_list; - spinlock_t i_prealloc_lock; -}; - -#endif /* _LINUX_EXT4_FS_I */ diff --git a/include/linux/ext4_fs_sb.h b/include/linux/ext4_fs_sb.h deleted file mode 100644 index abaae2c8cccf..000000000000 --- a/include/linux/ext4_fs_sb.h +++ /dev/null @@ -1,148 +0,0 @@ -/* - * linux/include/linux/ext4_fs_sb.h - * - * Copyright (C) 1992, 1993, 1994, 1995 - * Remy Card (card@masi.ibp.fr) - * Laboratoire MASI - Institut Blaise Pascal - * Universite Pierre et Marie Curie (Paris VI) - * - * from - * - * linux/include/linux/minix_fs_sb.h - * - * Copyright (C) 1991, 1992 Linus Torvalds - */ - -#ifndef _LINUX_EXT4_FS_SB -#define _LINUX_EXT4_FS_SB - -#ifdef __KERNEL__ -#include -#include -#include -#include -#endif -#include - -/* - * third extended-fs super-block data in memory - */ -struct ext4_sb_info { - unsigned long s_desc_size; /* Size of a group descriptor in bytes */ - unsigned long s_inodes_per_block;/* Number of inodes per block */ - unsigned long s_blocks_per_group;/* Number of blocks in a group */ - unsigned long s_inodes_per_group;/* Number of inodes in a group */ - unsigned long s_itb_per_group; /* Number of inode table blocks per group */ - unsigned long s_gdb_count; /* Number of group descriptor blocks */ - unsigned long s_desc_per_block; /* Number of group descriptors per block */ - ext4_group_t s_groups_count; /* Number of groups in the fs */ - unsigned long s_overhead_last; /* Last calculated overhead */ - unsigned long s_blocks_last; /* Last seen block count */ - loff_t s_bitmap_maxbytes; /* max bytes for bitmap files */ - struct buffer_head * s_sbh; /* Buffer containing the super block */ - struct ext4_super_block * s_es; /* Pointer to the super block in the buffer */ - struct buffer_head ** s_group_desc; - unsigned long s_mount_opt; - ext4_fsblk_t s_sb_block; - uid_t s_resuid; - gid_t s_resgid; - unsigned short s_mount_state; - unsigned short s_pad; - int s_addr_per_block_bits; - int s_desc_per_block_bits; - int s_inode_size; - int s_first_ino; - spinlock_t s_next_gen_lock; - u32 s_next_generation; - u32 s_hash_seed[4]; - int s_def_hash_version; - struct percpu_counter s_freeblocks_counter; - struct percpu_counter s_freeinodes_counter; - struct percpu_counter s_dirs_counter; - struct blockgroup_lock s_blockgroup_lock; - - /* root of the per fs reservation window tree */ - spinlock_t s_rsv_window_lock; - struct rb_root s_rsv_window_root; - struct ext4_reserve_window_node s_rsv_window_head; - - /* Journaling */ - struct inode * s_journal_inode; - struct journal_s * s_journal; - struct list_head s_orphan; - unsigned long s_commit_interval; - struct block_device *journal_bdev; -#ifdef CONFIG_JBD2_DEBUG - struct timer_list turn_ro_timer; /* For turning read-only (crash simulation) */ - wait_queue_head_t ro_wait_queue; /* For people waiting for the fs to go read-only */ -#endif -#ifdef CONFIG_QUOTA - char *s_qf_names[MAXQUOTAS]; /* Names of quota files with journalled quota */ - int s_jquota_fmt; /* Format of quota to use */ -#endif - unsigned int s_want_extra_isize; /* New inodes should reserve # bytes */ - -#ifdef EXTENTS_STATS - /* ext4 extents stats */ - unsigned long s_ext_min; - unsigned long s_ext_max; - unsigned long s_depth_max; - spinlock_t s_ext_stats_lock; - unsigned long s_ext_blocks; - unsigned long s_ext_extents; -#endif - - /* for buddy allocator */ - struct ext4_group_info ***s_group_info; - struct inode *s_buddy_cache; - long s_blocks_reserved; - spinlock_t s_reserve_lock; - struct list_head s_active_transaction; - struct list_head s_closed_transaction; - struct list_head s_committed_transaction; - spinlock_t s_md_lock; - tid_t s_last_transaction; - unsigned short *s_mb_offsets, *s_mb_maxs; - - /* tunables */ - unsigned long s_stripe; - unsigned long s_mb_stream_request; - unsigned long s_mb_max_to_scan; - unsigned long s_mb_min_to_scan; - unsigned long s_mb_stats; - unsigned long s_mb_order2_reqs; - unsigned long s_mb_group_prealloc; - /* where last allocation was done - for stream allocation */ - unsigned long s_mb_last_group; - unsigned long s_mb_last_start; - - /* history to debug policy */ - struct ext4_mb_history *s_mb_history; - int s_mb_history_cur; - int s_mb_history_max; - int s_mb_history_num; - struct proc_dir_entry *s_mb_proc; - spinlock_t s_mb_history_lock; - int s_mb_history_filter; - - /* stats for buddy allocator */ - spinlock_t s_mb_pa_lock; - atomic_t s_bal_reqs; /* number of reqs with len > 1 */ - atomic_t s_bal_success; /* we found long enough chunks */ - atomic_t s_bal_allocated; /* in blocks */ - atomic_t s_bal_ex_scanned; /* total extents scanned */ - atomic_t s_bal_goals; /* goal hits */ - atomic_t s_bal_breaks; /* too long searches */ - atomic_t s_bal_2orders; /* 2^order hits */ - spinlock_t s_bal_lock; - unsigned long s_mb_buddies_generated; - unsigned long long s_mb_generation_time; - atomic_t s_mb_lost_chunks; - atomic_t s_mb_preallocated; - atomic_t s_mb_discarded; - - /* locality groups */ - struct ext4_locality_group *s_locality_groups; -}; - -#endif /* _LINUX_EXT4_FS_SB */ diff --git a/include/linux/ext4_jbd2.h b/include/linux/ext4_jbd2.h deleted file mode 100644 index 38c71d3c8dbf..000000000000 --- a/include/linux/ext4_jbd2.h +++ /dev/null @@ -1,231 +0,0 @@ -/* - * linux/include/linux/ext4_jbd2.h - * - * Written by Stephen C. Tweedie , 1999 - * - * Copyright 1998--1999 Red Hat corp --- All Rights Reserved - * - * This file is part of the Linux kernel and is made available under - * the terms of the GNU General Public License, version 2, or at your - * option, any later version, incorporated herein by reference. - * - * Ext4-specific journaling extensions. - */ - -#ifndef _LINUX_EXT4_JBD2_H -#define _LINUX_EXT4_JBD2_H - -#include -#include -#include - -#define EXT4_JOURNAL(inode) (EXT4_SB((inode)->i_sb)->s_journal) - -/* Define the number of blocks we need to account to a transaction to - * modify one block of data. - * - * We may have to touch one inode, one bitmap buffer, up to three - * indirection blocks, the group and superblock summaries, and the data - * block to complete the transaction. - * - * For extents-enabled fs we may have to allocate and modify up to - * 5 levels of tree + root which are stored in the inode. */ - -#define EXT4_SINGLEDATA_TRANS_BLOCKS(sb) \ - (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS) \ - || test_opt(sb, EXTENTS) ? 27U : 8U) - -/* Extended attribute operations touch at most two data buffers, - * two bitmap buffers, and two group summaries, in addition to the inode - * and the superblock, which are already accounted for. */ - -#define EXT4_XATTR_TRANS_BLOCKS 6U - -/* Define the minimum size for a transaction which modifies data. This - * needs to take into account the fact that we may end up modifying two - * quota files too (one for the group, one for the user quota). The - * superblock only gets updated once, of course, so don't bother - * counting that again for the quota updates. */ - -#define EXT4_DATA_TRANS_BLOCKS(sb) (EXT4_SINGLEDATA_TRANS_BLOCKS(sb) + \ - EXT4_XATTR_TRANS_BLOCKS - 2 + \ - 2*EXT4_QUOTA_TRANS_BLOCKS(sb)) - -/* Delete operations potentially hit one directory's namespace plus an - * entire inode, plus arbitrary amounts of bitmap/indirection data. Be - * generous. We can grow the delete transaction later if necessary. */ - -#define EXT4_DELETE_TRANS_BLOCKS(sb) (2 * EXT4_DATA_TRANS_BLOCKS(sb) + 64) - -/* Define an arbitrary limit for the amount of data we will anticipate - * writing to any given transaction. For unbounded transactions such as - * write(2) and truncate(2) we can write more than this, but we always - * start off at the maximum transaction size and grow the transaction - * optimistically as we go. */ - -#define EXT4_MAX_TRANS_DATA 64U - -/* We break up a large truncate or write transaction once the handle's - * buffer credits gets this low, we need either to extend the - * transaction or to start a new one. Reserve enough space here for - * inode, bitmap, superblock, group and indirection updates for at least - * one block, plus two quota updates. Quota allocations are not - * needed. */ - -#define EXT4_RESERVE_TRANS_BLOCKS 12U - -#define EXT4_INDEX_EXTRA_TRANS_BLOCKS 8 - -#ifdef CONFIG_QUOTA -/* Amount of blocks needed for quota update - we know that the structure was - * allocated so we need to update only inode+data */ -#define EXT4_QUOTA_TRANS_BLOCKS(sb) (test_opt(sb, QUOTA) ? 2 : 0) -/* Amount of blocks needed for quota insert/delete - we do some block writes - * but inode, sb and group updates are done only once */ -#define EXT4_QUOTA_INIT_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_INIT_ALLOC*\ - (EXT4_SINGLEDATA_TRANS_BLOCKS(sb)-3)+3+DQUOT_INIT_REWRITE) : 0) -#define EXT4_QUOTA_DEL_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_DEL_ALLOC*\ - (EXT4_SINGLEDATA_TRANS_BLOCKS(sb)-3)+3+DQUOT_DEL_REWRITE) : 0) -#else -#define EXT4_QUOTA_TRANS_BLOCKS(sb) 0 -#define EXT4_QUOTA_INIT_BLOCKS(sb) 0 -#define EXT4_QUOTA_DEL_BLOCKS(sb) 0 -#endif - -int -ext4_mark_iloc_dirty(handle_t *handle, - struct inode *inode, - struct ext4_iloc *iloc); - -/* - * On success, We end up with an outstanding reference count against - * iloc->bh. This _must_ be cleaned up later. - */ - -int ext4_reserve_inode_write(handle_t *handle, struct inode *inode, - struct ext4_iloc *iloc); - -int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode); - -/* - * Wrapper functions with which ext4 calls into JBD. The intent here is - * to allow these to be turned into appropriate stubs so ext4 can control - * ext2 filesystems, so ext2+ext4 systems only nee one fs. This work hasn't - * been done yet. - */ - -static inline void ext4_journal_release_buffer(handle_t *handle, - struct buffer_head *bh) -{ - jbd2_journal_release_buffer(handle, bh); -} - -void ext4_journal_abort_handle(const char *caller, const char *err_fn, - struct buffer_head *bh, handle_t *handle, int err); - -int __ext4_journal_get_undo_access(const char *where, handle_t *handle, - struct buffer_head *bh); - -int __ext4_journal_get_write_access(const char *where, handle_t *handle, - struct buffer_head *bh); - -int __ext4_journal_forget(const char *where, handle_t *handle, - struct buffer_head *bh); - -int __ext4_journal_revoke(const char *where, handle_t *handle, - ext4_fsblk_t blocknr, struct buffer_head *bh); - -int __ext4_journal_get_create_access(const char *where, - handle_t *handle, struct buffer_head *bh); - -int __ext4_journal_dirty_metadata(const char *where, - handle_t *handle, struct buffer_head *bh); - -#define ext4_journal_get_undo_access(handle, bh) \ - __ext4_journal_get_undo_access(__FUNCTION__, (handle), (bh)) -#define ext4_journal_get_write_access(handle, bh) \ - __ext4_journal_get_write_access(__FUNCTION__, (handle), (bh)) -#define ext4_journal_revoke(handle, blocknr, bh) \ - __ext4_journal_revoke(__FUNCTION__, (handle), (blocknr), (bh)) -#define ext4_journal_get_create_access(handle, bh) \ - __ext4_journal_get_create_access(__FUNCTION__, (handle), (bh)) -#define ext4_journal_dirty_metadata(handle, bh) \ - __ext4_journal_dirty_metadata(__FUNCTION__, (handle), (bh)) -#define ext4_journal_forget(handle, bh) \ - __ext4_journal_forget(__FUNCTION__, (handle), (bh)) - -int ext4_journal_dirty_data(handle_t *handle, struct buffer_head *bh); - -handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks); -int __ext4_journal_stop(const char *where, handle_t *handle); - -static inline handle_t *ext4_journal_start(struct inode *inode, int nblocks) -{ - return ext4_journal_start_sb(inode->i_sb, nblocks); -} - -#define ext4_journal_stop(handle) \ - __ext4_journal_stop(__FUNCTION__, (handle)) - -static inline handle_t *ext4_journal_current_handle(void) -{ - return journal_current_handle(); -} - -static inline int ext4_journal_extend(handle_t *handle, int nblocks) -{ - return jbd2_journal_extend(handle, nblocks); -} - -static inline int ext4_journal_restart(handle_t *handle, int nblocks) -{ - return jbd2_journal_restart(handle, nblocks); -} - -static inline int ext4_journal_blocks_per_page(struct inode *inode) -{ - return jbd2_journal_blocks_per_page(inode); -} - -static inline int ext4_journal_force_commit(journal_t *journal) -{ - return jbd2_journal_force_commit(journal); -} - -/* super.c */ -int ext4_force_commit(struct super_block *sb); - -static inline int ext4_should_journal_data(struct inode *inode) -{ - if (!S_ISREG(inode->i_mode)) - return 1; - if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) - return 1; - if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) - return 1; - return 0; -} - -static inline int ext4_should_order_data(struct inode *inode) -{ - if (!S_ISREG(inode->i_mode)) - return 0; - if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) - return 0; - if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA) - return 1; - return 0; -} - -static inline int ext4_should_writeback_data(struct inode *inode) -{ - if (!S_ISREG(inode->i_mode)) - return 0; - if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) - return 0; - if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA) - return 1; - return 0; -} - -#endif /* _LINUX_EXT4_JBD2_H */ -- cgit v1.2.3 From c83617db76353ff30e825874be2c15c185b95759 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Tue, 29 Apr 2008 22:00:47 -0400 Subject: ext4: Don't do GFP_NOFS allocations after taking ext4_lock_group We can't do GFP_NOFS allocation after taking ext4_lock_group BUG: sleeping function called from invalid context at mm/slab.c:3054 in_atomic():1, irqs_disabled():0 1 lock held by vi/2426: #0: (&ei->i_data_sem){----}, at: [] ext4_release_file+0x23/0x66 Pid: 2426, comm: vi Not tainted 2.6.25-rc7 #24 [] __might_sleep+0xbe/0xc5 [] kmem_cache_alloc+0x22/0xa6 [] ext4_mb_release_inode_pa+0x73/0x1b3 [] ext4_mb_discard_inode_preallocations+0x22d/0x2d4 [] ? param_set_ushort+0x32/0x39 [] ext4_discard_reservation+0x27/0x6a [] ext4_release_file+0x2a/0x66 [] __fput+0xae/0x155 [] fput+0x17/0x19 [] filp_close+0x50/0x5a [] sys_close+0x71/0xad [] sysenter_past_esp+0x5f/0xa5 Signed-off-by: Aneesh Kumar K.V Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/mballoc.c | 29 +++++++++++++++-------------- 1 file changed, 15 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index f87471de3af7..d2f0b9661fb9 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -3730,9 +3730,9 @@ static int ext4_mb_new_preallocation(struct ext4_allocation_context *ac) */ static noinline_for_stack int ext4_mb_release_inode_pa(struct ext4_buddy *e4b, struct buffer_head *bitmap_bh, - struct ext4_prealloc_space *pa) + struct ext4_prealloc_space *pa, + struct ext4_allocation_context *ac) { - struct ext4_allocation_context *ac; struct super_block *sb = e4b->bd_sb; struct ext4_sb_info *sbi = EXT4_SB(sb); unsigned long end; @@ -3748,8 +3748,6 @@ ext4_mb_release_inode_pa(struct ext4_buddy *e4b, struct buffer_head *bitmap_bh, BUG_ON(group != e4b->bd_group && pa->pa_len != 0); end = bit + pa->pa_len; - ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS); - if (ac) { ac->ac_sb = sb; ac->ac_inode = pa->pa_inode; @@ -3794,23 +3792,19 @@ ext4_mb_release_inode_pa(struct ext4_buddy *e4b, struct buffer_head *bitmap_bh, */ } atomic_add(free, &sbi->s_mb_discarded); - if (ac) - kmem_cache_free(ext4_ac_cachep, ac); return err; } static noinline_for_stack int ext4_mb_release_group_pa(struct ext4_buddy *e4b, - struct ext4_prealloc_space *pa) + struct ext4_prealloc_space *pa, + struct ext4_allocation_context *ac) { - struct ext4_allocation_context *ac; struct super_block *sb = e4b->bd_sb; ext4_group_t group; ext4_grpblk_t bit; - ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS); - if (ac) ac->ac_op = EXT4_MB_HISTORY_DISCARD; @@ -3828,7 +3822,6 @@ ext4_mb_release_group_pa(struct ext4_buddy *e4b, ac->ac_b_ex.fe_len = pa->pa_len; ac->ac_b_ex.fe_logical = 0; ext4_mb_store_history(ac); - kmem_cache_free(ext4_ac_cachep, ac); } return 0; @@ -3850,6 +3843,7 @@ ext4_mb_discard_group_preallocations(struct super_block *sb, struct ext4_group_info *grp = ext4_get_group_info(sb, group); struct buffer_head *bitmap_bh = NULL; struct ext4_prealloc_space *pa, *tmp; + struct ext4_allocation_context *ac; struct list_head list; struct ext4_buddy e4b; int err; @@ -3877,6 +3871,7 @@ ext4_mb_discard_group_preallocations(struct super_block *sb, grp = ext4_get_group_info(sb, group); INIT_LIST_HEAD(&list); + ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS); repeat: ext4_lock_group(sb, group); list_for_each_entry_safe(pa, tmp, @@ -3931,9 +3926,9 @@ repeat: spin_unlock(pa->pa_obj_lock); if (pa->pa_linear) - ext4_mb_release_group_pa(&e4b, pa); + ext4_mb_release_group_pa(&e4b, pa, ac); else - ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa); + ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa, ac); list_del(&pa->u.pa_tmp_list); call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback); @@ -3941,6 +3936,8 @@ repeat: out: ext4_unlock_group(sb, group); + if (ac) + kmem_cache_free(ext4_ac_cachep, ac); ext4_mb_release_desc(&e4b); put_bh(bitmap_bh); return free; @@ -3961,6 +3958,7 @@ void ext4_mb_discard_inode_preallocations(struct inode *inode) struct super_block *sb = inode->i_sb; struct buffer_head *bitmap_bh = NULL; struct ext4_prealloc_space *pa, *tmp; + struct ext4_allocation_context *ac; ext4_group_t group = 0; struct list_head list; struct ext4_buddy e4b; @@ -3975,6 +3973,7 @@ void ext4_mb_discard_inode_preallocations(struct inode *inode) INIT_LIST_HEAD(&list); + ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS); repeat: /* first, collect all pa's in the inode */ spin_lock(&ei->i_prealloc_lock); @@ -4039,7 +4038,7 @@ repeat: ext4_lock_group(sb, group); list_del(&pa->pa_group_list); - ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa); + ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa, ac); ext4_unlock_group(sb, group); ext4_mb_release_desc(&e4b); @@ -4048,6 +4047,8 @@ repeat: list_del(&pa->u.pa_tmp_list); call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback); } + if (ac) + kmem_cache_free(ext4_ac_cachep, ac); } /* -- cgit v1.2.3 From ef7377289a1510d638004158e43878643bc75dc5 Mon Sep 17 00:00:00 2001 From: Solofo Ramangalahy Date: Tue, 29 Apr 2008 22:00:41 -0400 Subject: ext4: update ctime and mtime for truncate with extents. The recently announced "Linux POSIX file system test suite" caught a truncate issue when using extents: mtime and ctime are not updated when truncate is successful. This is the single issue caught with "default" ext4 (mkfs and mount with minimal options). The testsuite does not report failure with -o noextents. With the following patch, all tests of the testsuite pass. Signed-off-by: Solofo Ramangalahy Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index a472bc046363..47929c4e3dae 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -2809,6 +2809,8 @@ out_stop: ext4_orphan_del(handle, inode); up_write(&EXT4_I(inode)->i_data_sem); + inode->i_mtime = inode->i_ctime = ext4_current_time(inode); + ext4_mark_inode_dirty(handle, inode); ext4_journal_stop(handle); } -- cgit v1.2.3 From 8753e88f1b4345677620ec68f847222a6301e2fd Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Tue, 29 Apr 2008 22:00:36 -0400 Subject: ext4: mark inode dirty after initializing the extent tree We should mark the inode dirty only after initializing the extent tree. Also if we fail during extent initialization we need to call DQUOT_FREE_INODE. Signed-off-by: Aneesh Kumar K.V Signed-off-by: "Theodore Ts'o" --- fs/ext4/ialloc.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index d59bdf7233b5..c6efbab0c801 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -739,11 +739,6 @@ got: if (err) goto fail_free_drop; - err = ext4_mark_inode_dirty(handle, inode); - if (err) { - ext4_std_error(sb, err); - goto fail_free_drop; - } if (test_opt(sb, EXTENTS)) { /* set extent flag only for diretory, file and normal symlink*/ if (S_ISDIR(mode) || S_ISREG(mode) || S_ISLNK(mode)) { @@ -752,10 +747,16 @@ got: err = ext4_update_incompat_feature(handle, sb, EXT4_FEATURE_INCOMPAT_EXTENTS); if (err) - goto fail; + goto fail_free_drop; } } + err = ext4_mark_inode_dirty(handle, inode); + if (err) { + ext4_std_error(sb, err); + goto fail_free_drop; + } + ext4_debug("allocating inode %lu\n", inode->i_ino); goto really_out; fail: -- cgit v1.2.3 From c19204b0ae3f8a125118fd5d425d3c7a5f8fda9b Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Tue, 29 Apr 2008 22:00:28 -0400 Subject: ext4: don't use ext4_error in ext4_check_descriptors Because ext4_check_descriptors is called at mount time you can't use ext4_error as it calls ext4_commit_sb, which since the sb isn't all the way initialized causes bad things to happen (ie a panic). This patch changes the ext4_error's to printk's to keep this problem from happening. Thanks much, Signed-off-by: Josef Bacik Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/super.c | 29 +++++++++++++---------------- 1 file changed, 13 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 3435184114c4..52dd0679a4e2 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1484,36 +1484,33 @@ static int ext4_check_descriptors(struct super_block *sb) block_bitmap = ext4_block_bitmap(sb, gdp); if (block_bitmap < first_block || block_bitmap > last_block) { - ext4_error (sb, "ext4_check_descriptors", - "Block bitmap for group %lu" - " not in group (block %llu)!", - i, block_bitmap); + printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: " + "Block bitmap for group %lu not in group " + "(block %llu)!", i, block_bitmap); return 0; } inode_bitmap = ext4_inode_bitmap(sb, gdp); if (inode_bitmap < first_block || inode_bitmap > last_block) { - ext4_error (sb, "ext4_check_descriptors", - "Inode bitmap for group %lu" - " not in group (block %llu)!", - i, inode_bitmap); + printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: " + "Inode bitmap for group %lu not in group " + "(block %llu)!", i, inode_bitmap); return 0; } inode_table = ext4_inode_table(sb, gdp); if (inode_table < first_block || inode_table + sbi->s_itb_per_group - 1 > last_block) { - ext4_error (sb, "ext4_check_descriptors", - "Inode table for group %lu" - " not in group (block %llu)!", - i, inode_table); + printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: " + "Inode table for group %lu not in group " + "(block %llu)!", i, inode_table); return 0; } if (!ext4_group_desc_csum_verify(sbi, i, gdp)) { - ext4_error(sb, __func__, - "Checksum for group %lu failed (%u!=%u)\n", - i, le16_to_cpu(ext4_group_desc_csum(sbi, i, - gdp)), le16_to_cpu(gdp->bg_checksum)); + printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: " + "Checksum for group %lu failed (%u!=%u)\n", + i, le16_to_cpu(ext4_group_desc_csum(sbi, i, + gdp)), le16_to_cpu(gdp->bg_checksum)); return 0; } if (!flexbg_flag) -- cgit v1.2.3 From 60bd63d1928c65abd71d8b9b45672cf6e3101845 Mon Sep 17 00:00:00 2001 From: Solofo Ramangalahy Date: Tue, 29 Apr 2008 21:59:59 -0400 Subject: ext4: cleanup for compiling mballoc with verification and debugging #defines This patch allows compiling mballoc with: #define AGGRESSIVE_CHECK #define DOUBLE_CHECK #define MB_DEBUG It fixes: Compilation errors: fs/ext4/mballoc.c: In function '__mb_check_buddy': fs/ext4/mballoc.c:605: error: 'struct ext4_prealloc_space' has no member named 'group_list' fs/ext4/mballoc.c:606: error: 'struct ext4_prealloc_space' has no member named 'pstart' fs/ext4/mballoc.c:608: error: 'struct ext4_prealloc_space' has no member named 'len' Compilation warnings: fs/ext4/mballoc.c: In function 'ext4_mb_normalize_group_request': fs/ext4/mballoc.c:2863: warning: format '%lu' expects type 'long unsigned int', but argument 3 has type 'int' fs/ext4/mballoc.c: In function 'ext4_mb_use_inode_pa': fs/ext4/mballoc.c:3103: warning: format '%lu' expects type 'long unsigned int', but argument 3 has type 'int' Sparse check: fs/ext4/mballoc.c:3818:2: warning: context imbalance in 'ext4_mb_show_ac' - different lock contexts for basic block Signed-off-by: Solofo Ramangalahy Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/mballoc.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index d2f0b9661fb9..d4ae948606e8 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -896,10 +896,10 @@ static int __mb_check_buddy(struct ext4_buddy *e4b, char *file, list_for_each(cur, &grp->bb_prealloc_list) { ext4_group_t groupnr; struct ext4_prealloc_space *pa; - pa = list_entry(cur, struct ext4_prealloc_space, group_list); - ext4_get_group_no_and_offset(sb, pa->pstart, &groupnr, &k); + pa = list_entry(cur, struct ext4_prealloc_space, pa_group_list); + ext4_get_group_no_and_offset(sb, pa->pa_pstart, &groupnr, &k); MB_CHECK_ASSERT(groupnr == e4b->bd_group); - for (i = 0; i < pa->len; i++) + for (i = 0; i < pa->pa_len; i++) MB_CHECK_ASSERT(mb_test_bit(k + i, buddy)); } return 0; @@ -3131,7 +3131,7 @@ static void ext4_mb_normalize_group_request(struct ext4_allocation_context *ac) ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_stripe; else ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_mb_group_prealloc; - mb_debug("#%u: goal %lu blocks for locality group\n", + mb_debug("#%u: goal %u blocks for locality group\n", current->pid, ac->ac_g_ex.fe_len); } @@ -3371,7 +3371,7 @@ static void ext4_mb_use_inode_pa(struct ext4_allocation_context *ac, BUG_ON(pa->pa_free < len); pa->pa_free -= len; - mb_debug("use %llu/%lu from inode pa %p\n", start, len, pa); + mb_debug("use %llu/%u from inode pa %p\n", start, len, pa); } /* @@ -4108,7 +4108,7 @@ static void ext4_mb_show_ac(struct ext4_allocation_context *ac) printk(KERN_ERR "PA:%lu:%d:%u \n", i, start, pa->pa_len); } - ext4_lock_group(sb, i); + ext4_unlock_group(sb, i); if (grp->bb_free == 0) continue; -- cgit v1.2.3 From 8f6e39a7ade8a5329c5651a2bc07010b3011da6a Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Tue, 29 Apr 2008 22:01:31 -0400 Subject: ext4: Move mballoc headers/structures to a seperate header file mballoc.h Move function and structure definiations out of mballoc.c and put it under a new header file mballoc.h Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/mballoc.c | 296 +--------------------------------------------------- fs/ext4/mballoc.h | 304 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 305 insertions(+), 295 deletions(-) create mode 100644 fs/ext4/mballoc.h (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index d4ae948606e8..11e1fd59acbd 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -21,21 +21,7 @@ * mballoc.c contains the multiblocks allocation routines */ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "ext4_jbd2.h" -#include "ext4.h" -#include "group.h" - +#include "mballoc.h" /* * MUSTDO: * - test ext4_ext_search_left() and ext4_ext_search_right() @@ -345,286 +331,6 @@ * */ -/* - * with AGGRESSIVE_CHECK allocator runs consistency checks over - * structures. these checks slow things down a lot - */ -#define AGGRESSIVE_CHECK__ - -/* - * with DOUBLE_CHECK defined mballoc creates persistent in-core - * bitmaps, maintains and uses them to check for double allocations - */ -#define DOUBLE_CHECK__ - -/* - */ -#define MB_DEBUG__ -#ifdef MB_DEBUG -#define mb_debug(fmt, a...) printk(fmt, ##a) -#else -#define mb_debug(fmt, a...) -#endif - -/* - * with EXT4_MB_HISTORY mballoc stores last N allocations in memory - * and you can monitor it in /proc/fs/ext4//mb_history - */ -#define EXT4_MB_HISTORY -#define EXT4_MB_HISTORY_ALLOC 1 /* allocation */ -#define EXT4_MB_HISTORY_PREALLOC 2 /* preallocated blocks used */ -#define EXT4_MB_HISTORY_DISCARD 4 /* preallocation discarded */ -#define EXT4_MB_HISTORY_FREE 8 /* free */ - -#define EXT4_MB_HISTORY_DEFAULT (EXT4_MB_HISTORY_ALLOC | \ - EXT4_MB_HISTORY_PREALLOC) - -/* - * How long mballoc can look for a best extent (in found extents) - */ -#define MB_DEFAULT_MAX_TO_SCAN 200 - -/* - * How long mballoc must look for a best extent - */ -#define MB_DEFAULT_MIN_TO_SCAN 10 - -/* - * How many groups mballoc will scan looking for the best chunk - */ -#define MB_DEFAULT_MAX_GROUPS_TO_SCAN 5 - -/* - * with 'ext4_mb_stats' allocator will collect stats that will be - * shown at umount. The collecting costs though! - */ -#define MB_DEFAULT_STATS 1 - -/* - * files smaller than MB_DEFAULT_STREAM_THRESHOLD are served - * by the stream allocator, which purpose is to pack requests - * as close each to other as possible to produce smooth I/O traffic - * We use locality group prealloc space for stream request. - * We can tune the same via /proc/fs/ext4//stream_req - */ -#define MB_DEFAULT_STREAM_THRESHOLD 16 /* 64K */ - -/* - * for which requests use 2^N search using buddies - */ -#define MB_DEFAULT_ORDER2_REQS 2 - -/* - * default group prealloc size 512 blocks - */ -#define MB_DEFAULT_GROUP_PREALLOC 512 - -static struct kmem_cache *ext4_pspace_cachep; -static struct kmem_cache *ext4_ac_cachep; - -#ifdef EXT4_BB_MAX_BLOCKS -#undef EXT4_BB_MAX_BLOCKS -#endif -#define EXT4_BB_MAX_BLOCKS 30 - -struct ext4_free_metadata { - ext4_group_t group; - unsigned short num; - ext4_grpblk_t blocks[EXT4_BB_MAX_BLOCKS]; - struct list_head list; -}; - -struct ext4_group_info { - unsigned long bb_state; - unsigned long bb_tid; - struct ext4_free_metadata *bb_md_cur; - unsigned short bb_first_free; - unsigned short bb_free; - unsigned short bb_fragments; - struct list_head bb_prealloc_list; -#ifdef DOUBLE_CHECK - void *bb_bitmap; -#endif - unsigned short bb_counters[]; -}; - -#define EXT4_GROUP_INFO_NEED_INIT_BIT 0 -#define EXT4_GROUP_INFO_LOCKED_BIT 1 - -#define EXT4_MB_GRP_NEED_INIT(grp) \ - (test_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &((grp)->bb_state))) - - -struct ext4_prealloc_space { - struct list_head pa_inode_list; - struct list_head pa_group_list; - union { - struct list_head pa_tmp_list; - struct rcu_head pa_rcu; - } u; - spinlock_t pa_lock; - atomic_t pa_count; - unsigned pa_deleted; - ext4_fsblk_t pa_pstart; /* phys. block */ - ext4_lblk_t pa_lstart; /* log. block */ - unsigned short pa_len; /* len of preallocated chunk */ - unsigned short pa_free; /* how many blocks are free */ - unsigned short pa_linear; /* consumed in one direction - * strictly, for grp prealloc */ - spinlock_t *pa_obj_lock; - struct inode *pa_inode; /* hack, for history only */ -}; - - -struct ext4_free_extent { - ext4_lblk_t fe_logical; - ext4_grpblk_t fe_start; - ext4_group_t fe_group; - int fe_len; -}; - -/* - * Locality group: - * we try to group all related changes together - * so that writeback can flush/allocate them together as well - */ -struct ext4_locality_group { - /* for allocator */ - struct mutex lg_mutex; /* to serialize allocates */ - struct list_head lg_prealloc_list;/* list of preallocations */ - spinlock_t lg_prealloc_lock; -}; - -struct ext4_allocation_context { - struct inode *ac_inode; - struct super_block *ac_sb; - - /* original request */ - struct ext4_free_extent ac_o_ex; - - /* goal request (after normalization) */ - struct ext4_free_extent ac_g_ex; - - /* the best found extent */ - struct ext4_free_extent ac_b_ex; - - /* copy of the bext found extent taken before preallocation efforts */ - struct ext4_free_extent ac_f_ex; - - /* number of iterations done. we have to track to limit searching */ - unsigned long ac_ex_scanned; - __u16 ac_groups_scanned; - __u16 ac_found; - __u16 ac_tail; - __u16 ac_buddy; - __u16 ac_flags; /* allocation hints */ - __u8 ac_status; - __u8 ac_criteria; - __u8 ac_repeats; - __u8 ac_2order; /* if request is to allocate 2^N blocks and - * N > 0, the field stores N, otherwise 0 */ - __u8 ac_op; /* operation, for history only */ - struct page *ac_bitmap_page; - struct page *ac_buddy_page; - struct ext4_prealloc_space *ac_pa; - struct ext4_locality_group *ac_lg; -}; - -#define AC_STATUS_CONTINUE 1 -#define AC_STATUS_FOUND 2 -#define AC_STATUS_BREAK 3 - -struct ext4_mb_history { - struct ext4_free_extent orig; /* orig allocation */ - struct ext4_free_extent goal; /* goal allocation */ - struct ext4_free_extent result; /* result allocation */ - unsigned pid; - unsigned ino; - __u16 found; /* how many extents have been found */ - __u16 groups; /* how many groups have been scanned */ - __u16 tail; /* what tail broke some buddy */ - __u16 buddy; /* buddy the tail ^^^ broke */ - __u16 flags; - __u8 cr:3; /* which phase the result extent was found at */ - __u8 op:4; - __u8 merged:1; -}; - -struct ext4_buddy { - struct page *bd_buddy_page; - void *bd_buddy; - struct page *bd_bitmap_page; - void *bd_bitmap; - struct ext4_group_info *bd_info; - struct super_block *bd_sb; - __u16 bd_blkbits; - ext4_group_t bd_group; -}; -#define EXT4_MB_BITMAP(e4b) ((e4b)->bd_bitmap) -#define EXT4_MB_BUDDY(e4b) ((e4b)->bd_buddy) - -#ifndef EXT4_MB_HISTORY -static inline void ext4_mb_store_history(struct ext4_allocation_context *ac) -{ - return; -} -#else -static void ext4_mb_store_history(struct ext4_allocation_context *ac); -#endif - -#define in_range(b, first, len) ((b) >= (first) && (b) <= (first) + (len) - 1) - -static struct proc_dir_entry *proc_root_ext4; -struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t); - -static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap, - ext4_group_t group); -static void ext4_mb_poll_new_transaction(struct super_block *, handle_t *); -static void ext4_mb_free_committed_blocks(struct super_block *); -static void ext4_mb_return_to_preallocation(struct inode *inode, - struct ext4_buddy *e4b, sector_t block, - int count); -static void ext4_mb_put_pa(struct ext4_allocation_context *, - struct super_block *, struct ext4_prealloc_space *pa); -static int ext4_mb_init_per_dev_proc(struct super_block *sb); -static int ext4_mb_destroy_per_dev_proc(struct super_block *sb); - - -static inline void ext4_lock_group(struct super_block *sb, ext4_group_t group) -{ - struct ext4_group_info *grinfo = ext4_get_group_info(sb, group); - - bit_spin_lock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state)); -} - -static inline void ext4_unlock_group(struct super_block *sb, - ext4_group_t group) -{ - struct ext4_group_info *grinfo = ext4_get_group_info(sb, group); - - bit_spin_unlock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state)); -} - -static inline int ext4_is_group_locked(struct super_block *sb, - ext4_group_t group) -{ - struct ext4_group_info *grinfo = ext4_get_group_info(sb, group); - - return bit_spin_is_locked(EXT4_GROUP_INFO_LOCKED_BIT, - &(grinfo->bb_state)); -} - -static ext4_fsblk_t ext4_grp_offs_to_block(struct super_block *sb, - struct ext4_free_extent *fex) -{ - ext4_fsblk_t block; - - block = (ext4_fsblk_t) fex->fe_group * EXT4_BLOCKS_PER_GROUP(sb) - + fex->fe_start - + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); - return block; -} - static inline void *mb_correct_addr_and_bit(int *bit, void *addr) { #if BITS_PER_LONG == 64 diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h new file mode 100644 index 000000000000..bfe6add46bcf --- /dev/null +++ b/fs/ext4/mballoc.h @@ -0,0 +1,304 @@ +/* + * fs/ext4/mballoc.h + * + * Written by: Alex Tomas + * + */ +#ifndef _EXT4_MBALLOC_H +#define _EXT4_MBALLOC_H + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "ext4_jbd2.h" +#include "ext4.h" +#include "group.h" + +/* + * with AGGRESSIVE_CHECK allocator runs consistency checks over + * structures. these checks slow things down a lot + */ +#define AGGRESSIVE_CHECK__ + +/* + * with DOUBLE_CHECK defined mballoc creates persistent in-core + * bitmaps, maintains and uses them to check for double allocations + */ +#define DOUBLE_CHECK__ + +/* + */ +#define MB_DEBUG__ +#ifdef MB_DEBUG +#define mb_debug(fmt, a...) printk(fmt, ##a) +#else +#define mb_debug(fmt, a...) +#endif + +/* + * with EXT4_MB_HISTORY mballoc stores last N allocations in memory + * and you can monitor it in /proc/fs/ext4//mb_history + */ +#define EXT4_MB_HISTORY +#define EXT4_MB_HISTORY_ALLOC 1 /* allocation */ +#define EXT4_MB_HISTORY_PREALLOC 2 /* preallocated blocks used */ +#define EXT4_MB_HISTORY_DISCARD 4 /* preallocation discarded */ +#define EXT4_MB_HISTORY_FREE 8 /* free */ + +#define EXT4_MB_HISTORY_DEFAULT (EXT4_MB_HISTORY_ALLOC | \ + EXT4_MB_HISTORY_PREALLOC) + +/* + * How long mballoc can look for a best extent (in found extents) + */ +#define MB_DEFAULT_MAX_TO_SCAN 200 + +/* + * How long mballoc must look for a best extent + */ +#define MB_DEFAULT_MIN_TO_SCAN 10 + +/* + * How many groups mballoc will scan looking for the best chunk + */ +#define MB_DEFAULT_MAX_GROUPS_TO_SCAN 5 + +/* + * with 'ext4_mb_stats' allocator will collect stats that will be + * shown at umount. The collecting costs though! + */ +#define MB_DEFAULT_STATS 1 + +/* + * files smaller than MB_DEFAULT_STREAM_THRESHOLD are served + * by the stream allocator, which purpose is to pack requests + * as close each to other as possible to produce smooth I/O traffic + * We use locality group prealloc space for stream request. + * We can tune the same via /proc/fs/ext4//stream_req + */ +#define MB_DEFAULT_STREAM_THRESHOLD 16 /* 64K */ + +/* + * for which requests use 2^N search using buddies + */ +#define MB_DEFAULT_ORDER2_REQS 2 + +/* + * default group prealloc size 512 blocks + */ +#define MB_DEFAULT_GROUP_PREALLOC 512 + +static struct kmem_cache *ext4_pspace_cachep; +static struct kmem_cache *ext4_ac_cachep; + +#ifdef EXT4_BB_MAX_BLOCKS +#undef EXT4_BB_MAX_BLOCKS +#endif +#define EXT4_BB_MAX_BLOCKS 30 + +struct ext4_free_metadata { + ext4_group_t group; + unsigned short num; + ext4_grpblk_t blocks[EXT4_BB_MAX_BLOCKS]; + struct list_head list; +}; + +struct ext4_group_info { + unsigned long bb_state; + unsigned long bb_tid; + struct ext4_free_metadata *bb_md_cur; + unsigned short bb_first_free; + unsigned short bb_free; + unsigned short bb_fragments; + struct list_head bb_prealloc_list; +#ifdef DOUBLE_CHECK + void *bb_bitmap; +#endif + unsigned short bb_counters[]; +}; + +#define EXT4_GROUP_INFO_NEED_INIT_BIT 0 +#define EXT4_GROUP_INFO_LOCKED_BIT 1 + +#define EXT4_MB_GRP_NEED_INIT(grp) \ + (test_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &((grp)->bb_state))) + + +struct ext4_prealloc_space { + struct list_head pa_inode_list; + struct list_head pa_group_list; + union { + struct list_head pa_tmp_list; + struct rcu_head pa_rcu; + } u; + spinlock_t pa_lock; + atomic_t pa_count; + unsigned pa_deleted; + ext4_fsblk_t pa_pstart; /* phys. block */ + ext4_lblk_t pa_lstart; /* log. block */ + unsigned short pa_len; /* len of preallocated chunk */ + unsigned short pa_free; /* how many blocks are free */ + unsigned short pa_linear; /* consumed in one direction + * strictly, for grp prealloc */ + spinlock_t *pa_obj_lock; + struct inode *pa_inode; /* hack, for history only */ +}; + + +struct ext4_free_extent { + ext4_lblk_t fe_logical; + ext4_grpblk_t fe_start; + ext4_group_t fe_group; + int fe_len; +}; + +/* + * Locality group: + * we try to group all related changes together + * so that writeback can flush/allocate them together as well + */ +struct ext4_locality_group { + /* for allocator */ + struct mutex lg_mutex; /* to serialize allocates */ + struct list_head lg_prealloc_list;/* list of preallocations */ + spinlock_t lg_prealloc_lock; +}; + +struct ext4_allocation_context { + struct inode *ac_inode; + struct super_block *ac_sb; + + /* original request */ + struct ext4_free_extent ac_o_ex; + + /* goal request (after normalization) */ + struct ext4_free_extent ac_g_ex; + + /* the best found extent */ + struct ext4_free_extent ac_b_ex; + + /* copy of the bext found extent taken before preallocation efforts */ + struct ext4_free_extent ac_f_ex; + + /* number of iterations done. we have to track to limit searching */ + unsigned long ac_ex_scanned; + __u16 ac_groups_scanned; + __u16 ac_found; + __u16 ac_tail; + __u16 ac_buddy; + __u16 ac_flags; /* allocation hints */ + __u8 ac_status; + __u8 ac_criteria; + __u8 ac_repeats; + __u8 ac_2order; /* if request is to allocate 2^N blocks and + * N > 0, the field stores N, otherwise 0 */ + __u8 ac_op; /* operation, for history only */ + struct page *ac_bitmap_page; + struct page *ac_buddy_page; + struct ext4_prealloc_space *ac_pa; + struct ext4_locality_group *ac_lg; +}; + +#define AC_STATUS_CONTINUE 1 +#define AC_STATUS_FOUND 2 +#define AC_STATUS_BREAK 3 + +struct ext4_mb_history { + struct ext4_free_extent orig; /* orig allocation */ + struct ext4_free_extent goal; /* goal allocation */ + struct ext4_free_extent result; /* result allocation */ + unsigned pid; + unsigned ino; + __u16 found; /* how many extents have been found */ + __u16 groups; /* how many groups have been scanned */ + __u16 tail; /* what tail broke some buddy */ + __u16 buddy; /* buddy the tail ^^^ broke */ + __u16 flags; + __u8 cr:3; /* which phase the result extent was found at */ + __u8 op:4; + __u8 merged:1; +}; + +struct ext4_buddy { + struct page *bd_buddy_page; + void *bd_buddy; + struct page *bd_bitmap_page; + void *bd_bitmap; + struct ext4_group_info *bd_info; + struct super_block *bd_sb; + __u16 bd_blkbits; + ext4_group_t bd_group; +}; +#define EXT4_MB_BITMAP(e4b) ((e4b)->bd_bitmap) +#define EXT4_MB_BUDDY(e4b) ((e4b)->bd_buddy) + +#ifndef EXT4_MB_HISTORY +static inline void ext4_mb_store_history(struct ext4_allocation_context *ac) +{ + return; +} +#else +static void ext4_mb_store_history(struct ext4_allocation_context *ac); +#endif + +#define in_range(b, first, len) ((b) >= (first) && (b) <= (first) + (len) - 1) + +static struct proc_dir_entry *proc_root_ext4; +struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t); + +static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap, + ext4_group_t group); +static void ext4_mb_poll_new_transaction(struct super_block *, handle_t *); +static void ext4_mb_free_committed_blocks(struct super_block *); +static void ext4_mb_return_to_preallocation(struct inode *inode, + struct ext4_buddy *e4b, sector_t block, + int count); +static void ext4_mb_put_pa(struct ext4_allocation_context *, + struct super_block *, struct ext4_prealloc_space *pa); +static int ext4_mb_init_per_dev_proc(struct super_block *sb); +static int ext4_mb_destroy_per_dev_proc(struct super_block *sb); + + +static inline void ext4_lock_group(struct super_block *sb, ext4_group_t group) +{ + struct ext4_group_info *grinfo = ext4_get_group_info(sb, group); + + bit_spin_lock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state)); +} + +static inline void ext4_unlock_group(struct super_block *sb, + ext4_group_t group) +{ + struct ext4_group_info *grinfo = ext4_get_group_info(sb, group); + + bit_spin_unlock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state)); +} + +static inline int ext4_is_group_locked(struct super_block *sb, + ext4_group_t group) +{ + struct ext4_group_info *grinfo = ext4_get_group_info(sb, group); + + return bit_spin_is_locked(EXT4_GROUP_INFO_LOCKED_BIT, + &(grinfo->bb_state)); +} + +static ext4_fsblk_t ext4_grp_offs_to_block(struct super_block *sb, + struct ext4_free_extent *fex) +{ + ext4_fsblk_t block; + + block = (ext4_fsblk_t) fex->fe_group * EXT4_BLOCKS_PER_GROUP(sb) + + fex->fe_start + + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); + return block; +} +#endif -- cgit v1.2.3 From 7c2f3d6f89aab04c5c66a0a757888d3a77a5e899 Mon Sep 17 00:00:00 2001 From: Roel Kluin <12o3l@tiscali.nl> Date: Tue, 29 Apr 2008 22:01:27 -0400 Subject: ext3: fix test ext_generic_write_end() copied return value 'copied' is unsigned, whereas 'ret2' is not. The test (copied < 0) fails Signed-off-by: Roel Kluin <12o3l@tiscali.nl> Signed-off-by: "Theodore Ts'o" --- fs/ext3/inode.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index cc47b76091bf..6ae4ecf3ce40 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -1261,10 +1261,11 @@ static int ext3_ordered_write_end(struct file *file, new_i_size = pos + copied; if (new_i_size > EXT3_I(inode)->i_disksize) EXT3_I(inode)->i_disksize = new_i_size; - copied = ext3_generic_write_end(file, mapping, pos, len, copied, + ret2 = ext3_generic_write_end(file, mapping, pos, len, copied, page, fsdata); - if (copied < 0) - ret = copied; + copied = ret2; + if (ret2 < 0) + ret = ret2; } ret2 = ext3_journal_stop(handle); if (!ret) @@ -1289,10 +1290,11 @@ static int ext3_writeback_write_end(struct file *file, if (new_i_size > EXT3_I(inode)->i_disksize) EXT3_I(inode)->i_disksize = new_i_size; - copied = ext3_generic_write_end(file, mapping, pos, len, copied, + ret2 = ext3_generic_write_end(file, mapping, pos, len, copied, page, fsdata); - if (copied < 0) - ret = copied; + copied = ret2; + if (ret2 < 0) + ret = ret2; ret2 = ext3_journal_stop(handle); if (!ret) -- cgit v1.2.3 From f8a87d89304c1eea8e4a8dc02d134f57590913c6 Mon Sep 17 00:00:00 2001 From: Roel Kluin <12o3l@tiscali.nl> Date: Tue, 29 Apr 2008 22:01:18 -0400 Subject: ext4: fix test ext_generic_write_end() copied return value 'copied' is unsigned, whereas 'ret2' is not. The test (copied < 0) fails Signed-off-by: Roel Kluin <12o3l@tiscali.nl> Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 0c94db462c2f..8d9707746413 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1311,10 +1311,11 @@ static int ext4_ordered_write_end(struct file *file, new_i_size = pos + copied; if (new_i_size > EXT4_I(inode)->i_disksize) EXT4_I(inode)->i_disksize = new_i_size; - copied = ext4_generic_write_end(file, mapping, pos, len, copied, + ret2 = ext4_generic_write_end(file, mapping, pos, len, copied, page, fsdata); - if (copied < 0) - ret = copied; + copied = ret2; + if (ret2 < 0) + ret = ret2; } ret2 = ext4_journal_stop(handle); if (!ret) @@ -1339,10 +1340,11 @@ static int ext4_writeback_write_end(struct file *file, if (new_i_size > EXT4_I(inode)->i_disksize) EXT4_I(inode)->i_disksize = new_i_size; - copied = ext4_generic_write_end(file, mapping, pos, len, copied, + ret2 = ext4_generic_write_end(file, mapping, pos, len, copied, page, fsdata); - if (copied < 0) - ret = copied; + copied = ret2; + if (ret2 < 0) + ret = ret2; ret2 = ext4_journal_stop(handle); if (!ret) -- cgit v1.2.3 From f1fa3342e271029f93d323ca664809b94594fe04 Mon Sep 17 00:00:00 2001 From: Roel Kluin <12o3l@tiscali.nl> Date: Tue, 29 Apr 2008 22:01:15 -0400 Subject: ext4: fix hot spins in mballoc after err_freebuddy and err_freemeta In ext4_mb_init_backend() 'i' is of type ext4_group_t. Since unsigned, i >= 0 is always true, so fix hot spins after err_freebuddy: and -meta: and prevent decrements when zero. Signed-off-by: Roel Kluin <12o3l@tiscali.nl> Signed-off-by: Mingming Cao Signed-off-by: "Theodore Ts'o" --- fs/ext4/mballoc.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 11e1fd59acbd..fbec2ef93797 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2272,13 +2272,13 @@ static int ext4_mb_init_backend(struct super_block *sb) meta_group_info[j] = kzalloc(len, GFP_KERNEL); if (meta_group_info[j] == NULL) { printk(KERN_ERR "EXT4-fs: can't allocate buddy mem\n"); - i--; goto err_freebuddy; } desc = ext4_get_group_desc(sb, i, NULL); if (desc == NULL) { printk(KERN_ERR "EXT4-fs: can't read descriptor %lu\n", i); + i++; goto err_freebuddy; } memset(meta_group_info[j], 0, len); @@ -2318,13 +2318,11 @@ static int ext4_mb_init_backend(struct super_block *sb) return 0; err_freebuddy: - while (i >= 0) { + while (i-- > 0) kfree(ext4_get_group_info(sb, i)); - i--; - } i = num_meta_group_infos; err_freemeta: - while (--i >= 0) + while (i-- > 0) kfree(sbi->s_group_info[i]); iput(sbi->s_buddy_cache); err_freesgi: -- cgit v1.2.3 From 53492b1de46a7576170e865062ffcfc93bb5650b Mon Sep 17 00:00:00 2001 From: Gerald Schaefer Date: Wed, 30 Apr 2008 13:38:46 +0200 Subject: [S390] System z large page support. This adds hugetlbfs support on System z, using both hardware large page support if available and software large page emulation on older hardware. Shared (large) page tables are implemented in software emulation mode, by using page->index of the first tail page from a compound large page to store page table information. Signed-off-by: Gerald Schaefer Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/early.c | 16 ++++ arch/s390/kernel/head64.S | 2 +- arch/s390/kernel/setup.c | 10 ++- arch/s390/mm/Makefile | 2 +- arch/s390/mm/fault.c | 3 + arch/s390/mm/hugetlbpage.c | 134 ++++++++++++++++++++++++++++++++ arch/s390/mm/init.c | 23 ------ arch/s390/mm/vmem.c | 55 +++++++++++-- fs/Kconfig | 3 +- include/asm-s390/hugetlb.h | 183 ++++++++++++++++++++++++++++++++++++++++++++ include/asm-s390/page.h | 29 +++++-- include/asm-s390/pgtable.h | 12 +++ include/asm-s390/setup.h | 6 ++ include/asm-s390/tlbflush.h | 1 + 14 files changed, 437 insertions(+), 42 deletions(-) create mode 100644 arch/s390/mm/hugetlbpage.c create mode 100644 include/asm-s390/hugetlb.h (limited to 'fs') diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index bd188f6bd0e2..d0e09684b9ce 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -268,6 +268,19 @@ static noinline __init void setup_lowcore_early(void) s390_base_pgm_handler_fn = early_pgm_check_handler; } +static noinline __init void setup_hpage(void) +{ +#ifndef CONFIG_DEBUG_PAGEALLOC + unsigned int facilities; + + facilities = stfl(); + if (!(facilities & (1UL << 23)) || !(facilities & (1UL << 29))) + return; + machine_flags |= MACHINE_FLAG_HPAGE; + __ctl_set_bit(0, 23); +#endif +} + static __init void detect_mvpg(void) { #ifndef CONFIG_64BIT @@ -360,6 +373,8 @@ static __init void detect_machine_facilities(void) facilities = stfl(); if (facilities & (1 << 28)) machine_flags |= MACHINE_FLAG_IDTE; + if (facilities & (1 << 23)) + machine_flags |= MACHINE_FLAG_PFMF; if (facilities & (1 << 4)) machine_flags |= MACHINE_FLAG_MVCOS; #endif @@ -388,6 +403,7 @@ void __init startup_init(void) detect_diag9c(); detect_diag44(); detect_machine_facilities(); + setup_hpage(); sclp_read_info_early(); sclp_facilities_detect(); memsize = sclp_memory_detect(); diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S index 9c2c6f7d37e7..1d06961e87b3 100644 --- a/arch/s390/kernel/head64.S +++ b/arch/s390/kernel/head64.S @@ -129,7 +129,7 @@ startup_continue: # virtual and never return ... .align 16 .Lentry:.quad 0x0000000180000000,_stext -.Lctl: .quad 0x04b50002 # cr0: various things +.Lctl: .quad 0x04350002 # cr0: various things .quad 0 # cr1: primary space segment table .quad .Lduct # cr2: dispatchable unit control table .quad 0 # cr3: instruction authorization diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 694c6546ce64..2bc70b6e876a 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -749,6 +749,9 @@ static void __init setup_hwcaps(void) elf_hwcap |= 1UL << 6; } + if (MACHINE_HAS_HPAGE) + elf_hwcap |= 1UL << 7; + switch (cpuinfo->cpu_id.machine) { case 0x9672: #if !defined(CONFIG_64BIT) @@ -872,8 +875,9 @@ void __cpuinit print_cpu_info(struct cpuinfo_S390 *cpuinfo) static int show_cpuinfo(struct seq_file *m, void *v) { - static const char *hwcap_str[7] = { - "esan3", "zarch", "stfle", "msa", "ldisp", "eimm", "dfp" + static const char *hwcap_str[8] = { + "esan3", "zarch", "stfle", "msa", "ldisp", "eimm", "dfp", + "edat" }; struct cpuinfo_S390 *cpuinfo; unsigned long n = (unsigned long) v - 1; @@ -888,7 +892,7 @@ static int show_cpuinfo(struct seq_file *m, void *v) num_online_cpus(), loops_per_jiffy/(500000/HZ), (loops_per_jiffy/(5000/HZ))%100); seq_puts(m, "features\t: "); - for (i = 0; i < 7; i++) + for (i = 0; i < 8; i++) if (hwcap_str[i] && (elf_hwcap & (1UL << i))) seq_printf(m, "%s ", hwcap_str[i]); seq_puts(m, "\n"); diff --git a/arch/s390/mm/Makefile b/arch/s390/mm/Makefile index 66401930f83e..fb988a48a754 100644 --- a/arch/s390/mm/Makefile +++ b/arch/s390/mm/Makefile @@ -4,4 +4,4 @@ obj-y := init.o fault.o extmem.o mmap.o vmem.o pgtable.o obj-$(CONFIG_CMM) += cmm.o - +obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c index 2650f46001d0..4d537205e83c 100644 --- a/arch/s390/mm/fault.c +++ b/arch/s390/mm/fault.c @@ -28,6 +28,7 @@ #include #include #include +#include #include #include #include @@ -367,6 +368,8 @@ good_area: } survive: + if (is_vm_hugetlb_page(vma)) + address &= HPAGE_MASK; /* * If for any reason at all we couldn't handle the fault, * make sure we exit gracefully rather than endlessly redo diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c new file mode 100644 index 000000000000..f4b6124fdb75 --- /dev/null +++ b/arch/s390/mm/hugetlbpage.c @@ -0,0 +1,134 @@ +/* + * IBM System z Huge TLB Page Support for Kernel. + * + * Copyright 2007 IBM Corp. + * Author(s): Gerald Schaefer + */ + +#include +#include + + +void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, + pte_t *pteptr, pte_t pteval) +{ + pmd_t *pmdp = (pmd_t *) pteptr; + pte_t shadow_pteval = pteval; + unsigned long mask; + + if (!MACHINE_HAS_HPAGE) { + pteptr = (pte_t *) pte_page(pteval)[1].index; + mask = pte_val(pteval) & + (_SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO); + pte_val(pteval) = (_SEGMENT_ENTRY + __pa(pteptr)) | mask; + if (mm->context.noexec) { + pteptr += PTRS_PER_PTE; + pte_val(shadow_pteval) = + (_SEGMENT_ENTRY + __pa(pteptr)) | mask; + } + } + + pmd_val(*pmdp) = pte_val(pteval); + if (mm->context.noexec) { + pmdp = get_shadow_table(pmdp); + pmd_val(*pmdp) = pte_val(shadow_pteval); + } +} + +int arch_prepare_hugepage(struct page *page) +{ + unsigned long addr = page_to_phys(page); + pte_t pte; + pte_t *ptep; + int i; + + if (MACHINE_HAS_HPAGE) + return 0; + + ptep = (pte_t *) pte_alloc_one(&init_mm, address); + if (!ptep) + return -ENOMEM; + + pte = mk_pte(page, PAGE_RW); + for (i = 0; i < PTRS_PER_PTE; i++) { + set_pte_at(&init_mm, addr + i * PAGE_SIZE, ptep + i, pte); + pte_val(pte) += PAGE_SIZE; + } + page[1].index = (unsigned long) ptep; + return 0; +} + +void arch_release_hugepage(struct page *page) +{ + pte_t *ptep; + + if (MACHINE_HAS_HPAGE) + return; + + ptep = (pte_t *) page[1].index; + if (!ptep) + return; + pte_free(&init_mm, ptep); + page[1].index = 0; +} + +pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) +{ + pgd_t *pgdp; + pud_t *pudp; + pmd_t *pmdp = NULL; + + pgdp = pgd_offset(mm, addr); + pudp = pud_alloc(mm, pgdp, addr); + if (pudp) + pmdp = pmd_alloc(mm, pudp, addr); + return (pte_t *) pmdp; +} + +pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) +{ + pgd_t *pgdp; + pud_t *pudp; + pmd_t *pmdp = NULL; + + pgdp = pgd_offset(mm, addr); + if (pgd_present(*pgdp)) { + pudp = pud_offset(pgdp, addr); + if (pud_present(*pudp)) + pmdp = pmd_offset(pudp, addr); + } + return (pte_t *) pmdp; +} + +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + +struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, + int write) +{ + return ERR_PTR(-EINVAL); +} + +int pmd_huge(pmd_t pmd) +{ + if (!MACHINE_HAS_HPAGE) + return 0; + + return !!