From 0533400b7813df6c22a171499434d30bd57e799c Mon Sep 17 00:00:00 2001 From: Stoyan Gaydarov Date: Mon, 7 Jul 2008 07:45:59 -0500 Subject: [JFFS2] Use .unlocked_ioctl This changes the .ioctl to the .unlocked_ioctl version. Signed-off-by: Stoyan Gaydarov Signed-off-by: David Woodhouse --- fs/jffs2/dir.c | 2 +- fs/jffs2/file.c | 2 +- fs/jffs2/ioctl.c | 3 +-- fs/jffs2/os-linux.h | 2 +- 4 files changed, 4 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index c0c141f6fde1..cd219ef55254 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -38,7 +38,7 @@ const struct file_operations jffs2_dir_operations = { .read = generic_read_dir, .readdir = jffs2_readdir, - .ioctl = jffs2_ioctl, + .unlocked_ioctl=jffs2_ioctl, .fsync = jffs2_fsync }; diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c index 5e920343b2c5..5a98aa87c853 100644 --- a/fs/jffs2/file.c +++ b/fs/jffs2/file.c @@ -46,7 +46,7 @@ const struct file_operations jffs2_file_operations = .aio_read = generic_file_aio_read, .write = do_sync_write, .aio_write = generic_file_aio_write, - .ioctl = jffs2_ioctl, + .unlocked_ioctl=jffs2_ioctl, .mmap = generic_file_readonly_mmap, .fsync = jffs2_fsync, .splice_read = generic_file_splice_read, diff --git a/fs/jffs2/ioctl.c b/fs/jffs2/ioctl.c index e2177210f621..9d41f43e47bb 100644 --- a/fs/jffs2/ioctl.c +++ b/fs/jffs2/ioctl.c @@ -12,8 +12,7 @@ #include #include "nodelist.h" -int jffs2_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, - unsigned long arg) +long jffs2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { /* Later, this will provide for lsattr.jffs2 and chattr.jffs2, which will include compression support etc. */ diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h index 2cc866cf134f..5e194a5c8e29 100644 --- a/fs/jffs2/os-linux.h +++ b/fs/jffs2/os-linux.h @@ -167,7 +167,7 @@ int jffs2_fsync(struct file *, struct dentry *, int); int jffs2_do_readpage_unlock (struct inode *inode, struct page *pg); /* ioctl.c */ -int jffs2_ioctl(struct inode *, struct file *, unsigned int, unsigned long); +long jffs2_ioctl(struct file *, unsigned int, unsigned long); /* symlink.c */ extern const struct inode_operations jffs2_symlink_inode_operations; -- cgit v1.2.3 From 04e1e0cccade330ab3715ce59234f7e3b087e246 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Tue, 22 Jul 2008 13:04:18 +0000 Subject: [CIFS] Fix compiler warning on 64-bit Signed-off-by: Steve French --- fs/cifs/asn1.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/cifs/asn1.c b/fs/cifs/asn1.c index f58e41d3ba48..42765465701c 100644 --- a/fs/cifs/asn1.c +++ b/fs/cifs/asn1.c @@ -400,7 +400,7 @@ asn1_oid_decode(struct asn1_ctx *ctx, size = eoc - ctx->pointer + 1; /* first subid actually encodes first two subids */ - if (size < 2 || size > ULONG_MAX/sizeof(unsigned long)) + if (size < 2 || size > UINT_MAX/sizeof(unsigned long)) return 0; *oid = kmalloc(size * sizeof(unsigned long), GFP_ATOMIC); -- cgit v1.2.3 From 09e50d55a9490e9b7a6fdfbf8fc078924b25ecb5 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Wed, 23 Jul 2008 10:11:19 -0400 Subject: lockdep: annotate cifs in-kernel sockets Put CIFS sockets in their own class to avoid some lockdep warnings. CIFS sockets are not exposed to user-space, and so are not subject to the same deadlock scenarios. A similar change was made a couple of years ago for RPC sockets in commit ed07536ed6731775219c1df7fa26a7588753e693. This patch should prevent lockdep false-positives like this one: ======================================================= [ INFO: possible circular locking dependency detected ] 2.6.18-98.el5.jtltest.38.bz456320.1debug #1 ------------------------------------------------------- test5/2483 is trying to acquire lock: (sk_lock-AF_INET){--..}, at: [] tcp_sendmsg+0x1c/0xb2f but task is already holding lock: (&inode->i_alloc_sem){--..}, at: [] notify_change+0xf5/0x2e0 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #3 (&inode->i_alloc_sem){--..}: [] __lock_acquire+0x9a9/0xadf [] lock_acquire+0x55/0x70 [] notify_change+0xf5/0x2e0 [] down_write+0x3c/0x68 [] notify_change+0xf5/0x2e0 [] do_truncate+0x50/0x6b [] get_write_access+0x40/0x46 [] may_open+0x1d3/0x22e [] open_namei+0x2c6/0x6dd [] do_filp_open+0x1c/0x38 [] _spin_unlock+0x17/0x20 [] get_unused_fd+0xf9/0x107 [] do_sys_open+0x44/0xbe [] system_call+0x7e/0x83 [] 0xffffffffffffffff -> #2 (&sysfs_inode_imutex_key){--..}: [] __lock_acquire+0x9a9/0xadf [] create_dir+0x26/0x1d7 [] lock_acquire+0x55/0x70 [] create_dir+0x26/0x1d7 [] mutex_lock_nested+0x104/0x29c [] __lock_acquire+0x9ca/0xadf [] create_dir+0x26/0x1d7 [] sysfs_create_dir+0x58/0x76 [] kobject_add+0xdb/0x198 [] class_device_add+0xb2/0x465 [] kobject_get+0x12/0x17 [] register_netdevice+0x270/0x33e [] register_netdev+0x59/0x67 [] net_olddevs_init+0xb/0xac [] init+0x1f9/0x2fc [] _spin_unlock_irq+0x24/0x27 [] trace_hardirqs_on_thunk+0x35/0x37 [] child_rip+0xa/0x11 [] _spin_unlock_irq+0x24/0x27 [] restore_args+0x0/0x30 [] acpi_ds_init_one_object+0x0/0x80 [] init+0x0/0x2fc [] child_rip+0x0/0x11 [] 0xffffffffffffffff -> #1 (rtnl_mutex){--..}: [] __lock_acquire+0x9a9/0xadf [] ip_mc_leave_group+0x23/0xb7 [] lock_acquire+0x55/0x70 [] ip_mc_leave_group+0x23/0xb7 [] mutex_lock_nested+0x104/0x29c [] ip_mc_leave_group+0x23/0xb7 [] do_ip_setsockopt+0x6d1/0x9bf [] lock_release_holdtime+0x27/0x48 [] lock_release_holdtime+0x27/0x48 [] do_page_fault+0x503/0x835 [] socket_has_perm+0x5b/0x68 [] ip_setsockopt+0x22/0x78 [] sys_setsockopt+0x91/0xb7 [] tracesys+0xd5/0xdf [] 0xffffffffffffffff -> #0 (sk_lock-AF_INET){--..}: [] print_stack_trace+0x59/0x68 [] __lock_acquire+0x8bf/0xadf [] lock_acquire+0x55/0x70 [] tcp_sendmsg+0x1c/0xb2f [] lock_sock+0xd4/0xe4 [] _local_bh_enable+0xcb/0xe0 [] restore_args+0x0/0x30 [] tcp_sendmsg+0x1c/0xb2f [] sock_sendmsg+0xf3/0x110 [] autoremove_wake_function+0x0/0x2e [] kernel_text_address+0x1a/0x26 [] dump_trace+0x211/0x23a [] find_usage_backwards+0x5f/0x88 [] MD5Final+0xaf/0xc2 [cifs] [] cifs_calculate_signature+0x55/0x69 [cifs] [] kernel_sendmsg+0x35/0x47 [] smb_send+0xa3/0x151 [cifs] [] SendReceive+0x1a2/0x448 [cifs] [] __lock_acquire+0x95c/0xadf [] CIFSSMBSetEOF+0x20d/0x25b [cifs] [] cifs_set_file_size+0x110/0x3b7 [cifs] [] cifs_setattr+0x3b2/0x6f6 [cifs] [] notify_change+0xf5/0x2e0 [] notify_change+0x145/0x2e0 [] do_truncate+0x50/0x6b [] get_write_access+0x40/0x46 [] may_open+0x1d3/0x22e [] open_namei+0x2c6/0x6dd [] do_filp_open+0x1c/0x38 [] _spin_unlock+0x17/0x20 [] get_unused_fd+0xf9/0x107 [] do_sys_open+0x44/0xbe [] tracesys+0xd5/0xdf [] 0xffffffffffffffff other info that might help us debug this: 2 locks held by test5/2483: #0: (&inode->i_mutex){--..}, at: [] do_truncate+0x45/0x6b #1: (&inode->i_alloc_sem){--..}, at: [] notify_change+0xf5/0x2e0 stack backtrace: Call Trace: [] print_circular_bug_tail+0x65/0x6e [] print_stack_trace+0x59/0x68 [] __lock_acquire+0x8bf/0xadf [] lock_acquire+0x55/0x70 [] tcp_sendmsg+0x1c/0xb2f [] lock_sock+0xd4/0xe4 [] _local_bh_enable+0xcb/0xe0 [] restore_args+0x0/0x30 [] tcp_sendmsg+0x1c/0xb2f [] sock_sendmsg+0xf3/0x110 [] autoremove_wake_function+0x0/0x2e [] kernel_text_address+0x1a/0x26 [] dump_trace+0x211/0x23a [] find_usage_backwards+0x5f/0x88 [] :cifs:MD5Final+0xaf/0xc2 [] :cifs:cifs_calculate_signature+0x55/0x69 [] kernel_sendmsg+0x35/0x47 [] :cifs:smb_send+0xa3/0x151 [] :cifs:SendReceive+0x1a2/0x448 [] __lock_acquire+0x95c/0xadf [] :cifs:CIFSSMBSetEOF+0x20d/0x25b [] :cifs:cifs_set_file_size+0x110/0x3b7 [] :cifs:cifs_setattr+0x3b2/0x6f6 [] notify_change+0xf5/0x2e0 [] notify_change+0x145/0x2e0 [] do_truncate+0x50/0x6b [] get_write_access+0x40/0x46 [] may_open+0x1d3/0x22e [] open_namei+0x2c6/0x6dd [] do_filp_open+0x1c/0x38 [] _spin_unlock+0x17/0x20 [] get_unused_fd+0xf9/0x107 [] do_sys_open+0x44/0xbe [] tracesys+0xd5/0xdf Signed-off-by: Jeff Layton Acked-by: Peter Zijlstra Signed-off-by: Steve French --- fs/cifs/connect.c | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) (limited to 'fs') diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index e8fa46c7cff2..644462729387 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -1461,6 +1461,39 @@ get_dfs_path(int xid, struct cifsSesInfo *pSesInfo, const char *old_path, return rc; } +#ifdef CONFIG_DEBUG_LOCK_ALLOC +static struct lock_class_key cifs_key[2]; +static struct lock_class_key cifs_slock_key[2]; + +static inline void +cifs_reclassify_socket4(struct socket *sock) +{ + struct sock *sk = sock->sk; + BUG_ON(sock_owned_by_user(sk)); + sock_lock_init_class_and_name(sk, "slock-AF_INET-CIFS", + &cifs_slock_key[0], "sk_lock-AF_INET-CIFS", &cifs_key[0]); +} + +static inline void +cifs_reclassify_socket6(struct socket *sock) +{ + struct sock *sk = sock->sk; + BUG_ON(sock_owned_by_user(sk)); + sock_lock_init_class_and_name(sk, "slock-AF_INET6-CIFS", + &cifs_slock_key[1], "sk_lock-AF_INET6-CIFS", &cifs_key[1]); +} +#else +static inline void +cifs_reclassify_socket4(struct socket *sock) +{ +} + +static inline void +cifs_reclassify_socket6(struct socket *sock) +{ +} +#endif + /* See RFC1001 section 14 on representation of Netbios names */ static void rfc1002mangle(char *target, char *source, unsigned int length) { @@ -1495,6 +1528,7 @@ ipv4_connect(struct sockaddr_in *psin_server, struct socket **csocket, /* BB other socket options to set KEEPALIVE, NODELAY? */ cFYI(1, ("Socket created")); (*csocket)->sk->sk_allocation = GFP_NOFS; + cifs_reclassify_socket4(*csocket); } } @@ -1627,6 +1661,7 @@ ipv6_connect(struct sockaddr_in6 *psin_server, struct socket **csocket) /* BB other socket options to set KEEPALIVE, NODELAY? */ cFYI(1, ("ipv6 Socket created")); (*csocket)->sk->sk_allocation = GFP_NOFS; + cifs_reclassify_socket6(*csocket); } } -- cgit v1.2.3 From 8efdbde647f542ce0d303273df7ad4157caa03d0 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Wed, 23 Jul 2008 21:28:12 +0000 Subject: [CIFS] break ATTR_SIZE changes out into their own function Move the code that handles ATTR_SIZE changes to its own function. This makes for a smaller function and reduces the level of indentation. Signed-off-by: Jeff Layton Signed-off-by: Steve French --- fs/cifs/inode.c | 151 +++++++++++++++++++++++++++++--------------------------- 1 file changed, 78 insertions(+), 73 deletions(-) (limited to 'fs') diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 2e904bd111c8..46e54d39461d 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -1413,6 +1413,82 @@ out_busy: return -ETXTBSY; } +static int +cifs_set_file_size(struct inode *inode, struct iattr *attrs, + int xid, char *full_path) +{ + int rc; + struct cifsFileInfo *open_file; + struct cifsInodeInfo *cifsInode = CIFS_I(inode); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifsTconInfo *pTcon = cifs_sb->tcon; + + /* + * To avoid spurious oplock breaks from server, in the case of + * inodes that we already have open, avoid doing path based + * setting of file size if we can do it by handle. + * This keeps our caching token (oplock) and avoids timeouts + * when the local oplock break takes longer to flush + * writebehind data than the SMB timeout for the SetPathInfo + * request would allow + */ + open_file = find_writable_file(cifsInode); + if (open_file) { + __u16 nfid = open_file->netfid; + __u32 npid = open_file->pid; + rc = CIFSSMBSetFileSize(xid, pTcon, attrs->ia_size, nfid, + npid, false); + atomic_dec(&open_file->wrtPending); + cFYI(1, ("SetFSize for attrs rc = %d", rc)); + if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) { + unsigned int bytes_written; + rc = CIFSSMBWrite(xid, pTcon, nfid, 0, attrs->ia_size, + &bytes_written, NULL, NULL, 1); + cFYI(1, ("Wrt seteof rc %d", rc)); + } + } else + rc = -EINVAL; + + if (rc != 0) { + /* Set file size by pathname rather than by handle + either because no valid, writeable file handle for + it was found or because there was an error setting + it by handle */ + rc = CIFSSMBSetEOF(xid, pTcon, full_path, attrs->ia_size, + false, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + cFYI(1, ("SetEOF by path (setattrs) rc = %d", rc)); + if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) { + __u16 netfid; + int oplock = 0; + + rc = SMBLegacyOpen(xid, pTcon, full_path, + FILE_OPEN, GENERIC_WRITE, + CREATE_NOT_DIR, &netfid, &oplock, NULL, + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + if (rc == 0) { + unsigned int bytes_written; + rc = CIFSSMBWrite(xid, pTcon, netfid, 0, + attrs->ia_size, + &bytes_written, NULL, + NULL, 1); + cFYI(1, ("wrt seteof rc %d", rc)); + CIFSSMBClose(xid, pTcon, netfid); + } + } + } + + if (rc == 0) { + rc = cifs_vmtruncate(inode, attrs->ia_size); + cifs_truncate_page(inode->i_mapping, inode->i_size); + } + + return rc; +} + int cifs_setattr(struct dentry *direntry, struct iattr *attrs) { int xid; @@ -1420,7 +1496,6 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) struct cifsTconInfo *pTcon; char *full_path = NULL; int rc = -EACCES; - struct cifsFileInfo *open_file = NULL; FILE_BASIC_INFO time_buf; bool set_time = false; bool set_dosattr = false; @@ -1472,78 +1547,8 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) } if (attrs->ia_valid & ATTR_SIZE) { - /* To avoid spurious oplock breaks from server, in the case of - inodes that we already have open, avoid doing path based - setting of file size if we can do it by handle. - This keeps our caching token (oplock) and avoids timeouts - when the local oplock break takes longer to flush - writebehind data than the SMB timeout for the SetPathInfo - request would allow */ - - open_file = find_writable_file(cifsInode); - if (open_file) { - __u16 nfid = open_file->netfid; - __u32 npid = open_file->pid; - rc = CIFSSMBSetFileSize(xid, pTcon, attrs->ia_size, - nfid, npid, false); - atomic_dec(&open_file->wrtPending); - cFYI(1, ("SetFSize for attrs rc = %d", rc)); - if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) { - unsigned int bytes_written; - rc = CIFSSMBWrite(xid, pTcon, - nfid, 0, attrs->ia_size, - &bytes_written, NULL, NULL, - 1 /* 45 seconds */); - cFYI(1, ("Wrt seteof rc %d", rc)); - } - } else - rc = -EINVAL; - - if (rc != 0) { - /* Set file size by pathname rather than by handle - either because no valid, writeable file handle for - it was found or because there was an error setting - it by handle */ - rc = CIFSSMBSetEOF(xid, pTcon, full_path, - attrs->ia_size, false, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - cFYI(1, ("SetEOF by path (setattrs) rc = %d", rc)); - if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) { - __u16 netfid; - int oplock = 0; - - rc = SMBLegacyOpen(xid, pTcon, full_path, - FILE_OPEN, GENERIC_WRITE, - CREATE_NOT_DIR, &netfid, &oplock, - NULL, cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (rc == 0) { - unsigned int bytes_written; - rc = CIFSSMBWrite(xid, pTcon, - netfid, 0, - attrs->ia_size, - &bytes_written, NULL, - NULL, 1 /* 45 sec */); - cFYI(1, ("wrt seteof rc %d", rc)); - CIFSSMBClose(xid, pTcon, netfid); - } - - } - } - - /* Server is ok setting allocation size implicitly - no need - to call: - CIFSSMBSetEOF(xid, pTcon, full_path, attrs->ia_size, true, - cifs_sb->local_nls); - */ - - if (rc == 0) { - rc = cifs_vmtruncate(inode, attrs->ia_size); - cifs_truncate_page(inode->i_mapping, inode->i_size); - } else + rc = cifs_set_file_size(inode, attrs, xid, full_path); + if (rc != 0) goto cifs_setattr_exit; } -- cgit v1.2.3 From 5ca33c6ac30596fc1403a092f46ea48c406734e4 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Wed, 23 Jul 2008 17:45:58 -0700 Subject: cifs: assorted endian annotations fs/cifs/cifssmb.c:3917:13: warning: incorrect type in assignment (different base types) fs/cifs/cifssmb.c:3917:13: expected bool [unsigned] [usertype] is_unicode fs/cifs/cifssmb.c:3917:13: got restricted __le16 The comment explains why __force is used here. fs/cifs/connect.c:458:16: warning: cast to restricted __be32 fs/cifs/connect.c:458:16: warning: cast to restricted __be32 fs/cifs/connect.c:458:16: warning: cast to restricted __be32 fs/cifs/connect.c:458:16: warning: cast to restricted __be32 fs/cifs/connect.c:458:16: warning: cast to restricted __be32 fs/cifs/connect.c:458:16: warning: cast to restricted __be32 Signed-off-by: Harvey Harrison Signed-off-by: Steve French --- fs/cifs/cifssmb.c | 5 ++++- fs/cifs/connect.c | 2 +- 2 files changed, 5 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 4511b708f0f3..7e175e1399d1 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -3914,7 +3914,10 @@ parse_DFS_referrals(TRANSACTION2_GET_DFS_REFER_RSP *pSMBr, bool is_unicode; struct dfs_referral_level_3 *ref; - is_unicode = pSMBr->hdr.Flags2 & SMBFLG2_UNICODE; + if (pSMBr->hdr.Flags2 & SMBFLG2_UNICODE) + is_unicode = true; + else + is_unicode = false; *num_of_nodes = le16_to_cpu(pSMBr->NumberOfReferrals); if (*num_of_nodes < 1) { diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 644462729387..b51d5777cde6 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -455,7 +455,7 @@ incomplete_rcv: /* Note that FC 1001 length is big endian on the wire, but we convert it here so it is always manipulated as host byte order */ - pdu_length = ntohl(smb_buffer->smb_buf_length); + pdu_length = be32_to_cpu((__force __be32)smb_buffer->smb_buf_length); smb_buffer->smb_buf_length = pdu_length; cFYI(1, ("rfc1002 length 0x%x", pdu_length+4)); -- cgit v1.2.3 From f984c7b98253c541ed6e702ee521f6a84c8113d4 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 24 Jul 2008 02:34:24 +0000 Subject: Signed-off-by: Alexey Dobriyan Cc: Steven French Signed-off-by: Andrew Morton Signed-off-by: Steve French --- fs/cifs/cifs_debug.c | 636 +++++++++++++++++++-------------------------------- 1 file changed, 241 insertions(+), 395 deletions(-) (limited to 'fs') diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index cc950f69e51e..73925a77375e 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -107,9 +107,7 @@ void cifs_dump_mids(struct TCP_Server_Info *server) #endif /* CONFIG_CIFS_DEBUG2 */ #ifdef CONFIG_PROC_FS -static int -cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, - int count, int *eof, void *data) +static int cifs_debug_data_proc_show(struct seq_file *m, void *v) { struct list_head *tmp; struct list_head *tmp1; @@ -117,23 +115,13 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, struct cifsSesInfo *ses; struct cifsTconInfo *tcon; int i; - int length = 0; - char *original_buf = buf; - *beginBuffer = buf + offset; - - length = - sprintf(buf, + seq_puts(m, "Display Internal CIFS Data Structures for Debugging\n" "---------------------------------------------------\n"); - buf += length; - length = sprintf(buf, "CIFS Version %s\n", CIFS_VERSION); - buf += length; - length = sprintf(buf, - "Active VFS Requests: %d\n", GlobalTotalActiveXid); - buf += length; - length = sprintf(buf, "Servers:"); - buf += length; + seq_printf(m, "CIFS Version %s\n", CIFS_VERSION); + seq_printf(m, "Active VFS Requests: %d\n", GlobalTotalActiveXid); + seq_printf(m, "Servers:"); i = 0; read_lock(&GlobalSMBSeslock); @@ -142,11 +130,10 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, ses = list_entry(tmp, struct cifsSesInfo, cifsSessionList); if ((ses->serverDomain == NULL) || (ses->serverOS == NULL) || (ses->serverNOS == NULL)) { - buf += sprintf(buf, "\nentry for %s not fully " + seq_printf(m, "\nentry for %s not fully " "displayed\n\t", ses->serverName); } else { - length = - sprintf(buf, + seq_printf(m, "\n%d) Name: %s Domain: %s Mounts: %d OS:" " %s \n\tNOS: %s\tCapability: 0x%x\n\tSMB" " session status: %d\t", @@ -154,10 +141,9 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, atomic_read(&ses->inUse), ses->serverOS, ses->serverNOS, ses->capabilities, ses->status); - buf += length; } if (ses->server) { - buf += sprintf(buf, "TCP status: %d\n\tLocal Users To " + seq_printf(m, "TCP status: %d\n\tLocal Users To " "Server: %d SecMode: 0x%x Req On Wire: %d", ses->server->tcpStatus, atomic_read(&ses->server->socketUseCount), @@ -165,13 +151,12 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, atomic_read(&ses->server->inFlight)); #ifdef CONFIG_CIFS_STATS2 - buf += sprintf(buf, " In Send: %d In MaxReq Wait: %d", + seq_printf(m, " In Send: %d In MaxReq Wait: %d", atomic_read(&ses->server->inSend), atomic_read(&ses->server->num_waiters)); #endif - length = sprintf(buf, "\nMIDs:\n"); - buf += length; + seq_puts(m, "\nMIDs:\n"); spin_lock(&GlobalMid_Lock); list_for_each(tmp1, &ses->server->pending_mid_q) { @@ -179,7 +164,7 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, mid_q_entry, qhead); if (mid_entry) { - length = sprintf(buf, + seq_printf(m, "State: %d com: %d pid:" " %d tsk: %p mid %d\n", mid_entry->midState, @@ -187,7 +172,6 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, mid_entry->pid, mid_entry->tsk, mid_entry->mid); - buf += length; } } spin_unlock(&GlobalMid_Lock); @@ -195,11 +179,9 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, } read_unlock(&GlobalSMBSeslock); - sprintf(buf, "\n"); - buf++; + seq_putc(m, '\n'); - length = sprintf(buf, "Shares:"); - buf += length; + seq_puts(m, "Shares:"); i = 0; read_lock(&GlobalSMBSeslock); @@ -208,62 +190,52 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, i++; tcon = list_entry(tmp, struct cifsTconInfo, cifsConnectionList); dev_type = le32_to_cpu(tcon->fsDevInfo.DeviceType); - length = sprintf(buf, "\n%d) %s Uses: %d ", i, + seq_printf(m, "\n%d) %s Uses: %d ", i, tcon->treeName, atomic_read(&tcon->useCount)); - buf += length; if (tcon->nativeFileSystem) { - length = sprintf(buf, "Type: %s ", + seq_printf(m, "Type: %s ", tcon->nativeFileSystem); - buf += length; } - length = sprintf(buf, "DevInfo: 0x%x Attributes: 0x%x" + seq_printf(m, "DevInfo: 0x%x Attributes: 0x%x" "\nPathComponentMax: %d Status: %d", le32_to_cpu(tcon->fsDevInfo.DeviceCharacteristics), le32_to_cpu(tcon->fsAttrInfo.Attributes), le32_to_cpu(tcon->fsAttrInfo.MaxPathNameComponentLength), tcon->tidStatus); - buf += length; if (dev_type == FILE_DEVICE_DISK) - length = sprintf(buf, " type: DISK "); + seq_puts(m, " type: DISK "); else if (dev_type == FILE_DEVICE_CD_ROM) - length = sprintf(buf, " type: CDROM "); + seq_puts(m, " type: CDROM "); else - length = - sprintf(buf, " type: %d ", dev_type); - buf += length; + seq_printf(m, " type: %d ", dev_type); if (tcon->tidStatus == CifsNeedReconnect) { - buf += sprintf(buf, "\tDISCONNECTED "); - length += 14; + seq_puts(m, "\tDISCONNECTED "); } } read_unlock(&GlobalSMBSeslock); - length = sprintf(buf, "\n"); - buf += length; + seq_putc(m, '\n'); /* BB add code to dump additional info such as TCP session info now */ - /* Now calculate total size of returned data */ - length = buf - original_buf; - - if (offset + count >= length) - *eof = 1; - if (length < offset) { - *eof = 1; - return 0; - } else { - length = length - offset; - } - if (length > count) - length = count; + return 0; +} - return length; +static int cifs_debug_data_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_debug_data_proc_show, NULL); } -#ifdef CONFIG_CIFS_STATS +static const struct file_operations cifs_debug_data_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_debug_data_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; -static int -cifs_stats_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) +#ifdef CONFIG_CIFS_STATS +static ssize_t cifs_stats_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { char c; int rc; @@ -307,236 +279,127 @@ cifs_stats_write(struct file *file, const char __user *buffer, return count; } -static int -cifs_stats_read(char *buf, char **beginBuffer, off_t offset, - int count, int *eof, void *data) +static int cifs_stats_proc_show(struct seq_file *m, void *v) { - int item_length, i, length; + int i; struct list_head *tmp; struct cifsTconInfo *tcon; - *beginBuffer = buf + offset; - - length = sprintf(buf, + seq_printf(m, "Resources in use\nCIFS Session: %d\n", sesInfoAllocCount.counter); - buf += length; - item_length = - sprintf(buf, "Share (unique mount targets): %d\n", + seq_printf(m, "Share (unique mount targets): %d\n", tconInfoAllocCount.counter); - length += item_length; - buf += item_length; - item_length = - sprintf(buf, "SMB Request/Response Buffer: %d Pool size: %d\n", + seq_printf(m, "SMB Request/Response Buffer: %d Pool size: %d\n", bufAllocCount.counter, cifs_min_rcv + tcpSesAllocCount.counter); - length += item_length; - buf += item_length; - item_length = - sprintf(buf, "SMB Small Req/Resp Buffer: %d Pool size: %d\n", + seq_printf(m, "SMB Small Req/Resp Buffer: %d Pool size: %d\n", smBufAllocCount.counter, cifs_min_small); - length += item_length; - buf += item_length; #ifdef CONFIG_CIFS_STATS2 - item_length = sprintf(buf, "Total Large %d Small %d Allocations\n", + seq_printf(m, "Total Large %d Small %d Allocations\n", atomic_read(&totBufAllocCount), atomic_read(&totSmBufAllocCount)); - length += item_length; - buf += item_length; #endif /* CONFIG_CIFS_STATS2 */ - item_length = - sprintf(buf, "Operations (MIDs): %d\n", - midCount.counter); - length += item_length; - buf += item_length; - item_length = sprintf(buf, + seq_printf(m, "Operations (MIDs): %d\n", midCount.counter); + seq_printf(m, "\n%d session %d share reconnects\n", tcpSesReconnectCount.counter, tconInfoReconnectCount.counter); - length += item_length; - buf += item_length; - item_length = sprintf(buf, + seq_printf(m, "Total vfs operations: %d maximum at one time: %d\n", GlobalCurrentXid, GlobalMaxActiveXid); - length += item_length; - buf += item_length; i = 0; read_lock(&GlobalSMBSeslock); list_for_each(tmp, &GlobalTreeConnectionList) { i++; tcon = list_entry(tmp, struct cifsTconInfo, cifsConnectionList); - item_length = sprintf(buf, "\n%d) %s", i, tcon->treeName); - buf += item_length; - length += item_length; - if (tcon->tidStatus == CifsNeedReconnect) { - buf += sprintf(buf, "\tDISCONNECTED "); - length += 14; - } - item_length = sprintf(buf, "\nSMBs: %d Oplock Breaks: %d", + seq_printf(m, "\n%d) %s", i, tcon->treeName); + if (tcon->tidStatus == CifsNeedReconnect) + seq_puts(m, "\tDISCONNECTED "); + seq_printf(m, "\nSMBs: %d Oplock Breaks: %d", atomic_read(&tcon->num_smbs_sent), atomic_read(&tcon->num_oplock_brks)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, "\nReads: %d Bytes: %lld", + seq_printf(m, "\nReads: %d Bytes: %lld", atomic_read(&tcon->num_reads), (long long)(tcon->bytes_read)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, "\nWrites: %d Bytes: %lld", + seq_printf(m, "\nWrites: %d Bytes: %lld", atomic_read(&tcon->num_writes), (long long)(tcon->bytes_written)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, + seq_printf(m, "\nLocks: %d HardLinks: %d Symlinks: %d", atomic_read(&tcon->num_locks), atomic_read(&tcon->num_hardlinks), atomic_read(&tcon->num_symlinks)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, "\nOpens: %d Closes: %d Deletes: %d", + seq_printf(m, "\nOpens: %d Closes: %d Deletes: %d", atomic_read(&tcon->num_opens), atomic_read(&tcon->num_closes), atomic_read(&tcon->num_deletes)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, "\nMkdirs: %d Rmdirs: %d", + seq_printf(m, "\nMkdirs: %d Rmdirs: %d", atomic_read(&tcon->num_mkdirs), atomic_read(&tcon->num_rmdirs)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, "\nRenames: %d T2 Renames %d", + seq_printf(m, "\nRenames: %d T2 Renames %d", atomic_read(&tcon->num_renames), atomic_read(&tcon->num_t2renames)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, "\nFindFirst: %d FNext %d FClose %d", + seq_printf(m, "\nFindFirst: %d FNext %d FClose %d", atomic_read(&tcon->num_ffirst), atomic_read(&tcon->num_fnext), atomic_read(&tcon->num_fclose)); - buf += item_length; - length += item_length; } read_unlock(&GlobalSMBSeslock); - buf += sprintf(buf, "\n"); - length++; - - if (offset + count >= length) - *eof = 1; - if (length < offset) { - *eof = 1; - return 0; - } else { - length = length - offset; - } - if (length > count) - length = count; + seq_putc(m, '\n'); + return 0; +} - return length; +static int cifs_stats_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_stats_proc_show, NULL); } + +static const struct file_operations cifs_stats_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_stats_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_stats_proc_write, +}; #endif /* STATS */ static struct proc_dir_entry *proc_fs_cifs; -read_proc_t cifs_txanchor_read; -static read_proc_t cifsFYI_read; -static write_proc_t cifsFYI_write; -static read_proc_t oplockEnabled_read; -static write_proc_t oplockEnabled_write; -static read_proc_t lookupFlag_read; -static write_proc_t lookupFlag_write; -static read_proc_t traceSMB_read; -static write_proc_t traceSMB_write; -static read_proc_t multiuser_mount_read; -static write_proc_t multiuser_mount_write; -static read_proc_t security_flags_read; -static write_proc_t security_flags_write; -/* static read_proc_t ntlmv2_enabled_read; -static write_proc_t ntlmv2_enabled_write; -static read_proc_t packet_signing_enabled_read; -static write_proc_t packet_signing_enabled_write;*/ -static read_proc_t experimEnabled_read; -static write_proc_t experimEnabled_write; -static read_proc_t linuxExtensionsEnabled_read; -static write_proc_t linuxExtensionsEnabled_write; +static const struct file_operations cifsFYI_proc_fops; +static const struct file_operations cifs_oplock_proc_fops; +static const struct file_operations cifs_lookup_cache_proc_fops; +static const struct file_operations traceSMB_proc_fops; +static const struct file_operations cifs_multiuser_mount_proc_fops; +static const struct file_operations cifs_security_flags_proc_fops; +static const struct file_operations cifs_experimental_proc_fops; +static const struct file_operations cifs_linux_ext_proc_fops; void cifs_proc_init(void) { - struct proc_dir_entry *pde; - proc_fs_cifs = proc_mkdir("fs/cifs", NULL); if (proc_fs_cifs == NULL) return; proc_fs_cifs->owner = THIS_MODULE; - create_proc_read_entry("DebugData", 0, proc_fs_cifs, - cifs_debug_data_read, NULL); + proc_create("DebugData", 0, proc_fs_cifs, &cifs_debug_data_proc_fops); #ifdef CONFIG_CIFS_STATS - pde = create_proc_read_entry("Stats", 0, proc_fs_cifs, - cifs_stats_read, NULL); - if (pde) - pde->write_proc = cifs_stats_write; + proc_create("Stats", 0, proc_fs_cifs, &cifs_stats_proc_fops); #endif /* STATS */ - pde = create_proc_read_entry("cifsFYI", 0, proc_fs_cifs, - cifsFYI_read, NULL); - if (pde) - pde->write_proc = cifsFYI_write; - - pde = - create_proc_read_entry("traceSMB", 0, proc_fs_cifs, - traceSMB_read, NULL); - if (pde) - pde->write_proc = traceSMB_write; - - pde = create_proc_read_entry("OplockEnabled", 0, proc_fs_cifs, - oplockEnabled_read, NULL); - if (pde) - pde->write_proc = oplockEnabled_write; - - pde = create_proc_read_entry("Experimental", 0, proc_fs_cifs, - experimEnabled_read, NULL); - if (pde) - pde->write_proc = experimEnabled_write; - - pde = create_proc_read_entry("LinuxExtensionsEnabled", 0, proc_fs_cifs, - linuxExtensionsEnabled_read, NULL); - if (pde) - pde->write_proc = linuxExtensionsEnabled_write; - - pde = - create_proc_read_entry("MultiuserMount", 0, proc_fs_cifs, - multiuser_mount_read, NULL); - if (pde) - pde->write_proc = multiuser_mount_write; - - pde = - create_proc_read_entry("SecurityFlags", 0, proc_fs_cifs, - security_flags_read, NULL); - if (pde) - pde->write_proc = security_flags_write; - - pde = - create_proc_read_entry("LookupCacheEnabled", 0, proc_fs_cifs, - lookupFlag_read, NULL); - if (pde) - pde->write_proc = lookupFlag_write; - -/* pde = - create_proc_read_entry("NTLMV2Enabled", 0, proc_fs_cifs, - ntlmv2_enabled_read, NULL); - if (pde) - pde->write_proc = ntlmv2_enabled_write; - - pde = - create_proc_read_entry("PacketSigningEnabled", 0, proc_fs_cifs, - packet_signing_enabled_read, NULL); - if (pde) - pde->write_proc = packet_signing_enabled_write;*/ + proc_create("cifsFYI", 0, proc_fs_cifs, &cifsFYI_proc_fops); + proc_create("traceSMB", 0, proc_fs_cifs, &traceSMB_proc_fops); + proc_create("OplockEnabled", 0, proc_fs_cifs, &cifs_oplock_proc_fops); + proc_create("Experimental", 0, proc_fs_cifs, &cifs_experimental_proc_fops); + proc_create("LinuxExtensionsEnabled", 0, proc_fs_cifs, &cifs_linux_ext_proc_fops); + proc_create("MultiuserMount", 0, proc_fs_cifs, &cifs_multiuser_mount_proc_fops); + proc_create("SecurityFlags", 0, proc_fs_cifs, &cifs_security_flags_proc_fops); + proc_create("LookupCacheEnabled", 0, proc_fs_cifs, &cifs_lookup_cache_proc_fops); } void @@ -553,39 +416,26 @@ cifs_proc_clean(void) #endif remove_proc_entry("MultiuserMount", proc_fs_cifs); remove_proc_entry("OplockEnabled", proc_fs_cifs); -/* remove_proc_entry("NTLMV2Enabled",proc_fs_cifs); */ remove_proc_entry("SecurityFlags", proc_fs_cifs); -/* remove_proc_entry("PacketSigningEnabled", proc_fs_cifs); */ remove_proc_entry("LinuxExtensionsEnabled", proc_fs_cifs); remove_proc_entry("Experimental", proc_fs_cifs); remove_proc_entry("LookupCacheEnabled", proc_fs_cifs); remove_proc_entry("fs/cifs", NULL); } -static int -cifsFYI_read(char *page, char **start, off_t off, int count, - int *eof, void *data) +static int cifsFYI_proc_show(struct seq_file *m, void *v) { - int len; - - len = sprintf(page, "%d\n", cifsFYI); - - len -= off; - *start = page + off; - - if (len > count) - len = count; - else - *eof = 1; - - if (len < 0) - len = 0; + seq_printf(m, "%d\n", cifsFYI); + return 0; +} - return len; +static int cifsFYI_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifsFYI_proc_show, NULL); } -static int -cifsFYI_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifsFYI_proc_write(struct file *file, const char __user *buffer, + size_t count, loff_t *ppos) { char c; int rc; @@ -603,30 +453,28 @@ cifsFYI_write(struct file *file, const char __user *buffer, return count; } -static int -oplockEnabled_read(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - int len; - - len = sprintf(page, "%d\n", oplockEnabled); +static const struct file_operations cifsFYI_proc_fops = { + .owner = THIS_MODULE, + .open = cifsFYI_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifsFYI_proc_write, +}; - len -= off; - *start = page + off; - - if (len > count) - len = count; - else - *eof = 1; - - if (len < 0) - len = 0; +static int cifs_oplock_proc_show(struct seq_file *m, void *v) +{ + seq_printf(m, "%d\n", oplockEnabled); + return 0; +} - return len; +static int cifs_oplock_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_oplock_proc_show, NULL); } -static int -oplockEnabled_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifs_oplock_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { char c; int rc; @@ -642,30 +490,28 @@ oplockEnabled_write(struct file *file, const char __user *buffer, return count; } -static int -experimEnabled_read(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - int len; - - len = sprintf(page, "%d\n", experimEnabled); - - len -= off; - *start = page + off; - - if (len > count) - len = count; - else - *eof = 1; +static const struct file_operations cifs_oplock_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_oplock_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_oplock_proc_write, +}; - if (len < 0) - len = 0; +static int cifs_experimental_proc_show(struct seq_file *m, void *v) +{ + seq_printf(m, "%d\n", experimEnabled); + return 0; +} - return len; +static int cifs_experimental_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_experimental_proc_show, NULL); } -static int -experimEnabled_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifs_experimental_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { char c; int rc; @@ -683,29 +529,28 @@ experimEnabled_write(struct file *file, const char __user *buffer, return count; } -static int -linuxExtensionsEnabled_read(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - int len; - - len = sprintf(page, "%d\n", linuxExtEnabled); - len -= off; - *start = page + off; - - if (len > count) - len = count; - else - *eof = 1; +static const struct file_operations cifs_experimental_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_experimental_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_experimental_proc_write, +}; - if (len < 0) - len = 0; +static int cifs_linux_ext_proc_show(struct seq_file *m, void *v) +{ + seq_printf(m, "%d\n", linuxExtEnabled); + return 0; +} - return len; +static int cifs_linux_ext_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_linux_ext_proc_show, NULL); } -static int -linuxExtensionsEnabled_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifs_linux_ext_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { char c; int rc; @@ -721,31 +566,28 @@ linuxExtensionsEnabled_write(struct file *file, const char __user *buffer, return count; } +static const struct file_operations cifs_linux_ext_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_linux_ext_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_linux_ext_proc_write, +}; -static int -lookupFlag_read(char *page, char **start, off_t off, - int count, int *eof, void *data) +static int cifs_lookup_cache_proc_show(struct seq_file *m, void *v) { - int len; - - len = sprintf(page, "%d\n", lookupCacheEnabled); - - len -= off; - *start = page + off; - - if (len > count) - len = count; - else - *eof = 1; - - if (len < 0) - len = 0; + seq_printf(m, "%d\n", lookupCacheEnabled); + return 0; +} - return len; +static int cifs_lookup_cache_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_lookup_cache_proc_show, NULL); } -static int -lookupFlag_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifs_lookup_cache_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { char c; int rc; @@ -760,30 +602,29 @@ lookupFlag_write(struct file *file, const char __user *buffer, return count; } -static int -traceSMB_read(char *page, char **start, off_t off, int count, - int *eof, void *data) -{ - int len; - len = sprintf(page, "%d\n", traceSMB); +static const struct file_operations cifs_lookup_cache_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_lookup_cache_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_lookup_cache_proc_write, +}; - len -= off; - *start = page + off; - - if (len > count) - len = count; - else - *eof = 1; - - if (len < 0) - len = 0; +static int traceSMB_proc_show(struct seq_file *m, void *v) +{ + seq_printf(m, "%d\n", traceSMB); + return 0; +} - return len; +static int traceSMB_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, traceSMB_proc_show, NULL); } -static int -traceSMB_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t traceSMB_proc_write(struct file *file, const char __user *buffer, + size_t count, loff_t *ppos) { char c; int rc; @@ -799,30 +640,28 @@ traceSMB_write(struct file *file, const char __user *buffer, return count; } -static int -multiuser_mount_read(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - int len; - - len = sprintf(page, "%d\n", multiuser_mount); - - len -= off; - *start = page + off; +static const struct file_operations traceSMB_proc_fops = { + .owner = THIS_MODULE, + .open = traceSMB_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = traceSMB_proc_write, +}; - if (len > count) - len = count; - else - *eof = 1; - - if (len < 0) - len = 0; +static int cifs_multiuser_mount_proc_show(struct seq_file *m, void *v) +{ + seq_printf(m, "%d\n", multiuser_mount); + return 0; +} - return len; +static int cifs_multiuser_mount_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_multiuser_mount_proc_show, NULL); } -static int -multiuser_mount_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifs_multiuser_mount_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { char c; int rc; @@ -838,30 +677,28 @@ multiuser_mount_write(struct file *file, const char __user *buffer, return count; } -static int -security_flags_read(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - int len; - - len = sprintf(page, "0x%x\n", extended_security); - - len -= off; - *start = page + off; +static const struct file_operations cifs_multiuser_mount_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_multiuser_mount_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_multiuser_mount_proc_write, +}; - if (len > count) - len = count; - else - *eof = 1; - - if (len < 0) - len = 0; +static int cifs_security_flags_proc_show(struct seq_file *m, void *v) +{ + seq_printf(m, "0x%x\n", extended_security); + return 0; +} - return len; +static int cifs_security_flags_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_security_flags_proc_show, NULL); } -static int -security_flags_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifs_security_flags_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { unsigned int flags; char flags_string[12]; @@ -917,6 +754,15 @@ security_flags_write(struct file *file, const char __user *buffer, /* BB should we turn on MAY flags for other MUST options? */ return count; } + +static const struct file_operations cifs_security_flags_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_security_flags_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_security_flags_proc_write, +}; #else inline void cifs_proc_init(void) { -- cgit v1.2.3 From 99b1f5b2f6cd2f65cce02c5f63302df5878a5fbc Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 24 Jul 2008 02:37:45 +0000 Subject: [CIFS] remove checkpatch warning Signed-off-by: Steve French --- fs/cifs/cifs_debug.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index 73925a77375e..c213aa40064e 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -395,11 +395,16 @@ cifs_proc_init(void) proc_create("cifsFYI", 0, proc_fs_cifs, &cifsFYI_proc_fops); proc_create("traceSMB", 0, proc_fs_cifs, &traceSMB_proc_fops); proc_create("OplockEnabled", 0, proc_fs_cifs, &cifs_oplock_proc_fops); - proc_create("Experimental", 0, proc_fs_cifs, &cifs_experimental_proc_fops); - proc_create("LinuxExtensionsEnabled", 0, proc_fs_cifs, &cifs_linux_ext_proc_fops); - proc_create("MultiuserMount", 0, proc_fs_cifs, &cifs_multiuser_mount_proc_fops); - proc_create("SecurityFlags", 0, proc_fs_cifs, &cifs_security_flags_proc_fops); - proc_create("LookupCacheEnabled", 0, proc_fs_cifs, &cifs_lookup_cache_proc_fops); + proc_create("Experimental", 0, proc_fs_cifs, + &cifs_experimental_proc_fops); + proc_create("LinuxExtensionsEnabled", 0, proc_fs_cifs, + &cifs_linux_ext_proc_fops); + proc_create("MultiuserMount", 0, proc_fs_cifs, + &cifs_multiuser_mount_proc_fops); + proc_create("SecurityFlags", 0, proc_fs_cifs, + &cifs_security_flags_proc_fops); + proc_create("LookupCacheEnabled", 0, proc_fs_cifs, + &cifs_lookup_cache_proc_fops); } void @@ -655,9 +660,9 @@ static int cifs_multiuser_mount_proc_show(struct seq_file *m, void *v) return 0; } -static int cifs_multiuser_mount_proc_open(struct inode *inode, struct file *file) +static int cifs_multiuser_mount_proc_open(struct inode *inode, struct file *fh) { - return single_open(file, cifs_multiuser_mount_proc_show, NULL); + return single_open(fh, cifs_multiuser_mount_proc_show, NULL); } static ssize_t cifs_multiuser_mount_proc_write(struct file *file, -- cgit v1.2.3 From 76c510ad2e7d56cfe8f2cc7b23783e5c687cf704 Mon Sep 17 00:00:00 2001 From: Shirish Pargaonkar Date: Thu, 24 Jul 2008 14:48:33 +0000 Subject: [CIFS] Fix possible double free if search immediately after search rewind fails Signed-off-by: Shirish Pargaonkar Signed-off-by: Steve French --- fs/cifs/readdir.