(pmd_val(pmd) & _SEGMENT_ENTRY_LARGE); +} + +struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, + pmd_t *pmdp, int write) +{ + struct page *page; + + if (!MACHINE_HAS_HPAGE) + return NULL; + + page = pmd_page(*pmdp); + if (page) + page += ((address & ~HPAGE_MASK) >> PAGE_SHIFT); + return page; +} diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c index 202c952a29b4..acc92f46a096 100644 --- a/arch/s390/mm/init.c +++ b/arch/s390/mm/init.c @@ -77,28 +77,6 @@ void show_mem(void) printk("%lu pages pagetables\n", global_page_state(NR_PAGETABLE)); } -static void __init setup_ro_region(void) -{ - pgd_t *pgd; - pud_t *pud; - pmd_t *pmd; - pte_t *pte; - pte_t new_pte; - unsigned long address, end; - - address = ((unsigned long)&_stext) & PAGE_MASK; - end = PFN_ALIGN((unsigned long)&_eshared); - - for (; address < end; address += PAGE_SIZE) { - pgd = pgd_offset_k(address); - pud = pud_offset(pgd, address); - pmd = pmd_offset(pud, address); - pte = pte_offset_kernel(pmd, address); - new_pte = mk_pte_phys(address, __pgprot(_PAGE_RO)); - *pte = new_pte; - } -} - /* * paging_init() sets up the page tables */ @@ -121,7 +99,6 @@ void __init paging_init(void) clear_table((unsigned long *) init_mm.pgd, pgd_type, sizeof(unsigned long)*2048); vmem_map_init(); - setup_ro_region(); /* enable virtual mapping in kernel mode */ __ctl_load(S390_lowcore.kernel_asce, 1, 1); diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c index 3ffc0211dc85..97bce6c97574 100644 --- a/arch/s390/mm/vmem.c +++ b/arch/s390/mm/vmem.c @@ -10,10 +10,12 @@ #include #include #include +#include #include #include #include #include +#include static DEFINE_MUTEX(vmem_mutex); @@ -113,7 +115,7 @@ static pte_t __init_refok *vmem_pte_alloc(void) /* * Add a physical memory range to the 1:1 mapping. */ -static int vmem_add_range(unsigned long start, unsigned long size) +static int vmem_add_range(unsigned long start, unsigned long size, int ro) { unsigned long address; pgd_t *pg_dir; @@ -140,7 +142,19 @@ static int vmem_add_range(unsigned long start, unsigned long size) pud_populate_kernel(&init_mm, pu_dir, pm_dir); } + pte = mk_pte_phys(address, __pgprot(ro ? _PAGE_RO : 0)); pm_dir = pmd_offset(pu_dir, address); + +#ifdef __s390x__ + if (MACHINE_HAS_HPAGE && !(address & ~HPAGE_MASK) && + (address + HPAGE_SIZE <= start + size) && + (address >= HPAGE_SIZE)) { + pte_val(pte) |= _SEGMENT_ENTRY_LARGE; + pmd_val(*pm_dir) = pte_val(pte); + address += HPAGE_SIZE - PAGE_SIZE; + continue; + } +#endif if (pmd_none(*pm_dir)) { pt_dir = vmem_pte_alloc(); if (!pt_dir) @@ -149,7 +163,6 @@ static int vmem_add_range(unsigned long start, unsigned long size) } pt_dir = pte_offset_kernel(pm_dir, address); - pte = pfn_pte(address >> PAGE_SHIFT, PAGE_KERNEL); *pt_dir = pte; } ret = 0; @@ -180,6 +193,13 @@ static void vmem_remove_range(unsigned long start, unsigned long size) pm_dir = pmd_offset(pu_dir, address); if (pmd_none(*pm_dir)) continue; + + if (pmd_huge(*pm_dir)) { + pmd_clear_kernel(pm_dir); + address += HPAGE_SIZE - PAGE_SIZE; + continue; + } + pt_dir = pte_offset_kernel(pm_dir, address); *pt_dir = pte; } @@ -248,14 +268,14 @@ out: return ret; } -static int vmem_add_mem(unsigned long start, unsigned long size) +static int vmem_add_mem(unsigned long start, unsigned long size, int ro) { int ret; ret = vmem_add_mem_map(start, size); if (ret) return ret; - return vmem_add_range(start, size); + return vmem_add_range(start, size, ro); } /* @@ -338,7 +358,7 @@ int add_shared_memory(unsigned long start, unsigned long size) if (ret) goto out_free; - ret = vmem_add_mem(start, size); + ret = vmem_add_mem(start, size, 0); if (ret) goto out_remove; @@ -374,14 +394,35 @@ out: */ void __init vmem_map_init(void) { + unsigned long ro_start, ro_end; + unsigned long start, end; int i; INIT_LIST_HEAD(&init_mm.context.crst_list); INIT_LIST_HEAD(&init_mm.context.pgtable_list); init_mm.context.noexec = 0; NODE_DATA(0)->node_mem_map = VMEM_MAP; - for (i = 0; i < MEMORY_CHUNKS && memory_chunk[i].size > 0; i++) - vmem_add_mem(memory_chunk[i].addr, memory_chunk[i].size); + ro_start = ((unsigned long)&_stext) & PAGE_MASK; + ro_end = PFN_ALIGN((unsigned long)&_eshared); + for (i = 0; i < MEMORY_CHUNKS && memory_chunk[i].size > 0; i++) { + start = memory_chunk[i].addr; + end = memory_chunk[i].addr + memory_chunk[i].size; + if (start >= ro_end || end <= ro_start) + vmem_add_mem(start, end - start, 0); + else if (start >= ro_start && end <= ro_end) + vmem_add_mem(start, end - start, 1); + else if (start >= ro_start) { + vmem_add_mem(start, ro_end - start, 1); + vmem_add_mem(ro_end, end - ro_end, 0); + } else if (end < ro_end) { + vmem_add_mem(start, ro_start - start, 0); + vmem_add_mem(ro_start, end - ro_start, 1); + } else { + vmem_add_mem(start, ro_start - start, 0); + vmem_add_mem(ro_start, ro_end - ro_start, 1); + vmem_add_mem(ro_end, end - ro_end, 0); + } + } } /* diff --git a/fs/Kconfig b/fs/Kconfig index 2e43d46f65d6..cf12c403b8c7 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -1005,7 +1005,8 @@ config TMPFS_POSIX_ACL config HUGETLBFS bool "HugeTLB file system support" - depends on X86 || IA64 || PPC64 || SPARC64 || (SUPERH && MMU) || BROKEN + depends on X86 || IA64 || PPC64 || SPARC64 || (SUPERH && MMU) || \ + (S390 && 64BIT) || BROKEN help hugetlbfs is a filesystem backing for HugeTLB pages, based on ramfs. For architectures that support it, say Y here and read diff --git a/include/asm-s390/hugetlb.h b/include/asm-s390/hugetlb.h new file mode 100644 index 000000000000..600a776f8f75 --- /dev/null +++ b/include/asm-s390/hugetlb.h @@ -0,0 +1,183 @@ +/* + * IBM System z Huge TLB Page Support for Kernel. + * + * Copyright IBM Corp. 2008 + * Author(s): Gerald Schaefer + */ + +#ifndef _ASM_S390_HUGETLB_H +#define _ASM_S390_HUGETLB_H + +#include +#include + + +#define is_hugepage_only_range(mm, addr, len) 0 +#define hugetlb_free_pgd_range free_pgd_range + +void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t pte); + +/* + * If the arch doesn't supply something else, assume that hugepage + * size aligned regions are ok without further preparation. + */ +static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) +{ + if (len & ~HPAGE_MASK) + return -EINVAL; + if (addr & ~HPAGE_MASK) + return -EINVAL; + return 0; +} + +#define hugetlb_prefault_arch_hook(mm) do { } while (0) + +int arch_prepare_hugepage(struct page *page); +void arch_release_hugepage(struct page *page); + +static inline pte_t pte_mkhuge(pte_t pte) +{ + /* + * PROT_NONE needs to be remapped from the pte type to the ste type. + * The HW invalid bit is also different for pte and ste. The pte + * invalid bit happens to be the same as the ste _SEGMENT_ENTRY_LARGE + * bit, so we don't have to clear it. + */ + if (pte_val(pte) & _PAGE_INVALID) { + if (pte_val(pte) & _PAGE_SWT) + pte_val(pte) |= _HPAGE_TYPE_NONE; + pte_val(pte) |= _SEGMENT_ENTRY_INV; + } + /* + * Clear SW pte bits SWT and SWX, there are no SW bits in a segment + * table entry. + */ + pte_val(pte) &= ~(_PAGE_SWT | _PAGE_SWX); + /* + * Also set the change-override bit because we don't need dirty bit + * tracking for hugetlbfs pages. + */ + pte_val(pte) |= (_SEGMENT_ENTRY_LARGE | _SEGMENT_ENTRY_CO); + return pte; +} + +static inline pte_t huge_pte_wrprotect(pte_t pte) +{ + pte_val(pte) |= _PAGE_RO; + return pte; +} + +static inline int huge_pte_none(pte_t pte) +{ + return (pte_val(pte) & _SEGMENT_ENTRY_INV) && + !(pte_val(pte) & _SEGMENT_ENTRY_RO); +} + +static inline pte_t huge_ptep_get(pte_t *ptep) +{ + pte_t pte = *ptep; + unsigned long mask; + + if (!MACHINE_HAS_HPAGE) { + ptep = (pte_t *) (pte_val(pte) & _SEGMENT_ENTRY_ORIGIN); + if (ptep) { + mask = pte_val(pte) & + (_SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO); + pte = pte_mkhuge(*ptep); + pte_val(pte) |= mask; + } + } + return pte; +} + +static inline pte_t huge_ptep_get_and_clear(struct mm_struct *mm, + unsigned long addr, pte_t *ptep) +{ + pte_t pte = huge_ptep_get(ptep); + + pmd_clear((pmd_t *) ptep); + return pte; +} + +static inline void __pmd_csp(pmd_t *pmdp) +{ + register unsigned long reg2 asm("2") = pmd_val(*pmdp); + register unsigned long reg3 asm("3") = pmd_val(*pmdp) | + _SEGMENT_ENTRY_INV; + register unsigned long reg4 asm("4") = ((unsigned long) pmdp) + 5; + + asm volatile( + " csp %1,%3" + : "=m" (*pmdp) + : "d" (reg2), "d" (reg3), "d" (reg4), "m" (*pmdp) : "cc"); + pmd_val(*pmdp) = _SEGMENT_ENTRY_INV | _SEGMENT_ENTRY; +} + +static inline void __pmd_idte(unsigned long address, pmd_t *pmdp) +{ + unsigned long sto = (unsigned long) pmdp - + pmd_index(address) * sizeof(pmd_t); + + if (!(pmd_val(*pmdp) & _SEGMENT_ENTRY_INV)) { + asm volatile( + " .insn rrf,0xb98e0000,%2,%3,0,0" + : "=m" (*pmdp) + : "m" (*pmdp), "a" (sto), + "a" ((address & HPAGE_MASK)) + ); + } + pmd_val(*pmdp) = _SEGMENT_ENTRY_INV | _SEGMENT_ENTRY; +} + +static inline void huge_ptep_invalidate(struct mm_struct *mm, + unsigned long address, pte_t *ptep) +{ + pmd_t *pmdp = (pmd_t *) ptep; + + if (!MACHINE_HAS_IDTE) { + __pmd_csp(pmdp); + if (mm->context.noexec) { + pmdp = get_shadow_table(pmdp); + __pmd_csp(pmdp); + } + return; + } + + __pmd_idte(address, pmdp); + if (mm->context.noexec) { + pmdp = get_shadow_table(pmdp); + __pmd_idte(address, pmdp); + } + return; +} + +#define huge_ptep_set_access_flags(__vma, __addr, __ptep, __entry, __dirty) \ +({ \ + int __changed = !pte_same(huge_ptep_get(__ptep), __entry); \ + if (__changed) { \ + huge_ptep_invalidate((__vma)->vm_mm, __addr, __ptep); \ + set_huge_pte_at((__vma)->vm_mm, __addr, __ptep, __entry); \ + } \ + __changed; \ +}) + +#define huge_ptep_set_wrprotect(__mm, __addr, __ptep) \ +({ \ + pte_t __pte = huge_ptep_get(__ptep); \ + if (pte_write(__pte)) { \ + if (atomic_read(&(__mm)->mm_users) > 1 || \ + (__mm) != current->active_mm) \ + huge_ptep_invalidate(__mm, __addr, __ptep); \ + set_huge_pte_at(__mm, __addr, __ptep, \ + huge_pte_wrprotect(__pte)); \ + } \ +}) + +static inline void huge_ptep_clear_flush(struct vm_area_struct *vma, + unsigned long address, pte_t *ptep) +{ + huge_ptep_invalidate(vma->vm_mm, address, ptep); +} + +#endif /* _ASM_S390_HUGETLB_H */ diff --git a/include/asm-s390/page.h b/include/asm-s390/page.h index fe7f92b6ae6d..b01e6fc9a295 100644 --- a/include/asm-s390/page.h +++ b/include/asm-s390/page.h @@ -19,17 +19,34 @@ #define PAGE_DEFAULT_ACC 0 #define PAGE_DEFAULT_KEY (PAGE_DEFAULT_ACC << 4) +#define HPAGE_SHIFT 20 +#define HPAGE_SIZE (1UL << HPAGE_SHIFT) +#define HPAGE_MASK (~(HPAGE_SIZE - 1)) +#define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT) + +#define ARCH_HAS_SETCLEAR_HUGE_PTE +#define ARCH_HAS_HUGE_PTE_TYPE +#define ARCH_HAS_PREPARE_HUGEPAGE +#define ARCH_HAS_HUGEPAGE_CLEAR_FLUSH + #include #ifndef __ASSEMBLY__ static inline void clear_page(void *page) { - register unsigned long reg1 asm ("1") = 0; - register void *reg2 asm ("2") = page; - register unsigned long reg3 asm ("3") = 4096; - asm volatile( - " mvcl 2,0" - : "+d" (reg2), "+d" (reg3) : "d" (reg1) : "memory", "cc"); + if (MACHINE_HAS_PFMF) { + asm volatile( + " .insn rre,0xb9af0000,%0,%1" + : : "d" (0x10000), "a" (page) : "memory", "cc"); + } else { + register unsigned long reg1 asm ("1") = 0; + register void *reg2 asm ("2") = page; + register unsigned long reg3 asm ("3") = 4096; + asm volatile( + " mvcl 2,0" + : "+d" (reg2), "+d" (reg3) : "d" (reg1) + : "memory", "cc"); + } } static inline void copy_page(void *to, void *from) diff --git a/include/asm-s390/pgtable.h b/include/asm-s390/pgtable.h index f8347ce9c5a1..fd336f2e2a7a 100644 --- a/include/asm-s390/pgtable.h +++ b/include/asm-s390/pgtable.h @@ -233,6 +233,15 @@ extern char empty_zero_page[PAGE_SIZE]; #define _PAGE_TYPE_EX_RO 0x202 #define _PAGE_TYPE_EX_RW 0x002 +/* + * Only four types for huge pages, using the invalid bit and protection bit + * of a segment table entry. + */ +#define _HPAGE_TYPE_EMPTY 0x020 /* _SEGMENT_ENTRY_INV */ +#define _HPAGE_TYPE_NONE 0x220 +#define _HPAGE_TYPE_RO 0x200 /* _SEGMENT_ENTRY_RO */ +#define _HPAGE_TYPE_RW 0x000 + /* * PTE type bits are rather complicated. handle_pte_fault uses pte_present, * pte_none and pte_file to find out the pte type WITHOUT holding the page @@ -325,6 +334,9 @@ extern char empty_zero_page[PAGE_SIZE]; #define _SEGMENT_ENTRY (0) #define _SEGMENT_ENTRY_EMPTY (_SEGMENT_ENTRY_INV) +#define _SEGMENT_ENTRY_LARGE 0x400 /* STE-format control, large page */ +#define _SEGMENT_ENTRY_CO 0x100 /* change-recording override */ + #endif /* __s390x__ */ /* diff --git a/include/asm-s390/setup.h b/include/asm-s390/setup.h index 3a9e458fd8c3..ba69674012a7 100644 --- a/include/asm-s390/setup.h +++ b/include/asm-s390/setup.h @@ -69,6 +69,8 @@ extern unsigned long machine_flags; #define MACHINE_FLAG_DIAG9C (1UL << 7) #define MACHINE_FLAG_MVCOS (1UL << 8) #define MACHINE_FLAG_KVM (1UL << 9) +#define MACHINE_FLAG_HPAGE (1UL << 10) +#define MACHINE_FLAG_PFMF (1UL << 11) #define MACHINE_IS_VM (machine_flags & MACHINE_FLAG_VM) #define MACHINE_IS_KVM (machine_flags & MACHINE_FLAG_KVM) @@ -82,6 +84,8 @@ extern unsigned long machine_flags; #define MACHINE_HAS_DIAG44 (1) #define MACHINE_HAS_MVPG (machine_flags & MACHINE_FLAG_MVPG) #define MACHINE_HAS_MVCOS (0) +#define MACHINE_HAS_HPAGE (0) +#define MACHINE_HAS_PFMF (0) #else /* __s390x__ */ #define MACHINE_HAS_IEEE (1) #define MACHINE_HAS_CSP (1) @@ -89,6 +93,8 @@ extern unsigned long machine_flags; #define MACHINE_HAS_DIAG44 (machine_flags & MACHINE_FLAG_DIAG44) #define MACHINE_HAS_MVPG (1) #define MACHINE_HAS_MVCOS (machine_flags & MACHINE_FLAG_MVCOS) +#define MACHINE_HAS_HPAGE (machine_flags & MACHINE_FLAG_HPAGE) +#define MACHINE_HAS_PFMF (machine_flags & MACHINE_FLAG_PFMF) #endif /* __s390x__ */ #define MACHINE_HAS_SCLP (!MACHINE_IS_P390) diff --git a/include/asm-s390/tlbflush.h b/include/asm-s390/tlbflush.h index 9e57a93d7de1..d60394b9745e 100644 --- a/include/asm-s390/tlbflush.h +++ b/include/asm-s390/tlbflush.h @@ -2,6 +2,7 @@ #define _S390_TLBFLUSH_H #include +#include #include #include -- cgit v1.2.3 From 64275ea4f33636de198da5c78d0dbe31522555b0 Mon Sep 17 00:00:00 2001 From: David Chinner Date: Wed, 30 Apr 2008 17:11:16 +1000 Subject: [XFS] Include linux/random.h in all builds, not just debug. Noted-by: Stephen Rothwell Signed-off-by: Dave Chinner Signed-off-by: Linus Torvalds --- fs/xfs/linux-2.6/xfs_linux.h | 1 + fs/xfs/support/debug.h | 2 -- 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h index 1bc9f600365f..4edc46915b57 100644 --- a/fs/xfs/linux-2.6/xfs_linux.h +++ b/fs/xfs/linux-2.6/xfs_linux.h @@ -75,6 +75,7 @@ #include #include #include +#include #include #include diff --git a/fs/xfs/support/debug.h b/fs/xfs/support/debug.h index 855da0408647..75845f950814 100644 --- a/fs/xfs/support/debug.h +++ b/fs/xfs/support/debug.h @@ -49,8 +49,6 @@ extern void assfail(char *expr, char *f, int l); #else /* DEBUG */ -#include - #define ASSERT(expr) \ (unlikely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__)) -- cgit v1.2.3 From 2deb1acc653cbd5384b107d050d2deba089db2bd Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Wed, 30 Apr 2008 00:52:33 -0700 Subject: isofs: fix access to unallocated memory when reading corrupted filesystem When a directory on isofs is corrupted, we did not check whether length of the name in a directory entry and the length of the directory entry itself are consistent. This could lead to possible access beyond the end of buffer when the length of the name was too big. Add this sanity check to directory reading code. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/isofs/dir.c | 8 ++++++++ fs/isofs/namei.c | 7 +++++++ 2 files changed, 15 insertions(+) (limited to 'fs') diff --git a/fs/isofs/dir.c b/fs/isofs/dir.c index 1ba407c64df1..2f0dc5a14633 100644 --- a/fs/isofs/dir.c +++ b/fs/isofs/dir.c @@ -145,6 +145,14 @@ static int do_isofs_readdir(struct inode *inode, struct file *filp, } de = tmpde; } + /* Basic sanity check, whether name doesn't exceed dir entry */ + if (de_len < de->name_len[0] + + sizeof(struct iso_directory_record)) { + printk(KERN_NOTICE "iso9660: Corrupted directory entry" + " in block %lu of inode %lu\n", block, + inode->i_ino); + return -EIO; + } if (first_de) { isofs_normalize_block_and_offset(de, diff --git a/fs/isofs/namei.c b/fs/isofs/namei.c index 344b247bc29a..8299889a835e 100644 --- a/fs/isofs/namei.c +++ b/fs/isofs/namei.c @@ -111,6 +111,13 @@ isofs_find_entry(struct inode *dir, struct dentry *dentry, dlen = de->name_len[0]; dpnt = de->name; + /* Basic sanity check, whether name doesn't exceed dir entry */ + if (de_len < dlen + sizeof(struct iso_directory_record)) { + printk(KERN_NOTICE "iso9660: Corrupted directory entry" + " in block %lu of inode %lu\n", block, + dir->i_ino); + return 0; + } if (sbi->s_rock && ((i = get_rock_ridge_filename(de, tmpname, dir)))) { -- cgit v1.2.3 From 06fffb1267c9d986687b69d74a46ee332a50575e Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Wed, 30 Apr 2008 00:52:38 -0700 Subject: do_task_stat: don't take rcu_read_lock() lock_task_sighand() was changed, and do_task_stat() doesn't need rcu_read_lock any longer. sighand->siglock protects all "interesting" fields. Except: it doesn't protect ->tty->pgrp, but neither does rcu_read_lock(), this should be fixed. Signed-off-by: Oleg Nesterov Cc: "Eric W. Biederman" Cc: "Paul E. McKenney" Cc: Roland McGrath Cc: Alan Cox Cc: Pavel Emelyanov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/array.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs') diff --git a/fs/proc/array.c b/fs/proc/array.c index 07d6c4853fe8..b07a71002f2f 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -425,7 +425,6 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, cutime = cstime = utime = stime = cputime_zero; cgtime = gtime = cputime_zero; - rcu_read_lock(); if (lock_task_sighand(task, &flags)) { struct signal_struct *sig = task->signal; @@ -469,7 +468,6 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, unlock_task_sighand(task, &flags); } - rcu_read_unlock(); if (!whole || num_threads < 2) wchan = get_wchan(task); -- cgit v1.2.3 From 7a5e873f096e04e6d8719e4ecb7b70d2decca503 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Wed, 30 Apr 2008 00:53:04 -0700 Subject: signals: de_thread: simplify the ->child_reaper switching Now that we rely on SIGNAL_UNKILLABLE flag, de_thread() doesn't need the nasty hack to kill the old ->child_reaper during the mt-exec. This also means we can avoid taking tasklist_lock around zap_other_threads(). Signed-off-by: Oleg Nesterov Cc: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 15 ++------------- 1 file changed, 2 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/exec.c b/fs/exec.c index a13883903ee9..8fccc276d40d 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -766,9 +766,7 @@ static int de_thread(struct task_struct *tsk) /* * Kill all other threads in the thread group. - * We must hold tasklist_lock to call zap_other_threads. */ - read_lock(&tasklist_lock); spin_lock_irq(lock); if (signal_group_exit(sig)) { /* @@ -776,21 +774,10 @@ static int de_thread(struct task_struct *tsk) * return so that the signal is processed. */ spin_unlock_irq(lock); - read_unlock(&tasklist_lock); return -EAGAIN; } - - /* - * child_reaper ignores SIGKILL, change it now. - * Reparenting needs write_lock on tasklist_lock, - * so it is safe to do it under read_lock. - */ - if (unlikely(tsk->group_leader == task_child_reaper(tsk))) - task_active_pid_ns(tsk)->child_reaper = tsk; - sig->group_exit_task = tsk; zap_other_threads(tsk); - read_unlock(&tasklist_lock); /* Account for the thread group leader hanging around: */ count = thread_group_leader(tsk) ? 1 : 2; @@ -821,6 +808,8 @@ static int de_thread(struct task_struct *tsk) schedule(); } + if (unlikely(task_child_reaper(tsk) == leader)) + task_active_pid_ns(tsk)->child_reaper = tsk; /* * The only record we have of the real-time age of a * process, regardless of execs it's done, is start_time. -- cgit v1.2.3 From 4e4c22c71144c1b2e22c257ec6cf08ccb5be1165 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Wed, 30 Apr 2008 00:53:06 -0700 Subject: signals: add set_restore_sigmask This adds the set_restore_sigmask() inline in and replaces every set_thread_flag(TIF_RESTORE_SIGMASK) with a call to it. No change, but abstracts the details of the flag protocol from all the calls. Signed-off-by: Roland McGrath Cc: Oleg Nesterov Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Martin Schwidefsky Cc: Heiko Carstens Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/compat.c | 6 +++--- fs/eventpoll.c | 3 +-- fs/select.c | 4 ++-- include/linux/thread_info.h | 15 ++++++++++++++- kernel/compat.c | 3 +-- kernel/signal.c | 2 +- 6 files changed, 22 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/compat.c b/fs/compat.c index 2ce4456aad30..9964d542ae9e 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1720,7 +1720,7 @@ sticky: if (sigmask) { memcpy(¤t->saved_sigmask, &sigsaved, sizeof(sigsaved)); - set_thread_flag(TIF_RESTORE_SIGMASK); + set_restore_sigmask(); } } else if (sigmask) sigprocmask(SIG_SETMASK, &sigsaved, NULL); @@ -1791,7 +1791,7 @@ asmlinkage long compat_sys_ppoll(struct pollfd __user *ufds, if (sigmask) { memcpy(¤t->saved_sigmask, &sigsaved, sizeof(sigsaved)); - set_thread_flag(TIF_RESTORE_SIGMASK); + set_restore_sigmask(); } ret = -ERESTARTNOHAND; } else if (sigmask) @@ -2117,7 +2117,7 @@ asmlinkage long compat_sys_epoll_pwait(int epfd, if (err == -EINTR) { memcpy(¤t->saved_sigmask, &sigsaved, sizeof(sigsaved)); - set_thread_flag(TIF_RESTORE_SIGMASK); + set_restore_sigmask(); } else sigprocmask(SIG_SETMASK, &sigsaved, NULL); } diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 0d237182d721..71af2fc0041e 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1279,7 +1279,7 @@ asmlinkage long sys_epoll_pwait(int epfd, struct epoll_event __user *events, if (error == -EINTR) { memcpy(¤t->saved_sigmask, &sigsaved, sizeof(sigsaved)); - set_thread_flag(TIF_RESTORE_SIGMASK); + set_restore_sigmask(); } else sigprocmask(SIG_SETMASK, &sigsaved, NULL); } @@ -1309,4 +1309,3 @@ static int __init eventpoll_init(void) return 0; } fs_initcall(eventpoll_init); - diff --git a/fs/select.c b/fs/select.c index 00f58c5c7e05..32ce2b32fad1 100644 --- a/fs/select.c +++ b/fs/select.c @@ -498,7 +498,7 @@ sticky: if (sigmask) { memcpy(¤t->saved_sigmask, &sigsaved, sizeof(sigsaved)); - set_thread_flag(TIF_RESTORE_SIGMASK); + set_restore_sigmask(); } } else if (sigmask) sigprocmask(SIG_SETMASK, &sigsaved, NULL); @@ -805,7 +805,7 @@ asmlinkage long sys_ppoll(struct pollfd __user *ufds, unsigned int nfds, if (sigmask) { memcpy(¤t->saved_sigmask, &sigsaved, sizeof(sigsaved)); - set_thread_flag(TIF_RESTORE_SIGMASK); + set_restore_sigmask(); } ret = -ERESTARTNOHAND; } else if (sigmask) diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h index accd7bad35b0..43d8162c696e 100644 --- a/include/linux/thread_info.h +++ b/include/linux/thread_info.h @@ -92,6 +92,19 @@ static inline int test_ti_thread_flag(struct thread_info *ti, int flag) #define set_need_resched() set_thread_flag(TIF_NEED_RESCHED) #define clear_need_resched() clear_thread_flag(TIF_NEED_RESCHED) -#endif +#ifdef TIF_RESTORE_SIGMASK +/** + * set_restore_sigmask() - make sure saved_sigmask processing gets done + * + * This sets TIF_RESTORE_SIGMASK and ensures that the arch signal code + * will run before returning to user mode, to process the flag. + */ +static inline void set_restore_sigmask(void) +{ + set_thread_flag(TIF_RESTORE_SIGMASK); +} +#endif /* TIF_RESTORE_SIGMASK */ + +#endif /* __KERNEL__ */ #endif /* _LINUX_THREAD_INFO_H */ diff --git a/kernel/compat.c b/kernel/compat.c index e1ef04870c2a..4a856a3643bb 100644 --- a/kernel/compat.c +++ b/kernel/compat.c @@ -898,7 +898,7 @@ asmlinkage long compat_sys_rt_sigsuspend(compat_sigset_t __user *unewset, compat current->state = TASK_INTERRUPTIBLE; schedule(); - set_thread_flag(TIF_RESTORE_SIGMASK); + set_restore_sigmask(); return -ERESTARTNOHAND; } #endif /* __ARCH_WANT_COMPAT_SYS_RT_SIGSUSPEND */ @@ -1080,4 +1080,3 @@ compat_sys_sysinfo(struct compat_sysinfo __user *info) return 0; } - diff --git a/kernel/signal.c b/kernel/signal.c index 9ac737e53df1..72bb4f51f963 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -2541,7 +2541,7 @@ asmlinkage long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize) current->state = TASK_INTERRUPTIBLE; schedule(); - set_thread_flag(TIF_RESTORE_SIGMASK); + set_restore_sigmask(); return -ERESTARTNOHAND; } #endif /* __ARCH_WANT_SYS_RT_SIGSUSPEND */ -- cgit v1.2.3 From f3de272b821accbc8387211977c2de4f38468d05 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Wed, 30 Apr 2008 00:53:09 -0700 Subject: signals: use HAVE_SET_RESTORE_SIGMASK Change all the #ifdef TIF_RESTORE_SIGMASK conditionals in non-arch code to #ifdef HAVE_SET_RESTORE_SIGMASK. If arch code defines it first, the generic set_restore_sigmask() using TIF_RESTORE_SIGMASK is not defined. Signed-off-by: Roland McGrath Cc: Oleg Nesterov Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Martin Schwidefsky Cc: Heiko Carstens Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/compat.c | 8 ++++---- fs/eventpoll.c | 4 ++-- fs/select.c | 8 ++++---- include/linux/sched.h | 2 +- include/linux/thread_info.h | 10 ++++++++-- 5 files changed, 19 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/compat.c b/fs/compat.c index 9964d542ae9e..139dc93c092d 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1634,7 +1634,7 @@ sticky: return ret; } -#ifdef TIF_RESTORE_SIGMASK +#ifdef HAVE_SET_RESTORE_SIGMASK asmlinkage long compat_sys_pselect7(int n, compat_ulong_t __user *inp, compat_ulong_t __user *outp, compat_ulong_t __user *exp, struct compat_timespec __user *tsp, compat_sigset_t __user *sigmask, @@ -1825,7 +1825,7 @@ sticky: return ret; } -#endif /* TIF_RESTORE_SIGMASK */ +#endif /* HAVE_SET_RESTORE_SIGMASK */ #if defined(CONFIG_NFSD) || defined(CONFIG_NFSD_MODULE) /* Stuff for NFS server syscalls... */ @@ -2080,7 +2080,7 @@ long asmlinkage compat_sys_nfsservctl(int cmd, void *notused, void *notused2) #ifdef CONFIG_EPOLL -#ifdef TIF_RESTORE_SIGMASK +#ifdef HAVE_SET_RESTORE_SIGMASK asmlinkage long compat_sys_epoll_pwait(int epfd, struct compat_epoll_event __user *events, int maxevents, int timeout, @@ -2124,7 +2124,7 @@ asmlinkage long compat_sys_epoll_pwait(int epfd, return err; } -#endif /* TIF_RESTORE_SIGMASK */ +#endif /* HAVE_SET_RESTORE_SIGMASK */ #endif /* CONFIG_EPOLL */ diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 71af2fc0041e..221086fef174 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1241,7 +1241,7 @@ error_return: return error; } -#ifdef TIF_RESTORE_SIGMASK +#ifdef HAVE_SET_RESTORE_SIGMASK /* * Implement the event wait interface for the eventpoll file. It is the kernel @@ -1287,7 +1287,7 @@ asmlinkage long sys_epoll_pwait(int epfd, struct epoll_event __user *events, return error; } -#endif /* #ifdef TIF_RESTORE_SIGMASK */ +#endif /* HAVE_SET_RESTORE_SIGMASK */ static int __init eventpoll_init(void) { diff --git a/fs/select.c b/fs/select.c index 32ce2b32fad1..2c292146e246 100644 --- a/fs/select.c +++ b/fs/select.c @@ -425,7 +425,7 @@ sticky: return ret; } -#ifdef TIF_RESTORE_SIGMASK +#ifdef HAVE_SET_RESTORE_SIGMASK asmlinkage long sys_pselect7(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, struct timespec __user *tsp, const sigset_t __user *sigmask, size_t sigsetsize) @@ -528,7 +528,7 @@ asmlinkage long sys_pselect6(int n, fd_set __user *inp, fd_set __user *outp, return sys_pselect7(n, inp, outp, exp, tsp, up, sigsetsize); } -#endif /* TIF_RESTORE_SIGMASK */ +#endif /* HAVE_SET_RESTORE_SIGMASK */ struct poll_list { struct poll_list *next; @@ -759,7 +759,7 @@ asmlinkage long sys_poll(struct pollfd __user *ufds, unsigned int nfds, return ret; } -#ifdef TIF_RESTORE_SIGMASK +#ifdef HAVE_SET_RESTORE_SIGMASK asmlinkage long sys_ppoll(struct pollfd __user *ufds, unsigned int nfds, struct timespec __user *tsp, const sigset_t __user *sigmask, size_t sigsetsize) @@ -839,4 +839,4 @@ asmlinkage long sys_ppoll(struct pollfd __user *ufds, unsigned int nfds, return ret; } -#endif /* TIF_RESTORE_SIGMASK */ +#endif /* HAVE_SET_RESTORE_SIGMASK */ diff --git a/include/linux/sched.h b/include/linux/sched.h index fe970cdca83c..86e60796db62 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1175,7 +1175,7 @@ struct task_struct { struct sighand_struct *sighand; sigset_t blocked, real_blocked; - sigset_t saved_sigmask; /* To be restored with TIF_RESTORE_SIGMASK */ + sigset_t saved_sigmask; /* restored if set_restore_sigmask() was used */ struct sigpending pending; unsigned long sas_ss_sp; diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h index 81c5f82f0663..38a56477f27a 100644 --- a/include/linux/thread_info.h +++ b/include/linux/thread_info.h @@ -92,7 +92,13 @@ static inline int test_ti_thread_flag(struct thread_info *ti, int flag) #define set_need_resched() set_thread_flag(TIF_NEED_RESCHED) #define clear_need_resched() clear_thread_flag(TIF_NEED_RESCHED) -#ifdef TIF_RESTORE_SIGMASK +#if defined TIF_RESTORE_SIGMASK && !defined HAVE_SET_RESTORE_SIGMASK +/* + * An arch can define its own version of set_restore_sigmask() to get the + * job done however works, with or without TIF_RESTORE_SIGMASK. + */ +#define HAVE_SET_RESTORE_SIGMASK 1 + /** * set_restore_sigmask() - make sure saved_sigmask processing gets done * @@ -109,7 +115,7 @@ static inline void set_restore_sigmask(void) set_thread_flag(TIF_RESTORE_SIGMASK); set_thread_flag(TIF_SIGPENDING); } -#endif /* TIF_RESTORE_SIGMASK */ +#endif /* TIF_RESTORE_SIGMASK && !HAVE_SET_RESTORE_SIGMASK */ #endif /* __KERNEL__ */ -- cgit v1.2.3 From 2800d8d19e51414403df8144eaa214bb03400b87 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Wed, 30 Apr 2008 00:53:12 -0700 Subject: document de_thread() with exit_notify() connection Add a couple of small comments, it is not easy to see what this code does. Signed-off-by: Oleg Nesterov Cc: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 2 +- kernel/exit.c | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/exec.c b/fs/exec.c index 8fccc276d40d..9f9f931ef949 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -798,7 +798,7 @@ static int de_thread(struct task_struct *tsk) if (!thread_group_leader(tsk)) { leader = tsk->group_leader; - sig->notify_count = -1; + sig->notify_count = -1; /* for exit_notify() */ for (;;) { write_lock_irq(&tasklist_lock); if (likely(leader->exit_state)) diff --git a/kernel/exit.c b/kernel/exit.c index 413c81ec858e..879ed6e1c883 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -877,6 +877,7 @@ static void exit_notify(struct task_struct *tsk, int group_dead) state = EXIT_DEAD; tsk->exit_state = state; + /* mt-exec, de_thread() is waiting for us */ if (thread_group_leader(tsk) && tsk->signal->notify_count < 0 && tsk->signal->group_exit_task) -- cgit v1.2.3 From 04f378b198da233ca0aca341b113dc6579d46123 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 30 Apr 2008 00:53:29 -0700 Subject: tty: BKL pushdown - Push the BKL down into the line disciplines - Switch the tty layer to unlocked_ioctl - Introduce a new ctrl_lock spin lock for the control bits - Eliminate much of the lock_kernel use in n_tty - Prepare to (but don't yet) call the drivers with the lock dropped on the paths that historically held the lock BKL now primarily protects open/close/ldisc change in the tty layer [jirislaby@gmail.com: a couple of fixes] Signed-off-by: Alan Cox Signed-off-by: Jiri Slaby Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/n_hdlc.c | 24 ++++++++--- drivers/char/n_r3964.c | 16 ++++++- drivers/char/n_tty.c | 32 ++++++++++---- drivers/char/pty.c | 3 ++ drivers/char/tty_io.c | 107 ++++++++++++++++++++++++++++++++--------------- drivers/char/tty_ioctl.c | 6 +++ drivers/char/vt.c | 8 +++- fs/compat_ioctl.c | 2 +- include/linux/tty.h | 4 +- 9 files changed, 149 insertions(+), 53 deletions(-) (limited to 'fs') diff --git a/drivers/char/n_hdlc.c b/drivers/char/n_hdlc.c index 06803ed5568c..a07c0af4819e 100644 --- a/drivers/char/n_hdlc.c +++ b/drivers/char/n_hdlc.c @@ -578,26 +578,36 @@ static ssize_t n_hdlc_tty_read(struct tty_struct *tty, struct file *file, return -EFAULT; } + lock_kernel(); + for (;;) { - if (test_bit(TTY_OTHER_CLOSED, &tty->flags)) + if (test_bit(TTY_OTHER_CLOSED, &tty->flags)) { + unlock_kernel(); return -EIO; + } n_hdlc = tty2n_hdlc (tty); if (!n_hdlc || n_hdlc->magic != HDLC_MAGIC || - tty != n_hdlc->tty) + tty != n_hdlc->tty) { + unlock_kernel(); return 0; + } rbuf = n_hdlc_buf_get(&n_hdlc->rx_buf_list); if (rbuf) break; /* no data */ - if (file->f_flags & O_NONBLOCK) + if (file->f_flags & O_NONBLOCK) { + unlock_kernel(); return -EAGAIN; + } interruptible_sleep_on (&tty->read_wait); - if (signal_pending(current)) + if (signal_pending(current)) { + unlock_kernel(); return -EINTR; + } } if (rbuf->count > nr) @@ -618,7 +628,7 @@ static ssize_t n_hdlc_tty_read(struct tty_struct *tty, struct file *file, kfree(rbuf); else n_hdlc_buf_put(&n_hdlc->rx_free_buf_list,rbuf); - + unlock_kernel(); return ret; } /* end of n_hdlc_tty_read() */ @@ -661,6 +671,8 @@ static ssize_t n_hdlc_tty_write(struct tty_struct *tty, struct file *file, count = maxframe; } + lock_kernel(); + add_wait_queue(&tty->write_wait, &wait); set_current_state(TASK_INTERRUPTIBLE); @@ -695,7 +707,7 @@ static ssize_t n_hdlc_tty_write(struct tty_struct *tty, struct file *file, n_hdlc_buf_put(&n_hdlc->tx_buf_list,tbuf); n_hdlc_send_frames(n_hdlc,tty); } - + unlock_kernel(); return error; } /* end of n_hdlc_tty_write() */ diff --git a/drivers/char/n_r3964.c b/drivers/char/n_r3964.c index 6b918b80f73e..3f6486e9f1ec 100644 --- a/drivers/char/n_r3964.c +++ b/drivers/char/n_r3964.c @@ -1075,12 +1075,15 @@ static ssize_t r3964_read(struct tty_struct *tty, struct file *file, TRACE_L("read()"); + lock_kernel(); + pClient = findClient(pInfo, task_pid(current)); if (pClient) { pMsg = remove_msg(pInfo, pClient); if (pMsg == NULL) { /* no messages available. */ if (file->f_flags & O_NONBLOCK) { + unlock_kernel(); return -EAGAIN; } /* block until there is a message: */ @@ -1090,8 +1093,10 @@ static ssize_t r3964_read(struct tty_struct *tty, struct file *file, /* If we still haven't got a message, we must have been signalled */ - if (!pMsg) + if (!pMsg) { + unlock_kernel(); return -EINTR; + } /* deliver msg to client process: */ theMsg.msg_id = pMsg->msg_id; @@ -1102,12 +1107,15 @@ static ssize_t r3964_read(struct tty_struct *tty, struct file *file, kfree(pMsg); TRACE_M("r3964_read - msg kfree %p", pMsg); - if (copy_to_user(buf, &theMsg, count)) + if (copy_to_user(buf, &theMsg, count)) { + unlock_kernel(); return -EFAULT; + } TRACE_PS("read - return %d", count); return count; } + unlock_kernel(); return -EPERM; } @@ -1156,6 +1164,8 @@ static ssize_t r3964_write(struct tty_struct *tty, struct file *file, pHeader->locks = 0; pHeader->owner = NULL; + lock_kernel(); + pClient = findClient(pInfo, task_pid(current)); if (pClient) { pHeader->owner = pClient; @@ -1173,6 +1183,8 @@ static ssize_t r3964_write(struct tty_struct *tty, struct file *file, add_tx_queue(pInfo, pHeader); trigger_transmit(pInfo); + unlock_kernel(); + return 0; } diff --git a/drivers/char/n_tty.c b/drivers/char/n_tty.c index 0c09409fa45d..001d9d875387 100644 --- a/drivers/char/n_tty.c +++ b/drivers/char/n_tty.c @@ -183,22 +183,24 @@ static void reset_buffer_flags(struct tty_struct *tty) * at hangup) or when the N_TTY line discipline internally has to * clean the pending queue (for example some signals). * - * FIXME: tty->ctrl_status is not spinlocked and relies on - * lock_kernel() still. + * Locking: ctrl_lock */ static void n_tty_flush_buffer(struct tty_struct *tty) { + unsigned long flags; /* clear everything and unthrottle the driver */ reset_buffer_flags(tty); if (!tty->link) return; + spin_lock_irqsave(&tty->ctrl_lock, flags); if (tty->link->packet) { tty->ctrl_status |= TIOCPKT_FLUSHREAD; wake_up_interruptible(&tty->link->read_wait); } + spin_unlock_irqrestore(&tty->ctrl_lock, flags); } /** @@ -264,7 +266,7 @@ static inline int is_continuation(unsigned char c, struct tty_struct *tty) * relevant in the world today. If you ever need them, add them here. * * Called from both the receive and transmit sides and can be called - * re-entrantly. Relies on lock_kernel() still. + * re-entrantly. Relies on lock_kernel() for tty->column state. */ static int opost(unsigned char c, struct tty_struct *tty) @@ -275,6 +277,7 @@ static int opost(unsigned char c, struct tty_struct *tty) if (!space) return -1; + lock_kernel(); if (O_OPOST(tty)) { switch (c) { case '\n': @@ -323,6 +326,7 @@ static int opost(unsigned char c, struct tty_struct *tty) } } tty->driver->put_char(tty, c); + unlock_kernel(); return 0; } @@ -337,7 +341,8 @@ static int opost(unsigned char c, struct tty_struct *tty) * the simple cases normally found and helps to generate blocks of * symbols for the console driver and thus improve performance. * - * Called from write_chan under the tty layer write lock. + * Called from write_chan under the tty layer write lock. Relies + * on lock_kernel for the tty->column state. */ static ssize_t opost_block(struct tty_struct *tty, @@ -353,6 +358,7 @@ static ssize_t opost_block(struct tty_struct *tty, if (nr > space) nr = space; + lock_kernel(); for (i = 0, cp = buf; i < nr; i++, cp++) { switch (*cp) { case '\n': @@ -387,6 +393,7 @@ break_out: if (tty->driver->flush_chars) tty->driver->flush_chars(tty); i = tty->driver->write(tty, buf, i); + unlock_kernel(); return i; } @@ -1194,6 +1201,11 @@ extern ssize_t redirected_tty_write(struct file *, const char __user *, * Perform job control management checks on this file/tty descriptor * and if appropriate send any needed signals and return a negative * error code if action should be taken. + * + * FIXME: + * Locking: None - redirected write test is safe, testing + * current->signal should possibly lock current->sighand + * pgrp locking ? */ static int job_control(struct tty_struct *tty, struct file *file) @@ -1246,6 +1258,7 @@ static ssize_t read_chan(struct tty_struct *tty, struct file *file, ssize_t size; long timeout; unsigned long flags; + int packet; do_it_again: @@ -1289,16 +1302,19 @@ do_it_again: if (mutex_lock_interruptible(&tty->atomic_read_lock)) return -ERESTARTSYS; } + packet = tty->packet; add_wait_queue(&tty->read_wait, &wait); while (nr) { /* First test for status change. */ - if (tty->packet && tty->link->ctrl_status) { + if (packet && tty->link->ctrl_status) { unsigned char cs; if (b != buf) break; + spin_lock_irqsave(&tty->link->ctrl_lock, flags); cs = tty->link->ctrl_status; tty->link->ctrl_status = 0; + spin_unlock_irqrestore(&tty->link->ctrl_lock, flags); if (tty_put_user(tty, cs, b++)) { retval = -EFAULT; b--; @@ -1333,6 +1349,7 @@ do_it_again: retval = -ERESTARTSYS; break; } + /* FIXME: does n_tty_set_room need locking ? */ n_tty_set_room(tty); timeout = schedule_timeout(timeout); continue; @@ -1340,7 +1357,7 @@ do_it_again: __set_current_state(TASK_RUNNING); /* Deal with packet mode. */ - if (tty->packet && b == buf) { + if (packet && b == buf) { if (tty_put_user(tty, TIOCPKT_DATA, b++)) { retval = -EFAULT; b--; @@ -1388,6 +1405,8 @@ do_it_again: break; } else { int uncopied; + /* The copy function takes the read lock and handles + locking internally for this case */ uncopied = copy_from_read_buf(tty, &b, &nr); uncopied += copy_from_read_buf(tty, &b, &nr); if (uncopied) { @@ -1429,7 +1448,6 @@ do_it_again: goto do_it_again; n_tty_set_room(tty); - return retval; } diff --git a/drivers/char/pty.c b/drivers/char/pty.c index 706ff34728f1..6288356b769d 100644 --- a/drivers/char/pty.c +++ b/drivers/char/pty.c @@ -181,6 +181,7 @@ static int pty_set_lock(struct tty_struct *tty, int __user * arg) static void pty_flush_buffer(struct tty_struct *tty) { struct tty_struct *to = tty->link; + unsigned long flags; if (!to) return; @@ -189,8 +190,10 @@ static void pty_flush_buffer(struct tty_struct *tty) to->ldisc.flush_buffer(to); if (to->packet) { + spin_lock_irqsave(&tty->ctrl_lock, flags); tty->ctrl_status |= TIOCPKT_FLUSHWRITE; wake_up_interruptible(&to->read_wait); + spin_unlock_irqrestore(&tty->ctrl_lock, flags); } } diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index 2fa6856706ab..0b0354bc28d6 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -152,8 +152,7 @@ ssize_t redirected_tty_write(struct file *, const char __user *, static unsigned int tty_poll(struct file *, poll_table *); static int tty_open(struct inode *, struct file *); static int tty_release(struct inode *, struct file *); -int tty_ioctl(struct inode *inode, struct file *file, - unsigned int cmd, unsigned long arg); +long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg); #ifdef CONFIG_COMPAT static long tty_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg); @@ -1205,7 +1204,7 @@ EXPORT_SYMBOL_GPL(tty_find_polling_driver); * not in the foreground, send a SIGTTOU. If the signal is blocked or * ignored, go ahead and perform the operation. (POSIX 7.2) * - * Locking: none + * Locking: none - FIXME: review this */ int tty_check_change(struct tty_struct *tty) @@ -1247,8 +1246,8 @@ static unsigned int hung_up_tty_poll(struct file *filp, poll_table *wait) return POLLIN | POLLOUT | POLLERR | POLLHUP | POLLRDNORM | POLLWRNORM; } -static int hung_up_tty_ioctl(struct inode *inode, struct file *file, - unsigned int cmd, unsigned long arg) +static long hung_up_tty_ioctl(struct file *file, unsigned int cmd, + unsigned long arg) { return cmd == TIOCSPGRP ? -ENOTTY : -EIO; } @@ -1264,7 +1263,7 @@ static const struct file_operations tty_fops = { .read = tty_read, .write = tty_write, .poll = tty_poll, - .ioctl = tty_ioctl, + .unlocked_ioctl = tty_ioctl, .compat_ioctl = tty_compat_ioctl, .open = tty_open, .release = tty_release, @@ -1277,7 +1276,7 @@ static const struct file_operations ptmx_fops = { .read = tty_read, .write = tty_write, .poll = tty_poll, - .ioctl = tty_ioctl, + .unlocked_ioctl = tty_ioctl, .compat_ioctl = tty_compat_ioctl, .open = ptmx_open, .release = tty_release, @@ -1290,7 +1289,7 @@ static const struct file_operations console_fops = { .read = tty_read, .write = redirected_tty_write, .poll = tty_poll, - .ioctl = tty_ioctl, + .unlocked_ioctl = tty_ioctl, .compat_ioctl = tty_compat_ioctl, .open = tty_open, .release = tty_release, @@ -1302,7 +1301,7 @@ static const struct file_operations hung_up_tty_fops = { .read = hung_up_tty_read, .write = hung_up_tty_write, .poll = hung_up_tty_poll, - .ioctl = hung_up_tty_ioctl, + .unlocked_ioctl = hung_up_tty_ioctl, .compat_ioctl = hung_up_tty_compat_ioctl, .release = tty_release, }; @@ -1626,16 +1625,17 @@ void disassociate_ctty(int on_exit) struct tty_struct *tty; struct pid *tty_pgrp = NULL; - lock_kernel(); mutex_lock(&tty_mutex); tty = get_current_tty(); if (tty) { tty_pgrp = get_pid(tty->pgrp); mutex_unlock(&tty_mutex); + lock_kernel(); /* XXX: here we race, there is nothing protecting tty */ if (on_exit && tty->driver->type != TTY_DRIVER_TYPE_PTY) tty_vhangup(tty); + unlock_kernel(); } else if (on_exit) { struct pid *old_pgrp; spin_lock_irq(¤t->sighand->siglock); @@ -1648,7 +1648,6 @@ void disassociate_ctty(int on_exit) put_pid(old_pgrp); } mutex_unlock(&tty_mutex); - unlock_kernel(); return; } if (tty_pgrp) { @@ -1683,7 +1682,6 @@ void disassociate_ctty(int on_exit) read_lock(&tasklist_lock); session_clear_tty(task_session(current)); read_unlock(&tasklist_lock); - unlock_kernel(); } /** @@ -1693,8 +1691,10 @@ void disassociate_ctty(int on_exit) void no_tty(void) { struct task_struct *tsk = current; + lock_kernel(); if (tsk->signal->leader) disassociate_ctty(0); + unlock_kernel(); proc_clear_tty(tsk); } @@ -1714,19 +1714,24 @@ void no_tty(void) * but not always. * * Locking: - * Broken. Relies on BKL which is unsafe here. + * Uses the tty control lock internally */ void stop_tty(struct tty_struct *tty) { - if (tty->stopped) + unsigned long flags; + spin_lock_irqsave(&tty->ctrl_lock, flags); + if (tty->stopped) { + spin_unlock_irqrestore(&tty->ctrl_lock, flags); return; + } tty->stopped = 1; if (tty->link && tty->link->packet) { tty->ctrl_status &= ~TIOCPKT_START; tty->ctrl_status |= TIOCPKT_STOP; wake_up_interruptible(&tty->link->read_wait); } + spin_unlock_irqrestore(&tty->ctrl_lock, flags); if (tty->driver->stop) (tty->driver->stop)(tty); } @@ -1743,19 +1748,24 @@ EXPORT_SYMBOL(stop_tty); * driver start method is invoked and the line discipline woken. * * Locking: - * Broken. Relies on BKL which is unsafe here. + * ctrl_lock */ void start_tty(struct tty_struct *tty) { - if (!tty->stopped || tty->flow_stopped) + unsigned long flags; + spin_lock_irqsave(&tty->ctrl_lock, flags); + if (!tty->stopped || tty->flow_stopped) { + spin_unlock_irqrestore(&tty->ctrl_lock, flags); return; + } tty->stopped = 0; if (tty->link && tty->link->packet) { tty->ctrl_status &= ~TIOCPKT_STOP; tty->ctrl_status |= TIOCPKT_START; wake_up_interruptible(&tty->link->read_wait); } + spin_unlock_irqrestore(&tty->ctrl_lock, flags); if (tty->driver->start) (tty->driver->start)(tty); /* If we have a running line discipline it may need kicking */ @@ -1799,13 +1809,11 @@ static ssize_t tty_read(struct file *file, char __user *buf, size_t count, /* We want to wait for the line discipline to sort out in this situation */ ld = tty_ldisc_ref_wait(tty); - lock_kernel(); if (ld->read) i = (ld->read)(tty, file, buf, count); else i = -EIO; tty_ldisc_deref(ld); - unlock_kernel(); if (i > 0) inode->i_atime = current_fs_time(inode->i_sb); return i; @@ -1893,9 +1901,7 @@ static inline ssize_t do_tty_write( ret = -EFAULT; if (copy_from_user(tty->write_buf, buf, size)) break; - lock_kernel(); ret = write(tty, file, tty->write_buf, size); - unlock_kernel(); if (ret <= 0) break; written += ret; @@ -3070,10 +3076,13 @@ static int fionbio(struct file *file, int __user *p) if (get_user(nonblock, p)) return -EFAULT; + /* file->f_flags is still BKL protected in the fs layer - vomit */ + lock_kernel(); if (nonblock) file->f_flags |= O_NONBLOCK; else file->f_flags &= ~O_NONBLOCK; + unlock_kernel(); return 0; } @@ -3162,7 +3171,7 @@ static int tiocgpgrp(struct tty_struct *tty, struct tty_struct *real_tty, pid_t * Set the process group of the tty to the session passed. Only * permitted where the tty session is our session. * - * Locking: None + * Locking: RCU */ static int tiocspgrp(struct tty_struct *tty, struct tty_struct *real_tty, pid_t __user *p) @@ -3237,10 +3246,16 @@ static int tiocgsid(struct tty_struct *tty, struct tty_struct *real_tty, pid_t _ static int tiocsetd(struct tty_struct *tty, int __user *p) { int ldisc; + int ret; if (get_user(ldisc, p)) return -EFAULT; - return tty_set_ldisc(tty, ldisc); + + lock_kernel(); + ret = tty_set_ldisc(tty, ldisc); + unlock_kernel(); + + return ret; } /** @@ -3258,16 +3273,21 @@ static int tiocsetd(struct tty_struct *tty, int __user *p) static int send_break(struct tty_struct *tty, unsigned int duration) { + int retval = -EINTR; + + lock_kernel(); if (tty_write_lock(tty, 0) < 0) - return -EINTR; + goto out; tty->driver->break_ctl(tty, -1); if (!signal_pending(current)) msleep_interruptible(duration); tty->driver->break_ctl(tty, 0); tty_write_unlock(tty); - if (signal_pending(current)) - return -EINTR; - return 0; + if (!signal_pending(current)) + retval = 0; +out: + unlock_kernel(); + return retval; } /** @@ -3287,7 +3307,9 @@ static int tty_tiocmget(struct tty_struct *tty, struct file *file, int __user *p int retval = -EINVAL; if (tty->driver->tiocmget) { + lock_kernel(); retval = tty->driver->tiocmget(tty, file); + unlock_kernel(); if (retval >= 0) retval = put_user(retval, p); @@ -3337,7 +3359,9 @@ static int tty_tiocmset(struct tty_struct *tty, struct file *file, unsigned int set &= TIOCM_DTR|TIOCM_RTS|TIOCM_OUT1|TIOCM_OUT2|TIOCM_LOOP; clear &= TIOCM_DTR|TIOCM_RTS|TIOCM_OUT1|TIOCM_OUT2|TIOCM_LOOP; + lock_kernel(); retval = tty->driver->tiocmset(tty, file, set, clear); + unlock_kernel(); } return retval; } @@ -3345,20 +3369,18 @@ static int tty_tiocmset(struct tty_struct *tty, struct file *file, unsigned int /* * Split this up, as gcc can choke on it otherwise.. */ -int tty_ioctl(struct inode *inode, struct file *file, - unsigned int cmd, unsigned long arg) +long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { struct tty_struct *tty, *real_tty; void __user *p = (void __user *)arg; int retval; struct tty_ldisc *ld; + struct inode *inode = file->f_dentry->d_inode; tty = (struct tty_struct *)file->private_data; if (tty_paranoia_check(tty, inode, "tty_ioctl")) return -EINVAL; - /* CHECKME: is this safe as one end closes ? */ - real_tty = tty; if (tty->driver->type == TTY_DRIVER_TYPE_PTY && tty->driver->subtype == PTY_TYPE_MASTER) @@ -3367,13 +3389,19 @@ int tty_ioctl(struct inode *inode, struct file *file, /* * Break handling by driver */ + + retval = -EINVAL; + if (!tty->driver->break_ctl) { switch (cmd) { case TIOCSBRK: case TIOCCBRK: - if (tty->driver->ioctl) - return tty->driver->ioctl(tty, file, cmd, arg); - return -EINVAL; + if (tty->driver->ioctl) { + lock_kernel(); + retval = tty->driver->ioctl(tty, file, cmd, arg); + unlock_kernel(); + } + return retval; /* These two ioctl's always return success; even if */ /* the driver doesn't support them. */ @@ -3381,7 +3409,9 @@ int tty_ioctl(struct inode *inode, struct file *file, case TCSBRKP: if (!tty->driver->ioctl) return 0; + lock_kernel(); retval = tty->driver->ioctl(tty, file, cmd, arg); + unlock_kernel(); if (retval == -ENOIOCTLCMD) retval = 0; return retval; @@ -3401,7 +3431,9 @@ int tty_ioctl(struct inode *inode, struct file *file, if (retval) return retval; if (cmd != TIOCCBRK) { + lock_kernel(); tty_wait_until_sent(tty, 0); + unlock_kernel(); if (signal_pending(current)) return -EINTR; } @@ -3451,11 +3483,15 @@ int tty_ioctl(struct inode *inode, struct file *file, * Break handling */ case TIOCSBRK: /* Turn break on, unconditionally */ + lock_kernel(); tty->driver->break_ctl(tty, -1); + unlock_kernel(); return 0; case TIOCCBRK: /* Turn break off, unconditionally */ + lock_kernel(); tty->driver->break_ctl(tty, 0); + unlock_kernel(); return 0; case TCSBRK: /* SVID version: non-zero arg --> no break */ /* non-zero arg means wait for all output data @@ -3485,14 +3521,18 @@ int tty_ioctl(struct inode *inode, struct file *file, break; } if (tty->driver->ioctl) { + lock_kernel(); retval = (tty->driver->ioctl)(tty, file, cmd, arg); + unlock_kernel(); if (retval != -ENOIOCTLCMD) return retval; } ld = tty_ldisc_ref_wait(tty); retval = -EINVAL; if (ld->ioctl) { + lock_kernel(); retval = ld->ioctl(tty, file, cmd, arg); + unlock_kernel(); if (retval == -ENOIOCTLCMD) retval = -EINVAL; } @@ -3770,6 +3810,7 @@ static void initialize_tty_struct(struct tty_struct *tty) mutex_init(&tty->atomic_read_lock); mutex_init(&tty->atomic_write_lock); spin_lock_init(&tty->read_lock); + spin_lock_init(&tty->ctrl_lock); INIT_LIST_HEAD(&tty->tty_files); INIT_WORK(&tty->SAK_work, do_SAK_work); } diff --git a/drivers/char/tty_ioctl.c b/drivers/char/tty_ioctl.c index f95a80b2265f..d6353d89b451 100644 --- a/drivers/char/tty_ioctl.c +++ b/drivers/char/tty_ioctl.c @@ -395,6 +395,7 @@ static void change_termios(struct tty_struct *tty, struct ktermios *new_termios) int canon_change; struct ktermios old_termios = *tty->termios; struct tty_ldisc *ld; + unsigned long flags; /* * Perform the actual termios internal changes under lock. @@ -429,11 +430,13 @@ static void change_termios(struct tty_struct *tty, struct ktermios *new_termios) STOP_CHAR(tty) == '\023' && START_CHAR(tty) == '\021'); if (old_flow != new_flow) { + spin_lock_irqsave(&tty->ctrl_lock, flags); tty->ctrl_status &= ~(TIOCPKT_DOSTOP | TIOCPKT_NOSTOP); if (new_flow) tty->ctrl_status |= TIOCPKT_DOSTOP; else tty->ctrl_status |= TIOCPKT_NOSTOP; + spin_unlock_irqrestore(&tty->ctrl_lock, flags); wake_up_interruptible(&tty->link->read_wait); } } @@ -905,6 +908,7 @@ int n_tty_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd, unsigned long arg) { struct tty_struct *real_tty; + unsigned long flags; int retval; if (tty->driver->type == TTY_DRIVER_TYPE_PTY && @@ -963,6 +967,7 @@ int n_tty_ioctl(struct tty_struct *tty, struct file *file, return -ENOTTY; if (get_user(pktmode, (int __user *) arg)) return -EFAULT; + spin_lock_irqsave(&tty->ctrl_lock, flags); if (pktmode) { if (!tty->packet) { tty->packet = 1; @@ -970,6 +975,7 @@ int n_tty_ioctl(struct tty_struct *tty, struct file *file, } } else tty->packet = 0; + spin_unlock_irqrestore(&tty->ctrl_lock, flags); return 0; } default: diff --git a/drivers/char/vt.c b/drivers/char/vt.c index 1c2660477135..e64f0bf3624e 100644 --- a/drivers/char/vt.c +++ b/drivers/char/vt.c @@ -2541,6 +2541,9 @@ int tioclinux(struct tty_struct *tty, unsigned long arg) if (get_user(type, p)) return -EFAULT; ret = 0; + + lock_kernel(); + switch (type) { case TIOCL_SETSEL: @@ -2560,7 +2563,7 @@ int tioclinux(struct tty_struct *tty, unsigned long arg) ret = sel_loadlut(p); break; case TIOCL_GETSHIFTSTATE: - + /* * Make it possible to react to Shift+Mousebutton. * Note that 'shift_state' is an undocumented @@ -2615,6 +2618,7 @@ int tioclinux(struct tty_struct *tty, unsigned long arg) ret = -EINVAL; break; } + unlock_kernel(); return ret; } @@ -3829,7 +3833,7 @@ static int con_font_get(struct vc_data *vc, struct console_font_op *op) goto out; c = (font.width+7)/8 * 32 * font.charcount; - + if (op->data && font.charcount > op->charcount) rc = -ENOSPC; if (!(op->flags & KD_FONT_FLAG_OLD)) { diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index c6e72aebd16b..9663e8776724 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -1046,7 +1046,7 @@ static int vt_check(struct file *file) struct inode *inode = file->f_path.dentry->d_inode; struct vc_data *vc; - if (file->f_op->ioctl != tty_ioctl) + if (file->f_op->unlocked_ioctl != tty_ioctl) return -EINVAL; tty = (struct tty_struct *)file->private_data; diff --git a/include/linux/tty.h b/include/linux/tty.h index 265831ccaa88..4d3702bade03 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -183,6 +183,7 @@ struct tty_struct { int index; struct tty_ldisc ldisc; struct mutex termios_mutex; + spinlock_t ctrl_lock; struct ktermios *termios, *termios_locked; char name[64]; struct pid *pgrp; @@ -323,8 +324,7 @@ extern void tty_ldisc_put(int); extern void tty_wakeup(struct tty_struct *tty); extern void tty_ldisc_flush(struct tty_struct *tty); -extern int tty_ioctl(struct inode *inode, struct file *file, unsigned int cmd, - unsigned long arg); +extern long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg); extern int tty_mode_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd, unsigned long arg); extern int tty_perform_flush(struct tty_struct *tty, unsigned long arg); -- cgit v1.2.3 From 5d0fdf1e01899805b6c2c0b789a707dcb731b1ea Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 30 Apr 2008 00:53:31 -0700 Subject: tty_io: fix remaining pid struct locking This fixes the last couple of pid struct locking failures I know about. [oleg@tv-sign.ru: clean up do_task_stat()] Signed-off-by: Alan Cox Signed-off-by: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/tty_io.c | 28 +++++++++++++++++++++++++++- fs/proc/array.c | 4 +++- include/linux/tty.h | 1 + 3 files changed, 31 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index c8aa318eaa18..2460c4c76161 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -3173,6 +3173,27 @@ unlock: return ret; } +/** + * tty_get_pgrp - return a ref counted pgrp pid + * @tty: tty to read + * + * Returns a refcounted instance of the pid struct for the process + * group controlling the tty. + */ + +struct pid *tty_get_pgrp(struct tty_struct *tty) +{ + unsigned long flags; + struct pid *pgrp; + + spin_lock_irqsave(&tty->ctrl_lock, flags); + pgrp = get_pid(tty->pgrp); + spin_unlock_irqrestore(&tty->ctrl_lock, flags); + + return pgrp; +} +EXPORT_SYMBOL_GPL(tty_get_pgrp); + /** * tiocgpgrp - get process group * @tty: tty passed by user @@ -3187,13 +3208,18 @@ unlock: static int tiocgpgrp(struct tty_struct *tty, struct tty_struct *real_tty, pid_t __user *p) { + struct pid *pid; + int ret; /* * (tty == real_tty) is a cheap way of * testing if the tty is NOT a master pty. */ if (tty == real_tty && current->signal->tty != real_tty) return -ENOTTY; - return put_user(pid_vnr(real_tty->pgrp), p); + pid = tty_get_pgrp(real_tty); + ret = put_user(pid_vnr(pid), p); + put_pid(pid); + return ret; } /** diff --git a/fs/proc/array.c b/fs/proc/array.c index b07a71002f2f..c135cbdd9127 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -429,7 +429,9 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, struct signal_struct *sig = task->signal; if (sig->tty) { - tty_pgrp = pid_nr_ns(sig->tty->pgrp, ns); + struct pid *pgrp = tty_get_pgrp(sig->tty); + tty_pgrp = pid_nr_ns(pgrp, ns); + put_pid(pgrp); tty_nr = new_encode_dev(tty_devnum(sig->tty)); } diff --git a/include/linux/tty.h b/include/linux/tty.h index 381085e45cca..2699298b00ef 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -297,6 +297,7 @@ extern int tty_read_raw_data(struct tty_struct *tty, unsigned char *bufp, extern void tty_write_message(struct tty_struct *tty, char *msg); extern int is_current_pgrp_orphaned(void); +extern struct pid *tty_get_pgrp(struct tty_struct *tty); extern int is_ignored(int sig); extern int tty_signal(int sig, struct tty_struct *tty); extern void tty_hangup(struct tty_struct * tty); -- cgit v1.2.3 From f34d7a5b7010b82fe97da95496b9971435530062 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 30 Apr 2008 00:54:13 -0700 Subject: tty: The big operations rework - Operations are now a shared const function block as with most other Linux objects - Introduce wrappers for some optional functions to get consistent behaviour - Wrap put_char which used to be patched by the tty layer - Document which functions are needed/optional - Make put_char report success/fail - Cache the driver->ops pointer in the tty as tty->ops - Remove various surplus lock calls we no longer need - Remove proc_write method as noted by Alexey Dobriyan - Introduce some missing sanity checks where certain driver/ldisc combinations would oops as they didn't check needed methods were present [akpm@linux-foundation.org: fix fs/compat_ioctl.c build] [akpm@linux-foundation.org: fix isicom] [akpm@linux-foundation.org: fix arch/ia64/hp/sim/simserial.c build] [akpm@linux-foundation.org: fix kgdb] Signed-off-by: Alan Cox Acked-by: Greg Kroah-Hartman Cc: Jason Wessel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/hp/sim/simserial.c | 11 +- drivers/bluetooth/hci_ldisc.c | 13 +- drivers/char/ip2/ip2main.c | 12 +- drivers/char/isicom.c | 15 +- drivers/char/keyboard.c | 2 +- drivers/char/n_hdlc.c | 11 +- drivers/char/n_r3964.c | 17 +-- drivers/char/n_tty.c | 103 ++++++------- drivers/char/tty_io.c | 174 +++++++++------------- drivers/char/tty_ioctl.c | 62 +++++--- drivers/input/serio/serport.c | 2 +- drivers/isdn/gigaset/ser-gigaset.c | 15 +- drivers/net/hamradio/6pack.c | 36 ++--- drivers/net/hamradio/mkiss.c | 15 +- drivers/net/irda/irtty-sir.c | 95 ++++-------- drivers/net/ppp_async.c | 9 +- drivers/net/ppp_synctty.c | 9 +- drivers/net/slip.c | 13 +- drivers/net/wan/x25_asy.c | 279 +++++++++++++++++------------------ drivers/serial/kgdboc.c | 6 +- drivers/serial/serial_core.c | 38 +++-- drivers/usb/serial/digi_acceleport.c | 3 +- drivers/usb/serial/usb-serial.c | 129 +++------------- drivers/usb/serial/whiteheat.c | 4 +- fs/compat_ioctl.c | 2 +- fs/proc/proc_tty.c | 6 +- include/linux/tty.h | 8 + include/linux/tty_driver.h | 102 +++++++------ kernel/printk.c | 4 +- net/irda/ircomm/ircomm_tty.c | 6 +- 30 files changed, 537 insertions(+), 664 deletions(-) (limited to 'fs') diff --git a/arch/ia64/hp/sim/simserial.c b/arch/ia64/hp/sim/simserial.c index eb0c32a85fd7..23cafc80d2a4 100644 --- a/arch/ia64/hp/sim/simserial.c +++ b/arch/ia64/hp/sim/simserial.c @@ -210,21 +210,23 @@ static void do_softint(struct work_struct *private_) printk(KERN_ERR "simserial: do_softint called\n"); } -static void rs_put_char(struct tty_struct *tty, unsigned char ch) +static int rs_put_char(struct tty_struct *tty, unsigned char ch) { struct async_struct *info = (struct async_struct *)tty->driver_data; unsigned long flags; - if (!tty || !info->xmit.buf) return; + if (!tty || !info->xmit.buf) + return 0; local_irq_save(flags); if (CIRC_SPACE(info->xmit.head, info->xmit.tail, SERIAL_XMIT_SIZE) == 0) { local_irq_restore(flags); - return; + return 0; } info->xmit.buf[info->xmit.head] = ch; info->xmit.head = (info->xmit.head + 1) & (SERIAL_XMIT_SIZE-1); local_irq_restore(flags); + return 1; } static void transmit_chars(struct async_struct *info, int *intr_done) @@ -621,7 +623,8 @@ static void rs_close(struct tty_struct *tty, struct file * filp) * the line discipline to only process XON/XOFF characters. */ shutdown(info); - if (tty->driver->flush_buffer) tty->driver->flush_buffer(tty); + if (tty->ops->flush_buffer) + tty->ops->flush_buffer(tty); if (tty->ldisc.flush_buffer) tty->ldisc.flush_buffer(tty); info->event = 0; info->tty = NULL; diff --git a/drivers/bluetooth/hci_ldisc.c b/drivers/bluetooth/hci_ldisc.c index 7e31d5f1bc8a..a6c2619ec782 100644 --- a/drivers/bluetooth/hci_ldisc.c +++ b/drivers/bluetooth/hci_ldisc.c @@ -143,7 +143,7 @@ restart: int len; set_bit(TTY_DO_WRITE_WAKEUP, &tty->flags); - len = tty->driver->write(tty, skb->data, skb->len); + len = tty->ops->write(tty, skb->data, skb->len); hdev->stat.byte_tx += len; skb_pull(skb, len); @@ -190,8 +190,7 @@ static int hci_uart_flush(struct hci_dev *hdev) /* Flush any pending characters in the driver and discipline. */ tty_ldisc_flush(tty); - if (tty->driver && tty->driver->flush_buffer) - tty->driver->flush_buffer(tty); + tty_driver_flush_buffer(tty); if (test_bit(HCI_UART_PROTO_SET, &hu->flags)) hu->proto->flush(hu); @@ -285,9 +284,7 @@ static int hci_uart_tty_open(struct tty_struct *tty) if (tty->ldisc.flush_buffer) tty->ldisc.flush_buffer(tty); - - if (tty->driver && tty->driver->flush_buffer) - tty->driver->flush_buffer(tty); + tty_driver_flush_buffer(tty); return 0; } @@ -374,8 +371,8 @@ static void hci_uart_tty_receive(struct tty_struct *tty, const u8 *data, char *f spin_unlock(&hu->rx_lock); if (test_and_clear_bit(TTY_THROTTLED, &tty->flags) && - tty->driver->unthrottle) - tty->driver->unthrottle(tty); + tty->ops->unthrottle) + tty->ops->unthrottle(tty); } static int hci_uart_register_dev(struct hci_uart *hu) diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c index c4f4ca31f7c0..5ef69dcd2588 100644 --- a/drivers/char/ip2/ip2main.c +++ b/drivers/char/ip2/ip2main.c @@ -169,7 +169,7 @@ static int Fip_firmware_size; static int ip2_open(PTTY, struct file *); static void ip2_close(PTTY, struct file *); static int ip2_write(PTTY, const unsigned char *, int); -static void ip2_putchar(PTTY, unsigned char); +static int ip2_putchar(PTTY, unsigned char); static void ip2_flush_chars(PTTY); static int ip2_write_room(PTTY); static int ip2_chars_in_buf(PTTY); @@ -1616,10 +1616,9 @@ ip2_close( PTTY tty, struct file *pFile ) serviceOutgoingFifo ( pCh->pMyBord ); - if ( tty->driver->flush_buffer ) - tty->driver->flush_buffer(tty); - if ( tty->ldisc.flush_buffer ) - tty->ldisc.flush_buffer(tty); + if ( tty->driver->ops->flush_buffer ) + tty->driver->ops->flush_buffer(tty); + tty_ldisc_flush(tty); tty->closing = 0; pCh->pTTY = NULL; @@ -1738,7 +1737,7 @@ ip2_write( PTTY tty, const unsigned char *pData, int count) /* */ /* */ /******************************************************************************/ -static void +static int ip2_putchar( PTTY tty, unsigned char ch ) { i2ChanStrPtr pCh = tty->driver_data; @@ -1753,6 +1752,7 @@ ip2_putchar( PTTY tty, unsigned char ch ) ip2_flush_chars( tty ); } else write_unlock_irqrestore(&pCh->Pbuf_spinlock, flags); + return 1; // ip2trace (CHANN, ITRC_PUTC, ITRC_RETURN, 1, ch ); } diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c index 57b115272aaa..9c6be8da220c 100644 --- a/drivers/char/isicom.c +++ b/drivers/char/isicom.c @@ -1140,28 +1140,29 @@ static int isicom_write(struct tty_struct *tty, const unsigned char *buf, } /* put_char et all */ -static void isicom_put_char(struct tty_struct *tty, unsigned char ch) +static int isicom_put_char(struct tty_struct *tty, unsigned char ch) { struct isi_port *port = tty->driver_data; struct isi_board *card = port->card; unsigned long flags; if (isicom_paranoia_check(port, tty->name, "isicom_put_char")) - return; + return 0; if (!port->xmit_buf) - return; + return 0; spin_lock_irqsave(&card->card_lock, flags); - if (port->xmit_cnt >= SERIAL_XMIT_SIZE - 1) - goto out; + if (port->xmit_cnt >= SERIAL_XMIT_SIZE - 1) { + spin_unlock_irqrestore(&card->card_lock, flags); + return 0; + } port->xmit_buf[port->xmit_head++] = ch; port->xmit_head &= (SERIAL_XMIT_SIZE - 1); port->xmit_cnt++; spin_unlock_irqrestore(&card->card_lock, flags); -out: - return; + return 1; } /* flush_chars et all */ diff --git a/drivers/char/keyboard.c b/drivers/char/keyboard.c index 60b934adea65..d1c50b3302e5 100644 --- a/drivers/char/keyboard.c +++ b/drivers/char/keyboard.c @@ -1230,7 +1230,7 @@ static void kbd_keycode(unsigned int keycode, int down, int hw_raw) if (rep && (!vc_kbd_mode(kbd, VC_REPEAT) || - (tty && !L_ECHO(tty) && tty->driver->chars_in_buffer(tty)))) { + (tty && !L_ECHO(tty) && tty_chars_in_buffer(tty)))) { /* * Don't repeat a key if the input buffers are not empty and the * characters get aren't echoed locally. This makes key repeat diff --git a/drivers/char/n_hdlc.c b/drivers/char/n_hdlc.c index a07c0af4819e..a35bfd7ee80e 100644 --- a/drivers/char/n_hdlc.c +++ b/drivers/char/n_hdlc.c @@ -342,12 +342,10 @@ static int n_hdlc_tty_open (struct tty_struct *tty) #endif /* Flush any pending characters in the driver and discipline. */ - if (tty->ldisc.flush_buffer) - tty->ldisc.flush_buffer (tty); + tty->ldisc.flush_buffer(tty); - if (tty->driver->flush_buffer) - tty->driver->flush_buffer (tty); + tty_driver_flush_buffer(tty); if (debuglevel >= DEBUG_LEVEL_INFO) printk("%s(%d)n_hdlc_tty_open() success\n",__FILE__,__LINE__); @@ -399,7 +397,7 @@ static void n_hdlc_send_frames(struct n_hdlc *n_hdlc, struct tty_struct *tty) /* Send the next block of data to device */ tty->flags |= (1 << TTY_DO_WRITE_WAKEUP); - actual = tty->driver->write(tty, tbuf->buf, tbuf->count); + actual = tty->ops->write(tty, tbuf->buf, tbuf->count); /* rollback was possible and has been done */ if (actual == -ERESTARTSYS) { @@ -752,8 +750,7 @@ static int n_hdlc_tty_ioctl(struct tty_struct *tty, struct file *file, case TIOCOUTQ: /* get the pending tx byte count in the driver */ - count = tty->driver->chars_in_buffer ? - tty->driver->chars_in_buffer(tty) : 0; + count = tty_chars_in_buffer(tty); /* add size of next output frame in queue */ spin_lock_irqsave(&n_hdlc->tx_buf_list.spinlock,flags); if (n_hdlc->tx_buf_list.head) diff --git a/drivers/char/n_r3964.c b/drivers/char/n_r3964.c index 3f6486e9f1ec..902169062332 100644 --- a/drivers/char/n_r3964.c +++ b/drivers/char/n_r3964.c @@ -376,8 +376,9 @@ static void put_char(struct r3964_info *pInfo, unsigned char ch) if (tty == NULL) return; - if (tty->driver->put_char) { - tty->driver->put_char(tty, ch); + /* FIXME: put_char should not be called from an IRQ */ + if (tty->ops->put_char) { + tty->ops->put_char(tty, ch); } pInfo->bcc ^= ch; } @@ -386,12 +387,9 @@ static void flush(struct r3964_info *pInfo) { struct tty_struct *tty = pInfo->tty; - if (tty == NULL) + if (tty == NULL || tty->ops->flush_chars == NULL) return; - - if (tty->driver->flush_chars) { - tty->driver->flush_chars(tty); - } + tty->ops->flush_chars(tty); } static void trigger_transmit(struct r3964_info *pInfo) @@ -449,12 +447,11 @@ static void transmit_block(struct r3964_info *pInfo) struct r3964_block_header *pBlock = pInfo->tx_first; int room = 0; - if ((tty == NULL) || (pBlock == NULL)) { + if (tty == NULL || pBlock == NULL) { return; } - if (tty->driver->write_room) - room = tty->driver->write_room(tty); + room = tty_write_room(tty); TRACE_PS("transmit_block %p, room %d, length %d", pBlock, room, pBlock->length); diff --git a/drivers/char/n_tty.c b/drivers/char/n_tty.c index e1518e17e09d..abc93a93dcdd 100644 --- a/drivers/char/n_tty.c +++ b/drivers/char/n_tty.c @@ -149,8 +149,8 @@ static void check_unthrottle(struct tty_struct *tty) { if (tty->count && test_and_clear_bit(TTY_THROTTLED, &tty->flags) && - tty->driver->unthrottle) - tty->driver->unthrottle(tty); + tty->ops->unthrottle) + tty->ops->unthrottle(tty); } /** @@ -273,7 +273,7 @@ static int opost(unsigned char c, struct tty_struct *tty) { int space, spaces; - space = tty->driver->write_room(tty); + space = tty_write_room(tty); if (!space) return -1; @@ -286,7 +286,7 @@ static int opost(unsigned char c, struct tty_struct *tty) if (O_ONLCR(tty)) { if (space < 2) return -1; - tty->driver->put_char(tty, '\r'); + tty_put_char(tty, '\r'); tty->column = 0; } tty->canon_column = tty->column; @@ -308,7 +308,7 @@ static int opost(unsigned char c, struct tty_struct *tty) if (space < spaces) return -1; tty->column += spaces; - tty->driver->write(tty, " ", spaces); + tty->ops->write(tty, " ", spaces); return 0; } tty->column += spaces; @@ -325,7 +325,7 @@ static int opost(unsigned char c, struct tty_struct *tty) break; } } - tty->driver->put_char(tty, c); + tty_put_char(tty, c); unlock_kernel(); return 0; } @@ -352,7 +352,7 @@ static ssize_t opost_block(struct tty_struct *tty, int i; const unsigned char *cp; - space = tty->driver->write_room(tty); + space = tty_write_room(tty); if (!space) return 0; if (nr > space) @@ -390,27 +390,14 @@ static ssize_t opost_block(struct tty_struct *tty, } } break_out: - if (tty->driver->flush_chars) - tty->driver->flush_chars(tty); - i = tty->driver->write(tty, buf, i); + if (tty->ops->flush_chars) + tty->ops->flush_chars(tty); + i = tty->ops->write(tty, buf, i); unlock_kernel(); return i; } -/** - * put_char - write character to driver - * @c: character (or part of unicode symbol) - * @tty: terminal device - * - * Queue a byte to the driver layer for output - */ - -static inline void put_char(unsigned char c, struct tty_struct *tty) -{ - tty->driver->put_char(tty, c); -} - /** * echo_char - echo characters * @c: unicode byte to echo @@ -423,8 +410,8 @@ static inline void put_char(unsigned char c, struct tty_struct *tty) static void echo_char(unsigned char c, struct tty_struct *tty) { if (L_ECHOCTL(tty) && iscntrl(c) && c != '\t') { - put_char('^', tty); - put_char(c ^ 0100, tty); + tty_put_char(tty, '^'); + tty_put_char(tty, c ^ 0100); tty->column += 2; } else opost(c, tty); @@ -433,7 +420,7 @@ static void echo_char(unsigned char c, struct tty_struct *tty) static inline void finish_erasing(struct tty_struct *tty) { if (tty->erasing) { - put_char('/', tty); + tty_put_char(tty, '/'); tty->column++; tty->erasing = 0; } @@ -517,7 +504,7 @@ static void eraser(unsigned char c, struct tty_struct *tty) if (L_ECHO(tty)) { if (L_ECHOPRT(tty)) { if (!tty->erasing) { - put_char('\\', tty); + tty_put_char(tty, '\\'); tty->column++; tty->erasing = 1; } @@ -525,7 +512,7 @@ static void eraser(unsigned char c, struct tty_struct *tty) echo_char(c, tty); while (--cnt > 0) { head = (head+1) & (N_TTY_BUF_SIZE-1); - put_char(tty->read_buf[head], tty); + tty_put_char(tty, tty->read_buf[head]); } } else if (kill_type == ERASE && !L_ECHOE(tty)) { echo_char(ERASE_CHAR(tty), tty); @@ -553,22 +540,22 @@ static void eraser(unsigned char c, struct tty_struct *tty) /* Now backup to that column. */ while (tty->column > col) { /* Can't use opost here. */ - put_char('\b', tty); + tty_put_char(tty, '\b'); if (tty->column > 0) tty->column--; } } else { if (iscntrl(c) && L_ECHOCTL(tty)) { - put_char('\b', tty); - put_char(' ', tty); - put_char('\b', tty); + tty_put_char(tty, '\b'); + tty_put_char(tty, ' '); + tty_put_char(tty, '\b'); if (tty->column > 0) tty->column--; } if (!iscntrl(c) || L_ECHOCTL(tty)) { - put_char('\b', tty); - put_char(' ', tty); - put_char('\b', tty); + tty_put_char(tty, '\b'); + tty_put_char(tty, ' '); + tty_put_char(tty, '\b'); if (tty->column > 0) tty->column--; } @@ -599,8 +586,7 @@ static inline void isig(int sig, struct tty_struct *tty, int flush) kill_pgrp(tty->pgrp, sig, 1); if (flush || !L_NOFLSH(tty)) { n_tty_flush_buffer(tty); - if (tty->driver->flush_buffer) - tty->driver->flush_buffer(tty); + tty_driver_flush_buffer(tty); } } @@ -732,7 +718,7 @@ static inline void n_tty_receive_char(struct tty_struct *tty, unsigned char c) tty->lnext = 0; if (L_ECHO(tty)) { if (tty->read_cnt >= N_TTY_BUF_SIZE-1) { - put_char('\a', tty); /* beep if no space */ + tty_put_char(tty, '\a'); /* beep if no space */ return; } /* Record the column of first canon char. */ @@ -776,8 +762,7 @@ send_signal: */ if (!L_NOFLSH(tty)) { n_tty_flush_buffer(tty); - if (tty->driver->flush_buffer) - tty->driver->flush_buffer(tty); + tty_driver_flush_buffer(tty); } if (L_ECHO(tty)) echo_char(c, tty); @@ -806,8 +791,8 @@ send_signal: if (L_ECHO(tty)) { finish_erasing(tty); if (L_ECHOCTL(tty)) { - put_char('^', tty); - put_char('\b', tty); + tty_put_char(tty, '^'); + tty_put_char(tty, '\b'); } } return; @@ -828,7 +813,7 @@ send_signal: if (c == '\n') { if (L_ECHO(tty) || L_ECHONL(tty)) { if (tty->read_cnt >= N_TTY_BUF_SIZE-1) - put_char('\a', tty); + tty_put_char(tty, '\a'); opost('\n', tty); } goto handle_newline; @@ -846,7 +831,7 @@ send_signal: */ if (L_ECHO(tty)) { if (tty->read_cnt >= N_TTY_BUF_SIZE-1) - put_char('\a', tty); + tty_put_char(tty, '\a'); /* Record the column of first canon char. */ if (tty->canon_head == tty->read_head) tty->canon_column = tty->column; @@ -876,7 +861,7 @@ handle_newline: finish_erasing(tty); if (L_ECHO(tty)) { if (tty->read_cnt >= N_TTY_BUF_SIZE-1) { - put_char('\a', tty); /* beep if no space */ + tty_put_char(tty, '\a'); /* beep if no space */ return; } if (c == '\n') @@ -980,8 +965,8 @@ static void n_tty_receive_buf(struct tty_struct *tty, const unsigned char *cp, break; } } - if (tty->driver->flush_chars) - tty->driver->flush_chars(tty); + if (tty->ops->flush_chars) + tty->ops->flush_chars(tty); } n_tty_set_room(tty); @@ -1000,8 +985,8 @@ static void n_tty_receive_buf(struct tty_struct *tty, const unsigned char *cp, if (tty->receive_room < TTY_THRESHOLD_THROTTLE) { /* check TTY_THROTTLED first so it indicates our state */ if (!test_and_set_bit(TTY_THROTTLED, &tty->flags) && - tty->driver->throttle) - tty->driver->throttle(tty); + tty->ops->throttle) + tty->ops->throttle(tty); } } @@ -1086,6 +1071,9 @@ static void n_tty_set_termios(struct tty_struct *tty, struct ktermios *old) tty->real_raw = 0; } n_tty_set_room(tty); + /* The termios change make the tty ready for I/O */ + wake_up_interruptible(&tty->write_wait); + wake_up_interruptible(&tty->read_wait); } /** @@ -1513,11 +1501,11 @@ static ssize_t write_chan(struct tty_struct *tty, struct file *file, break; b++; nr--; } - if (tty->driver->flush_chars) - tty->driver->flush_chars(tty); + if (tty->ops->flush_chars) + tty->ops->flush_chars(tty); } else { while (nr > 0) { - c = tty->driver->write(tty, b, nr); + c = tty->ops->write(tty, b, nr); if (c < 0) { retval = c; goto break_out; @@ -1554,11 +1542,6 @@ break_out: * * This code must be sure never to sleep through a hangup. * Called without the kernel lock held - fine - * - * FIXME: if someone changes the VMIN or discipline settings for the - * terminal while another process is in poll() the poll does not - * recompute the new limits. Possibly set_termios should issue - * a read wakeup to fix this bug. */ static unsigned int normal_poll(struct tty_struct *tty, struct file *file, @@ -1582,9 +1565,9 @@ static unsigned int normal_poll(struct tty_struct *tty, struct file *file, else tty->minimum_to_wake = 1; } - if (!tty_is_writelocked(tty) && - tty->driver->chars_in_buffer(tty) < WAKEUP_CHARS && - tty->driver->write_room(tty) > 0) + if (tty->ops->write && !tty_is_writelocked(tty) && + tty_chars_in_buffer(tty) < WAKEUP_CHARS && + tty_write_room(tty) > 0) mask |= POLLOUT | POLLWRNORM; return mask; } diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index b1692afd797e..f69fb8d7a680 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -1108,8 +1108,8 @@ restart: a reference to the old ldisc. If we ended up flipping back to the existing ldisc we have two references to it */ - if (tty->ldisc.num != o_ldisc.num && tty->driver->set_ldisc) - tty->driver->set_ldisc(tty); + if (tty->ldisc.num != o_ldisc.num && tty->ops->set_ldisc) + tty->ops->set_ldisc(tty); tty_ldisc_put(o_ldisc.num); @@ -1181,9 +1181,8 @@ struct tty_driver *tty_find_polling_driver(char *name, int *line) if (*str == '\0') str = NULL; - if (tty_line >= 0 && tty_line <= p->num && p->poll_init && - !p->poll_init(p, tty_line, str)) { - + if (tty_line >= 0 && tty_line <= p->num && p->ops && + p->ops->poll_init && !p->ops->poll_init(p, tty_line, str)) { res = p; *line = tty_line; break; @@ -1452,8 +1451,7 @@ static void do_tty_hangup(struct work_struct *work) /* We may have no line discipline at this point */ if (ld->flush_buffer) ld->flush_buffer(tty); - if (tty->driver->flush_buffer) - tty->driver->flush_buffer(tty); + tty_driver_flush_buffer(tty); if ((test_bit(TTY_DO_WRITE_WAKEUP, &tty->flags)) && ld->write_wakeup) ld->write_wakeup(tty); @@ -1516,11 +1514,11 @@ static void do_tty_hangup(struct work_struct *work) * So we just call close() the right number of times. */ if (cons_filp) { - if (tty->driver->close) + if (tty->ops->close) for (n = 0; n < closecount; n++) - tty->driver->close(tty, cons_filp); - } else if (tty->driver->hangup) - (tty->driver->hangup)(tty); + tty->ops->close(tty, cons_filp); + } else if (tty->ops->hangup) + (tty->ops->hangup)(tty); /* * We don't want to have driver/ldisc interactions beyond * the ones we did here. The driver layer expects no @@ -1752,8 +1750,8 @@ void stop_tty(struct tty_struct *tty) wake_up_interruptible(&tty->link->read_wait); } spin_unlock_irqrestore(&tty->ctrl_lock, flags); - if (tty->driver->stop) - (tty->driver->stop)(tty); + if (tty->ops->stop) + (tty->ops->stop)(tty); } EXPORT_SYMBOL(stop_tty); @@ -1786,8 +1784,8 @@ void start_tty(struct tty_struct *tty) wake_up_interruptible(&tty->link->read_wait); } spin_unlock_irqrestore(&tty->ctrl_lock, flags); - if (tty->driver->start) - (tty->driver->start)(tty); + if (tty->ops->start) + (tty->ops->start)(tty); /* If we have a running line discipline it may need kicking */ tty_wakeup(tty); } @@ -1972,10 +1970,13 @@ static ssize_t tty_write(struct file *file, const char __user *buf, tty = (struct tty_struct *)file->private_data; if (tty_paranoia_check(tty, inode, "tty_write")) return -EIO; - if (!tty || !tty->driver->write || + if (!tty || !tty->ops->write || (test_bit(TTY_IO_ERROR, &tty->flags))) return -EIO; - + /* Short term debug to catch buggy drivers */ + if (tty->ops->write_room == NULL) + printk(KERN_ERR "tty driver %s lacks a write_room method.\n", + tty->driver->name); ld = tty_ldisc_ref_wait(tty); if (!ld->write) ret = -EIO; @@ -2122,6 +2123,7 @@ static int init_dev(struct tty_driver *driver, int idx, goto fail_no_mem; initialize_tty_struct(tty); tty->driver = driver; + tty->ops = driver->ops; tty->index = idx; tty_line_name(driver, idx, tty->name); @@ -2152,6 +2154,7 @@ static int init_dev(struct tty_driver *driver, int idx, goto free_mem_out; initialize_tty_struct(o_tty); o_tty->driver = driver->other; + o_tty->ops = driver->ops; o_tty->index = idx; tty_line_name(driver->other, idx, o_tty->name); @@ -2456,8 +2459,8 @@ static void release_dev(struct file *filp) } } #endif - if (tty->driver->close) - tty->driver->close(tty, filp); + if (tty->ops->close) + tty->ops->close(tty, filp); /* * Sanity check: if tty->count is going to zero, there shouldn't be @@ -2740,8 +2743,8 @@ got_driver: printk(KERN_DEBUG "opening %s...", tty->name); #endif if (!retval) { - if (tty->driver->open) - retval = tty->driver->open(tty, filp); + if (tty->ops->open) + retval = tty->ops->open(tty, filp); else retval = -ENODEV; } @@ -2840,7 +2843,7 @@ static int ptmx_open(struct inode *inode, struct file *filp) goto out1; check_tty_count(tty, "tty_open"); - retval = ptm_driver->open(tty, filp); + retval = ptm_driver->ops->open(tty, filp); if (!retval) return 0; out1: @@ -3336,25 +3339,20 @@ static int tiocsetd(struct tty_struct *tty, int __user *p) static int send_break(struct tty_struct *tty, unsigned int duration) { - int retval = -EINTR; - - lock_kernel(); if (tty_write_lock(tty, 0) < 0) - goto out; - tty->driver->break_ctl(tty, -1); + return -EINTR; + tty->ops->break_ctl(tty, -1); if (!signal_pending(current)) msleep_interruptible(duration); - tty->driver->break_ctl(tty, 0); + tty->ops->break_ctl(tty, 0); tty_write_unlock(tty); if (!signal_pending(current)) - retval = 0; -out: - unlock_kernel(); - return retval; + return -EINTR; + return 0; } /** - * tiocmget - get modem status + * tty_tiocmget - get modem status * @tty: tty device * @file: user file pointer * @p: pointer to result @@ -3369,10 +3367,8 @@ static int tty_tiocmget(struct tty_struct *tty, struct file *file, int __user *p { int retval = -EINVAL; - if (tty->driver->tiocmget) { - lock_kernel(); - retval = tty->driver->tiocmget(tty, file); - unlock_kernel(); + if (tty->ops->tiocmget) { + retval = tty->ops->tiocmget(tty, file); if (retval >= 0) retval = put_user(retval, p); @@ -3381,7 +3377,7 @@ static int tty_tiocmget(struct tty_struct *tty, struct file *file, int __user *p } /** - * tiocmset - set modem status + * tty_tiocmset - set modem status * @tty: tty device * @file: user file pointer * @cmd: command - clear bits, set bits or set all @@ -3398,7 +3394,7 @@ static int tty_tiocmset(struct tty_struct *tty, struct file *file, unsigned int { int retval = -EINVAL; - if (tty->driver->tiocmset) { + if (tty->ops->tiocmset) { unsigned int set, clear, val; retval = get_user(val, p); @@ -3422,9 +3418,7 @@ static int tty_tiocmset(struct tty_struct *tty, struct file *file, unsigned int set &= TIOCM_DTR|TIOCM_RTS|TIOCM_OUT1|TIOCM_OUT2|TIOCM_LOOP; clear &= TIOCM_DTR|TIOCM_RTS|TIOCM_OUT1|TIOCM_OUT2|TIOCM_LOOP; - lock_kernel(); - retval = tty->driver->tiocmset(tty, file, set, clear); - unlock_kernel(); + retval = tty->ops->tiocmset(tty, file, set, clear); } return retval; } @@ -3455,23 +3449,25 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg) retval = -EINVAL; - if (!tty->driver->break_ctl) { + if (!tty->ops->break_ctl) { switch (cmd) { case TIOCSBRK: case TIOCCBRK: - if (tty->driver->ioctl) - retval = tty->driver->ioctl(tty, file, cmd, arg); + if (tty->ops->ioctl) + retval = tty->ops->ioctl(tty, file, cmd, arg); + if (retval != -EINVAL && retval != -ENOIOCTLCMD) + printk(KERN_WARNING "tty: driver %s needs updating to use break_ctl\n", tty->driver->name); return retval; /* These two ioctl's always return success; even if */ /* the driver doesn't support them. */ case TCSBRK: case TCSBRKP: - if (!tty->driver->ioctl) + if (!tty->ops->ioctl) return 0; - lock_kernel(); - retval = tty->driver->ioctl(tty, file, cmd, arg); - unlock_kernel(); + retval = tty->ops->ioctl(tty, file, cmd, arg); + if (retval != -EINVAL && retval != -ENOIOCTLCMD) + printk(KERN_WARNING "tty: driver %s needs updating to use break_ctl\n", tty->driver->name); if (retval == -ENOIOCTLCMD) retval = 0; return retval; @@ -3491,9 +3487,7 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg) if (retval) return retval; if (cmd != TIOCCBRK) { - lock_kernel(); tty_wait_until_sent(tty, 0); - unlock_kernel(); if (signal_pending(current)) return -EINTR; } @@ -3531,7 +3525,6 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg) case TIOCGSID: return tiocgsid(tty, real_tty, p); case TIOCGETD: - /* FIXME: check this is ok */ return put_user(tty->ldisc.num, (int __user *)p); case TIOCSETD: return tiocsetd(tty, p); @@ -3543,15 +3536,13 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg) * Break handling */ case TIOCSBRK: /* Turn break on, unconditionally */ - lock_kernel(); - tty->driver->break_ctl(tty, -1); - unlock_kernel(); + if (tty->ops->break_ctl) + tty->ops->break_ctl(tty, -1); return 0; case TIOCCBRK: /* Turn break off, unconditionally */ - lock_kernel(); - tty->driver->break_ctl(tty, 0); - unlock_kernel(); + if (tty->ops->break_ctl) + tty->ops->break_ctl(tty, 0); return 0; case TCSBRK: /* SVID version: non-zero arg --> no break */ /* non-zero arg means wait for all output data @@ -3580,8 +3571,8 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg) } break; } - if (tty->driver->ioctl) { - retval = (tty->driver->ioctl)(tty, file, cmd, arg); + if (tty->ops->ioctl) { + retval = (tty->ops->ioctl)(tty, file, cmd, arg); if (retval != -ENOIOCTLCMD) return retval; } @@ -3608,8 +3599,8 @@ static long tty_compat_ioctl(struct file *file, unsigned int cmd, if (tty_paranoia_check(tty, inode, "tty_ioctl")) return -EINVAL; - if (tty->driver->compat_ioctl) { - retval = (tty->driver->compat_ioctl)(tty, file, cmd, arg); + if (tty->ops->compat_ioctl) { + retval = (tty->ops->compat_ioctl)(tty, file, cmd, arg); if (retval != -ENOIOCTLCMD) return retval; } @@ -3659,8 +3650,7 @@ void __do_SAK(struct tty_struct *tty) tty_ldisc_flush(tty); - if (tty->driver->flush_buffer) - tty->driver->flush_buffer(tty); + tty_driver_flush_buffer(tty); read_lock(&tasklist_lock); /* Kill the entire session */ @@ -3871,15 +3861,27 @@ static void initialize_tty_struct(struct tty_struct *tty) INIT_WORK(&tty->SAK_work, do_SAK_work); } -/* - * The default put_char routine if the driver did not define one. +/** + * tty_put_char - write one character to a tty + * @tty: tty + * @ch: character + * + * Write one byte to the tty using the provided put_char method + * if present. Returns the number of characters successfully output. + * + * Note: the specific put_char operation in the driver layer may go + * away soon. Don't call it directly, use this method */ -static void tty_default_put_char(struct tty_struct *tty, unsigned char ch) +int tty_put_char(struct tty_struct *tty, unsigned char ch) { - tty->driver->write(tty, &ch, 1); + if (tty->ops->put_char) + return tty->ops->put_char(tty, ch); + return tty->ops->write(tty, &ch, 1); } +EXPORT_SYMBOL_GPL(tty_put_char); + static struct class *tty_class; /** @@ -3962,37 +3964,8 @@ void put_tty_driver(struct tty_driver *driver) void tty_set_operations(struct tty_driver *driver, const struct tty_operations *op) { - driver->open = op->open; - driver->close = op->close; - driver->write = op->write; - driver->put_char = op->put_char; - driver->flush_chars = op->flush_chars; - driver->write_room = op->write_room; - driver->chars_in_buffer = op->chars_in_buffer; - driver->ioctl = op->ioctl; - driver->compat_ioctl = op->compat_ioctl; - driver->set_termios = op->set_termios; - driver->throttle = op->throttle; - driver->unthrottle = op->unthrottle; - driver->stop = op->stop; - driver->start = op->start; - driver->hangup = op->hangup; - driver->break_ctl = op->break_ctl; - driver->flush_buffer = op->flush_buffer; - driver->set_ldisc = op->set_ldisc; - driver->wait_until_sent = op->wait_until_sent; - driver->send_xchar = op->send_xchar; - driver->read_proc = op->read_proc; - driver->write_proc = op->write_proc; - driver->tiocmget = op->tiocmget; - driver->tiocmset = op->tiocmset; -#ifdef CONFIG_CONSOLE_POLL - driver->poll_init = op->poll_init; - driver->poll_get_char = op->poll_get_char; - driver->poll_put_char = op->poll_put_char; -#endif -} - + driver->ops = op; +}; EXPORT_SYMBOL(alloc_tty_driver); EXPORT_SYMBOL(put_tty_driver); @@ -4055,9 +4028,6 @@ int tty_register_driver(struct tty_driver *driver) return error; } - if (!driver->put_char) - driver->put_char = tty_default_put_char; - mutex_lock(&tty_mutex); list_add(&driver->tty_drivers, &tty_drivers); mutex_unlock(&tty_mutex); diff --git a/drivers/char/tty_ioctl.c b/drivers/char/tty_ioctl.c index 8c4bf3e48d5b..c10d40c4c5ca 100644 --- a/drivers/char/tty_ioctl.c +++ b/drivers/char/tty_ioctl.c @@ -40,6 +40,34 @@ #define TERMIOS_OLD 8 +int tty_chars_in_buffer(struct tty_struct *tty) +{ + if (tty->ops->chars_in_buffer) + return tty->ops->chars_in_buffer(tty); + else + return 0; +} + +EXPORT_SYMBOL(tty_chars_in_buffer); + +int tty_write_room(struct tty_struct *tty) +{ + if (tty->ops->write_room) + return tty->ops->write_room(tty); + return 2048; +} + +EXPORT_SYMBOL(tty_write_room); + +void tty_driver_flush_buffer(struct tty_struct *tty) +{ + if (tty->ops->flush_buffer) + tty->ops->flush_buffer(tty); +} + +EXPORT_SYMBOL(tty_driver_flush_buffer); + + /** * tty_wait_until_sent - wait for I/O to finish * @tty: tty we are waiting for @@ -58,17 +86,13 @@ void tty_wait_until_sent(struct tty_struct *tty, long timeout) printk(KERN_DEBUG "%s wait until sent...\n", tty_name(tty, buf)); #endif - if (!tty->driver->chars_in_buffer) - return; if (!timeout) timeout = MAX_SCHEDULE_TIMEOUT; - lock_kernel(); if (wait_event_interruptible_timeout(tty->write_wait, - !tty->driver->chars_in_buffer(tty), timeout) >= 0) { - if (tty->driver->wait_until_sent) - tty->driver->wait_until_sent(tty, timeout); + !tty_chars_in_buffer(tty), timeout) >= 0) { + if (tty->ops->wait_until_sent) + tty->ops->wait_until_sent(tty, timeout); } - unlock_kernel(); } EXPORT_SYMBOL(tty_wait_until_sent); @@ -444,8 +468,8 @@ static void change_termios(struct tty_struct *tty, struct ktermios *new_termios) } } - if (tty->driver->set_termios) - (*tty->driver->set_termios)(tty, &old_termios); + if (tty->ops->set_termios) + (*tty->ops->set_termios)(tty, &old_termios); else tty_termios_copy_hw(tty->termios, &old_termios); @@ -748,8 +772,8 @@ static int send_prio_char(struct tty_struct *tty, char ch) { int was_stopped = tty->stopped; - if (tty->driver->send_xchar) { - tty->driver->send_xchar(tty, ch); + if (tty->ops->send_xchar) { + tty->ops->send_xchar(tty, ch); return 0; } @@ -758,7 +782,7 @@ static int send_prio_char(struct tty_struct *tty, char ch) if (was_stopped) start_tty(tty); - tty->driver->write(tty, &ch, 1); + tty->ops->write(tty, &ch, 1); if (was_stopped) stop_tty(tty); tty_write_unlock(tty); @@ -778,13 +802,14 @@ static int tty_change_softcar(struct tty_struct *tty, int arg) { int ret = 0; int bit = arg ? CLOCAL : 0; - struct ktermios old = *tty->termios; + struct ktermios old; mutex_lock(&tty->termios_mutex); + old = *tty->termios; tty->termios->c_cflag &= ~CLOCAL; tty->termios->c_cflag |= bit; - if (tty->driver->set_termios) - tty->driver->set_termios(tty, &old); + if (tty->ops->set_termios) + tty->ops->set_termios(tty, &old); if ((tty->termios->c_cflag & CLOCAL) != bit) ret = -EINVAL; mutex_unlock(&tty->termios_mutex); @@ -926,8 +951,7 @@ int tty_perform_flush(struct tty_struct *tty, unsigned long arg) ld->flush_buffer(tty); /* fall through */ case TCOFLUSH: - if (tty->driver->flush_buffer) - tty->driver->flush_buffer(tty); + tty_driver_flush_buffer(tty); break; default: tty_ldisc_deref(ld); @@ -984,9 +1008,7 @@ int n_tty_ioctl(struct tty_struct *tty, struct file *file, case TCFLSH: return tty_perform_flush(tty, arg); case TIOCOUTQ: - return put_user(tty->driver->chars_in_buffer ? - tty->driver->chars_in_buffer(tty) : 0, - (int __user *) arg); + return put_user(tty_chars_in_buffer(tty), (int __user *) arg); case TIOCINQ: retval = tty->read_cnt; if (L_ICANON(tty)) diff --git a/drivers/input/serio/serport.c b/drivers/input/serio/serport.c index e1a3a79ab3f9..7ff71ba7b7c9 100644 --- a/drivers/input/serio/serport.c +++ b/drivers/input/serio/serport.c @@ -46,7 +46,7 @@ struct serport { static int serport_serio_write(struct serio *serio, unsigned char data) { struct serport *serport = serio->port_data; - return -(serport->tty->driver->write(serport->tty, &data, 1) != 1); + return -(serport->tty->ops->write(serport->tty, &data, 1) != 1); } static int serport_serio_open(struct serio *serio) diff --git a/drivers/isdn/gigaset/ser-gigaset.c b/drivers/isdn/gigaset/ser-gigaset.c index fceeb1d57682..45d1ee93cd39 100644 --- a/drivers/isdn/gigaset/ser-gigaset.c +++ b/drivers/isdn/gigaset/ser-gigaset.c @@ -68,10 +68,10 @@ static int write_modem(struct cardstate *cs) struct tty_struct *tty = cs->hw.ser->tty; struct bc_state *bcs = &cs->bcs[0]; /* only one channel */ struct sk_buff *skb = bcs->tx_skb; - int sent; + int sent = -EOPNOTSUPP; if (!tty || !tty->driver || !skb) - return -EFAULT; + return -EINVAL; if (!skb->len) { dev_kfree_skb_any(skb); @@ -80,7 +80,8 @@ static int write_modem(struct cardstate *cs) } set_bit(TTY_DO_WRITE_WAKEUP, &tty->flags); - sent = tty->driver->write(tty, skb->data, skb->len); + if (tty->ops->write) + sent = tty->ops->write(tty, skb->data, skb->len); gig_dbg(DEBUG_OUTPUT, "write_modem: sent %d", sent); if (sent < 0) { /* error */ @@ -120,7 +121,7 @@ static int send_cb(struct cardstate *cs) if (cb->len) { set_bit(TTY_DO_WRITE_WAKEUP, &tty->flags); - sent = tty->driver->write(tty, cb->buf + cb->offset, cb->len); + sent = tty->ops->write(tty, cb->buf + cb->offset, cb->len); if (sent < 0) { /* error */ gig_dbg(DEBUG_OUTPUT, "send_cb: write error %d", sent); @@ -440,14 +441,14 @@ static int gigaset_set_modem_ctrl(struct cardstate *cs, unsigned old_state, unsi struct tty_struct *tty = cs->hw.ser->tty; unsigned int set, clear; - if (!tty || !tty->driver || !tty->driver->tiocmset) - return -EFAULT; + if (!tty || !tty->driver || !tty->ops->tiocmset) + return -EINVAL; set = new_state & ~old_state; clear = old_state & ~new_state; if (!set && !clear) return 0; gig_dbg(DEBUG_IF, "tiocmset set %x clear %x", set, clear); - return tty->driver->tiocmset(tty, NULL, set, clear); + return tty->ops->tiocmset(tty, NULL, set, clear); } static int gigaset_baud_rate(struct cardstate *cs, unsigned cflag) diff --git a/drivers/net/hamradio/6pack.c b/drivers/net/hamradio/6pack.c index 1da55dd2a5a0..82a36266dfc9 100644 --- a/drivers/net/hamradio/6pack.c +++ b/drivers/net/hamradio/6pack.c @@ -148,13 +148,13 @@ static void sp_xmit_on_air(unsigned long channel) if (((sp->status1 & SIXP_DCD_MASK) == 0) && (random < sp->persistence)) { sp->led_state = 0x70; - sp->tty->driver->write(sp->tty, &sp->led_state, 1); + sp->tty->ops->write(sp->tty, &sp->led_state, 1); sp->tx_enable = 1; - actual = sp->tty->driver->write(sp->tty, sp->xbuff, sp->status2); + actual = sp->tty->ops->write(sp->tty, sp->xbuff, sp->status2); sp->xleft -= actual; sp->xhead += actual; sp->led_state = 0x60; - sp->tty->driver->write(sp->tty, &sp->led_state, 1); + sp->tty->ops->write(sp->tty, &sp->led_state, 1); sp->status2 = 0; } else mod_timer(&sp->tx_t, jiffies + ((when + 1) * HZ) / 100); @@ -220,13 +220,13 @@ static void sp_encaps(struct sixpack *sp, unsigned char *icp, int len) */ if (sp->duplex == 1) { sp->led_state = 0x70; - sp->tty->driver->write(sp->tty, &sp->led_state, 1); + sp->tty->ops->write(sp->tty, &sp->led_state, 1); sp->tx_enable = 1; - actual = sp->tty->driver->write(sp->tty, sp->xbuff, count); + actual = sp->tty->ops->write(sp->tty, sp->xbuff, count); sp->xleft = count - actual; sp->xhead = sp->xbuff + actual; sp->led_state = 0x60; - sp->tty->driver->write(sp->tty, &sp->led_state, 1); + sp->tty->ops->write(sp->tty, &sp->led_state, 1); } else { sp->xleft = count; sp->xhead = sp->xbuff; @@ -444,7 +444,7 @@ static void sixpack_write_wakeup(struct tty_struct *tty) } if (sp->tx_enable) { - actual = tty->driver->write(tty, sp->xhead, sp->xleft); + actual = tty->ops->write(tty, sp->xhead, sp->xleft); sp->xleft -= actual; sp->xhead += actual; } @@ -492,8 +492,8 @@ static void sixpack_receive_buf(struct tty_struct *tty, sp_put(sp); if (test_and_clear_bit(TTY_THROTTLED, &tty->flags) - && tty->driver->unthrottle) - tty->driver->unthrottle(tty); + && tty->ops->unthrottle) + tty->ops->unthrottle(tty); } /* @@ -554,8 +554,8 @@ static void resync_tnc(unsigned long channel) /* resync the TNC */ sp->led_state = 0x60; - sp->tty->driver->write(sp->tty, &sp->led_state, 1); - sp->tty->driver->write(sp->tty, &resync_cmd, 1); + sp->tty->ops->write(sp->tty, &sp->led_state, 1); + sp->tty->ops->write(sp->tty, &resync_cmd, 1); /* Start resync timer again -- the TNC might be still absent */ @@ -573,7 +573,7 @@ static inline int tnc_init(struct sixpack *sp) tnc_set_sync_state(sp, TNC_UNSYNC_STARTUP); - sp->tty->driver->write(sp->tty, &inbyte, 1); + sp->tty->ops->write(sp->tty, &inbyte, 1); del_timer(&sp->resync_t); sp->resync_t.data = (unsigned long) sp; @@ -601,6 +601,8 @@ static int sixpack_open(struct tty_struct *tty) if (!capable(CAP_NET_ADMIN)) return -EPERM; + if (tty->ops->write == NULL) + return -EOPNOTSUPP; dev = alloc_netdev(sizeof(struct sixpack), "sp%d", sp_setup); if (!dev) { @@ -914,9 +916,9 @@ static void decode_prio_command(struct sixpack *sp, unsigned char cmd) } else { /* output watchdog char if idle */ if ((sp->status2 != 0) && (sp->duplex == 1)) { sp->led_state = 0x70; - sp->tty->driver->write(sp->tty, &sp->led_state, 1); + sp->tty->ops->write(sp->tty, &sp->led_state, 1); sp->tx_enable = 1; - actual = sp->tty->driver->write(sp->tty, sp->xbuff, sp->status2); + actual = sp->tty->ops->write(sp->tty, sp->xbuff, sp->status2); sp->xleft -= actual; sp->xhead += actual; sp->led_state = 0x60; @@ -926,7 +928,7 @@ static void decode_prio_command(struct sixpack *sp, unsigned char cmd) } /* needed to trigger the TNC watchdog */ - sp->tty->driver->write(sp->tty, &sp->led_state, 1); + sp->tty->ops->write(sp->tty, &sp->led_state, 1); /* if the state byte has been received, the TNC is present, so the resync timer can be reset. */ @@ -956,12 +958,12 @@ static void decode_std_command(struct sixpack *sp, unsigned char cmd) if ((sp->status & SIXP_RX_DCD_MASK) == SIXP_RX_DCD_MASK) { sp->led_state = 0x68; - sp->tty->driver->write(sp->tty, &sp->led_state, 1); + sp->tty->ops->write(sp->tty, &sp->led_state, 1); } } else { sp->led_state = 0x60; /* fill trailing bytes with zeroes */ - sp->tty->driver->write(sp->tty, &sp->led_state, 1); + sp->tty->ops->write(sp->tty, &sp->led_state, 1); rest = sp->rx_count; if (rest != 0) for (i = rest; i <= 3; i++) diff --git a/drivers/net/hamradio/mkiss.c b/drivers/net/hamradio/mkiss.c index 30c9b3b0d131..ebcc5adee7cc 100644 --- a/drivers/net/hamradio/mkiss.c +++ b/drivers/net/hamradio/mkiss.c @@ -516,7 +516,7 @@ static void ax_encaps(struct net_device *dev, unsigned char *icp, int len) spin_unlock_bh(&ax->buflock); set_bit(TTY_DO_WRITE_WAKEUP, &ax->tty->flags); - actual = ax->tty->driver->write(ax->tty, ax->xbuff, count); + actual = ax->tty->ops->write(ax->tty, ax->xbuff, count); ax->stats.tx_packets++; ax->stats.tx_bytes += actual; @@ -546,7 +546,7 @@ static int ax_xmit(struct sk_buff *skb, struct net_device *dev) } printk(KERN_ERR "mkiss: %s: transmit timed out, %s?