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 83f306954883..5f40ed3473f5 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -690,6 +690,7 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, else cifs_buf_release(cifsFile->srch_inf. ntwrk_buf_start); + cifsFile->srch_inf.ntwrk_buf_start = NULL; } rc = initiate_cifs_search(xid, file); if (rc) { -- cgit v1.2.3 From b1910ad6222a705650dc991c003af43b94cdb3e1 Mon Sep 17 00:00:00 2001 From: Shirish Pargaonkar Date: Thu, 24 Jul 2008 14:53:20 +0000 Subject: [CIFS] Fix improper endian conversion of ACL subauth field In mode_to_acl when converting a Unix mode to a Windows ACL the subauth fields of the SID in the ACL were translated incorrectly on bigendian architectures Signed-off-by: Steve French --- fs/cifs/cifsacl.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c index 0e9fc2ba90ee..18faf65b9114 100644 --- a/fs/cifs/cifsacl.c +++ b/fs/cifs/cifsacl.c @@ -169,8 +169,7 @@ static void copy_sec_desc(const struct cifs_ntsd *pntsd, for (i = 0; i < 6; i++) ngroup_sid_ptr->authority[i] = group_sid_ptr->authority[i]; for (i = 0; i < 5; i++) - ngroup_sid_ptr->sub_auth[i] = - cpu_to_le32(group_sid_ptr->sub_auth[i]); + ngroup_sid_ptr->sub_auth[i] = group_sid_ptr->sub_auth[i]; return; } -- cgit v1.2.3 From ef571cadd516e7ffcdeac6bb8054e5908fcccfcf Mon Sep 17 00:00:00 2001 From: Shirish Pargaonkar Date: Thu, 24 Jul 2008 15:56:05 +0000 Subject: [CIFS] Fix warnings from checkpatch Signed-off-by: Steve French --- fs/cifs/cifs_debug.c | 4 ++-- fs/cifs/cifsacl.c | 38 +++++++++++++++++++------------------- fs/cifs/cifsencrypt.c | 3 +-- fs/cifs/cifsglob.h | 6 +++--- fs/cifs/cifspdu.h | 8 ++++---- fs/cifs/cifssmb.c | 5 ++--- 6 files changed, 31 insertions(+), 33 deletions(-) (limited to 'fs') diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index c213aa40064e..688a2d42153f 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -208,9 +208,9 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v) seq_puts(m, " type: CDROM "); else seq_printf(m, " type: %d ", dev_type); - if (tcon->tidStatus == CifsNeedReconnect) { + + if (tcon->tidStatus == CifsNeedReconnect) seq_puts(m, "\tDISCONNECTED "); - } } read_unlock(&GlobalSMBSeslock); diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c index 18faf65b9114..57ecdc83c26f 100644 --- a/fs/cifs/cifsacl.c +++ b/fs/cifs/cifsacl.c @@ -56,7 +56,7 @@ int match_sid(struct cifs_sid *ctsid) struct cifs_sid *cwsid; if (!ctsid) - return (-1); + return -1; for (i = 0; i < NUM_WK_SIDS; ++i) { cwsid = &(wksidarr[i].cifssid); @@ -87,11 +87,11 @@ int match_sid(struct cifs_sid *ctsid) } cFYI(1, ("matching sid: %s\n", wksidarr[i].sidname)); - return (0); /* sids compare/match */ + return 0; /* sids compare/match */ } cFYI(1, ("No matching sid")); - return (-1); + return -1; } /* if the two SIDs (roughly equivalent to a UUID for a user or group) are @@ -102,16 +102,16 @@ int compare_sids(const struct cifs_sid *ctsid, const struct cifs_sid *cwsid) int num_subauth, num_sat, num_saw; if ((!ctsid) || (!cwsid)) - return (0); + return 0; /* compare the revision */ if (ctsid->revision != cwsid->revision) - return (0); + return 0; /* compare all of the six auth values */ for (i = 0; i < 6; ++i) { if (ctsid->authority[i] != cwsid->authority[i]) - return (0); + return 0; } /* compare all of the subauth values if any */ @@ -121,11 +121,11 @@ int compare_sids(const struct cifs_sid *ctsid, const struct cifs_sid *cwsid) if (num_subauth) { for (i = 0; i < num_subauth; ++i) { if (ctsid->sub_auth[i] != cwsid->sub_auth[i]) - return (0); + return 0; } } - return (1); /* sids compare/match */ + return 1; /* sids compare/match */ } @@ -284,7 +284,7 @@ static __u16 fill_ace_for_sid(struct cifs_ace *pntace, size = 1 + 1 + 2 + 4 + 1 + 1 + 6 + (psid->num_subauth * 4); pntace->size = cpu_to_le16(size); - return (size); + return size; } @@ -425,7 +425,7 @@ static int set_chmod_dacl(struct cifs_acl *pndacl, struct cifs_sid *pownersid, pndacl->size = cpu_to_le16(size + sizeof(struct cifs_acl)); pndacl->num_aces = cpu_to_le32(3); - return (0); + return 0; } @@ -509,7 +509,7 @@ static int parse_sec_desc(struct cifs_ntsd *pntsd, int acl_len, sizeof(struct cifs_sid)); */ - return (0); + return 0; } @@ -526,7 +526,7 @@ static int build_sec_desc(struct cifs_ntsd *pntsd, struct cifs_ntsd *pnntsd, struct cifs_acl *ndacl_ptr = NULL; /* no need for SACL ptr */ if ((inode == NULL) || (pntsd == NULL) || (pnntsd == NULL)) - return (-EIO); + return -EIO; owner_sid_ptr = (struct cifs_sid *)((char *)pntsd + le32_to_cpu(pntsd->osidoffset)); @@ -549,7 +549,7 @@ static int build_sec_desc(struct cifs_ntsd *pntsd, struct cifs_ntsd *pnntsd, /* copy security descriptor control portion and owner and group sid */ copy_sec_desc(pntsd, pnntsd, sidsoffset); - return (rc); + return rc; } @@ -628,11 +628,11 @@ static int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen, cFYI(DBG2, ("set ACL for %s from mode 0x%x", path, inode->i_mode)); if (!inode) - return (rc); + return rc; sb = inode->i_sb; if (sb == NULL) - return (rc); + return rc; cifs_sb = CIFS_SB(sb); xid = GetXid(); @@ -651,7 +651,7 @@ static int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen, if (rc != 0) { cERROR(1, ("Unable to open file to set ACL")); FreeXid(xid); - return (rc); + return rc; } } @@ -664,7 +664,7 @@ static int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen, FreeXid(xid); - return (rc); + return rc; } /* Translate the CIFS ACL (simlar to NTFS ACL) for a file into mode bits */ @@ -714,7 +714,7 @@ int mode_to_acl(struct inode *inode, const char *path, __u64 nmode) if (!pnntsd) { cERROR(1, ("Unable to allocate security descriptor")); kfree(pntsd); - return (-ENOMEM); + return -ENOMEM; } rc = build_sec_desc(pntsd, pnntsd, inode, nmode); @@ -731,6 +731,6 @@ int mode_to_acl(struct inode *inode, const char *path, __u64 nmode) kfree(pntsd); } - return (rc); + return rc; } #endif /* CONFIG_CIFS_EXPERIMENTAL */ diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c index 4ff8939c6cc7..83fd40dc1ef0 100644 --- a/fs/cifs/cifsencrypt.c +++ b/fs/cifs/cifsencrypt.c @@ -310,9 +310,8 @@ void calc_lanman_hash(struct cifsSesInfo *ses, char *lnm_session_key) utf8 and other multibyte codepages each need their own strupper function since a byte at a time will ont work. */ - for (i = 0; i < CIFS_ENCPWD_SIZE; i++) { + for (i = 0; i < CIFS_ENCPWD_SIZE; i++) password_with_pad[i] = toupper(password_with_pad[i]); - } SMBencrypt(password_with_pad, ses->server->cryptKey, lnm_session_key); /* clear password before we return/free memory */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 9cfcf326ead3..7e1cf262effe 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -27,7 +27,7 @@ #define MAX_SES_INFO 2 #define MAX_TCON_INFO 4 -#define MAX_TREE_SIZE 2 + MAX_SERVER_SIZE + 1 + MAX_SHARE_SIZE + 1 +#define MAX_TREE_SIZE (2 + MAX_SERVER_SIZE + 1 + MAX_SHARE_SIZE + 1) #define MAX_SERVER_SIZE 15 #define MAX_SHARE_SIZE 64 /* used to be 20, this should still be enough */ #define MAX_USERNAME_SIZE 32 /* 32 is to allow for 15 char names + null @@ -537,8 +537,8 @@ require use of the stronger protocol */ #endif /* WEAK_PW_HASH */ #define CIFSSEC_MUST_SEAL 0x40040 /* not supported yet */ -#define CIFSSEC_DEF CIFSSEC_MAY_SIGN | CIFSSEC_MAY_NTLM | CIFSSEC_MAY_NTLMV2 -#define CIFSSEC_MAX CIFSSEC_MUST_SIGN | CIFSSEC_MUST_NTLMV2 +#define CIFSSEC_DEF (CIFSSEC_MAY_SIGN | CIFSSEC_MAY_NTLM | CIFSSEC_MAY_NTLMV2) +#define CIFSSEC_MAX (CIFSSEC_MUST_SIGN | CIFSSEC_MUST_NTLMV2) #define CIFSSEC_AUTH_MASK (CIFSSEC_MAY_NTLM | CIFSSEC_MAY_NTLMV2 | CIFSSEC_MAY_LANMAN | CIFSSEC_MAY_PLNTXT | CIFSSEC_MAY_KRB5) /* ***************************************************************** diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index 0f327c224da3..409abce12732 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -31,7 +31,7 @@ #else #define CIFS_PROT 0 #endif -#define POSIX_PROT CIFS_PROT+1 +#define POSIX_PROT (CIFS_PROT+1) #define BAD_PROT 0xFFFF /* SMB command codes */ @@ -341,7 +341,7 @@ #define CREATE_COMPLETE_IF_OPLK 0x00000100 /* should be zero */ #define CREATE_NO_EA_KNOWLEDGE 0x00000200 #define CREATE_EIGHT_DOT_THREE 0x00000400 /* doc says this is obsolete - "open for recovery" flag - should + "open for recovery" flag should be zero in any case */ #define CREATE_OPEN_FOR_RECOVERY 0x00000400 #define CREATE_RANDOM_ACCESS 0x00000800 @@ -414,8 +414,8 @@ struct smb_hdr { __u8 WordCount; } __attribute__((packed)); /* given a pointer to an smb_hdr retrieve the value of byte count */ -#define BCC(smb_var) ( *(__u16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount))) -#define BCC_LE(smb_var) ( *(__le16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount))) +#define BCC(smb_var) (*(__u16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount))) +#define BCC_LE(smb_var) (*(__le16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount))) /* given a pointer to an smb_hdr retrieve the pointer to the byte area */ #define pByteArea(smb_var) ((unsigned char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount) + 2) diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 7e175e1399d1..c621ffa2ca90 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -686,11 +686,10 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) SecurityBlob, count - 16, &server->secType); - if (rc == 1) { + if (rc == 1) rc = 0; - } else { + else rc = -EINVAL; - } } } else server->capabilities &= ~CAP_EXTENDED_SECURITY; -- cgit v1.2.3 From e2d2867ff8700d7431c68c089ff5f5ed7f2d5b40 Mon Sep 17 00:00:00 2001 From: Chris Wright Date: Thu, 24 Jul 2008 20:43:34 +0000 Subject: When verifying the decoded header before decoding the object identifier (expecting a SPNEGO pseudo-mechanism oid), the test to verify it is a primitive encoding is compared against the asn1 class. Primitive is not a class. This brings check in line with similar check for krb/ntlmssp oid. Signed-off-by: Chris Wright Cc: Steven French Signed-off-by: Andrew Morton Signed-off-by: Steve French --- fs/cifs/asn1.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/cifs/asn1.c b/fs/cifs/asn1.c index 42765465701c..6bb440b257b0 100644 --- a/fs/cifs/asn1.c +++ b/fs/cifs/asn1.c @@ -494,7 +494,7 @@ decode_negTokenInit(unsigned char *security_blob, int length, /* remember to free obj->oid */ rc = asn1_header_decode(&ctx, &end, &cls, &con, &tag); if (rc) { - if ((tag == ASN1_OJI) && (cls == ASN1_PRI)) { + if ((tag == ASN1_OJI) && (con == ASN1_PRI)) { rc = asn1_oid_decode(&ctx, end, &oid, &oidlen); if (rc) { rc = compare_oid(oid, oidlen, -- cgit v1.2.3 From f3c6ba986ab4527b6dfacf9f3b9e40f72466a8b2 Mon Sep 17 00:00:00 2001 From: Kentaro Makita Date: Fri, 25 Jul 2008 19:44:40 -0700 Subject: vfs: add cond_resched_lock while scanning dentry LRU lists Add cond_resched_lock(&dcache_lock) while scanning LRU lists on superblocks in __shrink_dcache_sb() Signed-off-by: Kentaro Makita Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/dcache.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/dcache.c b/fs/dcache.c index 3818d6ab76ca..f2584d22cb45 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -487,6 +487,7 @@ restart: if (!cnt) break; } + cond_resched_lock(&dcache_lock); } } while (!list_empty(&tmp)) { -- cgit v1.2.3 From 7d135a5d50a08bbc53b189d79d8bdb03136f5303 Mon Sep 17 00:00:00 2001 From: Matthias Kaehlcke Date: Fri, 25 Jul 2008 19:44:51 -0700 Subject: affs: convert s_bmlock into a mutex The semaphore s_bmlock is used as a mutex. Convert it to the mutex API. Signed-off-by: Matthias Kaehlcke Cc: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/affs/affs.h | 3 ++- fs/affs/bitmap.c | 18 +++++++++--------- fs/affs/super.c | 2 +- 3 files changed, 12 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/affs/affs.h b/fs/affs/affs.h index 223b1917093e..e9ec915f7553 100644 --- a/fs/affs/affs.h +++ b/fs/affs/affs.h @@ -2,6 +2,7 @@ #include #include #include +#include /* AmigaOS allows file names with up to 30 characters length. * Names longer than that will be silently truncated. If you @@ -98,7 +99,7 @@ struct affs_sb_info { gid_t s_gid; /* gid to override */ umode_t s_mode; /* mode to override */ struct buffer_head *s_root_bh; /* Cached root block. */ - struct semaphore s_bmlock; /* Protects bitmap access. */ + struct mutex s_bmlock; /* Protects bitmap access. */ struct affs_bm_info *s_bitmap; /* Bitmap infos. */ u32 s_bmap_count; /* # of bitmap blocks. */ u32 s_bmap_bits; /* # of bits in one bitmap blocks */ diff --git a/fs/affs/bitmap.c b/fs/affs/bitmap.c index c4a5ad09ddf2..dc5ef14bdc1c 100644 --- a/fs/affs/bitmap.c +++ b/fs/affs/bitmap.c @@ -45,14 +45,14 @@ affs_count_free_blocks(struct super_block *sb) if (sb->s_flags & MS_RDONLY) return 0; - down(&AFFS_SB(sb)->s_bmlock); + mutex_lock(&AFFS_SB(sb)->s_bmlock); bm = AFFS_SB(sb)->s_bitmap; free = 0; for (i = AFFS_SB(sb)->s_bmap_count; i > 0; bm++, i--) free += bm->bm_free; - up(&AFFS_SB(sb)->s_bmlock); + mutex_unlock(&AFFS_SB(sb)->s_bmlock); return free; } @@ -76,7 +76,7 @@ affs_free_block(struct super_block *sb, u32 block) bit = blk % sbi->s_bmap_bits; bm = &sbi->s_bitmap[bmap]; - down(&sbi->s_bmlock); + mutex_lock(&sbi->s_bmlock); bh = sbi->s_bmap_bh; if (sbi->s_last_bmap != bmap) { @@ -105,19 +105,19 @@ affs_free_block(struct super_block *sb, u32 block) sb->s_dirt = 1; bm->bm_free++; - up(&sbi->s_bmlock); + mutex_unlock(&sbi->s_bmlock); return; err_free: affs_warning(sb,"affs_free_block","Trying to free block %u which is already free", block); - up(&sbi->s_bmlock); + mutex_unlock(&sbi->s_bmlock); return; err_bh_read: affs_error(sb,"affs_free_block","Cannot read bitmap block %u", bm->bm_key); sbi->s_bmap_bh = NULL; sbi->s_last_bmap = ~0; - up(&sbi->s_bmlock); + mutex_unlock(&sbi->s_bmlock); return; err_range: @@ -168,7 +168,7 @@ affs_alloc_block(struct inode *inode, u32 goal) bmap = blk / sbi->s_bmap_bits; bm = &sbi->s_bitmap[bmap]; - down(&sbi->s_bmlock); + mutex_lock(&sbi->s_bmlock); if (bm->bm_free) goto find_bmap_bit; @@ -249,7 +249,7 @@ find_bit: mark_buffer_dirty(bh); sb->s_dirt = 1; - up(&sbi->s_bmlock); + mutex_unlock(&sbi->s_bmlock); pr_debug("%d\n", blk); return blk; @@ -259,7 +259,7 @@ err_bh_read: sbi->s_bmap_bh = NULL; sbi->s_last_bmap = ~0; err_full: - up(&sbi->s_bmlock); + mutex_unlock(&sbi->s_bmlock); pr_debug("failed\n"); return 0; } diff --git a/fs/affs/super.c b/fs/affs/super.c index d214837d5e42..4e0309566406 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -290,7 +290,7 @@ static int affs_fill_super(struct super_block *sb, void *data, int silent) if (!sbi) return -ENOMEM; sb->s_fs_info = sbi; - init_MUTEX(&sbi->s_bmlock); + mutex_init(&sbi->s_bmlock); if (!parse_options(data,&uid,&gid,&i,&reserved,&root_block, &blocksize,&sbi->s_prefix, -- cgit v1.2.3 From 75b25b4cabb7ce956c36442bf8225659b1864866 Mon Sep 17 00:00:00 2001 From: Dmitri Vorobiev Date: Fri, 25 Jul 2008 19:44:52 -0700 Subject: bfs: assorted cleanups This patch makes the following cleanups: o removing an unused variable from bfs_fill_super(); o removing unneeded blank spaces from pointer definitions. Signed-off-by: Dmitri Vorobiev Cc: Tigran Aivazian Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/bfs/bfs.h | 4 ++-- fs/bfs/inode.c | 3 +-- 2 files changed, 3 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bfs/bfs.h b/fs/bfs/bfs.h index 70f5d3a8eede..d1a6cd10afce 100644 --- a/fs/bfs/bfs.h +++ b/fs/bfs/bfs.h @@ -16,8 +16,8 @@ struct bfs_sb_info { unsigned long si_freei; unsigned long si_lf_eblk; unsigned long si_lasti; - unsigned long * si_imap; - struct buffer_head * si_sbh; /* buffer header w/superblock */ + unsigned long *si_imap; + struct buffer_head *si_sbh; /* buffer header w/superblock */ }; /* diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index 8db623838b50..c8b3982e11ca 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -380,7 +380,7 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent) struct bfs_inode *di; int block = (i - BFS_ROOT_INO) / BFS_INODES_PER_BLOCK + 1; int off = (i - BFS_ROOT_INO) % BFS_INODES_PER_BLOCK; - unsigned long sblock, eblock; + unsigned long eblock; if (!off) { brelse(bh); @@ -399,7 +399,6 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent) set_bit(i, info->si_imap); info->si_freeb -= BFS_FILEBLOCKS(di); - sblock = le32_to_cpu(di->i_sblock); eblock = le32_to_cpu(di->i_eblock); if (eblock > info->si_lf_eblk) info->si_lf_eblk = eblock; -- cgit v1.2.3 From 3f165e4cf2af042af7d2440d688299c0d2a48b1f Mon Sep 17 00:00:00 2001 From: Dmitri Vorobiev Date: Fri, 25 Jul 2008 19:44:54 -0700 Subject: bfs: kill BKL Replace the BKL-based locking scheme used in the bfs driver by a private filesystem-wide mutex. Signed-off-by: Dmitri Vorobiev Cc: Tigran Aivazian Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/bfs/bfs.h | 1 + fs/bfs/dir.c | 46 ++++++++++++++++++++++++++-------------------- fs/bfs/file.c | 4 ++-- fs/bfs/inode.c | 24 +++++++++++++++--------- 4 files changed, 44 insertions(+), 31 deletions(-) (limited to 'fs') diff --git a/fs/bfs/bfs.h b/fs/bfs/bfs.h index d1a6cd10afce..7109e451abf7 100644 --- a/fs/bfs/bfs.h +++ b/fs/bfs/bfs.h @@ -18,6 +18,7 @@ struct bfs_sb_info { unsigned long si_lasti; unsigned long *si_imap; struct buffer_head *si_sbh; /* buffer header w/superblock */ + struct mutex bfs_lock; }; /* diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c index 034950cb3cbe..87ee5ccee348 100644 --- a/fs/bfs/dir.c +++ b/fs/bfs/dir.c @@ -32,16 +32,17 @@ static int bfs_readdir(struct file *f, void *dirent, filldir_t filldir) struct inode *dir = f->f_path.dentry->d_inode; struct buffer_head *bh; struct bfs_dirent *de; + struct bfs_sb_info *info = BFS_SB(dir->i_sb); unsigned int offset; int block; - lock_kernel(); + mutex_lock(&info->bfs_lock); if (f->f_pos & (BFS_DIRENT_SIZE - 1)) { printf("Bad f_pos=%08lx for %s:%08lx\n", (unsigned long)f->f_pos, dir->i_sb->s_id, dir->i_ino); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return -EBADF; } @@ -61,7 +62,7 @@ static int bfs_readdir(struct file *f, void *dirent, filldir_t filldir) le16_to_cpu(de->ino), DT_UNKNOWN) < 0) { brelse(bh); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return 0; } } @@ -71,7 +72,7 @@ static int bfs_readdir(struct file *f, void *dirent, filldir_t filldir) brelse(bh); } - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return 0; } @@ -95,10 +96,10 @@ static int bfs_create(struct inode *dir, struct dentry *dentry, int mode, inode = new_inode(s); if (!inode) return -ENOSPC; - lock_kernel(); + mutex_lock(&info->bfs_lock); ino = find_first_zero_bit(info->si_imap, info->si_lasti); if (ino > info->si_lasti) { - unlock_kernel(); + mutex_unlock(&info->bfs_lock); iput(inode); return -ENOSPC; } @@ -125,10 +126,10 @@ static int bfs_create(struct inode *dir, struct dentry *dentry, int mode, if (err) { inode_dec_link_count(inode); iput(inode); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return err; } - unlock_kernel(); + mutex_unlock(&info->bfs_lock); d_instantiate(dentry, inode); return 0; } @@ -139,22 +140,23 @@ static struct dentry *bfs_lookup(struct inode *dir, struct dentry *dentry, struct inode *inode = NULL; struct buffer_head *bh; struct bfs_dirent *de; + struct bfs_sb_info *info = BFS_SB(dir->i_sb); if (dentry->d_name.len > BFS_NAMELEN) return ERR_PTR(-ENAMETOOLONG); - lock_kernel(); + mutex_lock(&info->bfs_lock); bh = bfs_find_entry(dir, dentry->d_name.name, dentry->d_name.len, &de); if (bh) { unsigned long ino = (unsigned long)le16_to_cpu(de->ino); brelse(bh); inode = bfs_iget(dir->i_sb, ino); if (IS_ERR(inode)) { - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return ERR_CAST(inode); } } - unlock_kernel(); + mutex_unlock(&info->bfs_lock); d_add(dentry, inode); return NULL; } @@ -163,13 +165,14 @@ static int bfs_link(struct dentry *old, struct inode *dir, struct dentry *new) { struct inode *inode = old->d_inode; + struct bfs_sb_info *info = BFS_SB(inode->i_sb); int err; - lock_kernel(); + mutex_lock(&info->bfs_lock); err = bfs_add_entry(dir, new->d_name.name, new->d_name.len, inode->i_ino); if (err) { - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return err; } inc_nlink(inode); @@ -177,19 +180,19 @@ static int bfs_link(struct dentry *old, struct inode *dir, mark_inode_dirty(inode); atomic_inc(&inode->i_count); d_instantiate(new, inode); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return 0; } static int bfs_unlink(struct inode *dir, struct dentry *dentry) { int error = -ENOENT; - struct inode *inode; + struct inode *inode = dentry->d_inode; struct buffer_head *bh; struct bfs_dirent *de; + struct bfs_sb_info *info = BFS_SB(inode->i_sb); - inode = dentry->d_inode; - lock_kernel(); + mutex_lock(&info->bfs_lock); bh = bfs_find_entry(dir, dentry->d_name.name, dentry->d_name.len, &de); if (!bh || (le16_to_cpu(de->ino) != inode->i_ino)) goto out_brelse; @@ -210,7 +213,7 @@ static int bfs_unlink(struct inode *dir, struct dentry *dentry) out_brelse: brelse(bh); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return error; } @@ -220,6 +223,7 @@ static int bfs_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *old_inode, *new_inode; struct buffer_head *old_bh, *new_bh; struct bfs_dirent *old_de, *new_de; + struct bfs_sb_info *info; int error = -ENOENT; old_bh = new_bh = NULL; @@ -227,7 +231,9 @@ static int bfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (S_ISDIR(old_inode->i_mode)) return -EINVAL; - lock_kernel(); + info = BFS_SB(old_inode->i_sb); + + mutex_lock(&info->bfs_lock); old_bh = bfs_find_entry(old_dir, old_dentry->d_name.name, old_dentry->d_name.len, &old_de); @@ -264,7 +270,7 @@ static int bfs_rename(struct inode *old_dir, struct dentry *old_dentry, error = 0; end_rename: - unlock_kernel(); + mutex_unlock(&info->bfs_lock); brelse(old_bh); brelse(new_bh); return error; diff --git a/fs/bfs/file.c b/fs/bfs/file.c index b11e63e8fbcd..6a021265f018 100644 --- a/fs/bfs/file.c +++ b/fs/bfs/file.c @@ -99,7 +99,7 @@ static int bfs_get_block(struct inode *inode, sector_t block, return -ENOSPC; /* The rest has to be protected against itself. */ - lock_kernel(); + mutex_lock(&info->bfs_lock); /* * If the last data block for this file is the last allocated @@ -151,7 +151,7 @@ static int bfs_get_block(struct inode *inode, sector_t block, mark_buffer_dirty(sbh); map_bh(bh_result, sb, phys); out: - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return err; } diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index c8b3982e11ca..053e690ec9ed 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -104,6 +104,7 @@ static int bfs_write_inode(struct inode *inode, int unused) struct bfs_inode *di; struct buffer_head *bh; int block, off; + struct bfs_sb_info *info = BFS_SB(inode->i_sb); dprintf("ino=%08x\n", ino); @@ -112,13 +113,13 @@ static int bfs_write_inode(struct inode *inode, int unused) return -EIO; } - lock_kernel(); + mutex_lock(&info->bfs_lock); block = (ino - BFS_ROOT_INO) / BFS_INODES_PER_BLOCK + 1; bh = sb_bread(inode->i_sb, block); if (!bh) { printf("Unable to read inode %s:%08x\n", inode->i_sb->s_id, ino); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return -EIO; } @@ -145,7 +146,7 @@ static int bfs_write_inode(struct inode *inode, int unused) mark_buffer_dirty(bh); brelse(bh); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return 0; } @@ -170,7 +171,7 @@ static void bfs_delete_inode(struct inode *inode) inode->i_size = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; - lock_kernel(); + mutex_lock(&info->bfs_lock); mark_inode_dirty(inode); block = (ino - BFS_ROOT_INO) / BFS_INODES_PER_BLOCK + 1; @@ -178,7 +179,7 @@ static void bfs_delete_inode(struct inode *inode) if (!bh) { printf("Unable to read inode %s:%08lx\n", inode->i_sb->s_id, ino); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return; } off = (ino - BFS_ROOT_INO) % BFS_INODES_PER_BLOCK; @@ -204,14 +205,16 @@ static void bfs_delete_inode(struct inode *inode) info->si_lf_eblk = bi->i_sblock - 1; mark_buffer_dirty(info->si_sbh); } - unlock_kernel(); + mutex_unlock(&info->bfs_lock); clear_inode(inode); } static void bfs_put_super(struct super_block *s) { struct bfs_sb_info *info = BFS_SB(s); + brelse(info->si_sbh); + mutex_destroy(&info->bfs_lock); kfree(info->si_imap); kfree(info); s->s_fs_info = NULL; @@ -236,11 +239,13 @@ static int bfs_statfs(struct dentry *dentry, struct kstatfs *buf) static void bfs_write_super(struct super_block *s) { - lock_kernel(); + struct bfs_sb_info *info = BFS_SB(s); + + mutex_lock(&info->bfs_lock); if (!(s->s_flags & MS_RDONLY)) - mark_buffer_dirty(BFS_SB(s)->si_sbh); + mark_buffer_dirty(info->si_sbh); s->s_dirt = 0; - unlock_kernel(); + mutex_unlock(&info->bfs_lock); } static struct kmem_cache *bfs_inode_cachep; @@ -409,6 +414,7 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent) s->s_dirt = 1; } dump_imap("read_super", s); + mutex_init(&info->bfs_lock); return 0; out: -- cgit v1.2.3 From 1b002d7b173ae7cc15ed90d3c07f6d106babc510 Mon Sep 17 00:00:00 2001 From: Bob Copeland Date: Fri, 25 Jul 2008 19:45:15 -0700 Subject: omfs: define filesystem structures Add header files containing OMFS on-disk and memory structures. Signed-off-by: Bob Copeland Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/omfs/omfs.h | 67 ++++++++++++++++++++++++++++++++++++++++++++++ fs/omfs/omfs_fs.h | 80 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 147 insertions(+) create mode 100644 fs/omfs/omfs.h create mode 100644 fs/omfs/omfs_fs.h (limited to 'fs') diff --git a/fs/omfs/omfs.h b/fs/omfs/omfs.h new file mode 100644 index 000000000000..2bc0f0670406 --- /dev/null +++ b/fs/omfs/omfs.h @@ -0,0 +1,67 @@ +#ifndef _OMFS_H +#define _OMFS_H + +#include +#include + +#include "omfs_fs.h" + +/* In-memory structures */ +struct omfs_sb_info { + u64 s_num_blocks; + u64 s_bitmap_ino; + u64 s_root_ino; + u32 s_blocksize; + u32 s_mirrors; + u32 s_sys_blocksize; + u32 s_clustersize; + int s_block_shift; + unsigned long **s_imap; + int s_imap_size; + struct mutex s_bitmap_lock; + int s_uid; + int s_gid; + int s_dmask; + int s_fmask; +}; + +/* convert a cluster number to a scaled block number */ +static inline sector_t clus_to_blk(struct omfs_sb_info *sbi, sector_t block) +{ + return block << sbi->s_block_shift; +} + +static inline struct omfs_sb_info *OMFS_SB(struct super_block *sb) +{ + return sb->s_fs_info; +} + +/* bitmap.c */ +extern unsigned long omfs_count_free(struct super_block *sb); +extern int omfs_allocate_block(struct super_block *sb, u64 block); +extern int omfs_allocate_range(struct super_block *sb, int min_request, + int max_request, u64 *return_block, int *return_size); +extern int omfs_clear_range(struct super_block *sb, u64 block, int count); + +/* dir.c */ +extern struct file_operations omfs_dir_operations; +extern struct inode_operations omfs_dir_inops; +extern int omfs_make_empty(struct inode *inode, struct super_block *sb); +extern int omfs_is_bad(struct omfs_sb_info *sbi, struct omfs_header *header, + u64 fsblock); + +/* file.c */ +extern struct file_operations omfs_file_operations; +extern struct inode_operations omfs_file_inops; +extern struct address_space_operations omfs_aops; +extern void omfs_make_empty_table(struct buffer_head *bh, int offset); +extern int omfs_shrink_inode(struct inode *inode); + +/* inode.c */ +extern struct inode *omfs_iget(struct super_block *sb, ino_t inode); +extern struct inode *omfs_new_inode(struct inode *dir, int mode); +extern int omfs_reserve_block(struct super_block *sb, sector_t block); +extern int omfs_find_empty_block(struct super_block *sb, int mode, ino_t *ino); +extern int omfs_sync_inode(struct inode *inode); + +#endif diff --git a/fs/omfs/omfs_fs.h b/fs/omfs/omfs_fs.h new file mode 100644 index 000000000000..12cca245d6e8 --- /dev/null +++ b/fs/omfs/omfs_fs.h @@ -0,0 +1,80 @@ +#ifndef _OMFS_FS_H +#define _OMFS_FS_H + +/* OMFS On-disk structures */ + +#define OMFS_MAGIC 0xC2993D87 +#define OMFS_IMAGIC 0xD2 + +#define OMFS_DIR 'D' +#define OMFS_FILE 'F' +#define OMFS_INODE_NORMAL 'e' +#define OMFS_INODE_CONTINUATION 'c' +#define OMFS_INODE_SYSTEM 's' +#define OMFS_NAMELEN 256 +#define OMFS_DIR_START 0x1b8 +#define OMFS_EXTENT_START 0x1d0 +#define OMFS_EXTENT_CONT 0x40 +#define OMFS_XOR_COUNT 19 +#define OMFS_MAX_BLOCK_SIZE 8192 + +struct omfs_super_block { + char s_fill1[256]; + __be64 s_root_block; /* block number of omfs_root_block */ + __be64 s_num_blocks; /* total number of FS blocks */ + __be32 s_magic; /* OMFS_MAGIC */ + __be32 s_blocksize; /* size of a block */ + __be32 s_mirrors; /* # of mirrors of system blocks */ + __be32 s_sys_blocksize; /* size of non-data blocks */ +}; + +struct omfs_header { + __be64 h_self; /* FS block where this is located */ + __be32 h_body_size; /* size of useful data after header */ + __be16 h_crc; /* crc-ccitt of body_size bytes */ + char h_fill1[2]; + u8 h_version; /* version, always 1 */ + char h_type; /* OMFS_INODE_X */ + u8 h_magic; /* OMFS_IMAGIC */ + u8 h_check_xor; /* XOR of header bytes before this */ + __be32 h_fill2; +}; + +struct omfs_root_block { + struct omfs_header r_head; /* header */ + __be64 r_fill1; + __be64 r_num_blocks; /* total number of FS blocks */ + __be64 r_root_dir; /* block # of root directory */ + __be64 r_bitmap; /* block # of free space bitmap */ + __be32 r_blocksize; /* size of a block */ + __be32 r_clustersize; /* size allocated for data blocks */ + __be64 r_mirrors; /* # of mirrors of system blocks */ + char r_name[OMFS_NAMELEN]; /* partition label */ +}; + +struct omfs_inode { + struct omfs_header i_head; /* header */ + __be64 i_parent; /* parent containing this inode */ + __be64 i_sibling; /* next inode in hash bucket */ + __be64 i_ctime; /* ctime, in milliseconds */ + char i_fill1[35]; + char i_type; /* OMFS_[DIR,FILE] */ + __be32 i_fill2; + char i_fill3[64]; + char i_name[OMFS_NAMELEN]; /* filename */ + __be64 i_size; /* size of file, in bytes */ +}; + +struct omfs_extent_entry { + __be64 e_cluster; /* start location of a set of blocks */ + __be64 e_blocks; /* number of blocks after e_cluster */ +}; + +struct omfs_extent { + __be64 e_next; /* next extent table location */ + __be32 e_extent_count; /* total # extents in this table */ + __be32 e_fill; + struct omfs_extent_entry e_entry; /* start of extent entries */ +}; + +#endif -- cgit v1.2.3 From 555e3775ced1d05203934fc6529bbf0560dd8733 Mon Sep 17 00:00:00 2001 From: Bob Copeland Date: Fri, 25 Jul 2008 19:45:15 -0700 Subject: omfs: add inode routines Add basic superblock and inode handling routines for OMFS Signed-off-by: Bob Copeland Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/omfs/inode.c | 553 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 553 insertions(+) create mode 100644 fs/omfs/inode.c (limited to 'fs') diff --git a/fs/omfs/inode.c b/fs/omfs/inode.c new file mode 100644 index 000000000000..d865f5535436 --- /dev/null +++ b/fs/omfs/inode.c @@ -0,0 +1,553 @@ +/* + * Optimized MPEG FS - inode and super operations. + * Copyright (C) 2006 Bob Copeland + * Released under GPL v2. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "omfs.h" + +MODULE_AUTHOR("Bob Copeland "); +MODULE_DESCRIPTION("OMFS (ReplayTV/Karma) Filesystem for Linux"); +MODULE_LICENSE("GPL"); + +struct inode *omfs_new_inode(struct inode *dir, int mode) +{ + struct inode *inode; + u64 new_block; + int err; + int len; + struct omfs_sb_info *sbi = OMFS_SB(dir->i_sb); + + inode = new_inode(dir->i_sb); + if (!inode) + return ERR_PTR(-ENOMEM); + + err = omfs_allocate_range(dir->i_sb, sbi->s_mirrors, sbi->s_mirrors, + &new_block, &len); + if (err) + goto fail; + + inode->i_ino = new_block; + inode->i_mode = mode; + inode->i_uid = current->fsuid; + inode->i_gid = current->fsgid; + inode->i_blocks = 0; + inode->i_mapping->a_ops = &omfs_aops; + + inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; + switch (mode & S_IFMT) { + case S_IFDIR: + inode->i_op = &omfs_dir_inops; + inode->i_fop = &omfs_dir_operations; + inode->i_size = sbi->s_sys_blocksize; + inc_nlink(inode); + break; + case S_IFREG: + inode->i_op = &omfs_file_inops; + inode->i_fop = &omfs_file_operations; + inode->i_size = 0; + break; + } + + insert_inode_hash(inode); + mark_inode_dirty(inode); + return inode; +fail: + make_bad_inode(inode); + iput(inode); + return ERR_PTR(err); +} + +/* + * Update the header checksums for a dirty inode based on its contents. + * Caller is expected to hold the buffer head underlying oi and mark it + * dirty. + */ +static void omfs_update_checksums(struct omfs_inode *oi) +{ + int xor, i, ofs = 0, count; + u16 crc = 0; + unsigned char *ptr = (unsigned char *) oi; + + count = be32_to_cpu(oi->i_head.h_body_size); + ofs = sizeof(struct omfs_header); + + crc = crc_itu_t(crc, ptr + ofs, count); + oi->i_head.h_crc = cpu_to_be16(crc); + + xor = ptr[0]; + for (i = 1; i < OMFS_XOR_COUNT; i++) + xor ^= ptr[i]; + + oi->i_head.h_check_xor = xor; +} + +static int omfs_write_inode(struct inode *inode, int wait) +{ + struct omfs_inode *oi; + struct omfs_sb_info *sbi = OMFS_SB(inode->i_sb); + struct buffer_head *bh, *bh2; + unsigned int block; + u64 ctime; + int i; + int ret = -EIO; + int sync_failed = 0; + + /* get current inode since we may have written sibling ptrs etc. */ + block = clus_to_blk(sbi, inode->i_ino); + bh = sb_bread(inode->i_sb, block); + if (!bh) + goto out; + + oi = (struct omfs_inode *) bh->b_data; + + oi->i_head.h_self = cpu_to_be64(inode->i_ino); + if (S_ISDIR(inode->i_mode)) + oi->i_type = OMFS_DIR; + else if (S_ISREG(inode->i_mode)) + oi->i_type = OMFS_FILE; + else { + printk(KERN_WARNING "omfs: unknown file type: %d\n", + inode->i_mode); + goto out_brelse; + } + + oi->i_head.h_body_size = cpu_to_be32(sbi->s_sys_blocksize - + sizeof(struct omfs_header)); + oi->i_head.h_version = 1; + oi->i_head.h_type = OMFS_INODE_NORMAL; + oi->i_head.h_magic = OMFS_IMAGIC; + oi->i_size = cpu_to_be64(inode->i_size); + + ctime = inode->i_ctime.tv_sec * 1000LL + + ((inode->i_ctime.tv_nsec + 999)/1000); + oi->i_ctime = cpu_to_be64(ctime); + + omfs_update_checksums(oi); + + mark_buffer_dirty(bh); + if (wait) { + sync_dirty_buffer(bh); + if (buffer_req(bh) && !buffer_uptodate(bh)) + sync_failed = 1; + } + + /* if mirroring writes, copy to next fsblock */ + for (i = 1; i < sbi->s_mirrors; i++) { + bh2 = sb_bread(inode->i_sb, block + i * + (sbi->s_blocksize / sbi->s_sys_blocksize)); + if (!bh2) + goto out_brelse; + + memcpy(bh2->b_data, bh->b_data, bh->b_size); + mark_buffer_dirty(bh2); + if (wait) { + sync_dirty_buffer(bh2); + if (buffer_req(bh2) && !buffer_uptodate(bh2)) + sync_failed = 1; + } + brelse(bh2); + } + ret = (sync_failed) ? -EIO : 0; +out_brelse: + brelse(bh); +out: + return ret; +} + +int omfs_sync_inode(struct inode *inode) +{ + return omfs_write_inode(inode, 1); +} + +/* + * called when an entry is deleted, need to clear the bits in the + * bitmaps. + */ +static void omfs_delete_inode(struct inode *inode) +{ + truncate_inode_pages(&inode->i_data, 0); + + if (S_ISREG(inode->i_mode)) { + inode->i_size = 0; + omfs_shrink_inode(inode); + } + + omfs_clear_range(inode->i_sb, inode->i_ino, 2); + clear_inode(inode); +} + +struct inode *omfs_iget(struct super_block *sb, ino_t ino) +{ + struct omfs_sb_info *sbi = OMFS_SB(sb); + struct omfs_inode *oi; + struct buffer_head *bh; + unsigned int block; + u64 ctime; + unsigned long nsecs; + struct inode *inode; + + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + + block = clus_to_blk(sbi, ino); + bh = sb_bread(inode->i_sb, block); + if (!bh) + goto iget_failed; + + oi = (struct omfs_inode *)bh->b_data; + + /* check self */ + if (ino != be64_to_cpu(oi->i_head.h_self)) + goto fail_bh; + + inode->i_uid = sbi->s_uid; + inode->i_gid = sbi->s_gid; + + ctime = be64_to_cpu(oi->i_ctime); + nsecs = do_div(ctime, 1000) * 1000L; + + inode->i_atime.tv_sec = ctime; + inode->i_mtime.tv_sec = ctime; + inode->i_ctime.tv_sec = ctime; + inode->i_atime.tv_nsec = nsecs; + inode->i_mtime.tv_nsec = nsecs; + inode->i_ctime.tv_nsec = nsecs; + + inode->i_mapping->a_ops = &omfs_aops; + + switch (oi->i_type) { + case OMFS_DIR: + inode->i_mode = S_IFDIR | (S_IRWXUGO & ~sbi->s_dmask); + inode->i_op = &omfs_dir_inops; + inode->i_fop = &omfs_dir_operations; + inode->i_size = be32_to_cpu(oi->i_head.h_body_size) + + sizeof(struct omfs_header); + inc_nlink(inode); + break; + case OMFS_FILE: + inode->i_mode = S_IFREG | (S_IRWXUGO & ~sbi->s_fmask); + inode->i_fop = &omfs_file_operations; + inode->i_size = be64_to_cpu(oi->i_size); + break; + } + brelse(bh); + unlock_new_inode(inode); + return inode; +fail_bh: + brelse(bh); +iget_failed: + iget_failed(inode); + return ERR_PTR(-EIO); +} + +static void omfs_put_super(struct super_block *sb) +{ + struct omfs_sb_info *sbi = OMFS_SB(sb); + kfree(sbi->s_imap); + kfree(sbi); + sb->s_fs_info = NULL; +} + +static int omfs_statfs(struct dentry *dentry, struct kstatfs *buf) +{ + struct super_block *s = dentry->d_sb; + struct omfs_sb_info *sbi = OMFS_SB(s); + buf->f_type = OMFS_MAGIC; + buf->f_bsize = sbi->s_blocksize; + buf->f_blocks = sbi->s_num_blocks; + buf->f_files = sbi->s_num_blocks; + buf->f_namelen = OMFS_NAMELEN; + + buf->f_bfree = buf->f_bavail = buf->f_ffree = + omfs_count_free(s); + return 0; +} + +static struct super_operations omfs_sops = { + .write_inode = omfs_write_inode, + .delete_inode = omfs_delete_inode, + .put_super = omfs_put_super, + .statfs = omfs_statfs, + .show_options = generic_show_options, +}; + +/* + * For Rio Karma, there is an on-disk free bitmap whose location is + * stored in the root block. For ReplayTV, there is no such free bitmap + * so we have to walk the tree. Both inodes and file data are allocated + * from the same map. This array can be big (300k) so we allocate + * in units of the blocksize. + */ +static int omfs_get_imap(struct super_block *sb) +{ + int bitmap_size; + int array_size; + int count; + struct omfs_sb_info *sbi = OMFS_SB(sb); + struct buffer_head *bh; + unsigned long **ptr; + sector_t block; + + bitmap_size = DIV_ROUND_UP(sbi->s_num_blocks, 8); + array_size = DIV_ROUND_UP(bitmap_size, sb->s_blocksize); + + if (sbi->s_bitmap_ino == ~0ULL) + goto out; + + sbi->s_imap_size = array_size; + sbi->s_imap = kzalloc(array_size * sizeof(unsigned long *), GFP_KERNEL); + if (!sbi->s_imap) + goto nomem; + + block = clus_to_blk(sbi, sbi->s_bitmap_ino); + ptr = sbi->s_imap; + for (count = bitmap_size; count > 0; count -= sb->s_blocksize) { + bh = sb_bread(sb, block++); + if (!bh) + goto nomem_free; + *ptr = kmalloc(sb->s_blocksize, GFP_KERNEL); + if (!*ptr) { + brelse(bh); + goto nomem_free; + } + memcpy(*ptr, bh->b_data, sb->s_blocksize); + if (count < sb->s_blocksize) + memset((void *)*ptr + count, 0xff, + sb->s_blocksize - count); + brelse(bh); + ptr++; + } +out: + return 0; + +nomem_free: + for (count = 0; count < array_size; count++) + kfree(sbi->s_imap[count]); + + kfree(sbi->s_imap); +nomem: + sbi->s_imap = NULL; + sbi->s_imap_size = 0; + return -ENOMEM; +} + +enum { + Opt_uid, Opt_gid, Opt_umask, Opt_dmask, Opt_fmask +}; + +static match_table_t tokens = { + {Opt_uid, "uid=%u"}, + {Opt_gid, "gid=%u"}, + {Opt_umask, "umask=%o"}, + {Opt_dmask, "dmask=%o"}, + {Opt_fmask, "fmask=%o"}, +}; + +static int parse_options(char *options, struct omfs_sb_info *sbi) +{ + char *p; + substring_t args[MAX_OPT_ARGS]; + int option; + + if (!options) + return 1; + + while ((p = strsep(&options, ",")) != NULL) { + int token; + if (!