\n", dev->name, - (ax->tty->driver->chars_in_buffer(ax->tty) || ax->xleft) ? + (ax->tty->ops->chars_in_buffer(ax->tty) || ax->xleft) ? "bad line quality" : "driver error"); ax->xleft = 0; @@ -736,6 +736,8 @@ static int mkiss_open(struct tty_struct *tty) if (!capable(CAP_NET_ADMIN)) return -EPERM; + if (tty->ops->write == NULL) + return -EOPNOTSUPP; dev = alloc_netdev(sizeof(struct mkiss), "ax%d", ax_setup); if (!dev) { @@ -754,8 +756,7 @@ static int mkiss_open(struct tty_struct *tty) tty->disc_data = ax; tty->receive_room = 65535; - if (tty->driver->flush_buffer) - tty->driver->flush_buffer(tty); + tty_driver_flush_buffer(tty); /* Restore default settings */ dev->type = ARPHRD_AX25; @@ -936,8 +937,8 @@ static void mkiss_receive_buf(struct tty_struct *tty, const unsigned char *cp, mkiss_put(ax); if (test_and_clear_bit(TTY_THROTTLED, &tty->flags) - && tty->driver->unthrottle) - tty->driver->unthrottle(tty); + && tty->ops->unthrottle) + tty->ops->unthrottle(tty); } /* @@ -962,7 +963,7 @@ static void mkiss_write_wakeup(struct tty_struct *tty) goto out; } - actual = tty->driver->write(tty, ax->xhead, ax->xleft); + actual = tty->ops->write(tty, ax->xhead, ax->xleft); ax->xleft -= actual; ax->xhead += actual; diff --git a/drivers/net/irda/irtty-sir.c b/drivers/net/irda/irtty-sir.c index fc753d7f674e..e6f40b7f9041 100644 --- a/drivers/net/irda/irtty-sir.c +++ b/drivers/net/irda/irtty-sir.c @@ -64,7 +64,7 @@ static int irtty_chars_in_buffer(struct sir_dev *dev) IRDA_ASSERT(priv != NULL, return -1;); IRDA_ASSERT(priv->magic == IRTTY_MAGIC, return -1;); - return priv->tty->driver->chars_in_buffer(priv->tty); + return tty_chars_in_buffer(priv->tty); } /* Wait (sleep) until underlaying hardware finished transmission @@ -93,10 +93,8 @@ static void irtty_wait_until_sent(struct sir_dev *dev) IRDA_ASSERT(priv->magic == IRTTY_MAGIC, return;); tty = priv->tty; - if (tty->driver->wait_until_sent) { - lock_kernel(); - tty->driver->wait_until_sent(tty, msecs_to_jiffies(100)); - unlock_kernel(); + if (tty->ops->wait_until_sent) { + tty->ops->wait_until_sent(tty, msecs_to_jiffies(100)); } else { msleep(USBSERIAL_TX_DONE_DELAY); @@ -125,48 +123,14 @@ static int irtty_change_speed(struct sir_dev *dev, unsigned speed) tty = priv->tty; - lock_kernel(); + mutex_lock(&tty->termios_mutex); old_termios = *(tty->termios); cflag = tty->termios->c_cflag; - - cflag &= ~CBAUD; - - IRDA_DEBUG(2, "%s(), Setting speed to %d\n", __FUNCTION__, speed); - - switch (speed) { - case 1200: - cflag |= B1200; - break; - case 2400: - cflag |= B2400; - break; - case 4800: - cflag |= B4800; - break; - case 19200: - cflag |= B19200; - break; - case 38400: - cflag |= B38400; - break; - case 57600: - cflag |= B57600; - break; - case 115200: - cflag |= B115200; - break; - case 9600: - default: - cflag |= B9600; - break; - } - - tty->termios->c_cflag = cflag; - if (tty->driver->set_termios) - tty->driver->set_termios(tty, &old_termios); - unlock_kernel(); - + tty_encode_baud_rate(tty, speed, speed); + if (tty->ops->set_termios) + tty->ops->set_termios(tty, &old_termios); priv->io.speed = speed; + mutex_unlock(&tty->termios_mutex); return 0; } @@ -202,8 +166,8 @@ static int irtty_set_dtr_rts(struct sir_dev *dev, int dtr, int rts) * This function is not yet defined for all tty driver, so * let's be careful... Jean II */ - IRDA_ASSERT(priv->tty->driver->tiocmset != NULL, return -1;); - priv->tty->driver->tiocmset(priv->tty, NULL, set, clear); + IRDA_ASSERT(priv->tty->ops->tiocmset != NULL, return -1;); + priv->tty->ops->tiocmset(priv->tty, NULL, set, clear); return 0; } @@ -225,17 +189,13 @@ static int irtty_do_write(struct sir_dev *dev, const unsigned char *ptr, size_t IRDA_ASSERT(priv->magic == IRTTY_MAGIC, return -1;); tty = priv->tty; - if (!tty->driver->write) + if (!tty->ops->write) return 0; tty->flags |= (1 << TTY_DO_WRITE_WAKEUP); - if (tty->driver->write_room) { - writelen = tty->driver->write_room(tty); - if (writelen > len) - writelen = len; - } - else + writelen = tty_write_room(tty); + if (writelen > len) writelen = len; - return tty->driver->write(tty, ptr, writelen); + return tty->ops->write(tty, ptr, writelen); } /* ------------------------------------------------------- */ @@ -321,7 +281,7 @@ static inline void irtty_stop_receiver(struct tty_struct *tty, int stop) struct ktermios old_termios; int cflag; - lock_kernel(); + mutex_lock(&tty->termios_mutex); old_termios = *(tty->termios); cflag = tty->termios->c_cflag; @@ -331,9 +291,9 @@ static inline void irtty_stop_receiver(struct tty_struct *tty, int stop) cflag |= CREAD; tty->termios->c_cflag = cflag; - if (tty->driver->set_termios) - tty->driver->set_termios(tty, &old_termios); - unlock_kernel(); + if (tty->ops->set_termios) + tty->ops->set_termios(tty, &old_termios); + mutex_unlock(&tty->termios_mutex); } /*****************************************************************/ @@ -359,8 +319,8 @@ static int irtty_start_dev(struct sir_dev *dev) tty = priv->tty; - if (tty->driver->start) - tty->driver->start(tty); + if (tty->ops->start) + tty->ops->start(tty); /* Make sure we can receive more data */ irtty_stop_receiver(tty, FALSE); @@ -388,8 +348,8 @@ static int irtty_stop_dev(struct sir_dev *dev) /* Make sure we don't receive more data */ irtty_stop_receiver(tty, TRUE); - if (tty->driver->stop) - tty->driver->stop(tty); + if (tty->ops->stop) + tty->ops->stop(tty); mutex_unlock(&irtty_mutex); @@ -483,11 +443,10 @@ static int irtty_open(struct tty_struct *tty) /* stop the underlying driver */ irtty_stop_receiver(tty, TRUE); - if (tty->driver->stop) - tty->driver->stop(tty); + if (tty->ops->stop) + tty->ops->stop(tty); - if (tty->driver->flush_buffer) - tty->driver->flush_buffer(tty); + tty_driver_flush_buffer(tty); /* apply mtt override */ sir_tty_drv.qos_mtt_bits = qos_mtt_bits; @@ -564,8 +523,8 @@ static void irtty_close(struct tty_struct *tty) /* Stop tty */ irtty_stop_receiver(tty, TRUE); tty->flags &= ~(1 << TTY_DO_WRITE_WAKEUP); - if (tty->driver->stop) - tty->driver->stop(tty); + if (tty->ops->stop) + tty->ops->stop(tty); kfree(priv); diff --git a/drivers/net/ppp_async.c b/drivers/net/ppp_async.c index f023d5b67e6e..1c4b7e37912c 100644 --- a/drivers/net/ppp_async.c +++ b/drivers/net/ppp_async.c @@ -158,6 +158,9 @@ ppp_asynctty_open(struct tty_struct *tty) struct asyncppp *ap; int err; + if (tty->ops->write == NULL) + return -EOPNOTSUPP; + err = -ENOMEM; ap = kzalloc(sizeof(*ap), GFP_KERNEL); if (!ap) @@ -359,8 +362,8 @@ ppp_asynctty_receive(struct tty_struct *tty, const unsigned char *buf, tasklet_schedule(&ap->tsk); ap_put(ap); if (test_and_clear_bit(TTY_THROTTLED, &tty->flags) - && tty->driver->unthrottle) - tty->driver->unthrottle(tty); + && tty->ops->unthrottle) + tty->ops->unthrottle(tty); } static void @@ -676,7 +679,7 @@ ppp_async_push(struct asyncppp *ap) if (!tty_stuffed && ap->optr < ap->olim) { avail = ap->olim - ap->optr; set_bit(TTY_DO_WRITE_WAKEUP, &tty->flags); - sent = tty->driver->write(tty, ap->optr, avail); + sent = tty->ops->write(tty, ap->optr, avail); if (sent < 0) goto flush; /* error, e.g. loss of CD */ ap->optr += sent; diff --git a/drivers/net/ppp_synctty.c b/drivers/net/ppp_synctty.c index 0d80fa546719..48ed5fdbfe18 100644 --- a/drivers/net/ppp_synctty.c +++ b/drivers/net/ppp_synctty.c @@ -207,6 +207,9 @@ ppp_sync_open(struct tty_struct *tty) struct syncppp *ap; int err; + if (tty->ops->write == NULL) + return -EOPNOTSUPP; + ap = kzalloc(sizeof(*ap), GFP_KERNEL); err = -ENOMEM; if (!ap) @@ -399,8 +402,8 @@ ppp_sync_receive(struct tty_struct *tty, const unsigned char *buf, tasklet_schedule(&ap->tsk); sp_put(ap); if (test_and_clear_bit(TTY_THROTTLED, &tty->flags) - && tty->driver->unthrottle) - tty->driver->unthrottle(tty); + && tty->ops->unthrottle) + tty->ops->unthrottle(tty); } static void @@ -653,7 +656,7 @@ ppp_sync_push(struct syncppp *ap) tty_stuffed = 0; if (!tty_stuffed && ap->tpkt) { set_bit(TTY_DO_WRITE_WAKEUP, &tty->flags); - sent = tty->driver->write(tty, ap->tpkt->data, ap->tpkt->len); + sent = tty->ops->write(tty, ap->tpkt->data, ap->tpkt->len); if (sent < 0) goto flush; /* error, e.g. loss of CD */ if (sent < ap->tpkt->len) { diff --git a/drivers/net/slip.c b/drivers/net/slip.c index 5a55ede352f4..84af68fdb6c2 100644 --- a/drivers/net/slip.c +++ b/drivers/net/slip.c @@ -396,14 +396,14 @@ static void sl_encaps(struct slip *sl, unsigned char *icp, int len) /* Order of next two lines is *very* important. * When we are sending a little amount of data, - * the transfer may be completed inside driver.write() + * the transfer may be completed inside the ops->write() * routine, because it's running with interrupts enabled. * In this case we *never* got WRITE_WAKEUP event, * if we did not request it before write operation. * 14 Oct 1994 Dmitry Gorodchanin. */ sl->tty->flags |= (1 << TTY_DO_WRITE_WAKEUP); - actual = sl->tty->driver->write(sl->tty, sl->xbuff, count); + actual = sl->tty->ops->write(sl->tty, sl->xbuff, count); #ifdef SL_CHECK_TRANSMIT sl->dev->trans_start = jiffies; #endif @@ -437,7 +437,7 @@ static void slip_write_wakeup(struct tty_struct *tty) return; } - actual = tty->driver->write(tty, sl->xhead, sl->xleft); + actual = tty->ops->write(tty, sl->xhead, sl->xleft); sl->xleft -= actual; sl->xhead += actual; } @@ -462,7 +462,7 @@ static void sl_tx_timeout(struct net_device *dev) } printk(KERN_WARNING "%s: transmit timed out, %s?\n", dev->name, - (sl->tty->driver->chars_in_buffer(sl->tty) || sl->xleft) ? + (tty_chars_in_buffer(sl->tty) || sl->xleft) ? "bad line quality" : "driver error"); sl->xleft = 0; sl->tty->flags &= ~(1 << TTY_DO_WRITE_WAKEUP); @@ -830,6 +830,9 @@ static int slip_open(struct tty_struct *tty) if (!capable(CAP_NET_ADMIN)) return -EPERM; + if (tty->ops->write == NULL) + return -EOPNOTSUPP; + /* RTnetlink lock is misused here to serialize concurrent opens of slip channels. There are better ways, but it is the simplest one. @@ -1432,7 +1435,7 @@ static void sl_outfill(unsigned long sls) /* put END into tty queue. Is it right ??? */ if (!netif_queue_stopped(sl->dev)) { /* if device busy no outfill */ - sl->tty->driver->write(sl->tty, &s, 1); + sl->tty->ops->write(sl->tty, &s, 1); } } else set_bit(SLF_OUTWAIT, &sl->flags); diff --git a/drivers/net/wan/x25_asy.c b/drivers/net/wan/x25_asy.c index 0f8aca8a4d43..249e18053d5f 100644 --- a/drivers/net/wan/x25_asy.c +++ b/drivers/net/wan/x25_asy.c @@ -17,7 +17,7 @@ #include #include -#include +#include #include #include #include @@ -95,7 +95,7 @@ static struct x25_asy *x25_asy_alloc(void) x25_asy_devs[i] = dev; return sl; } else { - printk("x25_asy_alloc() - register_netdev() failure.\n"); + printk(KERN_WARNING "x25_asy_alloc() - register_netdev() failure.\n"); free_netdev(dev); } } @@ -112,23 +112,22 @@ static void x25_asy_free(struct x25_asy *sl) kfree(sl->xbuff); sl->xbuff = NULL; - if (!test_and_clear_bit(SLF_INUSE, &sl->flags)) { - printk("%s: x25_asy_free for already free unit.\n", sl->dev->name); - } + if (!test_and_clear_bit(SLF_INUSE, &sl->flags)) + printk(KERN_ERR "%s: x25_asy_free for already free unit.\n", + sl->dev->name); } static int x25_asy_change_mtu(struct net_device *dev, int newmtu) { struct x25_asy *sl = dev->priv; unsigned char *xbuff, *rbuff; - int len = 2* newmtu; + int len = 2 * newmtu; xbuff = kmalloc(len + 4, GFP_ATOMIC); rbuff = kmalloc(len + 4, GFP_ATOMIC); - if (xbuff == NULL || rbuff == NULL) - { - printk("%s: unable to grow X.25 buffers, MTU change cancelled.\n", + if (xbuff == NULL || rbuff == NULL) { + printk(KERN_WARNING "%s: unable to grow X.25 buffers, MTU change cancelled.\n", dev->name); kfree(xbuff); kfree(rbuff); @@ -193,25 +192,23 @@ static void x25_asy_bump(struct x25_asy *sl) int err; count = sl->rcount; - sl->stats.rx_bytes+=count; - + sl->stats.rx_bytes += count; + skb = dev_alloc_skb(count+1); - if (skb == NULL) - { - printk("%s: memory squeeze, dropping packet.\n", sl->dev->name); + if (skb == NULL) { + printk(KERN_WARNING "%s: memory squeeze, dropping packet.\n", + sl->dev->name); sl->stats.rx_dropped++; return; } - skb_push(skb,1); /* LAPB internal control */ - memcpy(skb_put(skb,count), sl->rbuff, count); + skb_push(skb, 1); /* LAPB internal control */ + memcpy(skb_put(skb, count), sl->rbuff, count); skb->protocol = x25_type_trans(skb, sl->dev); - if((err=lapb_data_received(skb->dev, skb))!=LAPB_OK) - { + err = lapb_data_received(skb->dev, skb); + if (err != LAPB_OK) { kfree_skb(skb); - printk(KERN_DEBUG "x25_asy: data received err - %d\n",err); - } - else - { + printk(KERN_DEBUG "x25_asy: data received err - %d\n", err); + } else { netif_rx(skb); sl->dev->last_rx = jiffies; sl->stats.rx_packets++; @@ -224,10 +221,11 @@ static void x25_asy_encaps(struct x25_asy *sl, unsigned char *icp, int len) unsigned char *p; int actual, count, mtu = sl->dev->mtu; - if (len > mtu) - { /* Sigh, shouldn't occur BUT ... */ + if (len > mtu) { + /* Sigh, shouldn't occur BUT ... */ len = mtu; - printk ("%s: truncating oversized transmit packet!\n", sl->dev->name); + printk(KERN_DEBUG "%s: truncating oversized transmit packet!\n", + sl->dev->name); sl->stats.tx_dropped++; x25_asy_unlock(sl); return; @@ -245,7 +243,7 @@ static void x25_asy_encaps(struct x25_asy *sl, unsigned char *icp, int len) * 14 Oct 1994 Dmitry Gorodchanin. */ sl->tty->flags |= (1 << TTY_DO_WRITE_WAKEUP); - actual = sl->tty->driver->write(sl->tty, sl->xbuff, count); + actual = sl->tty->ops->write(sl->tty, sl->xbuff, count); sl->xleft = count - actual; sl->xhead = sl->xbuff + actual; /* VSV */ @@ -265,8 +263,7 @@ static void x25_asy_write_wakeup(struct tty_struct *tty) if (!sl || sl->magic != X25_ASY_MAGIC || !netif_running(sl->dev)) return; - if (sl->xleft <= 0) - { + if (sl->xleft <= 0) { /* Now serial buffer is almost free & we can start * transmission of another packet */ sl->stats.tx_packets++; @@ -275,14 +272,14 @@ static void x25_asy_write_wakeup(struct tty_struct *tty) return; } - actual = tty->driver->write(tty, sl->xhead, sl->xleft); + actual = tty->ops->write(tty, sl->xhead, sl->xleft); sl->xleft -= actual; sl->xhead += actual; } static void x25_asy_timeout(struct net_device *dev) { - struct x25_asy *sl = (struct x25_asy*)(dev->priv); + struct x25_asy *sl = dev->priv; spin_lock(&sl->lock); if (netif_queue_stopped(dev)) { @@ -290,7 +287,7 @@ static void x25_asy_timeout(struct net_device *dev) * 14 Oct 1994 Dmitry Gorodchanin. */ printk(KERN_WARNING "%s: transmit timed out, %s?\n", dev->name, - (sl->tty->driver->chars_in_buffer(sl->tty) || sl->xleft) ? + (tty_chars_in_buffer(sl->tty) || sl->xleft) ? "bad line quality" : "driver error"); sl->xleft = 0; sl->tty->flags &= ~(1 << TTY_DO_WRITE_WAKEUP); @@ -303,31 +300,34 @@ static void x25_asy_timeout(struct net_device *dev) static int x25_asy_xmit(struct sk_buff *skb, struct net_device *dev) { - struct x25_asy *sl = (struct x25_asy*)(dev->priv); + struct x25_asy *sl = dev->priv; int err; if (!netif_running(sl->dev)) { - printk("%s: xmit call when iface is down\n", dev->name); + printk(KERN_ERR "%s: xmit call when iface is down\n", + dev->name); kfree_skb(skb); return 0; } - - switch(skb->data[0]) - { - case 0x00:break; - case 0x01: /* Connection request .. do nothing */ - if((err=lapb_connect_request(dev))!=LAPB_OK) - printk(KERN_ERR "x25_asy: lapb_connect_request error - %d\n", err); - kfree_skb(skb); - return 0; - case 0x02: /* Disconnect request .. do nothing - hang up ?? */ - if((err=lapb_disconnect_request(dev))!=LAPB_OK) - printk(KERN_ERR "x25_asy: lapb_disconnect_request error - %d\n", err); - default: - kfree_skb(skb); - return 0; + + switch (skb->data[0]) { + case 0x00: + break; + case 0x01: /* Connection request .. do nothing */ + err = lapb_connect_request(dev); + if (err != LAPB_OK) + printk(KERN_ERR "x25_asy: lapb_connect_request error - %d\n", err); + kfree_skb(skb); + return 0; + case 0x02: /* Disconnect request .. do nothing - hang up ?? */ + err = lapb_disconnect_request(dev); + if (err != LAPB_OK) + printk(KERN_ERR "x25_asy: lapb_disconnect_request error - %d\n", err); + default: + kfree_skb(skb); + return 0; } - skb_pull(skb,1); /* Remove control byte */ + skb_pull(skb, 1); /* Remove control byte */ /* * If we are busy already- too bad. We ought to be able * to queue things at this point, to allow for a little @@ -338,10 +338,10 @@ static int x25_asy_xmit(struct sk_buff *skb, struct net_device *dev) * So, no queues ! * 14 Oct 1994 Dmitry Gorodchanin. */ - - if((err=lapb_data_request(dev,skb))!=LAPB_OK) - { - printk(KERN_ERR "lapbeth: lapb_data_request error - %d\n", err); + + err = lapb_data_request(dev, skb); + if (err != LAPB_OK) { + printk(KERN_ERR "x25_asy: lapb_data_request error - %d\n", err); kfree_skb(skb); return 0; } @@ -357,7 +357,7 @@ static int x25_asy_xmit(struct sk_buff *skb, struct net_device *dev) * Called when I frame data arrives. We did the work above - throw it * at the net layer. */ - + static int x25_asy_data_indication(struct net_device *dev, struct sk_buff *skb) { skb->dev->last_rx = jiffies; @@ -369,24 +369,22 @@ static int x25_asy_data_indication(struct net_device *dev, struct sk_buff *skb) * busy cases too well. Its tricky to see how to do this nicely - * perhaps lapb should allow us to bounce this ? */ - + static void x25_asy_data_transmit(struct net_device *dev, struct sk_buff *skb) { - struct x25_asy *sl=dev->priv; - + struct x25_asy *sl = dev->priv; + spin_lock(&sl->lock); - if (netif_queue_stopped(sl->dev) || sl->tty == NULL) - { + if (netif_queue_stopped(sl->dev) || sl->tty == NULL) { spin_unlock(&sl->lock); printk(KERN_ERR "x25_asy: tbusy drop\n"); kfree_skb(skb); return; } /* We were not busy, so we are now... :-) */ - if (skb != NULL) - { + if (skb != NULL) { x25_asy_lock(sl); - sl->stats.tx_bytes+=skb->len; + sl->stats.tx_bytes += skb->len; x25_asy_encaps(sl, skb->data, skb->len); dev_kfree_skb(skb); } @@ -396,15 +394,16 @@ static void x25_asy_data_transmit(struct net_device *dev, struct sk_buff *skb) /* * LAPB connection establish/down information. */ - + static void x25_asy_connected(struct net_device *dev, int reason) { struct x25_asy *sl = dev->priv; struct sk_buff *skb; unsigned char *ptr; - if ((skb = dev_alloc_skb(1)) == NULL) { - printk(KERN_ERR "lapbeth: out of memory\n"); + skb = dev_alloc_skb(1); + if (skb == NULL) { + printk(KERN_ERR "x25_asy: out of memory\n"); return; } @@ -422,7 +421,8 @@ static void x25_asy_disconnected(struct net_device *dev, int reason) struct sk_buff *skb; unsigned char *ptr; - if ((skb = dev_alloc_skb(1)) == NULL) { + skb = dev_alloc_skb(1); + if (skb == NULL) { printk(KERN_ERR "x25_asy: out of memory\n"); return; } @@ -449,7 +449,7 @@ static struct lapb_register_struct x25_asy_callbacks = { /* Open the low-level part of the X.25 channel. Easy! */ static int x25_asy_open(struct net_device *dev) { - struct x25_asy *sl = (struct x25_asy*)(dev->priv); + struct x25_asy *sl = dev->priv; unsigned long len; int err; @@ -466,13 +466,11 @@ static int x25_asy_open(struct net_device *dev) len = dev->mtu * 2; sl->rbuff = kmalloc(len + 4, GFP_KERNEL); - if (sl->rbuff == NULL) { + if (sl->rbuff == NULL) goto norbuff; - } sl->xbuff = kmalloc(len + 4, GFP_KERNEL); - if (sl->xbuff == NULL) { + if (sl->xbuff == NULL) goto noxbuff; - } sl->buffsize = len; sl->rcount = 0; @@ -480,11 +478,12 @@ static int x25_asy_open(struct net_device *dev) sl->flags &= (1 << SLF_INUSE); /* Clear ESCAPE & ERROR flags */ netif_start_queue(dev); - + /* * Now attach LAPB */ - if((err=lapb_register(dev, &x25_asy_callbacks))==LAPB_OK) + err = lapb_register(dev, &x25_asy_callbacks); + if (err == LAPB_OK) return 0; /* Cleanup */ @@ -499,18 +498,20 @@ norbuff: /* Close the low-level part of the X.25 channel. Easy! */ static int x25_asy_close(struct net_device *dev) { - struct x25_asy *sl = (struct x25_asy*)(dev->priv); + struct x25_asy *sl = dev->priv; int err; spin_lock(&sl->lock); - if (sl->tty) + if (sl->tty) sl->tty->flags &= ~(1 << TTY_DO_WRITE_WAKEUP); netif_stop_queue(dev); sl->rcount = 0; sl->xleft = 0; - if((err=lapb_unregister(dev))!=LAPB_OK) - printk(KERN_ERR "x25_asy_close: lapb_unregister error -%d\n",err); + err = lapb_unregister(dev); + if (err != LAPB_OK) + printk(KERN_ERR "x25_asy_close: lapb_unregister error -%d\n", + err); spin_unlock(&sl->lock); return 0; } @@ -521,8 +522,9 @@ static int x25_asy_close(struct net_device *dev) * a block of X.25 data has been received, which can now be decapsulated * and sent on to some IP layer for further processing. */ - -static void x25_asy_receive_buf(struct tty_struct *tty, const unsigned char *cp, char *fp, int count) + +static void x25_asy_receive_buf(struct tty_struct *tty, + const unsigned char *cp, char *fp, int count) { struct x25_asy *sl = (struct x25_asy *) tty->disc_data; @@ -533,9 +535,8 @@ static void x25_asy_receive_buf(struct tty_struct *tty, const unsigned char *cp, /* Read the characters out of the buffer */ while (count--) { if (fp && *fp++) { - if (!test_and_set_bit(SLF_ERROR, &sl->flags)) { + if (!test_and_set_bit(SLF_ERROR, &sl->flags)) sl->stats.rx_errors++; - } cp++; continue; } @@ -556,31 +557,31 @@ static int x25_asy_open_tty(struct tty_struct *tty) struct x25_asy *sl = (struct x25_asy *) tty->disc_data; int err; + if (tty->ops->write == NULL) + return -EOPNOTSUPP; + /* First make sure we're not already connected. */ - if (sl && sl->magic == X25_ASY_MAGIC) { + if (sl && sl->magic == X25_ASY_MAGIC) return -EEXIST; - } /* OK. Find a free X.25 channel to use. */ - if ((sl = x25_asy_alloc()) == NULL) { + sl = x25_asy_alloc(); + if (sl == NULL) return -ENFILE; - } sl->tty = tty; tty->disc_data = sl; tty->receive_room = 65536; - if (tty->driver->flush_buffer) { - tty->driver->flush_buffer(tty); - } + tty_driver_flush_buffer(tty); tty_ldisc_flush(tty); /* Restore default settings */ sl->dev->type = ARPHRD_X25; - + /* Perform the low-level X.25 async init */ - if ((err = x25_asy_open(sl->dev))) + err = x25_asy_open(sl->dev); + if (err) return err; - /* Done. We have linked the TTY line to a channel. */ return sl->dev->base_addr; } @@ -601,9 +602,7 @@ static void x25_asy_close_tty(struct tty_struct *tty) return; if (sl->dev->flags & IFF_UP) - { - (void) dev_close(sl->dev); - } + dev_close(sl->dev); tty->disc_data = NULL; sl->tty = NULL; @@ -613,8 +612,7 @@ static void x25_asy_close_tty(struct tty_struct *tty) static struct net_device_stats *x25_asy_get_stats(struct net_device *dev) { - struct x25_asy *sl = (struct x25_asy*)(dev->priv); - + struct x25_asy *sl = dev->priv; return &sl->stats; } @@ -641,21 +639,19 @@ int x25_asy_esc(unsigned char *s, unsigned char *d, int len) * character sequence, according to the X.25 protocol. */ - while (len-- > 0) - { - switch(c = *s++) - { - case X25_END: - *ptr++ = X25_ESC; - *ptr++ = X25_ESCAPE(X25_END); - break; - case X25_ESC: - *ptr++ = X25_ESC; - *ptr++ = X25_ESCAPE(X25_ESC); - break; - default: - *ptr++ = c; - break; + while (len-- > 0) { + switch (c = *s++) { + case X25_END: + *ptr++ = X25_ESC; + *ptr++ = X25_ESCAPE(X25_END); + break; + case X25_ESC: + *ptr++ = X25_ESC; + *ptr++ = X25_ESCAPE(X25_ESC); + break; + default: + *ptr++ = c; + break; } } *ptr++ = X25_END; @@ -665,31 +661,25 @@ int x25_asy_esc(unsigned char *s, unsigned char *d, int len) static void x25_asy_unesc(struct x25_asy *sl, unsigned char s) { - switch(s) - { - case X25_END: - if (!test_and_clear_bit(SLF_ERROR, &sl->flags) && (sl->rcount > 2)) - { - x25_asy_bump(sl); - } - clear_bit(SLF_ESCAPE, &sl->flags); - sl->rcount = 0; - return; - - case X25_ESC: - set_bit(SLF_ESCAPE, &sl->flags); - return; - - case X25_ESCAPE(X25_ESC): - case X25_ESCAPE(X25_END): - if (test_and_clear_bit(SLF_ESCAPE, &sl->flags)) - s = X25_UNESCAPE(s); - break; - } - if (!test_bit(SLF_ERROR, &sl->flags)) - { - if (sl->rcount < sl->buffsize) - { + switch (s) { + case X25_END: + if (!test_and_clear_bit(SLF_ERROR, &sl->flags) + && sl->rcount > 2) + x25_asy_bump(sl); + clear_bit(SLF_ESCAPE, &sl->flags); + sl->rcount = 0; + return; + case X25_ESC: + set_bit(SLF_ESCAPE, &sl->flags); + return; + case X25_ESCAPE(X25_ESC): + case X25_ESCAPE(X25_END): + if (test_and_clear_bit(SLF_ESCAPE, &sl->flags)) + s = X25_UNESCAPE(s); + break; + } + if (!test_bit(SLF_ERROR, &sl->flags)) { + if (sl->rcount < sl->buffsize) { sl->rbuff[sl->rcount++] = s; return; } @@ -709,7 +699,7 @@ static int x25_asy_ioctl(struct tty_struct *tty, struct file *file, if (!sl || sl->magic != X25_ASY_MAGIC) return -EINVAL; - switch(cmd) { + switch (cmd) { case SIOCGIFNAME: if (copy_to_user((void __user *)arg, sl->dev->name, strlen(sl->dev->name) + 1)) @@ -724,8 +714,8 @@ static int x25_asy_ioctl(struct tty_struct *tty, struct file *file, static int x25_asy_open_dev(struct net_device *dev) { - struct x25_asy *sl = (struct x25_asy*)(dev->priv); - if(sl->tty==NULL) + struct x25_asy *sl = dev->priv; + if (sl->tty == NULL) return -ENODEV; return 0; } @@ -741,9 +731,9 @@ static void x25_asy_setup(struct net_device *dev) set_bit(SLF_INUSE, &sl->flags); /* - * Finish setting up the DEVICE info. + * Finish setting up the DEVICE info. */ - + dev->mtu = SL_MTU; dev->hard_start_xmit = x25_asy_xmit; dev->tx_timeout = x25_asy_timeout; @@ -778,9 +768,10 @@ static int __init init_x25_asy(void) x25_asy_maxdev = 4; /* Sanity */ printk(KERN_INFO "X.25 async: version 0.00 ALPHA " - "(dynamic channels, max=%d).\n", x25_asy_maxdev ); + "(dynamic channels, max=%d).\n", x25_asy_maxdev); - x25_asy_devs = kcalloc(x25_asy_maxdev, sizeof(struct net_device*), GFP_KERNEL); + x25_asy_devs = kcalloc(x25_asy_maxdev, sizeof(struct net_device *), + GFP_KERNEL); if (!x25_asy_devs) { printk(KERN_WARNING "X25 async: Can't allocate x25_asy_ctrls[] " "array! Uaargh! (-> No X.25 available)\n"); @@ -802,7 +793,7 @@ static void __exit exit_x25_asy(void) struct x25_asy *sl = dev->priv; spin_lock_bh(&sl->lock); - if (sl->tty) + if (sl->tty) tty_hangup(sl->tty); spin_unlock_bh(&sl->lock); diff --git a/drivers/serial/kgdboc.c b/drivers/serial/kgdboc.c index 9cf03327386a..eadc1ab6bbce 100644 --- a/drivers/serial/kgdboc.c +++ b/drivers/serial/kgdboc.c @@ -96,12 +96,14 @@ static void cleanup_kgdboc(void) static int kgdboc_get_char(void) { - return kgdb_tty_driver->poll_get_char(kgdb_tty_driver, kgdb_tty_line); + return kgdb_tty_driver->ops->poll_get_char(kgdb_tty_driver, + kgdb_tty_line); } static void kgdboc_put_char(u8 chr) { - kgdb_tty_driver->poll_put_char(kgdb_tty_driver, kgdb_tty_line, chr); + kgdb_tty_driver->ops->poll_put_char(kgdb_tty_driver, + kgdb_tty_line, chr); } static int param_set_kgdboc_var(const char *kmessage, struct kernel_param *kp) diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c index 6c7a5cf76582..1e2b9d826f69 100644 --- a/drivers/serial/serial_core.c +++ b/drivers/serial/serial_core.c @@ -532,15 +532,25 @@ uart_write(struct tty_struct *tty, const unsigned char *buf, int count) static int uart_write_room(struct tty_struct *tty) { struct uart_state *state = tty->driver_data; + unsigned long flags; + int ret; - return uart_circ_chars_free(&state->info->xmit); + spin_lock_irqsave(&state->port->lock, flags); + ret = uart_circ_chars_free(&state->info->xmit); + spin_unlock_irqrestore(&state->port->lock, flags); + return ret; } static int uart_chars_in_buffer(struct tty_struct *tty) { struct uart_state *state = tty->driver_data; + unsigned long flags; + int ret; - return uart_circ_chars_pending(&state->info->xmit); + spin_lock_irqsave(&state->port->lock, flags); + ret = uart_circ_chars_pending(&state->info->xmit); + spin_unlock_irqrestore(&state->port->lock, flags); + return ret; } static void uart_flush_buffer(struct tty_struct *tty) @@ -622,6 +632,11 @@ static int uart_get_info(struct uart_state *state, struct serial_struct tmp; memset(&tmp, 0, sizeof(tmp)); + + /* Ensure the state we copy is consistent and no hardware changes + occur as we go */ + mutex_lock(&state->mutex); + tmp.type = port->type; tmp.line = port->line; tmp.port = port->iobase; @@ -641,6 +656,8 @@ static int uart_get_info(struct uart_state *state, tmp.iomem_reg_shift = port->regshift; tmp.iomem_base = (void *)(unsigned long)port->mapbase; + mutex_unlock(&state->mutex); + if (copy_to_user(retinfo, &tmp, sizeof(*retinfo))) return -EFAULT; return 0; @@ -918,14 +935,12 @@ static void uart_break_ctl(struct tty_struct *tty, int break_state) struct uart_state *state = tty->driver_data; struct uart_port *port = state->port; - lock_kernel(); mutex_lock(&state->mutex); if (port->type != PORT_UNKNOWN) port->ops->break_ctl(port, break_state); mutex_unlock(&state->mutex); - unlock_kernel(); } static int uart_do_autoconfig(struct uart_state *state) @@ -1074,7 +1089,6 @@ uart_ioctl(struct tty_struct *tty, struct file *filp, unsigned int cmd, int ret = -ENOIOCTLCMD; - lock_kernel(); /* * These ioctls don't rely on the hardware to be present. */ @@ -1144,10 +1158,9 @@ uart_ioctl(struct tty_struct *tty, struct file *filp, unsigned int cmd, break; } } - out_up: +out_up: mutex_unlock(&state->mutex); - out: - unlock_kernel(); +out: return ret; } @@ -1173,7 +1186,6 @@ static void uart_set_termios(struct tty_struct *tty, return; } - lock_kernel(); uart_change_speed(state, old_termios); /* Handle transition to B0 status */ @@ -1206,7 +1218,6 @@ static void uart_set_termios(struct tty_struct *tty, } spin_unlock_irqrestore(&state->port->lock, flags); } - unlock_kernel(); #if 0 /* * No need to wake up processes in open wait, since they @@ -1322,11 +1333,11 @@ static void uart_wait_until_sent(struct tty_struct *tty, int timeout) struct uart_port *port = state->port; unsigned long char_time, expire; - BUG_ON(!kernel_locked()); - if (port->type == PORT_UNKNOWN || port->fifosize == 0) return; + lock_kernel(); + /* * Set the check interval to be 1/5 of the estimated time to * send a single character, and make it at least 1. The check @@ -1372,6 +1383,7 @@ static void uart_wait_until_sent(struct tty_struct *tty, int timeout) break; } set_current_state(TASK_RUNNING); /* might not be needed */ + unlock_kernel(); } /* @@ -2085,7 +2097,9 @@ int uart_resume_port(struct uart_driver *drv, struct uart_port *port) int ret; uart_change_pm(state, 0); + spin_lock_irq(&port->lock); ops->set_mctrl(port, 0); + spin_unlock_irq(&port->lock); ret = ops->startup(port); if (ret == 0) { uart_change_speed(state, NULL); diff --git a/drivers/usb/serial/digi_acceleport.c b/drivers/usb/serial/digi_acceleport.c index d17d1645714f..04a56f300ea6 100644 --- a/drivers/usb/serial/digi_acceleport.c +++ b/drivers/usb/serial/digi_acceleport.c @@ -1421,8 +1421,7 @@ static void digi_close(struct usb_serial_port *port, struct file *filp) tty_wait_until_sent(tty, DIGI_CLOSE_TIMEOUT); /* flush driver and line discipline buffers */ - if (tty->driver->flush_buffer) - tty->driver->flush_buffer(tty); + tty_driver_flush_buffer(tty); tty_ldisc_flush(tty); if (port->serial->dev) { diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c index a9934a3f9845..0cb0d77dc429 100644 --- a/drivers/usb/serial/usb-serial.c +++ b/drivers/usb/serial/usb-serial.c @@ -296,16 +296,14 @@ static int serial_write (struct tty_struct * tty, const unsigned char *buf, int struct usb_serial_port *port = tty->driver_data; int retval = -ENODEV; - if (!port || port->serial->dev->state == USB_STATE_NOTATTACHED) + if (port->serial->dev->state == USB_STATE_NOTATTACHED) goto exit; dbg("%s - port %d, %d byte(s)", __func__, port->number, count); - if (!port->open_count) { - retval = -EINVAL; - dbg("%s - port not opened", __func__); - goto exit; - } + /* open_count is managed under the mutex lock for the tty so cannot + drop to zero until after the last close completes */ + WARN_ON(!port->open_count); /* pass on to the driver specific version of this function */ retval = port->serial->type->write(port, buf, count); @@ -317,61 +315,28 @@ exit: static int serial_write_room (struct tty_struct *tty) { struct usb_serial_port *port = tty->driver_data; - int retval = -ENODEV; - - if (!port) - goto exit; - dbg("%s - port %d", __func__, port->number); - - if (!port->open_count) { - dbg("%s - port not open", __func__); - goto exit; - } - + WARN_ON(!port->open_count); /* pass on to the driver specific version of this function */ - retval = port->serial->type->write_room(port); - -exit: - return retval; + return port->serial->type->write_room(port); } static int serial_chars_in_buffer (struct tty_struct *tty) { struct usb_serial_port *port = tty->driver_data; - int retval = -ENODEV; - - if (!port) - goto exit; - dbg("%s = port %d", __func__, port->number); - if (!port->open_count) { - dbg("%s - port not open", __func__); - goto exit; - } - + WARN_ON(!port->open_count); /* pass on to the driver specific version of this function */ - retval = port->serial->type->chars_in_buffer(port); - -exit: - return retval; + return port->serial->type->chars_in_buffer(port); } static void serial_throttle (struct tty_struct * tty) { struct usb_serial_port *port = tty->driver_data; - - if (!port) - return; - dbg("%s - port %d", __func__, port->number); - if (!port->open_count) { - dbg ("%s - port not open", __func__); - return; - } - + WARN_ON(!port->open_count); /* pass on to the driver specific version of this function */ if (port->serial->type->throttle) port->serial->type->throttle(port); @@ -380,17 +345,9 @@ static void serial_throttle (struct tty_struct * tty) static void serial_unthrottle (struct tty_struct * tty) { struct usb_serial_port *port = tty->driver_data; - - if (!port) - return; - dbg("%s - port %d", __func__, port->number); - if (!port->open_count) { - dbg("%s - port not open", __func__); - return; - } - + WARN_ON(!port->open_count); /* pass on to the driver specific version of this function */ if (port->serial->type->unthrottle) port->serial->type->unthrottle(port); @@ -401,42 +358,27 @@ static int serial_ioctl (struct tty_struct *tty, struct file * file, unsigned in struct usb_serial_port *port = tty->driver_data; int retval = -ENODEV; - lock_kernel(); - if (!port) - goto exit; - dbg("%s - port %d, cmd 0x%.4x", __func__, port->number, cmd); - /* Caution - port->open_count is BKL protected */ - if (!port->open_count) { - dbg ("%s - port not open", __func__); - goto exit; - } + WARN_ON(!port->open_count); /* pass on to the driver specific version of this function if it is available */ - if (port->serial->type->ioctl) + if (port->serial->type->ioctl) { + lock_kernel(); retval = port->serial->type->ioctl(port, file, cmd, arg); + unlock_kernel(); + } else retval = -ENOIOCTLCMD; -exit: - unlock_kernel(); return retval; } static void serial_set_termios (struct tty_struct *tty, struct ktermios * old) { struct usb_serial_port *port = tty->driver_data; - - if (!port) - return; - dbg("%s - port %d", __func__, port->number); - if (!port->open_count) { - dbg("%s - port not open", __func__); - return; - } - + WARN_ON(!port->open_count); /* pass on to the driver specific version of this function if it is available */ if (port->serial->type->set_termios) port->serial->type->set_termios(port, old); @@ -448,24 +390,15 @@ static void serial_break (struct tty_struct *tty, int break_state) { struct usb_serial_port *port = tty->driver_data; - lock_kernel(); - if (!port) { - unlock_kernel(); - return; - } - dbg("%s - port %d", __func__, port->number); - if (!port->open_count) { - dbg("%s - port not open", __func__); - unlock_kernel(); - return; - } - + WARN_ON(!port->open_count); /* pass on to the driver specific version of this function if it is available */ - if (port->serial->type->break_ctl) + if (port->serial->type->break_ctl) { + lock_kernel(); port->serial->type->break_ctl(port, break_state); - unlock_kernel(); + unlock_kernel(); + } } static int serial_read_proc (char *page, char **start, off_t off, int count, int *eof, void *data) @@ -519,19 +452,11 @@ static int serial_tiocmget (struct tty_struct *tty, struct file *file) { struct usb_serial_port *port = tty->driver_data; - if (!port) - return -ENODEV; - dbg("%s - port %d", __func__, port->number); - if (!port->open_count) { - dbg("%s - port not open", __func__); - return -ENODEV; - } - + WARN_ON(!port->open_count); if (port->serial->type->tiocmget) return port->serial->type->tiocmget(port, file); - return -EINVAL; } @@ -540,19 +465,11 @@ static int serial_tiocmset (struct tty_struct *tty, struct file *file, { struct usb_serial_port *port = tty->driver_data; - if (!port) - return -ENODEV; - dbg("%s - port %d", __func__, port->number); - if (!port->open_count) { - dbg("%s - port not open", __func__); - return -ENODEV; - } - + WARN_ON(!port->open_count); if (port->serial->type->tiocmset) return port->serial->type->tiocmset(port, file, set, clear); - return -EINVAL; } diff --git a/drivers/usb/serial/whiteheat.c b/drivers/usb/serial/whiteheat.c index e96bf8663ffc..f07e8a4c1f3d 100644 --- a/drivers/usb/serial/whiteheat.c +++ b/drivers/usb/serial/whiteheat.c @@ -673,15 +673,13 @@ static void whiteheat_close(struct usb_serial_port *port, struct file * filp) } */ - if (port->tty->driver->flush_buffer) - port->tty->driver->flush_buffer(port->tty); + tty_driver_flush_buffer(port->tty); tty_ldisc_flush(port->tty); firm_report_tx_done(port); firm_close(port); -printk(KERN_ERR"Before processing rx_urbs_submitted.\n"); /* shutdown our bulk reads and writes */ mutex_lock(&info->deathwarrant); spin_lock_irq(&info->lock); diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 9663e8776724..97dba0d92348 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -1053,7 +1053,7 @@ static int vt_check(struct file *file) if (tty_paranoia_check(tty, inode, "tty_ioctl")) return -EINVAL; - if (tty->driver->ioctl != vt_ioctl) + if (tty->ops->ioctl != vt_ioctl) return -EINVAL; vc = (struct vc_data *)tty->driver_data; diff --git a/fs/proc/proc_tty.c b/fs/proc/proc_tty.c index ac26ccc25f42..21f490f5d65c 100644 --- a/fs/proc/proc_tty.c +++ b/fs/proc/proc_tty.c @@ -192,16 +192,14 @@ void proc_tty_register_driver(struct tty_driver *driver) { struct proc_dir_entry *ent; - if ((!driver->read_proc && !driver->write_proc) || - !driver->driver_name || + if (!driver->ops->read_proc || !driver->driver_name || driver->proc_entry) return; ent = create_proc_entry(driver->driver_name, 0, proc_tty_driver); if (!ent) return; - ent->read_proc = driver->read_proc; - ent->write_proc = driver->write_proc; + ent->read_proc = driver->ops->read_proc; ent->owner = driver->owner; ent->data = driver; diff --git a/include/linux/tty.h b/include/linux/tty.h index 2699298b00ef..c36a76da2ae2 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -177,9 +177,13 @@ struct signal_struct; * size each time the window is created or resized anyway. * - TYT, 9/14/92 */ + +struct tty_operations; + struct tty_struct { int magic; struct tty_driver *driver; + const struct tty_operations *ops; int index; struct tty_ldisc ldisc; struct mutex termios_mutex; @@ -295,6 +299,10 @@ extern void tty_unregister_device(struct tty_driver *driver, unsigned index); extern int tty_read_raw_data(struct tty_struct *tty, unsigned char *bufp, int buflen); extern void tty_write_message(struct tty_struct *tty, char *msg); +extern int tty_put_char(struct tty_struct *tty, unsigned char c); +extern int tty_chars_in_buffer(struct tty_struct *tty); +extern int tty_write_room(struct tty_struct *tty); +extern void tty_driver_flush_buffer(struct tty_struct *tty); extern int is_current_pgrp_orphaned(void); extern struct pid *tty_get_pgrp(struct tty_struct *tty); diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h index 21f69aca4505..f80d73b690f6 100644 --- a/include/linux/tty_driver.h +++ b/include/linux/tty_driver.h @@ -12,11 +12,15 @@ * This routine is called when a particular tty device is opened. * This routine is mandatory; if this routine is not filled in, * the attempted open will fail with ENODEV. + * + * Required method. * * void (*close)(struct tty_struct * tty, struct file * filp); * * This routine is called when a particular tty device is closed. * + * Required method. + * * int (*write)(struct tty_struct * tty, * const unsigned char *buf, int count); * @@ -26,7 +30,9 @@ * number of characters actually accepted for writing. This * routine is mandatory. * - * void (*put_char)(struct tty_struct *tty, unsigned char ch); + * Optional: Required for writable devices. + * + * int (*put_char)(struct tty_struct *tty, unsigned char ch); * * This routine is called by the kernel to write a single * character to the tty device. If the kernel uses this routine, @@ -34,10 +40,18 @@ * done stuffing characters into the driver. If there is no room * in the queue, the character is ignored. * + * Optional: Kernel will use the write method if not provided. + * + * Note: Do not call this function directly, call tty_put_char + * * void (*flush_chars)(struct tty_struct *tty); * * This routine is called by the kernel after it has written a * series of characters to the tty device using put_char(). + * + * Optional: + * + * Note: Do not call this function directly, call tty_driver_flush_chars * * int (*write_room)(struct tty_struct *tty); * @@ -45,6 +59,10 @@ * will accept for queuing to be written. This number is subject * to change as output buffers get emptied, or if the output flow * control is acted. + * + * Required if write method is provided else not needed. + * + * Note: Do not call this function directly, call tty_write_room * * int (*ioctl)(struct tty_struct *tty, struct file * file, * unsigned int cmd, unsigned long arg); @@ -53,22 +71,29 @@ * device-specific ioctl's. If the ioctl number passed in cmd * is not recognized by the driver, it should return ENOIOCTLCMD. * + * Optional + * * long (*compat_ioctl)(struct tty_struct *tty, struct file * file, * unsigned int cmd, unsigned long arg); * * implement ioctl processing for 32 bit process on 64 bit system + * + * Optional * * void (*set_termios)(struct tty_struct *tty, struct ktermios * old); * * This routine allows the tty driver to be notified when - * device's termios settings have changed. Note that a - * well-designed tty driver should be prepared to accept the case - * where old == NULL, and try to do something rational. + * device's termios settings have changed. + * + * Optional: Called under the termios lock + * * * void (*set_ldisc)(struct tty_struct *tty); * * This routine allows the tty driver to be notified when the * device's termios settings have changed. + * + * Optional: Called under BKL (currently) * * void (*throttle)(struct tty_struct * tty); * @@ -86,17 +111,27 @@ * * This routine notifies the tty driver that it should stop * outputting characters to the tty device. + * + * Optional: + * + * Note: Call stop_tty not this method. * * void (*start)(struct tty_struct *tty); * * This routine notifies the tty driver that it resume sending * characters to the tty device. + * + * Optional: + * + * Note: Call start_tty not this method. * * void (*hangup)(struct tty_struct *tty); * * This routine notifies the tty driver that it should hangup the * tty device. * + * Required: + * * void (*break_ctl)(struct tty_stuct *tty, int state); * * This optional routine requests the tty driver to turn on or @@ -106,18 +141,26 @@ * * If this routine is implemented, the high-level tty driver will * handle the following ioctls: TCSBRK, TCSBRKP, TIOCSBRK, - * TIOCCBRK. Otherwise, these ioctls will be passed down to the - * driver to handle. + * TIOCCBRK. + * + * Optional: Required for TCSBRK/BRKP/etc handling. * * void (*wait_until_sent)(struct tty_struct *tty, int timeout); * * This routine waits until the device has written out all of the * characters in its transmitter FIFO. * + * Optional: If not provided the device is assumed to have no FIFO + * + * Note: Usually correct to call tty_wait_until_sent + * * void (*send_xchar)(struct tty_struct *tty, char ch); * * This routine is used to send a high-priority XON/XOFF * character to the device. + * + * Optional: If not provided then the write method is called under + * the atomic write lock to keep it serialized with the ldisc. */ #include @@ -132,7 +175,7 @@ struct tty_operations { void (*close)(struct tty_struct * tty, struct file * filp); int (*write)(struct tty_struct * tty, const unsigned char *buf, int count); - void (*put_char)(struct tty_struct *tty, unsigned char ch); + int (*put_char)(struct tty_struct *tty, unsigned char ch); void (*flush_chars)(struct tty_struct *tty); int (*write_room)(struct tty_struct *tty); int (*chars_in_buffer)(struct tty_struct *tty); @@ -153,8 +196,6 @@ struct tty_operations { void (*send_xchar)(struct tty_struct *tty, char ch); int (*read_proc)(char *page, char **start, off_t off, int count, int *eof, void *data); - int (*write_proc)(struct file *file, const char __user *buffer, - unsigned long count, void *data); int (*tiocmget)(struct tty_struct *tty, struct file *file); int (*tiocmset)(struct tty_struct *tty, struct file *file, unsigned int set, unsigned int clear); @@ -190,48 +231,13 @@ struct tty_driver { struct tty_struct **ttys; struct ktermios **termios; struct ktermios **termios_locked; - void *driver_state; /* only used for the PTY driver */ - + void *driver_state; + /* - * Interface routines from the upper tty layer to the tty - * driver. Will be replaced with struct tty_operations. + * Driver methods */ - int (*open)(struct tty_struct * tty, struct file * filp); - void (*close)(struct tty_struct * tty, struct file * filp); - int (*write)(struct tty_struct * tty, - const unsigned char *buf, int count); - void (*put_char)(struct tty_struct *tty, unsigned char ch); - void (*flush_chars)(struct tty_struct *tty); - int (*write_room)(struct tty_struct *tty); - int (*chars_in_buffer)(struct tty_struct *tty); - int (*ioctl)(struct tty_struct *tty, struct file * file, - unsigned int cmd, unsigned long arg); - long (*compat_ioctl)(struct tty_struct *tty, struct file * file, - unsigned int cmd, unsigned long arg); - void (*set_termios)(struct tty_struct *tty, struct ktermios * old); - void (*throttle)(struct tty_struct * tty); - void (*unthrottle)(struct tty_struct * tty); - void (*stop)(struct tty_struct *tty); - void (*start)(struct tty_struct *tty); - void (*hangup)(struct tty_struct *tty); - void (*break_ctl)(struct tty_struct *tty, int state); - void (*flush_buffer)(struct tty_struct *tty); - void (*set_ldisc)(struct tty_struct *tty); - void (*wait_until_sent)(struct tty_struct *tty, int timeout); - void (*send_xchar)(struct tty_struct *tty, char ch); - int (*read_proc)(char *page, char **start, off_t off, - int count, int *eof, void *data); - int (*write_proc)(struct file *file, const char __user *buffer, - unsigned long count, void *data); - int (*tiocmget)(struct tty_struct *tty, struct file *file); - int (*tiocmset)(struct tty_struct *tty, struct file *file, - unsigned int set, unsigned int clear); -#ifdef CONFIG_CONSOLE_POLL - int (*poll_init)(struct tty_driver *driver, int line, char *options); - int (*poll_get_char)(struct tty_driver *driver, int line); - void (*poll_put_char)(struct tty_driver *driver, int line, char ch); -#endif + const struct tty_operations *ops; struct list_head tty_drivers; }; diff --git a/kernel/printk.c b/kernel/printk.c index d3f9c0f788bf..0d232589a923 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -1272,8 +1272,8 @@ late_initcall(disable_boot_consoles); */ void tty_write_message(struct tty_struct *tty, char *msg) { - if (tty && tty->driver->write) - tty->driver->write(tty, msg, strlen(msg)); + if (tty && tty->ops->write) + tty->ops->write(tty, msg, strlen(msg)); return; } diff --git a/net/irda/ircomm/ircomm_tty.c b/net/irda/ircomm/ircomm_tty.c index d2620410cb0a..76c3057d0179 100644 --- a/net/irda/ircomm/ircomm_tty.c +++ b/net/irda/ircomm/ircomm_tty.c @@ -555,10 +555,8 @@ static void ircomm_tty_close(struct tty_struct *tty, struct file *filp) ircomm_tty_shutdown(self); - if (tty->driver->flush_buffer) - tty->driver->flush_buffer(tty); - if (tty->ldisc.flush_buffer) - tty->ldisc.flush_buffer(tty); + tty_driver_flush_buffer(tty); + tty_ldisc_flush(tty); tty->closing = 0; self->tty = NULL; -- cgit v1.2.3 From 718a916338e821a10961e6a7a17430c18e5e58d9 Mon Sep 17 00:00:00 2001 From: Sukadev Bhattiprolu Date: Wed, 30 Apr 2008 00:54:21 -0700 Subject: devpts: factor out PTY index allocation Factor out the code used to allocate/free a pts index into new interfaces, devpts_new_index() and devpts_kill_index(). This localizes the external data structures used in managing the pts indices. [akpm@linux-foundation.org: undo accidental mutex2sem conversion] Signed-off-by: Sukadev Bhattiprolu Signed-off-by: Serge Hallyn Signed-off-by: Matt Helsley Acked-by: H. Peter Anvin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/tty_io.c | 40 ++++++---------------------------------- fs/devpts/inode.c | 43 ++++++++++++++++++++++++++++++++++++++++++- include/linux/devpts_fs.h | 4 ++++ 3 files changed, 52 insertions(+), 35 deletions(-) (limited to 'fs') diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index edcb7e471f02..1d298c2cf930 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -91,7 +91,6 @@ #include #include #include -#include #include #include #include @@ -137,9 +136,6 @@ EXPORT_SYMBOL(tty_mutex); #ifdef CONFIG_UNIX98_PTYS extern struct tty_driver *ptm_driver; /* Unix98 pty masters; for /dev/ptmx */ -extern int pty_limit; /* Config limit on Unix98 ptys */ -static DEFINE_IDR(allocated_ptys); -static DEFINE_MUTEX(allocated_ptys_lock); static int ptmx_open(struct inode *, struct file *); #endif @@ -2639,15 +2635,9 @@ static void release_dev(struct file *filp) */ release_tty(tty, idx); -#ifdef CONFIG_UNIX98_PTYS /* Make this pty number available for reallocation */ - if (devpts) { - mutex_lock(&allocated_ptys_lock); - idr_remove(&allocated_ptys, idx); - mutex_unlock(&allocated_ptys_lock); - } -#endif - + if (devpts) + devpts_kill_index(idx); } /** @@ -2803,29 +2793,13 @@ static int ptmx_open(struct inode *inode, struct file *filp) struct tty_struct *tty; int retval; int index; - int idr_ret; nonseekable_open(inode, filp); /* find a device that is not in use. */ - mutex_lock(&allocated_ptys_lock); - if (!idr_pre_get(&allocated_ptys, GFP_KERNEL)) { - mutex_unlock(&allocated_ptys_lock); - return -ENOMEM; - } - idr_ret = idr_get_new(&allocated_ptys, NULL, &index); - if (idr_ret < 0) { - mutex_unlock(&allocated_ptys_lock); - if (idr_ret == -EAGAIN) - return -ENOMEM; - return -EIO; - } - if (index >= pty_limit) { - idr_remove(&allocated_ptys, index); - mutex_unlock(&allocated_ptys_lock); - return -EIO; - } - mutex_unlock(&allocated_ptys_lock); + index = devpts_new_index(); + if (index < 0) + return index; mutex_lock(&tty_mutex); retval = init_dev(ptm_driver, index, &tty); @@ -2850,9 +2824,7 @@ out1: release_dev(filp); return retval; out: - mutex_lock(&allocated_ptys_lock); - idr_remove(&allocated_ptys, index); - mutex_unlock(&allocated_ptys_lock); + devpts_kill_index(index); return retval; } #endif diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index f120e1207874..285b64a8b06e 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -17,6 +17,8 @@ #include #include #include +#include +#include #include #include #include @@ -26,6 +28,10 @@ #define DEVPTS_DEFAULT_MODE 0600 +extern int pty_limit; /* Config limit on Unix98 ptys */ +static DEFINE_IDR(allocated_ptys); +static DEFINE_MUTEX(allocated_ptys_lock); + static struct vfsmount *devpts_mnt; static struct dentry *devpts_root; @@ -171,9 +177,44 @@ static struct dentry *get_node(int num) return lookup_one_len(s, root, sprintf(s, "%d", num)); } +int devpts_new_index(void) +{ + int index; + int idr_ret; + +retry: + if (!idr_pre_get(&allocated_ptys, GFP_KERNEL)) { + return -ENOMEM; + } + + mutex_lock(&allocated_ptys_lock); + idr_ret = idr_get_new(&allocated_ptys, NULL, &index); + if (idr_ret < 0) { + mutex_unlock(&allocated_ptys_lock); + if (idr_ret == -EAGAIN) + goto retry; + return -EIO; + } + + if (index >= pty_limit) { + idr_remove(&allocated_ptys, index); + mutex_unlock(&allocated_ptys_lock); + return -EIO; + } + mutex_unlock(&allocated_ptys_lock); + return index; +} + +void devpts_kill_index(int idx) +{ + mutex_lock(&allocated_ptys_lock); + idr_remove(&allocated_ptys, idx); + mutex_unlock(&allocated_ptys_lock); +} + int devpts_pty_new(struct tty_struct *tty) { - int number = tty->index; + int number = tty->index; /* tty layer puts index from devpts_new_index() in here */ struct tty_driver *driver = tty->driver; dev_t device = MKDEV(driver->major, driver->minor_start+number); struct dentry *dentry; diff --git a/include/linux/devpts_fs.h b/include/linux/devpts_fs.h index b672ddc00735..154769cad3f3 100644 --- a/include/linux/devpts_fs.h +++ b/include/linux/devpts_fs.h @@ -17,6 +17,8 @@ #ifdef CONFIG_UNIX98_PTYS +int devpts_new_index(void); +void devpts_kill_index(int idx); int devpts_pty_new(struct tty_struct *tty); /* mknod in devpts */ struct tty_struct *devpts_get_tty(int number); /* get tty structure */ void devpts_pty_kill(int number); /* unlink */ @@ -24,6 +26,8 @@ void devpts_pty_kill(int number); /* unlink */ #else /* Dummy stubs in the no-pty case */ +static inline int devpts_new_index(void) { return -EINVAL; } +static inline void devpts_kill_index(int idx) { } static inline int devpts_pty_new(struct tty_struct *tty) { return -EINVAL; } static inline struct tty_struct *devpts_get_tty(int number) { return NULL; } static inline void devpts_pty_kill(int number) { } -- cgit v1.2.3 From fa799759f9801137f665dbedda2c0815f1bf6f1b Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 30 Apr 2008 00:54:33 -0700 Subject: mm: bdi: expose the BDI object in sysfs for NFS Register NFS' backing_dev_info under sysfs with the name "nfs-MAJOR:MINOR" Signed-off-by: Miklos Szeredi Cc: Peter Zijlstra Cc: Trond Myklebust Cc: "J. Bruce Fields" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfs/super.c | 26 ++++++++++++++++++++++++++ 1 file changed, 26 insertions(+) (limited to 'fs') diff --git a/fs/nfs/super.c b/fs/nfs/super.c index fa220dc74609..7226a506f3ca 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -1575,6 +1575,11 @@ static int nfs_compare_super(struct super_block *sb, void *data) return nfs_compare_mount_options(sb, server, mntflags); } +static int nfs_bdi_register(struct nfs_server *server) +{ + return bdi_register_dev(&server->backing_dev_info, server->s_dev); +} + static int nfs_get_sb(struct file_system_type *fs_type, int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt) { @@ -1617,6 +1622,10 @@ static int nfs_get_sb(struct file_system_type *fs_type, if (s->s_fs_info != server) { nfs_free_server(server); server = NULL; + } else { + error = nfs_bdi_register(server); + if (error) + goto error_splat_super; } if (!s->s_root) { @@ -1664,6 +1673,7 @@ static void nfs_kill_super(struct super_block *s) { struct nfs_server *server = NFS_SB(s); + bdi_unregister(&server->backing_dev_info); kill_anon_super(s); nfs_free_server(server); } @@ -1708,6 +1718,10 @@ static int nfs_xdev_get_sb(struct file_system_type *fs_type, int flags, if (s->s_fs_info != server) { nfs_free_server(server); server = NULL; + } else { + error = nfs_bdi_register(server); + if (error) + goto error_splat_super; } if (!s->s_root) { @@ -1984,6 +1998,10 @@ static int nfs4_get_sb(struct file_system_type *fs_type, if (s->s_fs_info != server) { nfs_free_server(server); server = NULL; + } else { + error = nfs_bdi_register(server); + if (error) + goto error_splat_super; } if (!s->s_root) { @@ -2070,6 +2088,10 @@ static int nfs4_xdev_get_sb(struct file_system_type *fs_type, int flags, if (s->s_fs_info != server) { nfs_free_server(server); server = NULL; + } else { + error = nfs_bdi_register(server); + if (error) + goto error_splat_super; } if (!s->s_root) { @@ -2149,6 +2171,10 @@ static int nfs4_referral_get_sb(struct file_system_type *fs_type, int flags, if (s->s_fs_info != server) { nfs_free_server(server); server = NULL; + } else { + error = nfs_bdi_register(server); + if (error) + goto error_splat_super; } if (!s->s_root) { -- cgit v1.2.3 From b6f2fcbcfca9db2bd7aa24940224fcd3bbdbb8aa Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 30 Apr 2008 00:54:34 -0700 Subject: mm: bdi: expose the BDI object in sysfs for FUSE Register FUSE's backing_dev_info under sysfs with the name "fuse-MAJOR:MINOR" Make the fuse control filesystem use s_dev instead of a fuse specific ID. This makes it easier to match directories under /sys/fs/fuse/connections/ with directories under /sys/class/bdi, and with actual mounts. Signed-off-by: Miklos Szeredi Cc: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/control.c | 2 +- fs/fuse/fuse_i.h | 4 ++-- fs/fuse/inode.c | 30 +++++++++++++++--------------- 3 files changed, 18 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/fuse/control.c b/fs/fuse/control.c index 105d4a271e07..4f3cab321415 100644 --- a/fs/fuse/control.c +++ b/fs/fuse/control.c @@ -117,7 +117,7 @@ int fuse_ctl_add_conn(struct fuse_conn *fc) parent = fuse_control_sb->s_root; inc_nlink(parent->d_inode); - sprintf(name, "%llu", (unsigned long long) fc->id); + sprintf(name, "%u", fc->dev); parent = fuse_ctl_add_dentry(parent, fc, name, S_IFDIR | 0500, 2, &simple_dir_inode_operations, &simple_dir_operations); diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 67aaf6ee38ea..c0481e48d161 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -390,8 +390,8 @@ struct fuse_conn { /** Entry on the fuse_conn_list */ struct list_head entry; - /** Unique ID */ - u64 id; + /** Device ID from super block */ + dev_t dev; /** Dentries in the control filesystem */ struct dentry *ctl_dentry[FUSE_CTL_NUM_DENTRIES]; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 4df34da2284a..c4fcfd59cd80 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -447,7 +447,7 @@ static int fuse_show_options(struct seq_file *m, struct vfsmount *mnt) return 0; } -static struct fuse_conn *new_conn(void) +static struct fuse_conn *new_conn(struct super_block *sb) { struct fuse_conn *fc; int err; @@ -468,19 +468,26 @@ static struct fuse_conn *new_conn(void) atomic_set(&fc->num_waiting, 0); fc->bdi.