*p) + continue; + + token = match_token(p, tokens, args); + switch (token) { + case Opt_uid: + if (match_int(&args[0], &option)) + return 0; + sbi->s_uid = option; + break; + case Opt_gid: + if (match_int(&args[0], &option)) + return 0; + sbi->s_gid = option; + break; + case Opt_umask: + if (match_octal(&args[0], &option)) + return 0; + sbi->s_fmask = sbi->s_dmask = option; + break; + case Opt_dmask: + if (match_octal(&args[0], &option)) + return 0; + sbi->s_dmask = option; + break; + case Opt_fmask: + if (match_octal(&args[0], &option)) + return 0; + sbi->s_fmask = option; + break; + default: + return 0; + } + } + return 1; +} + +static int omfs_fill_super(struct super_block *sb, void *data, int silent) +{ + struct buffer_head *bh, *bh2; + struct omfs_super_block *omfs_sb; + struct omfs_root_block *omfs_rb; + struct omfs_sb_info *sbi; + struct inode *root; + sector_t start; + int ret = -EINVAL; + + save_mount_options(sb, (char *) data); + + sbi = kzalloc(sizeof(struct omfs_sb_info), GFP_KERNEL); + if (!sbi) + return -ENOMEM; + + sb->s_fs_info = sbi; + + sbi->s_uid = current->uid; + sbi->s_gid = current->gid; + sbi->s_dmask = sbi->s_fmask = current->fs->umask; + + if (!parse_options((char *) data, sbi)) + goto end; + + sb->s_maxbytes = 0xffffffff; + + sb_set_blocksize(sb, 0x200); + + bh = sb_bread(sb, 0); + if (!bh) + goto end; + + omfs_sb = (struct omfs_super_block *)bh->b_data; + + if (omfs_sb->s_magic != cpu_to_be32(OMFS_MAGIC)) { + if (!silent) + printk(KERN_ERR "omfs: Invalid superblock (%x)\n", + omfs_sb->s_magic); + goto out_brelse_bh; + } + sb->s_magic = OMFS_MAGIC; + + sbi->s_num_blocks = be64_to_cpu(omfs_sb->s_num_blocks); + sbi->s_blocksize = be32_to_cpu(omfs_sb->s_blocksize); + sbi->s_mirrors = be32_to_cpu(omfs_sb->s_mirrors); + sbi->s_root_ino = be64_to_cpu(omfs_sb->s_root_block); + sbi->s_sys_blocksize = be32_to_cpu(omfs_sb->s_sys_blocksize); + mutex_init(&sbi->s_bitmap_lock); + + if (sbi->s_sys_blocksize > PAGE_SIZE) { + printk(KERN_ERR "omfs: sysblock size (%d) is out of range\n", + sbi->s_sys_blocksize); + goto out_brelse_bh; + } + + if (sbi->s_blocksize < sbi->s_sys_blocksize || + sbi->s_blocksize > OMFS_MAX_BLOCK_SIZE) { + printk(KERN_ERR "omfs: block size (%d) is out of range\n", + sbi->s_blocksize); + goto out_brelse_bh; + } + + /* + * Use sys_blocksize as the fs block since it is smaller than a + * page while the fs blocksize can be larger. + */ + sb_set_blocksize(sb, sbi->s_sys_blocksize); + + /* + * ...and the difference goes into a shift. sys_blocksize is always + * a power of two factor of blocksize. + */ + sbi->s_block_shift = get_bitmask_order(sbi->s_blocksize) - + get_bitmask_order(sbi->s_sys_blocksize); + + start = clus_to_blk(sbi, be64_to_cpu(omfs_sb->s_root_block)); + bh2 = sb_bread(sb, start); + if (!bh2) + goto out_brelse_bh; + + omfs_rb = (struct omfs_root_block *)bh2->b_data; + + sbi->s_bitmap_ino = be64_to_cpu(omfs_rb->r_bitmap); + sbi->s_clustersize = be32_to_cpu(omfs_rb->r_clustersize); + + if (sbi->s_num_blocks != be64_to_cpu(omfs_rb->r_num_blocks)) { + printk(KERN_ERR "omfs: block count discrepancy between " + "super and root blocks (%llx, %llx)\n", + sbi->s_num_blocks, be64_to_cpu(omfs_rb->r_num_blocks)); + goto out_brelse_bh2; + } + + ret = omfs_get_imap(sb); + if (ret) + goto out_brelse_bh2; + + sb->s_op = &omfs_sops; + + root = omfs_iget(sb, be64_to_cpu(omfs_rb->r_root_dir)); + if (IS_ERR(root)) { + ret = PTR_ERR(root); + goto out_brelse_bh2; + } + + sb->s_root = d_alloc_root(root); + if (!sb->s_root) { + iput(root); + goto out_brelse_bh2; + } + printk(KERN_DEBUG "omfs: Mounted volume %s\n", omfs_rb->r_name); + + ret = 0; +out_brelse_bh2: + brelse(bh2); +out_brelse_bh: + brelse(bh); +end: + return ret; +} + +static int omfs_get_sb(struct file_system_type *fs_type, + int flags, const char *dev_name, + void *data, struct vfsmount *m) +{ + return get_sb_bdev(fs_type, flags, dev_name, data, omfs_fill_super, m); +} + +static struct file_system_type omfs_fs_type = { + .owner = THIS_MODULE, + .name = "omfs", + .get_sb = omfs_get_sb, + .kill_sb = kill_block_super, + .fs_flags = FS_REQUIRES_DEV, +}; + +static int __init init_omfs_fs(void) +{ + return register_filesystem(&omfs_fs_type); +} + +static void __exit exit_omfs_fs(void) +{ + unregister_filesystem(&omfs_fs_type); +} + +module_init(init_omfs_fs); +module_exit(exit_omfs_fs); -- cgit v1.2.3 From a3ab7155ea21aadc8a4d5687e91b3d876973185e Mon Sep 17 00:00:00 2001 From: Bob Copeland Date: Fri, 25 Jul 2008 19:45:16 -0700 Subject: omfs: add directory routines Add lookup and directory management routines for OMFS. The filesystem uses hashing based on the filename and stores collisions, unordered, in siblings of files' inode structures. To support telldir, the current position in the hash table is encoded in fpos. Signed-off-by: Bob Copeland Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/omfs/dir.c | 504 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 504 insertions(+) create mode 100644 fs/omfs/dir.c (limited to 'fs') diff --git a/fs/omfs/dir.c b/fs/omfs/dir.c new file mode 100644 index 000000000000..05a5bc31e4bd --- /dev/null +++ b/fs/omfs/dir.c @@ -0,0 +1,504 @@ +/* + * OMFS (as used by RIO Karma) directory operations. + * Copyright (C) 2005 Bob Copeland + * Released under GPL v2. + */ + +#include +#include +#include +#include "omfs.h" + +static int omfs_hash(const char *name, int namelen, int mod) +{ + int i, hash = 0; + for (i = 0; i < namelen; i++) + hash ^= tolower(name[i]) << (i % 24); + return hash % mod; +} + +/* + * Finds the bucket for a given name and reads the containing block; + * *ofs is set to the offset of the first list entry. + */ +static struct buffer_head *omfs_get_bucket(struct inode *dir, + const char *name, int namelen, int *ofs) +{ + int nbuckets = (dir->i_size - OMFS_DIR_START)/8; + int block = clus_to_blk(OMFS_SB(dir->i_sb), dir->i_ino); + int bucket = omfs_hash(name, namelen, nbuckets); + + *ofs = OMFS_DIR_START + bucket * 8; + return sb_bread(dir->i_sb, block); +} + +static struct buffer_head *omfs_scan_list(struct inode *dir, u64 block, + const char *name, int namelen, + u64 *prev_block) +{ + struct buffer_head *bh; + struct omfs_inode *oi; + int err = -ENOENT; + *prev_block = ~0; + + while (block != ~0) { + bh = sb_bread(dir->i_sb, + clus_to_blk(OMFS_SB(dir->i_sb), block)); + if (!bh) { + err = -EIO; + goto err; + } + + oi = (struct omfs_inode *) bh->b_data; + if (omfs_is_bad(OMFS_SB(dir->i_sb), &oi->i_head, block)) { + brelse(bh); + goto err; + } + + if (strncmp(oi->i_name, name, namelen) == 0) + return bh; + + *prev_block = block; + block = be64_to_cpu(oi->i_sibling); + brelse(bh); + } +err: + return ERR_PTR(err); +} + +static struct buffer_head *omfs_find_entry(struct inode *dir, + const char *name, int namelen) +{ + struct buffer_head *bh; + int ofs; + u64 block, dummy; + + bh = omfs_get_bucket(dir, name, namelen, &ofs); + if (!bh) + return ERR_PTR(-EIO); + + block = be64_to_cpu(*((__be64 *) &bh->b_data[ofs])); + brelse(bh); + + return omfs_scan_list(dir, block, name, namelen, &dummy); +} + +int omfs_make_empty(struct inode *inode, struct super_block *sb) +{ + struct omfs_sb_info *sbi = OMFS_SB(sb); + int block = clus_to_blk(sbi, inode->i_ino); + struct buffer_head *bh; + struct omfs_inode *oi; + + bh = sb_bread(sb, block); + if (!bh) + return -ENOMEM; + + memset(bh->b_data, 0, sizeof(struct omfs_inode)); + + if (inode->i_mode & S_IFDIR) { + memset(&bh->b_data[OMFS_DIR_START], 0xff, + sbi->s_sys_blocksize - OMFS_DIR_START); + } else + omfs_make_empty_table(bh, OMFS_EXTENT_START); + + oi = (struct omfs_inode *) bh->b_data; + oi->i_head.h_self = cpu_to_be64(inode->i_ino); + oi->i_sibling = ~0ULL; + + mark_buffer_dirty(bh); + brelse(bh); + return 0; +} + +static int omfs_add_link(struct dentry *dentry, struct inode *inode) +{ + struct inode *dir = dentry->d_parent->d_inode; + const char *name = dentry->d_name.name; + int namelen = dentry->d_name.len; + struct omfs_inode *oi; + struct buffer_head *bh; + u64 block; + __be64 *entry; + int ofs; + + /* just prepend to head of queue in proper bucket */ + bh = omfs_get_bucket(dir, name, namelen, &ofs); + if (!bh) + goto out; + + entry = (__be64 *) &bh->b_data[ofs]; + block = be64_to_cpu(*entry); + *entry = cpu_to_be64(inode->i_ino); + mark_buffer_dirty(bh); + brelse(bh); + + /* now set the sibling and parent pointers on the new inode */ + bh = sb_bread(dir->i_sb, clus_to_blk(OMFS_SB(dir->i_sb), inode->i_ino)); + if (!bh) + goto out; + + oi = (struct omfs_inode *) bh->b_data; + memcpy(oi->i_name, name, namelen); + memset(oi->i_name + namelen, 0, OMFS_NAMELEN - namelen); + oi->i_sibling = cpu_to_be64(block); + oi->i_parent = cpu_to_be64(dir->i_ino); + mark_buffer_dirty(bh); + brelse(bh); + + dir->i_ctime = CURRENT_TIME_SEC; + + /* mark affected inodes dirty to rebuild checksums */ + mark_inode_dirty(dir); + mark_inode_dirty(inode); + return 0; +out: + return -ENOMEM; +} + +static int omfs_delete_entry(struct dentry *dentry) +{ + struct inode *dir = dentry->d_parent->d_inode; + struct inode *dirty; + const char *name = dentry->d_name.name; + int namelen = dentry->d_name.len; + struct omfs_inode *oi; + struct buffer_head *bh, *bh2; + __be64 *entry, next; + u64 block, prev; + int ofs; + int err = -ENOMEM; + + /* delete the proper node in the bucket's linked list */ + bh = omfs_get_bucket(dir, name, namelen, &ofs); + if (!bh) + goto out; + + entry = (__be64 *) &bh->b_data[ofs]; + block = be64_to_cpu(*entry); + + bh2 = omfs_scan_list(dir, block, name, namelen, &prev); + if (IS_ERR(bh2)) { + err = PTR_ERR(bh2); + goto out_free_bh; + } + + oi = (struct omfs_inode *) bh2->b_data; + next = oi->i_sibling; + brelse(bh2); + + if (prev != ~0) { + /* found in middle of list, get list ptr */ + brelse(bh); + bh = sb_bread(dir->i_sb, + clus_to_blk(OMFS_SB(dir->i_sb), prev)); + if (!bh) + goto out; + + oi = (struct omfs_inode *) bh->b_data; + entry = &oi->i_sibling; + } + + *entry = next; + mark_buffer_dirty(bh); + + if (prev != ~0) { + dirty = omfs_iget(dir->i_sb, prev); + if (!IS_ERR(dirty)) { + mark_inode_dirty(dirty); + iput(dirty); + } + } + + err = 0; +out_free_bh: + brelse(bh); +out: + return err; +} + +static int omfs_dir_is_empty(struct inode *inode) +{ + int nbuckets = (inode->i_size - OMFS_DIR_START) / 8; + struct buffer_head *bh; + u64 *ptr; + int i; + + bh = sb_bread(inode->i_sb, clus_to_blk(OMFS_SB(inode->i_sb), + inode->i_ino)); + + if (!bh) + return 0; + + ptr = (u64 *) &bh->b_data[OMFS_DIR_START]; + + for (i = 0; i < nbuckets; i++, ptr++) + if (*ptr != ~0) + break; + + brelse(bh); + return *ptr != ~0; +} + +static int omfs_unlink(struct inode *dir, struct dentry *dentry) +{ + int ret; + struct inode *inode = dentry->d_inode; + + ret = omfs_delete_entry(dentry); + if (ret) + goto end_unlink; + + inode_dec_link_count(inode); + mark_inode_dirty(dir); + +end_unlink: + return ret; +} + +static int omfs_rmdir(struct inode *dir, struct dentry *dentry) +{ + int err = -ENOTEMPTY; + struct inode *inode = dentry->d_inode; + + if (omfs_dir_is_empty(inode)) { + err = omfs_unlink(dir, dentry); + if (!err) + inode_dec_link_count(inode); + } + return err; +} + +static int omfs_add_node(struct inode *dir, struct dentry *dentry, int mode) +{ + int err; + struct inode *inode = omfs_new_inode(dir, mode); + + if (IS_ERR(inode)) + return PTR_ERR(inode); + + err = omfs_make_empty(inode, dir->i_sb); + if (err) + goto out_free_inode; + + err = omfs_add_link(dentry, inode); + if (err) + goto out_free_inode; + + d_instantiate(dentry, inode); + return 0; + +out_free_inode: + iput(inode); + return err; +} + +static int omfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +{ + return omfs_add_node(dir, dentry, mode | S_IFDIR); +} + +static int omfs_create(struct inode *dir, struct dentry *dentry, int mode, + struct nameidata *nd) +{ + return omfs_add_node(dir, dentry, mode | S_IFREG); +} + +static struct dentry *omfs_lookup(struct inode *dir, struct dentry *dentry, + struct nameidata *nd) +{ + struct buffer_head *bh; + struct inode *inode = NULL; + + if (dentry->d_name.len > OMFS_NAMELEN) + return ERR_PTR(-ENAMETOOLONG); + + bh = omfs_find_entry(dir, dentry->d_name.name, dentry->d_name.len); + if (!IS_ERR(bh)) { + struct omfs_inode *oi = (struct omfs_inode *)bh->b_data; + ino_t ino = be64_to_cpu(oi->i_head.h_self); + brelse(bh); + inode = omfs_iget(dir->i_sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); + } + d_add(dentry, inode); + return NULL; +} + +/* sanity check block's self pointer */ +int omfs_is_bad(struct omfs_sb_info *sbi, struct omfs_header *header, + u64 fsblock) +{ + int is_bad; + u64 ino = be64_to_cpu(header->h_self); + is_bad = ((ino != fsblock) || (ino < sbi->s_root_ino) || + (ino > sbi->s_num_blocks)); + + if (is_bad) + printk(KERN_WARNING "omfs: bad hash chain detected\n"); + + return is_bad; +} + +static int omfs_fill_chain(struct file *filp, void *dirent, filldir_t filldir, + u64 fsblock, int hindex) +{ + struct inode *dir = filp->f_dentry->d_inode; + struct buffer_head *bh; + struct omfs_inode *oi; + u64 self; + int res = 0; + unsigned char d_type; + + /* follow chain in this bucket */ + while (fsblock != ~0) { + bh = sb_bread(dir->i_sb, clus_to_blk(OMFS_SB(dir->i_sb), + fsblock)); + if (!bh) + goto out; + + oi = (struct omfs_inode *) bh->b_data; + if (omfs_is_bad(OMFS_SB(dir->i_sb), &oi->i_head, fsblock)) { + brelse(bh); + goto out; + } + + self = fsblock; + fsblock = be64_to_cpu(oi->i_sibling); + + /* skip visited nodes */ + if (hindex) { + hindex--; + brelse(bh); + continue; + } + + d_type = (oi->i_type == OMFS_DIR) ? DT_DIR : DT_REG; + + res = filldir(dirent, oi->i_name, strnlen(oi->i_name, + OMFS_NAMELEN), filp->f_pos, self, d_type); + if (res == 0) + filp->f_pos++; + brelse(bh); + } +out: + return res; +} + +static int omfs_rename(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry) +{ + struct inode *new_inode = new_dentry->d_inode; + struct inode *old_inode = old_dentry->d_inode; + struct buffer_head *bh; + int is_dir; + int err; + + is_dir = S_ISDIR(old_inode->i_mode); + + if (new_inode) { + /* overwriting existing file/dir */ + err = -ENOTEMPTY; + if (is_dir && !omfs_dir_is_empty(new_inode)) + goto out; + + err = -ENOENT; + bh = omfs_find_entry(new_dir, new_dentry->d_name.name, + new_dentry->d_name.len); + if (IS_ERR(bh)) + goto out; + brelse(bh); + + err = omfs_unlink(new_dir, new_dentry); + if (err) + goto out; + } + + /* since omfs locates files by name, we need to unlink _before_ + * adding the new link or we won't find the old one */ + inode_inc_link_count(old_inode); + err = omfs_unlink(old_dir, old_dentry); + if (err) { + inode_dec_link_count(old_inode); + goto out; + } + + err = omfs_add_link(new_dentry, old_inode); + if (err) + goto out; + + old_inode->i_ctime = CURRENT_TIME_SEC; +out: + return err; +} + +static int omfs_readdir(struct file *filp, void *dirent, filldir_t filldir) +{ + struct inode *dir = filp->f_dentry->d_inode; + struct buffer_head *bh; + loff_t offset, res; + unsigned int hchain, hindex; + int nbuckets; + u64 fsblock; + int ret = -EINVAL; + + if (filp->f_pos >> 32) + goto success; + + switch ((unsigned long) filp->f_pos) { + case 0: + if (filldir(dirent, ".", 1, 0, dir->i_ino, DT_DIR) < 0) + goto success; + filp->f_pos++; + /* fall through */ + case 1: + if (filldir(dirent, "..", 2, 1, + parent_ino(filp->f_dentry), DT_DIR) < 0) + goto success; + filp->f_pos = 1 << 20; + /* fall through */ + } + + nbuckets = (dir->i_size - OMFS_DIR_START) / 8; + + /* high 12 bits store bucket + 1 and low 20 bits store hash index */ + hchain = (filp->f_pos >> 20) - 1; + hindex = filp->f_pos & 0xfffff; + + bh = sb_bread(dir->i_sb, clus_to_blk(OMFS_SB(dir->i_sb), dir->i_ino)); + if (!bh) + goto out; + + offset = OMFS_DIR_START + hchain * 8; + + for (; hchain < nbuckets; hchain++, offset += 8) { + fsblock = be64_to_cpu(*((__be64 *) &bh->b_data[offset])); + + res = omfs_fill_chain(filp, dirent, filldir, fsblock, hindex); + hindex = 0; + if (res < 0) + break; + + filp->f_pos = (hchain+2) << 20; + } + brelse(bh); +success: + ret = 0; +out: + return ret; +} + +struct inode_operations omfs_dir_inops = { + .lookup = omfs_lookup, + .mkdir = omfs_mkdir, + .rename = omfs_rename, + .create = omfs_create, + .unlink = omfs_unlink, + .rmdir = omfs_rmdir, +}; + +struct file_operations omfs_dir_operations = { + .read = generic_read_dir, + .readdir = omfs_readdir, +}; -- cgit v1.2.3 From 8f09e98768c17287df076580c4cc72ac358312c6 Mon Sep 17 00:00:00 2001 From: Bob Copeland Date: Fri, 25 Jul 2008 19:45:16 -0700 Subject: omfs: add file routines Add functions for reading and manipulating the storage of file data in the extent-based OMFS. Signed-off-by: Bob Copeland Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/omfs/file.c | 346 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 346 insertions(+) create mode 100644 fs/omfs/file.c (limited to 'fs') diff --git a/fs/omfs/file.c b/fs/omfs/file.c new file mode 100644 index 000000000000..66e01fae4384 --- /dev/null +++ b/fs/omfs/file.c @@ -0,0 +1,346 @@ +/* + * OMFS (as used by RIO Karma) file operations. + * Copyright (C) 2005 Bob Copeland + * Released under GPL v2. + */ + +#include +#include +#include +#include +#include +#include "omfs.h" + +static int omfs_sync_file(struct file *file, struct dentry *dentry, + int datasync) +{ + struct inode *inode = dentry->d_inode; + int err; + + err = sync_mapping_buffers(inode->i_mapping); + if (!(inode->i_state & I_DIRTY)) + return err; + if (datasync && !(inode->i_state & I_DIRTY_DATASYNC)) + return err; + err |= omfs_sync_inode(inode); + return err ? -EIO : 0; +} + +void omfs_make_empty_table(struct buffer_head *bh, int offset) +{ + struct omfs_extent *oe = (struct omfs_extent *) &bh->b_data[offset]; + + oe->e_next = ~0ULL; + oe->e_extent_count = cpu_to_be32(1), + oe->e_fill = cpu_to_be32(0x22), + oe->e_entry.e_cluster = ~0ULL; + oe->e_entry.e_blocks = ~0ULL; +} + +int omfs_shrink_inode(struct inode *inode) +{ + struct omfs_sb_info *sbi = OMFS_SB(inode->i_sb); + struct omfs_extent *oe; + struct omfs_extent_entry *entry; + struct buffer_head *bh; + u64 next, last; + u32 extent_count; + int ret; + + /* traverse extent table, freeing each entry that is greater + * than inode->i_size; + */ + next = inode->i_ino; + + /* only support truncate -> 0 for now */ + ret = -EIO; + if (inode->i_size != 0) + goto out; + + bh = sb_bread(inode->i_sb, clus_to_blk(sbi, next)); + if (!bh) + goto out; + + oe = (struct omfs_extent *)(&bh->b_data[OMFS_EXTENT_START]); + + for (;;) { + + if (omfs_is_bad(sbi, (struct omfs_header *) bh->b_data, next)) { + brelse(bh); + goto out; + } + + extent_count = be32_to_cpu(oe->e_extent_count); + last = next; + next = be64_to_cpu(oe->e_next); + entry = &oe->e_entry; + + /* ignore last entry as it is the terminator */ + for (; extent_count > 1; extent_count--) { + u64 start, count; + start = be64_to_cpu(entry->e_cluster); + count = be64_to_cpu(entry->e_blocks); + + omfs_clear_range(inode->i_sb, start, (int) count); + entry++; + } + omfs_make_empty_table(bh, (char *) oe - bh->b_data); + mark_buffer_dirty(bh); + brelse(bh); + + if (last != inode->i_ino) + omfs_clear_range(inode->i_sb, last, sbi->s_mirrors); + + if (next == ~0) + break; + + bh = sb_bread(inode->i_sb, clus_to_blk(sbi, next)); + if (!bh) + goto out; + oe = (struct omfs_extent *) (&bh->b_data[OMFS_EXTENT_CONT]); + } + ret = 0; +out: + return ret; +} + +static void omfs_truncate(struct inode *inode) +{ + omfs_shrink_inode(inode); + mark_inode_dirty(inode); +} + +/* + * Add new blocks to the current extent, or create new entries/continuations + * as necessary. + */ +static int omfs_grow_extent(struct inode *inode, struct omfs_extent *oe, + u64 *ret_block) +{ + struct omfs_extent_entry *terminator; + struct omfs_extent_entry *entry = &oe->e_entry; + struct omfs_sb_info *sbi = OMFS_SB(inode->i_sb); + u32 extent_count = be32_to_cpu(oe->e_extent_count); + u64 new_block = 0; + u32 max_count; + int new_count; + int ret = 0; + + /* reached the end of the extent table with no blocks mapped. + * there are three possibilities for adding: grow last extent, + * add a new extent to the current extent table, and add a + * continuation inode. in last two cases need an allocator for + * sbi->s_cluster_size + */ + + /* TODO: handle holes */ + + /* should always have a terminator */ + if (extent_count < 1) + return -EIO; + + /* trivially grow current extent, if next block is not taken */ + terminator = entry + extent_count - 1; + if (extent_count > 1) { + entry = terminator-1; + new_block = be64_to_cpu(entry->e_cluster) + + be64_to_cpu(entry->e_blocks); + + if (omfs_allocate_block(inode->i_sb, new_block)) { + entry->e_blocks = + cpu_to_be64(be64_to_cpu(entry->e_blocks) + 1); + terminator->e_blocks = ~(cpu_to_be64( + be64_to_cpu(~terminator->e_blocks) + 1)); + goto out; + } + } + max_count = (sbi->s_sys_blocksize - OMFS_EXTENT_START - + sizeof(struct omfs_extent)) / + sizeof(struct omfs_extent_entry) + 1; + + /* TODO: add a continuation block here */ + if (be32_to_cpu(oe->e_extent_count) > max_count-1) + return -EIO; + + /* try to allocate a new cluster */ + ret = omfs_allocate_range(inode->i_sb, 1, sbi->s_clustersize, + &new_block, &new_count); + if (ret) + goto out_fail; + + /* copy terminator down an entry */ + entry = terminator; + terminator++; + memcpy(terminator, entry, sizeof(struct omfs_extent_entry)); + + entry->e_cluster = cpu_to_be64(new_block); + entry->e_blocks = cpu_to_be64((u64) new_count); + + terminator->e_blocks = ~(cpu_to_be64( + be64_to_cpu(~terminator->e_blocks) + (u64) new_count)); + + /* write in new entry */ + oe->e_extent_count = cpu_to_be32(1 + be32_to_cpu(oe->e_extent_count)); + +out: + *ret_block = new_block; +out_fail: + return ret; +} + +/* + * Scans across the directory table for a given file block number. + * If block not found, return 0. + */ +static sector_t find_block(struct inode *inode, struct omfs_extent_entry *ent, + sector_t block, int count, int *left) +{ + /* count > 1 because of terminator */ + sector_t searched = 0; + for (; count > 1; count--) { + int numblocks = clus_to_blk(OMFS_SB(inode->i_sb), + be64_to_cpu(ent->e_blocks)); + + if (block >= searched && + block < searched + numblocks) { + /* + * found it at cluster + (block - searched) + * numblocks - (block - searched) is remainder + */ + *left = numblocks - (block - searched); + return clus_to_blk(OMFS_SB(inode->i_sb), + be64_to_cpu(ent->e_cluster)) + + block - searched; + } + searched += numblocks; + ent++; + } + return 0; +} + +static int omfs_get_block(struct inode *inode, sector_t block, + struct buffer_head *bh_result, int create) +{ + struct buffer_head *bh; + sector_t next, offset; + int ret; + u64 new_block; + int extent_count; + struct omfs_extent *oe; + struct omfs_extent_entry *entry; + struct omfs_sb_info *sbi = OMFS_SB(inode->i_sb); + int max_blocks = bh_result->b_size >> inode->i_blkbits; + int remain; + + ret = -EIO; + bh = sb_bread(inode->i_sb, clus_to_blk(sbi, inode->i_ino)); + if (!bh) + goto out; + + oe = (struct omfs_extent *)(&bh->b_data[OMFS_EXTENT_START]); + next = inode->i_ino; + + for (;;) { + + if (omfs_is_bad(sbi, (struct omfs_header *) bh->b_data, next)) + goto out_brelse; + + extent_count = be32_to_cpu(oe->e_extent_count); + next = be64_to_cpu(oe->e_next); + entry = &oe->e_entry; + + offset = find_block(inode, entry, block, extent_count, &remain); + if (offset > 0) { + ret = 0; + map_bh(bh_result, inode->i_sb, offset); + if (remain > max_blocks) + remain = max_blocks; + bh_result->b_size = (remain << inode->i_blkbits); + goto out_brelse; + } + if (next == ~0) + break; + + brelse(bh); + bh = sb_bread(inode->i_sb, clus_to_blk(sbi, next)); + if (!bh) + goto out; + oe = (struct omfs_extent *) (&bh->b_data[OMFS_EXTENT_CONT]); + } + if (create) { + ret = omfs_grow_extent(inode, oe, &new_block); + if (ret == 0) { + mark_buffer_dirty(bh); + mark_inode_dirty(inode); + map_bh(bh_result, inode->i_sb, + clus_to_blk(sbi, new_block)); + } + } +out_brelse: + brelse(bh); +out: + return ret; +} + +static int omfs_readpage(struct file *file, struct page *page) +{ + return block_read_full_page(page, omfs_get_block); +} + +static int omfs_readpages(struct file *file, struct address_space *mapping, + struct list_head *pages, unsigned nr_pages) +{ + return mpage_readpages(mapping, pages, nr_pages, omfs_get_block); +} + +static int omfs_writepage(struct page *page, struct writeback_control *wbc) +{ + return block_write_full_page(page, omfs_get_block, wbc); +} + +static int +omfs_writepages(struct address_space *mapping, struct writeback_control *wbc) +{ + return mpage_writepages(mapping, wbc, omfs_get_block); +} + +static int omfs_write_begin(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata) +{ + *pagep = NULL; + return block_write_begin(file, mapping, pos, len, flags, + pagep, fsdata, omfs_get_block); +} + +static sector_t omfs_bmap(struct address_space *mapping, sector_t block) +{ + return generic_block_bmap(mapping, block, omfs_get_block); +} + +struct file_operations omfs_file_operations = { + .llseek = generic_file_llseek, + .read = do_sync_read, + .write = do_sync_write, + .aio_read = generic_file_aio_read, + .aio_write = generic_file_aio_write, + .mmap = generic_file_mmap, + .fsync = omfs_sync_file, + .splice_read = generic_file_splice_read, +}; + +struct inode_operations omfs_file_inops = { + .truncate = omfs_truncate +}; + +struct address_space_operations omfs_aops = { + .readpage = omfs_readpage, + .readpages = omfs_readpages, + .writepage = omfs_writepage, + .writepages = omfs_writepages, + .sync_page = block_sync_page, + .write_begin = omfs_write_begin, + .write_end = generic_write_end, + .bmap = omfs_bmap, +}; + -- cgit v1.2.3 From 36cc410a6799a205bfc6ccc38abd9d52f2afba64 Mon Sep 17 00:00:00 2001 From: Bob Copeland Date: Fri, 25 Jul 2008 19:45:17 -0700 Subject: omfs: add bitmap routines Add block allocation and block bitmap management routines for OMFS. Signed-off-by: Bob Copeland Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/omfs/bitmap.c | 192 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 192 insertions(+) create mode 100644 fs/omfs/bitmap.c (limited to 'fs') diff --git a/fs/omfs/bitmap.c b/fs/omfs/bitmap.c new file mode 100644 index 000000000000..dc75f22be3f2 --- /dev/null +++ b/fs/omfs/bitmap.c @@ -0,0 +1,192 @@ +#include +#include +#include +#include +#include "omfs.h" + +unsigned long omfs_count_free(struct super_block *sb) +{ + unsigned int i; + unsigned long sum = 0; + struct omfs_sb_info *sbi = OMFS_SB(sb); + int nbits = sb->s_blocksize * 8; + + for (i = 0; i < sbi->s_imap_size; i++) + sum += nbits - bitmap_weight(sbi->s_imap[i], nbits); + + return sum; +} + +/* + * Counts the run of zero bits starting at bit up to max. + * It handles the case where a run might spill over a buffer. + * Called with bitmap lock. + */ +static int count_run(unsigned long **addr, int nbits, + int addrlen, int bit, int max) +{ + int count = 0; + int x; + + for (; addrlen > 0; addrlen--, addr++) { + x = find_next_bit(*addr, nbits, bit); + count += x - bit; + + if (x < nbits || count > max) + return min(count, max); + + bit = 0; + } + return min(count, max); +} + +/* + * Sets or clears the run of count bits starting with bit. + * Called with bitmap lock. + */ +static int set_run(struct super_block *sb, int map, + int nbits, int bit, int count, int set) +{ + int i; + int err; + struct buffer_head *bh; + struct omfs_sb_info *sbi = OMFS_SB(sb); + + err = -ENOMEM; + bh = sb_bread(sb, clus_to_blk(sbi, sbi->s_bitmap_ino) + map); + if (!bh) + goto out; + + for (i = 0; i < count; i++, bit++) { + if (bit >= nbits) { + bit = 0; + map++; + + mark_buffer_dirty(bh); + brelse(bh); + bh = sb_bread(sb, + clus_to_blk(sbi, sbi->s_bitmap_ino) + map); + if (!bh) + goto out; + } + if (set) { + set_bit(bit, sbi->s_imap[map]); + set_bit(bit, (long *) bh->b_data); + } else { + clear_bit(bit, sbi->s_imap[map]); + clear_bit(bit, (long *) bh->b_data); + } + } + mark_buffer_dirty(bh); + brelse(bh); + err = 0; +out: + return err; +} + +/* + * Tries to allocate exactly one block. Returns true if sucessful. + */ +int omfs_allocate_block(struct super_block *sb, u64 block) +{ + struct buffer_head *bh; + struct omfs_sb_info *sbi = OMFS_SB(sb); + int bits_per_entry = 8 * sb->s_blocksize; + int map, bit; + int ret = 0; + u64 tmp; + + tmp = block; + bit = do_div(tmp, bits_per_entry); + map = tmp; + + mutex_lock(&sbi->s_bitmap_lock); + if (map >= sbi->s_imap_size || test_and_set_bit(bit, sbi->s_imap[map])) + goto out; + + if (sbi->s_bitmap_ino > 0) { + bh = sb_bread(sb, clus_to_blk(sbi, sbi->s_bitmap_ino) + map); + if (!bh) + goto out; + + set_bit(bit, (long *) bh->b_data); + mark_buffer_dirty(bh); + brelse(bh); + } + ret = 1; +out: + mutex_unlock(&sbi->s_bitmap_lock); + return ret; +} + + +/* + * Tries to allocate a set of blocks. The request size depends on the + * type: for inodes, we must allocate sbi->s_mirrors blocks, and for file + * blocks, we try to allocate sbi->s_clustersize, but can always get away + * with just one block. + */ +int omfs_allocate_range(struct super_block *sb, + int min_request, + int max_request, + u64 *return_block, + int *return_size) +{ + struct omfs_sb_info *sbi = OMFS_SB(sb); + int bits_per_entry = 8 * sb->s_blocksize; + int ret = 0; + int i, run, bit; + + mutex_lock(&sbi->s_bitmap_lock); + for (i = 0; i < sbi->s_imap_size; i++) { + bit = 0; + while (bit < bits_per_entry) { + bit = find_next_zero_bit(sbi->s_imap[i], bits_per_entry, + bit); + + if (bit == bits_per_entry) + break; + + run = count_run(&sbi->s_imap[i], bits_per_entry, + sbi->s_imap_size-i, bit, max_request); + + if (run >= min_request) + goto found; + bit += run; + } + } + ret = -ENOSPC; + goto out; + +found: + *return_block = i * bits_per_entry + bit; + *return_size = run; + ret = set_run(sb, i, bits_per_entry, bit, run, 1); + +out: + mutex_unlock(&sbi->s_bitmap_lock); + return ret; +} + +/* + * Clears count bits starting at a given block. + */ +int omfs_clear_range(struct super_block *sb, u64 block, int count) +{ + struct omfs_sb_info *sbi = OMFS_SB(sb); + int bits_per_entry = 8 * sb->s_blocksize; + u64 tmp; + int map, bit, ret; + + tmp = block; + bit = do_div(tmp, bits_per_entry); + map = tmp; + + if (map >= sbi->s_imap_size) + return 0; + + mutex_lock(&sbi->s_bitmap_lock); + ret = set_run(sb, map, bits_per_entry, bit, count, 0); + mutex_unlock(&sbi->s_bitmap_lock); + return ret; +} -- cgit v1.2.3 From 63ca8ce2a2641f9cb5f0add33ced4591681d1cd7 Mon Sep 17 00:00:00 2001 From: Bob Copeland Date: Fri, 25 Jul 2008 19:45:17 -0700 Subject: omfs: update kbuild to include OMFS Adds OMFS to the fs Kconfig and Makefile Signed-off-by: Bob Copeland Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/Kconfig | 13 +++++++++++++ fs/Makefile | 1 + fs/omfs/Makefile | 4 ++++ 3 files changed, 18 insertions(+) create mode 100644 fs/omfs/Makefile (limited to 'fs') diff --git a/fs/Kconfig b/fs/Kconfig index 97e3bdedb1e6..d3873583360b 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -1383,6 +1383,19 @@ config MINIX_FS partition (the one containing the directory /) cannot be compiled as a module. +config OMFS_FS + tristate "SonicBlue Optimized MPEG File System support" + depends on BLOCK + select CRC_ITU_T + help + This is the proprietary file system used by the Rio Karma music + player and ReplayTV DVR. Despite the name, this filesystem is not + more efficient than a standard FS for MPEG files, in fact likely + the opposite is true. Say Y if you have either of these devices + and wish to mount its disk. + + To compile this file system support as a module, choose M here: the + module will be called omfs. If unsure, say N. config HPFS_FS tristate "OS/2 HPFS file system support" diff --git a/fs/Makefile b/fs/Makefile index 3b2178b4bb66..a1482a5eff15 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -111,6 +111,7 @@ obj-$(CONFIG_ADFS_FS) += adfs/ obj-$(CONFIG_FUSE_FS) += fuse/ obj-$(CONFIG_UDF_FS) += udf/ obj-$(CONFIG_SUN_OPENPROMFS) += openpromfs/ +obj-$(CONFIG_OMFS_FS) += omfs/ obj-$(CONFIG_JFS_FS) += jfs/ obj-$(CONFIG_XFS_FS) += xfs/ obj-$(CONFIG_9P_FS) += 9p/ diff --git a/fs/omfs/Makefile b/fs/omfs/Makefile new file mode 100644 index 000000000000..8b82b63f1129 --- /dev/null +++ b/fs/omfs/Makefile @@ -0,0 +1,4 @@ + +obj-$(CONFIG_OMFS_FS) += omfs.o + +omfs-y := bitmap.o dir.o file.o inode.o -- cgit v1.2.3 From f5dd33c494a427b1d1a3b574de5c9e511c888864 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Fri, 25 Jul 2008 19:45:25 -0700 Subject: dio: use get_user_pages_fast Use get_user_pages_fast in the common/generic block and fs direct IO paths. Signed-off-by: Nick Piggin Cc: Dave Kleikamp Cc: Andy Whitcroft Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Andi Kleen Cc: Dave Kleikamp Cc: Badari Pulavarty Cc: Zach Brown Cc: Jens Axboe Reviewed-by: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/bio.c | 8 ++------ fs/direct-io.c | 10 ++-------- 2 files changed, 4 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/bio.c b/fs/bio.c index 88322b066acb..25f1af0d81e5 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -721,12 +721,8 @@ static struct bio *__bio_map_user_iov(struct request_queue *q, const int local_nr_pages = end - start; const int page_limit = cur_page + local_nr_pages; - down_read(¤t->mm->mmap_sem); - ret = get_user_pages(current, current->mm, uaddr, - local_nr_pages, - write_to_vm, 0, &pages[cur_page], NULL); - up_read(¤t->mm->mmap_sem); - + ret = get_user_pages_fast(uaddr, local_nr_pages, + write_to_vm, &pages[cur_page]); if (ret < local_nr_pages) { ret = -EFAULT; goto out_unmap; diff --git a/fs/direct-io.c b/fs/direct-io.c index 9e81addbd6ea..9606ee848fd8 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -150,17 +150,11 @@ static int dio_refill_pages(struct dio *dio) int nr_pages; nr_pages = min(dio->total_pages - dio->curr_page, DIO_PAGES); - down_read(¤t->mm->mmap_sem); - ret = get_user_pages( - current, /* Task for fault acounting */ - current->mm, /* whose pages? */ + ret = get_user_pages_fast( dio->curr_user_address, /* Where from? */ nr_pages, /* How many pages? */ dio->rw == READ, /* Write to memory? */ - 0, /* force (?) */ - &dio->pages[0], - NULL); /* vmas */ - up_read(¤t->mm->mmap_sem); + &dio->pages[0]); /* Put results here */ if (ret < 0 && dio->blocks_available && (dio->rw & WRITE)) { struct page *page = ZERO_PAGE(0); -- cgit v1.2.3 From bc40d73c950146725e9e768e856a416ec8949065 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Fri, 25 Jul 2008 19:45:26 -0700 Subject: splice: use get_user_pages_fast Use get_user_pages_fast in splice. This reverts some mmap_sem batching there, however the biggest problem with mmap_sem tends to be hold times blocking out other threads rather than cacheline bouncing. Further: on architectures that implement get_user_pages_fast without locks, mmap_sem can be avoided completely anyway. Signed-off-by: Nick Piggin Cc: Dave Kleikamp Cc: Andy Whitcroft Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Andi Kleen Cc: Dave Kleikamp Cc: Badari Pulavarty Cc: Zach Brown Cc: Jens Axboe Reviewed-by: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/splice.c | 41 +++-------------------------------------- 1 file changed, 3 insertions(+), 38 deletions(-) (limited to 'fs') diff --git a/fs/splice.c b/fs/splice.c index 399442179d89..47dc1a445d1f 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -1160,36 +1160,6 @@ static long do_splice(struct file *in, loff_t __user *off_in, return -EINVAL; } -/* - * Do a copy-from-user while holding the mmap_semaphore for reading, in a - * manner safe from deadlocking with simultaneous mmap() (grabbing mmap_sem - * for writing) and page faulting on the user memory pointed to by src. - * This assumes that we will very rarely hit the partial != 0 path, or this - * will not be a win. - */ -static int copy_from_user_mmap_sem(void *dst, const void __user *src, size_t n) -{ - int partial; - - if (!access_ok(VERIFY_READ, src, n)) - return -EFAULT; - - pagefault_disable(); - partial = __copy_from_user_inatomic(dst, src, n); - pagefault_enable(); - - /* - * Didn't copy everything, drop the mmap_sem and do a faulting copy - */ - if (unlikely(partial)) { - up_read(¤t->mm->mmap_sem); - partial = copy_from_user(dst, src, n); - down_read(¤t->mm->mmap_sem); - } - - return partial; -} - /* * Map an iov into an array of pages and offset/length tupples. With the * partial_page structure, we can map several non-contiguous ranges into @@ -1203,8 +1173,6 @@ static int get_iovec_page_array(const struct iovec __user *iov, { int buffers = 0, error = 0; - down_read(¤t->mm->mmap_sem); - while (nr_vecs) { unsigned long off, npages; struct iovec entry; @@ -1213,7 +1181,7 @@ static int get_iovec_page_array(const struct iovec __user *iov, int i; error = -EFAULT; - if (copy_from_user_mmap_sem(&entry, iov, sizeof(entry))) + if (copy_from_user(&entry, iov, sizeof(entry))) break; base = entry.iov_base; @@ -1247,9 +1215,8 @@ static int get_iovec_page_array(const struct iovec __user *iov, if (npages > PIPE_BUFFERS - buffers) npages = PIPE_BUFFERS - buffers; - error = get_user_pages(current, current->mm, - (unsigned long) base, npages, 0, 0, - &pages[buffers], NULL); + error = get_user_pages_fast((unsigned long)base, npages, + 0, &pages[buffers]); if (unlikely(error <= 0)) break; @@ -1288,8 +1255,6 @@ static int get_iovec_page_array(const struct iovec __user *iov, iov++; } - up_read(¤t->mm->mmap_sem); - if (buffers) return buffers; -- cgit v1.2.3 From 19fd6231279be3c3bdd02ed99f9b0eb195978064 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Fri, 25 Jul 2008 19:45:32 -0700 Subject: mm: spinlock tree_lock mapping->tree_lock has no read lockers. convert the lock from an rwlock to a spinlock. Signed-off-by: Nick Piggin Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Hugh Dickins Cc: "Paul E. McKenney" Reviewed-by: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 4 ++-- fs/inode.c | 2 +- include/asm-arm/cacheflush.h | 4 ++-- include/asm-parisc/cacheflush.h | 4 ++-- include/linux/fs.h | 2 +- mm/filemap.c | 10 +++++----- mm/migrate.c | 11 +++++------ mm/page-writeback.c | 12 ++++++------ mm/swap_state.c | 10 +++++----- mm/swapfile.c | 4 ++-- mm/truncate.c | 6 +++--- mm/vmscan.c | 8 ++++---- 12 files changed, 38 insertions(+), 39 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index d48caee12e2a..109b261192d9 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -706,7 +706,7 @@ static int __set_page_dirty(struct page *page, if (TestSetPageDirty(page)) return 0; - write_lock_irq(&mapping->tree_lock); + spin_lock_irq(&mapping->tree_lock); if (page->mapping) { /* Race with truncate? */ WARN_ON_ONCE(warn && !PageUptodate(page)); @@ -719,7 +719,7 @@ static int __set_page_dirty(struct page *page, radix_tree_tag_set(&mapping->page_tree, page_index(page), PAGECACHE_TAG_DIRTY); } - write_unlock_irq(&mapping->tree_lock); + spin_unlock_irq(&mapping->tree_lock); __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); return 1; diff --git a/fs/inode.c b/fs/inode.c index c36d9480335c..35b6414522ea 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -209,7 +209,7 @@ void inode_init_once(struct inode *inode) INIT_LIST_HEAD(&inode->i_dentry); INIT_LIST_HEAD(&inode->i_devices); INIT_RADIX_TREE(&inode->i_data.page_tree, GFP_ATOMIC); - rwlock_init(&inode->i_data.tree_lock); + spin_lock_init(&inode->i_data.tree_lock); spin_lock_init(&inode->i_data.i_mmap_lock); INIT_LIST_HEAD(&inode->i_data.private_list); spin_lock_init(&inode->i_data.private_lock); diff --git a/include/asm-arm/cacheflush.h b/include/asm-arm/cacheflush.h index 70b0fe724b62..03cf1ee977b7 100644 --- a/include/asm-arm/cacheflush.h +++ b/include/asm-arm/cacheflush.h @@ -424,9 +424,9 @@ static inline void flush_anon_page(struct vm_area_struct *vma, } #define flush_dcache_mmap_lock(mapping) \ - write_lock_irq(&(mapping)->tree_lock) + spin_lock_irq(&(mapping)->tree_lock) #define flush_dcache_mmap_unlock(mapping) \ - write_unlock_irq(&(mapping)->tree_lock) + spin_unlock_irq(&(mapping)->tree_lock) #define flush_icache_user_range(vma,page,addr,len) \ flush_dcache_page(page) diff --git a/include/asm-parisc/cacheflush.h b/include/asm-parisc/cacheflush.h index 2f1e1b05440a..b7ca6dc7fddc 100644 --- a/include/asm-parisc/cacheflush.h +++ b/include/asm-parisc/cacheflush.h @@ -45,9 +45,9 @@ void flush_cache_mm(struct mm_struct *mm); extern void flush_dcache_page(struct page *page); #define flush_dcache_mmap_lock(mapping) \ - write_lock_irq(&(mapping)->tree_lock) + spin_lock_irq(&(mapping)->tree_lock) #define flush_dcache_mmap_unlock(mapping) \ - write_unlock_irq(&(mapping)->tree_lock) + spin_unlock_irq(&(mapping)->tree_lock) #define flush_icache_page(vma,page) do { \ flush_kernel_dcache_page(page); \ diff --git a/include/linux/fs.h b/include/linux/fs.h index 49d8eb7a71be..53d2edb709b3 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -499,7 +499,7 @@ struct backing_dev_info; struct address_space { struct inode *host; /* owner: inode, block_device */ struct radix_tree_root page_tree; /* radix tree of all pages */ - rwlock_t tree_lock; /* and rwlock protecting it */ + spinlock_t tree_lock; /* and lock protecting it */ unsigned int i_mmap_writable;/* count VM_SHARED mappings */ struct prio_tree_root i_mmap; /* tree of private and shared mappings */ struct list_head i_mmap_nonlinear;/*list VM_NONLINEAR mappings */ diff --git a/mm/filemap.c b/mm/filemap.c index feb8448d8618..2ed8b0389c51 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -109,7 +109,7 @@ /* * Remove a page from the page cache and free it. Caller has to make * sure the page is locked and that nobody else uses it - or that usage - * is safe. The caller must hold a write_lock on the mapping's tree_lock. + * is safe. The caller must hold the mapping's tree_lock. */ void __remove_from_page_cache(struct page *page) { @@ -141,9 +141,9 @@ void remove_from_page_cache(struct page *page) BUG_ON(!PageLocked(page)); - write_lock_irq(&mapping->tree_lock); + spin_lock_irq(&mapping->tree_lock); __remove_from_page_cache(page); - write_unlock_irq(&mapping->tree_lock); + spin_unlock_irq(&mapping->tree_lock); } static int sync_page(void *word) @@ -469,7 +469,7 @@ int add_to_page_cache_locked(struct page *page, struct address_space *mapping, page->mapping = mapping; page->index = offset; - write_lock_irq(&mapping->tree_lock); + spin_lock_irq(&mapping->tree_lock); error = radix_tree_insert(&mapping->page_tree, offset, page); if (likely(!error)) { mapping->nrpages++; @@ -480,7 +480,7 @@ int add_to_page_cache_locked(struct page *page, struct address_space *mapping, page_cache_release(page); } - write_unlock_irq(&mapping->tree_lock); + spin_unlock_irq(&mapping->tree_lock); radix_tree_preload_end(); } else mem_cgroup_uncharge_cache_page(page); diff --git a/mm/migrate.c b/mm/migrate.c index 3ca6392e82cc..153572fb60b8 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -323,7 +323,7 @@ static int migrate_page_move_mapping(struct address_space *mapping, return 0; } - write_lock_irq(&mapping->tree_lock); + spin_lock_irq(&mapping->tree_lock); pslot = radix_tree_lookup_slot(&mapping->page_tree, page_index(page)); @@ -331,12 +331,12 @@ static int migrate_page_move_mapping(struct address_space *mapping, expected_count = 2 + !!PagePrivate(page); if (page_count(page) != expected_count || (struct page *)radix_tree_deref_slot(pslot) != page) { - write_unlock_irq(&mapping->tree_lock); + spin_unlock_irq(&mapping->tree_lock); return -EAGAIN; } if (!page_freeze_refs(page, expected_count)) { - write_unlock_irq(&mapping->tree_lock); + spin_unlock_irq(&mapping->tree_lock); return -EAGAIN; } @@ -373,10 +373,9 @@ static int migrate_page_move_mapping(struct address_space *mapping, __dec_zone_page_state(page, NR_FILE_PAGES); __inc_zone_page_state(newpage, NR_FILE_PAGES); - write_unlock_irq(&mapping->tree_lock); - if (!PageSwapCache(newpage)) { + spin_unlock_irq(&mapping->tree_lock); + if (!PageSwapCache(newpage)) mem_cgroup_uncharge_cache_page(page); - } return 0; } diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 94c6d8988ab3..24de8b65fdbd 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -1088,7 +1088,7 @@ int __set_page_dirty_nobuffers(struct page *page) if (!mapping) return 1; - write_lock_irq(&mapping->tree_lock); + spin_lock_irq(&mapping->tree_lock); mapping2 = page_mapping(page); if (mapping2) { /* Race with truncate? */ BUG_ON(mapping2 != mapping); @@ -1102,7 +1102,7 @@ int __set_page_dirty_nobuffers(struct page *page) radix_tree_tag_set(&mapping->page_tree, page_index(page), PAGECACHE_TAG_DIRTY); } - write_unlock_irq(&mapping->tree_lock); + spin_unlock_irq(&mapping->tree_lock); if (mapping->host) { /* !PageAnon && !swapper_space */ __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); @@ -1258,7 +1258,7 @@ int test_clear_page_writeback(struct page *page) struct backing_dev_info *bdi = mapping->backing_dev_info; unsigned long flags; - write_lock_irqsave(&mapping->tree_lock, flags); + spin_lock_irqsave(&mapping->tree_lock, flags); ret = TestClearPageWriteback(page); if (ret) { radix_tree_tag_clear(&mapping->page_tree, @@ -1269,7 +1269,7 @@ int test_clear_page_writeback(struct page *page) __bdi_writeout_inc(bdi); } } - write_unlock_irqrestore(&mapping->tree_lock, flags); + spin_unlock_irqrestore(&mapping->tree_lock, flags); } else { ret = TestClearPageWriteback(page); } @@ -1287,7 +1287,7 @@ int test_set_page_writeback(struct page *page) struct backing_dev_info *bdi = mapping->backing_dev_info; unsigned long flags; - write_lock_irqsave(&mapping->tree_lock, flags); + spin_lock_irqsave(&mapping->tree_lock, flags); ret = TestSetPageWriteback(page); if (!ret) { radix_tree_tag_set(&mapping->page_tree, @@ -1300,7 +1300,7 @@ int test_set_page_writeback(struct page *page) radix_tree_tag_clear(&mapping->page_tree, page_index(page), PAGECACHE_TAG_DIRTY); - write_unlock_irqrestore(&mapping->tree_lock, flags); + spin_unlock_irqrestore(&mapping->tree_lock, flags); } else { ret = TestSetPageWriteback(page); } diff --git a/mm/swap_state.c b/mm/swap_state.c index 3e3381d6c7ee..2c217e33d497 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -39,7 +39,7 @@ static struct backing_dev_info swap_backing_dev_info = { struct address_space swapper_space = { .page_tree = RADIX_TREE_INIT(GFP_ATOMIC|__GFP_NOWARN), - .tree_lock = __RW_LOCK_UNLOCKED(swapper_space.tree_lock), + .tree_lock = __SPIN_LOCK_UNLOCKED(swapper_space.tree_lock), .a_ops = &swap_aops, .i_mmap_nonlinear = LIST_HEAD_INIT(swapper_space.i_mmap_nonlinear), .backing_dev_info = &swap_backing_dev_info, @@ -80,7 +80,7 @@ int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask) SetPageSwapCache(page); set_page_private(page, entry.val); - write_lock_irq(&swapper_space.tree_lock); + spin_lock_irq(&swapper_space.tree_lock); error = radix_tree_insert(&swapper_space.page_tree, entry.val, page); if (likely(!error)) { @@ -88,7 +88,7 @@ int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask) __inc_zone_page_state(page, NR_FILE_PAGES); INC_CACHE_INFO(add_total); } - write_unlock_irq(&swapper_space.tree_lock); + spin_unlock_irq(&swapper_space.tree_lock); radix_tree_preload_end(); if (unlikely(error)) { @@ -182,9 +182,9 @@ void delete_from_swap_cache(struct page *page) entry.val = page_private(page); - write_lock_irq(&swapper_space.tree_lock); + spin_lock_irq(&swapper_space.tree_lock); __delete_from_swap_cache(page); - write_unlock_irq(&swapper_space.tree_lock); + spin_unlock_irq(&swapper_space.tree_lock); swap_free(entry); page_cache_release(page); diff --git a/mm/swapfile.c b/mm/swapfile.c index 2f33edb8bee9..af283933c14e 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -369,13 +369,13 @@ int remove_exclusive_swap_page(struct page *page) retval = 0; if (p->swap_map[swp_offset(entry)] == 1) { /* Recheck the page count with the swapcache lock held.. */ - write_lock_irq(&swapper_space.tree_lock); + spin_lock_irq(&swapper_space.tree_lock); if ((page_count(page) == 2) && !PageWriteback(page)) { __delete_from_swap_cache(page); SetPageDirty(page); retval = 1; } - write_unlock_irq(&swapper_space.tree_lock); + spin_unlock_irq(&swapper_space.tree_lock); } spin_unlock(&swap_lock); diff --git a/mm/truncate.c b/mm/truncate.c index b8961cb63414..e68443d74567 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -349,18 +349,18 @@ invalidate_complete_page2(struct address_space *mapping, struct page *page) if (PagePrivate(page) && !try_to_release_page(page, GFP_KERNEL)) return 0; - write_lock_irq(&mapping->tree_lock); + spin_lock_irq(&mapping->tree_lock); if (PageDirty(page)) goto failed; BUG_ON(PagePrivate(page)); __remove_from_page_cache(page); - write_unlock_irq(&mapping->tree_lock); + spin_unlock_irq(&mapping->tree_lock); ClearPageUptodate(page); page_cache_release(page); /* pagecache ref */ return 1; failed: - write_unlock_irq(&mapping->tree_lock); + spin_unlock_irq(&mapping->tree_lock); return 0; } diff --git a/mm/vmscan.c b/mm/vmscan.c index 0075eac1cd04..8f71761bc4b7 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -399,7 +399,7 @@ static int __remove_mapping(struct address_space *mapping, struct page *page) BUG_ON(!PageLocked(page)); BUG_ON(mapping != page_mapping(page)); - write_lock_irq(&mapping->tree_lock); + spin_lock_irq(&mapping->tree_lock); /* * The non racy check for a busy page. * @@ -436,17 +436,17 @@ static int __remove_mapping(struct address_space *mapping, struct page *page) if (PageSwapCache(page)) { swp_entry_t swap = { .val = page_private(page) }; __delete_from_swap_cache(page); - write_unlock_irq(&mapping->tree_lock); + spin_unlock_irq(&mapping->tree_lock); swap_free(swap); } else { __remove_from_page_cache(page); - write_unlock_irq(&mapping->tree_lock); + spin_unlock_irq(&mapping->tree_lock); } return 1; cannot_free: - write_unlock_irq(&mapping->tree_lock); + spin_unlock_irq(&mapping->tree_lock); return 0; } -- cgit v1.2.3 From 51cc50685a4275c6a02653670af9f108a64e01cf Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 25 Jul 2008 19:45:34 -0700 Subject: SL*B: drop kmem cache argument from constructor Kmem cache passed to constructor is only needed for constructors that are themselves multiplexeres. Nobody uses this "feature", nor does anybody uses passed kmem cache in non-trivial way, so pass only pointer to object. Non-trivial places are: arch/powerpc/mm/init_64.c arch/powerpc/mm/hugetlbpage.c This is flag day, yes. Signed-off-by: Alexey Dobriyan Acked-by: Pekka Enberg Acked-by: Christoph Lameter Cc: Jon Tollefson Cc: Nick Piggin Cc: Matt Mackall [akpm@linux-foundation.org: fix arch/powerpc/mm/hugetlbpage.c] [akpm@linux-foundation.org: fix mm/slab.c] [akpm@linux-foundation.org: fix ubifs] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/plat-s3c24xx/dma.c | 2 +- arch/powerpc/kernel/rtas_flash.c | 2 +- arch/powerpc/mm/hugetlbpage.c | 9 ++------- arch/powerpc/mm/init_64.c | 24 +++++++++--------------- arch/powerpc/platforms/cell/spufs/inode.c | 2 +- arch/sh/mm/pmb.c | 2 +- arch/xtensa/mm/init.c | 2 +- drivers/usb/mon/mon_text.c | 4 ++-- fs/adfs/super.c | 2 +- fs/affs/super.c | 2 +- fs/afs/super.c | 4 ++-- fs/befs/linuxvfs.c | 2 +- fs/bfs/inode.c | 2 +- fs/block_dev.c | 2 +- fs/buffer.c | 2 +- fs/cifs/cifsfs.c | 2 +- fs/coda/inode.c | 2 +- fs/ecryptfs/main.c | 4 ++-- fs/efs/super.c | 2 +- fs/ext2/super.c | 2 +- fs/ext3/super.c | 2 +- fs/ext4/super.c | 2 +- fs/fat/cache.c | 2 +- fs/fat/inode.c | 2 +- fs/fuse/inode.c | 2 +- fs/gfs2/main.c | 4 ++-- fs/hfs/super.c | 2 +- fs/hfsplus/super.c | 2 +- fs/hpfs/super.c | 2 +- fs/hugetlbfs/inode.c | 2 +- fs/inode.c | 2 +- fs/isofs/inode.c | 2 +- fs/jffs2/super.c | 2 +- fs/jfs/jfs_metapage.c | 2 +- fs/jfs/super.c | 2 +- fs/locks.c | 2 +- fs/minix/inode.c | 2 +- fs/ncpfs/inode.c | 2 +- fs/nfs/inode.c | 2 +- fs/ntfs/super.c | 2 +- fs/ocfs2/dlm/dlmfs.c | 3 +-- fs/ocfs2/super.c | 2 +- fs/openpromfs/inode.c | 2 +- fs/proc/inode.c | 2 +- fs/qnx4/inode.c | 2 +- fs/reiserfs/super.c | 2 +- fs/romfs/inode.c | 2 +- fs/smbfs/inode.c | 2 +- fs/sysv/inode.c | 2 +- fs/ubifs/super.c | 2 +- fs/udf/super.c | 2 +- fs/ufs/super.c | 2 +- fs/xfs/linux-2.6/kmem.h | 2 +- fs/xfs/linux-2.6/xfs_super.c | 1 - include/linux/slab.h | 2 +- include/linux/slub_def.h | 2 +- ipc/mqueue.c | 2 +- kernel/fork.c | 2 +- lib/idr.c | 2 +- lib/radix-tree.c | 2 +- mm/rmap.c | 2 +- mm/shmem.c | 2 +- mm/slab.c | 11 +++++------ mm/slob.c | 7 +++---- mm/slub.c | 13 ++++++------- net/socket.c | 2 +- net/sunrpc/rpc_pipe.c | 2 +- 67 files changed, 90 insertions(+), 106 deletions(-) (limited to 'fs') diff --git a/arch/arm/plat-s3c24xx/dma.c b/arch/arm/plat-s3c24xx/dma.c index 60f162dc4fad..8c5e656d5d8c 100644 --- a/arch/arm/plat-s3c24xx/dma.c +++ b/arch/arm/plat-s3c24xx/dma.c @@ -1304,7 +1304,7 @@ struct sysdev_class dma_sysclass = { /* kmem cache implementation */ -static void s3c2410_dma_cache_ctor(struct kmem_cache *c, void *p) +static void s3c2410_dma_cache_ctor(void *p) { memset(p, 0, sizeof(struct s3c2410_dma_buf)); } diff --git a/arch/powerpc/kernel/rtas_flash.c b/arch/powerpc/kernel/rtas_flash.c index 09ded5c424a9..149cb112cd1a 100644 --- a/arch/powerpc/kernel/rtas_flash.c +++ b/arch/powerpc/kernel/rtas_flash.c @@ -286,7 +286,7 @@ static ssize_t rtas_flash_read(struct file *file, char __user *buf, } /* constructor for flash_block_cache */ -void rtas_block_ctor(struct kmem_cache *cache, void *ptr) +void rtas_block_ctor(void *ptr) { memset(ptr, 0, RTAS_BLK_SIZE); } diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index fb42c4dd3217..ed0aab0208a6 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -113,7 +113,7 @@ static inline pte_t *hugepte_offset(hugepd_t *hpdp, unsigned long addr, static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp, unsigned long address, unsigned int psize) { - pte_t *new = kmem_cache_alloc(huge_pgtable_cache(psize), + pte_t *new = kmem_cache_zalloc(huge_pgtable_cache(psize), GFP_KERNEL|__GFP_REPEAT); if (! new) @@ -730,11 +730,6 @@ static int __init hugepage_setup_sz(char *str) } __setup("hugepagesz=", hugepage_setup_sz); -static void zero_ctor(struct kmem_cache *cache, void *addr) -{ - memset(addr, 0, kmem_cache_size(cache)); -} - static int __init hugetlbpage_init(void) { unsigned int psize; @@ -756,7 +751,7 @@ static int __init hugetlbpage_init(void) HUGEPTE_TABLE_SIZE(psize), HUGEPTE_TABLE_SIZE(psize), 0, - zero_ctor); + NULL); if (!huge_pgtable_cache(psize)) panic("hugetlbpage_init(): could not create %s"\ "\n", HUGEPTE_CACHE_NAME(psize)); diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c index a41bc5aa2043..4f7df85129d8 100644 --- a/arch/powerpc/mm/init_64.c +++ b/arch/powerpc/mm/init_64.c @@ -136,9 +136,14 @@ static int __init setup_kcore(void) module_init(setup_kcore); #endif -static void zero_ctor(struct kmem_cache *cache, void *addr) +static void pgd_ctor(void *addr) { - memset(addr, 0, kmem_cache_size(cache)); + memset(addr, 0, PGD_TABLE_SIZE); +} + +static void pmd_ctor(void *addr) +{ + memset(addr, 0, PMD_TABLE_SIZE); } static const unsigned int pgtable_cache_size[2] = { @@ -163,19 +168,8 @@ struct kmem_cache *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)]; void pgtable_cache_init(void) { - int i; - - for (i = 0; i < ARRAY_SIZE(pgtable_cache_size); i++) { - int size = pgtable_cache_size[i]; - const char *name = pgtable_cache_name[i]; - - pr_debug("Allocating page table cache %s (#%d) " - "for size: %08x...\n", name, i, size); - pgtable_cache[i] = kmem_cache_create(name, - size, size, - SLAB_PANIC, - zero_ctor); - } + pgtable_cache[0] = kmem_cache_create(pgtable_cache_name[0], PGD_TABLE_SIZE, PGD_TABLE_SIZE, SLAB_PANIC, pgd_ctor); + pgtable_cache[1] = kmem_cache_create(pgtable_cache_name[1], PMD_TABLE_SIZE, PMD_TABLE_SIZE, SLAB_PANIC, pmd_ctor); } #ifdef CONFIG_SPARSEMEM_VMEMMAP diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c index 7123472801d9..690ca7b0dcf6 100644 --- a/arch/powerpc/platforms/cell/spufs/inode.c +++ b/arch/powerpc/platforms/cell/spufs/inode.c @@ -78,7 +78,7 @@ spufs_destroy_inode(struct inode *inode) } static void -spufs_init_once(struct kmem_cache *cachep, void *p) +spufs_init_once(void *p) { struct spufs_inode_info *ei = p; diff --git a/arch/sh/mm/pmb.c b/arch/sh/mm/pmb.c index 0b0ec6e04753..46911bcbf17b 100644 --- a/arch/sh/mm/pmb.c +++ b/arch/sh/mm/pmb.c @@ -293,7 +293,7 @@ void pmb_unmap(unsigned long addr) } while (pmbe); } -static void pmb_cache_ctor(struct kmem_cache *cachep, void *pmb) +static void pmb_cache_ctor(void *pmb) { struct pmb_entry *pmbe = pmb; diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c index 81d0560eaea2..ee261005b363 100644 --- a/arch/xtensa/mm/init.c +++ b/arch/xtensa/mm/init.c @@ -309,7 +309,7 @@ void show_mem(void) struct kmem_cache *pgtable_cache __read_mostly; -static void pgd_ctor(struct kmem_cache *cache, void* addr) +static void pgd_ctor(void* addr) { pte_t* ptep = (pte_t*)addr; int i; diff --git a/drivers/usb/mon/mon_text.c b/drivers/usb/mon/mon_text.c index 5e3e4e9b6c77..1f715436d6d3 100644 --- a/drivers/usb/mon/mon_text.c +++ b/drivers/usb/mon/mon_text.c @@ -87,7 +87,7 @@ struct mon_reader_text { static struct dentry *mon_dir; /* Usually /sys/kernel/debug/usbmon */ -static void mon_text_ctor(struct kmem_cache *, void *); +static void mon_text_ctor(void *); struct mon_text_ptr { int cnt, limit; @@ -720,7 +720,7 @@ void mon_text_del(struct mon_bus *mbus) /* * Slab interface: constructor. */ -static void mon_text_ctor(struct kmem_cache *slab, void *mem) +static void mon_text_ctor(void *mem) { /* * Nothing to initialize. No, really! diff --git a/fs/adfs/super.c b/fs/adfs/super.c index 9e421eeb672b..26f3b43726bb 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -249,7 +249,7 @@ static void adfs_destroy_inode(struct inode *inode) kmem_cache_free(adfs_inode_cachep, ADFS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct adfs_inode_info *ei = (struct adfs_inode_info *) foo; diff --git a/fs/affs/super.c b/fs/affs/super.c index 4e0309566406..3a89094f93d0 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -90,7 +90,7 @@ static void affs_destroy_inode(struct inode *inode) kmem_cache_free(affs_inode_cachep, AFFS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct affs_inode_info *ei = (struct affs_inode_info *) foo; diff --git a/fs/afs/super.c b/fs/afs/super.c index 7e3faeef6818..250d8c4d66e4 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -27,7 +27,7 @@ #define AFS_FS_MAGIC 0x6B414653 /* 'kAFS' */ -static void afs_i_init_once(struct kmem_cache *cachep, void *foo); +static void afs_i_init_once(void *foo); static int afs_get_sb(struct file_system_type *fs_type, int flags, const char *dev_name, void *data, struct vfsmount *mnt); @@ -449,7 +449,7 @@ static void afs_put_super(struct super_block *sb) /* * initialise an inode cache slab element prior to any use */ -static void afs_i_init_once(struct kmem_cache *cachep, void *_vnode) +static void afs_i_init_once(void *_vnode) { struct afs_vnode *vnode = _vnode; diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index e8717de3bab3..02c6e62b72f8 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -289,7 +289,7 @@ befs_destroy_inode(struct inode *inode) kmem_cache_free(befs_inode_cachep, BEFS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct befs_inode_info *bi = (struct befs_inode_info *) foo; diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index 053e690ec9ed..0ed57b5ee012 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -264,7 +264,7 @@ static void bfs_destroy_inode(struct inode *inode) kmem_cache_free(bfs_inode_cachep, BFS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct bfs_inode_info *bi = foo; diff --git a/fs/block_dev.c b/fs/block_dev.c index 10d8a0aa871a..dcf37cada369 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -271,7 +271,7 @@ static void bdev_destroy_inode(struct inode *inode) kmem_cache_free(bdev_cachep, bdi); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct bdev_inode *ei = (struct bdev_inode *) foo; struct block_device *bdev = &ei->bdev; diff --git a/fs/buffer.c b/fs/buffer.c index 109b261192d9..5fd497cdd6f3 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -3272,7 +3272,7 @@ int bh_submit_read(struct buffer_head *bh) EXPORT_SYMBOL(bh_submit_read); static void -init_buffer_head(struct kmem_cache *cachep, void *data) +init_buffer_head(void *data) { struct buffer_head *bh = data; diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 22857c639df5..fe5f6809cba6 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -766,7 +766,7 @@ const struct file_operations cifs_dir_ops = { }; static void -cifs_init_once(struct kmem_cache *cachep, void *inode) +cifs_init_once(void *inode) { struct cifsInodeInfo *cifsi = inode; diff --git a/fs/coda/inode.c b/fs/coda/inode.c index 2f58dfc70083..830f51abb971 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -58,7 +58,7 @@ static void coda_destroy_inode(struct inode *inode) kmem_cache_free(coda_inode_cachep, ITOC(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct coda_inode_info *ei = (struct coda_inode_info *) foo; diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index 6f403cfba14f..448dfd597b5f 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -578,7 +578,7 @@ static struct file_system_type ecryptfs_fs_type = { * Initializes the ecryptfs_inode_info_cache when it is created */ static void -inode_info_init_once(struct kmem_cache *cachep, void *vptr) +inode_info_init_once(void *vptr) { struct ecryptfs_inode_info *ei = (struct ecryptfs_inode_info *)vptr; @@ -589,7 +589,7 @@ static struct ecryptfs_cache_info { struct kmem_cache **cache; const char *name; size_t size; - void (*ctor)(struct kmem_cache *cache, void *obj); + void (*ctor)(void *obj); } ecryptfs_cache_infos[] = { { .cache = &ecryptfs_auth_tok_list_item_cache, diff --git a/fs/efs/super.c b/fs/efs/super.c index d733531b55e2..567b134fa1f1 100644 --- a/fs/efs/super.c +++ b/fs/efs/super.c @@ -70,7 +70,7 @@ static void efs_destroy_inode(struct inode *inode) kmem_cache_free(efs_inode_cachep, INODE_INFO(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct efs_inode_info *ei = (struct efs_inode_info *) foo; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 31308a3b0b8b..fd88c7b43e66 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -159,7 +159,7 @@ static void ext2_destroy_inode(struct inode *inode) kmem_cache_free(ext2_inode_cachep, EXT2_I(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct ext2_inode_info *ei = (struct ext2_inode_info *) foo; diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 615788c6843a..8ddced384674 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -472,7 +472,7 @@ static void ext3_destroy_inode(struct inode *inode) kmem_cache_free(ext3_inode_cachep, EXT3_I(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct ext3_inode_info *ei = (struct ext3_inode_info *) foo; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 1cb371dcd609..b5479b1dff14 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -595,7 +595,7 @@ static void ext4_destroy_inode(struct inode *inode) kmem_cache_free(ext4_inode_cachep, EXT4_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct ext4_inode_info *ei = (struct ext4_inode_info *) foo; diff --git a/fs/fat/cache.c b/fs/fat/cache.c index 3a9ecac8d61f..3222f51c41cf 100644 --- a/fs/fat/cache.c +++ b/fs/fat/cache.c @@ -36,7 +36,7 @@ static inline int fat_max_cache(struct inode *inode) static struct kmem_cache *fat_cache_cachep; -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct fat_cache *cache = (struct fat_cache *)foo; diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 23676f9d79ce..6d266d793e2c 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -498,7 +498,7 @@ static void fat_destroy_inode(struct inode *inode) kmem_cache_free(fat_inode_cachep, MSDOS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct msdos_inode_info *ei = (struct msdos_inode_info *)foo; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 7d2f7d6e22e2..d2249f174e20 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -956,7 +956,7 @@ static inline void unregister_fuseblk(void) } #endif -static void fuse_inode_init_once(struct kmem_cache *cachep, void *foo) +static void fuse_inode_init_once(void *foo) { struct inode * inode = foo; diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c index bcc668d0fadd..bb2cc303ac29 100644 --- a/fs/gfs2/main.c +++ b/fs/gfs2/main.c @@ -24,7 +24,7 @@ #include "util.h" #include "glock.h" -static void gfs2_init_inode_once(struct kmem_cache *cachep, void *foo) +static void gfs2_init_inode_once(void *foo) { struct gfs2_inode *ip = foo; @@ -33,7 +33,7 @@ static void gfs2_init_inode_once(struct kmem_cache *cachep, void *foo) ip->i_alloc = NULL; } -static void gfs2_init_glock_once(struct kmem_cache *cachep, void *foo) +static void gfs2_init_glock_once(void *foo) { struct gfs2_glock *gl = foo; diff --git a/fs/hfs/super.c b/fs/hfs/super.c index ac2ec5ef66e4..4abb1047c689 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -432,7 +432,7 @@ static struct file_system_type hfs_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void hfs_init_once(struct kmem_cache *cachep, void *p) +static void hfs_init_once(void *p) { struct hfs_inode_info *i = p; diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 3859118531c7..e834e578c93f 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -485,7 +485,7 @@ static struct file_system_type hfsplus_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void hfsplus_init_once(struct kmem_cache *cachep, void *p) +static void hfsplus_init_once(void *p) { struct hfsplus_inode_info *i = p; diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index f63a699ec659..b8ae9c90ada0 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -173,7 +173,7 @@ static void hpfs_destroy_inode(struct inode *inode) kmem_cache_free(hpfs_inode_cachep, hpfs_i(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct hpfs_inode_info *ei = (struct hpfs_inode_info *) foo; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index dbd01d262ca4..3f58923fb39b 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -705,7 +705,7 @@ static const struct address_space_operations hugetlbfs_aops = { }; -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct hugetlbfs_inode_info *ei = (struct hugetlbfs_inode_info *)foo; diff --git a/fs/inode.c b/fs/inode.c index 35b6414522ea..b6726f644530 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -224,7 +224,7 @@ void inode_init_once(struct inode *inode) EXPORT_SYMBOL(inode_init_once); -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct inode * inode = (struct inode *) foo; diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index 044a254d526b..26948a6033b6 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -73,7 +73,7 @@ static void isofs_destroy_inode(struct inode *inode) kmem_cache_free(isofs_inode_cachep, ISOFS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct iso_inode_info *ei = foo; diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index 7da69eae49e4..efd401257ed9 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -44,7 +44,7 @@ static void jffs2_destroy_inode(struct inode *inode) kmem_cache_free(jffs2_inode_cachep, JFFS2_INODE_INFO(inode)); } -static void jffs2_i_init_once(struct kmem_cache *cachep, void *foo) +static void jffs2_i_init_once(void *foo) { struct jffs2_inode_info *f = foo; diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c index 854ff0ec574f..c350057087dd 100644 --- a/fs/jfs/jfs_metapage.c +++ b/fs/jfs/jfs_metapage.c @@ -182,7 +182,7 @@ static inline void remove_metapage(struct page *page, struct metapage *mp) #endif -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct metapage *mp = (struct metapage *)foo; diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 359c091d8965..3630718be395 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -760,7 +760,7 @@ static struct file_system_type jfs_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct jfs_inode_info *jfs_ip = (struct jfs_inode_info *) foo; diff --git a/fs/locks.c b/fs/locks.c index 01490300f7cb..5eb259e3cd38 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -201,7 +201,7 @@ EXPORT_SYMBOL(locks_init_lock); * Initialises the fields of the file lock which are invariant for * free file_locks. */ -static void init_once(struct kmem_cache *cache, void *foo) +static void init_once(void *foo) { struct file_lock *lock = (struct file_lock *) foo; diff --git a/fs/minix/inode.c b/fs/minix/inode.c index 523d73713418..d1d1eb84679d 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -68,7 +68,7 @@ static void minix_destroy_inode(struct inode *inode) kmem_cache_free(minix_inode_cachep, minix_i(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct minix_inode_info *ei = (struct minix_inode_info *) foo; diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 2e5ab1204dec..d642f0e5b365 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -64,7 +64,7 @@ static void ncp_destroy_inode(struct inode *inode) kmem_cache_free(ncp_inode_cachep, NCP_FINFO(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct ncp_inode_info *ei = (struct ncp_inode_info *) foo; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index df23f987da6b..52daefa2f521 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1242,7 +1242,7 @@ static inline void nfs4_init_once(struct nfs_inode *nfsi) #endif } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct nfs_inode *nfsi = (struct nfs_inode *) foo; diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c index 3e76f3b216bc..4a46743b5077 100644 --- a/fs/ntfs/super.c +++ b/fs/ntfs/super.c @@ -3080,7 +3080,7 @@ struct kmem_cache *ntfs_inode_cache; struct kmem_cache *ntfs_big_inode_cache; /* Init once constructor for the inode slab cache. */ -static void ntfs_big_inode_init_once(struct kmem_cache *cachep, void *foo) +static void ntfs_big_inode_init_once(void *foo) { ntfs_inode *ni = (ntfs_inode *)foo; diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c index e48aba698b77..533a789c3ef8 100644 --- a/fs/ocfs2/dlm/dlmfs.c +++ b/fs/ocfs2/dlm/dlmfs.c @@ -267,8 +267,7 @@ static ssize_t dlmfs_file_write(struct file *filp, return writelen; } -static void dlmfs_init_once(struct kmem_cache *cachep, - void *foo) +static void dlmfs_init_once(void *foo) { struct dlmfs_inode_private *ip = (struct dlmfs_inode_private *) foo; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index ccecfe5094fa..2560b33889aa 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -1118,7 +1118,7 @@ bail: return status; } -static void ocfs2_inode_init_once(struct kmem_cache *cachep, void *data) +static void ocfs2_inode_init_once(void *data) { struct ocfs2_inode_info *oi = data; diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index d17b4fd204e1..9f5b054f06b9 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -430,7 +430,7 @@ static struct file_system_type openprom_fs_type = { .kill_sb = kill_anon_super, }; -static void op_inode_init_once(struct kmem_cache * cachep, void *data) +static void op_inode_init_once(void *data) { struct op_inode_info *oi = (struct op_inode_info *) data; diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 02eca2ed9dd7..b37f25dc45a5 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -94,7 +94,7 @@ static void proc_destroy_inode(struct inode *inode) kmem_cache_free(proc_inode_cachep, PROC_I(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct proc_inode *ei = (struct proc_inode *) foo; diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index b31ab78052b3..2aad1044b84c 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -553,7 +553,7 @@ static void qnx4_destroy_inode(struct inode *inode) kmem_cache_free(qnx4_inode_cachep, qnx4_i(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct qnx4_inode_info *ei = (struct qnx4_inode_info *) foo; diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 2ec748ba0bd3..879e54d35c2d 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -521,7 +521,7 @@ static void reiserfs_destroy_inode(struct inode *inode) kmem_cache_free(reiserfs_inode_cachep, REISERFS_I(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct reiserfs_inode_info *ei = (struct reiserfs_inode_info *)foo; diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c index 3f13d491c7c7..8e51a2aaa977 100644 --- a/fs/romfs/inode.c +++ b/fs/romfs/inode.c @@ -577,7 +577,7 @@ static void romfs_destroy_inode(struct inode *inode) kmem_cache_free(romfs_inode_cachep, ROMFS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct romfs_inode_info *ei = foo; diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c index 376ef3ee6ed7..3528f40ffb0f 100644 --- a/fs/smbfs/inode.c +++ b/fs/smbfs/inode.c @@ -67,7 +67,7 @@ static void smb_destroy_inode(struct inode *inode) kmem_cache_free(smb_inode_cachep, SMB_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct smb_inode_info *ei = (struct smb_inode_info *) foo; diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index c5d60de0658f..df0d435baa48 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -326,7 +326,7 @@ static void sysv_destroy_inode(struct inode *inode) kmem_cache_free(sysv_inode_cachep, SYSV_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *p) +static void init_once(void *p) { struct sysv_inode_info *si = (struct sysv_inode_info *)p; diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 00eb9c68ad03..ca1e2d4e03cc 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -1841,7 +1841,7 @@ static struct file_system_type ubifs_fs_type = { /* * Inode slab cache constructor. */ -static void inode_slab_ctor(struct kmem_cache *cachep, void *obj) +static void inode_slab_ctor(void *obj) { struct ubifs_inode *ui = obj; inode_init_once(&ui->vfs_inode); diff --git a/fs/udf/super.c b/fs/udf/super.c index 44cc702f96cc..5698bbf83bbf 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -148,7 +148,7 @@ static void udf_destroy_inode(struct inode *inode) kmem_cache_free(udf_inode_cachep, UDF_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct udf_inode_info *ei = (struct udf_inode_info *)foo; diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 227c9d700040..3e30e40aa24d 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -1302,7 +1302,7 @@ static void ufs_destroy_inode(struct inode *inode) kmem_cache_free(ufs_inode_cachep, UFS_I(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct ufs_inode_info *ei = (struct ufs_inode_info *) foo; diff --git a/fs/xfs/linux-2.6/kmem.h b/fs/xfs/linux-2.6/kmem.h index 5e9564902976..a20683cf74dd 100644 --- a/fs/xfs/linux-2.6/kmem.h +++ b/fs/xfs/linux-2.6/kmem.h @@ -79,7 +79,7 @@ kmem_zone_init(int size, char *zone_name) static inline kmem_zone_t * kmem_zone_init_flags(int size, char *zone_name, unsigned long flags, - void (*construct)(kmem_zone_t *, void *)) + void (*construct)(void *)) { return kmem_cache_create(zone_name, size, 0, flags, construct); } diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 742b2c7852c1..943381284e2e 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -843,7 +843,6 @@ xfs_fs_destroy_inode( STATIC void xfs_fs_inode_init_once( - kmem_zone_t *zonep, void *vnode) { inode_init_once(vn_to_inode((bhv_vnode_t *)vnode)); diff --git a/include/linux/slab.h b/include/linux/slab.h index 41103910f8a2..9ff8e8499403 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -58,7 +58,7 @@ int slab_is_available(void); struct kmem_cache *kmem_cache_create(const char *, size_t, size_t, unsigned long, - void (*)(struct kmem_cache *, void *)); + void (*)(void *)); void kmem_cache_destroy(struct kmem_cache *); int kmem_cache_shrink(struct kmem_cache *); void kmem_cache_free(struct kmem_cache *, void *); diff --git a/include/linux/slub_def.h b/include/linux/slub_def.h index d117ea2825a9..5bad61a93f65 100644 --- a/include/linux/slub_def.h +++ b/include/linux/slub_def.h @@ -85,7 +85,7 @@ struct kmem_cache { struct kmem_cache_order_objects min; gfp_t allocflags; /* gfp flags to use on each alloc */ int refcount; /* Refcount for slab cache destroy */ - void (*ctor)(struct kmem_cache *, void *); + void (*ctor)(void *); int inuse; /* Offset to metadata */ int align; /* Alignment */ const char *name; /* Name (only for display!) */ diff --git a/ipc/mqueue.c b/ipc/mqueue.c index 1fdc2eb2f6d8..474984f9e032 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -207,7 +207,7 @@ static int mqueue_get_sb(struct file_system_type *fs_type, return get_sb_single(fs_type, flags, data, mqueue_fill_super, mnt); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct mqueue_inode_info *p = (struct mqueue_inode_info *) foo; diff --git a/kernel/fork.c b/kernel/fork.c index b99d73e971a4..80e83e459b17 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1442,7 +1442,7 @@ long do_fork(unsigned long clone_flags, #define ARCH_MIN_MMSTRUCT_ALIGN 0 #endif -static void sighand_ctor(struct kmem_cache *cachep, void *data) +static void sighand_ctor(void *data) { struct sighand_struct *sighand = data; diff --git a/lib/idr.c b/lib/idr.c index 3476f8203e97..e728c7fccc4d 100644 --- a/lib/idr.c +++ b/lib/idr.c @@ -607,7 +607,7 @@ void *idr_replace(struct idr *idp, void *ptr, int id) } EXPORT_SYMBOL(idr_replace); -static void idr_cache_ctor(struct kmem_cache *idr_layer_cache, void *idr_layer) +static void idr_cache_ctor(void *idr_layer) { memset(idr_layer, 0, sizeof(struct idr_layer)); } diff --git a/lib/radix-tree.c b/lib/radix-tree.c index 9c4f1ffa2864..be86b32bc874 100644 --- a/lib/radix-tree.c +++ b/lib/radix-tree.c @@ -1183,7 +1183,7 @@ int radix_tree_tagged(struct radix_tree_root *root, unsigned int tag) EXPORT_SYMBOL(radix_tree_tagged); static void -radix_tree_node_ctor(struct kmem_cache *cachep, void *node) +radix_tree_node_ctor(void *node) { memset(node, 0, sizeof(struct radix_tree_node)); } diff --git a/mm/rmap.c b/mm/rmap.c index abbd29f7c43f..39ae5a9bf382 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -138,7 +138,7 @@ void anon_vma_unlink(struct vm_area_struct *vma) anon_vma_free(anon_vma); } -static void anon_vma_ctor(struct kmem_cache *cachep, void *data) +static void anon_vma_ctor(void *data) { struct anon_vma *anon_vma = data; diff --git a/mm/shmem.c b/mm/shmem.c index 1089092aecaf..952d361774bb 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2352,7 +2352,7 @@ static void shmem_destroy_inode(struct inode *inode) kmem_cache_free(shmem_inode_cachep, SHMEM_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct shmem_inode_info *p = (struct shmem_inode_info *) foo; diff --git a/mm/slab.c b/mm/slab.c index 052e7d64537e..918f04f7fef1 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -406,7 +406,7 @@ struct kmem_cache { unsigned int dflags; /* dynamic flags */ /* constructor func */ - void (*ctor)(struct kmem_cache *, void *); + void (*ctor)(void *obj); /* 5) cache creation/removal */ const char *name; @@ -2137,8 +2137,7 @@ static int __init_refok setup_cpu_cache(struct kmem_cache *cachep) */ struct kmem_cache * kmem_cache_create (const char *name, size_t size, size_t align, - unsigned long flags, - void (*ctor)(struct kmem_cache *, void *)) + unsigned long flags, void (*ctor)(void *)) { size_t left_over, slab_size, ralign; struct kmem_cache *cachep = NULL, *pc; @@ -2653,7 +2652,7 @@ static void cache_init_objs(struct kmem_cache *cachep, * They must also be threaded. */ if (cachep->ctor && !(cachep->flags & SLAB_POISON)) - cachep->ctor(cachep, objp + obj_offset(cachep)); + cachep->ctor(objp + obj_offset(cachep)); if (cachep->flags & SLAB_RED_ZONE) { if (*dbg_redzone2(cachep, objp) != RED_INACTIVE) @@ -2669,7 +2668,7 @@ static void cache_init_objs(struct kmem_cache *cachep, cachep->buffer_size / PAGE_SIZE, 0); #else if (cachep->ctor) - cachep->ctor(cachep, objp); + cachep->ctor(objp); #endif slab_bufctl(slabp)[i] = i + 1; } @@ -3093,7 +3092,7 @@ static void *cache_alloc_debugcheck_after(struct kmem_cache *cachep, #endif objp += obj_offset(cachep); if (cachep->ctor && cachep->flags & SLAB_POISON) - cachep->ctor(cachep, objp); + cachep->ctor(objp); #if ARCH_SLAB_MINALIGN if ((u32)objp & (ARCH_SLAB_MINALIGN-1)) { printk(KERN_ERR "0x%p: not aligned to ARCH_SLAB_MINALIGN=%d\n", diff --git a/mm/slob.c b/mm/slob.c index de268eb7ac70..d8fbd4d1bfa7 100644 --- a/mm/slob.c +++ b/mm/slob.c @@ -525,12 +525,11 @@ struct kmem_cache { unsigned int size, align; unsigned long flags; const char *name; - void (*ctor)(struct kmem_cache *, void *); + void (*ctor)(void *); }; struct kmem_cache *kmem_cache_create(const char *name, size_t size, - size_t align, unsigned long flags, - void (*ctor)(struct kmem_cache *, void *)) + size_t align, unsigned long flags, void (*ctor)(void *)) { struct kmem_cache *c; @@ -575,7 +574,7 @@ void *kmem_cache_alloc_node(struct kmem_cache *c, gfp_t flags, int node) b = slob_new_page(flags, get_order(c->size), node); if (c->ctor) - c->ctor(c, b); + c->ctor(b); return b; } diff --git a/mm/slub.c b/mm/slub.c index 77c21cf53ff9..b7e2cd5d82db 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -1012,7 +1012,7 @@ __setup("slub_debug", setup_slub_debug); static unsigned long kmem_cache_flags(unsigned long objsize, unsigned long flags, const char *name, - void (*ctor)(struct kmem_cache *, void *)) + void (*ctor)(void *)) { /* * Enable debugging if selected on the kernel commandline. @@ -1040,7 +1040,7 @@ static inline int check_object(struct kmem_cache *s, struct page *page, static inline void add_full(struct kmem_cache_node *n, struct page *page) {} static inline unsigned long kmem_cache_flags(unsigned long objsize, unsigned long flags, const char *name, - void (*ctor)(struct kmem_cache *, void *)) + void (*ctor)(void *)) { return flags; } @@ -1103,7 +1103,7 @@ static void setup_object(struct kmem_cache *s, struct page *page, { setup_object_debug(s, page, object); if (unlikely(s->ctor)) - s->ctor(s, object); + s->ctor(object); } static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node) @@ -2286,7 +2286,7 @@ static int calculate_sizes(struct kmem_cache *s, int forced_order) static int kmem_cache_open(struct kmem_cache *s, gfp_t gfpflags, const char *name, size_t size, size_t align, unsigned long flags, - void (*ctor)(struct kmem_cache *, void *)) + void (*ctor)(void *)) { memset(s, 0, kmem_size); s->name = name; @@ -3042,7 +3042,7 @@ static int slab_unmergeable(struct kmem_cache *s) static struct kmem_cache *find_mergeable(size_t size, size_t align, unsigned long flags, const char *name, - void (*ctor)(struct kmem_cache *, void *)) + void (*ctor)(void *)) { struct kmem_cache *s; @@ -3082,8 +3082,7 @@ static struct kmem_cache *find_mergeable(size_t size, } struct kmem_cache *kmem_cache_create(const char *name, size_t size, - size_t align, unsigned long flags, - void (*ctor)(struct kmem_cache *, void *)) + size_t align, unsigned long flags, void (*ctor)(void *)) { struct kmem_cache *s; diff --git a/net/socket.c b/net/socket.c index 1310a82cbba7..8ef8ba81b9e2 100644 --- a/net/socket.c +++ b/net/socket.c @@ -265,7 +265,7 @@ static void sock_destroy_inode(struct inode *inode) container_of(inode, struct socket_alloc, vfs_inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct socket_alloc *ei = (struct socket_alloc *)foo; diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index 5a9b0e7828cd..23a2b8f6dc49 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -897,7 +897,7 @@ static struct file_system_type rpc_pipe_fs_type = { }; static void -init_once(struct kmem_cache * cachep, void *foo) +init_once(void *foo) { struct rpc_inode *rpci = (struct rpc_inode *) foo; -- cgit v1.2.3 From 5c752ad9f35910ff1912b3f3ae82878178ddc432 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Fri, 25 Jul 2008 19:45:40 -0700 Subject: Use WARN() in fs/ Use WARN() instead of a printk+WARN_ON() pair; this way the message becomes part of the warning section for better reporting/collection. Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 3 +-- fs/namespace.c | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 5fd497cdd6f3..f95805019639 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1214,8 +1214,7 @@ void __brelse(struct buffer_head * buf) put_bh(buf); return; } - printk(KERN_ERR "VFS: brelse: Trying to free free buffer\n"); - WARN_ON(1); + WARN(1, KERN_ERR "VFS: brelse: Trying to free free buffer\n"); } /* diff --git a/fs/namespace.c b/fs/namespace.c index 4f6f7635b59c..f30b11e2240e 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -309,10 +309,9 @@ static void handle_write_count_underflow(struct vfsmount *mnt) */ if ((atomic_read(&mnt->__mnt_writers) < 0) && !(mnt->mnt_flags & MNT_IMBALANCED_WRITE_COUNT)) { - printk(KERN_DEBUG "leak detected on mount(%p) writers " + WARN(1, KERN_DEBUG "leak detected on mount(%p) writers " "count: %d\n", mnt, atomic_read(&mnt->__mnt_writers)); - WARN_ON(1); /* use the flag to keep the dmesg spam down */ mnt->mnt_flags |= MNT_IMBALANCED_WRITE_COUNT; } -- cgit v1.2.3 From 99fcd77d15357e8ba51005c25cc750b9c28b2688 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Fri, 25 Jul 2008 19:45:41 -0700 Subject: Use WARN() in fs/sysfs Use WARN() instead of a printk+WARN_ON() pair; this way the message becomes part of the warning section for better reporting/collection. Also, with this, one fo the if() sections collapses entirely into the WARN(). Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/sysfs/dir.c | 5 +---- fs/sysfs/file.c | 3 +-- fs/sysfs/group.c | 3 +-- 3 files changed, 3 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index c1a7efb310bf..aedaeba82ae5 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -459,11 +459,8 @@ int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd) int ret; ret = __sysfs_add_one(acxt, sd); - if (ret == -EEXIST) { - printk(KERN_WARNING "sysfs: duplicate filename '%s' " + WARN(ret == -EEXIST, KERN_WARNING "sysfs: duplicate filename '%s' " "can not be created\n", sd->s_name); - WARN_ON(1); - } return ret; } diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 3f07893ff896..c9e4e5091da1 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -337,9 +337,8 @@ static int sysfs_open_file(struct inode *inode, struct file *file) if (kobj->ktype && kobj->ktype->sysfs_ops) ops = kobj->ktype->sysfs_ops; else { - printk(KERN_ERR "missing sysfs attribute operations for " + WARN(1, KERN_ERR "missing sysfs attribute operations for " "kobject: %s\n", kobject_name(kobj)); - WARN_ON(1); goto err_out; } diff --git a/fs/sysfs/group.c b/fs/sysfs/group.c index eeba38417b1d..fe611949a7f7 100644 --- a/fs/sysfs/group.c +++ b/fs/sysfs/group.c @@ -134,9 +134,8 @@ void sysfs_remove_group(struct kobject * kobj, if (grp->name) { sd = sysfs_get_dirent(dir_sd, grp->name); if (!sd) { - printk(KERN_WARNING "sysfs group %p not found for " + WARN(!sd, KERN_WARNING "sysfs group %p not found for " "kobject '%s'\n", grp, kobject_name(kobj)); - WARN_ON(!sd); return; } } else -- cgit v1.2.3 From 267e2a9c71b8e088ac307f9549f71468e86e26c1 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Fri, 25 Jul 2008 19:45:41 -0700 Subject: Use WARN() in fs/proc/ Use WARN() instead of a printk+WARN_ON() pair; this way the message becomes part of the warning section for better reporting/collection. This way, the entire if() {} section can collapse into the WARN() as well. Signed-off-by: Arjan van de Ven Acked-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/generic.