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE; fc->bdi.unplug_io_fn = default_unplug_io_fn; + fc->dev = sb->s_dev; err = bdi_init(&fc->bdi); - if (err) { - kfree(fc); - fc = NULL; - goto out; - } + if (err) + goto error_kfree; + err = bdi_register_dev(&fc->bdi, fc->dev); + if (err) + goto error_bdi_destroy; fc->reqctr = 0; fc->blocked = 1; fc->attr_version = 1; get_random_bytes(&fc->scramble_key, sizeof(fc->scramble_key)); } -out: return fc; + +error_bdi_destroy: + bdi_destroy(&fc->bdi); +error_kfree: + mutex_destroy(&fc->inst_mutex); + kfree(fc); + return NULL; } void fuse_conn_put(struct fuse_conn *fc) @@ -578,12 +585,6 @@ static void fuse_send_init(struct fuse_conn *fc, struct fuse_req *req) request_send_background(fc, req); } -static u64 conn_id(void) -{ - static u64 ctr = 1; - return ctr++; -} - static int fuse_fill_super(struct super_block *sb, void *data, int silent) { struct fuse_conn *fc; @@ -621,7 +622,7 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) if (file->f_op != &fuse_dev_operations) return -EINVAL; - fc = new_conn(); + fc = new_conn(sb); if (!fc) return -ENOMEM; @@ -659,7 +660,6 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) if (file->private_data) goto err_unlock; - fc->id = conn_id(); err = fuse_ctl_add_conn(fc); if (err) goto err_unlock; -- cgit v1.2.3 From e4ad08fe64afca4ef79ecc4c624e6e871688da0d Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 30 Apr 2008 00:54:37 -0700 Subject: mm: bdi: add separate writeback accounting capability Add a new BDI capability flag: BDI_CAP_NO_ACCT_WB. If this flag is set, then don't update the per-bdi writeback stats from test_set_page_writeback() and test_clear_page_writeback(). Misc cleanups: - convert bdi_cap_writeback_dirty() and friends to static inline functions - create a flag that includes all three dirty/writeback related flags, since almst all users will want to have them toghether Signed-off-by: Miklos Szeredi Cc: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/configfs/inode.c | 2 +- fs/hugetlbfs/inode.c | 2 +- fs/ocfs2/dlm/dlmfs.c | 2 +- fs/ramfs/inode.c | 2 +- fs/sysfs/inode.c | 2 +- include/linux/backing-dev.h | 77 +++++++++++++++++++++++++++++++++------------ kernel/cgroup.c | 2 +- mm/page-writeback.c | 4 +-- mm/shmem.c | 2 +- mm/swap_state.c | 2 +- 10 files changed, 67 insertions(+), 30 deletions(-) (limited to 'fs') diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c index 4c1ebff778ee..b9a1d810346d 100644 --- a/fs/configfs/inode.c +++ b/fs/configfs/inode.c @@ -47,7 +47,7 @@ static const struct address_space_operations configfs_aops = { static struct backing_dev_info configfs_backing_dev_info = { .ra_pages = 0, /* No readahead */ - .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK, + .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK, }; static const struct inode_operations configfs_inode_operations ={ diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 9783723e8ffe..aeabf80f81a5 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -45,7 +45,7 @@ static const struct inode_operations hugetlbfs_inode_operations; static struct backing_dev_info hugetlbfs_backing_dev_info = { .ra_pages = 0, /* No readahead */ - .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK, + .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK, }; int sysctl_hugetlb_shm_group; diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c index 61a000f8524c..e48aba698b77 100644 --- a/fs/ocfs2/dlm/dlmfs.c +++ b/fs/ocfs2/dlm/dlmfs.c @@ -327,7 +327,7 @@ clear_fields: static struct backing_dev_info dlmfs_backing_dev_info = { .ra_pages = 0, /* No readahead */ - .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK, + .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK, }; static struct inode *dlmfs_get_root_inode(struct super_block *sb) diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c index 8428d5b2711d..b13123424e49 100644 --- a/fs/ramfs/inode.c +++ b/fs/ramfs/inode.c @@ -44,7 +44,7 @@ static const struct inode_operations ramfs_dir_inode_operations; static struct backing_dev_info ramfs_backing_dev_info = { .ra_pages = 0, /* No readahead */ - .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK | + .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK | BDI_CAP_MAP_DIRECT | BDI_CAP_MAP_COPY | BDI_CAP_READ_MAP | BDI_CAP_WRITE_MAP | BDI_CAP_EXEC_MAP, }; diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index d9262f74f94e..f8b82e73b3bf 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -30,7 +30,7 @@ static const struct address_space_operations sysfs_aops = { static struct backing_dev_info sysfs_backing_dev_info = { .ra_pages = 0, /* No readahead */ - .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK, + .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK, }; static const struct inode_operations sysfs_inode_operations ={ diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index c49a2d045e11..13ab79d99268 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -12,6 +12,7 @@ #include #include #include +#include #include struct page; @@ -151,22 +152,43 @@ int bdi_set_max_ratio(struct backing_dev_info *bdi, unsigned int max_ratio); /* * Flags in backing_dev_info::capability - * - The first two flags control whether dirty pages will contribute to the - * VM's accounting and whether writepages() should be called for dirty pages - * (something that would not, for example, be appropriate for ramfs) - * - These flags let !MMU mmap() govern direct device mapping vs immediate - * copying more easily for MAP_PRIVATE, especially for ROM filesystems + * + * The first three flags control whether dirty pages will contribute to the + * VM's accounting and whether writepages() should be called for dirty pages + * (something that would not, for example, be appropriate for ramfs) + * + * WARNING: these flags are closely related and should not normally be + * used separately. The BDI_CAP_NO_ACCT_AND_WRITEBACK combines these + * three flags into a single convenience macro. + * + * BDI_CAP_NO_ACCT_DIRTY: Dirty pages shouldn't contribute to accounting + * BDI_CAP_NO_WRITEBACK: Don't write pages back + * BDI_CAP_NO_ACCT_WB: Don't automatically account writeback pages + * + * These flags let !MMU mmap() govern direct device mapping vs immediate + * copying more easily for MAP_PRIVATE, especially for ROM filesystems. + * + * BDI_CAP_MAP_COPY: Copy can be mapped (MAP_PRIVATE) + * BDI_CAP_MAP_DIRECT: Can be mapped directly (MAP_SHARED) + * BDI_CAP_READ_MAP: Can be mapped for reading + * BDI_CAP_WRITE_MAP: Can be mapped for writing + * BDI_CAP_EXEC_MAP: Can be mapped for execution */ -#define BDI_CAP_NO_ACCT_DIRTY 0x00000001 /* Dirty pages shouldn't contribute to accounting */ -#define BDI_CAP_NO_WRITEBACK 0x00000002 /* Don't write pages back */ -#define BDI_CAP_MAP_COPY 0x00000004 /* Copy can be mapped (MAP_PRIVATE) */ -#define BDI_CAP_MAP_DIRECT 0x00000008 /* Can be mapped directly (MAP_SHARED) */ -#define BDI_CAP_READ_MAP 0x00000010 /* Can be mapped for reading */ -#define BDI_CAP_WRITE_MAP 0x00000020 /* Can be mapped for writing */ -#define BDI_CAP_EXEC_MAP 0x00000040 /* Can be mapped for execution */ +#define BDI_CAP_NO_ACCT_DIRTY 0x00000001 +#define BDI_CAP_NO_WRITEBACK 0x00000002 +#define BDI_CAP_MAP_COPY 0x00000004 +#define BDI_CAP_MAP_DIRECT 0x00000008 +#define BDI_CAP_READ_MAP 0x00000010 +#define BDI_CAP_WRITE_MAP 0x00000020 +#define BDI_CAP_EXEC_MAP 0x00000040 +#define BDI_CAP_NO_ACCT_WB 0x00000080 + #define BDI_CAP_VMFLAGS \ (BDI_CAP_READ_MAP | BDI_CAP_WRITE_MAP | BDI_CAP_EXEC_MAP) +#define BDI_CAP_NO_ACCT_AND_WRITEBACK \ + (BDI_CAP_NO_WRITEBACK | BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_ACCT_WB) + #if defined(VM_MAYREAD) && \ (BDI_CAP_READ_MAP != VM_MAYREAD || \ BDI_CAP_WRITE_MAP != VM_MAYWRITE || \ @@ -206,17 +228,32 @@ void clear_bdi_congested(struct backing_dev_info *bdi, int rw); void set_bdi_congested(struct backing_dev_info *bdi, int rw); long congestion_wait(int rw, long timeout); -#define bdi_cap_writeback_dirty(bdi) \ - (!((bdi)->capabilities & BDI_CAP_NO_WRITEBACK)) -#define bdi_cap_account_dirty(bdi) \ - (!((bdi)->capabilities & BDI_CAP_NO_ACCT_DIRTY)) +static inline bool bdi_cap_writeback_dirty(struct backing_dev_info *bdi) +{ + return !(bdi->capabilities & BDI_CAP_NO_WRITEBACK); +} + +static inline bool bdi_cap_account_dirty(struct backing_dev_info *bdi) +{ + return !(bdi->capabilities & BDI_CAP_NO_ACCT_DIRTY); +} -#define mapping_cap_writeback_dirty(mapping) \ - bdi_cap_writeback_dirty((mapping)->backing_dev_info) +static inline bool bdi_cap_account_writeback(struct backing_dev_info *bdi) +{ + /* Paranoia: BDI_CAP_NO_WRITEBACK implies BDI_CAP_NO_ACCT_WB */ + return !(bdi->capabilities & (BDI_CAP_NO_ACCT_WB | + BDI_CAP_NO_WRITEBACK)); +} -#define mapping_cap_account_dirty(mapping) \ - bdi_cap_account_dirty((mapping)->backing_dev_info) +static inline bool mapping_cap_writeback_dirty(struct address_space *mapping) +{ + return bdi_cap_writeback_dirty(mapping->backing_dev_info); +} +static inline bool mapping_cap_account_dirty(struct address_space *mapping) +{ + return bdi_cap_account_dirty(mapping->backing_dev_info); +} #endif /* _LINUX_BACKING_DEV_H */ diff --git a/kernel/cgroup.c b/kernel/cgroup.c index b9d467d83fc1..fbc6fc8949b4 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -575,7 +575,7 @@ static struct inode_operations cgroup_dir_inode_operations; static struct file_operations proc_cgroupstats_operations; static struct backing_dev_info cgroup_backing_dev_info = { - .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK, + .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK, }; static struct inode *cgroup_new_inode(mode_t mode, struct super_block *sb) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 2a9942f5387c..bbcb916190c9 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -1246,7 +1246,7 @@ int test_clear_page_writeback(struct page *page) radix_tree_tag_clear(&mapping->page_tree, page_index(page), PAGECACHE_TAG_WRITEBACK); - if (bdi_cap_writeback_dirty(bdi)) { + if (bdi_cap_account_writeback(bdi)) { __dec_bdi_stat(bdi, BDI_WRITEBACK); __bdi_writeout_inc(bdi); } @@ -1275,7 +1275,7 @@ int test_set_page_writeback(struct page *page) radix_tree_tag_set(&mapping->page_tree, page_index(page), PAGECACHE_TAG_WRITEBACK); - if (bdi_cap_writeback_dirty(bdi)) + if (bdi_cap_account_writeback(bdi)) __inc_bdi_stat(bdi, BDI_WRITEBACK); } if (!PageDirty(page)) diff --git a/mm/shmem.c b/mm/shmem.c index e6d9298aa22a..e2a6ae1a44e9 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -201,7 +201,7 @@ static struct vm_operations_struct shmem_vm_ops; static struct backing_dev_info shmem_backing_dev_info __read_mostly = { .ra_pages = 0, /* No readahead */ - .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK, + .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK, .unplug_io_fn = default_unplug_io_fn, }; diff --git a/mm/swap_state.c b/mm/swap_state.c index 50757ee3f9f3..d8aadaf2a0ba 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -33,7 +33,7 @@ static const struct address_space_operations swap_aops = { }; static struct backing_dev_info swap_backing_dev_info = { - .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK, + .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK, .unplug_io_fn = swap_unplug_io_fn, }; -- cgit v1.2.3 From fc3ba692a4d19019387c5acaea63131f9eab05dd Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 30 Apr 2008 00:54:38 -0700 Subject: mm: Add NR_WRITEBACK_TEMP counter Fuse will use temporary buffers to write back dirty data from memory mappings (normal writes are done synchronously). This is needed, because there cannot be any guarantee about the time in which a write will complete. By using temporary buffers, from the MM's point if view the page is written back immediately. If the writeout was due to memory pressure, this effectively migrates data from a full zone to a less full zone. This patch adds a new counter (NR_WRITEBACK_TEMP) for the number of pages used as temporary buffers. [Lee.Schermerhorn@hp.com: add vmstat_text for NR_WRITEBACK_TEMP] Signed-off-by: Miklos Szeredi Cc: Christoph Lameter Signed-off-by: Lee Schermerhorn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/node.c | 2 ++ fs/proc/proc_misc.c | 2 ++ include/linux/mmzone.h | 1 + mm/page-writeback.c | 3 ++- mm/vmstat.c | 1 + 5 files changed, 8 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/drivers/base/node.c b/drivers/base/node.c index 12fde2d03d69..39f3d1b3a213 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -77,6 +77,7 @@ static ssize_t node_read_meminfo(struct sys_device * dev, char * buf) "Node %d PageTables: %8lu kB\n" "Node %d NFS_Unstable: %8lu kB\n" "Node %d Bounce: %8lu kB\n" + "Node %d WritebackTmp: %8lu kB\n" "Node %d Slab: %8lu kB\n" "Node %d SReclaimable: %8lu kB\n" "Node %d SUnreclaim: %8lu kB\n", @@ -99,6 +100,7 @@ static ssize_t node_read_meminfo(struct sys_device * dev, char * buf) nid, K(node_page_state(nid, NR_PAGETABLE)), nid, K(node_page_state(nid, NR_UNSTABLE_NFS)), nid, K(node_page_state(nid, NR_BOUNCE)), + nid, K(node_page_state(nid, NR_WRITEBACK_TEMP)), nid, K(node_page_state(nid, NR_SLAB_RECLAIMABLE) + node_page_state(nid, NR_SLAB_UNRECLAIMABLE)), nid, K(node_page_state(nid, NR_SLAB_RECLAIMABLE)), diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index 48bcf20cec2f..74a323d2b850 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -179,6 +179,7 @@ static int meminfo_read_proc(char *page, char **start, off_t off, "PageTables: %8lu kB\n" "NFS_Unstable: %8lu kB\n" "Bounce: %8lu kB\n" + "WritebackTmp: %8lu kB\n" "CommitLimit: %8lu kB\n" "Committed_AS: %8lu kB\n" "VmallocTotal: %8lu kB\n" @@ -210,6 +211,7 @@ static int meminfo_read_proc(char *page, char **start, off_t off, K(global_page_state(NR_PAGETABLE)), K(global_page_state(NR_UNSTABLE_NFS)), K(global_page_state(NR_BOUNCE)), + K(global_page_state(NR_WRITEBACK_TEMP)), K(allowed), K(committed), (unsigned long)VMALLOC_TOTAL >> 10, diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index aad98003176f..ceb675d83a56 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -97,6 +97,7 @@ enum zone_stat_item { NR_UNSTABLE_NFS, /* NFS unstable pages */ NR_BOUNCE, NR_VMSCAN_WRITE, + NR_WRITEBACK_TEMP, /* Writeback using temporary buffers */ #ifdef CONFIG_NUMA NUMA_HIT, /* allocated in intended node */ NUMA_MISS, /* allocated in non intended node */ diff --git a/mm/page-writeback.c b/mm/page-writeback.c index c90a1e8e479f..789b6adbef37 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -211,7 +211,8 @@ clip_bdi_dirty_limit(struct backing_dev_info *bdi, long dirty, long *pbdi_dirty) avail_dirty = dirty - (global_page_state(NR_FILE_DIRTY) + global_page_state(NR_WRITEBACK) + - global_page_state(NR_UNSTABLE_NFS)); + global_page_state(NR_UNSTABLE_NFS) + + global_page_state(NR_WRITEBACK_TEMP)); if (avail_dirty < 0) avail_dirty = 0; diff --git a/mm/vmstat.c b/mm/vmstat.c index 280a7ed549f2..1a32130b958c 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -612,6 +612,7 @@ static const char * const vmstat_text[] = { "nr_unstable", "nr_bounce", "nr_vmscan_write", + "nr_writeback_temp", #ifdef CONFIG_NUMA "numa_hit", -- cgit v1.2.3 From 3be5a52b30aa5cf9d795b7634f728f612197b1c4 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 30 Apr 2008 00:54:41 -0700 Subject: fuse: support writable mmap Quoting Linus (3 years ago, FUSE inclusion discussions): "User-space filesystems are hard to get right. I'd claim that they are almost impossible, unless you limit them somehow (shared writable mappings are the nastiest part - if you don't have those, you can reasonably limit your problems by limiting the number of dirty pages you accept through normal "write()" calls)." Instead of attempting the impossible, I've just waited for the dirty page accounting infrastructure to materialize (thanks to Peter Zijlstra and others). This nicely solved the biggest problem: limiting the number of pages used for write caching. Some small details remained, however, which this largish patch attempts to address. It provides a page writeback implementation for fuse, which is completely safe against VM related deadlocks. Performance may not be very good for certain usage patterns, but generally it should be acceptable. It has been tested extensively with fsx-linux and bash-shared-mapping. Fuse page writeback design -------------------------- fuse_writepage() allocates a new temporary page with GFP_NOFS|__GFP_HIGHMEM. It copies the contents of the original page, and queues a WRITE request to the userspace filesystem using this temp page. The writeback is finished instantly from the MM's point of view: the page is removed from the radix trees, and the PageDirty and PageWriteback flags are cleared. For the duration of the actual write, the NR_WRITEBACK_TEMP counter is incremented. The per-bdi writeback count is not decremented until the actual write completes. On dirtying the page, fuse waits for a previous write to finish before proceeding. This makes sure, there can only be one temporary page used at a time for one cached page. This approach is wasteful in both memory and CPU bandwidth, so why is this complication needed? The basic problem is that there can be no guarantee about the time in which the userspace filesystem will complete a write. It may be buggy or even malicious, and fail to complete WRITE requests. We don't want unrelated parts of the system to grind to a halt in such cases. Also a filesystem may need additional resources (particularly memory) to complete a WRITE request. There's a great danger of a deadlock if that allocation may wait for the writepage to finish. Currently there are several cases where the kernel can block on page writeback: - allocation order is larger than PAGE_ALLOC_COSTLY_ORDER - page migration - throttle_vm_writeout (through NR_WRITEBACK) - sync(2) Of course in some cases (fsync, msync) we explicitly want to allow blocking. So for these cases new code has to be added to fuse, since the VM is not tracking writeback pages for us any more. As an extra safetly measure, the maximum dirty ratio allocated to a single fuse filesystem is set to 1% by default. This way one (or several) buggy or malicious fuse filesystems cannot slow down the rest of the system by hogging dirty memory. With appropriate privileges, this limit can be raised through '/sys/class/bdi//max_ratio'. Signed-off-by: Miklos Szeredi Cc: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/dev.c | 19 ++++ fs/fuse/dir.c | 84 ++++++++++++++- fs/fuse/file.c | 321 ++++++++++++++++++++++++++++++++++++++++++++++++++++--- fs/fuse/fuse_i.h | 37 +++++++ fs/fuse/inode.c | 49 +++++++-- 5 files changed, 481 insertions(+), 29 deletions(-) (limited to 'fs') diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index af639807524e..bba83762c484 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -47,6 +47,14 @@ struct fuse_req *fuse_request_alloc(void) return req; } +struct fuse_req *fuse_request_alloc_nofs(void) +{ + struct fuse_req *req = kmem_cache_alloc(fuse_req_cachep, GFP_NOFS); + if (req) + fuse_request_init(req); + return req; +} + void fuse_request_free(struct fuse_req *req) { kmem_cache_free(fuse_req_cachep, req); @@ -429,6 +437,17 @@ void request_send_background(struct fuse_conn *fc, struct fuse_req *req) request_send_nowait(fc, req); } +/* + * Called under fc->lock + * + * fc->connected must have been checked previously + */ +void request_send_background_locked(struct fuse_conn *fc, struct fuse_req *req) +{ + req->isreply = 1; + request_send_nowait_locked(fc, req); +} + /* * Lock the request. Up to the next unlock_request() there mustn't be * anything that could cause a page-fault. If the request was already diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index c4807b3fc8a3..48b9971ecd97 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -1106,6 +1106,50 @@ static void iattr_to_fattr(struct iattr *iattr, struct fuse_setattr_in *arg) } } +/* + * Prevent concurrent writepages on inode + * + * This is done by adding a negative bias to the inode write counter + * and waiting for all pending writes to finish. + */ +void fuse_set_nowrite(struct inode *inode) +{ + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_inode *fi = get_fuse_inode(inode); + + BUG_ON(!mutex_is_locked(&inode->i_mutex)); + + spin_lock(&fc->lock); + BUG_ON(fi->writectr < 0); + fi->writectr += FUSE_NOWRITE; + spin_unlock(&fc->lock); + wait_event(fi->page_waitq, fi->writectr == FUSE_NOWRITE); +} + +/* + * Allow writepages on inode + * + * Remove the bias from the writecounter and send any queued + * writepages. + */ +static void __fuse_release_nowrite(struct inode *inode) +{ + struct fuse_inode *fi = get_fuse_inode(inode); + + BUG_ON(fi->writectr != FUSE_NOWRITE); + fi->writectr = 0; + fuse_flush_writepages(inode); +} + +void fuse_release_nowrite(struct inode *inode) +{ + struct fuse_conn *fc = get_fuse_conn(inode); + + spin_lock(&fc->lock); + __fuse_release_nowrite(inode); + spin_unlock(&fc->lock); +} + /* * Set attributes, and at the same time refresh them. * @@ -1122,6 +1166,8 @@ static int fuse_do_setattr(struct dentry *entry, struct iattr *attr, struct fuse_req *req; struct fuse_setattr_in inarg; struct fuse_attr_out outarg; + bool is_truncate = false; + loff_t oldsize; int err; if (!fuse_allow_task(fc, current)) @@ -1145,12 +1191,16 @@ static int fuse_do_setattr(struct dentry *entry, struct iattr *attr, send_sig(SIGXFSZ, current, 0); return -EFBIG; } + is_truncate = true; } req = fuse_get_req(fc); if (IS_ERR(req)) return PTR_ERR(req); + if (is_truncate) + fuse_set_nowrite(inode); + memset(&inarg, 0, sizeof(inarg)); memset(&outarg, 0, sizeof(outarg)); iattr_to_fattr(attr, &inarg); @@ -1181,16 +1231,44 @@ static int fuse_do_setattr(struct dentry *entry, struct iattr *attr, if (err) { if (err == -EINTR) fuse_invalidate_attr(inode); - return err; + goto error; } if ((inode->i_mode ^ outarg.attr.mode) & S_IFMT) { make_bad_inode(inode); - return -EIO; + err = -EIO; + goto error; + } + + spin_lock(&fc->lock); + fuse_change_attributes_common(inode, &outarg.attr, + attr_timeout(&outarg)); + oldsize = inode->i_size; + i_size_write(inode, outarg.attr.size); + + if (is_truncate) { + /* NOTE: this may release/reacquire fc->lock */ + __fuse_release_nowrite(inode); + } + spin_unlock(&fc->lock); + + /* + * Only call invalidate_inode_pages2() after removing + * FUSE_NOWRITE, otherwise fuse_launder_page() would deadlock. + */ + if (S_ISREG(inode->i_mode) && oldsize != outarg.attr.size) { + if (outarg.attr.size < oldsize) + fuse_truncate(inode->i_mapping, outarg.attr.size); + invalidate_inode_pages2(inode->i_mapping); } - fuse_change_attributes(inode, &outarg.attr, attr_timeout(&outarg), 0); return 0; + +error: + if (is_truncate) + fuse_release_nowrite(inode); + + return err; } static int fuse_setattr(struct dentry *entry, struct iattr *attr) diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 676b0bc8a86d..68051f3bdf91 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -210,6 +210,49 @@ u64 fuse_lock_owner_id(struct fuse_conn *fc, fl_owner_t id) return (u64) v0 + ((u64) v1 << 32); } +/* + * Check if page is under writeback + * + * This is currently done by walking the list of writepage requests + * for the inode, which can be pretty inefficient. + */ +static bool fuse_page_is_writeback(struct inode *inode, pgoff_t index) +{ + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_inode *fi = get_fuse_inode(inode); + struct fuse_req *req; + bool found = false; + + spin_lock(&fc->lock); + list_for_each_entry(req, &fi->writepages, writepages_entry) { + pgoff_t curr_index; + + BUG_ON(req->inode != inode); + curr_index = req->misc.write.in.offset >> PAGE_CACHE_SHIFT; + if (curr_index == index) { + found = true; + break; + } + } + spin_unlock(&fc->lock); + + return found; +} + +/* + * Wait for page writeback to be completed. + * + * Since fuse doesn't rely on the VM writeback tracking, this has to + * use some other means. + */ +static int fuse_wait_on_page_writeback(struct inode *inode, pgoff_t index) +{ + struct fuse_inode *fi = get_fuse_inode(inode); + + wait_event(fi->page_waitq, !fuse_page_is_writeback(inode, index)); + return 0; +} + static int fuse_flush(struct file *file, fl_owner_t id) { struct inode *inode = file->f_path.dentry->d_inode; @@ -245,6 +288,21 @@ static int fuse_flush(struct file *file, fl_owner_t id) return err; } +/* + * Wait for all pending writepages on the inode to finish. + * + * This is currently done by blocking further writes with FUSE_NOWRITE + * and waiting for all sent writes to complete. + * + * This must be called under i_mutex, otherwise the FUSE_NOWRITE usage + * could conflict with truncation. + */ +static void fuse_sync_writes(struct inode *inode) +{ + fuse_set_nowrite(inode); + fuse_release_nowrite(inode); +} + int fuse_fsync_common(struct file *file, struct dentry *de, int datasync, int isdir) { @@ -261,6 +319,17 @@ int fuse_fsync_common(struct file *file, struct dentry *de, int datasync, if ((!isdir && fc->no_fsync) || (isdir && fc->no_fsyncdir)) return 0; + /* + * Start writeback against all dirty pages of the inode, then + * wait for all outstanding writes, before sending the FSYNC + * request. + */ + err = write_inode_now(inode, 0); + if (err) + return err; + + fuse_sync_writes(inode); + req = fuse_get_req(fc); if (IS_ERR(req)) return PTR_ERR(req); @@ -340,6 +409,13 @@ static int fuse_readpage(struct file *file, struct page *page) if (is_bad_inode(inode)) goto out; + /* + * Page writeback can extend beyond the liftime of the + * page-cache page, so make sure we read a properly synced + * page. + */ + fuse_wait_on_page_writeback(inode, page->index); + req = fuse_get_req(fc); err = PTR_ERR(req); if (IS_ERR(req)) @@ -411,6 +487,8 @@ static int fuse_readpages_fill(void *_data, struct page *page) struct inode *inode = data->inode; struct fuse_conn *fc = get_fuse_conn(inode); + fuse_wait_on_page_writeback(inode, page->index); + if (req->num_pages && (req->num_pages == FUSE_MAX_PAGES_PER_REQ || (req->num_pages + 1) * PAGE_CACHE_SIZE > fc->max_read || @@ -477,11 +555,10 @@ static ssize_t fuse_file_aio_read(struct kiocb *iocb, const struct iovec *iov, } static void fuse_write_fill(struct fuse_req *req, struct file *file, - struct inode *inode, loff_t pos, size_t count, - int writepage) + struct fuse_file *ff, struct inode *inode, + loff_t pos, size_t count, int writepage) { struct fuse_conn *fc = get_fuse_conn(inode); - struct fuse_file *ff = file->private_data; struct fuse_write_in *inarg = &req->misc.write.in; struct fuse_write_out *outarg = &req->misc.write.out; @@ -490,7 +567,7 @@ static void fuse_write_fill(struct fuse_req *req, struct file *file, inarg->offset = pos; inarg->size = count; inarg->write_flags = writepage ? FUSE_WRITE_CACHE : 0; - inarg->flags = file->f_flags; + inarg->flags = file ? file->f_flags : 0; req->in.h.opcode = FUSE_WRITE; req->in.h.nodeid = get_node_id(inode); req->in.argpages = 1; @@ -511,7 +588,7 @@ static size_t fuse_send_write(struct fuse_req *req, struct file *file, fl_owner_t owner) { struct fuse_conn *fc = get_fuse_conn(inode); - fuse_write_fill(req, file, inode, pos, count, 0); + fuse_write_fill(req, file, file->private_data, inode, pos, count, 0); if (owner != NULL) { struct fuse_write_in *inarg = &req->misc.write.in; inarg->write_flags |= FUSE_WRITE_LOCKOWNER; @@ -546,6 +623,12 @@ static int fuse_buffered_write(struct file *file, struct inode *inode, if (is_bad_inode(inode)) return -EIO; + /* + * Make sure writepages on the same page are not mixed up with + * plain writes. + */ + fuse_wait_on_page_writeback(inode, page->index); + req = fuse_get_req(fc); if (IS_ERR(req)) return PTR_ERR(req); @@ -716,21 +799,225 @@ static ssize_t fuse_direct_write(struct file *file, const char __user *buf, return res; } -static int fuse_file_mmap(struct file *file, struct vm_area_struct *vma) +static void fuse_writepage_free(struct fuse_conn *fc, struct fuse_req *req) { - if ((vma->vm_flags & VM_SHARED)) { - if ((vma->vm_flags & VM_WRITE)) - return -ENODEV; - else - vma->vm_flags &= ~VM_MAYWRITE; + __free_page(req->pages[0]); + fuse_file_put(req->ff); + fuse_put_request(fc, req); +} + +static void fuse_writepage_finish(struct fuse_conn *fc, struct fuse_req *req) +{ + struct inode *inode = req->inode; + struct fuse_inode *fi = get_fuse_inode(inode); + struct backing_dev_info *bdi = inode->i_mapping->backing_dev_info; + + list_del(&req->writepages_entry); + dec_bdi_stat(bdi, BDI_WRITEBACK); + dec_zone_page_state(req->pages[0], NR_WRITEBACK_TEMP); + bdi_writeout_inc(bdi); + wake_up(&fi->page_waitq); +} + +/* Called under fc->lock, may release and reacquire it */ +static void fuse_send_writepage(struct fuse_conn *fc, struct fuse_req *req) +{ + struct fuse_inode *fi = get_fuse_inode(req->inode); + loff_t size = i_size_read(req->inode); + struct fuse_write_in *inarg = &req->misc.write.in; + + if (!fc->connected) + goto out_free; + + if (inarg->offset + PAGE_CACHE_SIZE <= size) { + inarg->size = PAGE_CACHE_SIZE; + } else if (inarg->offset < size) { + inarg->size = size & (PAGE_CACHE_SIZE - 1); + } else { + /* Got truncated off completely */ + goto out_free; } - return generic_file_mmap(file, vma); + + req->in.args[1].size = inarg->size; + fi->writectr++; + request_send_background_locked(fc, req); + return; + + out_free: + fuse_writepage_finish(fc, req); + spin_unlock(&fc->lock); + fuse_writepage_free(fc, req); + spin_lock(&fc->lock); } -static int fuse_set_page_dirty(struct page *page) +/* + * If fi->writectr is positive (no truncate or fsync going on) send + * all queued writepage requests. + * + * Called with fc->lock + */ +void fuse_flush_writepages(struct inode *inode) { - printk("fuse_set_page_dirty: should not happen\n"); - dump_stack(); + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_inode *fi = get_fuse_inode(inode); + struct fuse_req *req; + + while (fi->writectr >= 0 && !list_empty(&fi->queued_writes)) { + req = list_entry(fi->queued_writes.next, struct fuse_req, list); + list_del_init(&req->list); + fuse_send_writepage(fc, req); + } +} + +static void fuse_writepage_end(struct fuse_conn *fc, struct fuse_req *req) +{ + struct inode *inode = req->inode; + struct fuse_inode *fi = get_fuse_inode(inode); + + mapping_set_error(inode->i_mapping, req->out.h.error); + spin_lock(&fc->lock); + fi->writectr--; + fuse_writepage_finish(fc, req); + spin_unlock(&fc->lock); + fuse_writepage_free(fc, req); +} + +static int fuse_writepage_locked(struct page *page) +{ + struct address_space *mapping = page->mapping; + struct inode *inode = mapping->host; + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_inode *fi = get_fuse_inode(inode); + struct fuse_req *req; + struct fuse_file *ff; + struct page *tmp_page; + + set_page_writeback(page); + + req = fuse_request_alloc_nofs(); + if (!req) + goto err; + + tmp_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM); + if (!tmp_page) + goto err_free; + + spin_lock(&fc->lock); + BUG_ON(list_empty(&fi->write_files)); + ff = list_entry(fi->write_files.next, struct fuse_file, write_entry); + req->ff = fuse_file_get(ff); + spin_unlock(&fc->lock); + + fuse_write_fill(req, NULL, ff, inode, page_offset(page), 0, 1); + + copy_highpage(tmp_page, page); + req->num_pages = 1; + req->pages[0] = tmp_page; + req->page_offset = 0; + req->end = fuse_writepage_end; + req->inode = inode; + + inc_bdi_stat(mapping->backing_dev_info, BDI_WRITEBACK); + inc_zone_page_state(tmp_page, NR_WRITEBACK_TEMP); + end_page_writeback(page); + + spin_lock(&fc->lock); + list_add(&req->writepages_entry, &fi->writepages); + list_add_tail(&req->list, &fi->queued_writes); + fuse_flush_writepages(inode); + spin_unlock(&fc->lock); + + return 0; + +err_free: + fuse_request_free(req); +err: + end_page_writeback(page); + return -ENOMEM; +} + +static int fuse_writepage(struct page *page, struct writeback_control *wbc) +{ + int err; + + err = fuse_writepage_locked(page); + unlock_page(page); + + return err; +} + +static int fuse_launder_page(struct page *page) +{ + int err = 0; + if (clear_page_dirty_for_io(page)) { + struct inode *inode = page->mapping->host; + err = fuse_writepage_locked(page); + if (!err) + fuse_wait_on_page_writeback(inode, page->index); + } + return err; +} + +/* + * Write back dirty pages now, because there may not be any suitable + * open files later + */ +static void fuse_vma_close(struct vm_area_struct *vma) +{ + filemap_write_and_wait(vma->vm_file->f_mapping); +} + +/* + * Wait for writeback against this page to complete before allowing it + * to be marked dirty again, and hence written back again, possibly + * before the previous writepage completed. + * + * Block here, instead of in ->writepage(), so that the userspace fs + * can only block processes actually operating on the filesystem. + * + * Otherwise unprivileged userspace fs would be able to block + * unrelated: + * + * - page migration + * - sync(2) + * - try_to_free_pages() with order > PAGE_ALLOC_COSTLY_ORDER + */ +static int fuse_page_mkwrite(struct vm_area_struct *vma, struct page *page) +{ + /* + * Don't use page->mapping as it may become NULL from a + * concurrent truncate. + */ + struct inode *inode = vma->vm_file->f_mapping->host; + + fuse_wait_on_page_writeback(inode, page->index); + return 0; +} + +static struct vm_operations_struct fuse_file_vm_ops = { + .close = fuse_vma_close, + .fault = filemap_fault, + .page_mkwrite = fuse_page_mkwrite, +}; + +static int fuse_file_mmap(struct file *file, struct vm_area_struct *vma) +{ + if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE)) { + struct inode *inode = file->f_dentry->d_inode; + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_inode *fi = get_fuse_inode(inode); + struct fuse_file *ff = file->private_data; + /* + * file may be written through mmap, so chain it onto the + * inodes's write_file list + */ + spin_lock(&fc->lock); + if (list_empty(&ff->write_entry)) + list_add(&ff->write_entry, &fi->write_files); + spin_unlock(&fc->lock); + } + file_accessed(file); + vma->vm_ops = &fuse_file_vm_ops; return 0; } @@ -940,10 +1227,12 @@ static const struct file_operations fuse_direct_io_file_operations = { static const struct address_space_operations fuse_file_aops = { .readpage = fuse_readpage, + .writepage = fuse_writepage, + .launder_page = fuse_launder_page, .write_begin = fuse_write_begin, .write_end = fuse_write_end, .readpages = fuse_readpages, - .set_page_dirty = fuse_set_page_dirty, + .set_page_dirty = __set_page_dirty_nobuffers, .bmap = fuse_bmap, }; diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index c0481e48d161..4b094fbc9c7f 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -15,6 +15,7 @@ #include #include #include +#include /** Max number of pages that can be used in a single read request */ #define FUSE_MAX_PAGES_PER_REQ 32 @@ -25,6 +26,9 @@ /** Congestion starts at 75% of maximum */ #define FUSE_CONGESTION_THRESHOLD (FUSE_MAX_BACKGROUND * 75 / 100) +/** Bias for fi->writectr, meaning new writepages must not be sent */ +#define FUSE_NOWRITE INT_MIN + /** It could be as large as PATH_MAX, but would that have any uses? */ #define FUSE_NAME_MAX 1024 @@ -73,6 +77,19 @@ struct fuse_inode { /** Files usable in writepage. Protected by fc->lock */ struct list_head write_files; + + /** Writepages pending on truncate or fsync */ + struct list_head queued_writes; + + /** Number of sent writes, a negative bias (FUSE_NOWRITE) + * means more writes are blocked */ + int writectr; + + /** Waitq for writepage completion */ + wait_queue_head_t page_waitq; + + /** List of writepage requestst (pending or sent) */ + struct list_head writepages; }; /** FUSE specific file data */ @@ -242,6 +259,12 @@ struct fuse_req { /** File used in the request (or NULL) */ struct fuse_file *ff; + /** Inode used in the request or NULL */ + struct inode *inode; + + /** Link on fi->writepages */ + struct list_head writepages_entry; + /** Request completion callback */ void (*end)(struct fuse_conn *, struct fuse_req *); @@ -504,6 +527,11 @@ void fuse_init_symlink(struct inode *inode); void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr, u64 attr_valid, u64 attr_version); +void fuse_change_attributes_common(struct inode *inode, struct fuse_attr *attr, + u64 attr_valid); + +void fuse_truncate(struct address_space *mapping, loff_t offset); + /** * Initialize the client device */ @@ -522,6 +550,8 @@ void fuse_ctl_cleanup(void); */ struct fuse_req *fuse_request_alloc(void); +struct fuse_req *fuse_request_alloc_nofs(void); + /** * Free a request */ @@ -558,6 +588,8 @@ void request_send_noreply(struct fuse_conn *fc, struct fuse_req *req); */ void request_send_background(struct fuse_conn *fc, struct fuse_req *req); +void request_send_background_locked(struct fuse_conn *fc, struct fuse_req *req); + /* Abort all requests */ void fuse_abort_conn(struct fuse_conn *fc); @@ -600,3 +632,8 @@ u64 fuse_lock_owner_id(struct fuse_conn *fc, fl_owner_t id); int fuse_update_attributes(struct inode *inode, struct kstat *stat, struct file *file, bool *refreshed); + +void fuse_flush_writepages(struct inode *inode); + +void fuse_set_nowrite(struct inode *inode); +void fuse_release_nowrite(struct inode *inode); diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index c4fcfd59cd80..7d01c68852a8 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -59,7 +59,11 @@ static struct inode *fuse_alloc_inode(struct super_block *sb) fi->nodeid = 0; fi->nlookup = 0; fi->attr_version = 0; + fi->writectr = 0; INIT_LIST_HEAD(&fi->write_files); + INIT_LIST_HEAD(&fi->queued_writes); + INIT_LIST_HEAD(&fi->writepages); + init_waitqueue_head(&fi->page_waitq); fi->forget_req = fuse_request_alloc(); if (!fi->forget_req) { kmem_cache_free(fuse_inode_cachep, inode); @@ -73,6 +77,7 @@ static void fuse_destroy_inode(struct inode *inode) { struct fuse_inode *fi = get_fuse_inode(inode); BUG_ON(!list_empty(&fi->write_files)); + BUG_ON(!list_empty(&fi->queued_writes)); if (fi->forget_req) fuse_request_free(fi->forget_req); kmem_cache_free(fuse_inode_cachep, inode); @@ -109,7 +114,7 @@ static int fuse_remount_fs(struct super_block *sb, int *flags, char *data) return 0; } -static void fuse_truncate(struct address_space *mapping, loff_t offset) +void fuse_truncate(struct address_space *mapping, loff_t offset) { /* See vmtruncate() */ unmap_mapping_range(mapping, offset + PAGE_SIZE - 1, 0, 1); @@ -117,19 +122,12 @@ static void fuse_truncate(struct address_space *mapping, loff_t offset) unmap_mapping_range(mapping, offset + PAGE_SIZE - 1, 0, 1); } - -void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr, - u64 attr_valid, u64 attr_version) +void fuse_change_attributes_common(struct inode *inode, struct fuse_attr *attr, + u64 attr_valid) { struct fuse_conn *fc = get_fuse_conn(inode); struct fuse_inode *fi = get_fuse_inode(inode); - loff_t oldsize; - spin_lock(&fc->lock); - if (attr_version != 0 && fi->attr_version > attr_version) { - spin_unlock(&fc->lock); - return; - } fi->attr_version = ++fc->attr_version; fi->i_time = attr_valid; @@ -159,6 +157,22 @@ void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr, fi->orig_i_mode = inode->i_mode; if (!(fc->flags & FUSE_DEFAULT_PERMISSIONS)) inode->i_mode &= ~S_ISVTX; +} + +void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr, + u64 attr_valid, u64 attr_version) +{ + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_inode *fi = get_fuse_inode(inode); + loff_t oldsize; + + spin_lock(&fc->lock); + if (attr_version != 0 && fi->attr_version > attr_version) { + spin_unlock(&fc->lock); + return; + } + + fuse_change_attributes_common(inode, attr, attr_valid); oldsize = inode->i_size; i_size_write(inode, attr->size); @@ -468,6 +482,8 @@ static struct fuse_conn *new_conn(struct super_block *sb) atomic_set(&fc->num_waiting, 0); fc->bdi.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE; fc->bdi.unplug_io_fn = default_unplug_io_fn; + /* fuse does it's own writeback accounting */ + fc->bdi.capabilities = BDI_CAP_NO_ACCT_WB; fc->dev = sb->s_dev; err = bdi_init(&fc->bdi); if (err) @@ -475,6 +491,19 @@ static struct fuse_conn *new_conn(struct super_block *sb) err = bdi_register_dev(&fc->bdi, fc->dev); if (err) goto error_bdi_destroy; + /* + * For a single fuse filesystem use max 1% of dirty + + * writeback threshold. + * + * This gives about 1M of write buffer for memory maps on a + * machine with 1G and 10% dirty_ratio, which should be more + * than enough. + * + * Privileged users can raise it by writing to + * + * /sys/class/bdi//max_ratio + */ + bdi_set_max_ratio(&fc->bdi, 1); fc->reqctr = 0; fc->blocked = 1; fc->attr_version = 1; -- cgit v1.2.3 From 854512ec358f291bcadd7daea10d6bf3704933de Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 30 Apr 2008 00:54:41 -0700 Subject: fuse: clean up setting i_size in write Extract common code for setting i_size in write functions into a common helper. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/file.c | 28 +++++++++++++++------------- 1 file changed, 15 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 68051f3bdf91..f0f0f278b4ea 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -610,13 +610,24 @@ static int fuse_write_begin(struct file *file, struct address_space *mapping, return 0; } +static void fuse_write_update_size(struct inode *inode, loff_t pos) +{ + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_inode *fi = get_fuse_inode(inode); + + spin_lock(&fc->lock); + fi->attr_version = ++fc->attr_version; + if (pos > inode->i_size) + i_size_write(inode, pos); + spin_unlock(&fc->lock); +} + static int fuse_buffered_write(struct file *file, struct inode *inode, loff_t pos, unsigned count, struct page *page) { int err; size_t nres; struct fuse_conn *fc = get_fuse_conn(inode); - struct fuse_inode *fi = get_fuse_inode(inode); unsigned offset = pos & (PAGE_CACHE_SIZE - 1); struct fuse_req *req; @@ -643,12 +654,7 @@ static int fuse_buffered_write(struct file *file, struct inode *inode, err = -EIO; if (!err) { pos += nres; - spin_lock(&fc->lock); - fi->attr_version = ++fc->attr_version; - if (pos > inode->i_size) - i_size_write(inode, pos); - spin_unlock(&fc->lock); - + fuse_write_update_size(inode, pos); if (count == PAGE_CACHE_SIZE) SetPageUptodate(page); } @@ -766,12 +772,8 @@ static ssize_t fuse_direct_io(struct file *file, const char __user *buf, } fuse_put_request(fc, req); if (res > 0) { - if (write) { - spin_lock(&fc->lock); - if (pos > inode->i_size) - i_size_write(inode, pos); - spin_unlock(&fc->lock); - } + if (write) + fuse_write_update_size(inode, pos); *ppos = pos; } fuse_invalidate_attr(inode); -- cgit v1.2.3 From ea9b9907b82a09bd1a708004454f7065de77c5b0 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 30 Apr 2008 00:54:42 -0700 Subject: fuse: implement perform_write Introduce fuse_perform_write. With fusexmp (a passthrough filesystem), large (1MB) writes into a backing tmpfs filesystem are sped up by almost 4 times (256MB/s vs 71MB/s). [mszeredi@suse.cz]: - split into smaller functions - testing - duplicate generic_file_aio_write(), so that there's no need to add a new ->perform_write() a_op. Comment from hch. Signed-off-by: Nick Piggin Signed-off-by: Miklos Szeredi Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/file.c | 194 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 193 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/fuse/file.c b/fs/fuse/file.c index f0f0f278b4ea..c5b5982bf386 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -677,6 +677,198 @@ static int fuse_write_end(struct file *file, struct address_space *mapping, return res; } +static size_t fuse_send_write_pages(struct fuse_req *req, struct file *file, + struct inode *inode, loff_t pos, + size_t count) +{ + size_t res; + unsigned offset; + unsigned i; + + for (i = 0; i < req->num_pages; i++) + fuse_wait_on_page_writeback(inode, req->pages[i]->index); + + res = fuse_send_write(req, file, inode, pos, count, NULL); + + offset = req->page_offset; + count = res; + for (i = 0; i < req->num_pages; i++) { + struct page *page = req->pages[i]; + + if (!req->out.h.error && !offset && count >= PAGE_CACHE_SIZE) + SetPageUptodate(page); + + if (count > PAGE_CACHE_SIZE - offset) + count -= PAGE_CACHE_SIZE - offset; + else + count = 0; + offset = 0; + + unlock_page(page); + page_cache_release(page); + } + + return res; +} + +static ssize_t fuse_fill_write_pages(struct fuse_req *req, + struct address_space *mapping, + struct iov_iter *ii, loff_t pos) +{ + struct fuse_conn *fc = get_fuse_conn(mapping->host); + unsigned offset = pos & (PAGE_CACHE_SIZE - 1); + size_t count = 0; + int err; + + req->page_offset = offset; + + do { + size_t tmp; + struct page *page; + pgoff_t index = pos >> PAGE_CACHE_SHIFT; + size_t bytes = min_t(size_t, PAGE_CACHE_SIZE - offset, + iov_iter_count(ii)); + + bytes = min_t(size_t, bytes, fc->max_write - count); + + again: + err = -EFAULT; + if (iov_iter_fault_in_readable(ii, bytes)) + break; + + err = -ENOMEM; + page = __grab_cache_page(mapping, index); + if (!page) + break; + + pagefault_disable(); + tmp = iov_iter_copy_from_user_atomic(page, ii, offset, bytes); + pagefault_enable(); + flush_dcache_page(page); + + if (!tmp) { + unlock_page(page); + page_cache_release(page); + bytes = min(bytes, iov_iter_single_seg_count(ii)); + goto again; + } + + err = 0; + req->pages[req->num_pages] = page; + req->num_pages++; + + iov_iter_advance(ii, tmp); + count += tmp; + pos += tmp; + offset += tmp; + if (offset == PAGE_CACHE_SIZE) + offset = 0; + + } while (iov_iter_count(ii) && count < fc->max_write && + req->num_pages < FUSE_MAX_PAGES_PER_REQ && offset == 0); + + return count > 0 ? count : err; +} + +static ssize_t fuse_perform_write(struct file *file, + struct address_space *mapping, + struct iov_iter *ii, loff_t pos) +{ + struct inode *inode = mapping->host; + struct fuse_conn *fc = get_fuse_conn(inode); + int err = 0; + ssize_t res = 0; + + if (is_bad_inode(inode)) + return -EIO; + + do { + struct fuse_req *req; + ssize_t count; + + req = fuse_get_req(fc); + if (IS_ERR(req)) { + err = PTR_ERR(req); + break; + } + + count = fuse_fill_write_pages(req, mapping, ii, pos); + if (count <= 0) { + err = count; + } else { + size_t num_written; + + num_written = fuse_send_write_pages(req, file, inode, + pos, count); + err = req->out.h.error; + if (!err) { + res += num_written; + pos += num_written; + + /* break out of the loop on short write */ + if (num_written != count) + err = -EIO; + } + } + fuse_put_request(fc, req); + } while (!err && iov_iter_count(ii)); + + if (res > 0) + fuse_write_update_size(inode, pos); + + fuse_invalidate_attr(inode); + + return res > 0 ? res : err; +} + +static ssize_t fuse_file_aio_write(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos) +{ + struct file *file = iocb->ki_filp; + struct address_space *mapping = file->f_mapping; + size_t count = 0; + ssize_t written = 0; + struct inode *inode = mapping->host; + ssize_t err; + struct iov_iter i; + + WARN_ON(iocb->ki_pos != pos); + + err = generic_segment_checks(iov, &nr_segs, &count, VERIFY_READ); + if (err) + return err; + + mutex_lock(&inode->i_mutex); + vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE); + + /* We can write back this queue in page reclaim */ + current->backing_dev_info = mapping->backing_dev_info; + + err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); + if (err) + goto out; + + if (count == 0) + goto out; + + err = remove_suid(file->f_path.dentry); + if (err) + goto out; + + file_update_time(file); + + iov_iter_init(&i, iov, nr_segs, count, 0); + written = fuse_perform_write(file, mapping, &i, pos); + if (written >= 0) + iocb->ki_pos = pos + written; + +out: + current->backing_dev_info = NULL; + mutex_unlock(&inode->i_mutex); + + return written ? written : err; +} + static void fuse_release_user_pages(struct fuse_req *req, int write) { unsigned i; @@ -1203,7 +1395,7 @@ static const struct file_operations fuse_file_operations = { .read = do_sync_read, .aio_read = fuse_file_aio_read, .write = do_sync_write, - .aio_write = generic_file_aio_write, + .aio_write = fuse_file_aio_write, .mmap = fuse_file_mmap, .open = fuse_open, .flush = fuse_flush, -- cgit v1.2.3 From 5c5c5e51b26413d50a9efae2ca7d6c5c6cd453ac Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 30 Apr 2008 00:54:43 -0700 Subject: fuse: update file size on short read If the READ request returned a short count, then either - cached size is incorrect - filesystem is buggy, as short reads are only allowed on EOF So assume that the size is wrong and refresh it, so that cached read() doesn't zero fill the missing chunk. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/dir.c | 2 +- fs/fuse/file.c | 52 ++++++++++++++++++++++++++++++++++++++++++++++------ fs/fuse/fuse_i.h | 7 ++++++- 3 files changed, 53 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 48b9971ecd97..2060bf06b906 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -132,7 +132,7 @@ static void fuse_lookup_init(struct fuse_req *req, struct inode *dir, req->out.args[0].value = outarg; } -static u64 fuse_get_attr_version(struct fuse_conn *fc) +u64 fuse_get_attr_version(struct fuse_conn *fc) { u64 curr_version; diff --git a/fs/fuse/file.c b/fs/fuse/file.c index c5b5982bf386..a02418c89d4b 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -363,7 +363,7 @@ static int fuse_fsync(struct file *file, struct dentry *de, int datasync) void fuse_read_fill(struct fuse_req *req, struct file *file, struct inode *inode, loff_t pos, size_t count, int opcode) { - struct fuse_read_in *inarg = &req->misc.read_in; + struct fuse_read_in *inarg = &req->misc.read.in; struct fuse_file *ff = file->private_data; inarg->fh = ff->fh; @@ -389,7 +389,7 @@ static size_t fuse_send_read(struct fuse_req *req, struct file *file, fuse_read_fill(req, file, inode, pos, count, FUSE_READ); if (owner != NULL) { - struct fuse_read_in *inarg = &req->misc.read_in; + struct fuse_read_in *inarg = &req->misc.read.in; inarg->read_flags |= FUSE_READ_LOCKOWNER; inarg->lock_owner = fuse_lock_owner_id(fc, owner); @@ -398,11 +398,29 @@ static size_t fuse_send_read(struct fuse_req *req, struct file *file, return req->out.args[0].size; } +static void fuse_read_update_size(struct inode *inode, loff_t size, + u64 attr_ver) +{ + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_inode *fi = get_fuse_inode(inode); + + spin_lock(&fc->lock); + if (attr_ver == fi->attr_version && size < inode->i_size) { + fi->attr_version = ++fc->attr_version; + i_size_write(inode, size); + } + spin_unlock(&fc->lock); +} + static int fuse_readpage(struct file *file, struct page *page) { struct inode *inode = page->mapping->host; struct fuse_conn *fc = get_fuse_conn(inode); struct fuse_req *req; + size_t num_read; + loff_t pos = page_offset(page); + size_t count = PAGE_CACHE_SIZE; + u64 attr_ver; int err; err = -EIO; @@ -421,15 +439,25 @@ static int fuse_readpage(struct file *file, struct page *page) if (IS_ERR(req)) goto out; + attr_ver = fuse_get_attr_version(fc); + req->out.page_zeroing = 1; req->num_pages = 1; req->pages[0] = page; - fuse_send_read(req, file, inode, page_offset(page), PAGE_CACHE_SIZE, - NULL); + num_read = fuse_send_read(req, file, inode, pos, count, NULL); err = req->out.h.error; fuse_put_request(fc, req); - if (!err) + + if (!err) { + /* + * Short read means EOF. If file size is larger, truncate it + */ + if (num_read < count) + fuse_read_update_size(inode, pos + num_read, attr_ver); + SetPageUptodate(page); + } + fuse_invalidate_attr(inode); /* atime changed */ out: unlock_page(page); @@ -439,8 +467,19 @@ static int fuse_readpage(struct file *file, struct page *page) static void fuse_readpages_end(struct fuse_conn *fc, struct fuse_req *req) { int i; + size_t count = req->misc.read.in.size; + size_t num_read = req->out.args[0].size; + struct inode *inode = req->pages[0]->mapping->host; - fuse_invalidate_attr(req->pages[0]->mapping->host); /* atime changed */ + /* + * Short read means EOF. If file size is larger, truncate it + */ + if (!req->out.h.error && num_read < count) { + loff_t pos = page_offset(req->pages[0]) + num_read; + fuse_read_update_size(inode, pos, req->misc.read.attr_ver); + } + + fuse_invalidate_attr(inode); /* atime changed */ for (i = 0; i < req->num_pages; i++) { struct page *page = req->pages[i]; @@ -463,6 +502,7 @@ static void fuse_send_readpages(struct fuse_req *req, struct file *file, size_t count = req->num_pages << PAGE_CACHE_SHIFT; req->out.page_zeroing = 1; fuse_read_fill(req, file, inode, pos, count, FUSE_READ); + req->misc.read.attr_ver = fuse_get_attr_version(fc); if (fc->async_read) { struct fuse_file *ff = file->private_data; req->ff = fuse_file_get(ff); diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 4b094fbc9c7f..934dd819a4ef 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -239,7 +239,10 @@ struct fuse_req { } release; struct fuse_init_in init_in; struct fuse_init_out init_out; - struct fuse_read_in read_in; + struct { + struct fuse_read_in in; + u64 attr_ver; + } read; struct { struct fuse_write_in in; struct fuse_write_out out; @@ -637,3 +640,5 @@ void fuse_flush_writepages(struct inode *inode); void fuse_set_nowrite(struct inode *inode); void fuse_release_nowrite(struct inode *inode); + +u64 fuse_get_attr_version(struct fuse_conn *fc); -- cgit v1.2.3 From e5d9a0df07484d6d191756878c974e4307fb24ce Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 30 Apr 2008 00:54:44 -0700 Subject: fuse: fix max i/o size calculation Fix a bug that Werner Baumann reported: fuse can send a bigger write request than the maximum specified. This only affected direct_io operation. In addition set a sane minimum for the max_read and max_write tunables, so I/O always makes some progress. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/file.c | 7 ++++--- fs/fuse/inode.c | 3 ++- 2 files changed, 6 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/fuse/file.c b/fs/fuse/file.c index a02418c89d4b..2d3649e42599 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -966,14 +966,15 @@ static ssize_t fuse_direct_io(struct file *file, const char __user *buf, while (count) { size_t nres; - size_t nbytes = min(count, nmax); - int err = fuse_get_user_pages(req, buf, nbytes, !write); + size_t nbytes_limit = min(count, nmax); + size_t nbytes; + int err = fuse_get_user_pages(req, buf, nbytes_limit, !write); if (err) { res = err; break; } nbytes = (req->num_pages << PAGE_SHIFT) - req->page_offset; - nbytes = min(count, nbytes); + nbytes = min(nbytes_limit, nbytes); if (write) nres = fuse_send_write(req, file, inode, pos, nbytes, current->files); diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 7d01c68852a8..0cef5ea319f3 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -584,6 +584,7 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_req *req) fc->bdi.ra_pages = min(fc->bdi.ra_pages, ra_pages); fc->minor = arg->minor; fc->max_write = arg->minor < 5 ? 4096 : arg->max_write; + fc->max_write = min_t(unsigned, 4096, fc->max_write); fc->conn_init = 1; } fuse_put_request(fc, req); @@ -658,7 +659,7 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) fc->flags = d.flags; fc->user_id = d.user_id; fc->group_id = d.group_id; - fc->max_read = d.max_read; + fc->max_read = min_t(unsigned, 4096, d.max_read); /* Used by get_root_inode() */ sb->s_fs_info = fc; -- cgit v1.2.3 From b48badf013018ef2aa4a46416454bdb18f77fb01 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 30 Apr 2008 00:54:44 -0700 Subject: fuse: fix node ID type Node ID is 64bit but it is passed as unsigned long to some functions. This breakage wasn't noticed, because libfuse uses unsigned long too. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/fuse_i.h | 4 ++-- fs/fuse/inode.c | 8 ++++---- 2 files changed, 6 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 934dd819a4ef..dadffa21a206 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -464,7 +464,7 @@ extern const struct file_operations fuse_dev_operations; /** * Get a filled in inode */ -struct inode *fuse_iget(struct super_block *sb, unsigned long nodeid, +struct inode *fuse_iget(struct super_block *sb, u64 nodeid, int generation, struct fuse_attr *attr, u64 attr_valid, u64 attr_version); @@ -472,7 +472,7 @@ struct inode *fuse_iget(struct super_block *sb, unsigned long nodeid, * Send FORGET command */ void fuse_send_forget(struct fuse_conn *fc, struct fuse_req *req, - unsigned long nodeid, u64 nlookup); + u64 nodeid, u64 nlookup); /** * Initialize READ or READDIR request diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 0cef5ea319f3..79b615873838 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -84,7 +84,7 @@ static void fuse_destroy_inode(struct inode *inode) } void fuse_send_forget(struct fuse_conn *fc, struct fuse_req *req, - unsigned long nodeid, u64 nlookup) + u64 nodeid, u64 nlookup) { struct fuse_forget_in *inarg = &req->misc.forget_in; inarg->nlookup = nlookup; @@ -207,7 +207,7 @@ static void fuse_init_inode(struct inode *inode, struct fuse_attr *attr) static int fuse_inode_eq(struct inode *inode, void *_nodeidp) { - unsigned long nodeid = *(unsigned long *) _nodeidp; + u64 nodeid = *(u64 *) _nodeidp; if (get_node_id(inode) == nodeid) return 1; else @@ -216,12 +216,12 @@ static int fuse_inode_eq(struct inode *inode, void *_nodeidp) static int fuse_inode_set(struct inode *inode, void *_nodeidp) { - unsigned long nodeid = *(unsigned long *) _nodeidp; + u64 nodeid = *(u64 *) _nodeidp; get_fuse_inode(inode)->nodeid = nodeid; return 0; } -struct inode *fuse_iget(struct super_block *sb, unsigned long nodeid, +struct inode *fuse_iget(struct super_block *sb, u64 nodeid, int generation, struct fuse_attr *attr, u64 attr_valid, u64 attr_version) { -- cgit v1.2.3 From 5559b8f4d1f630b8614b6c8e13b8bf6c9c45d7d7 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 30 Apr 2008 00:54:45 -0700 Subject: fuse: fix race in llseek Fuse doesn't use i_mutex to protect setting i_size, and so generic_file_llseek() can be racy: it doesn't use i_size_read(). So do a fuse specific llseek method, which does use i_size_read(). [akpm@linux-foundation.org: make `retval' loff_t] Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/file.c | 29 +++++++++++++++++++++++++++-- 1 file changed, 27 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 2d3649e42599..9ced35b00686 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -1431,8 +1431,33 @@ static sector_t fuse_bmap(struct address_space *mapping, sector_t block) return err ? 0 : outarg.block; } +static loff_t fuse_file_llseek(struct file *file, loff_t offset, int origin) +{ + loff_t retval; + struct inode *inode = file->f_path.dentry->d_inode; + + mutex_lock(&inode->i_mutex); + switch (origin) { + case SEEK_END: + offset += i_size_read(inode); + break; + case SEEK_CUR: + offset += file->f_pos; + } + retval = -EINVAL; + if (offset >= 0 && offset <= inode->i_sb->s_maxbytes) { + if (offset != file->f_pos) { + file->f_pos = offset; + file->f_version = 0; + } + retval = offset; + } + mutex_unlock(&inode->i_mutex); + return retval; +} + static const struct file_operations fuse_file_operations = { - .llseek = generic_file_llseek, + .llseek = fuse_file_llseek, .read = do_sync_read, .aio_read = fuse_file_aio_read, .write = do_sync_write, @@ -1448,7 +1473,7 @@ static const struct file_operations fuse_file_operations = { }; static const struct file_operations fuse_direct_io_file_operations = { - .llseek = generic_file_llseek, + .llseek = fuse_file_llseek, .read = fuse_direct_read, .write = fuse_direct_write, .open = fuse_open, -- cgit v1.2.3 From 4dbf930ed6c1f8aa992937d0461f8f70d4004aad Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 30 Apr 2008 00:54:45 -0700 Subject: fuse: fix sparse warnings fs/fuse/dev.c:306:2: warning: context imbalance in 'wait_answer_interruptible' - unexpected unlock fs/fuse/dev.c:361:2: warning: context imbalance in 'request_wait_answer' - unexpected unlock fs/fuse/dev.c:1002:4: warning: context imbalance in 'end_io_requests' - unexpected unlock Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/dev.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index bba83762c484..87250b6a8682 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -299,6 +299,7 @@ static void request_end(struct fuse_conn *fc, struct fuse_req *req) static void wait_answer_interruptible(struct fuse_conn *fc, struct fuse_req *req) + __releases(fc->lock) __acquires(fc->lock) { if (signal_pending(current)) return; @@ -315,8 +316,8 @@ static void queue_interrupt(struct fuse_conn *fc, struct fuse_req *req) kill_fasync(&fc->fasync, SIGIO, POLL_IN); } -/* Called with fc->lock held. Releases, and then reacquires it. */ static void request_wait_answer(struct fuse_conn *fc, struct fuse_req *req) + __releases(fc->lock) __acquires(fc->lock) { if (!fc->no_interrupt) { /* Any signal may interrupt this */ @@ -987,6 +988,7 @@ static void end_requests(struct fuse_conn *fc, struct list_head *head) * locked). */ static void end_io_requests(struct fuse_conn *fc) + __releases(fc->lock) __acquires(fc->lock) { while (!list_empty(&fc->io)) { struct fuse_req *req = -- cgit v1.2.3 From 86098fa0115358abf5159093d11ddb306ce4b0da Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 30 Apr 2008 00:54:46 -0700 Subject: reiserfs: use open_bdev_excl Use the proper helper to open a blockdevice by name for filesystem use, this makes sure it's properly claimed (also added for open-by-number) and gets rid of the struct file abuse. Tested by mounting a reiserfs filesystem with external journal. Signed-off-by: Christoph Hellwig Cc: Chris Mason Cc: Jeff Mahoney Acked-by: Edward Shishkin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/journal.c | 50 +++++++++++++++++++----------------------- include/linux/reiserfs_fs_sb.h | 1 - 2 files changed, 23 insertions(+), 28 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index da86042b3e03..e396b2fa4743 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -2574,11 +2574,9 @@ static int release_journal_dev(struct super_block *super, result = 0; - if (journal->j_dev_file != NULL) { - result = filp_close(journal->j_dev_file, NULL); - journal->j_dev_file = NULL; - journal->j_dev_bd = NULL; - } else if (journal->j_dev_bd != NULL) { + if (journal->j_dev_bd != NULL) { + if (journal->j_dev_bd->bd_dev != super->s_dev) + bd_release(journal->j_dev_bd); result = blkdev_put(journal->j_dev_bd); journal->j_dev_bd = NULL; } @@ -2603,7 +2601,6 @@ static int journal_init_dev(struct super_block *super, result = 0; journal->j_dev_bd = NULL; - journal->j_dev_file = NULL; jdev = SB_ONDISK_JOURNAL_DEVICE(super) ? new_decode_dev(SB_ONDISK_JOURNAL_DEVICE(super)) : super->s_dev; @@ -2620,35 +2617,34 @@ static int journal_init_dev(struct super_block *super, "cannot init journal device '%s': %i", __bdevname(jdev, b), result); return result; - } else if (jdev != super->s_dev) + } else if (jdev != super->s_dev) { + result = bd_claim(journal->j_dev_bd, journal); + if (result) { + blkdev_put(journal->j_dev_bd); + return result; + } + set_blocksize(journal->j_dev_bd, super->s_blocksize); + } + return 0; } - journal->j_dev_file = filp_open(jdev_name, 0, 0); - if (!IS_ERR(journal->j_dev_file)) { - struct inode *jdev_inode = journal->j_dev_file->f_mapping->host; - if (!S_ISBLK(jdev_inode->i_mode)) { - reiserfs_warning(super, "journal_init_dev: '%s' is " - "not a block device", jdev_name); - result = -ENOTBLK; - release_journal_dev(super, journal); - } else { - /* ok */ - journal->j_dev_bd = I_BDEV(jdev_inode); - set_blocksize(journal->j_dev_bd, super->s_blocksize); - reiserfs_info(super, - "journal_init_dev: journal device: %s\n", - bdevname(journal->j_dev_bd, b)); - } - } else { - result = PTR_ERR(journal->j_dev_file); - journal->j_dev_file = NULL; + journal->j_dev_bd = open_bdev_excl(jdev_name, 0, journal); + if (IS_ERR(journal->j_dev_bd)) { + result = PTR_ERR(journal->j_dev_bd); + journal->j_dev_bd = NULL; reiserfs_warning(super, "journal_init_dev: Cannot open '%s': %i", jdev_name, result); + return result; } - return result; + + set_blocksize(journal->j_dev_bd, super->s_blocksize); + reiserfs_info(super, + "journal_init_dev: journal device: %s\n", + bdevname(journal->j_dev_bd, b)); + return 0; } /** diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h index db5ef9b83c3f..336ee43ed7d8 100644 --- a/include/linux/reiserfs_fs_sb.h +++ b/include/linux/reiserfs_fs_sb.h @@ -177,7 +177,6 @@ struct reiserfs_journal { struct reiserfs_journal_cnode *j_last; /* newest journal block */ struct reiserfs_journal_cnode *j_first; /* oldest journal block. start here for traverse */ - struct file *j_dev_file; struct block_device *j_dev_bd; int j_1st_reserved_block; /* first block on s_dev of reserved area journal */ -- cgit v1.2.3 From 6369a4abb486692cd0f5fe592b48ec7419b7976c Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Wed, 30 Apr 2008 00:54:47 -0700 Subject: affs: be*_add_cpu conversion replace all: big_endian_variable = cpu_to_beX(beX_to_cpu(big_endian_variable) + expression_in_cpu_byteorder); with: beX_add_cpu(&big_endian_variable, expression_in_cpu_byteorder); generated with semantic patch Signed-off-by: Marcin Slusarz Cc: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/affs/file.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/affs/file.c b/fs/affs/file.c index e87ede608f77..1a4f092f24ef 100644 --- a/fs/affs/file.c +++ b/fs/affs/file.c @@ -539,7 +539,7 @@ affs_extent_file_ofs(struct inode *inode, u32 newsize) tmp = min(bsize - boff, newsize - size); BUG_ON(boff + tmp > bsize || tmp > bsize); memset(AFFS_DATA(bh) + boff, 0, tmp); - AFFS_DATA_HEAD(bh)->size = cpu_to_be32(be32_to_cpu(AFFS_DATA_HEAD(bh)->size) + tmp); + be32_add_cpu(&AFFS_DATA_HEAD(bh)->size, tmp); affs_fix_checksum(sb, bh); mark_buffer_dirty_inode(bh, inode); size += tmp; @@ -680,7 +680,7 @@ static int affs_write_end_ofs(struct file *file, struct address_space *mapping, tmp = min(bsize - boff, to - from); BUG_ON(boff + tmp > bsize || tmp > bsize); memcpy(AFFS_DATA(bh) + boff, data + from, tmp); - AFFS_DATA_HEAD(bh)->size = cpu_to_be32(be32_to_cpu(AFFS_DATA_HEAD(bh)->size) + tmp); + be32_add_cpu(&AFFS_DATA_HEAD(bh)->size, tmp); affs_fix_checksum(sb, bh); mark_buffer_dirty_inode(bh, inode); written += tmp; -- cgit v1.2.3 From 20c79e785ae3f813310261dde81b29ab0c3e28b4 Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Wed, 30 Apr 2008 00:54:47 -0700 Subject: hfs/hfsplus: be*_add_cpu conversion replace all: big_endian_variable = cpu_to_beX(beX_to_cpu(big_endian_variable) + expression_in_cpu_byteorder); with: beX_add_cpu(&big_endian_variable, expression_in_cpu_byteorder); generated with semantic patch Signed-off-by: Marcin Slusarz Cc: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hfs/mdb.c | 2 +- fs/hfsplus/super.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/hfs/mdb.c b/fs/hfs/mdb.c index b4651e128d7f..36ca2e1a4fa3 100644 --- a/fs/hfs/mdb.c +++ b/fs/hfs/mdb.c @@ -215,7 +215,7 @@ int hfs_mdb_get(struct super_block *sb) attrib &= cpu_to_be16(~HFS_SB_ATTRIB_UNMNT); attrib |= cpu_to_be16(HFS_SB_ATTRIB_INCNSTNT); mdb->drAtrb = attrib; - mdb->drWrCnt = cpu_to_be32(be32_to_cpu(mdb->drWrCnt) + 1); + be32_add_cpu(&mdb->drWrCnt, 1); mdb->drLsMod = hfs_mtime(); mark_buffer_dirty(HFS_SB(sb)->mdb_bh); diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 946466cd9f25..ce97a54518d8 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -423,7 +423,7 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) */ vhdr->last_mount_vers = cpu_to_be32(HFSP_MOUNT_VERSION); vhdr->modify_date = hfsp_now2mt(); - vhdr->write_count = cpu_to_be32(be32_to_cpu(vhdr->write_count) + 1); + be32_add_cpu(&vhdr->write_count, 1); vhdr->attributes &= cpu_to_be32(~HFSPLUS_VOL_UNMNT); vhdr->attributes |= cpu_to_be32(HFSPLUS_VOL_INCNSTNT); mark_buffer_dirty(HFSPLUS_SB(sb).s_vhbh); -- cgit v1.2.3 From e3592b12f507d2c12c883d9c18084b72a5710db3 Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Wed, 30 Apr 2008 00:54:48 -0700 Subject: quota: le*_add_cpu conversion replace all: little_endian_variable = cpu_to_leX(leX_to_cpu(little_endian_variable) + expression_in_cpu_byteorder); with: leX_add_cpu(&little_endian_variable, expression_in_cpu_byteorder); generated with semantic patch Signed-off-by: Marcin Slusarz Acked-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/quota_v2.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/quota_v2.c b/fs/quota_v2.c index 23b647f25d08..234ada903633 100644 --- a/fs/quota_v2.c +++ b/fs/quota_v2.c @@ -306,7 +306,7 @@ static uint find_free_dqentry(struct dquot *dquot, int *err) printk(KERN_ERR "VFS: find_free_dqentry(): Can't remove block (%u) from entry free list.\n", blk); goto out_buf; } - dh->dqdh_entries = cpu_to_le16(le16_to_cpu(dh->dqdh_entries)+1); + le16_add_cpu(&dh->dqdh_entries, 1); memset(&fakedquot, 0, sizeof(struct v2_disk_dqblk)); /* Find free structure in block */ for (i = 0; i < V2_DQSTRINBLK && memcmp(&fakedquot, ddquot+i, sizeof(struct v2_disk_dqblk)); i++); @@ -448,7 +448,7 @@ static int free_dqentry(struct dquot *dquot, uint blk) goto out_buf; } dh = (struct v2_disk_dqdbheader *)buf; - dh->dqdh_entries = cpu_to_le16(le16_to_cpu(dh->dqdh_entries)-1); + le16_add_cpu(&dh->dqdh_entries, -1); if (!le16_to_cpu(dh->dqdh_entries)) { /* Block got free? */ if ((ret = remove_free_dqentry(sb, type, buf, blk)) < 0 || (ret = put_free_dqblk(sb, type, buf, blk)) < 0) { -- cgit v1.2.3 From 07132922aac0caf807c56b9c2a388954b357a8c4 Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Wed, 30 Apr 2008 00:54:49 -0700 Subject: sysv: [bl]e*_add_cpu conversion replace all: big/little_endian_variable = cpu_to_[bl]eX([bl]eX_to_cpu(big/little_endian_variable) + expression_in_cpu_byteorder); with: [bl]eX_add_cpu(&big/little_endian_variable, expression_in_cpu_byteorder); generated with semantic patch Signed-off-by: Marcin Slusarz Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/sysv/sysv.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/sysv/sysv.h b/fs/sysv/sysv.h index 42d51d1c05cd..38ebe3f85b3d 100644 --- a/fs/sysv/sysv.h +++ b/fs/sysv/sysv.h @@ -217,9 +217,9 @@ static inline __fs32 fs32_add(struct sysv_sb_info *sbi, __fs32 *n, int d) if (sbi->s_bytesex == BYTESEX_PDP) *(__u32*)n = PDP_swab(PDP_swab(*(__u32*)n)+d); else if (sbi->s_bytesex == BYTESEX_LE) - *(__le32*)n = cpu_to_le32(le32_to_cpu(*(__le32*)n)+d); + le32_add_cpu((__le32 *)n, d); else - *(__be32*)n = cpu_to_be32(be32_to_cpu(*(__be32*)n)+d); + be32_add_cpu((__be32 *)n, d); return *n; } @@ -242,9 +242,9 @@ static inline __fs16 cpu_to_fs16(struct sysv_sb_info *sbi, __u16 n) static inline __fs16 fs16_add(struct sysv_sb_info *sbi, __fs16 *n, int d) { if (sbi->s_bytesex != BYTESEX_BE) - *(__le16*)n = cpu_to_le16(le16_to_cpu(*(__le16 *)n)+d); + le16_add_cpu((__le16 *)n, d); else - *(__be16*)n = cpu_to_be16(be16_to_cpu(*(__be16 *)n)+d); + be16_add_cpu((__be16 *)n, d); return *n; } -- cgit v1.2.3 From 3e5a5097303eedb4ffae2719843eb064221b1db4 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 30 Apr 2008 00:54:53 -0700 Subject: hfs: fix warning with 64k PAGE_SIZE fs/hfs/btree.c: In function 'hfs_bmap_alloc': fs/hfs/btree.c:263: warning: comparison is always false due to limited range of data type The patch makes the warning go away, but the code might actually be buggy? Cc: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hfs/btree.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/hfs/btree.c b/fs/hfs/btree.c index 24cf6fc43021..f6621a785202 100644 --- a/fs/hfs/btree.c +++ b/fs/hfs/btree.c @@ -208,7 +208,9 @@ struct hfs_bnode *hfs_bmap_alloc(struct hfs_btree *tree) struct hfs_bnode *node, *next_node; struct page **pagep; u32 nidx, idx; - u16 off, len; + unsigned off; + u16 off16; + u16 len; u8 *data, byte, m; int i; @@ -235,7 +237,8 @@ struct hfs_bnode *hfs_bmap_alloc(struct hfs_btree *tree) node = hfs_bnode_find(tree, nidx); if (IS_ERR(node)) return node; - len = hfs_brec_lenoff(node, 2, &off); + len = hfs_brec_lenoff(node, 2, &off16); + off = off16; off += node->page_offset; pagep = node->page + (off >> PAGE_CACHE_SHIFT); @@ -280,7 +283,8 @@ struct hfs_bnode *hfs_bmap_alloc(struct hfs_btree *tree) return next_node; node = next_node; - len = hfs_brec_lenoff(node, 0, &off); + len = hfs_brec_lenoff(node, 0, &off16); + off = off16; off += node->page_offset; pagep = node->page + (off >> PAGE_CACHE_SHIFT); data = kmap(*pagep); -- cgit v1.2.3 From 487798df6d25e76ed6558b3e17c44cf0458cc6f3 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 30 Apr 2008 00:54:54 -0700 Subject: hfsplus: fix warning with 64k PAGE_SIZE fs/hfsplus/btree.c: In function 'hfsplus_bmap_alloc': fs/hfsplus/btree.c:239: warning: comparison is always false due to limited range of data type But this might hide a real bug? Cc: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hfsplus/btree.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/hfsplus/btree.c b/fs/hfsplus/btree.c index bb5433608a42..e49fcee1e293 100644 --- a/fs/hfsplus/btree.c +++ b/fs/hfsplus/btree.c @@ -184,7 +184,9 @@ struct hfs_bnode *hfs_bmap_alloc(struct hfs_btree *tree) struct hfs_bnode *node, *next_node; struct page **pagep; u32 nidx, idx; - u16 off, len; + unsigned off; + u16 off16; + u16 len; u8 *data, byte, m; int i; @@ -211,7 +213,8 @@ struct hfs_bnode *hfs_bmap_alloc(struct hfs_btree *tree) node = hfs_bnode_find(tree, nidx); if (IS_ERR(node)) return node; - len = hfs_brec_lenoff(node, 2, &off); + len = hfs_brec_lenoff(node, 2, &off16); + off = off16; off += node->page_offset; pagep = node->page + (off >> PAGE_CACHE_SHIFT); @@ -256,7 +259,8 @@ struct hfs_bnode *hfs_bmap_alloc(struct hfs_btree *tree) return next_node; node = next_node; - len = hfs_brec_lenoff(node, 0, &off); + len = hfs_brec_lenoff(node, 0, &off16); + off = off16; off += node->page_offset; pagep = node->page + (off >> PAGE_CACHE_SHIFT); data = kmap(*pagep); -- cgit v1.2.3 From c6f3a97f86a5c97be0ca255976110bb9c3cfe669 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 30 Apr 2008 00:55:03 -0700 Subject: debugobjects: add timer specific object debugging code Add calls to the generic object debugging infrastructure and provide fixup functions which allow to keep the system alive when recoverable problems have been detected by the object debugging core code. Signed-off-by: Thomas Gleixner Acked-by: Ingo Molnar Cc: Greg KH Cc: Randy Dunlap Cc: Kay Sievers Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/parport/ieee1284.c | 4 +- fs/aio.c | 5 +- include/linux/poison.h | 7 +++ include/linux/timer.h | 23 ++++++- kernel/timer.c | 153 ++++++++++++++++++++++++++++++++++++++++++--- lib/Kconfig.debug | 8 +++ 6 files changed, 187 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/drivers/parport/ieee1284.c b/drivers/parport/ieee1284.c index 54a6ef72906e..0338b0912674 100644 --- a/drivers/parport/ieee1284.c +++ b/drivers/parport/ieee1284.c @@ -76,7 +76,7 @@ int parport_wait_event (struct parport *port, signed long timeout) semaphore. */ return 1; - init_timer (&timer); + init_timer_on_stack(&timer); timer.expires = jiffies + timeout; timer.function = timeout_waiting_on_port; port_from_cookie[port->number % PARPORT_MAX] = port; @@ -88,6 +88,8 @@ int parport_wait_event (struct parport *port, signed long timeout) /* Timed out. */ ret = 1; + destroy_timer_on_stack(&timer); + return ret; } diff --git a/fs/aio.c b/fs/aio.c index 99c2352906a0..b5253e77eb2f 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -1078,9 +1078,7 @@ static void timeout_func(unsigned long data) static inline void init_timeout(struct aio_timeout *to) { - init_timer(&to->timer); - to->timer.data = (unsigned long)to; - to->timer.function = timeout_func; + setup_timer_on_stack(&to->timer, timeout_func, (unsigned long) to); to->timed_out = 0; to->p = current; } @@ -1213,6 +1211,7 @@ retry: if (timeout) clear_timeout(&to); out: + destroy_timer_on_stack(&to.timer); return i ? i : ret; } diff --git a/include/linux/poison.h b/include/linux/poison.h index a9c31be7052c..9f31683728fd 100644 --- a/include/linux/poison.h +++ b/include/linux/poison.h @@ -10,6 +10,13 @@ #define LIST_POISON1 ((void *) 0x00100100) #define LIST_POISON2 ((void *) 0x00200200) +/********** include/linux/timer.h **********/ +/* + * Magic number "tsta" to indicate a static timer initializer + * for the object debugging code. + */ +#define TIMER_ENTRY_STATIC ((void *) 0x74737461) + /********** mm/slab.c **********/ /* * Magic nums for obj red zoning. diff --git a/include/linux/timer.h b/include/linux/timer.h index 979fefdeb862..d4ba79248a27 100644 --- a/include/linux/timer.h +++ b/include/linux/timer.h @@ -4,6 +4,7 @@ #include #include #include +#include struct tvec_base; @@ -25,6 +26,7 @@ struct timer_list { extern struct tvec_base boot_tvec_bases; #define TIMER_INITIALIZER(_function, _expires, _data) { \ + .entry = { .prev = TIMER_ENTRY_STATIC }, \ .function = (_function), \ .expires = (_expires), \ .data = (_data), \ @@ -38,6 +40,17 @@ extern struct tvec_base boot_tvec_bases; void init_timer(struct timer_list *timer); void init_timer_deferrable(struct timer_list *timer); +#ifdef CONFIG_DEBUG_OBJECTS_TIMERS +extern void init_timer_on_stack(struct timer_list *timer); +extern void destroy_timer_on_stack(struct timer_list *timer); +#else +static inline void destroy_timer_on_stack(struct timer_list *timer) { } +static inline void init_timer_on_stack(struct timer_list *timer) +{ + init_timer(timer); +} +#endif + static inline void setup_timer(struct timer_list * timer, void (*function)(unsigned long), unsigned long data) @@ -47,6 +60,15 @@ static inline void setup_timer(struct timer_list * timer, init_timer(timer); } +static inline void setup_timer_on_stack(struct timer_list *timer, + void (*function)(unsigned long), + unsigned long data) +{ + timer->function = function; + timer->data = data; + init_timer_on_stack(timer); +} + /** * timer_pending - is a timer pending? * @timer: the timer in question @@ -164,5 +186,4 @@ unsigned long __round_jiffies_relative(unsigned long j, int cpu); unsigned long round_jiffies(unsigned long j); unsigned long round_jiffies_relative(unsigned long j); - #endif diff --git a/kernel/timer.c b/kernel/timer.c index f3d35d4ea42e..ceacc6626572 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -320,14 +320,130 @@ static void timer_stats_account_timer(struct timer_list *timer) static void timer_stats_account_timer(struct timer_list *timer) {} #endif -/** - * init_timer - initialize a timer. - * @timer: the timer to be initialized - * - * init_timer() must be done to a timer prior calling *any* of the - * other timer functions. +#ifdef CONFIG_DEBUG_OBJECTS_TIMERS + +static struct debug_obj_descr timer_debug_descr; + +/* + * fixup_init is called when: + * - an active object is initialized */ -void init_timer(struct timer_list *timer) +static int timer_fixup_init(void *addr, enum debug_obj_state state) +{ + struct timer_list *timer = addr; + + switch (state) { + case ODEBUG_STATE_ACTIVE: + del_timer_sync(timer); + debug_object_init(timer, &timer_debug_descr); + return 1; + default: + return 0; + } +} + +/* + * fixup_activate is called when: + * - an active object is activated + * - an unknown object is activated (might be a statically initialized object) + */ +static int timer_fixup_activate(void *addr, enum debug_obj_state state) +{ + struct timer_list *timer = addr; + + switch (state) { + + case ODEBUG_STATE_NOTAVAILABLE: + /* + * This is not really a fixup. The timer was + * statically initialized. We just make sure that it + * is tracked in the object tracker. + */ + if (timer->entry.next == NULL && + timer->entry.prev == TIMER_ENTRY_STATIC) { + debug_object_init(timer, &timer_debug_descr); + debug_object_activate(timer, &timer_debug_descr); + return 0; + } else { + WARN_ON_ONCE(1); + } + return 0; + + case ODEBUG_STATE_ACTIVE: + WARN_ON(1); + + default: + return 0; + } +} + +/* + * fixup_free is called when: + * - an active object is freed + */ +static int timer_fixup_free(void *addr, enum debug_obj_state state) +{ + struct timer_list *timer = addr; + + switch (state) { + case ODEBUG_STATE_ACTIVE: + del_timer_sync(timer); + debug_object_free(timer, &timer_debug_descr); + return 1; + default: + return 0; + } +} + +static struct debug_obj_descr timer_debug_descr = { + .name = "timer_list", + .fixup_init = timer_fixup_init, + .fixup_activate = timer_fixup_activate, + .fixup_free = timer_fixup_free, +}; + +static inline void debug_timer_init(struct timer_list *timer) +{ + debug_object_init(timer, &timer_debug_descr); +} + +static inline void debug_timer_activate(struct timer_list *timer) +{ + debug_object_activate(timer, &timer_debug_descr); +} + +static inline void debug_timer_deactivate(struct timer_list *timer) +{ + debug_object_deactivate(timer, &timer_debug_descr); +} + +static inline void debug_timer_free(struct timer_list *timer) +{ + debug_object_free(timer, &timer_debug_descr); +} + +static void __init_timer(struct timer_list *timer); + +void init_timer_on_stack(struct timer_list *timer) +{ + debug_object_init_on_stack(timer, &timer_debug_descr); + __init_timer(timer); +} +EXPORT_SYMBOL_GPL(init_timer_on_stack); + +void destroy_timer_on_stack(struct timer_list *timer) +{ + debug_object_free(timer, &timer_debug_descr); +} +EXPORT_SYMBOL_GPL(destroy_timer_on_stack); + +#else +static inline void debug_timer_init(struct timer_list *timer) { } +static inline void debug_timer_activate(struct timer_list *timer) { } +static inline void debug_timer_deactivate(struct timer_list *timer) { } +#endif + +static void __init_timer(struct timer_list *timer) { timer->entry.next = NULL; timer->base = __raw_get_cpu_var(tvec_bases); @@ -337,6 +453,19 @@ void init_timer(struct timer_list *timer) memset(timer->start_comm, 0, TASK_COMM_LEN); #endif } + +/** + * init_timer - initialize a timer. + * @timer: the timer to be initialized + * + * init_timer() must be done to a timer prior calling *any* of the + * other timer functions. + */ +void init_timer(struct timer_list *timer) +{ + debug_timer_init(timer); + __init_timer(timer); +} EXPORT_SYMBOL(init_timer); void init_timer_deferrable(struct timer_list *timer) @@ -351,6 +480,8 @@ static inline void detach_timer(struct timer_list *timer, { struct list_head *entry = &timer->entry; + debug_timer_deactivate(timer); + __list_del(entry->prev, entry->next); if (clear_pending) entry->next = NULL; @@ -405,6 +536,8 @@ int __mod_timer(struct timer_list *timer, unsigned long expires) ret = 1; } + debug_timer_activate(timer); + new_base = __get_cpu_var(tvec_bases); if (base != new_base) { @@ -450,6 +583,7 @@ void add_timer_on(struct timer_list *timer, int cpu) BUG_ON(timer_pending(timer) || !timer->function); spin_lock_irqsave(&base->lock, flags); timer_set_base(timer, base); + debug_timer_activate(timer); internal_add_timer(base, timer); /* * Check whether the other CPU is idle and needs to be @@ -1086,11 +1220,14 @@ signed long __sched schedule_timeout(signed long timeout) expire = timeout + jiffies; - setup_timer(&timer, process_timeout, (unsigned long)current); + setup_timer_on_stack(&timer, process_timeout, (unsigned long)current); __mod_timer(&timer, expire); schedule(); del_singleshot_timer_sync(&timer); + /* Remove the timer from the object tracker */ + destroy_timer_on_stack(&timer); + timeout = expire - jiffies; out: diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 3e132b0a59cc..d2099f41aa1e 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -217,6 +217,14 @@ config DEBUG_OBJECTS_FREE properly. This can make kmalloc/kfree-intensive workloads much slower. +config DEBUG_OBJECTS_TIMERS + bool "Debug timer objects" + depends on DEBUG_OBJECTS + help + If you say Y here, additional code will be inserted into the + timer routines to track the life time of timer objects and + validate the timer operations. + config DEBUG_SLAB bool "Debug slab memory allocations" depends on DEBUG_KERNEL && SLAB -- cgit v1.2.3 From 530b6412786d7f83592c1a8e2445541ed73fca76 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Wed, 30 Apr 2008 00:55:09 -0700 Subject: afs: replace remaining __FUNCTION__ occurrences __FUNCTION__ is gcc-specific, use __func__ Signed-off-by: Harvey Harrison Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/afs/dir.c | 4 ++-- fs/afs/internal.h | 8 ++++---- 2 files changed, 6 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/afs/dir.c b/fs/afs/dir.c index b58af8f18bc4..dfda03d4397d 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -140,7 +140,7 @@ static inline void afs_dir_check_page(struct inode *dir, struct page *page) if (page->index == 0 && qty != ntohs(dbuf->blocks[0].pagehdr.npages)) { printk("kAFS: %s(%lu): wrong number of dir blocks %d!=%hu\n", - __FUNCTION__, dir->i_ino, qty, + __func__, dir->i_ino, qty, ntohs(dbuf->blocks[0].pagehdr.npages)); goto error; } @@ -159,7 +159,7 @@ static inline void afs_dir_check_page(struct inode *dir, struct page *page) for (tmp = 0; tmp < qty; tmp++) { if (dbuf->blocks[tmp].pagehdr.magic != AFS_DIR_MAGIC) { printk("kAFS: %s(%lu): bad magic %d/%d is %04hx\n", - __FUNCTION__, dir->i_ino, tmp, qty, + __func__, dir->i_ino, tmp, qty, ntohs(dbuf->blocks[tmp].pagehdr.magic)); goto error; } diff --git a/fs/afs/internal.h b/fs/afs/internal.h index eec41c76de72..7102824ba847 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -757,8 +757,8 @@ void _dbprintk(const char *fmt, ...) { } -#define kenter(FMT,...) dbgprintk("==> %s("FMT")",__FUNCTION__ ,##__VA_ARGS__) -#define kleave(FMT,...) dbgprintk("<== %s()"FMT"",__FUNCTION__ ,##__VA_ARGS__) +#define kenter(FMT,...) dbgprintk("==> %s("FMT")",__func__ ,##__VA_ARGS__) +#define kleave(FMT,...) dbgprintk("<== %s()"FMT"",__func__ ,##__VA_ARGS__) #define kdebug(FMT,...) dbgprintk(" "FMT ,##__VA_ARGS__) @@ -791,8 +791,8 @@ do { \ } while (0) #else -#define _enter(FMT,...) _dbprintk("==> %s("FMT")",__FUNCTION__ ,##__VA_ARGS__) -#define _leave(FMT,...) _dbprintk("<== %s()"FMT"",__FUNCTION__ ,##__VA_ARGS__) +#define _enter(FMT,...) _dbprintk("==> %s("FMT")",__func__ ,##__VA_ARGS__) +#define _leave(FMT,...) _dbprintk("<== %s()"FMT"",__func__ ,##__VA_ARGS__) #define _debug(FMT,...) _dbprintk(" "FMT ,##__VA_ARGS__) #endif -- cgit v1.2.3 From 8e24eea728068bbeb6a3c500b848f883a20bf225 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Wed, 30 Apr 2008 00:55:09 -0700 Subject: fs: replace remaining __FUNCTION__ occurrences __FUNCTION__ is gcc-specific, use __func__ Signed-off-by: Harvey Harrison Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/adfs/adfs.h | 2 +- fs/autofs4/autofs_i.h | 2 +- fs/bfs/bfs.h | 2 +- fs/buffer.c | 2 +- fs/configfs/file.c | 2 +- fs/configfs/mount.c | 2 +- fs/configfs/symlink.c | 4 ++-- fs/dlm/lockspace.c | 2 +- fs/exportfs/expfs.c | 10 +++++----- fs/fat/cache.c | 6 +++--- fs/fat/fatent.c | 2 +- fs/fat/file.c | 2 +- fs/gfs2/locking/dlm/sysfs.c | 2 +- fs/gfs2/util.h | 18 +++++++++--------- fs/jffs2/debug.h | 8 ++++---- fs/jffs2/xattr.c | 4 ++-- fs/lockd/clntproc.c | 2 +- fs/lockd/svclock.c | 2 +- fs/msdos/namei.c | 2 +- fs/namespace.c | 4 ++-- fs/nfsd/nfs4callback.c | 4 ++-- fs/ntfs/debug.h | 6 +++--- fs/partitions/ldm.c | 8 ++++---- fs/smbfs/smb_debug.h | 6 +++--- fs/sysfs/file.c | 2 +- fs/sysfs/mount.c | 2 +- fs/udf/super.c | 4 ++-- fs/vfat/namei.c | 2 +- 28 files changed, 57 insertions(+), 57 deletions(-) (limited to 'fs') diff --git a/fs/adfs/adfs.h b/fs/adfs/adfs.h index 936f2af39c43..831157502d5a 100644 --- a/fs/adfs/adfs.h +++ b/fs/adfs/adfs.h @@ -75,7 +75,7 @@ extern unsigned int adfs_map_free(struct super_block *sb); /* Misc */ void __adfs_error(struct super_block *sb, const char *function, const char *fmt, ...); -#define adfs_error(sb, fmt...) __adfs_error(sb, __FUNCTION__, fmt) +#define adfs_error(sb, fmt...) __adfs_error(sb, __func__, fmt) /* super.c */ diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index 2d4ae40718d9..c3d352d7fa93 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -35,7 +35,7 @@ /* #define DEBUG */ #ifdef DEBUG -#define DPRINTK(fmt,args...) do { printk(KERN_DEBUG "pid %d: %s: " fmt "\n" , current->pid , __FUNCTION__ , ##args); } while(0) +#define DPRINTK(fmt,args...) do { printk(KERN_DEBUG "pid %d: %s: " fmt "\n" , current->pid , __func__ , ##args); } while(0) #else #define DPRINTK(fmt,args...) do {} while(0) #endif diff --git a/fs/bfs/bfs.h b/fs/bfs/bfs.h index 71faf4d23908..70f5d3a8eede 100644 --- a/fs/bfs/bfs.h +++ b/fs/bfs/bfs.h @@ -42,7 +42,7 @@ static inline struct bfs_inode_info *BFS_I(struct inode *inode) #define printf(format, args...) \ - printk(KERN_ERR "BFS-fs: %s(): " format, __FUNCTION__, ## args) + printk(KERN_ERR "BFS-fs: %s(): " format, __func__, ## args) /* inode.c */ extern struct inode *bfs_iget(struct super_block *sb, unsigned long ino); diff --git a/fs/buffer.c b/fs/buffer.c index 189efa4efc6e..a073f3f4f013 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1101,7 +1101,7 @@ grow_buffers(struct block_device *bdev, sector_t block, int size) printk(KERN_ERR "%s: requested out-of-range block %llu for " "device %s\n", - __FUNCTION__, (unsigned long long)block, + __func__, (unsigned long long)block, bdevname(bdev, b)); return -EIO; } diff --git a/fs/configfs/file.c b/fs/configfs/file.c index 397cb503a180..2b6cb23dd14e 100644 --- a/fs/configfs/file.c +++ b/fs/configfs/file.c @@ -115,7 +115,7 @@ configfs_read_file(struct file *file, char __user *buf, size_t count, loff_t *pp goto out; } pr_debug("%s: count = %zd, ppos = %lld, buf = %s\n", - __FUNCTION__, count, *ppos, buffer->page); + __func__, count, *ppos, buffer->page); retval = simple_read_from_buffer(buf, count, ppos, buffer->page, buffer->count); out: diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c index de3b31d0a37d..8421cea7d8c7 100644 --- a/fs/configfs/mount.c +++ b/fs/configfs/mount.c @@ -92,7 +92,7 @@ static int configfs_fill_super(struct super_block *sb, void *data, int silent) root = d_alloc_root(inode); if (!root) { - pr_debug("%s: could not get root dentry!\n",__FUNCTION__); + pr_debug("%s: could not get root dentry!\n",__func__); iput(inode); return -ENOMEM; } diff --git a/fs/configfs/symlink.c b/fs/configfs/symlink.c index 78929ea84ff2..2a731ef5f305 100644 --- a/fs/configfs/symlink.c +++ b/fs/configfs/symlink.c @@ -210,13 +210,13 @@ static int configfs_get_target_path(struct config_item * item, struct config_ite if (size > PATH_MAX) return -ENAMETOOLONG; - pr_debug("%s: depth = %d, size = %d\n", __FUNCTION__, depth, size); + pr_debug("%s: depth = %d, size = %d\n", __func__, depth, size); for (s = path; depth--; s += 3) strcpy(s,"../"); fill_item_path(target, path, size); - pr_debug("%s: path = '%s'\n", __FUNCTION__, path); + pr_debug("%s: path = '%s'\n", __func__, path); return 0; } diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index b64e55e0515d..499e16759e96 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -200,7 +200,7 @@ int __init dlm_lockspace_init(void) dlm_kset = kset_create_and_add("dlm", NULL, kernel_kobj); if (!dlm_kset) { - printk(KERN_WARNING "%s: can not create kset\n", __FUNCTION__); + printk(KERN_WARNING "%s: can not create kset\n", __func__); return -ENOMEM; } return 0; diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c index 109ab5e44eca..cc91227d3bb8 100644 --- a/fs/exportfs/expfs.c +++ b/fs/exportfs/expfs.c @@ -150,12 +150,12 @@ reconnect_path(struct vfsmount *mnt, struct dentry *target_dir) if (IS_ERR(ppd)) { err = PTR_ERR(ppd); dprintk("%s: get_parent of %ld failed, err %d\n", - __FUNCTION__, pd->d_inode->i_ino, err); + __func__, pd->d_inode->i_ino, err); dput(pd); break; } - dprintk("%s: find name of %lu in %lu\n", __FUNCTION__, + dprintk("%s: find name of %lu in %lu\n", __func__, pd->d_inode->i_ino, ppd->d_inode->i_ino); err = exportfs_get_name(mnt, ppd, nbuf, pd); if (err) { @@ -168,14 +168,14 @@ reconnect_path(struct vfsmount *mnt, struct dentry *target_dir) continue; break; } - dprintk("%s: found name: %s\n", __FUNCTION__, nbuf); + dprintk("%s: found name: %s\n", __func__, nbuf); mutex_lock(&ppd->d_inode->i_mutex); npd = lookup_one_len(nbuf, ppd, strlen(nbuf)); mutex_unlock(&ppd->d_inode->i_mutex); if (IS_ERR(npd)) { err = PTR_ERR(npd); dprintk("%s: lookup failed: %d\n", - __FUNCTION__, err); + __func__, err); dput(ppd); dput(pd); break; @@ -188,7 +188,7 @@ reconnect_path(struct vfsmount *mnt, struct dentry *target_dir) if (npd == pd) noprogress = 0; else - printk("%s: npd != pd\n", __FUNCTION__); + printk("%s: npd != pd\n", __func__); dput(npd); dput(ppd); if (IS_ROOT(pd)) { diff --git a/fs/fat/cache.c b/fs/fat/cache.c index 639b3b4f86d1..fda25479af26 100644 --- a/fs/fat/cache.c +++ b/fs/fat/cache.c @@ -242,7 +242,7 @@ int fat_get_cluster(struct inode *inode, int cluster, int *fclus, int *dclus) /* prevent the infinite loop of cluster chain */ if (*fclus > limit) { fat_fs_panic(sb, "%s: detected the cluster chain loop" - " (i_pos %lld)", __FUNCTION__, + " (i_pos %lld)", __func__, MSDOS_I(inode)->i_pos); nr = -EIO; goto out; @@ -253,7 +253,7 @@ int fat_get_cluster(struct inode *inode, int cluster, int *fclus, int *dclus) goto out; else if (nr == FAT_ENT_FREE) { fat_fs_panic(sb, "%s: invalid cluster chain" - " (i_pos %lld)", __FUNCTION__, + " (i_pos %lld)", __func__, MSDOS_I(inode)->i_pos); nr = -EIO; goto out; @@ -286,7 +286,7 @@ static int fat_bmap_cluster(struct inode *inode, int cluster) return ret; else if (ret == FAT_ENT_EOF) { fat_fs_panic(sb, "%s: request beyond EOF (i_pos %lld)", - __FUNCTION__, MSDOS_I(inode)->i_pos); + __func__, MSDOS_I(inode)->i_pos); return -EIO; } return dclus; diff --git a/fs/fat/fatent.c b/fs/fat/fatent.c index 13ab763cc510..302e95c4af7e 100644 --- a/fs/fat/fatent.c +++ b/fs/fat/fatent.c @@ -546,7 +546,7 @@ int fat_free_clusters(struct inode *inode, int cluster) goto error; } else if (cluster == FAT_ENT_FREE) { fat_fs_panic(sb, "%s: deleting FAT entry beyond EOF", - __FUNCTION__); + __func__); err = -EIO; goto error; } diff --git a/fs/fat/file.c b/fs/fat/file.c index d604bb132422..27cc1164ec36 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -208,7 +208,7 @@ static int fat_free(struct inode *inode, int skip) } else if (ret == FAT_ENT_FREE) { fat_fs_panic(sb, "%s: invalid cluster chain (i_pos %lld)", - __FUNCTION__, MSDOS_I(inode)->i_pos); + __func__, MSDOS_I(inode)->i_pos); ret = -EIO; } else if (ret > 0) { err = fat_ent_write(inode, &fatent, FAT_ENT_EOF, wait); diff --git a/fs/gfs2/locking/dlm/sysfs.c b/fs/gfs2/locking/dlm/sysfs.c index 8479da47049c..a4ff271df9ee 100644 --- a/fs/gfs2/locking/dlm/sysfs.c +++ b/fs/gfs2/locking/dlm/sysfs.c @@ -212,7 +212,7 @@ int gdlm_sysfs_init(void) { gdlm_kset = kset_create_and_add("lock_dlm", NULL, kernel_kobj); if (!gdlm_kset) { - printk(KERN_WARNING "%s: can not create kset\n", __FUNCTION__); + printk(KERN_WARNING "%s: can not create kset\n", __func__); return -ENOMEM; } return 0; diff --git a/fs/gfs2/util.h b/fs/gfs2/util.h index 509c5d60bd80..7f48576289c9 100644 --- a/fs/gfs2/util.h +++ b/fs/gfs2/util.h @@ -41,7 +41,7 @@ int gfs2_assert_withdraw_i(struct gfs2_sbd *sdp, char *assertion, #define gfs2_assert_withdraw(sdp, assertion) \ ((likely(assertion)) ? 0 : gfs2_assert_withdraw_i((sdp), #assertion, \ - __FUNCTION__, __FILE__, __LINE__)) + __func__, __FILE__, __LINE__)) int gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion, @@ -49,28 +49,28 @@ int gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion, #define gfs2_assert_warn(sdp, assertion) \ ((likely(assertion)) ? 0 : gfs2_assert_warn_i((sdp), #assertion, \ - __FUNCTION__, __FILE__, __LINE__)) + __func__, __FILE__, __LINE__)) int gfs2_consist_i(struct gfs2_sbd *sdp, int cluster_wide, const char *function, char *file, unsigned int line); #define gfs2_consist(sdp) \ -gfs2_consist_i((sdp), 0, __FUNCTION__, __FILE__, __LINE__) +gfs2_consist_i((sdp), 0, __func__, __FILE__, __LINE__) int gfs2_consist_inode_i(struct gfs2_inode *ip, int cluster_wide, const char *function, char *file, unsigned int line); #define gfs2_consist_inode(ip) \ -gfs2_consist_inode_i((ip), 0, __FUNCTION__, __FILE__, __LINE__) +gfs2_consist_inode_i((ip), 0, __func__, __FILE__, __LINE__) int gfs2_consist_rgrpd_i(struct gfs2_rgrpd *rgd, int cluster_wide, const char *function, char *file, unsigned int line); #define gfs2_consist_rgrpd(rgd) \ -gfs2_consist_rgrpd_i((rgd), 0, __FUNCTION__, __FILE__, __LINE__) +gfs2_consist_rgrpd_i((rgd), 0, __func__, __FILE__, __LINE__) int gfs2_meta_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh, @@ -91,7 +91,7 @@ static inline int gfs2_meta_check_i(struct gfs2_sbd *sdp, } #define gfs2_meta_check(sdp, bh) \ -gfs2_meta_check_i((sdp), (bh), __FUNCTION__, __FILE__, __LINE__) +gfs2_meta_check_i((sdp), (bh), __func__, __FILE__, __LINE__) int gfs2_metatype_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh, @@ -118,7 +118,7 @@ static inline int gfs2_metatype_check_i(struct gfs2_sbd *sdp, } #define gfs2_metatype_check(sdp, bh, type) \ -gfs2_metatype_check_i((sdp), (bh), (type), __FUNCTION__, __FILE__, __LINE__) +gfs2_metatype_check_i((sdp), (bh), (type), __func__, __FILE__, __LINE__) static inline void gfs2_metatype_set(struct buffer_head *bh, u16 type, u16 format) @@ -134,14 +134,14 @@ int gfs2_io_error_i(struct gfs2_sbd *sdp, const char *function, char *file, unsigned int line); #define gfs2_io_error(sdp) \ -gfs2_io_error_i((sdp), __FUNCTION__, __FILE__, __LINE__); +gfs2_io_error_i((sdp), __func__, __FILE__, __LINE__); int gfs2_io_error_bh_i(struct gfs2_sbd *sdp, struct buffer_head *bh, const char *function, char *file, unsigned int line); #define gfs2_io_error_bh(sdp, bh) \ -gfs2_io_error_bh_i((sdp), (bh), __FUNCTION__, __FILE__, __LINE__); +gfs2_io_error_bh_i((sdp), (bh), __func__, __FILE__, __LINE__); extern struct kmem_cache *gfs2_glock_cachep; diff --git a/fs/jffs2/debug.h b/fs/jffs2/debug.h index 9645275023e6..a113ecc3bafe 100644 --- a/fs/jffs2/debug.h +++ b/fs/jffs2/debug.h @@ -82,28 +82,28 @@ do { \ printk(JFFS2_ERR_MSG_PREFIX \ " (%d) %s: " fmt, task_pid_nr(current), \ - __FUNCTION__ , ##__VA_ARGS__); \ + __func__ , ##__VA_ARGS__); \ } while(0) #define JFFS2_WARNING(fmt, ...) \ do { \ printk(JFFS2_WARN_MSG_PREFIX \ " (%d) %s: " fmt, task_pid_nr(current), \ - __FUNCTION__ , ##__VA_ARGS__); \ + __func__ , ##__VA_ARGS__); \ } while(0) #define JFFS2_NOTICE(fmt, ...) \ do { \ printk(JFFS2_NOTICE_MSG_PREFIX \ " (%d) %s: " fmt, task_pid_nr(current), \ - __FUNCTION__ , ##__VA_ARGS__); \ + __func__ , ##__VA_ARGS__); \ } while(0) #define JFFS2_DEBUG(fmt, ...) \ do { \ printk(JFFS2_DBG_MSG_PREFIX \ " (%d) %s: " fmt, task_pid_nr(current), \ - __FUNCTION__ , ##__VA_ARGS__); \ + __func__ , ##__VA_ARGS__); \ } while(0) /* diff --git a/fs/jffs2/xattr.c b/fs/jffs2/xattr.c index e48665984cb3..574cb7532d6c 100644 --- a/fs/jffs2/xattr.c +++ b/fs/jffs2/xattr.c @@ -82,7 +82,7 @@ static int is_xattr_datum_unchecked(struct jffs2_sb_info *c, struct jffs2_xattr_ static void unload_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd) { /* must be called under down_write(xattr_sem) */ - D1(dbg_xattr("%s: xid=%u, version=%u\n", __FUNCTION__, xd->xid, xd->version)); + D1(dbg_xattr("%s: xid=%u, version=%u\n", __func__, xd->xid, xd->version)); if (xd->xname) { c->xdatum_mem_usage -= (xd->name_len + 1 + xd->value_len); kfree(xd->xname); @@ -1252,7 +1252,7 @@ int jffs2_garbage_collect_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ rc = jffs2_reserve_space_gc(c, totlen, &length, JFFS2_SUMMARY_XREF_SIZE); if (rc) { JFFS2_WARNING("%s: jffs2_reserve_space_gc() = %d, request = %u\n", - __FUNCTION__, rc, totlen); + __func__, rc, totlen); rc = rc ? rc : -EBADFD; goto out; } diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index 40b16f23e49a..5df517b81f3f 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -573,7 +573,7 @@ again: /* Ensure the resulting lock will get added to granted list */ fl->fl_flags |= FL_SLEEP; if (do_vfs_lock(fl) < 0) - printk(KERN_WARNING "%s: VFS is out of sync with lock manager!\n", __FUNCTION__); + printk(KERN_WARNING "%s: VFS is out of sync with lock manager!\n", __func__); up_read(&host->h_rwsem); fl->fl_flags = fl_flags; status = 0; diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index 4d81553d2948..81aca859bfde 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -752,7 +752,7 @@ nlmsvc_grant_blocked(struct nlm_block *block) return; default: printk(KERN_WARNING "lockd: unexpected error %d in %s!\n", - -error, __FUNCTION__); + -error, __func__); nlmsvc_insert_block(block, 10 * HZ); nlmsvc_release_block(block); return; diff --git a/fs/msdos/namei.c b/fs/msdos/namei.c index 2d4358c59f68..05ff4f1d7026 100644 --- a/fs/msdos/namei.c +++ b/fs/msdos/namei.c @@ -609,7 +609,7 @@ error_inode: if (corrupt < 0) { fat_fs_panic(new_dir->i_sb, "%s: Filesystem corrupted (i_pos %lld)", - __FUNCTION__, sinfo.i_pos); + __func__, sinfo.i_pos); } goto out; } diff --git a/fs/namespace.c b/fs/namespace.c index 061e5edb4d27..4fc302c2a0e0 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -2329,10 +2329,10 @@ void __init mnt_init(void) err = sysfs_init(); if (err) printk(KERN_WARNING "%s: sysfs_init error: %d\n", - __FUNCTION__, err); + __func__, err); fs_kobj = kobject_create_and_add("fs", NULL); if (!fs_kobj) - printk(KERN_WARNING "%s: kobj create error\n", __FUNCTION__); + printk(KERN_WARNING "%s: kobj create error\n", __func__); init_rootfs(); init_mount_tree(); } diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 562abf3380d0..0b3ffa9840c2 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -104,7 +104,7 @@ xdr_writemem(__be32 *p, const void *ptr, int nbytes) } while (0) #define RESERVE_SPACE(nbytes) do { \ p = xdr_reserve_space(xdr, nbytes); \ - if (!p) dprintk("NFSD: RESERVE_SPACE(%d) failed in function %s\n", (int) (nbytes), __FUNCTION__); \ + if (!p) dprintk("NFSD: RESERVE_SPACE(%d) failed in function %s\n", (int) (nbytes), __func__); \ BUG_ON(!p); \ } while (0) @@ -134,7 +134,7 @@ xdr_error: \ p = xdr_inline_decode(xdr, nbytes); \ if (!p) { \ dprintk("NFSD: %s: reply buffer overflowed in line %d.\n", \ - __FUNCTION__, __LINE__); \ + __func__, __LINE__); \ return -EIO; \ } \ } while (0) diff --git a/fs/ntfs/debug.h b/fs/ntfs/debug.h index 8ac37c33d127..5e6724c1afd1 100644 --- a/fs/ntfs/debug.h +++ b/fs/ntfs/debug.h @@ -45,7 +45,7 @@ static void ntfs_debug(const char *f, ...); extern void __ntfs_debug (const char *file, int line, const char *function, const char *format, ...) __attribute__ ((format (printf, 4, 5))); #define ntfs_debug(f, a...) \ - __ntfs_debug(__FILE__, __LINE__, __FUNCTION__, f, ##a) + __ntfs_debug(__FILE__, __LINE__, __func__, f, ##a) extern void ntfs_debug_dump_runlist(const runlist_element *rl); @@ -58,10 +58,10 @@ extern void ntfs_debug_dump_runlist(const runlist_element *rl); extern void __ntfs_warning(const char *function, const struct super_block *sb, const char *fmt, ...) __attribute__ ((format (printf, 3, 4))); -#define ntfs_warning(sb, f, a...) __ntfs_warning(__FUNCTION__, sb, f, ##a) +#define ntfs_warning(sb, f, a...) __ntfs_warning(__func__, sb, f, ##a) extern void __ntfs_error(const char *function, const struct super_block *sb, const char *fmt, ...) __attribute__ ((format (printf, 3, 4))); -#define ntfs_error(sb, f, a...) __ntfs_error(__FUNCTION__, sb, f, ##a) +#define ntfs_error(sb, f, a...) __ntfs_error(__func__, sb, f, ##a) #endif /* _LINUX_NTFS_DEBUG_H */ diff --git a/fs/partitions/ldm.c b/fs/partitions/ldm.c index e7dd1d4e3473..0fdda2e8a4cc 100644 --- a/fs/partitions/ldm.c +++ b/fs/partitions/ldm.c @@ -41,12 +41,12 @@ #ifndef CONFIG_LDM_DEBUG #define ldm_debug(...) do {} while (0) #else -#define ldm_debug(f, a...) _ldm_printk (KERN_DEBUG, __FUNCTION__, f, ##a) +#define ldm_debug(f, a...) _ldm_printk (KERN_DEBUG, __func__, f, ##a) #endif -#define ldm_crit(f, a...) _ldm_printk (KERN_CRIT, __FUNCTION__, f, ##a) -#define ldm_error(f, a...) _ldm_printk (KERN_ERR, __FUNCTION__, f, ##a) -#define ldm_info(f, a...) _ldm_printk (KERN_INFO, __FUNCTION__, f, ##a) +#define ldm_crit(f, a...) _ldm_printk (KERN_CRIT, __func__, f, ##a) +#define ldm_error(f, a...) _ldm_printk (KERN_ERR, __func__, f, ##a) +#define ldm_info(f, a...) _ldm_printk (KERN_INFO, __func__, f, ##a) __attribute__ ((format (printf, 3, 4))) static void _ldm_printk (const char *level, const char *function, diff --git a/fs/smbfs/smb_debug.h b/fs/smbfs/smb_debug.h index 734972b92694..fc4b1a5dd755 100644 --- a/fs/smbfs/smb_debug.h +++ b/fs/smbfs/smb_debug.h @@ -11,14 +11,14 @@ * these are normally enabled. */ #ifdef SMBFS_PARANOIA -# define PARANOIA(f, a...) printk(KERN_NOTICE "%s: " f, __FUNCTION__ , ## a) +# define PARANOIA(f, a...) printk(KERN_NOTICE "%s: " f, __func__ , ## a) #else # define PARANOIA(f, a...) do { ; } while(0) #endif /* lots of debug messages */ #ifdef SMBFS_DEBUG_VERBOSE -# define VERBOSE(f, a...) printk(KERN_DEBUG "%s: " f, __FUNCTION__ , ## a) +# define VERBOSE(f, a...) printk(KERN_DEBUG "%s: " f, __func__ , ## a) #else # define VERBOSE(f, a...) do { ; } while(0) #endif @@ -28,7 +28,7 @@ * too common name. */ #ifdef SMBFS_DEBUG -#define DEBUG1(f, a...) printk(KERN_DEBUG "%s: " f, __FUNCTION__ , ## a) +#define DEBUG1(f, a...) printk(KERN_DEBUG "%s: " f, __func__ , ## a) #else #define DEBUG1(f, a...) do { ; } while(0) #endif diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index dbdfabbfd609..e7735f643cd1 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -135,7 +135,7 @@ sysfs_read_file(struct file *file, char __user *buf, size_t count, loff_t *ppos) goto out; } pr_debug("%s: count = %zd, ppos = %lld, buf = %s\n", - __FUNCTION__, count, *ppos, buffer->page); + __func__, count, *ppos, buffer->page); retval = simple_read_from_buffer(buf, count, ppos, buffer->page, buffer->count); out: diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index 74168266cd59..14f0023984d7 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -61,7 +61,7 @@ static int sysfs_fill_super(struct super_block *sb, void *data, int silent) /* instantiate and link root dentry */ root = d_alloc_root(inode); if (!root) { - pr_debug("%s: could not get root dentry!\n",__FUNCTION__); + pr_debug("%s: could not get root dentry!\n",__func__); iput(inode); return -ENOMEM; } diff --git a/fs/udf/super.c b/fs/udf/super.c index b564fc140fe4..9fb18a340fc1 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -240,7 +240,7 @@ static int udf_sb_alloc_partition_maps(struct super_block *sb, u32 count) sbi->s_partmaps = kcalloc(count, sizeof(struct udf_part_map), GFP_KERNEL); if (!sbi->s_partmaps) { - udf_error(sb, __FUNCTION__, + udf_error(sb, __func__, "Unable to allocate space for %d partition maps", count); sbi->s_partitions = 0; @@ -1086,7 +1086,7 @@ static struct udf_bitmap *udf_sb_alloc_bitmap(struct super_block *sb, u32 index) bitmap = vmalloc(size); /* TODO: get rid of vmalloc */ if (bitmap == NULL) { - udf_error(sb, __FUNCTION__, + udf_error(sb, __func__, "Unable to allocate space for bitmap " "and %d buffer_head pointers", nr_groups); return NULL; diff --git a/fs/vfat/namei.c b/fs/vfat/namei.c index 5b66162d0747..a3522727ea5b 100644 --- a/fs/vfat/namei.c +++ b/fs/vfat/namei.c @@ -986,7 +986,7 @@ error_inode: if (corrupt < 0) { fat_fs_panic(new_dir->i_sb, "%s: Filesystem corrupted (i_pos %lld)", - __FUNCTION__, sinfo.i_pos); + __func__, sinfo.i_pos); } goto out; } -- cgit v1.2.3 From 40a2159abf3d0107bba359246554bd7d56f2171b Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Mon, 28 Apr 2008 15:59:58 +0100 Subject: sysfs: Disallow truncation of files in sysfs sysfs allows attribute files to be truncated, e.g. using ftruncate(), with the expected effect on their inode. For most attributes, this doesn't change the "real" size of the file i.e. how much can be read from it. However, the parameter validation for reading and writing binary attribute files is based on the inode size and not the size specified in the file's bin_attribute, so it can be broken by this. For example, if we try using dd to write to such a file: # pwd /sys/bus/pci/devices/0000:08:00.0 # ls -l config -rw-r--r-- 1 root root 4096 Feb 1 17:35 config # dd if=/dev/zero of=config bs=4 count=1 1+0 records in 1+0 records out # ls -l config -rw-r--r-- 1 root root 0 Feb 1 17:50 config # dd if=/dev/zero of=config bs=4 count=1 seek=128 dd: writing `config': No space left on device 1+0 records in 0+0 records out Also, after truncation to 0, parameter validation for read and write is disabled. Most bin_attribute read and write methods also validate the size and offset, but for some this will allow out-of-range access. This may be a security issue, though access to such files is often limited to root. In any case, the validation should remain for safety's sake!) This was previously reported in Bugzilla as bug 9867. sysfs should ignore size changes or else refuse them (by returning -EINVAL). This patch makes it ignore them. Signed-off-by: Ben Hutchings Signed-off-by: Greg Kroah-Hartman --- fs/sysfs/inode.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index f8b82e73b3bf..eb53c632f856 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -59,6 +59,8 @@ int sysfs_setattr(struct dentry * dentry, struct iattr * iattr) if (error) return error; + iattr->ia_valid &= ~ATTR_SIZE; /* ignore size changes */ + error = inode_setattr(inode, iattr); if (error) return error; -- cgit v1.2.3 From 883ce42ec45c2dbef5be7c133ade9741ac978329 Mon Sep 17 00:00:00 2001 From: "Robert P. J. Day" Date: Fri, 25 Apr 2008 08:52:51 -0400 Subject: DEBUGFS: Correct location of debugfs API documentation. Signed-off-by: Robert P. J. Day Signed-off-by: Greg Kroah-Hartman --- fs/debugfs/file.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index fddffe4851f5..159a5efd6a8a 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -9,7 +9,7 @@ * 2 as published by the Free Software Foundation. * * debugfs is for people to use instead of /proc or /sys. - * See Documentation/DocBook/kernel-api for more details. + * See Documentation/DocBook/filesystems for more details. * */ -- cgit v1.2.3 From 9d80f7539a91c0154e40fc9e4ae5e818dd8f102e Mon Sep 17 00:00:00 2001 From: Joel Becker Date: Tue, 22 Apr 2008 11:46:44 -0700 Subject: ocfs2: Correct merge of 52f7c21 (Move /sys/o2cb to /sys/fs/o2cb) Commit 52f7c21b613f80cb425d115c9e5b4ed958a133c0 was intended to move /sys/o2cb to /sys/fs/o2cb, providing /sys/o2cb as a symlink for backwards compatibility. However, the merge apparently added the symlink but failed to move the directory, resulting in a duplicate filename error. It's a one-line change that was missing. Signed-off-by: Joel Becker Acked-by: Randy Dunlap Signed-off-by: Mark Fasheh --- fs/ocfs2/cluster/sys.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ocfs2/cluster/sys.c b/fs/ocfs2/cluster/sys.c index 98429fd68499..bc702dab5d1f 100644 --- a/fs/ocfs2/cluster/sys.c +++ b/fs/ocfs2/cluster/sys.c @@ -65,7 +65,7 @@ int o2cb_sys_init(void) { int ret; - o2cb_kset = kset_create_and_add("o2cb", NULL, NULL); + o2cb_kset = kset_create_and_add("o2cb", NULL, fs_kobj); if (!o2cb_kset) return -ENOMEM; -- cgit v1.2.3 From 4d8755b5e667df8f01647773ba744a5ac97e68e6 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 21 Apr 2008 11:49:26 +0300 Subject: ocfs2: make struct ocfs2_control_device static This patch makes the needlessly global struct ocfs2_control_device static. Signed-off-by: Adrian Bunk Acked-by: Joel Becker Signed-off-by: Mark Fasheh --- fs/ocfs2/stack_user.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ocfs2/stack_user.c b/fs/ocfs2/stack_user.c index 7428663f9cbb..b503772cd0ec 100644 --- a/fs/ocfs2/stack_user.c +++ b/fs/ocfs2/stack_user.c @@ -635,7 +635,7 @@ static const struct file_operations ocfs2_control_fops = { .owner = THIS_MODULE, }; -struct miscdevice ocfs2_control_device = { +static struct miscdevice ocfs2_control_device = { .minor = MISC_DYNAMIC_MINOR, .name = "ocfs2_control", .fops = &ocfs2_control_fops, -- cgit v1.2.3 From 4af694e672aaa85940d6e29d27b7eeea5f6eb258 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 21 Apr 2008 11:49:31 +0300 Subject: ocfs2: make struct o2cb_stack_ops static This patch makes the needlessly global struct o2cb_stack_ops static. Signed-off-by: Adrian Bunk Acked-by: Joel Becker Signed-off-by: Mark Fasheh --- fs/ocfs2/stack_o2cb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ocfs2/stack_o2cb.c b/fs/ocfs2/stack_o2cb.c index ac1d74c63bf5..bbd1667aa7d3 100644 --- a/fs/ocfs2/stack_o2cb.c +++ b/fs/ocfs2/stack_o2cb.c @@ -385,7 +385,7 @@ static int o2cb_cluster_this_node(unsigned int *node) return 0; } -struct ocfs2_stack_operations o2cb_stack_ops = { +static struct ocfs2_stack_operations o2cb_stack_ops = { .connect = o2cb_cluster_connect, .disconnect = o2cb_cluster_disconnect, .hangup = o2cb_cluster_hangup, -- cgit v1.2.3 From 95642e56647d84963428a1168baa8a73cb782ac3 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 21 Apr 2008 11:49:37 +0300 Subject: ocfs2/dlm: dlmdebug.c: make 2 functions static This patch makes the following needlessly global functions static: - stringify_lockname() - dlm_debug_put() Signed-off-by: Adrian Bunk Acked-by: Sunil Mushran Signed-off-by: Mark Fasheh --- fs/ocfs2/dlm/dlmdebug.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/ocfs2/dlm/dlmdebug.c b/fs/ocfs2/dlm/dlmdebug.c index 5f6d858770a2..1b81dcba175d 100644 --- a/fs/ocfs2/dlm/dlmdebug.c +++ b/fs/ocfs2/dlm/dlmdebug.c @@ -44,7 +44,8 @@ #define MLOG_MASK_PREFIX ML_DLM #include "cluster/masklog.h" -int stringify_lockname(const char *lockname, int locklen, char *buf, int len); +static int stringify_lockname(const char *lockname, int locklen, char *buf, + int len); void dlm_print_one_lock_resource(struct dlm_lock_resource *res) { @@ -251,7 +252,8 @@ EXPORT_SYMBOL_GPL(dlm_errname); * * For more on lockname formats, please refer to dlmglue.c and ocfs2_lockid.h. */ -int stringify_lockname(const char *lockname, int locklen, char *buf, int len) +static int stringify_lockname(const char *lockname, int locklen, char *buf, + int len) { int out = 0; __be64 inode_blkno_be; @@ -368,7 +370,7 @@ static void dlm_debug_free(struct kref *kref) kfree(dc); } -void dlm_debug_put(struct dlm_debug_ctxt *dc) +static void dlm_debug_put(struct dlm_debug_ctxt *dc) { if (dc) kref_put(&dc->debug_refcnt, dlm_debug_free); -- cgit v1.2.3 From bc535809c06ada210d89f5a43b335c68ecbb8e1b Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Fri, 18 Apr 2008 10:23:53 -0700 Subject: ocfs2: Allow uid/gid/perm changes of symlinks This patch adds the ability to change attributes of a symlink. Fixes oss bugzilla#963 http://oss.oracle.com/bugzilla/show_bug.cgi?id=963 Signed-off-by: Sunil Mushran Signed-off-by: Mark Fasheh --- fs/ocfs2/file.c | 4 ++++ fs/ocfs2/symlink.c | 2 ++ 2 files changed, 6 insertions(+) (limited to 'fs') diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 9154c82d3258..57e0d30cde98 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1048,6 +1048,10 @@ int ocfs2_setattr(struct dentry *dentry, struct iattr *attr) mlog_entry("(0x%p, '%.*s')\n", dentry, dentry->d_name.len, dentry->d_name.name); + /* ensuring we don't even attempt to truncate a symlink */ + if (S_ISLNK(inode->i_mode)) + attr->ia_valid &= ~ATTR_SIZE; + if (attr->ia_valid & ATTR_MODE) mlog(0, "mode change: %d\n", attr->ia_mode); if (attr->ia_valid & ATTR_UID) diff --git a/fs/ocfs2/symlink.c b/fs/ocfs2/symlink.c index 7134007ba22f..ba9dbb51d25b 100644 --- a/fs/ocfs2/symlink.c +++ b/fs/ocfs2/symlink.c @@ -167,9 +167,11 @@ const struct inode_operations ocfs2_symlink_inode_operations = { .readlink = page_readlink, .follow_link = ocfs2_follow_link, .getattr = ocfs2_getattr, + .setattr = ocfs2_setattr, }; const struct inode_operations ocfs2_fast_symlink_inode_operations = { .readlink = ocfs2_readlink, .follow_link = ocfs2_follow_link, .getattr = ocfs2_getattr, + .setattr = ocfs2_setattr, }; -- cgit v1.2.3 From 4ba1c5bfd2e5a6c9528eb7777b66c297e70f61ca Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Fri, 18 Apr 2008 15:03:59 -0700 Subject: ocfs2: Use GFP_NOFS in kmalloc during localalloc window move kmalloc() during a localalloc window move can trigger the mm to prune the dcache which inturn can trigger the fs to delete an inode causing it start a recursive transaction. The fix also makes the change in kmalloc during localalloc shutdown just to be safe. Fixes oss bugzilla#901 http://oss.oracle.com/bugzilla/show_bug.cgi?id=901 Signed-off-by: Sunil Mushran Signed-off-by: Mark Fasheh --- fs/ocfs2/localalloc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index ce0dc147602a..be774bdc8b36 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -260,7 +260,7 @@ void ocfs2_shutdown_local_alloc(struct ocfs2_super *osb) bh = osb->local_alloc_bh; alloc = (struct ocfs2_dinode *) bh->b_data; - alloc_copy = kmalloc(bh->b_size, GFP_KERNEL); + alloc_copy = kmalloc(bh->b_size, GFP_NOFS); if (!alloc_copy) { status = -ENOMEM; goto out_commit; @@ -931,7 +931,7 @@ static int ocfs2_local_alloc_slide_window(struct ocfs2_super *osb, * local alloc shutdown won't try to double free main bitmap * bits. Make a copy so the sync function knows which bits to * free. */ - alloc_copy = kmalloc(osb->local_alloc_bh->b_size, GFP_KERNEL); + alloc_copy = kmalloc(osb->local_alloc_bh->b_size, GFP_NOFS); if (!alloc_copy) { status = -ENOMEM; mlog_errno(status); -- cgit v1.2.3 From 214b7049a7929f03bbd2786aaef04b8b79db34e2 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 1 May 2008 03:52:22 +0100 Subject: Fix dnotify/close race We have a race between fcntl() and close() that can lead to dnotify_struct inserted into inode's list *after* the last descriptor had been gone from current->files. Since that's the only point where dnotify_struct gets evicted, we are screwed - it will stick around indefinitely. Even after struct file in question is gone and freed. Worse, we can trigger send_sigio() on it at any later point, which allows to send an arbitrary signal to arbitrary process if we manage to apply enough memory pressure to get the page that used to host that struct file and fill it with the right pattern... Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- fs/dnotify.c | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'fs') diff --git a/fs/dnotify.c b/fs/dnotify.c index 28d01ed66de0..eaecc4cfe540 100644 --- a/fs/dnotify.c +++ b/fs/dnotify.c @@ -20,6 +20,7 @@ #include #include #include +#include int dir_notify_enable __read_mostly = 1; @@ -66,6 +67,7 @@ int fcntl_dirnotify(int fd, struct file *filp, unsigned long arg) struct dnotify_struct **prev; struct inode *inode; fl_owner_t id = current->files; + struct file *f; int error = 0; if ((arg & ~DN_MULTISHOT) == 0) { @@ -92,6 +94,15 @@ int fcntl_dirnotify(int fd, struct file *filp, unsigned long arg) prev = &odn->dn_next; } + rcu_read_lock(); + f = fcheck(fd); + rcu_read_unlock(); + /* we'd lost the race with close(), sod off silently */ + /* note that inode->i_lock prevents reordering problems + * between accesses to descriptor table and ->i_dnotify */ + if (f != filp) + goto out_free; + error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0); if (error) goto out_free; -- cgit v1.2.3 From 4e571aba7bb25a3a069a7b88c0f63fe5a14c05c6 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Thu, 1 May 2008 12:28:04 +0100 Subject: [JFFS2] Clean up jffs2_alloc_inode() and jffs2_i_init_once() Ditch a couple of pointless casts from void *, and use the normal variable name 'f' for jffs2_inode_info pointers -- especially since it actually shows up in lockdep reports. Signed-off-by: David Woodhouse --- fs/jffs2/super.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index f3353df178e7..7da69eae49e4 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -31,11 +31,12 @@ static struct kmem_cache *jffs2_inode_cachep; static struct inode *jffs2_alloc_inode(struct super_block *sb) { - struct jffs2_inode_info *ei; - ei = (struct jffs2_inode_info *)kmem_cache_alloc(jffs2_inode_cachep, GFP_KERNEL); - if (!ei) + struct jffs2_inode_info *f; + + f = kmem_cache_alloc(jffs2_inode_cachep, GFP_KERNEL); + if (!f) return NULL; - return &ei->vfs_inode; + return &f->vfs_inode; } static void jffs2_destroy_inode(struct inode *inode) @@ -45,10 +46,10 @@ static void jffs2_destroy_inode(struct inode *inode) static void jffs2_i_init_once(struct kmem_cache *cachep, void *foo) { - struct jffs2_inode_info *ei = (struct jffs2_inode_info *) foo; + struct jffs2_inode_info *f = foo; - mutex_init(&ei->sem); - inode_init_once(&ei->vfs_inode); + mutex_init(&f->sem); + inode_init_once(&f->vfs_inode); } static int jffs2_sync_fs(struct super_block *sb, int wait) -- cgit v1.2.3 From 590fe34c47cb5c2d836ac76fabc5f160bf31a3f1 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Thu, 1 May 2008 15:53:28 +0100 Subject: [JFFS2] Quiet lockdep false positive. Don't hold f->sem while calling into jffs2_do_create(). It makes lockdep unhappy, and we don't really need it -- the _reason_ it's a false positive is because nobody else can see this inode yet and so nobody will be trying to lock it anyway. Signed-off-by: David Woodhouse --- fs/jffs2/dir.c | 7 +++++++ fs/jffs2/write.c | 6 +++--- 2 files changed, 10 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index c63e7a96af0d..2bba3d3435be 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -208,6 +208,13 @@ static int jffs2_create(struct inode *dir_i, struct dentry *dentry, int mode, f = JFFS2_INODE_INFO(inode); dir_f = JFFS2_INODE_INFO(dir_i); + /* jffs2_do_create() will want to lock it, _after_ reserving + space and taking c-alloc_sem. If we keep it locked here, + lockdep gets unhappy (although it's a false positive; + nothing else will be looking at this inode yet so there's + no chance of AB-BA deadlock involving its f->sem). */ + mutex_unlock(&f->sem); + ret = jffs2_do_create(c, dir_f, f, ri, dentry->d_name.name, dentry->d_name.len); if (ret) diff --git a/fs/jffs2/write.c b/fs/jffs2/write.c index 665fce9797d3..87891bdd7915 100644 --- a/fs/jffs2/write.c +++ b/fs/jffs2/write.c @@ -438,10 +438,10 @@ int jffs2_do_create(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, str ret = jffs2_reserve_space(c, sizeof(*ri), &alloclen, ALLOC_NORMAL, JFFS2_SUMMARY_INODE_SIZE); D1(printk(KERN_DEBUG "jffs2_do_create(): reserved 0x%x bytes\n", alloclen)); - if (ret) { - mutex_unlock(&f->sem); + if (ret) return ret; - } + + mutex_lock(&f->sem); ri->data_crc = cpu_to_je32(0); ri->node_crc = cpu_to_je32(crc32(0, ri, sizeof(*ri)-8)); -- cgit v1.2.3 From 02c6be615f1fcd37ac5ed93a3ad6692ad8991cd9 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Thu, 1 May 2008 04:34:45 -0700 Subject: vfs: fix permission checking in sys_utimensat If utimensat() is called with both times set to UTIME_NOW or one of them to UTIME_NOW and the other to UTIME_OMIT, then it will update the file time without any permission checking. I don't think this can be used for anything other than a local DoS, but could be quite bewildering at that (e.g. "Why was that large source tree rebuilt when I didn't modify anything???") This affects all kernels from 2.6.22, when the utimensat() syscall was introduced. Fix by doing the same permission checking as for the "times == NULL" case. Thanks to Michael Kerrisk, whose utimensat-non-conformances-and-fixes.patch in -mm also fixes this (and breaks other stuff), only he didn't realize the security implications of this bug. Signed-off-by: Miklos Szeredi Cc: Ulrich Drepper Cc: Michael Kerrisk Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/utimes.c | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/utimes.c b/fs/utimes.c index a2bef77dc9c9..af059d5cb485 100644 --- a/fs/utimes.c +++ b/fs/utimes.c @@ -40,9 +40,14 @@ asmlinkage long sys_utime(char __user *filename, struct utimbuf __user *times) #endif +static bool nsec_special(long nsec) +{ + return nsec == UTIME_OMIT || nsec == UTIME_NOW; +} + static bool nsec_valid(long nsec) { - if (nsec == UTIME_OMIT || nsec == UTIME_NOW) + if (nsec_special(nsec)) return true; return nsec >= 0 && nsec <= 999999999; @@ -119,7 +124,15 @@ long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags newattrs.ia_mtime.tv_nsec = times[1].tv_nsec; newattrs.ia_valid |= ATTR_MTIME_SET; } - } else { + } + + /* + * If times is NULL or both times are either UTIME_OMIT or + * UTIME_NOW, then need to check permissions, because + * inode_change_ok() won't do it. + */ + if (!times || (nsec_special(times[0].tv_nsec) && + nsec_special(times[1].tv_nsec))) { error = -EACCES; if (IS_IMMUTABLE(inode)) goto mnt_drop_write_and_out; -- cgit v1.2.3 From afec570c32a0d116e3c68af583ed1d11110f12fc Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Thu, 1 May 2008 04:35:06 -0700 Subject: autofs4: fix sparse warning in waitq.c:autofs4_expire_indirect() Re-order some code in expire.c:autofs4_expire_indirect() to avoid compile warning, reported by Harvey Harrison: CHECK fs/autofs4/expire.c fs/autofs4/expire.c:383:2: warning: context imbalance in 'autofs4_expire_indirect' - unexpected unlock Signed-off-by: Ian Kent Reviewed-by: Harvey Harrison Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/expire.c | 24 +++++++++++------------- 1 file changed, 11 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c index d96e5c14a9ca..cfa12db8f41b 100644 --- a/fs/autofs4/expire.c +++ b/fs/autofs4/expire.c @@ -333,7 +333,7 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb, /* Can we expire this guy */ if (autofs4_can_expire(dentry, timeout, do_now)) { expired = dentry; - break; + goto found; } goto next; } @@ -352,7 +352,7 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb, inf->flags |= AUTOFS_INF_EXPIRING; spin_unlock(&sbi->fs_lock); expired = dentry; - break; + goto found; } spin_unlock(&sbi->fs_lock); /* @@ -363,7 +363,7 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb, expired = autofs4_check_leaves(mnt, dentry, timeout, do_now); if (expired) { dput(dentry); - break; + goto found; } } next: @@ -371,18 +371,16 @@ next: spin_lock(&dcache_lock); next = next->next; } - - if (expired) { - DPRINTK("returning %p %.*s", - expired, (int)expired->d_name.len, expired->d_name.name); - spin_lock(&dcache_lock); - list_move(&expired->d_parent->d_subdirs, &expired->d_u.d_child); - spin_unlock(&dcache_lock); - return expired; - } spin_unlock(&dcache_lock); - return NULL; + +found: + DPRINTK("returning %p %.*s", + expired, (int)expired->d_name.len, expired->d_name.name); + spin_lock(&dcache_lock); + list_move(&expired->d_parent->d_subdirs, &expired->d_u.d_child); + spin_unlock(&dcache_lock); + return expired; } /* Perform an expiry operation */ -- cgit v1.2.3 From cab0936aac8aa907c6bb814c2cf26385478f254b Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Thu, 1 May 2008 04:35:07 -0700 Subject: autofs4: check for invalid dentry in getpath Catch invalid dentry when calculating its path. Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/waitq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index 1fe28e4754c2..75e5955c3f6d 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -171,7 +171,7 @@ static int autofs4_getpath(struct autofs_sb_info *sbi, for (tmp = dentry ; tmp != root ; tmp = tmp->d_parent) len += tmp->d_name.len + 1; - if (--len > NAME_MAX) { + if (!len || --len > NAME_MAX) { spin_unlock(&dcache_lock); return 0; } -- cgit v1.2.3 From 033790449ba9c4dcf8478a87693d33df625c23b5 Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Thu, 1 May 2008 04:35:08 -0700 Subject: autofs4: fix execution order race in mount request code Jeff Moyer has identified a race in due to an execution order dependency in the autofs4 function root.c:try_to_fill_dentry(). Jeff's description of this race is: "P1 does a lookup of /mount/submount/foo. Since the VFS can't find an entry for "foo" under /mount/submount, it calls into the autofs4 kernel module to allocate a new dentry, D1. The kernel creates a new waitq for this lookup and calls the daemon to perform the mount. The daemon performs a mkdir of the "foo" directory under /mount/submount, which ends up creating a *new* dentry, D2. Then, P2 does a lookup of /mount/submount/foo. The VFS path walking logic finds a dentry in the dcache, D2, and calls the revalidate function with this. In the autofs4 revalidate code, we then trigger a mount, since the dentry is an empty directory that isn't a mountpoint, and so set DCACHE_AUTOFS_PENDING and call into the wait code to trigger the mount. The wait code finds our existing waitq entry (since it is keyed off of the directory name) and adds itself to the list of waiters. After the daemon finishes the mount, it calls back into the kernel to release the waiters. When this happens, P1 is woken up and goes about clearing the DCACHE_AUTOFS_PENDING flag, but it does this in D1! So, given that P1 in our case is a program that will immediately try to access a file under /mount/submount/foo, we end up finding the dentry D2 which still has the pending flag set, and we set out to wait for a mount *again*! So, one way to address this is to re-do the lookup at the end of try_to_fill_dentry, and to clear the pending flag on the hashed dentry. This seems a sane approach to me." And Jeff's patch does this. Signed-off-by: Jeff Moyer Signed-off-by-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/root.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'fs') diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index aa4c5ff8a40d..0533d37c73ae 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -242,6 +242,7 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags) { struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb); struct autofs_info *ino = autofs4_dentry_ino(dentry); + struct dentry *new; int status = 0; /* Block on any pending expiry here; invalidate the dentry @@ -318,6 +319,27 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags) spin_lock(&dentry->d_lock); dentry->d_flags &= ~DCACHE_AUTOFS_PENDING; spin_unlock(&dentry->d_lock); + + /* + * The dentry that is passed in from lookup may not be the one + * we end up using, as mkdir can create a new one. If this + * happens, and another process tries the lookup at the same time, + * it will set the PENDING flag on this new dentry, but add itself + * to our waitq. Then, if after the lookup succeeds, the first + * process that requested the mount performs another lookup of the + * same directory, it will show up as still pending! So, we need + * to redo the lookup here and clear pending on that dentry. + */ + if (d_unhashed(dentry)) { + new = d_lookup(dentry->d_parent, &dentry->d_name); + if (new) { + spin_lock(&new->d_lock); + new->d_flags &= ~DCACHE_AUTOFS_PENDING; + spin_unlock(&new->d_lock); + dput(new); + } + } + return status; } -- cgit v1.2.3 From 9d2de6ad2a78bb8b60bf7a54e6043dca44e9a801 Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Thu, 1 May 2008 04:35:09 -0700 Subject: autofs4: fix incorrect return from root.c:try_to_fill_dentry() Jeff Moyer has identified a case where the autofs4 function root.c:try_to_fill_dentry() can return -EBUSY when it should return 0. Jeff's description of the way this happens is: "automount starts an expire for directory d. after the callout to the daemon, but before the rmdir, another process tries to walk into the same directory. It puts itself onto the waitq, pending the expiration. When the expire finishes, the second process is woken up. In try_to_fill_dentry, it does this check: status = d_invalidate(dentry); if (status != -EBUSY) return -EAGAIN; And status is EBUSY. The dentry still has a non-zero d_inode, and the flags do not contain LOOKUP_CONTINUE or LOOKUP_DIRECTORY So, we fall through and return -EBUSY to the caller." Signed-off-by: Jeff Moyer Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/root.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 0533d37c73ae..6e250030a690 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -243,7 +243,7 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags) struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb); struct autofs_info *ino = autofs4_dentry_ino(dentry); struct dentry *new; - int status = 0; + int status; /* Block on any pending expiry here; invalidate the dentry when expiration is done to trigger mount request with a new @@ -340,7 +340,7 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags) } } - return status; + return 0; } /* For autofs direct mounts the follow link triggers the mount */ -- cgit v1.2.3 From 868eb7a8539d3e8c494209be2b1f4084a274dfef Mon Sep 17 00:00:00 2001 From: Jan Blunck Date: Thu, 1 May 2008 04:35:10 -0700 Subject: autofs: path_{get,put}() cleanups Here are some more places where path_{get,put}() can be used instead of dput()/mntput() pair. Besides that it fixes a bug in autofs4_mount_busy() where mntput() was called before dput(). Signed-off-by: Jan Blunck Cc: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/expire.c | 2 +- fs/autofs4/root.c | 12 ++++++------ 2 files changed, 7 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c index cfa12db8f41b..894fee54d4d8 100644 --- a/fs/autofs4/expire.c +++ b/fs/autofs4/expire.c @@ -73,8 +73,8 @@ static int autofs4_mount_busy(struct vfsmount *mnt, struct dentry *dentry) status = 0; done: DPRINTK("returning = %d", status); - mntput(mnt); dput(dentry); + mntput(mnt); return status; } diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 6e250030a690..edf5b6bddb52 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -146,17 +146,17 @@ static int autofs4_dir_open(struct inode *inode, struct file *file) if (d_mountpoint(dentry)) { struct file *fp = NULL; - struct vfsmount *fp_mnt = mntget(mnt); - struct dentry *fp_dentry = dget(dentry); + struct path fp_path = { .dentry = dentry, .mnt = mnt }; - if (!autofs4_follow_mount(&fp_mnt, &fp_dentry)) { - dput(fp_dentry); - mntput(fp_mnt); + path_get(&fp_path); + + if (!autofs4_follow_mount(&fp_path.mnt, &fp_path.dentry)) { + path_put(&fp_path); dcache_dir_close(inode, file); goto out; } - fp = dentry_open(fp_dentry, fp_mnt, file->f_flags); + fp = dentry_open(fp_path.dentry, fp_path.mnt, file->f_flags); status = PTR_ERR(fp); if (IS_ERR(fp)) { dcache_dir_close(inode, file); -- cgit v1.2.3 From bd7309677c937bf23296f6c81027123c84c5cc5c Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Thu, 1 May 2008 04:35:15 -0700 Subject: fuse: use clamp() rather than nested min/max clamp() exists for this use. Signed-off-by: Harvey Harrison Cc: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/file.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 9ced35b00686..f28cf8b46f80 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -934,7 +934,7 @@ static int fuse_get_user_pages(struct fuse_req *req, const char __user *buf, nbytes = min(nbytes, (unsigned) FUSE_MAX_PAGES_PER_REQ << PAGE_SHIFT); npages = (nbytes + offset + PAGE_SIZE - 1) >> PAGE_SHIFT; - npages = min(max(npages, 1), FUSE_MAX_PAGES_PER_REQ); + npages = clamp(npages, 1, FUSE_MAX_PAGES_PER_REQ); down_read(¤t->mm->mmap_sem); npages = get_user_pages(current, current->mm, user_addr, npages, write, 0, req->pages, NULL); -- cgit v1.2.3 From 1b690b48786229571e590dd22fe01ecc22a8746b Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Thu, 1 May 2008 16:59:24 +0100 Subject: [JFFS2] Invert last argument of jffs2_gc_fetch_inode(), make it boolean. We don't actually care about nlink; we only care whether the inode in question is unlinked or not. Signed-off-by: David Woodhouse --- fs/jffs2/fs.c | 9 +++++---- fs/jffs2/gc.c | 2 +- fs/jffs2/os-linux.h | 2 +- fs/jffs2/wbuf.c | 2 +- 4 files changed, 8 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c index 3eb1c84b0a33..e14b185a80df 100644 --- a/fs/jffs2/fs.c +++ b/fs/jffs2/fs.c @@ -586,11 +586,12 @@ void jffs2_gc_release_inode(struct jffs2_sb_info *c, } struct jffs2_inode_info *jffs2_gc_fetch_inode(struct jffs2_sb_info *c, - int inum, int nlink) + int inum, int unlinked) { struct inode *inode; struct jffs2_inode_cache *ic; - if (!nlink) { + + if (unlinked) { /* The inode has zero nlink but its nodes weren't yet marked obsolete. This has to be because we're still waiting for the final (close() and) iput() to happen. @@ -638,8 +639,8 @@ struct jffs2_inode_info *jffs2_gc_fetch_inode(struct jffs2_sb_info *c, return ERR_CAST(inode); } if (is_bad_inode(inode)) { - printk(KERN_NOTICE "Eep. read_inode() failed for ino #%u. nlink %d\n", - inum, nlink); + printk(KERN_NOTICE "Eep. read_inode() failed for ino #%u. unlinked %d\n", + inum, unlinked); /* NB. This will happen again. We need to do something appropriate here. */ iput(inode); return ERR_PTR(-EIO); diff --git a/fs/jffs2/gc.c b/fs/jffs2/gc.c index bad005664e30..d4db0efeceaf 100644 --- a/fs/jffs2/gc.c +++ b/fs/jffs2/gc.c @@ -401,7 +401,7 @@ int jffs2_garbage_collect_pass(struct jffs2_sb_info *c) nlink = ic->nlink; spin_unlock(&c->inocache_lock); - f = jffs2_gc_fetch_inode(c, inum, nlink); + f = jffs2_gc_fetch_inode(c, inum, !nlink); if (IS_ERR(f)) { ret = PTR_ERR(f); goto release_sem; diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h index 1b10d2594092..2cc866cf134f 100644 --- a/fs/jffs2/os-linux.h +++ b/fs/jffs2/os-linux.h @@ -187,7 +187,7 @@ int jffs2_do_fill_super(struct super_block *sb, void *data, int silent); void jffs2_gc_release_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f); struct jffs2_inode_info *jffs2_gc_fetch_inode(struct jffs2_sb_info *c, - int inum, int nlink); + int inum, int unlinked); unsigned char *jffs2_gc_fetch_page(struct jffs2_sb_info *c, struct jffs2_inode_info *f, diff --git a/fs/jffs2/wbuf.c b/fs/jffs2/wbuf.c index 8de52b607678..92ab32a987ba 100644 --- a/fs/jffs2/wbuf.c +++ b/fs/jffs2/wbuf.c @@ -494,7 +494,7 @@ static void jffs2_wbuf_recover(struct jffs2_sb_info *c) /* If it's an in-core inode, then we have to adjust any full_dirent or full_dnode structure to point to the new version instead of the old */ - f = jffs2_gc_fetch_inode(c, ic->ino, ic->nlink); + f = jffs2_gc_fetch_inode(c, ic->ino, !ic->nlink); if (IS_ERR(f)) { /* Should never happen; it _must_ be present */ JFFS2_ERROR("Failed to iget() ino #%u, err %ld\n", -- cgit v1.2.3 From a2dcb44c3c5a8151d2d9f6ac8ad0789efcdbe184 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 23 Apr 2008 14:05:15 -0400 Subject: [PATCH] make osf_select() use core_sys_select() ... instead of open-coding it Signed-off-by: Al Viro --- arch/alpha/kernel/osf_sys.c | 69 +++------------------------------------------ fs/select.c | 2 +- include/linux/poll.h | 2 ++ 3 files changed, 7 insertions(+), 66 deletions(-) (limited to 'fs') diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index 9fee37e2596f..32ca1b927307 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -981,27 +981,18 @@ asmlinkage int osf_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, struct timeval32 __user *tvp) { - fd_set_bits fds; - char *bits; - size_t size; - long timeout; - int ret = -EINVAL; - struct fdtable *fdt; - int max_fds; - - timeout = MAX_SCHEDULE_TIMEOUT; + s64 timeout = MAX_SCHEDULE_TIMEOUT; if (tvp) { time_t sec, usec; if (!access_ok(VERIFY_READ, tvp, sizeof(*tvp)) || __get_user(sec, &tvp->tv_sec) || __get_user(usec, &tvp->tv_usec)) { - ret = -EFAULT; - goto out_nofds; + return -EFAULT; } if (sec < 0 || usec < 0) - goto out_nofds; + return -EINVAL; if ((unsigned long) sec < MAX_SELECT_SECONDS) { timeout = (usec + 1000000/HZ - 1) / (1000000/HZ); @@ -1009,60 +1000,8 @@ osf_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, } } - rcu_read_lock(); - fdt = files_fdtable(current->files); - max_fds = fdt->max_fds; - rcu_read_unlock(); - if (n < 0 || n > max_fds) - goto out_nofds; - - /* - * We need 6 bitmaps (in/out/ex for both incoming and outgoing), - * since we used fdset we need to allocate memory in units of - * long-words. - */ - ret = -ENOMEM; - size = FDS_BYTES(n); - bits = kmalloc(6 * size, GFP_KERNEL); - if (!bits) - goto out_nofds; - fds.in = (unsigned long *) bits; - fds.out = (unsigned long *) (bits + size); - fds.ex = (unsigned long *) (bits + 2*size); - fds.res_in = (unsigned long *) (bits + 3*size); - fds.res_out = (unsigned long *) (bits + 4*size); - fds.res_ex = (unsigned long *) (bits + 5*size); - - if ((ret = get_fd_set(n, inp->fds_bits, fds.in)) || - (ret = get_fd_set(n, outp->fds_bits, fds.out)) || - (ret = get_fd_set(n, exp->fds_bits, fds.ex))) - goto out; - zero_fd_set(n, fds.res_in); - zero_fd_set(n, fds.res_out); - zero_fd_set(n, fds.res_ex); - - ret = do_select(n, &fds, &timeout); - /* OSF does not copy back the remaining time. */ - - if (ret < 0) - goto out; - if (!ret) { - ret = -ERESTARTNOHAND; - if (signal_pending(current)) - goto out; - ret = 0; - } - - if (set_fd_set(n, inp->fds_bits, fds.res_in) || - set_fd_set(n, outp->fds_bits, fds.res_out) || - set_fd_set(n, exp->fds_bits, fds.res_ex)) - ret = -EFAULT; - - out: - kfree(bits); - out_nofds: - return ret; + return core_sys_select(n, inp, outp, exp, &timeout); } struct rusage32 { diff --git a/fs/select.c b/fs/select.c index 2c292146e246..80d70387e790 100644 --- a/fs/select.c +++ b/fs/select.c @@ -298,7 +298,7 @@ int do_select(int n, fd_set_bits *fds, s64 *timeout) #define MAX_SELECT_SECONDS \ ((unsigned long) (MAX_SCHEDULE_TIMEOUT / HZ)-1) -static int core_sys_select(int n, fd_set __user *inp, fd_set __user *outp, +int core_sys_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, s64 *timeout) { fd_set_bits fds; diff --git a/include/linux/poll.h b/include/linux/poll.h index 16d813b364ef..ef453828877a 100644 --- a/include/linux/poll.h +++ b/include/linux/poll.h @@ -117,6 +117,8 @@ void zero_fd_set(unsigned long nr, unsigned long *fdset) extern int do_select(int n, fd_set_bits *fds, s64 *timeout); extern int do_sys_poll(struct pollfd __user * ufds, unsigned int nfds, s64 *timeout); +extern int core_sys_select(int n, fd_set __user *inp, fd_set __user *outp, + fd_set __user *exp, s64 *timeout); #endif /* KERNEL */ -- cgit v1.2.3 From 9f3acc3140444a900ab280de942291959f0f615d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 24 Apr 2008 07:44:08 -0400 Subject: [PATCH] split linux/file.h Initial splitoff of the low-level stuff; taken to fdtable.h Signed-off-by: Al Viro --- arch/mips/kernel/kspd.c | 1 + arch/powerpc/platforms/cell/spufs/coredump.c | 1 + drivers/char/tty_audit.c | 1 + drivers/char/tty_io.c | 1 + fs/compat.c | 1 + fs/dnotify.c | 2 +- fs/exec.c | 1 + fs/fcntl.c | 1 + fs/file.c | 1 + fs/file_table.c | 1 + fs/locks.c | 1 + fs/open.c | 1 + fs/proc/array.c | 1 + fs/proc/base.c | 1 + fs/select.c | 1 + include/linux/fdtable.h | 99 ++++++++++++++++++++++++++++ include/linux/file.h | 86 +----------------------- include/linux/init_task.h | 2 +- kernel/exit.c | 1 + kernel/fork.c | 1 + kernel/kmod.c | 1 + security/selinux/hooks.c | 1 + 22 files changed, 121 insertions(+), 86 deletions(-) create mode 100644 include/linux/fdtable.h (limited to 'fs') diff --git a/arch/mips/kernel/kspd.c b/arch/mips/kernel/kspd.c index 998c4efcce88..ceb62dce1c9c 100644 --- a/arch/mips/kernel/kspd.c +++ b/arch/mips/kernel/kspd.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include #include diff --git a/arch/powerpc/platforms/cell/spufs/coredump.c b/arch/powerpc/platforms/cell/spufs/coredump.c index b962c3ab470c..af116aadba10 100644 --- a/arch/powerpc/platforms/cell/spufs/coredump.c +++ b/arch/powerpc/platforms/cell/spufs/coredump.c @@ -22,6 +22,7 @@ #include #include +#include #include #include #include diff --git a/drivers/char/tty_audit.c b/drivers/char/tty_audit.c index 6342b0534f4d..3582f43345a8 100644 --- a/drivers/char/tty_audit.c +++ b/drivers/char/tty_audit.c @@ -11,6 +11,7 @@ #include #include +#include #include struct tty_audit_buf { diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index 1d298c2cf930..49c1a2267a55 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -78,6 +78,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/compat.c b/fs/compat.c index 139dc93c092d..332a869d2c53 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/dnotify.c b/fs/dnotify.c index eaecc4cfe540..676073b8dda5 100644 --- a/fs/dnotify.c +++ b/fs/dnotify.c @@ -20,7 +20,7 @@ #include #include #include -#include +#include int dir_notify_enable __read_mostly = 1; diff --git a/fs/exec.c b/fs/exec.c index 9f9f931ef949..aeaa9791d8be 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -24,6 +24,7 @@ #include #include +#include #include #include #include diff --git a/fs/fcntl.c b/fs/fcntl.c index 3f3ac630ccde..bfd776509a72 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/file.c b/fs/file.c index 5110acb1c9ef..f6fbcb49faf7 100644 --- a/fs/file.c +++ b/fs/file.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/file_table.c b/fs/file_table.c index 7a0a9b872251..83084225b4c3 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/locks.c b/fs/locks.c index 44d9a6a7ec50..663c069b59b3 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -116,6 +116,7 @@ #include #include +#include #include #include #include diff --git a/fs/open.c b/fs/open.c index 7af1f05d5978..a1450086e92f 100644 --- a/fs/open.c +++ b/fs/open.c @@ -7,6 +7,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/proc/array.c b/fs/proc/array.c index c135cbdd9127..dca997a93bff 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -73,6 +73,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/proc/base.c b/fs/proc/base.c index fcf02f2deeba..808cbdc193d3 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -56,6 +56,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/select.c b/fs/select.c index 80d70387e790..8dda969614a9 100644 --- a/fs/select.c +++ b/fs/select.c @@ -21,6 +21,7 @@ #include #include /* for STICKY_TIMEOUTS */ #include +#include #include #include diff --git a/include/linux/fdtable.h b/include/linux/fdtable.h new file mode 100644 index 000000000000..a118f3c0b240 --- /dev/null +++ b/include/linux/fdtable.h @@ -0,0 +1,99 @@ +/* + * descriptor table internals; you almost certainly want file.h instead. + */ + +#ifndef __LINUX_FDTABLE_H +#define __LINUX_FDTABLE_H + +#include +#include +#include +#include +#include +#include + +/* + * The default fd array needs to be at least BITS_PER_LONG, + * as this is the granularity returned by copy_fdset(). + */ +#define NR_OPEN_DEFAULT BITS_PER_LONG + +/* + * The embedded_fd_set is a small fd_set, + * suitable for most tasks (which open <= BITS_PER_LONG files) + */ +struct embedded_fd_set { + unsigned long fds_bits[1]; +}; + +struct fdtable { + unsigned int max_fds; + struct file ** fd; /* current fd array */ + fd_set *close_on_exec; + fd_set *open_fds; + struct rcu_head rcu; + struct fdtable *next; +}; + +/* + * Open file table structure + */ +struct files_struct { + /* + * read mostly part + */ + atomic_t count; + struct fdtable *fdt; + struct fdtable fdtab; + /* + * written part on a separate cache line in SMP + */ + spinlock_t file_lock ____cacheline_aligned_in_smp; + int next_fd; + struct embedded_fd_set close_on_exec_init; + struct embedded_fd_set open_fds_init; + struct file * fd_array[NR_OPEN_DEFAULT]; +}; + +#define files_fdtable(files) (rcu_dereference((files)->fdt)) + +extern struct kmem_cache *filp_cachep; + +struct file_operations; +struct vfsmount; +struct dentry; + +extern int expand_files(struct files_struct *, int nr); +extern void free_fdtable_rcu(struct rcu_head *rcu); +extern void __init files_defer_init(void); + +static inline void free_fdtable(struct fdtable *fdt) +{ + call_rcu(&fdt->rcu, free_fdtable_rcu); +} + +static inline struct file * fcheck_files(struct files_struct *files, unsigned int fd) +{ + struct file * file = NULL; + struct fdtable *fdt = files_fdtable(files); + + if (fd < fdt->max_fds) + file = rcu_dereference(fdt->fd[fd]); + return file; +} + +/* + * Check whether the specified fd has an open file. + */ +#define fcheck(fd) fcheck_files(current->files, fd) + +struct task_struct; + +struct files_struct *get_files_struct(struct task_struct *); +void put_files_struct(struct files_struct *fs); +void reset_files_struct(struct files_struct *); +int unshare_files(struct files_struct **); + +extern struct kmem_cache *files_cachep; + +#endif /* __LINUX_FDTABLE_H */ diff --git a/include/linux/file.h b/include/linux/file.h index 69baf5a4f0a5..27c64bdc68c9 100644 --- a/include/linux/file.h +++ b/include/linux/file.h @@ -5,59 +5,11 @@ #ifndef __LINUX_FILE_H #define __LINUX_FILE_H -#include -#include #include -#include -#include #include +#include -/* - * The default fd array needs to be at least BITS_PER_LONG, - * as this is the granularity returned by copy_fdset(). - */ -#define NR_OPEN_DEFAULT BITS_PER_LONG - -/* - * The embedded_fd_set is a small fd_set, - * suitable for most tasks (which open <= BITS_PER_LONG files) - */ -struct embedded_fd_set { - unsigned long fds_bits[1]; -}; - -struct fdtable { - unsigned int max_fds; - struct file ** fd; /* current fd array */ - fd_set *close_on_exec; - fd_set *open_fds; - struct rcu_head rcu; - struct fdtable *next; -}; - -/* - * Open file table structure - */ -struct files_struct { - /* - * read mostly part - */ - atomic_t count; - struct fdtable *fdt; - struct fdtable fdtab; - /* - * written part on a separate cache line in SMP - */ - spinlock_t file_lock ____cacheline_aligned_in_smp; - int next_fd; - struct embedded_fd_set close_on_exec_init; - struct embedded_fd_set open_fds_init; - struct file * fd_array[NR_OPEN_DEFAULT]; -}; - -#define files_fdtable(files) (rcu_dereference((files)->fdt)) - -extern struct kmem_cache *filp_cachep; +struct file; extern void __fput(struct file *); extern void fput(struct file *); @@ -85,41 +37,7 @@ extern void put_filp(struct file *); extern int get_unused_fd(void); extern int get_unused_fd_flags(int flags); extern void put_unused_fd(unsigned int fd); -struct kmem_cache; - -extern int expand_files(struct files_struct *, int nr); -extern void free_fdtable_rcu(struct rcu_head *rcu); -extern void __init files_defer_init(void); - -static inline void free_fdtable(struct fdtable *fdt) -{ - call_rcu(&fdt->rcu, free_fdtable_rcu); -} - -static inline struct file * fcheck_files(struct files_struct *files, unsigned int fd) -{ - struct file * file = NULL; - struct fdtable *fdt = files_fdtable(files); - - if (fd < fdt->max_fds) - file = rcu_dereference(fdt->fd[fd]); - return file; -} - -/* - * Check whether the specified fd has an open file. - */ -#define fcheck(fd) fcheck_files(current->files, fd) extern void fd_install(unsigned int fd, struct file *file); -struct task_struct; - -struct files_struct *get_files_struct(struct task_struct *); -void put_files_struct(struct files_struct *fs); -void reset_files_struct(struct files_struct *); -int unshare_files(struct files_struct **); - -extern struct kmem_cache *files_cachep; - #endif /* __LINUX_FILE_H */ diff --git a/include/linux/init_task.h b/include/linux/init_task.h index bf6b8a61f8db..b24c2875aa05 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -1,7 +1,7 @@ #ifndef _LINUX__INIT_TASK_H #define _LINUX__INIT_TASK_H -#include +#include #include #include #include diff --git a/kernel/exit.c b/kernel/exit.c index d3ad54677f9c..1510f78a0ffa 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include #include diff --git a/kernel/fork.c b/kernel/fork.c index 2bb675af4de3..933e60ebccae 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include diff --git a/kernel/kmod.c b/kernel/kmod.c index e2764047ec03..8df97d3dfda8 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include #include diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 1b50a6ebc55f..1c864c0efe2b 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -39,6 +39,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3 From 2030a42cecd4dd1985a2ab03e25f3cd6106a5ca8 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 23 Feb 2008 06:46:49 -0500 Subject: [PATCH] sanitize anon_inode_getfd() a) none of the callers even looks at inode or file returned by anon_inode_getfd() b) any caller that would try to look at those would be racy, since by the time it returns we might have raced with close() from another thread and that file would be pining for fjords. Signed-off-by: Al Viro --- fs/anon_inodes.c | 13 +++---------- fs/eventfd.c | 15 +++++---------- fs/eventpoll.c | 23 ++++++++--------------- fs/signalfd.c | 17 ++++------------- fs/timerfd.c | 11 +++-------- include/linux/anon_inodes.h | 3 +-- virt/kvm/kvm_main.c | 21 +++++---------------- 7 files changed, 29 insertions(+), 74 deletions(-) (limited to 'fs') diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c index f42be069e085..977ef208c051 100644 --- a/fs/anon_inodes.c +++ b/fs/anon_inodes.c @@ -57,9 +57,6 @@ static struct dentry_operations anon_inodefs_dentry_operations = { * anonymous inode, and a dentry that describe the "class" * of the file * - * @pfd: [out] pointer to the file descriptor - * @dpinode: [out] pointer to the inode - * @pfile: [out] pointer to the file struct * @name: [in] name of the "class" of the new file * @fops [in] file operations for the new file * @priv [in] private data for the new file (will be file's private_data) @@ -68,10 +65,9 @@ static struct dentry_operations anon_inodefs_dentry_operations = { * that do not need to have a full-fledged inode in order to operate correctly. * All the files created with anon_inode_getfd() will share a single inode, * hence saving memory and avoiding code duplication for the file/inode/dentry - * setup. + * setup. Returns new descriptor