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/proc/generic.c b/fs/proc/generic.c index bc0a0dd2d844..cb4096cc3fb7 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -806,12 +806,9 @@ continue_removing: if (S_ISDIR(de->mode)) parent->nlink--; de->nlink = 0; - if (de->subdir) { - printk(KERN_WARNING "%s: removing non-empty directory " + WARN(de->subdir, 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); } -- cgit v1.2.3 From 6341c393fcc37d58727865f1ee2f65e632e9d4f0 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 25 Jul 2008 19:45:44 -0700 Subject: tracehook: exec This moves all the ptrace hooks related to exec into tracehook.h inlines. This also lifts the calls for tracing out of the binfmt load_binary hooks into search_binary_handler() after it calls into the binfmt module. This change has no effect, since all the binfmt modules' load_binary functions did the call at the end on success, and now search_binary_handler() does it immediately after return if successful. We consolidate the repeated code, and binfmt modules no longer need to import ptrace_notify(). Signed-off-by: Roland McGrath Cc: Oleg Nesterov Reviewed-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/ia32/ia32_aout.c | 6 ------ fs/binfmt_aout.c | 6 ------ fs/binfmt_elf.c | 6 ------ fs/binfmt_elf_fdpic.c | 7 ------- fs/binfmt_flat.c | 3 --- fs/binfmt_som.c | 2 -- fs/exec.c | 12 ++++-------- include/linux/tracehook.h | 46 ++++++++++++++++++++++++++++++++++++++++++++++ 8 files changed, 50 insertions(+), 38 deletions(-) (limited to 'fs') diff --git a/arch/x86/ia32/ia32_aout.c b/arch/x86/ia32/ia32_aout.c index 58cccb6483b0..a0e1dbe67dc1 100644 --- a/arch/x86/ia32/ia32_aout.c +++ b/arch/x86/ia32/ia32_aout.c @@ -441,12 +441,6 @@ beyond_if: regs->r8 = regs->r9 = regs->r10 = regs->r11 = regs->r12 = regs->r13 = regs->r14 = regs->r15 = 0; set_fs(USER_DS); - if (unlikely(current->ptrace & PT_PTRACED)) { - if (current->ptrace & PT_TRACE_EXEC) - ptrace_notify((PTRACE_EVENT_EXEC << 8) | SIGTRAP); - else - send_sig(SIGTRAP, current, 0); - } return 0; } diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index ba4cddb92f1d..204cfd1d7676 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -444,12 +444,6 @@ beyond_if: regs->gp = ex.a_gpvalue; #endif start_thread(regs, ex.a_entry, current->mm->start_stack); - if (unlikely(current->ptrace & PT_PTRACED)) { - if (current->ptrace & PT_TRACE_EXEC) - ptrace_notify ((PTRACE_EVENT_EXEC << 8) | SIGTRAP); - else - send_sig(SIGTRAP, current, 0); - } return 0; } diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 3b6ff854d983..655ed8d30a86 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1003,12 +1003,6 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) #endif start_thread(regs, elf_entry, bprm->p); - if (unlikely(current->ptrace & PT_PTRACED)) { - if (current->ptrace & PT_TRACE_EXEC) - ptrace_notify ((PTRACE_EVENT_EXEC << 8) | SIGTRAP); - else - send_sig(SIGTRAP, current, 0); - } retval = 0; out: kfree(loc); diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 1b59b1edf26d..fdeadab2f18b 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -433,13 +433,6 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, entryaddr = interp_params.entry_addr ?: exec_params.entry_addr; start_thread(regs, entryaddr, current->mm->start_stack); - if (unlikely(current->ptrace & PT_PTRACED)) { - if (current->ptrace & PT_TRACE_EXEC) - ptrace_notify((PTRACE_EVENT_EXEC << 8) | SIGTRAP); - else - send_sig(SIGTRAP, current, 0); - } - retval = 0; error: diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 2cb1acda3a82..56372ecf1690 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -920,9 +920,6 @@ static int load_flat_binary(struct linux_binprm * bprm, struct pt_regs * regs) start_thread(regs, start_addr, current->mm->start_stack); - if (current->ptrace & PT_PTRACED) - send_sig(SIGTRAP, current, 0); - return 0; } diff --git a/fs/binfmt_som.c b/fs/binfmt_som.c index fdc36bfd6a7b..68be580ba289 100644 --- a/fs/binfmt_som.c +++ b/fs/binfmt_som.c @@ -274,8 +274,6 @@ load_som_binary(struct linux_binprm * bprm, struct pt_regs * regs) map_hpux_gateway_page(current,current->mm); start_thread_som(regs, som_entry, bprm->p); - if (current->ptrace & PT_PTRACED) - send_sig(SIGTRAP, current, 0); return 0; /* error cleanup */ diff --git a/fs/exec.c b/fs/exec.c index 5e559013e303..b8792a131533 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -42,13 +42,13 @@ #include #include #include -#include #include #include #include #include #include #include +#include #include #include @@ -1071,13 +1071,8 @@ EXPORT_SYMBOL(prepare_binprm); static int unsafe_exec(struct task_struct *p) { - int unsafe = 0; - if (p->ptrace & PT_PTRACED) { - if (p->ptrace & PT_PTRACE_CAP) - unsafe |= LSM_UNSAFE_PTRACE_CAP; - else - unsafe |= LSM_UNSAFE_PTRACE; - } + int unsafe = tracehook_unsafe_exec(p); + if (atomic_read(&p->fs->count) > 1 || atomic_read(&p->files->count) > 1 || atomic_read(&p->sighand->count) > 1) @@ -1214,6 +1209,7 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs) read_unlock(&binfmt_lock); retval = fn(bprm, regs); if (retval >= 0) { + tracehook_report_exec(fmt, bprm, regs); put_binfmt(fmt); allow_write_access(bprm->file); if (bprm->file) diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h index bea0f3eeff54..6276353709c1 100644 --- a/include/linux/tracehook.h +++ b/include/linux/tracehook.h @@ -48,5 +48,51 @@ #include #include +#include +struct linux_binprm; + +/** + * tracehook_unsafe_exec - check for exec declared unsafe due to tracing + * @task: current task doing exec + * + * Return %LSM_UNSAFE_* bits applied to an exec because of tracing. + * + * Called with task_lock() held on @task. + */ +static inline int tracehook_unsafe_exec(struct task_struct *task) +{ + int unsafe = 0; + int ptrace = task_ptrace(task); + if (ptrace & PT_PTRACED) { + if (ptrace & PT_PTRACE_CAP) + unsafe |= LSM_UNSAFE_PTRACE_CAP; + else + unsafe |= LSM_UNSAFE_PTRACE; + } + return unsafe; +} + +/** + * tracehook_report_exec - a successful exec was completed + * @fmt: &struct linux_binfmt that performed the exec + * @bprm: &struct linux_binprm containing exec details + * @regs: user-mode register state + * + * An exec just completed, we are shortly going to return to user mode. + * The freshly initialized register state can be seen and changed in @regs. + * The name, file and other pointers in @bprm are still on hand to be + * inspected, but will be freed as soon as this returns. + * + * Called with no locks, but with some kernel resources held live + * and a reference on @fmt->module. + */ +static inline void tracehook_report_exec(struct linux_binfmt *fmt, + struct linux_binprm *bprm, + struct pt_regs *regs) +{ + if (!ptrace_event(PT_TRACE_EXEC, PTRACE_EVENT_EXEC, 0) && + unlikely(task_ptrace(current) & PT_PTRACED)) + send_sig(SIGTRAP, current, 0); +} #endif /* */ -- cgit v1.2.3 From 0d094efeb1e98010c6b99923f1eb7e17bf1e3a74 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 25 Jul 2008 19:45:49 -0700 Subject: tracehook: tracehook_tracer_task This adds the tracehook_tracer_task() hook to consolidate all forms of "Who is using ptrace on me?" logic. This is used for "TracerPid:" in /proc and for permission checks. We also clean up the selinux code the called an identical accessor. Signed-off-by: Roland McGrath Cc: Oleg Nesterov Reviewed-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/array.c | 9 +++++++-- fs/proc/base.c | 13 +++++++++---- include/linux/tracehook.h | 18 ++++++++++++++++++ security/selinux/hooks.c | 22 +++------------------- 4 files changed, 37 insertions(+), 25 deletions(-) (limited to 'fs') diff --git a/fs/proc/array.c b/fs/proc/array.c index 797d775e0354..0d6eb33597c6 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -80,6 +80,7 @@ #include #include #include +#include #include #include @@ -168,8 +169,12 @@ static inline void task_state(struct seq_file *m, struct pid_namespace *ns, rcu_read_lock(); ppid = pid_alive(p) ? task_tgid_nr_ns(rcu_dereference(p->real_parent), ns) : 0; - tpid = pid_alive(p) && p->ptrace ? - task_pid_nr_ns(rcu_dereference(p->parent), ns) : 0; + tpid = 0; + if (pid_alive(p)) { + struct task_struct *tracer = tracehook_tracer_task(p); + if (tracer) + tpid = task_pid_nr_ns(tracer, ns); + } seq_printf(m, "State:\t%s\n" "Tgid:\t%d\n" diff --git a/fs/proc/base.c b/fs/proc/base.c index a891fe4cb43b..4b74dba69a6d 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -69,6 +69,7 @@ #include #include #include +#include #include #include #include @@ -231,10 +232,14 @@ static int check_mem_permission(struct task_struct *task) * 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_access(task, PTRACE_MODE_ATTACH)) - return 0; + if (task_is_stopped_or_traced(task)) { + int match; + rcu_read_lock(); + match = (tracehook_tracer_task(task) == current); + rcu_read_unlock(); + if (match && ptrace_may_access(task, PTRACE_MODE_ATTACH)) + return 0; + } /* * Noone else is allowed. diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h index 9a5b3be2503a..6468ca0fe69b 100644 --- a/include/linux/tracehook.h +++ b/include/linux/tracehook.h @@ -72,6 +72,24 @@ static inline int tracehook_unsafe_exec(struct task_struct *task) return unsafe; } +/** + * tracehook_tracer_task - return the task that is tracing the given task + * @tsk: task to consider + * + * Returns NULL if noone is tracing @task, or the &struct task_struct + * pointer to its tracer. + * + * Must called under rcu_read_lock(). The pointer returned might be kept + * live only by RCU. During exec, this may be called with task_lock() + * held on @task, still held from when tracehook_unsafe_exec() was called. + */ +static inline struct task_struct *tracehook_tracer_task(struct task_struct *tsk) +{ + if (task_ptrace(tsk) & PT_PTRACED) + return rcu_dereference(tsk->parent); + return NULL; +} + /** * tracehook_report_exec - a successful exec was completed * @fmt: &struct linux_binfmt that performed the exec diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 63f131fc42e4..3481cde5bf15 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -25,7 +25,7 @@ #include #include -#include +#include #include #include #include @@ -1971,22 +1971,6 @@ static int selinux_vm_enough_memory(struct mm_struct *mm, long pages) return __vm_enough_memory(mm, pages, cap_sys_admin); } -/** - * task_tracer_task - return the task that is tracing the given task - * @task: task to consider - * - * Returns NULL if noone is tracing @task, or the &struct task_struct - * pointer to its tracer. - * - * Must be called under rcu_read_lock(). - */ -static struct task_struct *task_tracer_task(struct task_struct *task) -{ - if (task->ptrace & PT_PTRACED) - return rcu_dereference(task->parent); - return NULL; -} - /* binprm security operations */ static int selinux_bprm_alloc_security(struct linux_binprm *bprm) @@ -2238,7 +2222,7 @@ static void selinux_bprm_apply_creds(struct linux_binprm *bprm, int unsafe) u32 ptsid = 0; rcu_read_lock(); - tracer = task_tracer_task(current); + tracer = tracehook_tracer_task(current); if (likely(tracer != NULL)) { sec = tracer->security; ptsid = sec->sid; @@ -5247,7 +5231,7 @@ static int selinux_setprocattr(struct task_struct *p, Otherwise, leave SID unchanged and fail. */ task_lock(p); rcu_read_lock(); - tracer = task_tracer_task(p); + tracer = tracehook_tracer_task(p); if (tracer != NULL) { struct task_security_struct *ptsec = tracer->security; u32 ptsid = ptsec->sid; -- cgit v1.2.3 From ebcb67341fee34061430f3367f2e507e52ee051b Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 25 Jul 2008 19:46:00 -0700 Subject: /proc/PID/syscall This adds /proc/PID/syscall and /proc/PID/task/TID/syscall magic files. These use task_current_syscall() to show the task's current system call number and argument registers, stack pointer and PC. For a task blocked but not in a syscall, the file shows "-1" in place of the syscall number, followed by only the SP and PC. For a task that's not blocked, it shows "running". Signed-off-by: Roland McGrath Cc: Oleg Nesterov Reviewed-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/base.c | 26 ++++++++++++++++++++++++++ 1 file changed, 26 insertions(+) (limited to 'fs') diff --git a/fs/proc/base.c b/fs/proc/base.c index 4b74dba69a6d..81bce6791bfc 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -509,6 +509,26 @@ static int proc_pid_limits(struct task_struct *task, char *buffer) return count; } +#ifdef CONFIG_HAVE_ARCH_TRACEHOOK +static int proc_pid_syscall(struct task_struct *task, char *buffer) +{ + long nr; + unsigned long args[6], sp, pc; + + if (task_current_syscall(task, &nr, args, 6, &sp, &pc)) + return sprintf(buffer, "running\n"); + + if (nr < 0) + return sprintf(buffer, "%ld 0x%lx 0x%lx\n", nr, sp, pc); + + return sprintf(buffer, + "%ld 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx\n", + nr, + args[0], args[1], args[2], args[3], args[4], args[5], + sp, pc); +} +#endif /* CONFIG_HAVE_ARCH_TRACEHOOK */ + /************************************************************************/ /* Here the fs part begins */ /************************************************************************/ @@ -2477,6 +2497,9 @@ static const struct pid_entry tgid_base_stuff[] = { INF("limits", S_IRUSR, pid_limits), #ifdef CONFIG_SCHED_DEBUG REG("sched", S_IRUGO|S_IWUSR, pid_sched), +#endif +#ifdef CONFIG_HAVE_ARCH_TRACEHOOK + INF("syscall", S_IRUSR, pid_syscall), #endif INF("cmdline", S_IRUGO, pid_cmdline), ONE("stat", S_IRUGO, tgid_stat), @@ -2809,6 +2832,9 @@ static const struct pid_entry tid_base_stuff[] = { INF("limits", S_IRUSR, pid_limits), #ifdef CONFIG_SCHED_DEBUG REG("sched", S_IRUGO|S_IWUSR, pid_sched), +#endif +#ifdef CONFIG_HAVE_ARCH_TRACEHOOK + INF("syscall", S_IRUSR, pid_syscall), #endif INF("cmdline", S_IRUGO, pid_cmdline), ONE("stat", S_IRUGO, tid_stat), -- cgit v1.2.3 From d70b67c8bc72ee23b55381bd6a884f4796692f77 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 2 Jul 2008 21:30:15 +0200 Subject: [patch] vfs: fix lookup on deleted directory Lookup can install a child dentry for a deleted directory. This keeps the directory dentry alive, and the inode pinned in the cache and on disk, even after all external references have gone away. This isn't a big problem normally, since memory pressure or umount will clear out the directory dentry and its children, releasing the inode. But for UBIFS this causes problems because its orphan area can overflow. Fix this by returning ENOENT for all lookups on a S_DEAD directory before creating a child dentry. Thanks to Zoltan Sogor for noticing this while testing UBIFS, and Artem for the excellent analysis of the problem and testing. Reported-by: Artem Bityutskiy Tested-by: Artem Bityutskiy Signed-off-by: Miklos Szeredi Signed-off-by: Al Viro --- fs/namei.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index 01e67dddcc3d..3b26a240ade9 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -519,7 +519,14 @@ static struct dentry * real_lookup(struct dentry * parent, struct qstr * name, s */ result = d_lookup(parent, name); if (!result) { - struct dentry * dentry = d_alloc(parent, name); + struct dentry *dentry; + + /* Don't create child dentry for a dead directory. */ + result = ERR_PTR(-ENOENT); + if (IS_DEADDIR(dir)) + goto out_unlock; + + dentry = d_alloc(parent, name); result = ERR_PTR(-ENOMEM); if (dentry) { result = dir->i_op->lookup(dir, dentry, nd); @@ -528,6 +535,7 @@ static struct dentry * real_lookup(struct dentry * parent, struct qstr * name, s else result = dentry; } +out_unlock: mutex_unlock(&dir->i_mutex); return result; } @@ -1317,7 +1325,14 @@ static struct dentry *__lookup_hash(struct qstr *name, dentry = cached_lookup(base, name, nd); if (!dentry) { - struct dentry *new = d_alloc(base, name); + struct dentry *new; + + /* Don't create child dentry for a dead directory. */ + dentry = ERR_PTR(-ENOENT); + if (IS_DEADDIR(inode)) + goto out; + + new = d_alloc(base, name); dentry = ERR_PTR(-ENOMEM); if (!new) goto out; -- cgit v1.2.3 From d2d9648ec6858e19d16a0b16da62534e85888653 Mon Sep 17 00:00:00 2001 From: Denys Vlasenko Date: Tue, 1 Jul 2008 14:16:09 +0200 Subject: [PATCH] reuse xxx_fifo_fops for xxx_pipe_fops Merge fifo and pipe file_operations. Signed-off-by: Denys Vlasenko Signed-off-by: Al Viro --- fs/fifo.c | 8 ++++---- fs/pipe.c | 51 ++++++++------------------------------------------- include/linux/fs.h | 6 +++--- 3 files changed, 15 insertions(+), 50 deletions(-) (limited to 'fs') diff --git a/fs/fifo.c b/fs/fifo.c index 9785e36f81e7..987bf9411495 100644 --- a/fs/fifo.c +++ b/fs/fifo.c @@ -57,7 +57,7 @@ static int fifo_open(struct inode *inode, struct file *filp) * POSIX.1 says that O_NONBLOCK means return with the FIFO * opened, even when there is no process writing the FIFO. */ - filp->f_op = &read_fifo_fops; + filp->f_op = &read_pipefifo_fops; pipe->r_counter++; if (pipe->readers++ == 0) wake_up_partner(inode); @@ -86,7 +86,7 @@ static int fifo_open(struct inode *inode, struct file *filp) if ((filp->f_flags & O_NONBLOCK) && !pipe->readers) goto err; - filp->f_op = &write_fifo_fops; + filp->f_op = &write_pipefifo_fops; pipe->w_counter++; if (!pipe->writers++) wake_up_partner(inode); @@ -105,7 +105,7 @@ static int fifo_open(struct inode *inode, struct file *filp) * This implementation will NEVER block on a O_RDWR open, since * the process can at least talk to itself. */ - filp->f_op = &rdwr_fifo_fops; + filp->f_op = &rdwr_pipefifo_fops; pipe->readers++; pipe->writers++; @@ -151,5 +151,5 @@ err_nocleanup: * depending on the access mode of the file... */ const struct file_operations def_fifo_fops = { - .open = fifo_open, /* will set read or write pipe_fops */ + .open = fifo_open, /* will set read_ or write_pipefifo_fops */ }; diff --git a/fs/pipe.c b/fs/pipe.c index 10c4e9aa5c49..fcba6542b8d0 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -777,45 +777,10 @@ pipe_rdwr_open(struct inode *inode, struct file *filp) /* * The file_operations structs are not static because they * are also used in linux/fs/fifo.c to do operations on FIFOs. + * + * Pipes reuse fifos' file_operations structs. */ -const struct file_operations read_fifo_fops = { - .llseek = no_llseek, - .read = do_sync_read, - .aio_read = pipe_read, - .write = bad_pipe_w, - .poll = pipe_poll, - .unlocked_ioctl = pipe_ioctl, - .open = pipe_read_open, - .release = pipe_read_release, - .fasync = pipe_read_fasync, -}; - -const struct file_operations write_fifo_fops = { - .llseek = no_llseek, - .read = bad_pipe_r, - .write = do_sync_write, - .aio_write = pipe_write, - .poll = pipe_poll, - .unlocked_ioctl = pipe_ioctl, - .open = pipe_write_open, - .release = pipe_write_release, - .fasync = pipe_write_fasync, -}; - -const struct file_operations rdwr_fifo_fops = { - .llseek = no_llseek, - .read = do_sync_read, - .aio_read = pipe_read, - .write = do_sync_write, - .aio_write = pipe_write, - .poll = pipe_poll, - .unlocked_ioctl = pipe_ioctl, - .open = pipe_rdwr_open, - .release = pipe_rdwr_release, - .fasync = pipe_rdwr_fasync, -}; - -static const struct file_operations read_pipe_fops = { +const struct file_operations read_pipefifo_fops = { .llseek = no_llseek, .read = do_sync_read, .aio_read = pipe_read, @@ -827,7 +792,7 @@ static const struct file_operations read_pipe_fops = { .fasync = pipe_read_fasync, }; -static const struct file_operations write_pipe_fops = { +const struct file_operations write_pipefifo_fops = { .llseek = no_llseek, .read = bad_pipe_r, .write = do_sync_write, @@ -839,7 +804,7 @@ static const struct file_operations write_pipe_fops = { .fasync = pipe_write_fasync, }; -static const struct file_operations rdwr_pipe_fops = { +const struct file_operations rdwr_pipefifo_fops = { .llseek = no_llseek, .read = do_sync_read, .aio_read = pipe_read, @@ -927,7 +892,7 @@ static struct inode * get_pipe_inode(void) inode->i_pipe = pipe; pipe->readers = pipe->writers = 1; - inode->i_fop = &rdwr_pipe_fops; + inode->i_fop = &rdwr_pipefifo_fops; /* * Mark the inode dirty from the very beginning, @@ -978,7 +943,7 @@ struct file *create_write_pipe(int flags) d_instantiate(dentry, inode); err = -ENFILE; - f = alloc_file(pipe_mnt, dentry, FMODE_WRITE, &write_pipe_fops); + f = alloc_file(pipe_mnt, dentry, FMODE_WRITE, &write_pipefifo_fops); if (!f) goto err_dentry; f->f_mapping = inode->i_mapping; @@ -1020,7 +985,7 @@ struct file *create_read_pipe(struct file *wrf, int flags) f->f_pos = 0; f->f_flags = O_RDONLY | (flags & O_NONBLOCK); - f->f_op = &read_pipe_fops; + f->f_op = &read_pipefifo_fops; f->f_mode = FMODE_READ; f->f_version = 0; diff --git a/include/linux/fs.h b/include/linux/fs.h index 53d2edb709b3..7721a2ac9c0e 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1696,9 +1696,9 @@ extern void init_special_inode(struct inode *, umode_t, dev_t); extern void make_bad_inode(struct inode *); extern int is_bad_inode(struct inode *); -extern const struct file_operations read_fifo_fops; -extern const struct file_operations write_fifo_fops; -extern const struct file_operations rdwr_fifo_fops; +extern const struct file_operations read_pipefifo_fops; +extern const struct file_operations write_pipefifo_fops; +extern const struct file_operations rdwr_pipefifo_fops; extern int fs_may_remount_ro(struct super_block *); -- cgit v1.2.3 From 7ac6cd653d7c31ad6b7bb5b88c549c4ebf628c34 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 1 Jul 2008 23:07:54 +0200 Subject: [patch] hppfs: remove hppfs_permission hppfs_permission() is equivalent to the '.permission == NULL' case. Signed-off-by: Miklos Szeredi Signed-off-by: Al Viro --- fs/hppfs/hppfs.c | 7 ------- 1 file changed, 7 deletions(-) (limited to 'fs') diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c index 65077aa90f0a..2b3d1828db99 100644 --- a/fs/hppfs/hppfs.c +++ b/fs/hppfs/hppfs.c @@ -655,20 +655,13 @@ static void *hppfs_follow_link(struct dentry *dentry, struct nameidata *nd) return proc_dentry->d_inode->i_op->follow_link(proc_dentry, nd); } -int hppfs_permission(struct inode *inode, int mask, struct nameidata *nd) -{ - return generic_permission(inode, mask, NULL); -} - static const struct inode_operations hppfs_dir_iops = { .lookup = hppfs_lookup, - .permission = hppfs_permission, }; static const struct inode_operations hppfs_link_iops = { .readlink = hppfs_readlink, .follow_link = hppfs_follow_link, - .permission = hppfs_permission, }; static struct inode *get_inode(struct super_block *sb, struct dentry *dentry) -- cgit v1.2.3 From 9043476f726802f4b00c96d0c4f418dde48d1304 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 15 Jul 2008 08:54:06 -0400 Subject: [PATCH] sanitize proc_sysctl * keep references to ctl_table_head and ctl_table in /proc/sys inodes * grab the former during operations, use the latter for access to entry if that succeeds * have ->d_compare() check if table should be seen for one who does lookup; that allows us to avoid flipping inodes - if we have the same name resolve to different things, we'll just keep several dentries and ->d_compare() will reject the wrong ones. * have ->lookup() and ->readdir() scan the table of our inode first, then walk all ctl_table_header and scan ->attached_by for those that are attached to our directory. * implement ->getattr(). * get rid of insane amounts of tree-walking * get rid of the need to know dentry in ->permission() and of the contortions induced by that. Signed-off-by: Al Viro --- fs/proc/inode.c | 5 + fs/proc/proc_sysctl.c | 427 ++++++++++++++++++++++-------------------------- include/linux/proc_fs.h | 5 + include/linux/sysctl.h | 1 + kernel/sysctl.c | 15 ++ 5 files changed, 218 insertions(+), 235 deletions(-) (limited to 'fs') diff --git a/fs/proc/inode.c b/fs/proc/inode.c index b37f25dc45a5..8bb03f056c28 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include @@ -65,6 +66,8 @@ static void proc_delete_inode(struct inode *inode) module_put(de->owner); de_put(de); } + if (PROC_I(inode)->sysctl) + sysctl_head_put(PROC_I(inode)->sysctl); clear_inode(inode); } @@ -84,6 +87,8 @@ static struct inode *proc_alloc_inode(struct super_block *sb) ei->fd = 0; ei->op.proc_get_link = NULL; ei->pde = NULL; + ei->sysctl = NULL; + ei->sysctl_entry = NULL; inode = &ei->vfs_inode; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; return inode; diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 5acc001d49f6..fa1ec2433e44 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -10,149 +10,110 @@ static struct dentry_operations proc_sys_dentry_operations; static const struct file_operations proc_sys_file_operations; static const struct inode_operations proc_sys_inode_operations; +static const struct file_operations proc_sys_dir_file_operations; +static const struct inode_operations proc_sys_dir_operations; -static void proc_sys_refresh_inode(struct inode *inode, struct ctl_table *table) -{ - /* Refresh the cached information bits in the inode */ - if (table) { - inode->i_uid = 0; - inode->i_gid = 0; - inode->i_mode = table->mode; - if (table->proc_handler) { - inode->i_mode |= S_IFREG; - inode->i_nlink = 1; - } else { - inode->i_mode |= S_IFDIR; - inode->i_nlink = 0; /* It is too hard to figure out */ - } - } -} - -static struct inode *proc_sys_make_inode(struct inode *dir, struct ctl_table *table) +static struct inode *proc_sys_make_inode(struct super_block *sb, + struct ctl_table_header *head, struct ctl_table *table) { struct inode *inode; - struct proc_inode *dir_ei, *ei; - int depth; + struct proc_inode *ei; - inode = new_inode(dir->i_sb); + inode = new_inode(sb); if (!inode) goto out; - /* A directory is always one deeper than it's parent */ - dir_ei = PROC_I(dir); - depth = dir_ei->fd + 1; - + sysctl_head_get(head); ei = PROC_I(inode); - ei->fd = depth; + ei->sysctl = head; + ei->sysctl_entry = table; + inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; - inode->i_op = &proc_sys_inode_operations; - inode->i_fop = &proc_sys_file_operations; inode->i_flags |= S_PRIVATE; /* tell selinux to ignore this inode */ - proc_sys_refresh_inode(inode, table); + inode->i_mode = table->mode; + if (!table->child) { + inode->i_mode |= S_IFREG; + inode->i_op = &proc_sys_inode_operations; + inode->i_fop = &proc_sys_file_operations; + } else { + inode->i_mode |= S_IFDIR; + inode->i_nlink = 0; + inode->i_op = &proc_sys_dir_operations; + inode->i_fop = &proc_sys_dir_file_operations; + } out: return inode; } -static struct dentry *proc_sys_ancestor(struct dentry *dentry, int depth) -{ - for (;;) { - struct proc_inode *ei; - - ei = PROC_I(dentry->d_inode); - if (ei->fd == depth) - break; /* found */ - - dentry = dentry->d_parent; - } - return dentry; -} - -static struct ctl_table *proc_sys_lookup_table_one(struct ctl_table *table, - struct qstr *name) +static struct ctl_table *find_in_table(struct ctl_table *p, struct qstr *name) { int len; - for ( ; table->ctl_name || table->procname; table++) { + for ( ; p->ctl_name || p->procname; p++) { - if (!table->procname) + if (!p->procname) continue; - len = strlen(table->procname); + len = strlen(p->procname); if (len != name->len) continue; - if (memcmp(table->procname, name->name, len) != 0) + if (memcmp(p->procname, name->name, len) != 0) continue; /* I have a match */ - return table; + return p; } return NULL; } -static struct ctl_table *proc_sys_lookup_table(struct dentry *dentry, - struct ctl_table *table) +struct ctl_table_header *grab_header(struct inode *inode) { - struct dentry *ancestor; - struct proc_inode *ei; - int depth, i; + if (PROC_I(inode)->sysctl) + return sysctl_head_grab(PROC_I(inode)->sysctl); + else + return sysctl_head_next(NULL); +} - ei = PROC_I(dentry->d_inode); - depth = ei->fd; +static struct dentry *proc_sys_lookup(struct inode *dir, struct dentry *dentry, + struct nameidata *nd) +{ + struct ctl_table_header *head = grab_header(dir); + struct ctl_table *table = PROC_I(dir)->sysctl_entry; + struct ctl_table_header *h = NULL; + struct qstr *name = &dentry->d_name; + struct ctl_table *p; + struct inode *inode; + struct dentry *err = ERR_PTR(-ENOENT); - if (depth == 0) - return table; + if (IS_ERR(head)) + return ERR_CAST(head); - for (i = 1; table && (i <= depth); i++) { - ancestor = proc_sys_ancestor(dentry, i); - table = proc_sys_lookup_table_one(table, &ancestor->d_name); - if (table) - table = table->child; + if (table && !table->child) { + WARN_ON(1); + goto out; } - return table; - -} -static struct ctl_table *proc_sys_lookup_entry(struct dentry *dparent, - struct qstr *name, - struct ctl_table *table) -{ - table = proc_sys_lookup_table(dparent, table); - if (table) - table = proc_sys_lookup_table_one(table, name); - return table; -} -static struct ctl_table *do_proc_sys_lookup(struct dentry *parent, - struct qstr *name, - struct ctl_table_header **ptr) -{ - struct ctl_table_header *head; - struct ctl_table *table = NULL; + table = table ? table->child : head->ctl_table; - for (head = sysctl_head_next(NULL); head; - head = sysctl_head_next(head)) { - table = proc_sys_lookup_entry(parent, name, head->ctl_table); - if (table) - break; + p = find_in_table(table, name); + if (!p) { + for (h = sysctl_head_next(NULL); h; h = sysctl_head_next(h)) { + if (h->attached_to != table) + continue; + p = find_in_table(h->attached_by, name); + if (p) + break; + } } - *ptr = head; - return table; -} - -static struct dentry *proc_sys_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) -{ - struct ctl_table_header *head; - struct inode *inode; - struct dentry *err; - struct ctl_table *table; - err = ERR_PTR(-ENOENT); - table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head); - if (!table) + if (!p) goto out; err = ERR_PTR(-ENOMEM); - inode = proc_sys_make_inode(dir, table); + inode = proc_sys_make_inode(dir->i_sb, h ? h : head, p); + if (h) + sysctl_head_finish(h); + if (!inode) goto out; @@ -168,22 +129,14 @@ out: 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; - struct ctl_table *table; + struct inode *inode = filp->f_path.dentry->d_inode; + struct ctl_table_header *head = grab_header(inode); + struct ctl_table *table = PROC_I(inode)->sysctl_entry; 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; + if (IS_ERR(head)) + return PTR_ERR(head); /* * At this point we know that the sysctl was not unregistered @@ -193,6 +146,11 @@ static ssize_t proc_sys_call_handler(struct file *filp, void __user *buf, if (sysctl_perm(head->root, table, write ? MAY_WRITE : MAY_READ)) goto out; + /* if that can happen at all, it should be -EINVAL, not -EISDIR */ + error = -EINVAL; + if (!table->proc_handler) + goto out; + /* careful: calling conventions are nasty here */ res = count; error = table->proc_handler(table, write, filp, buf, &res, ppos); @@ -218,82 +176,86 @@ static ssize_t proc_sys_write(struct file *filp, const char __user *buf, static int proc_sys_fill_cache(struct file *filp, void *dirent, - filldir_t filldir, struct ctl_table *table) + filldir_t filldir, + struct ctl_table_header *head, + struct ctl_table *table) { - struct ctl_table_header *head; - struct ctl_table *child_table = NULL; struct dentry *child, *dir = filp->f_path.dentry; struct inode *inode; struct qstr qname; ino_t ino = 0; unsigned type = DT_UNKNOWN; - int ret; qname.name = table->procname; qname.len = strlen(table->procname); qname.hash = full_name_hash(qname.name, qname.len); - /* Suppress duplicates. - * Only fill a directory entry if it is the value that - * an ordinary lookup of that name returns. Hide all - * others. - * - * If we ever cache this translation in the dcache - * I should do a dcache lookup first. But for now - * it is just simpler not to. - */ - ret = 0; - child_table = do_proc_sys_lookup(dir, &qname, &head); - sysctl_head_finish(head); - if (child_table != table) - return 0; - child = d_lookup(dir, &qname); if (!child) { - struct dentry *new; - new = d_alloc(dir, &qname); - if (new) { - inode = proc_sys_make_inode(dir->d_inode, table); - if (!inode) - child = ERR_PTR(-ENOMEM); - else { - new->d_op = &proc_sys_dentry_operations; - d_add(new, inode); + child = d_alloc(dir, &qname); + if (child) { + inode = proc_sys_make_inode(dir->d_sb, head, table); + if (!inode) { + dput(child); + return -ENOMEM; + } else { + child->d_op = &proc_sys_dentry_operations; + d_add(child, inode); } - if (child) - dput(new); - else - child = new; + } else { + return -ENOMEM; } } - if (!child || IS_ERR(child) || !child->d_inode) - goto end_instantiate; inode = child->d_inode; - if (inode) { - ino = inode->i_ino; - type = inode->i_mode >> 12; - } + ino = inode->i_ino; + type = inode->i_mode >> 12; dput(child); -end_instantiate: - if (!ino) - ino= find_inode_number(dir, &qname); - if (!ino) - ino = 1; - return filldir(dirent, qname.name, qname.len, filp->f_pos, ino, type); + return !!filldir(dirent, qname.name, qname.len, filp->f_pos, ino, type); +} + +static int scan(struct ctl_table_header *head, ctl_table *table, + unsigned long *pos, struct file *file, + void *dirent, filldir_t filldir) +{ + + for (; table->ctl_name || table->procname; table++, (*pos)++) { + int res; + + /* Can't do anything without a proc name */ + if (!table->procname) + continue; + + if (*pos < file->f_pos) + continue; + + res = proc_sys_fill_cache(file, dirent, filldir, head, table); + if (res) + return res; + + file->f_pos = *pos + 1; + } + return 0; } static int proc_sys_readdir(struct file *filp, void *dirent, filldir_t filldir) { - struct dentry *dentry = filp->f_dentry; + struct dentry *dentry = filp->f_path.dentry; struct inode *inode = dentry->d_inode; - struct ctl_table_header *head = NULL; - struct ctl_table *table; + struct ctl_table_header *head = grab_header(inode); + struct ctl_table *table = PROC_I(inode)->sysctl_entry; + struct ctl_table_header *h = NULL; unsigned long pos; - int ret; + int ret = -EINVAL; + + if (IS_ERR(head)) + return PTR_ERR(head); - ret = -ENOTDIR; - if (!S_ISDIR(inode->i_mode)) + if (table && !table->child) { + WARN_ON(1); goto out; + } + + table = table ? table->child : head->ctl_table; ret = 0; /* Avoid a switch here: arm builds fail with missing __cmpdi2 */ @@ -311,30 +273,17 @@ static int proc_sys_readdir(struct file *filp, void *dirent, filldir_t filldir) } pos = 2; - /* - Find each instance of the directory - * - Read all entries in each instance - * - Before returning an entry to user space lookup the entry - * by name and if I find a different entry don't return - * this one because it means it is a buried dup. - * For sysctl this should only happen for directory entries. - */ - for (head = sysctl_head_next(NULL); head; head = sysctl_head_next(head)) { - table = proc_sys_lookup_table(dentry, head->ctl_table); + ret = scan(head, table, &pos, filp, dirent, filldir); + if (ret) + goto out; - if (!table) + for (h = sysctl_head_next(NULL); h; h = sysctl_head_next(h)) { + if (h->attached_to != table) continue; - - for (; table->ctl_name || table->procname; table++, pos++) { - /* Can't do anything without a proc name */ - if (!table->procname) - continue; - - if (pos < filp->f_pos) - continue; - - if (proc_sys_fill_cache(filp, dirent, filldir, table) < 0) - goto out; - filp->f_pos = pos + 1; + ret = scan(h, h->attached_by, &pos, filp, dirent, filldir); + if (ret) { + sysctl_head_finish(h); + break; } } ret = 1; @@ -349,47 +298,18 @@ static int proc_sys_permission(struct inode *inode, int mask, struct nameidata * * sysctl entries that are not writeable, * are _NOT_ writeable, capabilities or not. */ - struct ctl_table_header *head; - struct ctl_table *table; - struct dentry *dentry; - int mode; - int depth; + struct ctl_table_header *head = grab_header(inode); + struct ctl_table *table = PROC_I(inode)->sysctl_entry; int error; - head = NULL; - depth = PROC_I(inode)->fd; - - /* First check the cached permissions, in case we don't have - * enough information to lookup the sysctl table entry. - */ - error = -EACCES; - mode = inode->i_mode; - - if (current->euid == 0) - mode >>= 6; - else if (in_group_p(0)) - mode >>= 3; - - if ((mode & mask & (MAY_READ|MAY_WRITE|MAY_EXEC)) == mask) - error = 0; - - /* If we can't get a sysctl table entry the permission - * checks on the cached mode will have to be enough. - */ - if (!nd || !depth) - goto out; + if (IS_ERR(head)) + return PTR_ERR(head); - dentry = nd->path.dentry; - table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head); + if (!table) /* global root - r-xr-xr-x */ + error = mask & MAY_WRITE ? -EACCES : 0; + else /* Use the permissions on the sysctl table entry */ + error = sysctl_perm(head->root, table, mask); - /* If the entry does not exist deny permission */ - error = -EACCES; - if (!table) - goto out; - - /* Use the permissions on the sysctl table entry */ - error = sysctl_perm(head->root, table, mask); -out: sysctl_head_finish(head); return error; } @@ -409,33 +329,70 @@ static int proc_sys_setattr(struct dentry *dentry, struct iattr *attr) return error; } -/* I'm lazy and don't distinguish between files and directories, - * until access time. - */ +static int proc_sys_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) +{ + struct inode *inode = dentry->d_inode; + struct ctl_table_header *head = grab_header(inode); + struct ctl_table *table = PROC_I(inode)->sysctl_entry; + + if (IS_ERR(head)) + return PTR_ERR(head); + + generic_fillattr(inode, stat); + if (table) + stat->mode = (stat->mode & S_IFMT) | table->mode; + + sysctl_head_finish(head); + return 0; +} + static const struct file_operations proc_sys_file_operations = { .read = proc_sys_read, .write = proc_sys_write, +}; + +static const struct file_operations proc_sys_dir_file_operations = { .readdir = proc_sys_readdir, }; static const struct inode_operations proc_sys_inode_operations = { + .permission = proc_sys_permission, + .setattr = proc_sys_setattr, + .getattr = proc_sys_getattr, +}; + +static const struct inode_operations proc_sys_dir_operations = { .lookup = proc_sys_lookup, .permission = proc_sys_permission, .setattr = proc_sys_setattr, + .getattr = proc_sys_getattr, }; static int proc_sys_revalidate(struct dentry *dentry, struct nameidata *nd) { - struct ctl_table_header *head; - struct ctl_table *table; - table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head); - proc_sys_refresh_inode(dentry->d_inode, table); - sysctl_head_finish(head); - return !!table; + return !PROC_I(dentry->d_inode)->sysctl->unregistering; +} + +static int proc_sys_delete(struct dentry *dentry) +{ + return !!PROC_I(dentry->d_inode)->sysctl->unregistering; +} + +static int proc_sys_compare(struct dentry *dir, struct qstr *qstr, + struct qstr *name) +{ + struct dentry *dentry = container_of(qstr, struct dentry, d_name); + if (qstr->len != name->len) + return 1; + if (memcmp(qstr->name, name->name, name->len)) + return 1; + return !sysctl_is_seen(PROC_I(dentry->d_inode)->sysctl); } static struct dentry_operations proc_sys_dentry_operations = { .d_revalidate = proc_sys_revalidate, + .d_delete = proc_sys_delete, + .d_compare = proc_sys_compare, }; static struct proc_dir_entry *proc_sys_root; @@ -443,8 +400,8 @@ static struct proc_dir_entry *proc_sys_root; int proc_sys_init(void) { proc_sys_root = proc_mkdir("sys", NULL); - proc_sys_root->proc_iops = &proc_sys_inode_operations; - proc_sys_root->proc_fops = &proc_sys_file_operations; + proc_sys_root->proc_iops = &proc_sys_dir_operations; + proc_sys_root->proc_fops = &proc_sys_dir_file_operations; proc_sys_root->nlink = 0; return 0; } diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index f560d1705afe..fb61850d1cfc 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -282,11 +282,16 @@ union proc_op { struct task_struct *task); }; +struct ctl_table_header; +struct ctl_table; + struct proc_inode { struct pid *pid; int fd; union proc_op op; struct proc_dir_entry *pde; + struct ctl_table_header *sysctl; + struct ctl_table *sysctl_entry; struct inode vfs_inode; }; diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 3f6599aeb0db..d0437f36921f 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -961,6 +961,7 @@ struct ctl_table_header; extern void sysctl_head_get(struct ctl_table_header *); extern void sysctl_head_put(struct ctl_table_header *); +extern int sysctl_is_seen(struct ctl_table_header *); extern struct ctl_table_header *sysctl_head_grab(struct ctl_table_header *); extern struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev); extern struct ctl_table_header *__sysctl_head_next(struct nsproxy *namespaces, diff --git a/kernel/sysctl.c b/kernel/sysctl.c index c9a0af887033..ff5abcca5ddf 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1932,6 +1932,21 @@ void unregister_sysctl_table(struct ctl_table_header * header) spin_unlock(&sysctl_lock); } +int sysctl_is_seen(struct ctl_table_header *p) +{ + struct ctl_table_set *set = p->set; + int res; + spin_lock(&sysctl_lock); + if (p->unregistering) + res = 0; + else if (!set->is_seen) + res = 1; + else + res = set->is_seen(set); + spin_unlock(&sysctl_lock); + return res; +} + void setup_sysctl_set(struct ctl_table_set *p, struct ctl_table_set *parent, int (*is_seen)(struct ctl_table_set *)) -- cgit v1.2.3 From 1bd5191d9f5d1928c4efdf604c4164b04bb88dbe Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 21 May 2008 19:15:03 +0200 Subject: [patch 05/14] hpfs: dont call permission() hpfs_unlink() calls permission() prior to truncating the file. HPFS doesn't define a .permission method, so replace with explicit call to generic_permission(). This is equivalent, except that devcgroup_inode_permission() and security_inode_permission() are not called. The truncation is just an implementation detail of the unlink, so these security checks are unnecessary. I suspect that even calling generic_permission() is unnecessary, since we shouldn't mind if the file isn't writable. But I leave that to the maintainer to decide. Signed-off-by: Miklos Szeredi CC: Mikulas Patocka --- fs/hpfs/namei.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c index d256559b4104..d9c59a775449 100644 --- a/fs/hpfs/namei.c +++ b/fs/hpfs/namei.c @@ -415,7 +415,7 @@ again: d_drop(dentry); spin_lock(&dentry->d_lock); if (atomic_read(&dentry->d_count) > 1 || - permission(inode, MAY_WRITE, NULL) || + generic_permission(inode, MAY_WRITE, NULL) || !S_ISREG(inode->i_mode) || get_write_access(inode)) { spin_unlock(&dentry->d_lock); -- cgit v1.2.3 From e6305c43eda10ebfd2ad9e35d6e172ccc7bb3695 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 15 Jul 2008 21:03:57 -0400 Subject: [PATCH] sanitize ->permission() prototype * kill nameidata * argument; map the 3 bits in ->flags anybody cares about to new MAY_... ones and pass with the mask. * kill redundant gfs2_iop_permission() * sanitize ecryptfs_permission() * fix remaining places where ->permission() instances might barf on new MAY_... found in mask. The obvious next target in that direction is permission(9) folded fix for nfs_permission() breakage from Miklos Szeredi Signed-off-by: Al Viro --- fs/afs/internal.h | 4 +--- fs/afs/security.c | 2 +- fs/bad_inode.c | 3 +-- fs/cifs/cifsfs.c | 2 +- fs/coda/dir.c | 4 +++- fs/coda/pioctl.c | 6 ++---- fs/ecryptfs/inode.c | 17 ++--------------- fs/ext2/acl.c | 2 +- fs/ext2/acl.h | 2 +- fs/ext3/acl.c | 2 +- fs/ext3/acl.h | 2 +- fs/ext4/acl.c | 2 +- fs/ext4/acl.h | 2 +- fs/fuse/dir.c | 6 +++--- fs/gfs2/ops_inode.c | 12 +++--------- fs/hfs/inode.c | 3 +-- fs/hfsplus/inode.c | 2 +- fs/hostfs/hostfs_kern.c | 2 +- fs/jffs2/acl.c | 2 +- fs/jffs2/acl.h | 2 +- fs/jfs/acl.c | 2 +- fs/jfs/jfs_acl.h | 2 +- fs/namei.c | 23 +++++++++++++++++------ fs/nfs/dir.c | 11 +++++------ fs/ocfs2/file.c | 2 +- fs/ocfs2/file.h | 3 +-- fs/proc/base.c | 3 +-- fs/proc/proc_sysctl.c | 2 +- fs/reiserfs/xattr.c | 2 +- fs/smbfs/file.c | 4 ++-- fs/xfs/linux-2.6/xfs_iops.c | 3 +-- include/linux/coda_linux.h | 2 +- include/linux/fs.h | 5 ++++- include/linux/nfs_fs.h | 2 +- include/linux/reiserfs_xattr.h | 2 +- include/linux/shmem_fs.h | 2 +- kernel/sysctl.c | 10 +++++----- mm/shmem_acl.c | 2 +- 38 files changed, 74 insertions(+), 87 deletions(-) (limited to 'fs') diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 7102824ba847..3cb6920ff30b 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -469,8 +469,6 @@ extern bool afs_cm_incoming_call(struct afs_call *); extern const struct inode_operations afs_dir_inode_operations; extern const struct file_operations afs_dir_file_operations; -extern int afs_permission(struct inode *, int, struct nameidata *); - /* * file.c */ @@ -605,7 +603,7 @@ extern void afs_clear_permits(struct afs_vnode *); extern void afs_cache_permit(struct afs_vnode *, struct key *, long); extern void afs_zap_permits(struct rcu_head *); extern struct key *afs_request_key(struct afs_cell *); -extern int afs_permission(struct inode *, int, struct nameidata *); +extern int afs_permission(struct inode *, int); /* * server.c diff --git a/fs/afs/security.c b/fs/afs/security.c index 3bcbeceba1bb..3ef504370034 100644 --- a/fs/afs/security.c +++ b/fs/afs/security.c @@ -284,7 +284,7 @@ static int afs_check_permit(struct afs_vnode *vnode, struct key *key, * - AFS ACLs are attached to directories only, and a file is controlled by its * parent directory's ACL */ -int afs_permission(struct inode *inode, int mask, struct nameidata *nd) +int afs_permission(struct inode *inode, int mask) { struct afs_vnode *vnode = AFS_FS_I(inode); afs_access_t uninitialized_var(access); diff --git a/fs/bad_inode.c b/fs/bad_inode.c index f1c2ea8342f5..5f1538c03b1b 100644 --- a/fs/bad_inode.c +++ b/fs/bad_inode.c @@ -243,8 +243,7 @@ static int bad_inode_readlink(struct dentry *dentry, char __user *buffer, return -EIO; } -static int bad_inode_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int bad_inode_permission(struct inode *inode, int mask) { return -EIO; } diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index fe5f6809cba6..1ec7076f7b24 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -267,7 +267,7 @@ cifs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static int cifs_permission(struct inode *inode, int mask, struct nameidata *nd) +static int cifs_permission(struct inode *inode, int mask) { struct cifs_sb_info *cifs_sb; diff --git a/fs/coda/dir.c b/fs/coda/dir.c index 3d2580e00a3e..c5916228243c 100644 --- a/fs/coda/dir.c +++ b/fs/coda/dir.c @@ -137,9 +137,11 @@ exit: } -int coda_permission(struct inode *inode, int mask, struct nameidata *nd) +int coda_permission(struct inode *inode, int mask) { int error = 0; + + mask &= MAY_READ | MAY_WRITE | MAY_EXEC; if (!mask) return 0; diff --git a/fs/coda/pioctl.c b/fs/coda/pioctl.c index c21a1f552a63..c38a98974fb0 100644 --- a/fs/coda/pioctl.c +++ b/fs/coda/pioctl.c @@ -24,8 +24,7 @@ #include /* pioctl ops */ -static int coda_ioctl_permission(struct inode *inode, int mask, - struct nameidata *nd); +static int coda_ioctl_permission(struct inode *inode, int mask); static int coda_pioctl(struct inode * inode, struct file * filp, unsigned int cmd, unsigned long user_data); @@ -42,8 +41,7 @@ const struct file_operations coda_ioctl_operations = { }; /* the coda pioctl inode ops */ -static int coda_ioctl_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int coda_ioctl_permission(struct inode *inode, int mask) { return 0; } diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index d755455e3bff..32f4228efcd5 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -830,22 +830,9 @@ out: } static int -ecryptfs_permission(struct inode *inode, int mask, struct nameidata *nd) +ecryptfs_permission(struct inode *inode, int mask) { - int rc; - - if (nd) { - struct vfsmount *vfsmnt_save = nd->path.mnt; - struct dentry *dentry_save = nd->path.dentry; - - nd->path.mnt = ecryptfs_dentry_to_lower_mnt(nd->path.dentry); - nd->path.dentry = ecryptfs_dentry_to_lower(nd->path.dentry); - rc = permission(ecryptfs_inode_to_lower(inode), mask, nd); - nd->path.mnt = vfsmnt_save; - nd->path.dentry = dentry_save; - } else - rc = permission(ecryptfs_inode_to_lower(inode), mask, NULL); - return rc; + return permission(ecryptfs_inode_to_lower(inode), mask, NULL); } /** diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c index e58669e1b87c..ae8c4f850b27 100644 --- a/fs/ext2/acl.c +++ b/fs/ext2/acl.c @@ -294,7 +294,7 @@ ext2_check_acl(struct inode *inode, int mask) } int -ext2_permission(struct inode *inode, int mask, struct nameidata *nd) +ext2_permission(struct inode *inode, int mask) { return generic_permission(inode, mask, ext2_check_acl); } diff --git a/fs/ext2/acl.h b/fs/ext2/acl.h index 0bde85bafe38..b42cf578554b 100644 --- a/fs/ext2/acl.h +++ b/fs/ext2/acl.h @@ -58,7 +58,7 @@ static inline int ext2_acl_count(size_t size) #define EXT2_ACL_NOT_CACHED ((void *)-1) /* acl.c */ -extern int ext2_permission (struct inode *, int, struct nameidata *); +extern int ext2_permission (struct inode *, int); extern int ext2_acl_chmod (struct inode *); extern int ext2_init_acl (struct inode *, struct inode *); diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c index a754d1848173..b60bb241880c 100644 --- a/fs/ext3/acl.c +++ b/fs/ext3/acl.c @@ -299,7 +299,7 @@ ext3_check_acl(struct inode *inode, int mask) } int -ext3_permission(struct inode *inode, int mask, struct nameidata *nd) +ext3_permission(struct inode *inode, int mask) { return generic_permission(inode, mask, ext3_check_acl); } diff --git a/fs/ext3/acl.h b/fs/ext3/acl.h index 0d1e6279cbfd..42da16b8cac0 100644 --- a/fs/ext3/acl.h +++ b/fs/ext3/acl.h @@ -58,7 +58,7 @@ static inline int ext3_acl_count(size_t size) #define EXT3_ACL_NOT_CACHED ((void *)-1) /* acl.c */ -extern int ext3_permission (struct inode *, int, struct nameidata *); +extern int ext3_permission (struct inode *, int); extern int ext3_acl_chmod (struct inode *); extern int ext3_init_acl (handle_t *, struct inode *, struct inode *); diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c index 3c8dab880d91..c7d04e165446 100644 --- a/fs/ext4/acl.c +++ b/fs/ext4/acl.c @@ -299,7 +299,7 @@ ext4_check_acl(struct inode *inode, int mask) } int -ext4_permission(struct inode *inode, int mask, struct nameidata *nd) +ext4_permission(struct inode *inode, int mask) { return generic_permission(inode, mask, ext4_check_acl); } diff --git a/fs/ext4/acl.h b/fs/ext4/acl.h index 26a5c1abf147..cd2b855a07d6 100644 --- a/fs/ext4/acl.h +++ b/fs/ext4/acl.h @@ -58,7 +58,7 @@ static inline int ext4_acl_count(size_t size) #define EXT4_ACL_NOT_CACHED ((void *)-1) /* acl.c */ -extern int ext4_permission (struct inode *, int, struct nameidata *); +extern int ext4_permission (struct inode *, int); extern int ext4_acl_chmod (struct inode *); extern int ext4_init_acl (handle_t *, struct inode *, struct inode *); diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 51d0035ff07e..48a7934cb950 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -898,7 +898,7 @@ static int fuse_access(struct inode *inode, int mask) return PTR_ERR(req); memset(&inarg, 0, sizeof(inarg)); - inarg.mask = mask; + inarg.mask = mask & (MAY_READ | MAY_WRITE | MAY_EXEC); req->in.h.opcode = FUSE_ACCESS; req->in.h.nodeid = get_node_id(inode); req->in.numargs = 1; @@ -927,7 +927,7 @@ static int fuse_access(struct inode *inode, int mask) * access request is sent. Execute permission is still checked * locally based on file mode. */ -static int fuse_permission(struct inode *inode, int mask, struct nameidata *nd) +static int fuse_permission(struct inode *inode, int mask) { struct fuse_conn *fc = get_fuse_conn(inode); bool refreshed = false; @@ -962,7 +962,7 @@ static int fuse_permission(struct inode *inode, int mask, struct nameidata *nd) exist. So if permissions are revoked this won't be noticed immediately, only after the attribute timeout has expired */ - } else if (nd && (nd->flags & (LOOKUP_ACCESS | LOOKUP_CHDIR))) { + } else if (mask & (MAY_ACCESS | MAY_CHDIR)) { err = fuse_access(inode, mask); } else if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode)) { if (!(inode->i_mode & S_IXUGO)) { diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index 1e252dfc5294..4e982532f085 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -915,12 +915,6 @@ int gfs2_permission(struct inode *inode, int mask) return error; } -static int gfs2_iop_permission(struct inode *inode, int mask, - struct nameidata *nd) -{ - return gfs2_permission(inode, mask); -} - static int setattr_size(struct inode *inode, struct iattr *attr) { struct gfs2_inode *ip = GFS2_I(inode); @@ -1150,7 +1144,7 @@ static int gfs2_removexattr(struct dentry *dentry, const char *name) } const struct inode_operations gfs2_file_iops = { - .permission = gfs2_iop_permission, + .permission = gfs2_permission, .setattr = gfs2_setattr, .getattr = gfs2_getattr, .setxattr = gfs2_setxattr, @@ -1169,7 +1163,7 @@ const struct inode_operations gfs2_dir_iops = { .rmdir = gfs2_rmdir, .mknod = gfs2_mknod, .rename = gfs2_rename, - .permission = gfs2_iop_permission, + .permission = gfs2_permission, .setattr = gfs2_setattr, .getattr = gfs2_getattr, .setxattr = gfs2_setxattr, @@ -1181,7 +1175,7 @@ const struct inode_operations gfs2_dir_iops = { const struct inode_operations gfs2_symlink_iops = { .readlink = gfs2_readlink, .follow_link = gfs2_follow_link, - .permission = gfs2_iop_permission, + .permission = gfs2_permission, .setattr = gfs2_setattr, .getattr = gfs2_getattr, .setxattr = gfs2_setxattr, diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index dc4ec640e875..aa73f3fd5dd9 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -511,8 +511,7 @@ void hfs_clear_inode(struct inode *inode) } } -static int hfs_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int hfs_permission(struct inode *inode, int mask) { if (S_ISREG(inode->i_mode) && mask & MAY_EXEC) return 0; diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index cc3b5e24339b..d4014e3044d2 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -238,7 +238,7 @@ static void hfsplus_set_perms(struct inode *inode, struct hfsplus_perm *perms) perms->dev = cpu_to_be32(HFSPLUS_I(inode).dev); } -static int hfsplus_permission(struct inode *inode, int mask, struct nameidata *nd) +static int hfsplus_permission(struct inode *inode, int mask) { /* MAY_EXEC is also used for lookup, if no x bit is set allow lookup, * open_exec has the same test, so it's still not executable, if a x bit diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index 5222345ddccf..d6ecabf4d231 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -822,7 +822,7 @@ int hostfs_rename(struct inode *from_ino, struct dentry *from, return err; } -int hostfs_permission(struct inode *ino, int desired, struct nameidata *nd) +int hostfs_permission(struct inode *ino, int desired) { char *name; int r = 0, w = 0, x = 0, err; diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c index 4c80404a9aba..d98713777a1b 100644 --- a/fs/jffs2/acl.c +++ b/fs/jffs2/acl.c @@ -314,7 +314,7 @@ static int jffs2_check_acl(struct inode *inode, int mask) return -EAGAIN; } -int jffs2_permission(struct inode *inode, int mask, struct nameidata *nd) +int jffs2_permission(struct inode *inode, int mask) { return generic_permission(inode, mask, jffs2_check_acl); } diff --git a/fs/jffs2/acl.h b/fs/jffs2/acl.h index 0bb7f003fd80..8ca058aed384 100644 --- a/fs/jffs2/acl.h +++ b/fs/jffs2/acl.h @@ -28,7 +28,7 @@ struct jffs2_acl_header { #define JFFS2_ACL_NOT_CACHED ((void *)-1) -extern int jffs2_permission(struct inode *, int, struct nameidata *); +extern int jffs2_permission(struct inode *, int); extern int jffs2_acl_chmod(struct inode *); extern int jffs2_init_acl_pre(struct inode *, struct inode *, int *); extern int jffs2_init_acl_post(struct inode *); diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c index 4d84bdc88299..d3e5c33665de 100644 --- a/fs/jfs/acl.c +++ b/fs/jfs/acl.c @@ -140,7 +140,7 @@ static int jfs_check_acl(struct inode *inode, int mask) return -EAGAIN; } -int jfs_permission(struct inode *inode, int mask, struct nameidata *nd) +int jfs_permission(struct inode *inode, int mask) { return generic_permission(inode, mask, jfs_check_acl); } diff --git a/fs/jfs/jfs_acl.h b/fs/jfs/jfs_acl.h index 455fa4292045..88475f10a389 100644 --- a/fs/jfs/jfs_acl.h +++ b/fs/jfs/jfs_acl.h @@ -20,7 +20,7 @@ #ifdef CONFIG_JFS_POSIX_ACL -int jfs_permission(struct inode *, int, struct nameidata *); +int jfs_permission(struct inode *, int); int jfs_init_acl(tid_t, struct inode *, struct inode *); int jfs_setattr(struct dentry *, struct iattr *); diff --git a/fs/namei.c b/fs/namei.c index 3b26a240ade9..46af98ed136b 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -185,6 +185,8 @@ int generic_permission(struct inode *inode, int mask, { umode_t mode = inode->i_mode; + mask &= MAY_READ | MAY_WRITE | MAY_EXEC; + if (current->fsuid == inode->i_uid) mode >>= 6; else { @@ -203,7 +205,7 @@ int generic_permission(struct inode *inode, int mask, /* * If the DACs are ok we don't need any capability check. */ - if (((mode & mask & (MAY_READ|MAY_WRITE|MAY_EXEC)) == mask)) + if ((mask & ~mode) == 0) return 0; check_capabilities: @@ -228,7 +230,7 @@ int generic_permission(struct inode *inode, int mask, int permission(struct inode *inode, int mask, struct nameidata *nd) { - int retval, submask; + int retval; struct vfsmount *mnt = NULL; if (nd) @@ -261,9 +263,17 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) } /* Ordinary permission routines do not understand MAY_APPEND. */ - submask = mask & ~MAY_APPEND; if (inode->i_op && inode->i_op->permission) { - retval = inode->i_op->permission(inode, submask, nd); + int extra = 0; + if (nd) { + if (nd->flags & LOOKUP_ACCESS) + extra |= MAY_ACCESS; + if (nd->flags & LOOKUP_CHDIR) + extra |= MAY_CHDIR; + if (nd->flags & LOOKUP_OPEN) + extra |= MAY_OPEN; + } + retval = inode->i_op->permission(inode, mask | extra); if (!retval) { /* * Exec permission on a regular file is denied if none @@ -277,7 +287,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) return -EACCES; } } else { - retval = generic_permission(inode, submask, NULL); + retval = generic_permission(inode, mask, NULL); } if (retval) return retval; @@ -286,7 +296,8 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) if (retval) return retval; - return security_inode_permission(inode, mask, nd); + return security_inode_permission(inode, + mask & (MAY_READ|MAY_WRITE|MAY_EXEC), nd); } /** diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 28a238dab23a..74f92b717f78 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1884,7 +1884,7 @@ static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask) return status; nfs_access_add_cache(inode, &cache); out: - if ((cache.mask & mask) == mask) + if ((mask & ~cache.mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) == 0) return 0; return -EACCES; } @@ -1907,17 +1907,17 @@ int nfs_may_open(struct inode *inode, struct rpc_cred *cred, int openflags) return nfs_do_access(inode, cred, nfs_open_permission_mask(openflags)); } -int nfs_permission(struct inode *inode, int mask, struct nameidata *nd) +int nfs_permission(struct inode *inode, int mask) { struct rpc_cred *cred; int res = 0; nfs_inc_stats(inode, NFSIOS_VFSACCESS); - if (mask == 0) + if ((mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) == 0) goto out; /* Is this sys_access() ? */ - if (nd != NULL && (nd->flags & LOOKUP_ACCESS)) + if (mask & MAY_ACCESS) goto force_lookup; switch (inode->i_mode & S_IFMT) { @@ -1926,8 +1926,7 @@ int nfs_permission(struct inode *inode, int mask, struct nameidata *nd) case S_IFREG: /* NFSv4 has atomic_open... */ if (nfs_server_capable(inode, NFS_CAP_ATOMIC_OPEN) - && nd != NULL - && (nd->flags & LOOKUP_OPEN)) + && (mask & MAY_OPEN)) goto out; break; case S_IFDIR: diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index e8514e8b6ce8..be2dd95d3a1d 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1176,7 +1176,7 @@ bail: return err; } -int ocfs2_permission(struct inode *inode, int mask, struct nameidata *nd) +int ocfs2_permission(struct inode *inode, int mask) { int ret; diff --git a/fs/ocfs2/file.h b/fs/ocfs2/file.h index 048ddcaf5c80..1e27b4d017ea 100644 --- a/fs/ocfs2/file.h +++ b/fs/ocfs2/file.h @@ -62,8 +62,7 @@ int ocfs2_lock_allocators(struct inode *inode, struct ocfs2_dinode *di, int ocfs2_setattr(struct dentry *dentry, struct iattr *attr); int ocfs2_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); -int ocfs2_permission(struct inode *inode, int mask, - struct nameidata *nd); +int ocfs2_permission(struct inode *inode, int mask); int ocfs2_should_update_atime(struct inode *inode, struct vfsmount *vfsmnt); diff --git a/fs/proc/base.c b/fs/proc/base.c index 81bce6791bfc..d82d800389f6 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1859,8 +1859,7 @@ static const struct file_operations proc_fd_operations = { * /proc/pid/fd needs a special permission handler so that a process can still * access /proc/self/fd after it has executed a setuid(). */ -static int proc_fd_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int proc_fd_permission(struct inode *inode, int mask) { int rv; diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index fa1ec2433e44..f9a8b892718f 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -292,7 +292,7 @@ out: return ret; } -static int proc_sys_permission(struct inode *inode, int mask, struct nameidata *nd) +static int proc_sys_permission(struct inode *inode, int mask) { /* * sysctl entries that are not writeable, diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index d7c4935c1034..bb3cb5b7cdb2 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -1250,7 +1250,7 @@ static int reiserfs_check_acl(struct inode *inode, int mask) return error; } -int reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd) +int reiserfs_permission(struct inode *inode, int mask) { /* * We don't do permission checks on the internal objects. diff --git a/fs/smbfs/file.c b/fs/smbfs/file.c index 2294783320cb..e4f8d51a5553 100644 --- a/fs/smbfs/file.c +++ b/fs/smbfs/file.c @@ -408,7 +408,7 @@ smb_file_release(struct inode *inode, struct file * file) * privileges, so we need our own check for this. */ static int -smb_file_permission(struct inode *inode, int mask, struct nameidata *nd) +smb_file_permission(struct inode *inode, int mask) { int mode = inode->i_mode; int error = 0; @@ -417,7 +417,7 @@ smb_file_permission(struct inode *inode, int mask, struct nameidata *nd) /* Look at user permissions */ mode >>= 6; - if ((mode & 7 & mask) != mask) + if (mask & ~mode & (MAY_READ | MAY_WRITE | MAY_EXEC)) error = -EACCES; return error; } diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index 2bf287ef5489..5fc61c824bb9 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -589,8 +589,7 @@ xfs_check_acl( STATIC int xfs_vn_permission( struct inode *inode, - int mask, - struct nameidata *nd) + int mask) { return generic_permission(inode, mask, xfs_check_acl); } diff --git a/include/linux/coda_linux.h b/include/linux/coda_linux.h index 31b75311e2ca..dcc228aa335a 100644 --- a/include/linux/coda_linux.h +++ b/include/linux/coda_linux.h @@ -37,7 +37,7 @@ extern const struct file_operations coda_ioctl_operations; /* operations shared over more than one file */ int coda_open(struct inode *i, struct file *f); int coda_release(struct inode *i, struct file *f); -int coda_permission(struct inode *inode, int mask, struct nameidata *nd); +int coda_permission(struct inode *inode, int mask); int coda_revalidate_inode(struct dentry *); int coda_getattr(struct vfsmount *, struct dentry *, struct kstat *); int coda_setattr(struct dentry *, struct iattr *); diff --git a/include/linux/fs.h b/include/linux/fs.h index 7721a2ac9c0e..6c923c9b79bc 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -60,6 +60,9 @@ extern int dir_notify_enable; #define MAY_WRITE 2 #define MAY_READ 4 #define MAY_APPEND 8 +#define MAY_ACCESS 16 +#define MAY_CHDIR 32 +#define MAY_OPEN 64 #define FMODE_READ 1 #define FMODE_WRITE 2 @@ -1272,7 +1275,7 @@ struct inode_operations { void * (*follow_link) (struct dentry *, struct nameidata *); void (*put_link) (struct dentry *, struct nameidata *, void *); void (*truncate) (struct inode *); - int (*permission) (struct inode *, int, struct nameidata *); + int (*permission) (struct inode *, int); int (*setattr) (struct dentry *, struct iattr *); int (*getattr) (struct vfsmount *mnt, struct dentry *, struct kstat *); int (*setxattr) (struct dentry *, const char *,const void *,size_t,int); diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 29d261918734..f08f9ca602af 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -332,7 +332,7 @@ extern int nfs_refresh_inode(struct inode *, struct nfs_fattr *); extern int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr); extern int nfs_post_op_update_inode_force_wcc(struct inode *inode, struct nfs_fattr *fattr); extern int nfs_getattr(struct vfsmount *, struct dentry *, struct kstat *); -extern int nfs_permission(struct inode *, int, struct nameidata *); +extern int nfs_permission(struct inode *, int); extern int nfs_open(struct inode *, struct file *); extern int nfs_release(struct inode *, struct file *); extern int nfs_attribute_timeout(struct inode *inode); diff --git a/include/linux/reiserfs_xattr.h b/include/linux/reiserfs_xattr.h index 66a96814d614..af135ae895db 100644 --- a/include/linux/reiserfs_xattr.h +++ b/include/linux/reiserfs_xattr.h @@ -55,7 +55,7 @@ int reiserfs_removexattr(struct dentry *dentry, const char *name); int reiserfs_delete_xattrs(struct inode *inode); int reiserfs_chown_xattrs(struct inode *inode, struct iattr *attrs); int reiserfs_xattr_init(struct super_block *sb, int mount_flags); -int reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd); +int reiserfs_permission(struct inode *inode, int mask); int reiserfs_xattr_del(struct inode *, const char *); int reiserfs_xattr_get(const struct inode *, const char *, void *, size_t); diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h index f2d12d5a21b8..fd83f2584b15 100644 --- a/include/linux/shmem_fs.h +++ b/include/linux/shmem_fs.h @@ -43,7 +43,7 @@ static inline struct shmem_inode_info *SHMEM_I(struct inode *inode) } #ifdef CONFIG_TMPFS_POSIX_ACL -int shmem_permission(struct inode *, int, struct nameidata *); +int shmem_permission(struct inode *, int); int shmem_acl_init(struct inode *, struct inode *); void shmem_acl_destroy_inode(struct inode *); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index ff5abcca5ddf..911d846f0503 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1516,9 +1516,9 @@ static int do_sysctl_strategy(struct ctl_table_root *root, int op = 0, rc; if (oldval) - op |= 004; + op |= MAY_READ; if (newval) - op |= 002; + op |= MAY_WRITE; if (sysctl_perm(root, table, op)) return -EPERM; @@ -1560,7 +1560,7 @@ repeat: if (n == table->ctl_name) { int error; if (table->child) { - if (sysctl_perm(root, table, 001)) + if (sysctl_perm(root, table, MAY_EXEC)) return -EPERM; name++; nlen--; @@ -1635,7 +1635,7 @@ static int test_perm(int mode, int op) mode >>= 6; else if (in_egroup_p(0)) mode >>= 3; - if ((mode & op & 0007) == op) + if ((op & ~mode & (MAY_READ|MAY_WRITE|MAY_EXEC)) == 0) return 0; return -EACCES; } @@ -1645,7 +1645,7 @@ int sysctl_perm(struct ctl_table_root *root, struct ctl_table *table, int op) int error; int mode; - error = security_sysctl(table, op); + error = security_sysctl(table, op & (MAY_READ | MAY_WRITE | MAY_EXEC)); if (error) return error; diff --git a/mm/shmem_acl.c b/mm/shmem_acl.c index f5664c5b9eb1..8e5aadd7dcd6 100644 --- a/mm/shmem_acl.c +++ b/mm/shmem_acl.c @@ -191,7 +191,7 @@ shmem_check_acl(struct inode *inode, int mask) * shmem_permission - permission() inode operation */ int -shmem_permission(struct inode *inode, int mask, struct nameidata *nd) +shmem_permission(struct inode *inode, int mask) { return generic_permission(inode, mask, shmem_check_acl); } -- cgit v1.2.3 From c82e42da8a6b2f3a85dc4d4278cb8238702f8f64 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 24 Jun 2008 16:50:12 +0200 Subject: [patch 1/5] vfs: truncate: dont check immutable twice vfs_permission(MAY_WRITE) already checked for the inode being immutable, so no need to repeat it. Signed-off-by: Miklos Szeredi Acked-by: Christoph Hellwig --- fs/open.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/open.c b/fs/open.c index bb98d2fe809f..b2e4c93aed03 100644 --- a/fs/open.c +++ b/fs/open.c @@ -256,7 +256,7 @@ static long do_sys_truncate(const char __user * path, loff_t length) goto mnt_drop_write_and_out; error = -EPERM; - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) + if (IS_APPEND(inode)) goto mnt_drop_write_and_out; error = get_write_access(inode); -- cgit v1.2.3 From 2f1936b87783a3a56c9441b27b9ba7a747f11e8e Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 24 Jun 2008 16:50:14 +0200 Subject: [patch 3/5] vfs: change remove_suid() to file_remove_suid() All calls to remove_suid() are made with a file pointer, because (similarly to file_update_time) it is called when the file is written. Clean up callers by passing in a file instead of a dentry. Signed-off-by: Miklos Szeredi --- fs/fuse/file.c | 2 +- fs/ntfs/file.c | 2 +- fs/splice.c | 4 ++-- fs/xfs/linux-2.6/xfs_lrw.c | 2 +- include/linux/fs.h | 2 +- mm/filemap.c | 7 ++++--- mm/filemap_xip.c | 2 +- 7 files changed, 11 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 67ff2c6a8f63..2bada6bbc317 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -893,7 +893,7 @@ static ssize_t fuse_file_aio_write(struct kiocb *iocb, const struct iovec *iov, if (count == 0) goto out; - err = remove_suid(file->f_path.dentry); + err = file_remove_suid(file); if (err) goto out; diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c index 3c5550cd11d6..d020866d4232 100644 --- a/fs/ntfs/file.c +++ b/fs/ntfs/file.c @@ -2118,7 +2118,7 @@ static ssize_t ntfs_file_aio_write_nolock(struct kiocb *iocb, goto out; if (!count) goto out; - err = remove_suid(file->f_path.dentry); + err = file_remove_suid(file); if (err) goto out; file_update_time(file); diff --git a/fs/splice.c b/fs/splice.c index 47dc1a445d1f..b30311ba8af6 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -772,7 +772,7 @@ generic_file_splice_write_nolock(struct pipe_inode_info *pipe, struct file *out, ssize_t ret; int err; - err = remove_suid(out->f_path.dentry); + err = file_remove_suid(out); if (unlikely(err)) return err; @@ -830,7 +830,7 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, ssize_t ret; inode_double_lock(inode, pipe->inode); - ret = remove_suid(out->f_path.dentry); + ret = file_remove_suid(out); if (likely(!ret)) ret = __splice_from_pipe(pipe, &sd, pipe_to_file); inode_double_unlock(inode, pipe->inode); diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c index 5e3b57516ec7..82333b3e118e 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.c +++ b/fs/xfs/linux-2.6/xfs_lrw.c @@ -711,7 +711,7 @@ start: !capable(CAP_FSETID)) { error = xfs_write_clear_setuid(xip); if (likely(!error)) - error = -remove_suid(file->f_path.dentry); + error = -file_remove_suid(file); if (unlikely(error)) { goto out_unlock_internal; } diff --git a/include/linux/fs.h b/include/linux/fs.h index 6c923c9b79bc..1a3546e69f9e 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1834,7 +1834,7 @@ extern void clear_inode(struct inode *); extern void destroy_inode(struct inode *); extern struct inode *new_inode(struct super_block *); extern int should_remove_suid(struct dentry *); -extern int remove_suid(struct dentry *); +extern int file_remove_suid(struct file *); extern void __insert_inode_hash(struct inode *, unsigned long hashval); extern void remove_inode_hash(struct inode *); diff --git a/mm/filemap.c b/mm/filemap.c index 2ed8b0389c51..5de7633e1dbe 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1758,8 +1758,9 @@ static int __remove_suid(struct dentry *dentry, int kill) return notify_change(dentry, &newattrs); } -int remove_suid(struct dentry *dentry) +int file_remove_suid(struct file *file) { + struct dentry *dentry = file->f_path.dentry; int killsuid = should_remove_suid(dentry); int killpriv = security_inode_need_killpriv(dentry); int error = 0; @@ -1773,7 +1774,7 @@ int remove_suid(struct dentry *dentry) return error; } -EXPORT_SYMBOL(remove_suid); +EXPORT_SYMBOL(file_remove_suid); static size_t __iovec_copy_from_user_inatomic(char *vaddr, const struct iovec *iov, size_t base, size_t bytes) @@ -2529,7 +2530,7 @@ __generic_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov, if (count == 0) goto out; - err = remove_suid(file->f_path.dentry); + err = file_remove_suid(file); if (err) goto out; diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c index 3e744abcce9d..98a3f31ccd6a 100644 --- a/mm/filemap_xip.c +++ b/mm/filemap_xip.c @@ -380,7 +380,7 @@ xip_file_write(struct file *filp, const char __user *buf, size_t len, if (count == 0) goto out_backing; - ret = remove_suid(filp->f_path.dentry); + ret = file_remove_suid(filp); if (ret) goto out_backing; -- cgit v1.2.3 From 7e79eedb3b22200cc8b774baea3a7bf28d766101 Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Tue, 24 Jun 2008 16:50:15 +0200 Subject: [patch 4/5] vfs: reuse local variable in vfs_link() Why not reuse "inode" which is assigned as struct inode *inode = old_dentry->d_inode; in the beginning of vfs_link() ? Signed-off-by: Tetsuo Handa Signed-off-by: Miklos Szeredi --- fs/namei.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index 46af98ed136b..3b67be7631dc 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -2524,19 +2524,19 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de return -EPERM; if (!dir->i_op || !dir->i_op->link) return -EPERM; - if (S_ISDIR(old_dentry->d_inode->i_mode)) + if (S_ISDIR(inode->i_mode)) return -EPERM; error = security_inode_link(old_dentry, dir, new_dentry); if (error) return error; - mutex_lock(&old_dentry->d_inode->i_mutex); + mutex_lock(&inode->i_mutex); DQUOT_INIT(dir); error = dir->i_op->link(old_dentry, dir, new_dentry); - mutex_unlock(&old_dentry->d_inode->i_mutex); + mutex_unlock(&inode->i_mutex); if (!error) - fsnotify_link(dir, old_dentry->d_inode, new_dentry); + fsnotify_link(dir, inode, new_dentry); return error; } -- cgit v1.2.3 From db2e747b14991a4c6a5c98b0e5f552a193237c03 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 24 Jun 2008 16:50:16 +0200 Subject: [patch 5/5] vfs: remove mode parameter from vfs_symlink() Remove the unused mode parameter from vfs_symlink and callers. Thanks to Tetsuo Handa for noticing. CC: Tetsuo Handa Signed-off-by: Miklos Szeredi --- fs/ecryptfs/inode.c | 4 +--- fs/namei.c | 4 ++-- fs/nfsd/vfs.c | 10 ++-------- include/linux/fs.h | 2 +- 4 files changed, 6 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 32f4228efcd5..f25caf2b0887 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -465,7 +465,6 @@ static int ecryptfs_symlink(struct inode *dir, struct dentry *dentry, int rc; struct dentry *lower_dentry; struct dentry *lower_dir_dentry; - umode_t mode; char *encoded_symname; int encoded_symlen; struct ecryptfs_crypt_stat *crypt_stat = NULL; @@ -473,7 +472,6 @@ static int ecryptfs_symlink(struct inode *dir, struct dentry *dentry, lower_dentry = ecryptfs_dentry_to_lower(dentry); dget(lower_dentry); lower_dir_dentry = lock_parent(lower_dentry); - mode = S_IALLUGO; encoded_symlen = ecryptfs_encode_filename(crypt_stat, symname, strlen(symname), &encoded_symname); @@ -482,7 +480,7 @@ static int ecryptfs_symlink(struct inode *dir, struct dentry *dentry, goto out_lock; } rc = vfs_symlink(lower_dir_dentry->d_inode, lower_dentry, - encoded_symname, mode); + encoded_symname); kfree(encoded_symname); if (rc || !lower_dentry->d_inode) goto out_lock; diff --git a/fs/namei.c b/fs/namei.c index 3b67be7631dc..ae0e56fdb742 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -2434,7 +2434,7 @@ asmlinkage long sys_unlink(const char __user *pathname) return do_unlinkat(AT_FDCWD, pathname); } -int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname, int mode) +int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname) { int error = may_create(dir, dentry, NULL); @@ -2483,7 +2483,7 @@ asmlinkage long sys_symlinkat(const char __user *oldname, error = mnt_want_write(nd.path.mnt); if (error) goto out_dput; - error = vfs_symlink(nd.path.dentry->d_inode, dentry, from, S_IALLUGO); + error = vfs_symlink(nd.path.dentry->d_inode, dentry, from); mnt_drop_write(nd.path.mnt); out_dput: dput(dentry); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 0f4481e0502d..ad1ad59e3742 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1516,7 +1516,6 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, struct dentry *dentry, *dnew; __be32 err, cerr; int host_err; - umode_t mode; err = nfserr_noent; if (!flen || !plen) @@ -1535,11 +1534,6 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, if (IS_ERR(dnew)) goto out_nfserr; - mode = S_IALLUGO; - /* Only the MODE ATTRibute is even vaguely meaningful */ - if (iap && (iap->ia_valid & ATTR_MODE)) - mode = iap->ia_mode & S_IALLUGO; - host_err = mnt_want_write(fhp->fh_export->ex_path.mnt); if (host_err) goto out_nfserr; @@ -1551,11 +1545,11 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, else { strncpy(path_alloced, path, plen); path_alloced[plen] = 0; - host_err = vfs_symlink(dentry->d_inode, dnew, path_alloced, mode); + host_err = vfs_symlink(dentry->d_inode, dnew, path_alloced); kfree(path_alloced); } } else - host_err = vfs_symlink(dentry->d_inode, dnew, path, mode); + host_err = vfs_symlink(dentry->d_inode, dnew, path); if (!host_err) { if (EX_ISSYNC(fhp->fh_export)) diff --git a/include/linux/fs.h b/include/linux/fs.h index 1a3546e69f9e..25998e803fc2 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1139,7 +1139,7 @@ extern int vfs_permission(struct nameidata *, int); extern int vfs_create(struct inode *, struct dentry *, int, struct nameidata *); extern int vfs_mkdir(struct inode *, struct dentry *, int); extern int vfs_mknod(struct inode *, struct dentry *, int, dev_t); -extern int vfs_symlink(struct inode *, struct dentry *, const char *, int); +extern int vfs_symlink(struct inode *, struct dentry *, const char *); extern int vfs_link(struct dentry *, struct inode *, struct dentry *); extern int vfs_rmdir(struct inode *, struct dentry *); extern int vfs_unlink(struct inode *, struct dentry *); -- cgit v1.2.3 From 8bb79224b87aab92071e94d46e70bd160d89bf34 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 16 Jul 2008 09:51:03 -0400 Subject: [PATCH] permission checks for chdir need special treatment only on the last step ... so we ought to pass MAY_CHDIR to vfs_permission() instead of having it triggered on every step of preceding pathname resolution. LOOKUP_CHDIR is killed by that. Signed-off-by: Al Viro --- fs/namei.c | 2 -- fs/open.c | 5 ++--- include/linux/namei.h | 1 - 3 files changed, 2 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index ae0e56fdb742..6c76e1ee9c45 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -268,8 +268,6 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) if (nd) { if (nd->flags & LOOKUP_ACCESS) extra |= MAY_ACCESS; - if (nd->flags & LOOKUP_CHDIR) - extra |= MAY_CHDIR; if (nd->flags & LOOKUP_OPEN) extra |= MAY_OPEN; } diff --git a/fs/open.c b/fs/open.c index b2e4c93aed03..8e02d42bfe44 100644 --- a/fs/open.c +++ b/fs/open.c @@ -501,12 +501,11 @@ asmlinkage long sys_chdir(const char __user * filename) struct nameidata nd; int error; - error = __user_walk(filename, - LOOKUP_FOLLOW|LOOKUP_DIRECTORY|LOOKUP_CHDIR, &nd); + error = __user_walk(filename, LOOKUP_FOLLOW|LOOKUP_DIRECTORY, &nd); if (error) goto out; - error = vfs_permission(&nd, MAY_EXEC); + error = vfs_permission(&nd, MAY_EXEC | MAY_CHDIR); if (error) goto dput_and_out; diff --git a/include/linux/namei.h b/include/linux/namei.h index 24d88e98a626..3cf62d26d493 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -55,7 +55,6 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND}; #define LOOKUP_OPEN (0x0100) #define LOOKUP_CREATE (0x0200) #define LOOKUP_ACCESS (0x0400) -#define LOOKUP_CHDIR (0x0800) extern int __user_walk(const char __user *, unsigned, struct nameidata *); extern int __user_walk_fd(int dfd, const char __user *, unsigned, struct nameidata *); -- cgit v1.2.3 From 7f2da1e7d0330395e5e9e350b879b98a1ea495df Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 10 May 2008 20:44:54 -0400 Subject: [PATCH] kill altroot long overdue... Signed-off-by: Al Viro --- fs/namei.c | 89 +------------------------------------------ fs/namespace.c | 8 +--- fs/open.c | 3 +- include/asm-alpha/namei.h | 17 --------- include/asm-arm/namei.h | 25 ------------ include/asm-avr32/namei.h | 7 ---- include/asm-blackfin/namei.h | 19 --------- include/asm-cris/namei.h | 17 --------- include/asm-frv/namei.h | 18 --------- include/asm-h8300/namei.h | 17 --------- include/asm-ia64/namei.h | 25 ------------ include/asm-m32r/namei.h | 17 --------- include/asm-m68k/namei.h | 17 --------- include/asm-m68knommu/namei.h | 1 - include/asm-mips/namei.h | 11 ------ include/asm-mn10300/namei.h | 22 ----------- include/asm-parisc/namei.h | 17 --------- include/asm-powerpc/namei.h | 20 ---------- include/asm-s390/namei.h | 21 ---------- include/asm-sh/namei.h | 17 --------- include/asm-sparc/namei.h | 8 ---- include/asm-sparc64/namei.h | 1 - include/asm-um/namei.h | 6 --- include/asm-v850/namei.h | 17 --------- include/asm-x86/namei.h | 11 ------ include/asm-xtensa/namei.h | 26 ------------- include/linux/fs_struct.h | 3 +- include/linux/namei.h | 1 - kernel/exec_domain.c | 1 - kernel/exit.c | 2 - kernel/fork.c | 7 ---- 31 files changed, 5 insertions(+), 466 deletions(-) delete mode 100644 include/asm-alpha/namei.h delete mode 100644 include/asm-arm/namei.h delete mode 100644 include/asm-avr32/namei.h delete mode 100644 include/asm-blackfin/namei.h delete mode 100644 include/asm-cris/namei.h delete mode 100644 include/asm-frv/namei.h delete mode 100644 include/asm-h8300/namei.h delete mode 100644 include/asm-ia64/namei.h delete mode 100644 include/asm-m32r/namei.h delete mode 100644 include/asm-m68k/namei.h delete mode 100644 include/asm-m68knommu/namei.h delete mode 100644 include/asm-mips/namei.h delete mode 100644 include/asm-mn10300/namei.h delete mode 100644 include/asm-parisc/namei.h delete mode 100644 include/asm-powerpc/namei.h delete mode 100644 include/asm-s390/namei.h delete mode 100644 include/asm-sh/namei.h delete mode 100644 include/asm-sparc/namei.h delete mode 100644 include/asm-sparc64/namei.h delete mode 100644 include/asm-um/namei.h delete mode 100644 include/asm-v850/namei.h delete mode 100644 include/asm-x86/namei.h delete mode 100644 include/asm-xtensa/namei.h (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index 6c76e1ee9c45..095818089ac1 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -31,7 +31,6 @@ #include #include #include -#include #include #define ACC_MODE(x) ("\000\004\002\006"[(x)&O_ACCMODE]) @@ -562,27 +561,16 @@ out_unlock: return result; } -static int __emul_lookup_dentry(const char *, struct nameidata *); - /* SMP-safe */ -static __always_inline int +static __always_inline void walk_init_root(const char *name, struct nameidata *nd) { struct fs_struct *fs = current->fs; read_lock(&fs->lock); - if (fs->altroot.dentry && !(nd->flags & LOOKUP_NOALT)) { - nd->path = fs->altroot; - path_get(&fs->altroot); - read_unlock(&fs->lock); - if (__emul_lookup_dentry(name,nd)) - return 0; - read_lock(&fs->lock); - } nd->path = fs->root; path_get(&fs->root); read_unlock(&fs->lock); - return 1; } /* @@ -623,12 +611,9 @@ static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *l if (*link == '/') { path_put(&nd->path); - if (!walk_init_root(link, nd)) - /* weird __emul_prefix() stuff did it */ - goto out; + walk_init_root(link, nd); } res = link_path_walk(link, nd); -out: if (nd->depth || res || nd->last_type!=LAST_NORM) return res; /* @@ -1077,67 +1062,6 @@ static int path_walk(const char *name, struct nameidata *nd) return link_path_walk(name, nd); } -/* - * SMP-safe: Returns 1 and nd will have valid dentry and mnt, if - * everything is done. Returns 0 and drops input nd, if lookup failed; - */ -static int __emul_lookup_dentry(const char *name, struct nameidata *nd) -{ - if (path_walk(name, nd)) - return 0; /* something went wrong... */ - - if (!nd->path.dentry->d_inode || - S_ISDIR(nd->path.dentry->d_inode->i_mode)) { - struct path old_path = nd->path; - struct qstr last = nd->last; - int last_type = nd->last_type; - struct fs_struct *fs = current->fs; - - /* - * NAME was not found in alternate root or it's a directory. - * Try to find it in the normal root: - */ - nd->last_type = LAST_ROOT; - read_lock(&fs->lock); - nd->path = fs->root; - path_get(&fs->root); - read_unlock(&fs->lock); - if (path_walk(name, nd) == 0) { - if (nd->path.dentry->d_inode) { - path_put(&old_path); - return 1; - } - path_put(&nd->path); - } - nd->path = old_path; - nd->last = last; - nd->last_type = last_type; - } - return 1; -} - -void set_fs_altroot(void) -{ - char *emul = __emul_prefix(); - struct nameidata nd; - struct path path = {}, old_path; - int err; - struct fs_struct *fs = current->fs; - - if (!emul) - goto set_it; - err = path_lookup(emul, LOOKUP_FOLLOW|LOOKUP_DIRECTORY|LOOKUP_NOALT, &nd); - if (!err) - path = nd.path; -set_it: - write_lock(&fs->lock); - old_path = fs->altroot; - fs->altroot = path; - write_unlock(&fs->lock); - if (old_path.dentry) - path_put(&old_path); -} - /* Returns 0 and nd will be valid on success; Retuns error, otherwise. */ static int do_path_lookup(int dfd, const char *name, unsigned int flags, struct nameidata *nd) @@ -1153,14 +1077,6 @@ static int do_path_lookup(int dfd, const char *name, if (*name=='/') { read_lock(&fs->lock); - if (fs->altroot.dentry && !(nd->flags & LOOKUP_NOALT)) { - nd->path = fs->altroot; - path_get(&fs->altroot); - read_unlock(&fs->lock); - if (__emul_lookup_dentry(name,nd)) - goto out; /* found in altroot */ - read_lock(&fs->lock); - } nd->path = fs->root; path_get(&fs->root); read_unlock(&fs->lock); @@ -1194,7 +1110,6 @@ static int do_path_lookup(int dfd, const char *name, } retval = path_walk(name, nd); -out: if (unlikely(!retval && !audit_dummy_context() && nd->path.dentry && nd->path.dentry->d_inode)) audit_inode(name, nd->path.dentry); diff --git a/fs/namespace.c b/fs/namespace.c index f30b11e2240e..c4fcf48acef8 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1972,7 +1972,7 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, struct fs_struct *fs) { struct mnt_namespace *new_ns; - struct vfsmount *rootmnt = NULL, *pwdmnt = NULL, *altrootmnt = NULL; + struct vfsmount *rootmnt = NULL, *pwdmnt = NULL; struct vfsmount *p, *q; new_ns = kmalloc(sizeof(struct mnt_namespace), GFP_KERNEL); @@ -2015,10 +2015,6 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, pwdmnt = p; fs->pwd.mnt = mntget(q); } - if (p == fs->altroot.mnt) { - altrootmnt = p; - fs->altroot.mnt = mntget(q); - } } p = next_mnt(p, mnt_ns->root); q = next_mnt(q, new_ns->root); @@ -2029,8 +2025,6 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, mntput(rootmnt); if (pwdmnt) mntput(pwdmnt); - if (altrootmnt) - mntput(altrootmnt); return new_ns; } diff --git a/fs/open.c b/fs/open.c index 8e02d42bfe44..d3a2a00f52dc 100644 --- a/fs/open.c +++ b/fs/open.c @@ -548,7 +548,7 @@ asmlinkage long sys_chroot(const char __user * filename) struct nameidata nd; int error; - error = __user_walk(filename, LOOKUP_FOLLOW | LOOKUP_DIRECTORY | LOOKUP_NOALT, &nd); + error = __user_walk(filename, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &nd); if (error) goto out; @@ -561,7 +561,6 @@ asmlinkage long sys_chroot(const char __user * filename) goto dput_and_out; set_fs_root(current->fs, &nd.path); - set_fs_altroot(); error = 0; dput_and_out: path_put(&nd.path); diff --git a/include/asm-alpha/namei.h b/include/asm-alpha/namei.h deleted file mode 100644 index 5cc9bb39499d..000000000000 --- a/include/asm-alpha/namei.h +++ /dev/null @@ -1,17 +0,0 @@ -/* $Id: namei.h,v 1.1 1996/12/13 14:48:21 jj Exp $ - * linux/include/asm-alpha/namei.h - * - * Included from linux/fs/namei.c - */ - -#ifndef __ALPHA_NAMEI_H -#define __ALPHA_NAMEI_H - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif /* __ALPHA_NAMEI_H */ diff --git a/include/asm-arm/namei.h b/include/asm-arm/namei.h deleted file mode 100644 index a402d3b9d0f7..000000000000 --- a/include/asm-arm/namei.h +++ /dev/null @@ -1,25 +0,0 @@ -/* - * linux/include/asm-arm/namei.h - * - * Routines to handle famous /usr/gnemul - * Derived from the Sparc version of this file - * - * Included from linux/fs/namei.c - */ - -#ifndef __ASMARM_NAMEI_H -#define __ASMARM_NAMEI_H - -#define ARM_BSD_EMUL "usr/gnemul/bsd/" - -static inline char *__emul_prefix(void) -{ - switch (current->personality) { - case PER_BSD: - return ARM_BSD_EMUL; - default: - return NULL; - } -} - -#endif /* __ASMARM_NAMEI_H */ diff --git a/include/asm-avr32/namei.h b/include/asm-avr32/namei.h deleted file mode 100644 index f0a26de06cab..000000000000 --- a/include/asm-avr32/namei.h +++ /dev/null @@ -1,7 +0,0 @@ -#ifndef __ASM_AVR32_NAMEI_H -#define __ASM_AVR32_NAMEI_H - -/* This dummy routine may be changed to something useful */ -#define __emul_prefix() NULL - -#endif /* __ASM_AVR32_NAMEI_H */ diff --git a/include/asm-blackfin/namei.h b/include/asm-blackfin/namei.h deleted file mode 100644 index 8b89a2d65cb4..000000000000 --- a/include/asm-blackfin/namei.h +++ /dev/null @@ -1,19 +0,0 @@ -/* - * linux/include/asm/namei.h - * - * Included from linux/fs/namei.c - * - * Changes made by Lineo Inc. May 2001 - */ - -#ifndef __BFIN_NAMEI_H -#define __BFIN_NAMEI_H - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif diff --git a/include/asm-cris/namei.h b/include/asm-cris/namei.h deleted file mode 100644 index 8a3be7a6d9f6..000000000000 --- a/include/asm-cris/namei.h +++ /dev/null @@ -1,17 +0,0 @@ -/* $Id: namei.h,v 1.1 2000/07/10 16:32:31 bjornw Exp $ - * linux/include/asm-cris/namei.h - * - * Included from linux/fs/namei.c - */ - -#ifndef __CRIS_NAMEI_H -#define __CRIS_NAMEI_H - -/* used to find file-system prefixes for doing emulations - * see for example asm-sparc/namei.h - * we don't use it... - */ - -#define __emul_prefix() NULL - -#endif /* __CRIS_NAMEI_H */ diff --git a/include/asm-frv/namei.h b/include/asm-frv/namei.h deleted file mode 100644 index 4ea57171d951..000000000000 --- a/include/asm-frv/namei.h +++ /dev/null @@ -1,18 +0,0 @@ -/* - * include/asm-frv/namei.h - * - * Included from linux/fs/namei.c - */ - -#ifndef __ASM_NAMEI_H -#define __ASM_NAMEI_H - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif - diff --git a/include/asm-h8300/namei.h b/include/asm-h8300/namei.h deleted file mode 100644 index ab6f196db6e0..000000000000 --- a/include/asm-h8300/namei.h +++ /dev/null @@ -1,17 +0,0 @@ -/* - * linux/include/asm-h8300/namei.h - * - * Included from linux/fs/namei.c - */ - -#ifndef __H8300_NAMEI_H -#define __H8300_NAMEI_H - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif diff --git a/include/asm-ia64/namei.h b/include/asm-ia64/namei.h deleted file mode 100644 index 78e768079083..000000000000 --- a/include/asm-ia64/namei.h +++ /dev/null @@ -1,25 +0,0 @@ -#ifndef _ASM_IA64_NAMEI_H -#define _ASM_IA64_NAMEI_H - -/* - * Modified 1998, 1999, 2001 - * David Mosberger-Tang , Hewlett-Packard Co - */ - -#include -#include - -#define EMUL_PREFIX_LINUX_IA32 "/emul/ia32-linux/" - -static inline char * -__emul_prefix (void) -{ - switch (current->personality) { - case PER_LINUX32: - return EMUL_PREFIX_LINUX_IA32; - default: - return NULL; - } -} - -#endif /* _ASM_IA64_NAMEI_H */ diff --git a/include/asm-m32r/namei.h b/include/asm-m32r/namei.h deleted file mode 100644 index 210f8056b805..000000000000 --- a/include/asm-m32r/namei.h +++ /dev/null @@ -1,17 +0,0 @@ -#ifndef _ASM_M32R_NAMEI_H -#define _ASM_M32R_NAMEI_H - -/* - * linux/include/asm-m32r/namei.h - * - * Included from linux/fs/namei.c - */ - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif /* _ASM_M32R_NAMEI_H */ diff --git a/include/asm-m68k/namei.h b/include/asm-m68k/namei.h deleted file mode 100644 index f33f243b644a..000000000000 --- a/include/asm-m68k/namei.h +++ /dev/null @@ -1,17 +0,0 @@ -/* - * linux/include/asm-m68k/namei.h - * - * Included from linux/fs/namei.c - */ - -#ifndef __M68K_NAMEI_H -#define __M68K_NAMEI_H - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif diff --git a/include/asm-m68knommu/namei.h b/include/asm-m68knommu/namei.h deleted file mode 100644 index 31a85d27b931..000000000000 --- a/include/asm-m68knommu/namei.