or -error. */ -int anon_inode_getfd(int *pfd, struct inode **pinode, struct file **pfile, - const char *name, const struct file_operations *fops, +int anon_inode_getfd(const char *name, const struct file_operations *fops, void *priv) { struct qstr this; @@ -125,10 +121,7 @@ int anon_inode_getfd(int *pfd, struct inode **pinode, struct file **pfile, fd_install(fd, file); - *pfd = fd; - *pinode = anon_inode_inode; - *pfile = file; - return 0; + return fd; err_dput: dput(dentry); diff --git a/fs/eventfd.c b/fs/eventfd.c index a9f130cd50ac..343942deeec1 100644 --- a/fs/eventfd.c +++ b/fs/eventfd.c @@ -200,10 +200,8 @@ struct file *eventfd_fget(int fd) asmlinkage long sys_eventfd(unsigned int count) { - int error, fd; + int fd; struct eventfd_ctx *ctx; - struct file *file; - struct inode *inode; ctx = kmalloc(sizeof(*ctx), GFP_KERNEL); if (!ctx) @@ -216,12 +214,9 @@ asmlinkage long sys_eventfd(unsigned int count) * When we call this, the initialization must be complete, since * anon_inode_getfd() will install the fd. */ - error = anon_inode_getfd(&fd, &inode, &file, "[eventfd]", - &eventfd_fops, ctx); - if (!error) - return fd; - - kfree(ctx); - return error; + fd = anon_inode_getfd("[eventfd]", &eventfd_fops, ctx); + if (fd < 0) + kfree(ctx); + return fd; } diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 221086fef174..990c01d2d66b 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1050,8 +1050,6 @@ asmlinkage long sys_epoll_create(int size) { int error, fd = -1; struct eventpoll *ep; - struct inode *inode; - struct file *file; DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d)\n", current, size)); @@ -1061,29 +1059,24 @@ asmlinkage long sys_epoll_create(int size) * structure ( "struct eventpoll" ). */ error = -EINVAL; - if (size <= 0 || (error = ep_alloc(&ep)) != 0) + if (size <= 0 || (error = ep_alloc(&ep)) < 0) { + fd = error; goto error_return; + } /* * Creates all the items needed to setup an eventpoll file. That is, - * a file structure, and inode and a free file descriptor. + * a file structure and a free file descriptor. */ - error = anon_inode_getfd(&fd, &inode, &file, "[eventpoll]", - &eventpoll_fops, ep); - if (error) - goto error_free; + fd = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep); + if (fd < 0) + ep_free(ep); +error_return: DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d) = %d\n", current, size, fd)); return fd; - -error_free: - ep_free(ep); -error_return: - DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d) = %d\n", - current, size, error)); - return error; } /* diff --git a/fs/signalfd.c b/fs/signalfd.c index 8ead0db35933..619725644c75 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -207,11 +207,8 @@ static const struct file_operations signalfd_fops = { asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemask) { - int error; sigset_t sigmask; struct signalfd_ctx *ctx; - struct file *file; - struct inode *inode; if (sizemask != sizeof(sigset_t) || copy_from_user(&sigmask, user_mask, sizeof(sigmask))) @@ -230,12 +227,11 @@ asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemas * When we call this, the initialization must be complete, since * anon_inode_getfd() will install the fd. */ - error = anon_inode_getfd(&ufd, &inode, &file, "[signalfd]", - &signalfd_fops, ctx); - if (error) - goto err_fdalloc; + ufd = anon_inode_getfd("[signalfd]", &signalfd_fops, ctx); + if (ufd < 0) + kfree(ctx); } else { - file = fget(ufd); + struct file *file = fget(ufd); if (!file) return -EBADF; ctx = file->private_data; @@ -252,9 +248,4 @@ asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemas } return ufd; - -err_fdalloc: - kfree(ctx); - return error; } - diff --git a/fs/timerfd.c b/fs/timerfd.c index 5400524e9cb1..d87d354ec424 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -181,10 +181,8 @@ static struct file *timerfd_fget(int fd) asmlinkage long sys_timerfd_create(int clockid, int flags) { - int error, ufd; + int ufd; struct timerfd_ctx *ctx; - struct file *file; - struct inode *inode; if (flags) return -EINVAL; @@ -200,12 +198,9 @@ asmlinkage long sys_timerfd_create(int clockid, int flags) ctx->clockid = clockid; hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS); - error = anon_inode_getfd(&ufd, &inode, &file, "[timerfd]", - &timerfd_fops, ctx); - if (error) { + ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx); + if (ufd < 0) kfree(ctx); - return error; - } return ufd; } diff --git a/include/linux/anon_inodes.h b/include/linux/anon_inodes.h index b2e1ba325b9a..6129e58ca7c9 100644 --- a/include/linux/anon_inodes.h +++ b/include/linux/anon_inodes.h @@ -8,8 +8,7 @@ #ifndef _LINUX_ANON_INODES_H #define _LINUX_ANON_INODES_H -int anon_inode_getfd(int *pfd, struct inode **pinode, struct file **pfile, - const char *name, const struct file_operations *fops, +int anon_inode_getfd(const char *name, const struct file_operations *fops, void *priv); #endif /* _LINUX_ANON_INODES_H */ diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index c82cf15730a1..e89338e2b043 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -834,16 +834,9 @@ static const struct file_operations kvm_vcpu_fops = { */ static int create_vcpu_fd(struct kvm_vcpu *vcpu) { - int fd, r; - struct inode *inode; - struct file *file; - - r = anon_inode_getfd(&fd, &inode, &file, - "kvm-vcpu", &kvm_vcpu_fops, vcpu); - if (r) { + int fd = anon_inode_getfd("kvm-vcpu", &kvm_vcpu_fops, vcpu); + if (fd < 0) kvm_put_kvm(vcpu->kvm); - return r; - } return fd; } @@ -1168,19 +1161,15 @@ static const struct file_operations kvm_vm_fops = { static int kvm_dev_ioctl_create_vm(void) { - int fd, r; - struct inode *inode; - struct file *file; + int fd; struct kvm *kvm; kvm = kvm_create_vm(); if (IS_ERR(kvm)) return PTR_ERR(kvm); - r = anon_inode_getfd(&fd, &inode, &file, "kvm-vm", &kvm_vm_fops, kvm); - if (r) { + fd = anon_inode_getfd("kvm-vm", &kvm_vm_fops, kvm); + if (fd < 0) kvm_put_kvm(kvm); - return r; - } return fd; } -- cgit v1.2.3 From 5c598b3428c372a1209597cee99a70da20625876 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 27 Apr 2008 20:04:15 -0400 Subject: [PATCH] fix sysctl_nr_open bugs * if luser with root sets it to something that is not a multiple of BITS_PER_LONG, the system is screwed. * if it gets decreased at the wrong time, we can get expand_files() returning success and _not_ increasing the size of table as asked. Signed-off-by: Al Viro --- fs/file.c | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/file.c b/fs/file.c index f6fbcb49faf7..4c6f0ea12c41 100644 --- a/fs/file.c +++ b/fs/file.c @@ -150,8 +150,16 @@ static struct fdtable * alloc_fdtable(unsigned int nr) nr /= (1024 / sizeof(struct file *)); nr = roundup_pow_of_two(nr + 1); nr *= (1024 / sizeof(struct file *)); - if (nr > sysctl_nr_open) - nr = sysctl_nr_open; + /* + * Note that this can drive nr *below* what we had passed if sysctl_nr_open + * had been set lower between the check in expand_files() and here. Deal + * with that in caller, it's cheaper that way. + * + * We make sure that nr remains a multiple of BITS_PER_LONG - otherwise + * bitmaps handling below becomes unpleasant, to put it mildly... + */ + if (unlikely(nr > sysctl_nr_open)) + nr = ((sysctl_nr_open - 1) | (BITS_PER_LONG - 1)) + 1; fdt = kmalloc(sizeof(struct fdtable), GFP_KERNEL); if (!fdt) @@ -199,6 +207,16 @@ static int expand_fdtable(struct files_struct *files, int nr) spin_lock(&files->file_lock); if (!new_fdt) return -ENOMEM; + /* + * extremely unlikely race - sysctl_nr_open decreased between the check in + * caller and alloc_fdtable(). Cheaper to catch it here... + */ + if (unlikely(new_fdt->max_fds <= nr)) { + free_fdarr(new_fdt); + free_fdset(new_fdt); + kfree(new_fdt); + return -EMFILE; + } /* * Check again since another task may have expanded the fd table while * we dropped the lock -- cgit v1.2.3 From 27c72b040c0be8f3704ed0b6b84c12cbba24a7e8 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Thu, 1 May 2008 18:47:17 +0100 Subject: [JFFS2] Track parent inode for directories (for NFS export) To support NFS export, we need to know the parent inode of directories. Rather than growing the jffs2_inode_cache structure, share space with the nlink field -- which was always set to 1 for directories anyway. Signed-off-by: David Woodhouse --- fs/jffs2/build.c | 31 ++++++++++++++++++++----------- fs/jffs2/dir.c | 35 +++++++++++++++++++++++++---------- fs/jffs2/erase.c | 2 +- fs/jffs2/fs.c | 5 ++--- fs/jffs2/gc.c | 6 +++--- fs/jffs2/nodelist.h | 5 ++++- fs/jffs2/nodemgmt.c | 2 +- fs/jffs2/readinode.c | 7 ++++--- fs/jffs2/scan.c | 2 +- fs/jffs2/wbuf.c | 2 +- fs/jffs2/write.c | 11 ++++++----- fs/jffs2/xattr.c | 4 ++-- 12 files changed, 70 insertions(+), 42 deletions(-) (limited to 'fs') diff --git a/fs/jffs2/build.c b/fs/jffs2/build.c index d58f845ccb85..c5e1450d79f9 100644 --- a/fs/jffs2/build.c +++ b/fs/jffs2/build.c @@ -46,7 +46,7 @@ next_inode(int *i, struct jffs2_inode_cache *ic, struct jffs2_sb_info *c) static void jffs2_build_inode_pass1(struct jffs2_sb_info *c, - struct jffs2_inode_cache *ic) + struct jffs2_inode_cache *ic) { struct jffs2_full_dirent *fd; @@ -68,11 +68,17 @@ static void jffs2_build_inode_pass1(struct jffs2_sb_info *c, continue; } - if (child_ic->nlink++ && fd->type == DT_DIR) { - JFFS2_ERROR("child dir \"%s\" (ino #%u) of dir ino #%u appears to be a hard link\n", - fd->name, fd->ino, ic->ino); - /* TODO: What do we do about it? */ - } + if (fd->type == DT_DIR) { + if (child_ic->pino_nlink) { + JFFS2_ERROR("child dir \"%s\" (ino #%u) of dir ino #%u appears to be a hard link\n", + fd->name, fd->ino, ic->ino); + /* TODO: What do we do about it? */ + } else { + child_ic->pino_nlink = ic->ino; + } + } else + child_ic->pino_nlink++; + dbg_fsbuild("increased nlink for child \"%s\" (ino #%u)\n", fd->name, fd->ino); /* Can't free scan_dents so far. We might need them in pass 2 */ } @@ -125,7 +131,7 @@ static int jffs2_build_filesystem(struct jffs2_sb_info *c) dbg_fsbuild("pass 2 starting\n"); for_each_inode(i, c, ic) { - if (ic->nlink) + if (ic->pino_nlink) continue; jffs2_build_remove_unlinked_inode(c, ic, &dead_fds); @@ -232,16 +238,19 @@ static void jffs2_build_remove_unlinked_inode(struct jffs2_sb_info *c, /* Reduce nlink of the child. If it's now zero, stick it on the dead_fds list to be cleaned up later. Else just free the fd */ - child_ic->nlink--; + if (fd->type == DT_DIR) + child_ic->pino_nlink = 0; + else + child_ic->pino_nlink--; - if (!child_ic->nlink) { - dbg_fsbuild("inode #%u (\"%s\") has now got zero nlink, adding to dead_fds list.\n", + if (!child_ic->pino_nlink) { + dbg_fsbuild("inode #%u (\"%s\") now has no links; adding to dead_fds list.\n", fd->ino, fd->name); fd->next = *dead_fds; *dead_fds = fd; } else { dbg_fsbuild("inode #%u (\"%s\") has now got nlink %d. Ignoring.\n", - fd->ino, fd->name, child_ic->nlink); + fd->ino, fd->name, child_ic->pino_nlink); jffs2_free_full_dirent(fd); } } diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index 2bba3d3435be..c0c141f6fde1 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -226,7 +226,8 @@ static int jffs2_create(struct inode *dir_i, struct dentry *dentry, int mode, d_instantiate(dentry, inode); D1(printk(KERN_DEBUG "jffs2_create: Created ino #%lu with mode %o, nlink %d(%d). nrpages %ld\n", - inode->i_ino, inode->i_mode, inode->i_nlink, f->inocache->nlink, inode->i_mapping->nrpages)); + inode->i_ino, inode->i_mode, inode->i_nlink, + f->inocache->pino_nlink, inode->i_mapping->nrpages)); return 0; fail: @@ -250,7 +251,7 @@ static int jffs2_unlink(struct inode *dir_i, struct dentry *dentry) ret = jffs2_do_unlink(c, dir_f, dentry->d_name.name, dentry->d_name.len, dead_f, now); if (dead_f->inocache) - dentry->d_inode->i_nlink = dead_f->inocache->nlink; + dentry->d_inode->i_nlink = dead_f->inocache->pino_nlink; if (!ret) dir_i->i_mtime = dir_i->i_ctime = ITIME(now); return ret; @@ -283,7 +284,7 @@ static int jffs2_link (struct dentry *old_dentry, struct inode *dir_i, struct de if (!ret) { mutex_lock(&f->sem); - old_dentry->d_inode->i_nlink = ++f->inocache->nlink; + old_dentry->d_inode->i_nlink = ++f->inocache->pino_nlink; mutex_unlock(&f->sem); d_instantiate(dentry, old_dentry->d_inode); dir_i->i_mtime = dir_i->i_ctime = ITIME(now); @@ -500,11 +501,14 @@ static int jffs2_mkdir (struct inode *dir_i, struct dentry *dentry, int mode) inode->i_op = &jffs2_dir_inode_operations; inode->i_fop = &jffs2_dir_operations; - /* Directories get nlink 2 at start */ - inode->i_nlink = 2; f = JFFS2_INODE_INFO(inode); + /* Directories get nlink 2 at start */ + inode->i_nlink = 2; + /* but ic->pino_nlink is the parent ino# */ + f->inocache->pino_nlink = dir_i->i_ino; + ri->data_crc = cpu_to_je32(0); ri->node_crc = cpu_to_je32(crc32(0, ri, sizeof(*ri)-8)); @@ -601,17 +605,25 @@ static int jffs2_mkdir (struct inode *dir_i, struct dentry *dentry, int mode) static int jffs2_rmdir (struct inode *dir_i, struct dentry *dentry) { + struct jffs2_sb_info *c = JFFS2_SB_INFO(dir_i->i_sb); + struct jffs2_inode_info *dir_f = JFFS2_INODE_INFO(dir_i); struct jffs2_inode_info *f = JFFS2_INODE_INFO(dentry->d_inode); struct jffs2_full_dirent *fd; int ret; + uint32_t now = get_seconds(); for (fd = f->dents ; fd; fd = fd->next) { if (fd->ino) return -ENOTEMPTY; } - ret = jffs2_unlink(dir_i, dentry); - if (!ret) + + ret = jffs2_do_unlink(c, dir_f, dentry->d_name.name, + dentry->d_name.len, f, now); + if (!ret) { + dir_i->i_mtime = dir_i->i_ctime = ITIME(now); + clear_nlink(dentry->d_inode); drop_nlink(dir_i); + } return ret; } @@ -824,7 +836,10 @@ static int jffs2_rename (struct inode *old_dir_i, struct dentry *old_dentry, inode which didn't exist. */ if (victim_f->inocache) { mutex_lock(&victim_f->sem); - victim_f->inocache->nlink--; + if (S_ISDIR(new_dentry->d_inode->i_mode)) + victim_f->inocache->pino_nlink = 0; + else + victim_f->inocache->pino_nlink--; mutex_unlock(&victim_f->sem); } } @@ -845,8 +860,8 @@ static int jffs2_rename (struct inode *old_dir_i, struct dentry *old_dentry, struct jffs2_inode_info *f = JFFS2_INODE_INFO(old_dentry->d_inode); mutex_lock(&f->sem); inc_nlink(old_dentry->d_inode); - if (f->inocache) - f->inocache->nlink++; + if (f->inocache && !S_ISDIR(old_dentry->d_inode->i_mode)) + f->inocache->pino_nlink++; mutex_unlock(&f->sem); printk(KERN_NOTICE "jffs2_rename(): Link succeeded, unlink failed (err %d). You now have a hard link\n", ret); diff --git a/fs/jffs2/erase.c b/fs/jffs2/erase.c index 25a640e566d3..5e86f43616a1 100644 --- a/fs/jffs2/erase.c +++ b/fs/jffs2/erase.c @@ -294,7 +294,7 @@ static inline void jffs2_remove_node_refs_from_ino_list(struct jffs2_sb_info *c, break; #endif default: - if (ic->nodes == (void *)ic && ic->nlink == 0) + if (ic->nodes == (void *)ic && ic->pino_nlink == 0) jffs2_del_ino_cache(c, ic); } } diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c index e14b185a80df..086c43830221 100644 --- a/fs/jffs2/fs.c +++ b/fs/jffs2/fs.c @@ -273,7 +273,7 @@ struct inode *jffs2_iget(struct super_block *sb, unsigned long ino) inode->i_mtime = ITIME(je32_to_cpu(latest_node.mtime)); inode->i_ctime = ITIME(je32_to_cpu(latest_node.ctime)); - inode->i_nlink = f->inocache->nlink; + inode->i_nlink = f->inocache->pino_nlink; inode->i_blocks = (inode->i_size + 511) >> 9; @@ -286,13 +286,12 @@ struct inode *jffs2_iget(struct super_block *sb, unsigned long ino) case S_IFDIR: { struct jffs2_full_dirent *fd; + inode->i_nlink = 2; /* parent and '.' */ for (fd=f->dents; fd; fd = fd->next) { if (fd->type == DT_DIR && fd->ino) inc_nlink(inode); } - /* and '..' */ - inc_nlink(inode); /* Root dir gets i_nlink 3 for some reason */ if (inode->i_ino == 1) inc_nlink(inode); diff --git a/fs/jffs2/gc.c b/fs/jffs2/gc.c index d4db0efeceaf..090c556ffed2 100644 --- a/fs/jffs2/gc.c +++ b/fs/jffs2/gc.c @@ -161,8 +161,8 @@ int jffs2_garbage_collect_pass(struct jffs2_sb_info *c) continue; } - if (!ic->nlink) { - D1(printk(KERN_DEBUG "Skipping check of ino #%d with nlink zero\n", + if (!ic->pino_nlink) { + D1(printk(KERN_DEBUG "Skipping check of ino #%d with nlink/pino zero\n", ic->ino)); spin_unlock(&c->inocache_lock); jffs2_xattr_delete_inode(c, ic); @@ -398,7 +398,7 @@ int jffs2_garbage_collect_pass(struct jffs2_sb_info *c) it's vaguely possible. */ inum = ic->ino; - nlink = ic->nlink; + nlink = ic->pino_nlink; spin_unlock(&c->inocache_lock); f = jffs2_gc_fetch_inode(c, inum, !nlink); diff --git a/fs/jffs2/nodelist.h b/fs/jffs2/nodelist.h index 8219df6eb6d8..1750445556c3 100644 --- a/fs/jffs2/nodelist.h +++ b/fs/jffs2/nodelist.h @@ -177,7 +177,10 @@ struct jffs2_inode_cache { #ifdef CONFIG_JFFS2_FS_XATTR struct jffs2_xattr_ref *xref; #endif - int nlink; + uint32_t pino_nlink; /* Directories store parent inode + here; other inodes store nlink. + Zero always means that it's + completely unlinked. */ }; /* Inode states for 'state' above. We need the 'GC' state to prevent diff --git a/fs/jffs2/nodemgmt.c b/fs/jffs2/nodemgmt.c index 9df8f3ef20df..a9bf9603c1ba 100644 --- a/fs/jffs2/nodemgmt.c +++ b/fs/jffs2/nodemgmt.c @@ -709,7 +709,7 @@ void jffs2_mark_node_obsolete(struct jffs2_sb_info *c, struct jffs2_raw_node_ref break; #endif default: - if (ic->nodes == (void *)ic && ic->nlink == 0) + if (ic->nodes == (void *)ic && ic->pino_nlink == 0) jffs2_del_ino_cache(c, ic); break; } diff --git a/fs/jffs2/readinode.c b/fs/jffs2/readinode.c index 4cb4d76de07f..9fc4833c117c 100644 --- a/fs/jffs2/readinode.c +++ b/fs/jffs2/readinode.c @@ -1123,7 +1123,8 @@ static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, size_t retlen; int ret; - dbg_readinode("ino #%u nlink is %d\n", f->inocache->ino, f->inocache->nlink); + dbg_readinode("ino #%u pino/nlink is %d\n", f->inocache->ino, + f->inocache->pino_nlink); memset(&rii, 0, sizeof(rii)); @@ -1358,7 +1359,7 @@ int jffs2_do_read_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, } dbg_readinode("creating inocache for root inode\n"); memset(f->inocache, 0, sizeof(struct jffs2_inode_cache)); - f->inocache->ino = f->inocache->nlink = 1; + f->inocache->ino = f->inocache->pino_nlink = 1; f->inocache->nodes = (struct jffs2_raw_node_ref *)f->inocache; f->inocache->state = INO_STATE_READING; jffs2_add_ino_cache(c, f->inocache); @@ -1401,7 +1402,7 @@ void jffs2_do_clear_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f) jffs2_clear_acl(f); jffs2_xattr_delete_inode(c, f->inocache); mutex_lock(&f->sem); - deleted = f->inocache && !f->inocache->nlink; + deleted = f->inocache && !f->inocache->pino_nlink; if (f->inocache && f->inocache->state != INO_STATE_CHECKING) jffs2_set_inocache_state(c, f->inocache, INO_STATE_CLEARING); diff --git a/fs/jffs2/scan.c b/fs/jffs2/scan.c index 272872d27fd5..8c1e692bef79 100644 --- a/fs/jffs2/scan.c +++ b/fs/jffs2/scan.c @@ -940,7 +940,7 @@ struct jffs2_inode_cache *jffs2_scan_make_ino_cache(struct jffs2_sb_info *c, uin ic->nodes = (void *)ic; jffs2_add_ino_cache(c, ic); if (ino == 1) - ic->nlink = 1; + ic->pino_nlink = 1; return ic; } diff --git a/fs/jffs2/wbuf.c b/fs/jffs2/wbuf.c index 92ab32a987ba..0e78b00035e4 100644 --- a/fs/jffs2/wbuf.c +++ b/fs/jffs2/wbuf.c @@ -494,7 +494,7 @@ static void jffs2_wbuf_recover(struct jffs2_sb_info *c) /* If it's an in-core inode, then we have to adjust any full_dirent or full_dnode structure to point to the new version instead of the old */ - f = jffs2_gc_fetch_inode(c, ic->ino, !ic->nlink); + f = jffs2_gc_fetch_inode(c, ic->ino, !ic->pino_nlink); if (IS_ERR(f)) { /* Should never happen; it _must_ be present */ JFFS2_ERROR("Failed to iget() ino #%u, err %ld\n", diff --git a/fs/jffs2/write.c b/fs/jffs2/write.c index 87891bdd7915..ca29440e9435 100644 --- a/fs/jffs2/write.c +++ b/fs/jffs2/write.c @@ -19,7 +19,8 @@ #include "compr.h" -int jffs2_do_new_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, uint32_t mode, struct jffs2_raw_inode *ri) +int jffs2_do_new_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, + uint32_t mode, struct jffs2_raw_inode *ri) { struct jffs2_inode_cache *ic; @@ -31,7 +32,7 @@ int jffs2_do_new_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, uint memset(ic, 0, sizeof(*ic)); f->inocache = ic; - f->inocache->nlink = 1; + f->inocache->pino_nlink = 1; /* Will be overwritten shortly for directories */ f->inocache->nodes = (struct jffs2_raw_node_ref *)f->inocache; f->inocache->state = INO_STATE_PRESENT; @@ -635,9 +636,9 @@ int jffs2_do_unlink(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, jffs2_mark_node_obsolete(c, fd->raw); jffs2_free_full_dirent(fd); } - } - - dead_f->inocache->nlink--; + dead_f->inocache->pino_nlink = 0; + } else + dead_f->inocache->pino_nlink--; /* NB: Caller must set inode nlink if appropriate */ mutex_unlock(&dead_f->sem); } diff --git a/fs/jffs2/xattr.c b/fs/jffs2/xattr.c index e48665984cb3..05531f291bfa 100644 --- a/fs/jffs2/xattr.c +++ b/fs/jffs2/xattr.c @@ -592,7 +592,7 @@ void jffs2_xattr_delete_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache When an inode with XATTR is removed, those XATTRs must be removed. */ struct jffs2_xattr_ref *ref, *_ref; - if (!ic || ic->nlink > 0) + if (!ic || ic->pino_nlink > 0) return; down_write(&c->xattr_sem); @@ -829,7 +829,7 @@ void jffs2_build_xattr_subsystem(struct jffs2_sb_info *c) ref->xd and ref->ic are not valid yet. */ xd = jffs2_find_xattr_datum(c, ref->xid); ic = jffs2_get_ino_cache(c, ref->ino); - if (!xd || !ic || !ic->nlink) { + if (!xd || !ic || !ic->pino_nlink) { dbg_xattr("xref(ino=%u, xid=%u, xseqno=%u) is orphan.\n", ref->ino, ref->xid, ref->xseqno); ref->xseqno |= XREF_DELETE_MARKER; -- cgit v1.2.3 From a98889f3d8882995b5aa2255b931cf0202325cc0 Mon Sep 17 00:00:00 2001 From: Jared Hulbert Date: Tue, 29 Apr 2008 23:26:49 -0700 Subject: [MTD][NOR] Add physical address to point() method MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Adding the ability to get a physical address from point() in addition to virtual address. This physical address is required for XIP of userspace code from flash. Signed-off-by: Jared Hulbert Reviewed-by: Jörn Engel Acked-by: Nicolas Pitre Acked-by: Greg Ungerer Signed-off-by: David Woodhouse --- drivers/mtd/chips/cfi_cmdset_0001.c | 14 ++++++++------ drivers/mtd/devices/mtdram.c | 11 +++++++---- drivers/mtd/devices/phram.c | 13 +++++++------ drivers/mtd/devices/pmc551.c | 27 +++++++++++++++++---------- drivers/mtd/devices/slram.c | 15 ++++++++++----- drivers/mtd/maps/uclinux.c | 6 ++++-- drivers/mtd/mtdpart.c | 8 ++++---- fs/jffs2/erase.c | 7 ++++--- fs/jffs2/readinode.c | 9 +++++---- fs/jffs2/scan.c | 7 ++++--- include/linux/mtd/mtd.h | 6 ++++-- include/linux/mtd/pmc551.h | 5 +++-- 12 files changed, 77 insertions(+), 51 deletions(-) (limited to 'fs') diff --git a/drivers/mtd/chips/cfi_cmdset_0001.c b/drivers/mtd/chips/cfi_cmdset_0001.c index e812df607a5c..fcd1aeccdf93 100644 --- a/drivers/mtd/chips/cfi_cmdset_0001.c +++ b/drivers/mtd/chips/cfi_cmdset_0001.c @@ -82,9 +82,8 @@ static struct mtd_info *cfi_intelext_setup (struct mtd_info *); static int cfi_intelext_partition_fixup(struct mtd_info *, struct cfi_private **); static int cfi_intelext_point (struct mtd_info *mtd, loff_t from, size_t len, - size_t *retlen, u_char **mtdbuf); -static void cfi_intelext_unpoint (struct mtd_info *mtd, u_char *addr, loff_t from, - size_t len); + size_t *retlen, void **virt, resource_size_t *phys); +static void cfi_intelext_unpoint(struct mtd_info *mtd, loff_t from, size_t len); static int chip_ready (struct map_info *map, struct flchip *chip, unsigned long adr, int mode); static int get_chip(struct map_info *map, struct flchip *chip, unsigned long adr, int mode); @@ -1240,7 +1239,8 @@ static int do_point_onechip (struct map_info *map, struct flchip *chip, loff_t a return ret; } -static int cfi_intelext_point (struct mtd_info *mtd, loff_t from, size_t len, size_t *retlen, u_char **mtdbuf) +static int cfi_intelext_point(struct mtd_info *mtd, loff_t from, size_t len, + size_t *retlen, void **virt, resource_size_t *phys) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -1257,8 +1257,10 @@ static int cfi_intelext_point (struct mtd_info *mtd, loff_t from, size_t len, si chipnum = (from >> cfi->chipshift); ofs = from - (chipnum << cfi->chipshift); - *mtdbuf = (void *)map->virt + cfi->chips[chipnum].start + ofs; + *virt = map->virt + cfi->chips[chipnum].start + ofs; *retlen = 0; + if (phys) + *phys = map->phys + cfi->chips[chipnum].start + ofs; while (len) { unsigned long thislen; @@ -1291,7 +1293,7 @@ static int cfi_intelext_point (struct mtd_info *mtd, loff_t from, size_t len, si return 0; } -static void cfi_intelext_unpoint (struct mtd_info *mtd, u_char *addr, loff_t from, size_t len) +static void cfi_intelext_unpoint(struct mtd_info *mtd, loff_t from, size_t len) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; diff --git a/drivers/mtd/devices/mtdram.c b/drivers/mtd/devices/mtdram.c index bf485ff49457..0399be178620 100644 --- a/drivers/mtd/devices/mtdram.c +++ b/drivers/mtd/devices/mtdram.c @@ -48,18 +48,21 @@ static int ram_erase(struct mtd_info *mtd, struct erase_info *instr) } static int ram_point(struct mtd_info *mtd, loff_t from, size_t len, - size_t *retlen, u_char **mtdbuf) + size_t *retlen, void **virt, resource_size_t *phys) { if (from + len > mtd->size) return -EINVAL; - *mtdbuf = mtd->priv + from; + /* can we return a physical address with this driver? */ + if (phys) + return -EINVAL; + + *virt = mtd->priv + from; *retlen = len; return 0; } -static void ram_unpoint(struct mtd_info *mtd, u_char * addr, loff_t from, - size_t len) +static void ram_unpoint(struct mtd_info *mtd, loff_t from, size_t len) { } diff --git a/drivers/mtd/devices/phram.c b/drivers/mtd/devices/phram.c index 5f960182da95..c7987b1c5e01 100644 --- a/drivers/mtd/devices/phram.c +++ b/drivers/mtd/devices/phram.c @@ -57,20 +57,21 @@ static int phram_erase(struct mtd_info *mtd, struct erase_info *instr) } static int phram_point(struct mtd_info *mtd, loff_t from, size_t len, - size_t *retlen, u_char **mtdbuf) + size_t *retlen, void **virt, resource_size_t *phys) { - u_char *start = mtd->priv; - if (from + len > mtd->size) return -EINVAL; - *mtdbuf = start + from; + /* can we return a physical address with this driver? */ + if (phys) + return -EINVAL; + + *virt = mtd->priv + from; *retlen = len; return 0; } -static void phram_unpoint(struct mtd_info *mtd, u_char *addr, loff_t from, - size_t len) +static void phram_unpoint(struct mtd_info *mtd, loff_t from, size_t len) { } diff --git a/drivers/mtd/devices/pmc551.c b/drivers/mtd/devices/pmc551.c index 7060a0895ce2..bc9981749064 100644 --- a/drivers/mtd/devices/pmc551.c +++ b/drivers/mtd/devices/pmc551.c @@ -134,7 +134,8 @@ static int pmc551_erase(struct mtd_info *mtd, struct erase_info *instr) eoff_lo = end & (priv->asize - 1); soff_lo = instr->addr & (priv->asize - 1); - pmc551_point(mtd, instr->addr, instr->len, &retlen, &ptr); + pmc551_point(mtd, instr->addr, instr->len, &retlen, + (void **)&ptr, NULL); if (soff_hi == eoff_hi || mtd->size == priv->asize) { /* The whole thing fits within one access, so just one shot @@ -154,7 +155,8 @@ static int pmc551_erase(struct mtd_info *mtd, struct erase_info *instr) } soff_hi += priv->asize; pmc551_point(mtd, (priv->base_map0 | soff_hi), - priv->asize, &retlen, &ptr); + priv->asize, &retlen, + (void **)&ptr, NULL); } memset(ptr, 0xff, eoff_lo); } @@ -170,7 +172,7 @@ static int pmc551_erase(struct mtd_info *mtd, struct erase_info *instr) } static int pmc551_point(struct mtd_info *mtd, loff_t from, size_t len, - size_t * retlen, u_char ** mtdbuf) + size_t *retlen, void **virt, resource_size_t *phys) { struct mypriv *priv = mtd->priv; u32 soff_hi; @@ -188,6 +190,10 @@ static int pmc551_point(struct mtd_info *mtd, loff_t from, size_t len, return -EINVAL; } + /* can we return a physical address with this driver? */ + if (phys) + return -EINVAL; + soff_hi = from & ~(priv->asize - 1); soff_lo = from & (priv->asize - 1); @@ -198,13 +204,12 @@ static int pmc551_point(struct mtd_info *mtd, loff_t from, size_t len, priv->curr_map0 = soff_hi; } - *mtdbuf = priv->start + soff_lo; + *virt = priv->start + soff_lo; *retlen = len; return 0; } -static void pmc551_unpoint(struct mtd_info *mtd, u_char * addr, loff_t from, - size_t len) +static void pmc551_unpoint(struct mtd_info *mtd, loff_t from, size_t len) { #ifdef CONFIG_MTD_PMC551_DEBUG printk(KERN_DEBUG "pmc551_unpoint()\n"); @@ -242,7 +247,7 @@ static int pmc551_read(struct mtd_info *mtd, loff_t from, size_t len, soff_lo = from & (priv->asize - 1); eoff_lo = end & (priv->asize - 1); - pmc551_point(mtd, from, len, retlen, &ptr); + pmc551_point(mtd, from, len, retlen, (void **)&ptr, NULL); if (soff_hi == eoff_hi) { /* The whole thing fits within one access, so just one shot @@ -263,7 +268,8 @@ static int pmc551_read(struct mtd_info *mtd, loff_t from, size_t len, goto out; } soff_hi += priv->asize; - pmc551_point(mtd, soff_hi, priv->asize, retlen, &ptr); + pmc551_point(mtd, soff_hi, priv->asize, retlen, + (void **)&ptr, NULL); } memcpy(copyto, ptr, eoff_lo); copyto += eoff_lo; @@ -308,7 +314,7 @@ static int pmc551_write(struct mtd_info *mtd, loff_t to, size_t len, soff_lo = to & (priv->asize - 1); eoff_lo = end & (priv->asize - 1); - pmc551_point(mtd, to, len, retlen, &ptr); + pmc551_point(mtd, to, len, retlen, (void **)&ptr, NULL); if (soff_hi == eoff_hi) { /* The whole thing fits within one access, so just one shot @@ -329,7 +335,8 @@ static int pmc551_write(struct mtd_info *mtd, loff_t to, size_t len, goto out; } soff_hi += priv->asize; - pmc551_point(mtd, soff_hi, priv->asize, retlen, &ptr); + pmc551_point(mtd, soff_hi, priv->asize, retlen, + (void **)&ptr, NULL); } memcpy(ptr, copyfrom, eoff_lo); copyfrom += eoff_lo; diff --git a/drivers/mtd/devices/slram.c b/drivers/mtd/devices/slram.c index d293add1857c..cb86db746f28 100644 --- a/drivers/mtd/devices/slram.c +++ b/drivers/mtd/devices/slram.c @@ -76,8 +76,9 @@ static char *map; static slram_mtd_list_t *slram_mtdlist = NULL; static int slram_erase(struct mtd_info *, struct erase_info *); -static int slram_point(struct mtd_info *, loff_t, size_t, size_t *, u_char **); -static void slram_unpoint(struct mtd_info *, u_char *, loff_t, size_t); +static int slram_point(struct mtd_info *, loff_t, size_t, size_t *, void **, + resource_size_t *); +static void slram_unpoint(struct mtd_info *, loff_t, size_t); static int slram_read(struct mtd_info *, loff_t, size_t, size_t *, u_char *); static int slram_write(struct mtd_info *, loff_t, size_t, size_t *, const u_char *); @@ -104,19 +105,23 @@ static int slram_erase(struct mtd_info *mtd, struct erase_info *instr) } static int slram_point(struct mtd_info *mtd, loff_t from, size_t len, - size_t *retlen, u_char **mtdbuf) + size_t *retlen, void **virt, resource_size_t *phys) { slram_priv_t *priv = mtd->priv; + /* can we return a physical address with this driver? */ + if (phys) + return -EINVAL; + if (from + len > mtd->size) return -EINVAL; - *mtdbuf = priv->start + from; + *virt = priv->start + from; *retlen = len; return(0); } -static void slram_unpoint(struct mtd_info *mtd, u_char *addr, loff_t from, size_t len) +static void slram_unpoint(struct mtd_info *mtd, loff_t from, size_t len) { } diff --git a/drivers/mtd/maps/uclinux.c b/drivers/mtd/maps/uclinux.c index 14ffb1a9302a..c42f4b83f686 100644 --- a/drivers/mtd/maps/uclinux.c +++ b/drivers/mtd/maps/uclinux.c @@ -40,10 +40,12 @@ struct mtd_partition uclinux_romfs[] = { /****************************************************************************/ int uclinux_point(struct mtd_info *mtd, loff_t from, size_t len, - size_t *retlen, u_char **mtdbuf) + size_t *retlen, void **virt, resource_size_t *phys) { struct map_info *map = mtd->priv; - *mtdbuf = (u_char *) (map->virt + ((int) from)); + *virt = map->virt + from; + if (phys) + *phys = map->phys + from; *retlen = len; return(0); } diff --git a/drivers/mtd/mtdpart.c b/drivers/mtd/mtdpart.c index c66902df3171..07c701169344 100644 --- a/drivers/mtd/mtdpart.c +++ b/drivers/mtd/mtdpart.c @@ -68,7 +68,7 @@ static int part_read (struct mtd_info *mtd, loff_t from, size_t len, } static int part_point (struct mtd_info *mtd, loff_t from, size_t len, - size_t *retlen, u_char **buf) + size_t *retlen, void **virt, resource_size_t *phys) { struct mtd_part *part = PART(mtd); if (from >= mtd->size) @@ -76,14 +76,14 @@ static int part_point (struct mtd_info *mtd, loff_t from, size_t len, else if (from + len > mtd->size) len = mtd->size - from; return part->master->point (part->master, from + part->offset, - len, retlen, buf); + len, retlen, virt, phys); } -static void part_unpoint (struct mtd_info *mtd, u_char *addr, loff_t from, size_t len) +static void part_unpoint(struct mtd_info *mtd, loff_t from, size_t len) { struct mtd_part *part = PART(mtd); - part->master->unpoint (part->master, addr, from + part->offset, len); + part->master->unpoint(part->master, from + part->offset, len); } static int part_read_oob(struct mtd_info *mtd, loff_t from, diff --git a/fs/jffs2/erase.c b/fs/jffs2/erase.c index 5e86f43616a1..dddb2a6c9e2c 100644 --- a/fs/jffs2/erase.c +++ b/fs/jffs2/erase.c @@ -332,7 +332,8 @@ static int jffs2_block_check_erase(struct jffs2_sb_info *c, struct jffs2_erasebl if (c->mtd->point) { unsigned long *wordebuf; - ret = c->mtd->point(c->mtd, jeb->offset, c->sector_size, &retlen, (unsigned char **)&ebuf); + ret = c->mtd->point(c->mtd, jeb->offset, c->sector_size, + &retlen, &ebuf, NULL); if (ret) { D1(printk(KERN_DEBUG "MTD point failed %d\n", ret)); goto do_flash_read; @@ -340,7 +341,7 @@ static int jffs2_block_check_erase(struct jffs2_sb_info *c, struct jffs2_erasebl if (retlen < c->sector_size) { /* Don't muck about if it won't let us point to the whole erase sector */ D1(printk(KERN_DEBUG "MTD point returned len too short: 0x%zx\n", retlen)); - c->mtd->unpoint(c->mtd, ebuf, jeb->offset, retlen); + c->mtd->unpoint(c->mtd, jeb->offset, retlen); goto do_flash_read; } wordebuf = ebuf-sizeof(*wordebuf); @@ -349,7 +350,7 @@ static int jffs2_block_check_erase(struct jffs2_sb_info *c, struct jffs2_erasebl if (*++wordebuf != ~0) break; } while(--retlen); - c->mtd->unpoint(c->mtd, ebuf, jeb->offset, c->sector_size); + c->mtd->unpoint(c->mtd, jeb->offset, c->sector_size); if (retlen) { printk(KERN_WARNING "Newly-erased block contained word 0x%lx at offset 0x%08tx\n", *wordebuf, jeb->offset + c->sector_size-retlen*sizeof(*wordebuf)); diff --git a/fs/jffs2/readinode.c b/fs/jffs2/readinode.c index 9fc4833c117c..6ca08ad887c0 100644 --- a/fs/jffs2/readinode.c +++ b/fs/jffs2/readinode.c @@ -63,10 +63,11 @@ static int check_node_data(struct jffs2_sb_info *c, struct jffs2_tmp_dnode_info /* TODO: instead, incapsulate point() stuff to jffs2_flash_read(), * adding and jffs2_flash_read_end() interface. */ if (c->mtd->point) { - err = c->mtd->point(c->mtd, ofs, len, &retlen, &buffer); + err = c->mtd->point(c->mtd, ofs, len, &retlen, + (void **)&buffer, NULL); if (!err && retlen < len) { JFFS2_WARNING("MTD point returned len too short: %zu instead of %u.\n", retlen, tn->csize); - c->mtd->unpoint(c->mtd, buffer, ofs, retlen); + c->mtd->unpoint(c->mtd, ofs, retlen); } else if (err) JFFS2_WARNING("MTD point failed: error code %d.\n", err); else @@ -100,7 +101,7 @@ static int check_node_data(struct jffs2_sb_info *c, struct jffs2_tmp_dnode_info kfree(buffer); #ifndef __ECOS else - c->mtd->unpoint(c->mtd, buffer, ofs, len); + c->mtd->unpoint(c->mtd, ofs, len); #endif if (crc != tn->data_crc) { @@ -136,7 +137,7 @@ free_out: kfree(buffer); #ifndef __ECOS else - c->mtd->unpoint(c->mtd, buffer, ofs, len); + c->mtd->unpoint(c->mtd, ofs, len); #endif return err; } diff --git a/fs/jffs2/scan.c b/fs/jffs2/scan.c index 8c1e692bef79..1d437de1e9a8 100644 --- a/fs/jffs2/scan.c +++ b/fs/jffs2/scan.c @@ -97,11 +97,12 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) size_t pointlen; if (c->mtd->point) { - ret = c->mtd->point (c->mtd, 0, c->mtd->size, &pointlen, &flashbuf); + ret = c->mtd->point(c->mtd, 0, c->mtd->size, &pointlen, + (void **)&flashbuf, NULL); if (!ret && pointlen < c->mtd->size) { /* Don't muck about if it won't let us point to the whole flash */ D1(printk(KERN_DEBUG "MTD point returned len too short: 0x%zx\n", pointlen)); - c->mtd->unpoint(c->mtd, flashbuf, 0, pointlen); + c->mtd->unpoint(c->mtd, 0, pointlen); flashbuf = NULL; } if (ret) @@ -267,7 +268,7 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) kfree(flashbuf); #ifndef __ECOS else - c->mtd->unpoint(c->mtd, flashbuf, 0, c->mtd->size); + c->mtd->unpoint(c->mtd, 0, c->mtd->size); #endif if (s) kfree(s); diff --git a/include/linux/mtd/mtd.h b/include/linux/mtd/mtd.h index 0a13bb35f044..245f9098e171 100644 --- a/include/linux/mtd/mtd.h +++ b/include/linux/mtd/mtd.h @@ -143,10 +143,12 @@ struct mtd_info { int (*erase) (struct mtd_info *mtd, struct erase_info *instr); /* This stuff for eXecute-In-Place */ - int (*point) (struct mtd_info *mtd, loff_t from, size_t len, size_t *retlen, u_char **mtdbuf); + /* phys is optional and may be set to NULL */ + int (*point) (struct mtd_info *mtd, loff_t from, size_t len, + size_t *retlen, void **virt, resource_size_t *phys); /* We probably shouldn't allow XIP if the unpoint isn't a NULL */ - void (*unpoint) (struct mtd_info *mtd, u_char * addr, loff_t from, size_t len); + void (*unpoint) (struct mtd_info *mtd, loff_t from, size_t len); int (*read) (struct mtd_info *mtd, loff_t from, size_t len, size_t *retlen, u_char *buf); diff --git a/include/linux/mtd/pmc551.h b/include/linux/mtd/pmc551.h index a7f6d20ad407..5cc070c24d88 100644 --- a/include/linux/mtd/pmc551.h +++ b/include/linux/mtd/pmc551.h @@ -36,8 +36,9 @@ struct mypriv { * Function Prototypes */ static int pmc551_erase(struct mtd_info *, struct erase_info *); -static void pmc551_unpoint(struct mtd_info *, u_char *, loff_t, size_t); -static int pmc551_point (struct mtd_info *mtd, loff_t from, size_t len, size_t *retlen, u_char **mtdbuf); +static void pmc551_unpoint(struct mtd_info *, loff_t, size_t); +static int pmc551_point(struct mtd_info *mtd, loff_t from, size_t len, + size_t *retlen, void **virt, resource_size_t *phys); static int pmc551_read(struct mtd_info *, loff_t, size_t, size_t *, u_char *); static int pmc551_write(struct mtd_info *, loff_t, size_t, size_t *, const u_char *); -- cgit v1.2.3 From 78e92b99ec4eb73755abd4e357b0b211eadafd88 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Fri, 2 May 2008 04:12:41 -0700 Subject: netns: assign PDE->data before gluing entry into /proc tree In this unfortunate case, proc_mkdir_mode wrapper can't be used anymore and this is no way to reuse proc_create_data due to nlinks assignment. So, copy the code from proc_mkdir and assign PDE->data at the appropriate moment. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- fs/proc/generic.c | 17 +++++++++++++++++ fs/proc/proc_net.c | 11 ----------- 2 files changed, 17 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 9d53b39a9cf8..43e54e86cefd 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -641,6 +641,23 @@ struct proc_dir_entry *proc_mkdir_mode(const char *name, mode_t mode, return ent; } +struct proc_dir_entry *proc_net_mkdir(struct net *net, const char *name, + struct proc_dir_entry *parent) +{ + struct proc_dir_entry *ent; + + ent = __proc_create(&parent, name, S_IFDIR | S_IRUGO | S_IXUGO, 2); + if (ent) { + ent->data = net; + if (proc_register(parent, ent) < 0) { + kfree(ent); + ent = NULL; + } + } + return ent; +} +EXPORT_SYMBOL_GPL(proc_net_mkdir); + struct proc_dir_entry *proc_mkdir(const char *name, struct proc_dir_entry *parent) { diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index 13cd7835d0df..83f357b30d71 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -159,17 +159,6 @@ struct net *get_proc_net(const struct inode *inode) } EXPORT_SYMBOL_GPL(get_proc_net); -struct proc_dir_entry *proc_net_mkdir(struct net *net, const char *name, - struct proc_dir_entry *parent) -{ - struct proc_dir_entry *pde; - pde = proc_mkdir_mode(name, S_IRUGO | S_IXUGO, parent); - if (pde != NULL) - pde->data = net; - return pde; -} -EXPORT_SYMBOL_GPL(proc_net_mkdir); - static __net_init int proc_net_ns_init(struct net *net) { struct proc_dir_entry *netd, *net_statd; -- cgit v1.2.3 From d35c7b0e54a596c5a8134d75999b7f391a9c6550 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Sat, 3 May 2008 15:10:37 -0400 Subject: unified (weak) sys_pipe implementation This replaces the duplicated arch-specific versions of "sys_pipe()" with one unified implementation. This removes almost 250 lines of duplicated code. It's marked __weak, so that *if* an architecture wants to override the default implementation it can do so by simply having its own replacement version, since many architectures use alternate calling conventions for the 'pipe()' system call for legacy reasons (ie traditional UNIX implementations often return the two file descriptors in registers) I still haven't changed the cris version even though Linus says the BKL isn't needed. The arch maintainer can easily do it if there are really no obstacles. Signed-off-by: Ulrich Drepper Signed-off-by: Linus Torvalds --- arch/arm/kernel/sys_arm.c | 17 ----------------- arch/avr32/kernel/sys_avr32.c | 13 ------------- arch/blackfin/kernel/sys_bfin.c | 17 ----------------- arch/frv/kernel/sys_frv.c | 17 ----------------- arch/h8300/kernel/sys_h8300.c | 17 ----------------- arch/m68k/kernel/sys_m68k.c | 17 ----------------- arch/m68knommu/kernel/sys_m68k.c | 17 ----------------- arch/mn10300/kernel/sys_mn10300.c | 17 ----------------- arch/parisc/kernel/sys_parisc.c | 13 ------------- arch/powerpc/kernel/syscalls.c | 17 ----------------- arch/s390/kernel/sys_s390.c | 17 ----------------- arch/sh/kernel/sys_sh64.c | 17 ----------------- arch/um/kernel/syscall.c | 17 ----------------- arch/v850/kernel/syscalls.c | 17 ----------------- arch/x86/kernel/sys_i386_32.c | 17 ----------------- arch/x86/kernel/sys_x86_64.c | 17 ----------------- fs/pipe.c | 17 +++++++++++++++++ include/asm-powerpc/syscalls.h | 2 +- 18 files changed, 18 insertions(+), 265 deletions(-) (limited to 'fs') diff --git a/arch/arm/kernel/sys_arm.c b/arch/arm/kernel/sys_arm.c index 9bd1870d980e..0128687ba0f7 100644 --- a/arch/arm/kernel/sys_arm.c +++ b/arch/arm/kernel/sys_arm.c @@ -34,23 +34,6 @@ extern unsigned long do_mremap(unsigned long addr, unsigned long old_len, unsigned long new_len, unsigned long flags, unsigned long new_addr); -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - /* common code for old and new mmaps */ inline long do_mmap2( unsigned long addr, unsigned long len, diff --git a/arch/avr32/kernel/sys_avr32.c b/arch/avr32/kernel/sys_avr32.c index 8deb6003ee62..8e8911e55c8f 100644 --- a/arch/avr32/kernel/sys_avr32.c +++ b/arch/avr32/kernel/sys_avr32.c @@ -14,19 +14,6 @@ #include #include -asmlinkage int sys_pipe(unsigned long __user *filedes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(filedes, fd, sizeof(fd))) - error = -EFAULT; - } - return error; -} - asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, off_t offset) diff --git a/arch/blackfin/kernel/sys_bfin.c b/arch/blackfin/kernel/sys_bfin.c index efb7b25a2633..fce49d7cf001 100644 --- a/arch/blackfin/kernel/sys_bfin.c +++ b/arch/blackfin/kernel/sys_bfin.c @@ -45,23 +45,6 @@ #include #include -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2 * sizeof(int))) - error = -EFAULT; - } - return error; -} - /* common code for old and new mmaps */ static inline long do_mmap2(unsigned long addr, unsigned long len, diff --git a/arch/frv/kernel/sys_frv.c b/arch/frv/kernel/sys_frv.c index 04c6b1677ccf..49b2cf2c38f3 100644 --- a/arch/frv/kernel/sys_frv.c +++ b/arch/frv/kernel/sys_frv.c @@ -28,23 +28,6 @@ #include #include -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -asmlinkage long sys_pipe(unsigned long __user * fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff) diff --git a/arch/h8300/kernel/sys_h8300.c b/arch/h8300/kernel/sys_h8300.c index 00608be6d567..2745656dcc52 100644 --- a/arch/h8300/kernel/sys_h8300.c +++ b/arch/h8300/kernel/sys_h8300.c @@ -27,23 +27,6 @@ #include #include -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long * fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - /* common code for old and new mmaps */ static inline long do_mmap2( unsigned long addr, unsigned long len, diff --git a/arch/m68k/kernel/sys_m68k.c b/arch/m68k/kernel/sys_m68k.c index e892f17ba3fa..7f54efaf60bb 100644 --- a/arch/m68k/kernel/sys_m68k.c +++ b/arch/m68k/kernel/sys_m68k.c @@ -30,23 +30,6 @@ #include #include -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long __user * fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - /* common code for old and new mmaps */ static inline long do_mmap2( unsigned long addr, unsigned long len, diff --git a/arch/m68knommu/kernel/sys_m68k.c b/arch/m68knommu/kernel/sys_m68k.c index 65f7a95f056e..700281638629 100644 --- a/arch/m68knommu/kernel/sys_m68k.c +++ b/arch/m68knommu/kernel/sys_m68k.c @@ -28,23 +28,6 @@ #include #include -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long * fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - /* common code for old and new mmaps */ static inline long do_mmap2( unsigned long addr, unsigned long len, diff --git a/arch/mn10300/kernel/sys_mn10300.c b/arch/mn10300/kernel/sys_mn10300.c index 5f17a1ebc825..bca5a84dc72c 100644 --- a/arch/mn10300/kernel/sys_mn10300.c +++ b/arch/mn10300/kernel/sys_mn10300.c @@ -28,23 +28,6 @@ #define MIN_MAP_ADDR PAGE_SIZE /* minimum fixed mmap address */ -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way Unix traditionally does this, though. - */ -asmlinkage long sys_pipe(unsigned long __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2 * sizeof(int))) - error = -EFAULT; - } - return error; -} - /* * memory mapping syscall */ diff --git a/arch/parisc/kernel/sys_parisc.c b/arch/parisc/kernel/sys_parisc.c index 4f589216b39e..71b31957c8f1 100644 --- a/arch/parisc/kernel/sys_parisc.c +++ b/arch/parisc/kernel/sys_parisc.c @@ -33,19 +33,6 @@ #include #include -int sys_pipe(int __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - static unsigned long get_unshared_area(unsigned long addr, unsigned long len) { struct vm_area_struct *vma; diff --git a/arch/powerpc/kernel/syscalls.c b/arch/powerpc/kernel/syscalls.c index e722a4eeb5d0..4fe69ca24481 100644 --- a/arch/powerpc/kernel/syscalls.c +++ b/arch/powerpc/kernel/syscalls.c @@ -136,23 +136,6 @@ int sys_ipc(uint call, int first, unsigned long second, long third, return ret; } -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -int sys_pipe(int __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - static inline unsigned long do_mmap2(unsigned long addr, size_t len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long off, int shift) diff --git a/arch/s390/kernel/sys_s390.c b/arch/s390/kernel/sys_s390.c index 988d0d64c2c8..5fdb799062b7 100644 --- a/arch/s390/kernel/sys_s390.c +++ b/arch/s390/kernel/sys_s390.c @@ -32,23 +32,6 @@ #include #include "entry.h" -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way Unix traditionally does this, though. - */ -asmlinkage long sys_pipe(unsigned long __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - /* common code for old and new mmaps */ static inline long do_mmap2( unsigned long addr, unsigned long len, diff --git a/arch/sh/kernel/sys_sh64.c b/arch/sh/kernel/sys_sh64.c index 578004d71e02..91fb8445a5a0 100644 --- a/arch/sh/kernel/sys_sh64.c +++ b/arch/sh/kernel/sys_sh64.c @@ -30,23 +30,6 @@ #include #include -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way Unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long * fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - /* * Do a system call from kernel instead of calling sys_execve so we * end up with proper pt_regs. diff --git a/arch/um/kernel/syscall.c b/arch/um/kernel/syscall.c index 9cffc628a37e..128ee85bc8d9 100644 --- a/arch/um/kernel/syscall.c +++ b/arch/um/kernel/syscall.c @@ -73,23 +73,6 @@ long old_mmap(unsigned long addr, unsigned long len, out: return err; } -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -long sys_pipe(unsigned long __user * fildes) -{ - int fd[2]; - long error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, sizeof(fd))) - error = -EFAULT; - } - return error; -} - long sys_uname(struct old_utsname __user * name) { diff --git a/arch/v850/kernel/syscalls.c b/arch/v850/kernel/syscalls.c index 003db9c8c44a..1a83daf8e24f 100644 --- a/arch/v850/kernel/syscalls.c +++ b/arch/v850/kernel/syscalls.c @@ -132,23 +132,6 @@ sys_ipc (uint call, int first, int second, int third, void *ptr, long fifth) return ret; } -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -int sys_pipe (int *fildes) -{ - int fd[2]; - int error; - - error = do_pipe (fd); - if (!error) { - if (copy_to_user (fildes, fd, 2*sizeof (int))) - error = -EFAULT; - } - return error; -} - static inline unsigned long do_mmap2 (unsigned long addr, size_t len, unsigned long prot, unsigned long flags, diff --git a/arch/x86/kernel/sys_i386_32.c b/arch/x86/kernel/sys_i386_32.c index a86d26f036e1..d2ab52cc1d6b 100644 --- a/arch/x86/kernel/sys_i386_32.c +++ b/arch/x86/kernel/sys_i386_32.c @@ -22,23 +22,6 @@ #include #include -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way Unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long __user * fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff) diff --git a/arch/x86/kernel/sys_x86_64.c b/arch/x86/kernel/sys_x86_64.c index bd802a5e1aa3..3b360ef33817 100644 --- a/arch/x86/kernel/sys_x86_64.c +++ b/arch/x86/kernel/sys_x86_64.c @@ -17,23 +17,6 @@ #include #include -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way Unix traditionally does this, though. - */ -asmlinkage long sys_pipe(int __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - asmlinkage long sys_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long off) { diff --git a/fs/pipe.c b/fs/pipe.c index f73492b6817e..3499f9ff6316 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -1075,6 +1075,23 @@ int do_pipe(int *fd) return error; } +/* + * sys_pipe() is the normal C calling standard for creating + * a pipe. It's not the way Unix traditionally does this, though. + */ +asmlinkage long __weak sys_pipe(int __user *fildes) +{ + int fd[2]; + int error; + + error = do_pipe(fd); + if (!error) { + if (copy_to_user(fildes, fd, sizeof(fd))) + error = -EFAULT; + } + return error; +} + /* * pipefs should _never_ be mounted by userland - too much of security hassle, * no real gain from having the whole whorehouse mounted. So we don't need diff --git a/include/asm-powerpc/syscalls.h b/include/asm-powerpc/syscalls.h index b3ca41fc8bb1..2b8a458f990a 100644 --- a/include/asm-powerpc/syscalls.h +++ b/include/asm-powerpc/syscalls.h @@ -30,7 +30,7 @@ asmlinkage int sys_fork(unsigned long p1, unsigned long p2, asmlinkage int sys_vfork(unsigned long p1, unsigned long p2, unsigned long p3, unsigned long p4, unsigned long p5, unsigned long p6, struct pt_regs *regs); -asmlinkage int sys_pipe(int __user *fildes); +asmlinkage long sys_pipe(int __user *fildes); asmlinkage long sys_rt_sigaction(int sig, const struct sigaction __user *act, struct sigaction __user *oact, size_t sigsetsize); -- cgit v1.2.3 From eb28062f131b0a1da32b2554fd819af5221040de Mon Sep 17 00:00:00 2001 From: Bryan Wu Date: Sun, 4 May 2008 23:12:55 +0800 Subject: task_nommu: fix compile failing bug because of spilt file.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit CC fs/proc/task_nommu.o fs/proc/task_nommu.c: In function ‘task_mem’: fs/proc/task_nommu.c:55: error: dereferencing pointer to incomplete type make[2]: *** [fs/proc/task_nommu.o] Error 1 make[1]: *** [fs/proc] Error 2 make: *** [fs] Error 2 Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds --- fs/proc/task_nommu.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c index 4b733f108455..4b4f9cc2f186 100644 --- a/fs/proc/task_nommu.c +++ b/fs/proc/task_nommu.c @@ -1,6 +1,7 @@ #include #include +#include #include #include #include -- cgit v1.2.3