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-mips/namei.h b/include/asm-mips/namei.h deleted file mode 100644 index a6605a752469..000000000000 --- a/include/asm-mips/namei.h +++ /dev/null @@ -1,11 +0,0 @@ -#ifndef _ASM_NAMEI_H -#define _ASM_NAMEI_H - -/* - * This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - */ - -#define __emul_prefix() NULL - -#endif /* _ASM_NAMEI_H */ diff --git a/include/asm-mn10300/namei.h b/include/asm-mn10300/namei.h deleted file mode 100644 index bd9ce94aeb65..000000000000 --- a/include/asm-mn10300/namei.h +++ /dev/null @@ -1,22 +0,0 @@ -/* Emulation stuff - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. - */ - -#ifndef _ASM_NAMEI_H -#define _ASM_NAMEI_H - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif /* _ASM_NAMEI_H */ diff --git a/include/asm-parisc/namei.h b/include/asm-parisc/namei.h deleted file mode 100644 index 8d29b3d9fb33..000000000000 --- a/include/asm-parisc/namei.h +++ /dev/null @@ -1,17 +0,0 @@ -/* $Id: namei.h,v 1.1 1996/12/13 14:48:21 jj Exp $ - * linux/include/asm-parisc/namei.h - * - * Included from linux/fs/namei.c - */ - -#ifndef __PARISC_NAMEI_H -#define __PARISC_NAMEI_H - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif /* __PARISC_NAMEI_H */ diff --git a/include/asm-powerpc/namei.h b/include/asm-powerpc/namei.h deleted file mode 100644 index 657443474a6a..000000000000 --- a/include/asm-powerpc/namei.h +++ /dev/null @@ -1,20 +0,0 @@ -#ifndef _ASM_POWERPC_NAMEI_H -#define _ASM_POWERPC_NAMEI_H - -#ifdef __KERNEL__ - -/* - * Adapted from include/asm-alpha/namei.h - * - * Included from fs/namei.c - */ - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif /* __KERNEL__ */ -#endif /* _ASM_POWERPC_NAMEI_H */ diff --git a/include/asm-s390/namei.h b/include/asm-s390/namei.h deleted file mode 100644 index 3e286bdde4b0..000000000000 --- a/include/asm-s390/namei.h +++ /dev/null @@ -1,21 +0,0 @@ -/* - * include/asm-s390/namei.h - * - * S390 version - * - * Derived from "include/asm-i386/namei.h" - * - * Included from linux/fs/namei.c - */ - -#ifndef __S390_NAMEI_H -#define __S390_NAMEI_H - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif /* __S390_NAMEI_H */ diff --git a/include/asm-sh/namei.h b/include/asm-sh/namei.h deleted file mode 100644 index 338a5d947143..000000000000 --- a/include/asm-sh/namei.h +++ /dev/null @@ -1,17 +0,0 @@ -/* $Id: namei.h,v 1.3 2000/07/04 06:24:49 gniibe Exp $ - * linux/include/asm-sh/namei.h - * - * Included from linux/fs/namei.c - */ - -#ifndef __ASM_SH_NAMEI_H -#define __ASM_SH_NAMEI_H - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif /* __ASM_SH_NAMEI_H */ diff --git a/include/asm-sparc/namei.h b/include/asm-sparc/namei.h deleted file mode 100644 index eff944b8e321..000000000000 --- a/include/asm-sparc/namei.h +++ /dev/null @@ -1,8 +0,0 @@ -#ifndef ___ASM_SPARC_NAMEI_H -#define ___ASM_SPARC_NAMEI_H -#if defined(__sparc__) && defined(__arch64__) -#include -#else -#include -#endif -#endif diff --git a/include/asm-sparc64/namei.h b/include/asm-sparc64/namei.h deleted file mode 100644 index 1344a910ba2f..000000000000 --- a/include/asm-sparc64/namei.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-um/namei.h b/include/asm-um/namei.h deleted file mode 100644 index 002984d5bc85..000000000000 --- a/include/asm-um/namei.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __UM_NAMEI_H -#define __UM_NAMEI_H - -#include "asm/arch/namei.h" - -#endif diff --git a/include/asm-v850/namei.h b/include/asm-v850/namei.h deleted file mode 100644 index ee8339b23843..000000000000 --- a/include/asm-v850/namei.h +++ /dev/null @@ -1,17 +0,0 @@ -/* - * linux/include/asm-v850/namei.h - * - * Included from linux/fs/namei.c - */ - -#ifndef __V850_NAMEI_H__ -#define __V850_NAMEI_H__ - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif /* __V850_NAMEI_H__ */ diff --git a/include/asm-x86/namei.h b/include/asm-x86/namei.h deleted file mode 100644 index 415ef5d9550e..000000000000 --- a/include/asm-x86/namei.h +++ /dev/null @@ -1,11 +0,0 @@ -#ifndef _ASM_X86_NAMEI_H -#define _ASM_X86_NAMEI_H - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif /* _ASM_X86_NAMEI_H */ diff --git a/include/asm-xtensa/namei.h b/include/asm-xtensa/namei.h deleted file mode 100644 index 3fdff039d27d..000000000000 --- a/include/asm-xtensa/namei.h +++ /dev/null @@ -1,26 +0,0 @@ -/* - * include/asm-xtensa/namei.h - * - * Included from linux/fs/namei.c - * - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file "COPYING" in the main directory of this archive - * for more details. - * - * Copyright (C) 2001 - 2005 Tensilica Inc. - */ - -#ifndef _XTENSA_NAMEI_H -#define _XTENSA_NAMEI_H - -#ifdef __KERNEL__ - -/* This dummy routine maybe changed to something useful - * for /usr/gnemul/ emulation stuff. - * Look at asm-sparc/namei.h for details. - */ - -#define __emul_prefix() NULL - -#endif /* __KERNEL__ */ -#endif /* _XTENSA_NAMEI_H */ diff --git a/include/linux/fs_struct.h b/include/linux/fs_struct.h index 282f54219129..9e5a06e78d02 100644 --- a/include/linux/fs_struct.h +++ b/include/linux/fs_struct.h @@ -7,7 +7,7 @@ struct fs_struct { atomic_t count; rwlock_t lock; int umask; - struct path root, pwd, altroot; + struct path root, pwd; }; #define INIT_FS { \ @@ -19,7 +19,6 @@ struct fs_struct { extern struct kmem_cache *fs_cachep; extern void exit_fs(struct task_struct *); -extern void set_fs_altroot(void); extern void set_fs_root(struct fs_struct *, struct path *); extern void set_fs_pwd(struct fs_struct *, struct path *); extern struct fs_struct *copy_fs_struct(struct fs_struct *); diff --git a/include/linux/namei.h b/include/linux/namei.h index 3cf62d26d493..768773d57857 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -47,7 +47,6 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND}; #define LOOKUP_DIRECTORY 2 #define LOOKUP_CONTINUE 4 #define LOOKUP_PARENT 16 -#define LOOKUP_NOALT 32 #define LOOKUP_REVAL 64 /* * Intent data diff --git a/kernel/exec_domain.c b/kernel/exec_domain.c index c1ef192aa655..0d407e886735 100644 --- a/kernel/exec_domain.c +++ b/kernel/exec_domain.c @@ -168,7 +168,6 @@ __set_personality(u_long personality) current->personality = personality; oep = current_thread_info()->exec_domain; current_thread_info()->exec_domain = ep; - set_fs_altroot(); module_put(oep->module); return 0; diff --git a/kernel/exit.c b/kernel/exit.c index 6cdf60712bd2..0caf590548a0 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -565,8 +565,6 @@ void put_fs_struct(struct fs_struct *fs) if (atomic_dec_and_test(&fs->count)) { path_put(&fs->root); path_put(&fs->pwd); - if (fs->altroot.dentry) - path_put(&fs->altroot); kmem_cache_free(fs_cachep, fs); } } diff --git a/kernel/fork.c b/kernel/fork.c index abb3ed6298f6..5e050c1317c4 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -657,13 +657,6 @@ static struct fs_struct *__copy_fs_struct(struct fs_struct *old) path_get(&old->root); fs->pwd = old->pwd; path_get(&old->pwd); - if (old->altroot.dentry) { - fs->altroot = old->altroot; - path_get(&old->altroot); - } else { - fs->altroot.mnt = NULL; - fs->altroot.dentry = NULL; - } read_unlock(&old->lock); } return fs; -- cgit v1.2.3 From a110343f0d6d41f68b7cf8c00b57a3172c67f816 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 17 Jul 2008 09:19:08 -0400 Subject: [PATCH] fix MAY_CHDIR/MAY_ACCESS/LOOKUP_ACCESS mess * MAY_CHDIR is redundant - it's an equivalent of MAY_ACCESS * MAY_ACCESS on fuse should affect only the last step of pathname resolution * fchdir() and chroot() should pass MAY_ACCESS, for the same reason why chdir() needs that. * now that we pass MAY_ACCESS explicitly in all cases, LOOKUP_ACCESS can be removed; it has no business being in nameidata. Signed-off-by: Al Viro --- fs/fuse/dir.c | 2 +- fs/namei.c | 2 -- fs/open.c | 10 +++++----- include/linux/fs.h | 3 +-- include/linux/namei.h | 1 - 5 files changed, 7 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 48a7934cb950..fd03330cadeb 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -962,7 +962,7 @@ static int fuse_permission(struct inode *inode, int mask) exist. So if permissions are revoked this won't be noticed immediately, only after the attribute timeout has expired */ - } else if (mask & (MAY_ACCESS | MAY_CHDIR)) { + } else if (mask & MAY_ACCESS) { err = fuse_access(inode, mask); } else if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode)) { if (!(inode->i_mode & S_IXUGO)) { diff --git a/fs/namei.c b/fs/namei.c index 095818089ac1..33dcaf025c49 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -265,8 +265,6 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) if (inode->i_op && inode->i_op->permission) { int extra = 0; if (nd) { - if (nd->flags & LOOKUP_ACCESS) - extra |= MAY_ACCESS; if (nd->flags & LOOKUP_OPEN) extra |= MAY_OPEN; } diff --git a/fs/open.c b/fs/open.c index d3a2a00f52dc..3317e1909b2c 100644 --- a/fs/open.c +++ b/fs/open.c @@ -457,11 +457,11 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) old_cap = cap_set_effective(current->cap_permitted); } - res = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW|LOOKUP_ACCESS, &nd); + res = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd); if (res) goto out; - res = vfs_permission(&nd, mode); + res = vfs_permission(&nd, mode | MAY_ACCESS); /* SuS v2 requires we report a read only fs too */ if(res || !(mode & S_IWOTH) || special_file(nd.path.dentry->d_inode->i_mode)) @@ -505,7 +505,7 @@ asmlinkage long sys_chdir(const char __user * filename) if (error) goto out; - error = vfs_permission(&nd, MAY_EXEC | MAY_CHDIR); + error = vfs_permission(&nd, MAY_EXEC | MAY_ACCESS); if (error) goto dput_and_out; @@ -534,7 +534,7 @@ asmlinkage long sys_fchdir(unsigned int fd) if (!S_ISDIR(inode->i_mode)) goto out_putf; - error = file_permission(file, MAY_EXEC); + error = file_permission(file, MAY_EXEC | MAY_ACCESS); if (!error) set_fs_pwd(current->fs, &file->f_path); out_putf: @@ -552,7 +552,7 @@ asmlinkage long sys_chroot(const char __user * filename) if (error) goto out; - error = vfs_permission(&nd, MAY_EXEC); + error = vfs_permission(&nd, MAY_EXEC | MAY_ACCESS); if (error) goto dput_and_out; diff --git a/include/linux/fs.h b/include/linux/fs.h index 25998e803fc2..d8721e818b45 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -61,8 +61,7 @@ extern int dir_notify_enable; #define MAY_READ 4 #define MAY_APPEND 8 #define MAY_ACCESS 16 -#define MAY_CHDIR 32 -#define MAY_OPEN 64 +#define MAY_OPEN 32 #define FMODE_READ 1 #define FMODE_WRITE 2 diff --git a/include/linux/namei.h b/include/linux/namei.h index 768773d57857..60e35a02f6cb 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -53,7 +53,6 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND}; */ #define LOOKUP_OPEN (0x0100) #define LOOKUP_CREATE (0x0200) -#define LOOKUP_ACCESS (0x0400) extern int __user_walk(const char __user *, unsigned, struct nameidata *); extern int __user_walk_fd(int dfd, const char __user *, unsigned, struct nameidata *); -- cgit v1.2.3 From b77b0646ef4efe31a7449bb3d9360fd00f95433d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 17 Jul 2008 09:37:02 -0400 Subject: [PATCH] pass MAY_OPEN to vfs_permission() explicitly ... and get rid of the last "let's deduce mask from nameidata->flags" bit. Signed-off-by: Al Viro --- fs/exec.c | 4 ++-- fs/namei.c | 13 ++++--------- include/linux/security.h | 7 +++---- security/capability.c | 3 +-- security/security.c | 4 ++-- security/selinux/hooks.c | 5 ++--- security/smack/smack_lsm.c | 3 +-- 7 files changed, 15 insertions(+), 24 deletions(-) (limited to 'fs') diff --git a/fs/exec.c b/fs/exec.c index b8792a131533..0ba5d355c5a1 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -118,7 +118,7 @@ asmlinkage long sys_uselib(const char __user * library) if (!S_ISREG(nd.path.dentry->d_inode->i_mode)) goto exit; - error = vfs_permission(&nd, MAY_READ | MAY_EXEC); + error = vfs_permission(&nd, MAY_READ | MAY_EXEC | MAY_OPEN); if (error) goto exit; @@ -666,7 +666,7 @@ struct file *open_exec(const char *name) struct inode *inode = nd.path.dentry->d_inode; file = ERR_PTR(-EACCES); if (S_ISREG(inode->i_mode)) { - int err = vfs_permission(&nd, MAY_EXEC); + int err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN); file = ERR_PTR(err); if (!err) { file = nameidata_to_filp(&nd, diff --git a/fs/namei.c b/fs/namei.c index 33dcaf025c49..6b0e8e5e079e 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -263,12 +263,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) /* Ordinary permission routines do not understand MAY_APPEND. */ if (inode->i_op && inode->i_op->permission) { - int extra = 0; - if (nd) { - if (nd->flags & LOOKUP_OPEN) - extra |= MAY_OPEN; - } - retval = inode->i_op->permission(inode, mask | extra); + retval = inode->i_op->permission(inode, mask); if (!retval) { /* * Exec permission on a regular file is denied if none @@ -292,7 +287,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) return retval; return security_inode_permission(inode, - mask & (MAY_READ|MAY_WRITE|MAY_EXEC), nd); + mask & (MAY_READ|MAY_WRITE|MAY_EXEC)); } /** @@ -492,7 +487,7 @@ static int exec_permission_lite(struct inode *inode, return -EACCES; ok: - return security_inode_permission(inode, MAY_EXEC, nd); + return security_inode_permission(inode, MAY_EXEC); } /* @@ -1692,7 +1687,7 @@ struct file *do_filp_open(int dfd, const char *pathname, int will_write; int flag = open_to_namei_flags(open_flag); - acc_mode = ACC_MODE(flag); + acc_mode = MAY_OPEN | ACC_MODE(flag); /* O_TRUNC implies we need access checks for write permissions */ if (flag & O_TRUNC) diff --git a/include/linux/security.h b/include/linux/security.h index f0e9adb22ac2..fd96e7f8a6f9 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -1362,7 +1362,7 @@ struct security_operations { struct inode *new_dir, struct dentry *new_dentry); int (*inode_readlink) (struct dentry *dentry); int (*inode_follow_link) (struct dentry *dentry, struct nameidata *nd); - int (*inode_permission) (struct inode *inode, int mask, struct nameidata *nd); + int (*inode_permission) (struct inode *inode, int mask); int (*inode_setattr) (struct dentry *dentry, struct iattr *attr); int (*inode_getattr) (struct vfsmount *mnt, struct dentry *dentry); void (*inode_delete) (struct inode *inode); @@ -1628,7 +1628,7 @@ int security_inode_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry); int security_inode_readlink(struct dentry *dentry); int security_inode_follow_link(struct dentry *dentry, struct nameidata *nd); -int security_inode_permission(struct inode *inode, int mask, struct nameidata *nd); +int security_inode_permission(struct inode *inode, int mask); 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); @@ -2021,8 +2021,7 @@ static inline int security_inode_follow_link(struct dentry *dentry, return 0; } -static inline int security_inode_permission(struct inode *inode, int mask, - struct nameidata *nd) +static inline int security_inode_permission(struct inode *inode, int mask) { return 0; } diff --git a/security/capability.c b/security/capability.c index 5b01c0b02422..63d10da515a5 100644 --- a/security/capability.c +++ b/security/capability.c @@ -211,8 +211,7 @@ static int cap_inode_follow_link(struct dentry *dentry, return 0; } -static int cap_inode_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int cap_inode_permission(struct inode *inode, int mask) { return 0; } diff --git a/security/security.c b/security/security.c index 59f23b5918b3..78ed3ffde242 100644 --- a/security/security.c +++ b/security/security.c @@ -429,11 +429,11 @@ int security_inode_follow_link(struct dentry *dentry, struct nameidata *nd) return security_ops->inode_follow_link(dentry, nd); } -int security_inode_permission(struct inode *inode, int mask, struct nameidata *nd) +int security_inode_permission(struct inode *inode, int mask) { if (unlikely(IS_PRIVATE(inode))) return 0; - return security_ops->inode_permission(inode, mask, nd); + return security_ops->inode_permission(inode, mask); } int security_inode_setattr(struct dentry *dentry, struct iattr *attr) diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 3481cde5bf15..5ba13908b5b4 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -2624,12 +2624,11 @@ static int selinux_inode_follow_link(struct dentry *dentry, struct nameidata *na return dentry_has_perm(current, NULL, dentry, FILE__READ); } -static int selinux_inode_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int selinux_inode_permission(struct inode *inode, int mask) { int rc; - rc = secondary_ops->inode_permission(inode, mask, nd); + rc = secondary_ops->inode_permission(inode, mask); if (rc) return rc; diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c index ee5a51cbc5eb..1b40e558f983 100644 --- a/security/smack/smack_lsm.c +++ b/security/smack/smack_lsm.c @@ -522,8 +522,7 @@ static int smack_inode_rename(struct inode *old_inode, * * Returns 0 if access is permitted, -EACCES otherwise */ -static int smack_inode_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int smack_inode_permission(struct inode *inode, int mask) { /* * No permission to check. Existence test. Yup, it's there. -- cgit v1.2.3 From 672b16b2f66c149888bd876a4f92342112205fe1 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 17 Jul 2008 09:45:01 -0400 Subject: [PATCH] more nameidata removal: exec_permission_lite() doesn't need it Signed-off-by: Al Viro --- fs/namei.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index 6b0e8e5e079e..6d75430358ac 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -460,8 +460,7 @@ static struct dentry * cached_lookup(struct dentry * parent, struct qstr * name, * short-cut DAC fails, then call permission() to do more * complete permission check. */ -static int exec_permission_lite(struct inode *inode, - struct nameidata *nd) +static int exec_permission_lite(struct inode *inode) { umode_t mode = inode->i_mode; @@ -884,7 +883,7 @@ static int __link_path_walk(const char *name, struct nameidata *nd) unsigned int c; nd->flags |= LOOKUP_CONTINUE; - err = exec_permission_lite(inode, nd); + err = exec_permission_lite(inode); if (err == -EAGAIN) err = vfs_permission(nd, MAY_EXEC); if (err) -- cgit v1.2.3 From 88b387824fdaecb6ba0f471acf0aadf7d24739fd Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Mon, 21 Jul 2008 18:06:36 +0800 Subject: [PATCH] vfs: use kstrdup() and check failing allocation - use kstrdup() instead of kmalloc() + memcpy() - return NULL if allocating ->mnt_devname failed - mnt_devname should be const Signed-off-by: Li Zefan Acked-by: Cyrill Gorcunov Signed-off-by: Al Viro --- fs/namespace.c | 24 +++++++++++++----------- include/linux/mount.h | 2 +- 2 files changed, 14 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/namespace.c b/fs/namespace.c index c4fcf48acef8..26380f599534 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -112,9 +112,13 @@ struct vfsmount *alloc_vfsmnt(const char *name) int err; err = mnt_alloc_id(mnt); - if (err) { - kmem_cache_free(mnt_cache, mnt); - return NULL; + if (err) + goto out_free_cache; + + if (name) { + mnt->mnt_devname = kstrdup(name, GFP_KERNEL); + if (!mnt->mnt_devname) + goto out_free_id; } atomic_set(&mnt->mnt_count, 1); @@ -127,16 +131,14 @@ struct vfsmount *alloc_vfsmnt(const char *name) INIT_LIST_HEAD(&mnt->mnt_slave_list); INIT_LIST_HEAD(&mnt->mnt_slave); atomic_set(&mnt->__mnt_writers, 0); - if (name) { - int size = strlen(name) + 1; - char *newname = kmalloc(size, GFP_KERNEL); - if (newname) { - memcpy(newname, name, size); - mnt->mnt_devname = newname; - } - } } return mnt; + +out_free_id: + mnt_free_id(mnt); +out_free_cache: + kmem_cache_free(mnt_cache, mnt); + return NULL; } /* diff --git a/include/linux/mount.h b/include/linux/mount.h index 4374d1adeb4b..b5efaa2132ab 100644 --- a/include/linux/mount.h +++ b/include/linux/mount.h @@ -47,7 +47,7 @@ struct vfsmount { struct list_head mnt_child; /* and going through their mnt_child */ int mnt_flags; /* 4 bytes hole on 64bits arches */ - char *mnt_devname; /* Name of device e.g. /dev/dsk/hda1 */ + const char *mnt_devname; /* Name of device e.g. /dev/dsk/hda1 */ struct list_head mnt_list; struct list_head mnt_expire; /* link in fs-specific expiry list */ struct list_head mnt_share; /* circular list of shared mounts */ -- cgit v1.2.3 From 9767d74957450da6365c363d69e3d02d605d7375 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 1 Jul 2008 15:01:26 +0200 Subject: [patch 1/4] vfs: utimes: move owner check into inode_change_ok() Add a new ia_valid flag: ATTR_TIMES_SET, to handle the UTIMES_OMIT/UTIMES_NOW and UTIMES_NOW/UTIMES_OMIT cases. In these cases neither ATTR_MTIME_SET nor ATTR_ATIME_SET is in the flags, yet the POSIX draft specifies that permission checking is performed the same way as if one or both of the times was explicitly set to a timestamp. See the path "vfs: utimensat(): fix error checking for {UTIME_NOW,UTIME_OMIT} case" by Michael Kerrisk for the patch introducing this behavior. This is a cleanup, as well as allowing filesystems (NFS/fuse/...) to perform their own permission checking instead of the default. CC: Ulrich Drepper CC: Michael Kerrisk Signed-off-by: Miklos Szeredi Signed-off-by: Al Viro --- fs/attr.c | 2 +- fs/utimes.c | 17 ++++------------- include/linux/fs.h | 33 +++++++++++++++++---------------- 3 files changed, 22 insertions(+), 30 deletions(-) (limited to 'fs') diff --git a/fs/attr.c b/fs/attr.c index 966b73e25f82..765fc75fab3b 100644 --- a/fs/attr.c +++ b/fs/attr.c @@ -51,7 +51,7 @@ int inode_change_ok(struct inode *inode, struct iattr *attr) } /* Check for setting the inode time. */ - if (ia_valid & (ATTR_MTIME_SET | ATTR_ATIME_SET)) { + if (ia_valid & (ATTR_MTIME_SET | ATTR_ATIME_SET | ATTR_TIMES_SET)) { if (!is_owner_or_cap(inode)) goto error; } diff --git a/fs/utimes.c b/fs/utimes.c index b6b664e7145e..ecf8941ba34a 100644 --- a/fs/utimes.c +++ b/fs/utimes.c @@ -101,7 +101,6 @@ long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags times[1].tv_nsec == UTIME_NOW) times = NULL; - /* In most cases, the checks are done in inode_change_ok() */ newattrs.ia_valid = ATTR_CTIME | ATTR_MTIME | ATTR_ATIME; if (times) { error = -EPERM; @@ -123,21 +122,13 @@ 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; } - /* - * For the UTIME_OMIT/UTIME_NOW and UTIME_NOW/UTIME_OMIT - * cases, we need to make an extra check that is not done by - * inode_change_ok(). + * Tell inode_change_ok(), that this is an explicit time + * update, even if neither ATTR_ATIME_SET nor ATTR_MTIME_SET + * were used. */ - if (((times[0].tv_nsec == UTIME_NOW && - times[1].tv_nsec == UTIME_OMIT) - || - (times[0].tv_nsec == UTIME_OMIT && - times[1].tv_nsec == UTIME_NOW)) - && !is_owner_or_cap(inode)) - goto mnt_drop_write_and_out; + newattrs.ia_valid |= ATTR_TIMES_SET; } else { - /* * If times is NULL (or both times are UTIME_NOW), * then we need to check permissions, because diff --git a/include/linux/fs.h b/include/linux/fs.h index d8721e818b45..527b9e482f99 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -320,22 +320,23 @@ typedef void (dio_iodone_t)(struct kiocb *iocb, loff_t offset, * Attribute flags. These should be or-ed together to figure out what * has been changed! */ -#define ATTR_MODE 1 -#define ATTR_UID 2 -#define ATTR_GID 4 -#define ATTR_SIZE 8 -#define ATTR_ATIME 16 -#define ATTR_MTIME 32 -#define ATTR_CTIME 64 -#define ATTR_ATIME_SET 128 -#define ATTR_MTIME_SET 256 -#define ATTR_FORCE 512 /* Not a change, but a change it */ -#define ATTR_ATTR_FLAG 1024 -#define ATTR_KILL_SUID 2048 -#define ATTR_KILL_SGID 4096 -#define ATTR_FILE 8192 -#define ATTR_KILL_PRIV 16384 -#define ATTR_OPEN 32768 /* Truncating from open(O_TRUNC) */ +#define ATTR_MODE (1 << 0) +#define ATTR_UID (1 << 1) +#define ATTR_GID (1 << 2) +#define ATTR_SIZE (1 << 3) +#define ATTR_ATIME (1 << 4) +#define ATTR_MTIME (1 << 5) +#define ATTR_CTIME (1 << 6) +#define ATTR_ATIME_SET (1 << 7) +#define ATTR_MTIME_SET (1 << 8) +#define ATTR_FORCE (1 << 9) /* Not a change, but a change it */ +#define ATTR_ATTR_FLAG (1 << 10) +#define ATTR_KILL_SUID (1 << 11) +#define ATTR_KILL_SGID (1 << 12) +#define ATTR_FILE (1 << 13) +#define ATTR_KILL_PRIV (1 << 14) +#define ATTR_OPEN (1 << 15) /* Truncating from open(O_TRUNC) */ +#define ATTR_TIMES_SET (1 << 16) /* * This is the Inode Attributes structure, used for notify_change(). It -- cgit v1.2.3 From e9b76fedc61235da80b6b7f81dfd67ec224dfb49 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 1 Jul 2008 15:01:27 +0200 Subject: [patch 2/4] vfs: utimes cleanup Untange the mess that is do_utimes(). Add kerneldoc comment to do_utimes(). CC: Ulrich Drepper CC: Michael Kerrisk Signed-off-by: Miklos Szeredi Signed-off-by: Al Viro --- fs/utimes.c | 114 ++++++++++++++++++++++++++++++++++-------------------------- 1 file changed, 65 insertions(+), 49 deletions(-) (limited to 'fs') diff --git a/fs/utimes.c b/fs/utimes.c index ecf8941ba34a..8e09dbdfd7f5 100644 --- a/fs/utimes.c +++ b/fs/utimes.c @@ -48,54 +48,15 @@ static bool nsec_valid(long nsec) return nsec >= 0 && nsec <= 999999999; } -/* If times==NULL, set access and modification to current time, - * must be owner or have write permission. - * Else, update from *times, must be owner or super user. - */ -long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags) +static int utimes_common(struct path *path, struct timespec *times) { int error; - struct nameidata nd; - struct dentry *dentry; - struct inode *inode; struct iattr newattrs; - struct file *f = NULL; - struct vfsmount *mnt; - - error = -EINVAL; - if (times && (!nsec_valid(times[0].tv_nsec) || - !nsec_valid(times[1].tv_nsec))) { - goto out; - } - - if (flags & ~AT_SYMLINK_NOFOLLOW) - goto out; - - if (filename == NULL && dfd != AT_FDCWD) { - error = -EINVAL; - if (flags & AT_SYMLINK_NOFOLLOW) - goto out; - - error = -EBADF; - f = fget(dfd); - if (!f) - goto out; - dentry = f->f_path.dentry; - mnt = f->f_path.mnt; - } else { - error = __user_walk_fd(dfd, filename, (flags & AT_SYMLINK_NOFOLLOW) ? 0 : LOOKUP_FOLLOW, &nd); - if (error) - goto out; - - dentry = nd.path.dentry; - mnt = nd.path.mnt; - } - - inode = dentry->d_inode; + struct inode *inode = path->dentry->d_inode; - error = mnt_want_write(mnt); + error = mnt_want_write(path->mnt); if (error) - goto dput_and_out; + goto out; if (times && times[0].tv_nsec == UTIME_NOW && times[1].tv_nsec == UTIME_NOW) @@ -145,15 +106,70 @@ long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags } } mutex_lock(&inode->i_mutex); - error = notify_change(dentry, &newattrs); + error = notify_change(path->dentry, &newattrs); mutex_unlock(&inode->i_mutex); + mnt_drop_write_and_out: - mnt_drop_write(mnt); -dput_and_out: - if (f) - fput(f); - else + mnt_drop_write(path->mnt); +out: + return error; +} + +/* + * do_utimes - change times on filename or file descriptor + * @dfd: open file descriptor, -1 or AT_FDCWD + * @filename: path name or NULL + * @times: new times or NULL + * @flags: zero or more flags (only AT_SYMLINK_NOFOLLOW for the moment) + * + * If filename is NULL and dfd refers to an open file, then operate on + * the file. Otherwise look up filename, possibly using dfd as a + * starting point. + * + * If times==NULL, set access and modification to current time, + * must be owner or have write permission. + * Else, update from *times, must be owner or super user. + */ +long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags) +{ + int error = -EINVAL; + + if (times && (!nsec_valid(times[0].tv_nsec) || + !nsec_valid(times[1].tv_nsec))) { + goto out; + } + + if (flags & ~AT_SYMLINK_NOFOLLOW) + goto out; + + if (filename == NULL && dfd != AT_FDCWD) { + struct file *file; + + if (flags & AT_SYMLINK_NOFOLLOW) + goto out; + + file = fget(dfd); + error = -EBADF; + if (!file) + goto out; + + error = utimes_common(&file->f_path, times); + fput(file); + } else { + struct nameidata nd; + int lookup_flags = 0; + + if (!(flags & AT_SYMLINK_NOFOLLOW)) + lookup_flags |= LOOKUP_FOLLOW; + + error = __user_walk_fd(dfd, filename, lookup_flags, &nd); + if (error) + goto out; + + error = utimes_common(&nd.path, times); path_put(&nd.path); + } + out: return error; } -- cgit v1.2.3 From b1da47e29e467f1ec36dc78d009bfb109fd533c7 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 1 Jul 2008 15:01:28 +0200 Subject: [patch 3/4] fat: dont call notify_change The FAT_IOCTL_SET_ATTRIBUTES ioctl() calls notify_change() to change the file mode before changing the inode attributes. Replace with explicit calls to security_inode_setattr(), fat_setattr() and fsnotify_change(). This is equivalent to the original. The reason it is needed, is that later in the series we move the immutable check into notify_change(). That would break the FAT_IOCTL_SET_ATTRIBUTES ioctl, as it needs to perform the mode change regardless of the immutability of the file. [Fix error if fat is built as a module. Thanks to OGAWA Hirofumi for noticing.] Signed-off-by: Miklos Szeredi Acked-by: OGAWA Hirofumi Signed-off-by: Al Viro --- fs/fat/file.c | 15 ++++++++++++++- security/security.c | 1 + 2 files changed, 15 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/fat/file.c b/fs/fat/file.c index c672df4036e9..8707a8cfa02c 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -15,6 +15,8 @@ #include #include #include +#include +#include int fat_generic_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg) @@ -64,6 +66,7 @@ int fat_generic_ioctl(struct inode *inode, struct file *filp, /* Equivalent to a chmod() */ ia.ia_valid = ATTR_MODE | ATTR_CTIME; + ia.ia_ctime = current_fs_time(inode->i_sb); if (is_dir) { ia.ia_mode = MSDOS_MKMODE(attr, S_IRWXUGO & ~sbi->options.fs_dmask) @@ -90,11 +93,21 @@ int fat_generic_ioctl(struct inode *inode, struct file *filp, } } + /* + * The security check is questionable... We single + * out the RO attribute for checking by the security + * module, just because it maps to a file mode. + */ + err = security_inode_setattr(filp->f_path.dentry, &ia); + if (err) + goto up; + /* This MUST be done before doing anything irreversible... */ - err = notify_change(filp->f_path.dentry, &ia); + err = fat_setattr(filp->f_path.dentry, &ia); if (err) goto up; + fsnotify_change(filp->f_path.dentry, ia.ia_valid); if (sbi->options.sys_immutable) { if (attr & ATTR_SYS) inode->i_flags |= S_IMMUTABLE; diff --git a/security/security.c b/security/security.c index 78ed3ffde242..ff7068727757 100644 --- a/security/security.c +++ b/security/security.c @@ -442,6 +442,7 @@ int security_inode_setattr(struct dentry *dentry, struct iattr *attr) return 0; return security_ops->inode_setattr(dentry, attr); } +EXPORT_SYMBOL_GPL(security_inode_setattr); int security_inode_getattr(struct vfsmount *mnt, struct dentry *dentry) { -- cgit v1.2.3 From beb29e058c35ab69e96e455a12ccf7505f6de425 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 1 Jul 2008 15:01:29 +0200 Subject: [patch 4/4] vfs: immutable inode checking cleanup Move the immutable and append-only checks from chmod, chown and utimes into notify_change(). Checks for immutable and append-only files are always performed by the VFS and not by the filesystem (see permission() and may_...() in namei.c), so these belong in notify_change(), and not in inode_change_ok(). This should be completely equivalent. CC: Ulrich Drepper CC: Michael Kerrisk Signed-off-by: Miklos Szeredi Signed-off-by: Al Viro --- fs/attr.c | 5 +++++ fs/open.c | 24 ++---------------------- fs/utimes.c | 4 ---- 3 files changed, 7 insertions(+), 26 deletions(-) (limited to 'fs') diff --git a/fs/attr.c b/fs/attr.c index 765fc75fab3b..26c71ba1eed4 100644 --- a/fs/attr.c +++ b/fs/attr.c @@ -108,6 +108,11 @@ int notify_change(struct dentry * dentry, struct iattr * attr) struct timespec now; unsigned int ia_valid = attr->ia_valid; + if (ia_valid & (ATTR_MODE | ATTR_UID | ATTR_GID | ATTR_TIMES_SET)) { + if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) + return -EPERM; + } + now = current_fs_time(inode->i_sb); attr->ia_ctime = now; diff --git a/fs/open.c b/fs/open.c index 3317e1909b2c..3b3c43674be3 100644 --- a/fs/open.c +++ b/fs/open.c @@ -588,9 +588,6 @@ asmlinkage long sys_fchmod(unsigned int fd, mode_t mode) err = mnt_want_write(file->f_path.mnt); if (err) goto out_putf; - err = -EPERM; - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - goto out_drop_write; mutex_lock(&inode->i_mutex); if (mode == (mode_t) -1) mode = inode->i_mode; @@ -598,8 +595,6 @@ asmlinkage long sys_fchmod(unsigned int fd, mode_t mode) newattrs.ia_valid = ATTR_MODE | ATTR_CTIME; err = notify_change(dentry, &newattrs); mutex_unlock(&inode->i_mutex); - -out_drop_write: mnt_drop_write(file->f_path.mnt); out_putf: fput(file); @@ -623,11 +618,6 @@ asmlinkage long sys_fchmodat(int dfd, const char __user *filename, error = mnt_want_write(nd.path.mnt); if (error) goto dput_and_out; - - error = -EPERM; - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - goto out_drop_write; - mutex_lock(&inode->i_mutex); if (mode == (mode_t) -1) mode = inode->i_mode; @@ -635,8 +625,6 @@ asmlinkage long sys_fchmodat(int dfd, const char __user *filename, newattrs.ia_valid = ATTR_MODE | ATTR_CTIME; error = notify_change(nd.path.dentry, &newattrs); mutex_unlock(&inode->i_mutex); - -out_drop_write: mnt_drop_write(nd.path.mnt); dput_and_out: path_put(&nd.path); @@ -651,18 +639,10 @@ asmlinkage long sys_chmod(const char __user *filename, mode_t mode) static int chown_common(struct dentry * dentry, uid_t user, gid_t group) { - struct inode * inode; + struct inode *inode = dentry->d_inode; int error; struct iattr newattrs; - error = -ENOENT; - if (!(inode = dentry->d_inode)) { - printk(KERN_ERR "chown_common: NULL inode\n"); - goto out; - } - error = -EPERM; - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - goto out; newattrs.ia_valid = ATTR_CTIME; if (user != (uid_t) -1) { newattrs.ia_valid |= ATTR_UID; @@ -678,7 +658,7 @@ static int chown_common(struct dentry * dentry, uid_t user, gid_t group) mutex_lock(&inode->i_mutex); error = notify_change(dentry, &newattrs); mutex_unlock(&inode->i_mutex); -out: + return error; } diff --git a/fs/utimes.c b/fs/utimes.c index 8e09dbdfd7f5..dad679d3a158 100644 --- a/fs/utimes.c +++ b/fs/utimes.c @@ -64,10 +64,6 @@ static int utimes_common(struct path *path, struct timespec *times) newattrs.ia_valid = ATTR_CTIME | ATTR_MTIME | ATTR_ATIME; if (times) { - error = -EPERM; - if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) - goto mnt_drop_write_and_out; - if (times[0].tv_nsec == UTIME_OMIT) newattrs.ia_valid &= ~ATTR_ATIME; else if (times[0].tv_nsec != UTIME_NOW) { -- cgit v1.2.3 From e56b6a5dda1a36ffaa532df6f975ea324298fa4d Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 19 May 2008 07:53:34 +0200 Subject: Re: [PATCH 3/6] vfs: open_exec cleanup On Mon, May 19, 2008 at 12:01:49AM +0200, Marcin Slusarz wrote: > open_exec is needlessly indented, calls ERR_PTR with 0 argument > (which is not valid errno) and jumps into middle of function > just to return value. > So clean it up a bit. Still looks rather messy. See below for a better version. Signed-off-by: Christoph Hellwig Signed-off-by: Al Viro --- fs/exec.c | 58 ++++++++++++++++++++++++++++++---------------------------- 1 file changed, 30 insertions(+), 28 deletions(-) (limited to 'fs') diff --git a/fs/exec.c b/fs/exec.c index 0ba5d355c5a1..346e3f69c6e0 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -656,38 +656,40 @@ EXPORT_SYMBOL(setup_arg_pages); struct file *open_exec(const char *name) { struct nameidata nd; - int err; struct file *file; + int err; - err = path_lookup_open(AT_FDCWD, name, LOOKUP_FOLLOW, &nd, FMODE_READ|FMODE_EXEC); - file = ERR_PTR(err); - - if (!err) { - struct inode *inode = nd.path.dentry->d_inode; - file = ERR_PTR(-EACCES); - if (S_ISREG(inode->i_mode)) { - int err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN); - file = ERR_PTR(err); - if (!err) { - file = nameidata_to_filp(&nd, - O_RDONLY|O_LARGEFILE); - if (!IS_ERR(file)) { - err = deny_write_access(file); - if (err) { - fput(file); - file = ERR_PTR(err); - } - } -out: - return file; - } - } - release_open_intent(&nd); - path_put(&nd.path); + err = path_lookup_open(AT_FDCWD, name, LOOKUP_FOLLOW, &nd, + FMODE_READ|FMODE_EXEC); + if (err) + goto out; + + err = -EACCES; + if (!S_ISREG(nd.path.dentry->d_inode->i_mode)) + goto out_path_put; + + err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN); + if (err) + goto out_path_put; + + file = nameidata_to_filp(&nd, O_RDONLY|O_LARGEFILE); + if (IS_ERR(file)) + return file; + + err = deny_write_access(file); + if (err) { + fput(file); + goto out; } - goto out; -} + return file; + + out_path_put: + release_open_intent(&nd); + path_put(&nd.path); + out: + return ERR_PTR(err); +} EXPORT_SYMBOL(open_exec); int kernel_read(struct file *file, unsigned long offset, -- cgit v1.2.3 From 30524472c2f728c20d6bf35191042a5d455c0a64 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 22 Jul 2008 00:02:33 -0400 Subject: [PATCH] take noexec checks to very few callers that care Signed-off-by: Al Viro --- fs/exec.c | 7 +++++++ fs/namei.c | 9 --------- fs/open.c | 10 ++++++++++ 3 files changed, 17 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/exec.c b/fs/exec.c index 346e3f69c6e0..eca58c29eded 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -118,6 +118,10 @@ asmlinkage long sys_uselib(const char __user * library) if (!S_ISREG(nd.path.dentry->d_inode->i_mode)) goto exit; + error = -EACCES; + if (nd.path.mnt->mnt_flags & MNT_NOEXEC) + goto exit; + error = vfs_permission(&nd, MAY_READ | MAY_EXEC | MAY_OPEN); if (error) goto exit; @@ -668,6 +672,9 @@ struct file *open_exec(const char *name) if (!S_ISREG(nd.path.dentry->d_inode->i_mode)) goto out_path_put; + if (nd.path.mnt->mnt_flags & MNT_NOEXEC) + goto out_path_put; + err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN); if (err) goto out_path_put; diff --git a/fs/namei.c b/fs/namei.c index 6d75430358ac..396cb3e5c364 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -252,15 +252,6 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) return -EACCES; } - if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode)) { - /* - * MAY_EXEC on regular files is denied if the fs is mounted - * with the "noexec" flag. - */ - if (mnt && (mnt->mnt_flags & MNT_NOEXEC)) - return -EACCES; - } - /* Ordinary permission routines do not understand MAY_APPEND. */ if (inode->i_op && inode->i_op->permission) { retval = inode->i_op->permission(inode, mask); diff --git a/fs/open.c b/fs/open.c index 3b3c43674be3..d5e421ad0cf6 100644 --- a/fs/open.c +++ b/fs/open.c @@ -461,6 +461,16 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) if (res) goto out; + if ((mode & MAY_EXEC) && S_ISREG(nd.path.dentry->d_inode->i_mode)) { + /* + * MAY_EXEC on regular files is denied if the fs is mounted + * with the "noexec" flag. + */ + res = -EACCES; + if (nd.path.mnt->mnt_flags & MNT_NOEXEC) + goto out_path_release; + } + res = vfs_permission(&nd, mode | MAY_ACCESS); /* SuS v2 requires we report a read only fs too */ if(res || !(mode & S_IWOTH) || -- cgit v1.2.3 From f419a2e3b64def707e1384ee38abb77f99af5f6d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 22 Jul 2008 00:07:17 -0400 Subject: [PATCH] kill nameidata passing to permission(), rename to inode_permission() Incidentally, the name that gives hundreds of false positives on grep is not a good idea... Signed-off-by: Al Viro --- fs/ecryptfs/inode.c | 2 +- fs/namei.c | 22 +++++++++------------- fs/nfsd/nfsfh.c | 2 +- fs/nfsd/vfs.c | 4 ++-- fs/utimes.c | 2 +- fs/xattr.c | 2 +- include/linux/fs.h | 2 +- ipc/mqueue.c | 2 +- 8 files changed, 17 insertions(+), 21 deletions(-) (limited to 'fs') diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index f25caf2b0887..89209f00f9c7 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -830,7 +830,7 @@ out: static int ecryptfs_permission(struct inode *inode, int mask) { - return permission(ecryptfs_inode_to_lower(inode), mask, NULL); + return inode_permission(ecryptfs_inode_to_lower(inode), mask); } /** diff --git a/fs/namei.c b/fs/namei.c index 396cb3e5c364..5029b93ebbd5 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -227,13 +227,9 @@ int generic_permission(struct inode *inode, int mask, return -EACCES; } -int permission(struct inode *inode, int mask, struct nameidata *nd) +int inode_permission(struct inode *inode, int mask) { int retval; - struct vfsmount *mnt = NULL; - - if (nd) - mnt = nd->path.mnt; if (mask & MAY_WRITE) { umode_t mode = inode->i_mode; @@ -293,7 +289,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) */ int vfs_permission(struct nameidata *nd, int mask) { - return permission(nd->path.dentry->d_inode, mask, nd); + return inode_permission(nd->path.dentry->d_inode, mask); } /** @@ -310,7 +306,7 @@ int vfs_permission(struct nameidata *nd, int mask) */ int file_permission(struct file *file, int mask) { - return permission(file->f_path.dentry->d_inode, mask, NULL); + return inode_permission(file->f_path.dentry->d_inode, mask); } /* @@ -1262,7 +1258,7 @@ static struct dentry *lookup_hash(struct nameidata *nd) { int err; - err = permission(nd->path.dentry->d_inode, MAY_EXEC, nd); + err = inode_permission(nd->path.dentry->d_inode, MAY_EXEC); if (err) return ERR_PTR(err); return __lookup_hash(&nd->last, nd->path.dentry, nd); @@ -1310,7 +1306,7 @@ struct dentry *lookup_one_len(const char *name, struct dentry *base, int len) if (err) return ERR_PTR(err); - err = permission(base->d_inode, MAY_EXEC, NULL); + err = inode_permission(base->d_inode, MAY_EXEC); if (err) return ERR_PTR(err); return __lookup_hash(&this, base, NULL); @@ -1400,7 +1396,7 @@ static int may_delete(struct inode *dir,struct dentry *victim,int isdir) BUG_ON(victim->d_parent->d_inode != dir); audit_inode_child(victim->d_name.name, victim, dir); - error = permission(dir,MAY_WRITE | MAY_EXEC, NULL); + error = inode_permission(dir, MAY_WRITE | MAY_EXEC); if (error) return error; if (IS_APPEND(dir)) @@ -1437,7 +1433,7 @@ static inline int may_create(struct inode *dir, struct dentry *child, return -EEXIST; if (IS_DEADDIR(dir)) return -ENOENT; - return permission(dir,MAY_WRITE | MAY_EXEC, nd); + return inode_permission(dir, MAY_WRITE | MAY_EXEC); } /* @@ -2543,7 +2539,7 @@ static int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry, * we'll need to flip '..'. */ if (new_dir != old_dir) { - error = permission(old_dentry->d_inode, MAY_WRITE, NULL); + error = inode_permission(old_dentry->d_inode, MAY_WRITE); if (error) return error; } @@ -2897,7 +2893,7 @@ EXPORT_SYMBOL(page_symlink); EXPORT_SYMBOL(page_symlink_inode_operations); EXPORT_SYMBOL(path_lookup); EXPORT_SYMBOL(vfs_path_lookup); -EXPORT_SYMBOL(permission); +EXPORT_SYMBOL(inode_permission); EXPORT_SYMBOL(vfs_permission); EXPORT_SYMBOL(file_permission); EXPORT_SYMBOL(unlock_rename); diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index f45451eb1e38..ea37c96f0445 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -51,7 +51,7 @@ static int nfsd_acceptable(void *expv, struct dentry *dentry) /* make sure parents give x permission to user */ int err; parent = dget_parent(tdentry); - err = permission(parent->d_inode, MAY_EXEC, NULL); + err = inode_permission(parent->d_inode, MAY_EXEC); if (err < 0) { dput(parent); break; diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index ad1ad59e3742..18060bed5267 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1953,12 +1953,12 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp, return 0; /* This assumes NFSD_MAY_{READ,WRITE,EXEC} == MAY_{READ,WRITE,EXEC} */ - err = permission(inode, acc & (MAY_READ|MAY_WRITE|MAY_EXEC), NULL); + err = inode_permission(inode, acc & (MAY_READ|MAY_WRITE|MAY_EXEC)); /* Allow read access to binaries even when mode 111 */ if (err == -EACCES && S_ISREG(inode->i_mode) && acc == (NFSD_MAY_READ | NFSD_MAY_OWNER_OVERRIDE)) - err = permission(inode, MAY_EXEC, NULL); + err = inode_permission(inode, MAY_EXEC); return err? nfserrno(err) : 0; } diff --git a/fs/utimes.c b/fs/utimes.c index dad679d3a158..dc28b7826259 100644 --- a/fs/utimes.c +++ b/fs/utimes.c @@ -96,7 +96,7 @@ static int utimes_common(struct path *path, struct timespec *times) goto mnt_drop_write_and_out; if (!is_owner_or_cap(inode)) { - error = permission(inode, MAY_WRITE, NULL); + error = inode_permission(inode, MAY_WRITE); if (error) goto mnt_drop_write_and_out; } diff --git a/fs/xattr.c b/fs/xattr.c index 4706a8b1f495..b96222e05ba0 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -63,7 +63,7 @@ xattr_permission(struct inode *inode, const char *name, int mask) return -EPERM; } - return permission(inode, mask, NULL); + return inode_permission(inode, mask); } int diff --git a/include/linux/fs.h b/include/linux/fs.h index 527b9e482f99..9d2de4cadabd 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1770,7 +1770,7 @@ extern int do_remount_sb(struct super_block *sb, int flags, extern sector_t bmap(struct inode *, sector_t); #endif extern int notify_change(struct dentry *, struct iattr *); -extern int permission(struct inode *, int, struct nameidata *); +extern int inode_permission(struct inode *, int); extern int generic_permission(struct inode *, int, int (*check_acl)(struct inode *, int)); diff --git a/ipc/mqueue.c b/ipc/mqueue.c index 474984f9e032..96fb36cd9874 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -638,7 +638,7 @@ static int oflag2acc[O_ACCMODE] = { MAY_READ, MAY_WRITE, return ERR_PTR(-EINVAL); } - if (permission(dentry->d_inode, oflag2acc[oflag & O_ACCMODE], NULL)) { + if (inode_permission(dentry->d_inode, oflag2acc[oflag & O_ACCMODE])) { dput(dentry); mntput(mqueue_mnt); return ERR_PTR(-EACCES); -- cgit v1.2.3 From 256984a83880ff7ac78055cb87baea48137f0b77 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 22 Jul 2008 08:09:30 -0400 Subject: [PATCH] preparation to __user_walk_fd cleanup Almost all users __user_walk_fd() and friends care only about struct path. Get rid of the few that do not. Signed-off-by: Al Viro --- fs/inotify_user.c | 2 +- fs/open.c | 18 ++++++++++-------- 2 files changed, 11 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/inotify_user.c b/fs/inotify_user.c index fe79c25d95dc..9b99ebf28884 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -365,7 +365,7 @@ static int find_inode(const char __user *dirname, struct nameidata *nd, if (error) return error; /* you can only watch an inode if you have read permissions on it */ - error = vfs_permission(nd, MAY_READ); + error = inode_permission(nd->path.dentry->d_inode, MAY_READ); if (error) path_put(&nd->path); return error; diff --git a/fs/open.c b/fs/open.c index d5e421ad0cf6..e94266700eda 100644 --- a/fs/open.c +++ b/fs/open.c @@ -251,7 +251,7 @@ static long do_sys_truncate(const char __user * path, loff_t length) if (error) goto dput_and_out; - error = vfs_permission(&nd, MAY_WRITE); + error = inode_permission(inode, MAY_WRITE); if (error) goto mnt_drop_write_and_out; @@ -426,6 +426,7 @@ out: asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) { struct nameidata nd; + struct inode *inode; int old_fsuid, old_fsgid; kernel_cap_t uninitialized_var(old_cap); /* !SECURE_NO_SETUID_FIXUP */ int res; @@ -461,7 +462,9 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) if (res) goto out; - if ((mode & MAY_EXEC) && S_ISREG(nd.path.dentry->d_inode->i_mode)) { + inode = nd.path.dentry->d_inode; + + if ((mode & MAY_EXEC) && S_ISREG(inode->i_mode)) { /* * MAY_EXEC on regular files is denied if the fs is mounted * with the "noexec" flag. @@ -471,10 +474,9 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) goto out_path_release; } - res = vfs_permission(&nd, mode | MAY_ACCESS); + res = inode_permission(inode, mode | MAY_ACCESS); /* SuS v2 requires we report a read only fs too */ - if(res || !(mode & S_IWOTH) || - special_file(nd.path.dentry->d_inode->i_mode)) + if (res || !(mode & S_IWOTH) || special_file(inode->i_mode)) goto out_path_release; /* * This is a rare case where using __mnt_is_readonly() @@ -515,7 +517,7 @@ asmlinkage long sys_chdir(const char __user * filename) if (error) goto out; - error = vfs_permission(&nd, MAY_EXEC | MAY_ACCESS); + error = inode_permission(nd.path.dentry->d_inode, MAY_EXEC | MAY_ACCESS); if (error) goto dput_and_out; @@ -544,7 +546,7 @@ asmlinkage long sys_fchdir(unsigned int fd) if (!S_ISDIR(inode->i_mode)) goto out_putf; - error = file_permission(file, MAY_EXEC | MAY_ACCESS); + error = inode_permission(inode, MAY_EXEC | MAY_ACCESS); if (!error) set_fs_pwd(current->fs, &file->f_path); out_putf: @@ -562,7 +564,7 @@ asmlinkage long sys_chroot(const char __user * filename) if (error) goto out; - error = vfs_permission(&nd, MAY_EXEC | MAY_ACCESS); + error = inode_permission(nd.path.dentry->d_inode, MAY_EXEC | MAY_ACCESS); if (error) goto dput_and_out; -- cgit v1.2.3 From 2d8f30380ab8c706f4e0a8f1aaa22b5886e9ac8a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 22 Jul 2008 09:59:21 -0400 Subject: [PATCH] sanitize __user_walk_fd() et.al. * do not pass nameidata; struct path is all the callers want. * switch to new helpers: user_path_at(dfd, pathname, flags, &path) user_path(pathname, &path) user_lpath(pathname, &path) user_path_dir(pathname, &path) (fail if not a directory) The last 3 are trivial macro wrappers for the first one. * remove nameidata in callers. Signed-off-by: Al Viro --- arch/alpha/kernel/osf_sys.c | 10 ++-- arch/parisc/hpux/sys_hpux.c | 10 ++-- fs/coda/pioctl.c | 14 ++--- fs/compat.c | 20 +++---- fs/inotify_user.c | 22 ++++---- fs/namei.c | 36 ++++++------- fs/namespace.c | 74 +++++++++++++------------- fs/open.c | 124 +++++++++++++++++++++---------------------- fs/stat.c | 32 +++++------ fs/utimes.c | 8 +-- fs/xattr.c | 96 ++++++++++++++++----------------- fs/xfs/linux-2.6/xfs_ioctl.c | 14 +++-- include/linux/namei.h | 13 ++--- 13 files changed, 235 insertions(+), 238 deletions(-) (limited to 'fs') diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index 32ca1b927307..6e943135f0e0 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -253,15 +253,15 @@ do_osf_statfs(struct dentry * dentry, struct osf_statfs __user *buffer, } asmlinkage int -osf_statfs(char __user *path, struct osf_statfs __user *buffer, unsigned long bufsiz) +osf_statfs(char __user *pathname, struct osf_statfs __user *buffer, unsigned long bufsiz) { - struct nameidata nd; + struct path path; int retval; - retval = user_path_walk(path, &nd); + retval = user_path(pathname, &path); if (!retval) { - retval = do_osf_statfs(nd.path.dentry, buffer, bufsiz); - path_put(&nd.path); + retval = do_osf_statfs(path.dentry, buffer, bufsiz); + path_put(&path); } return retval; } diff --git a/arch/parisc/hpux/sys_hpux.c b/arch/parisc/hpux/sys_hpux.c index be255ebb609c..18072e03a019 100644 --- a/arch/parisc/hpux/sys_hpux.c +++ b/arch/parisc/hpux/sys_hpux.c @@ -210,19 +210,19 @@ static int vfs_statfs_hpux(struct dentry *dentry, struct hpux_statfs *buf) } /* hpux statfs */ -asmlinkage long hpux_statfs(const char __user *path, +asmlinkage long hpux_statfs(const char __user *pathname, struct hpux_statfs __user *buf) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (!error) { struct hpux_statfs tmp; - error = vfs_statfs_hpux(nd.path.dentry, &tmp); + error = vfs_statfs_hpux(path.dentry, &tmp); if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) error = -EFAULT; - path_put(&nd.path); + path_put(&path); } return error; } diff --git a/fs/coda/pioctl.c b/fs/coda/pioctl.c index c38a98974fb0..c51365422aa8 100644 --- a/fs/coda/pioctl.c +++ b/fs/coda/pioctl.c @@ -49,7 +49,7 @@ static int coda_ioctl_permission(struct inode *inode, int mask) static int coda_pioctl(struct inode * inode, struct file * filp, unsigned int cmd, unsigned long user_data) { - struct nameidata nd; + struct path path; int error; struct PioctlData data; struct inode *target_inode = NULL; @@ -64,21 +64,21 @@ static int coda_pioctl(struct inode * inode, struct file * filp, * Look up the pathname. Note that the pathname is in * user memory, and namei takes care of this */ - if ( data.follow ) { - error = user_path_walk(data.path, &nd); + if (data.follow) { + error = user_path(data.path, &path); } else { - error = user_path_walk_link(data.path, &nd); + error = user_lpath(data.path, &path); } if ( error ) { return error; } else { - target_inode = nd.path.dentry->d_inode; + target_inode = path.dentry->d_inode; } /* return if it is not a Coda inode */ if ( target_inode->i_sb != inode->i_sb ) { - path_put(&nd.path); + path_put(&path); return -EINVAL; } @@ -87,7 +87,7 @@ static int coda_pioctl(struct inode * inode, struct file * filp, error = venus_pioctl(inode->i_sb, &(cnp->c_fid), cmd, &data); - path_put(&nd.path); + path_put(&path); return error; } diff --git a/fs/compat.c b/fs/compat.c index 106eba28ec5a..c9d1472e65c5 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -234,18 +234,18 @@ static int put_compat_statfs(struct compat_statfs __user *ubuf, struct kstatfs * * The following statfs calls are copies of code from fs/open.c and * should be checked against those from time to time */ -asmlinkage long compat_sys_statfs(const char __user *path, struct compat_statfs __user *buf) +asmlinkage long compat_sys_statfs(const char __user *pathname, struct compat_statfs __user *buf) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (!error) { struct kstatfs tmp; - error = vfs_statfs(nd.path.dentry, &tmp); + error = vfs_statfs(path.dentry, &tmp); if (!error) error = put_compat_statfs(buf, &tmp); - path_put(&nd.path); + path_put(&path); } return error; } @@ -299,21 +299,21 @@ static int put_compat_statfs64(struct compat_statfs64 __user *ubuf, struct kstat return 0; } -asmlinkage long compat_sys_statfs64(const char __user *path, compat_size_t sz, struct compat_statfs64 __user *buf) +asmlinkage long compat_sys_statfs64(const char __user *pathname, compat_size_t sz, struct compat_statfs64 __user *buf) { - struct nameidata nd; + struct path path; int error; if (sz != sizeof(*buf)) return -EINVAL; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (!error) { struct kstatfs tmp; - error = vfs_statfs(nd.path.dentry, &tmp); + error = vfs_statfs(path.dentry, &tmp); if (!error) error = put_compat_statfs64(buf, &tmp); - path_put(&nd.path); + path_put(&path); } return error; } diff --git a/fs/inotify_user.c b/fs/inotify_user.c index 9b99ebf28884..60249429a253 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -354,20 +354,20 @@ static void inotify_dev_event_dequeue(struct inotify_device *dev) } /* - * find_inode - resolve a user-given path to a specific inode and return a nd + * find_inode - resolve a user-given path to a specific inode */ -static int find_inode(const char __user *dirname, struct nameidata *nd, +static int find_inode(const char __user *dirname, struct path *path, unsigned flags) { int error; - error = __user_walk(dirname, flags, nd); + error = user_path_at(AT_FDCWD, dirname, flags, path); if (error) return error; /* you can only watch an inode if you have read permissions on it */ - error = inode_permission(nd->path.dentry->d_inode, MAY_READ); + error = inode_permission(path->dentry->d_inode, MAY_READ); if (error) - path_put(&nd->path); + path_put(path); return error; } @@ -650,11 +650,11 @@ asmlinkage long sys_inotify_init(void) return sys_inotify_init1(0); } -asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask) +asmlinkage long sys_inotify_add_watch(int fd, const char __user *pathname, u32 mask) { struct inode *inode; struct inotify_device *dev; - struct nameidata nd; + struct path path; struct file *filp; int ret, fput_needed; unsigned flags = 0; @@ -674,12 +674,12 @@ asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask) if (mask & IN_ONLYDIR) flags |= LOOKUP_DIRECTORY; - ret = find_inode(path, &nd, flags); + ret = find_inode(pathname, &path, flags); if (unlikely(ret)) goto fput_and_out; - /* inode held in place by reference to nd; dev by fget on fd */ - inode = nd.path.dentry->d_inode; + /* inode held in place by reference to path; dev by fget on fd */ + inode = path.dentry->d_inode; dev = filp->private_data; mutex_lock(&dev->up_mutex); @@ -688,7 +688,7 @@ asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask) ret = create_watch(dev, inode, mask); mutex_unlock(&dev->up_mutex); - path_put(&nd.path); + path_put(&path); fput_and_out: fput_light(filp, fput_needed); return ret; diff --git a/fs/namei.c b/fs/namei.c index 5029b93ebbd5..edb5e973f9b3 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1334,24 +1334,24 @@ struct dentry *lookup_one_noperm(const char *name, struct dentry *base) return __lookup_hash(&this, base, NULL); } -int __user_walk_fd(int dfd, const char __user *name, unsigned flags, - struct nameidata *nd) +int user_path_at(int dfd, const char __user *name, unsigned flags, + struct path *path) { + struct nameidata nd; char *tmp = getname(name); int err = PTR_ERR(tmp); - if (!IS_ERR(tmp)) { - err = do_path_lookup(dfd, tmp, flags, nd); + + BUG_ON(flags & LOOKUP_PARENT); + + err = do_path_lookup(dfd, tmp, flags, &nd); putname(tmp); + if (!err) + *path = nd.path; } return err; } -int __user_walk(const char __user *name, unsigned flags, struct nameidata *nd) -{ - return __user_walk_fd(AT_FDCWD, name, flags, nd); -} - /* * It's inline, so penalty for filesystems that don't use sticky bit is * minimal. @@ -2446,7 +2446,8 @@ asmlinkage long sys_linkat(int olddfd, const char __user *oldname, int flags) { struct dentry *new_dentry; - struct nameidata nd, old_nd; + struct nameidata nd; + struct path old_path; int error; char * to; @@ -2457,16 +2458,16 @@ asmlinkage long sys_linkat(int olddfd, const char __user *oldname, if (IS_ERR(to)) return PTR_ERR(to); - error = __user_walk_fd(olddfd, oldname, - flags & AT_SYMLINK_FOLLOW ? LOOKUP_FOLLOW : 0, - &old_nd); + error = user_path_at(olddfd, oldname, + flags & AT_SYMLINK_FOLLOW ? LOOKUP_FOLLOW : 0, + &old_path); if (error) goto exit; error = do_path_lookup(newdfd, to, LOOKUP_PARENT, &nd); if (error) goto out; error = -EXDEV; - if (old_nd.path.mnt != nd.path.mnt) + if (old_path.mnt != nd.path.mnt) goto out_release; new_dentry = lookup_create(&nd, 0); error = PTR_ERR(new_dentry); @@ -2475,7 +2476,7 @@ asmlinkage long sys_linkat(int olddfd, const char __user *oldname, error = mnt_want_write(nd.path.mnt); if (error) goto out_dput; - error = vfs_link(old_nd.path.dentry, nd.path.dentry->d_inode, new_dentry); + error = vfs_link(old_path.dentry, nd.path.dentry->d_inode, new_dentry); mnt_drop_write(nd.path.mnt); out_dput: dput(new_dentry); @@ -2484,7 +2485,7 @@ out_unlock: out_release: path_put(&nd.path); out: - path_put(&old_nd.path); + path_put(&old_path); exit: putname(to); @@ -2877,8 +2878,7 @@ const struct inode_operations page_symlink_inode_operations = { .put_link = page_put_link, }; -EXPORT_SYMBOL(__user_walk); -EXPORT_SYMBOL(__user_walk_fd); +EXPORT_SYMBOL(user_path_at); EXPORT_SYMBOL(follow_down); EXPORT_SYMBOL(follow_up); EXPORT_SYMBOL(get_write_access); /* binfmt_aout */ diff --git a/fs/namespace.c b/fs/namespace.c index 26380f599534..411728c0c8bb 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1130,27 +1130,27 @@ static int do_umount(struct vfsmount *mnt, int flags) asmlinkage long sys_umount(char __user * name, int flags) { - struct nameidata nd; + struct path path; int retval; - retval = __user_walk(name, LOOKUP_FOLLOW, &nd); + retval = user_path(name, &path); if (retval) goto out; retval = -EINVAL; - if (nd.path.dentry != nd.path.mnt->mnt_root) + if (path.dentry != path.mnt->mnt_root) goto dput_and_out; - if (!check_mnt(nd.path.mnt)) + if (!check_mnt(path.mnt)) goto dput_and_out; retval = -EPERM; if (!capable(CAP_SYS_ADMIN)) goto dput_and_out; - retval = do_umount(nd.path.mnt, flags); + retval = do_umount(path.mnt, flags); dput_and_out: /* we mustn't call path_put() as that would clear mnt_expiry_mark */ - dput(nd.path.dentry); - mntput_no_expire(nd.path.mnt); + dput(path.dentry); + mntput_no_expire(path.mnt); out: return retval; } @@ -2179,28 +2179,26 @@ asmlinkage long sys_pivot_root(const char __user * new_root, const char __user * put_old) { struct vfsmount *tmp; - struct nameidata new_nd, old_nd; - struct path parent_path, root_parent, root; + struct path new, old, parent_path, root_parent, root; int error; if (!capable(CAP_SYS_ADMIN)) return -EPERM; - error = __user_walk(new_root, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, - &new_nd); + error = user_path_dir(new_root, &new); if (error) goto out0; error = -EINVAL; - if (!check_mnt(new_nd.path.mnt)) + if (!check_mnt(new.mnt)) goto out1; - error = __user_walk(put_old, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &old_nd); + error = user_path_dir(put_old, &old); if (error) goto out1; - error = security_sb_pivotroot(&old_nd.path, &new_nd.path); + error = security_sb_pivotroot(&old, &new); if (error) { - path_put(&old_nd.path); + path_put(&old); goto out1; } @@ -2209,69 +2207,69 @@ asmlinkage long sys_pivot_root(const char __user * new_root, path_get(¤t->fs->root); read_unlock(¤t->fs->lock); down_write(&namespace_sem); - mutex_lock(&old_nd.path.dentry->d_inode->i_mutex); + mutex_lock(&old.dentry->d_inode->i_mutex); error = -EINVAL; - if (IS_MNT_SHARED(old_nd.path.mnt) || - IS_MNT_SHARED(new_nd.path.mnt->mnt_parent) || + if (IS_MNT_SHARED(old.mnt) || + IS_MNT_SHARED(new.mnt->mnt_parent) || IS_MNT_SHARED(root.mnt->mnt_parent)) goto out2; if (!check_mnt(root.mnt)) goto out2; error = -ENOENT; - if (IS_DEADDIR(new_nd.path.dentry->d_inode)) + if (IS_DEADDIR(new.dentry->d_inode)) goto out2; - if (d_unhashed(new_nd.path.dentry) && !IS_ROOT(new_nd.path.dentry)) + if (d_unhashed(new.dentry) && !IS_ROOT(new.dentry)) goto out2; - if (d_unhashed(old_nd.path.dentry) && !IS_ROOT(old_nd.path.dentry)) + if (d_unhashed(old.dentry) && !IS_ROOT(old.dentry)) goto out2; error = -EBUSY; - if (new_nd.path.mnt == root.mnt || - old_nd.path.mnt == root.mnt) + if (new.mnt == root.mnt || + old.mnt == root.mnt) goto out2; /* loop, on the same file system */ error = -EINVAL; if (root.mnt->mnt_root != root.dentry) goto out2; /* not a mountpoint */ if (root.mnt->mnt_parent == root.mnt) goto out2; /* not attached */ - if (new_nd.path.mnt->mnt_root != new_nd.path.dentry) + if (new.mnt->mnt_root != new.dentry) goto out2; /* not a mountpoint */ - if (new_nd.path.mnt->mnt_parent == new_nd.path.mnt) + if (new.mnt->mnt_parent == new.mnt) goto out2; /* not attached */ /* make sure we can reach put_old from new_root */ - tmp = old_nd.path.mnt; + tmp = old.mnt; spin_lock(&vfsmount_lock); - if (tmp != new_nd.path.mnt) { + if (tmp != new.mnt) { for (;;) { if (tmp->mnt_parent == tmp) goto out3; /* already mounted on put_old */ - if (tmp->mnt_parent == new_nd.path.mnt) + if (tmp->mnt_parent == new.mnt) break; tmp = tmp->mnt_parent; } - if (!is_subdir(tmp->mnt_mountpoint, new_nd.path.dentry)) + if (!is_subdir(tmp->mnt_mountpoint, new.dentry)) goto out3; - } else if (!is_subdir(old_nd.path.dentry, new_nd.path.dentry)) + } else if (!is_subdir(old.dentry, new.dentry)) goto out3; - detach_mnt(new_nd.path.mnt, &parent_path); + detach_mnt(new.mnt, &parent_path); detach_mnt(root.mnt, &root_parent); /* mount old root on put_old */ - attach_mnt(root.mnt, &old_nd.path); + attach_mnt(root.mnt, &old); /* mount new_root on / */ - attach_mnt(new_nd.path.mnt, &root_parent); + attach_mnt(new.mnt, &root_parent); touch_mnt_namespace(current->nsproxy->mnt_ns); spin_unlock(&vfsmount_lock); - chroot_fs_refs(&root, &new_nd.path); - security_sb_post_pivotroot(&root, &new_nd.path); + chroot_fs_refs(&root, &new); + security_sb_post_pivotroot(&root, &new); error = 0; path_put(&root_parent); path_put(&parent_path); out2: - mutex_unlock(&old_nd.path.dentry->d_inode->i_mutex); + mutex_unlock(&old.dentry->d_inode->i_mutex); up_write(&namespace_sem); path_put(&root); - path_put(&old_nd.path); + path_put(&old); out1: - path_put(&new_nd.path); + path_put(&new); out0: return error; out3: diff --git a/fs/open.c b/fs/open.c index e94266700eda..3fe1a6857c75 100644 --- a/fs/open.c +++ b/fs/open.c @@ -122,37 +122,37 @@ static int vfs_statfs64(struct dentry *dentry, struct statfs64 *buf) return 0; } -asmlinkage long sys_statfs(const char __user * path, struct statfs __user * buf) +asmlinkage long sys_statfs(const char __user *pathname, struct statfs __user * buf) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (!error) { struct statfs tmp; - error = vfs_statfs_native(nd.path.dentry, &tmp); + error = vfs_statfs_native(path.dentry, &tmp); if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) error = -EFAULT; - path_put(&nd.path); + path_put(&path); } return error; } -asmlinkage long sys_statfs64(const char __user *path, size_t sz, struct statfs64 __user *buf) +asmlinkage long sys_statfs64(const char __user *pathname, size_t sz, struct statfs64 __user *buf) { - struct nameidata nd; + struct path path; long error; if (sz != sizeof(*buf)) return -EINVAL; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (!error) { struct statfs64 tmp; - error = vfs_statfs64(nd.path.dentry, &tmp); + error = vfs_statfs64(path.dentry, &tmp); if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) error = -EFAULT; - path_put(&nd.path); + path_put(&path); } return error; } @@ -223,20 +223,20 @@ int do_truncate(struct dentry *dentry, loff_t length, unsigned int time_attrs, return err; } -static long do_sys_truncate(const char __user * path, loff_t length) +static long do_sys_truncate(const char __user *pathname, loff_t length) { - struct nameidata nd; - struct inode * inode; + struct path path; + struct inode *inode; int error; error = -EINVAL; if (length < 0) /* sorry, but loff_t says... */ goto out; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (error) goto out; - inode = nd.path.dentry->d_inode; + inode = path.dentry->d_inode; /* For directories it's -EISDIR, for other non-regulars - -EINVAL */ error = -EISDIR; @@ -247,7 +247,7 @@ static long do_sys_truncate(const char __user * path, loff_t length) if (!S_ISREG(inode->i_mode)) goto dput_and_out; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (error) goto dput_and_out; @@ -274,15 +274,15 @@ static long do_sys_truncate(const char __user * path, loff_t length) error = locks_verify_truncate(inode, NULL, length); if (!error) { DQUOT_INIT(inode); - error = do_truncate(nd.path.dentry, length, 0, NULL); + error = do_truncate(path.dentry, length, 0, NULL); } put_write_and_out: put_write_access(inode); mnt_drop_write_and_out: - mnt_drop_write(nd.path.mnt); + mnt_drop_write(path.mnt); dput_and_out: - path_put(&nd.path); + path_put(&path); out: return error; } @@ -425,7 +425,7 @@ out: */ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) { - struct nameidata nd; + struct path path; struct inode *inode; int old_fsuid, old_fsgid; kernel_cap_t uninitialized_var(old_cap); /* !SECURE_NO_SETUID_FIXUP */ @@ -449,7 +449,7 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) * FIXME: There is a race here against sys_capset. The * capabilities can change yet we will restore the old * value below. We should hold task_capabilities_lock, - * but we cannot because user_path_walk can sleep. + * but we cannot because user_path_at can sleep. */ #endif /* ndef CONFIG_SECURITY_FILE_CAPABILITIES */ if (current->uid) @@ -458,11 +458,11 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) old_cap = cap_set_effective(current->cap_permitted); } - res = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd); + res = user_path_at(dfd, filename, LOOKUP_FOLLOW, &path); if (res) goto out; - inode = nd.path.dentry->d_inode; + inode = path.dentry->d_inode; if ((mode & MAY_EXEC) && S_ISREG(inode->i_mode)) { /* @@ -470,7 +470,7 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) * with the "noexec" flag. */ res = -EACCES; - if (nd.path.mnt->mnt_flags & MNT_NOEXEC) + if (path.mnt->mnt_flags & MNT_NOEXEC) goto out_path_release; } @@ -488,11 +488,11 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) * inherently racy and know that the fs may change * state before we even see this result. */ - if (__mnt_is_readonly(nd.path.mnt)) + if (__mnt_is_readonly(path.mnt)) res = -EROFS; out_path_release: - path_put(&nd.path); + path_put(&path); out: current->fsuid = old_fsuid; current->fsgid = old_fsgid; @@ -510,21 +510,21 @@ asmlinkage long sys_access(const char __user *filename, int mode) asmlinkage long sys_chdir(const char __user * filename) { - struct nameidata nd; + struct path path; int error; - error = __user_walk(filename, LOOKUP_FOLLOW|LOOKUP_DIRECTORY, &nd); + error = user_path_dir(filename, &path); if (error) goto out; - error = inode_permission(nd.path.dentry->d_inode, MAY_EXEC | MAY_ACCESS); + error = inode_permission(path.dentry->d_inode, MAY_EXEC | MAY_ACCESS); if (error) goto dput_and_out; - set_fs_pwd(current->fs, &nd.path); + set_fs_pwd(current->fs, &path); dput_and_out: - path_put(&nd.path); + path_put(&path); out: return error; } @@ -557,14 +557,14 @@ out: asmlinkage long sys_chroot(const char __user * filename) { - struct nameidata nd; + struct path path; int error; - error = __user_walk(filename, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &nd); + error = user_path_dir(filename, &path); if (error) goto out; - error = inode_permission(nd.path.dentry->d_inode, MAY_EXEC | MAY_ACCESS); + error = inode_permission(path.dentry->d_inode, MAY_EXEC | MAY_ACCESS); if (error) goto dput_and_out; @@ -572,10 +572,10 @@ asmlinkage long sys_chroot(const char __user * filename) if (!capable(CAP_SYS_CHROOT)) goto dput_and_out; - set_fs_root(current->fs, &nd.path); + set_fs_root(current->fs, &path); error = 0; dput_and_out: - path_put(&nd.path); + path_put(&path); out: return error; } @@ -617,17 +617,17 @@ out: asmlinkage long sys_fchmodat(int dfd, const char __user *filename, mode_t mode) { - struct nameidata nd; - struct inode * inode; + struct path path; + struct inode *inode; int error; struct iattr newattrs; - error = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd); + error = user_path_at(dfd, filename, LOOKUP_FOLLOW, &path); if (error) goto out; - inode = nd.path.dentry->d_inode; + inode = path.dentry->d_inode; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (error) goto dput_and_out; mutex_lock(&inode->i_mutex); @@ -635,11 +635,11 @@ asmlinkage long sys_fchmodat(int dfd, const char __user *filename, mode = inode->i_mode; newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO); newattrs.ia_valid = ATTR_MODE | ATTR_CTIME; - error = notify_change(nd.path.dentry, &newattrs); + error = notify_change(path.dentry, &newattrs); mutex_unlock(&inode->i_mutex); - mnt_drop_write(nd.path.mnt); + mnt_drop_write(path.mnt); dput_and_out: - path_put(&nd.path); + path_put(&path); out: return error; } @@ -676,19 +676,19 @@ static int chown_common(struct dentry * dentry, uid_t user, gid_t group) asmlinkage long sys_chown(const char __user * filename, uid_t user, gid_t group) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk(filename, &nd); + error = user_path(filename, &path); if (error) goto out; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (error) goto out_release; - error = chown_common(nd.path.dentry, user, group); - mnt_drop_write(nd.path.mnt); + error = chown_common(path.dentry, user, group); + mnt_drop_write(path.mnt); out_release: - path_put(&nd.path); + path_put(&path); out: return error; } @@ -696,7 +696,7 @@ out: asmlinkage long sys_fchownat(int dfd, const char __user *filename, uid_t user, gid_t group, int flag) { - struct nameidata nd; + struct path path; int error = -EINVAL; int follow; @@ -704,35 +704,35 @@ asmlinkage long sys_fchownat(int dfd, const char __user *filename, uid_t user, goto out; follow = (flag & AT_SYMLINK_NOFOLLOW) ? 0 : LOOKUP_FOLLOW; - error = __user_walk_fd(dfd, filename, follow, &nd); + error = user_path_at(dfd, filename, follow, &path); if (error) goto out; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (error) goto out_release; - error = chown_common(nd.path.dentry, user, group); - mnt_drop_write(nd.path.mnt); + error = chown_common(path.dentry, user, group); + mnt_drop_write(path.mnt); out_release: - path_put(&nd.path); + path_put(&path); out: return error; } asmlinkage long sys_lchown(const char __user * filename, uid_t user, gid_t group) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk_link(filename, &nd); + error = user_lpath(filename, &path); if (error) goto out; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (error) goto out_release; - error = chown_common(nd.path.dentry, user, group); - mnt_drop_write(nd.path.mnt); + error = chown_common(path.dentry, user, group); + mnt_drop_write(path.mnt); out_release: - path_put(&nd.path); + path_put(&path); out: return error; } diff --git a/fs/stat.c b/fs/stat.c index 9cf41f719d50..7c46fbeb8b76 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -57,13 +57,13 @@ EXPORT_SYMBOL(vfs_getattr); int vfs_stat_fd(int dfd, char __user *name, struct kstat *stat) { - struct nameidata nd; + struct path path; int error; - error = __user_walk_fd(dfd, name, LOOKUP_FOLLOW, &nd); + error = user_path_at(dfd, name, LOOKUP_FOLLOW, &path); if (!error) { - error = vfs_getattr(nd.path.mnt, nd.path.dentry, stat); - path_put(&nd.path); + error = vfs_getattr(path.mnt, path.dentry, stat); + path_put(&path); } return error; } @@ -77,13 +77,13 @@ EXPORT_SYMBOL(vfs_stat); int vfs_lstat_fd(int dfd, char __user *name, struct kstat *stat) { - struct nameidata nd; + struct path path; int error; - error = __user_walk_fd(dfd, name, 0, &nd); + error = user_path_at(dfd, name, 0, &path); if (!error) { - error = vfs_getattr(nd.path.mnt, nd.path.dentry, stat); - path_put(&nd.path); + error = vfs_getattr(path.mnt, path.dentry, stat); + path_put(&path); } return error; } @@ -291,29 +291,29 @@ asmlinkage long sys_newfstat(unsigned int fd, struct stat __user *statbuf) return error; } -asmlinkage long sys_readlinkat(int dfd, const char __user *path, +asmlinkage long sys_readlinkat(int dfd, const char __user *pathname, char __user *buf, int bufsiz) { - struct nameidata nd; + struct path path; int error; if (bufsiz <= 0) return -EINVAL; - error = __user_walk_fd(dfd, path, 0, &nd); + error = user_path_at(dfd, pathname, 0, &path); if (!error) { - struct inode *inode = nd.path.dentry->d_inode; + struct inode *inode = path.dentry->d_inode; error = -EINVAL; if (inode->i_op && inode->i_op->readlink) { - error = security_inode_readlink(nd.path.dentry); + error = security_inode_readlink(path.dentry); if (!error) { - touch_atime(nd.path.mnt, nd.path.dentry); - error = inode->i_op->readlink(nd.path.dentry, + touch_atime(path.mnt, path.dentry); + error = inode->i_op->readlink(path.dentry, buf, bufsiz); } } - path_put(&nd.path); + path_put(&path); } return error; } diff --git a/fs/utimes.c b/fs/utimes.c index dc28b7826259..6929e3e91d05 100644 --- a/fs/utimes.c +++ b/fs/utimes.c @@ -152,18 +152,18 @@ long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags error = utimes_common(&file->f_path, times); fput(file); } else { - struct nameidata nd; + struct path path; int lookup_flags = 0; if (!(flags & AT_SYMLINK_NOFOLLOW)) lookup_flags |= LOOKUP_FOLLOW; - error = __user_walk_fd(dfd, filename, lookup_flags, &nd); + error = user_path_at(dfd, filename, lookup_flags, &path); if (error) goto out; - error = utimes_common(&nd.path, times); - path_put(&nd.path); + error = utimes_common(&path, times); + path_put(&path); } out: diff --git a/fs/xattr.c b/fs/xattr.c index b96222e05ba0..468377e66531 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -252,40 +252,40 @@ setxattr(struct dentry *d, const char __user *name, const void __user *value, } asmlinkage long -sys_setxattr(const char __user *path, const char __user *name, +sys_setxattr(const char __user *pathname, const char __user *name, const void __user *value, size_t size, int flags) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (error) return error; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (!error) { - error = setxattr(nd.path.dentry, name, value, size, flags); - mnt_drop_write(nd.path.mnt); + error = setxattr(path.dentry, name, value, size, flags); + mnt_drop_write(path.mnt); } - path_put(&nd.path); + path_put(&path); return error; } asmlinkage long -sys_lsetxattr(const char __user *path, const char __user *name, +sys_lsetxattr(const char __user *pathname, const char __user *name, const void __user *value, size_t size, int flags) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk_link(path, &nd); + error = user_lpath(pathname, &path); if (error) return error; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (!error) { - error = setxattr(nd.path.dentry, name, value, size, flags); - mnt_drop_write(nd.path.mnt); + error = setxattr(path.dentry, name, value, size, flags); + mnt_drop_write(path.mnt); } - path_put(&nd.path); + path_put(&path); return error; } @@ -350,32 +350,32 @@ getxattr(struct dentry *d, const char __user *name, void __user *value, } asmlinkage ssize_t -sys_getxattr(const char __user *path, const char __user *name, +sys_getxattr(const char __user *pathname, const char __user *name, void __user *value, size_t size) { - struct nameidata nd; + struct path path; ssize_t error; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (error) return error; - error = getxattr(nd.path.dentry, name, value, size); - path_put(&nd.path); + error = getxattr(path.dentry, name, value, size); + path_put(&path); return error; } asmlinkage ssize_t -sys_lgetxattr(const char __user *path, const char __user *name, void __user *value, +sys_lgetxattr(const char __user *pathname, const char __user *name, void __user *value, size_t size) { - struct nameidata nd; + struct path path; ssize_t error; - error = user_path_walk_link(path, &nd); + error = user_lpath(pathname, &path); if (error) return error; - error = getxattr(nd.path.dentry, name, value, size); - path_put(&nd.path); + error = getxattr(path.dentry, name, value, size); + path_put(&path); return error; } @@ -425,30 +425,30 @@ listxattr(struct dentry *d, char __user *list, size_t size) } asmlinkage ssize_t -sys_listxattr(const char __user *path, char __user *list, size_t size) +sys_listxattr(const char __user *pathname, char __user *list, size_t size) { - struct nameidata nd; + struct path path; ssize_t error; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (error) return error; - error = listxattr(nd.path.dentry, list, size); - path_put(&nd.path); + error = listxattr(path.dentry, list, size); + path_put(&path); return error; } asmlinkage ssize_t -sys_llistxattr(const char __user *path, char __user *list, size_t size) +sys_llistxattr(const char __user *pathname, char __user *list, size_t size) { - struct nameidata nd; + struct path path; ssize_t error; - error = user_path_walk_link(path, &nd); + error = user_lpath(pathname, &path); if (error) return error; - error = listxattr(nd.path.dentry, list, size); - path_put(&nd.path); + error = listxattr(path.dentry, list, size); + path_put(&path); return error; } @@ -486,38 +486,38 @@ removexattr(struct dentry *d, const char __user *name) } asmlinkage long -sys_removexattr(const char __user *path, const char __user *name) +sys_removexattr(const char __user *pathname, const char __user *name) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (error) return error; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (!error) { - error = removexattr(nd.path.dentry, name); - mnt_drop_write(nd.path.mnt); + error = removexattr(path.dentry, name); + mnt_drop_write(path.mnt); } - path_put(&nd.path); + path_put(&path); return error; } asmlinkage long -sys_lremovexattr(const char __user *path, const char __user *name) +sys_lremovexattr(const char __user *pathname, const char __user *name) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk_link(path, &nd); + error = user_lpath(pathname, &path); if (error) return error; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (!error) { - error = removexattr(nd.path.dentry, name); - mnt_drop_write(nd.path.mnt); + error = removexattr(path.dentry, name); + mnt_drop_write(path.mnt); } - path_put(&nd.path); + path_put(&path); return error; } diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index a42ba9d71156..01939ba2d8de 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -84,17 +84,15 @@ xfs_find_handle( switch (cmd) { case XFS_IOC_PATH_TO_FSHANDLE: case XFS_IOC_PATH_TO_HANDLE: { - struct nameidata nd; - int error; - - error = user_path_walk_link((const char __user *)hreq.path, &nd); + struct path path; + int error = user_lpath((const char __user *)hreq.path, &path); if (error) return error; - ASSERT(nd.path.dentry); - ASSERT(nd.path.dentry->d_inode); - inode = igrab(nd.path.dentry->d_inode); - path_put(&nd.path); + ASSERT(path.dentry); + ASSERT(path.dentry->d_inode); + inode = igrab(path.dentry->d_inode); + path_put(&path); break; } diff --git a/include/linux/namei.h b/include/linux/namei.h index 60e35a02f6cb..00888ff69504 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -54,12 +54,13 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND}; #define LOOKUP_OPEN (0x0100) #define LOOKUP_CREATE (0x0200) -extern int __user_walk(const char __user *, unsigned, struct nameidata *); -extern int __user_walk_fd(int dfd, const char __user *, unsigned, struct nameidata *); -#define user_path_walk(name,nd) \ - __user_walk_fd(AT_FDCWD, name, LOOKUP_FOLLOW, nd) -#define user_path_walk_link(name,nd) \ - __user_walk_fd(AT_FDCWD, name, 0, nd) +extern int user_path_at(int, const char __user *, unsigned, struct path *); + +#define user_path(name, path) user_path_at(AT_FDCWD, name, LOOKUP_FOLLOW, path) +#define user_lpath(name, path) user_path_at(AT_FDCWD, name, 0, path) +#define user_path_dir(name, path) \ + user_path_at(AT_FDCWD, name, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, path) + extern int path_lookup(const char *, unsigned, struct nameidata *); extern int vfs_path_lookup(struct dentry *, struct vfsmount *, const char *, unsigned int, struct nameidata *); -- cgit v1.2.3 From 2ad94ae654f5eb72fd3260b706aea645cf4a7791 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 21 Jul 2008 09:32:51 -0400 Subject: [PATCH] new (local) helper: user_path_parent() Preparation to untangling intents mess: reduce the number of do_path_lookup() callers. Signed-off-by: Al Viro --- fs/namei.c | 137 +++++++++++++++++++++++++------------------------------------ 1 file changed, 57 insertions(+), 80 deletions(-) (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index edb5e973f9b3..38ceb6e06eba 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1352,6 +1352,24 @@ int user_path_at(int dfd, const char __user *name, unsigned flags, return err; } +static int user_path_parent(int dfd, const char __user *path, + struct nameidata *nd, char **name) +{ + char *s = getname(path); + int error; + + if (IS_ERR(s)) + return PTR_ERR(s); + + error = do_path_lookup(dfd, s, LOOKUP_PARENT, nd); + if (error) + putname(s); + else + *name = s; + + return error; +} + /* * It's inline, so penalty for filesystems that don't use sticky bit is * minimal. @@ -1989,20 +2007,18 @@ static int may_mknod(mode_t mode) asmlinkage long sys_mknodat(int dfd, const char __user *filename, int mode, unsigned dev) { - int error = 0; - char * tmp; - struct dentry * dentry; + int error; + char *tmp; + struct dentry *dentry; struct nameidata nd; if (S_ISDIR(mode)) return -EPERM; - tmp = getname(filename); - if (IS_ERR(tmp)) - return PTR_ERR(tmp); - error = do_path_lookup(dfd, tmp, LOOKUP_PARENT, &nd); + error = user_path_parent(dfd, filename, &nd, &tmp); if (error) - goto out; + return error; + dentry = lookup_create(&nd, 0); if (IS_ERR(dentry)) { error = PTR_ERR(dentry); @@ -2034,7 +2050,6 @@ out_dput: out_unlock: mutex_unlock(&nd.path.dentry->d_inode->i_mutex); path_put(&nd.path); -out: putname(tmp); return error; @@ -2074,14 +2089,10 @@ asmlinkage long sys_mkdirat(int dfd, const char __user *pathname, int mode) struct dentry *dentry; struct nameidata nd; - tmp = getname(pathname); - error = PTR_ERR(tmp); - if (IS_ERR(tmp)) + error = user_path_parent(dfd, pathname, &nd, &tmp); + if (error) goto out_err; - error = do_path_lookup(dfd, tmp, LOOKUP_PARENT, &nd); - if (error) - goto out; dentry = lookup_create(&nd, 1); error = PTR_ERR(dentry); if (IS_ERR(dentry)) @@ -2099,7 +2110,6 @@ out_dput: out_unlock: mutex_unlock(&nd.path.dentry->d_inode->i_mutex); path_put(&nd.path); -out: putname(tmp); out_err: return error; @@ -2177,13 +2187,9 @@ static long do_rmdir(int dfd, const char __user *pathname) struct dentry *dentry; struct nameidata nd; - name = getname(pathname); - if(IS_ERR(name)) - return PTR_ERR(name); - - error = do_path_lookup(dfd, name, LOOKUP_PARENT, &nd); + error = user_path_parent(dfd, pathname, &nd, &name); if (error) - goto exit; + return error; switch(nd.last_type) { case LAST_DOTDOT: @@ -2212,7 +2218,6 @@ exit2: mutex_unlock(&nd.path.dentry->d_inode->i_mutex); exit1: path_put(&nd.path); -exit: putname(name); return error; } @@ -2261,19 +2266,16 @@ int vfs_unlink(struct inode *dir, struct dentry *dentry) */ static long do_unlinkat(int dfd, const char __user *pathname) { - int error = 0; - char * name; + int error; + char *name; struct dentry *dentry; struct nameidata nd; struct inode *inode = NULL; - name = getname(pathname); - if(IS_ERR(name)) - return PTR_ERR(name); - - error = do_path_lookup(dfd, name, LOOKUP_PARENT, &nd); + error = user_path_parent(dfd, pathname, &nd, &name); if (error) - goto exit; + return error; + error = -EISDIR; if (nd.last_type != LAST_NORM) goto exit1; @@ -2300,7 +2302,6 @@ static long do_unlinkat(int dfd, const char __user *pathname) iput(inode); /* truncate the inode here */ exit1: path_put(&nd.path); -exit: putname(name); return error; @@ -2350,23 +2351,20 @@ int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname) asmlinkage long sys_symlinkat(const char __user *oldname, int newdfd, const char __user *newname) { - int error = 0; - char * from; - char * to; + int error; + char *from; + char *to; struct dentry *dentry; struct nameidata nd; from = getname(oldname); - if(IS_ERR(from)) + if (IS_ERR(from)) return PTR_ERR(from); - to = getname(newname); - error = PTR_ERR(to); - if (IS_ERR(to)) - goto out_putname; - error = do_path_lookup(newdfd, to, LOOKUP_PARENT, &nd); + error = user_path_parent(newdfd, newname, &nd, &to); if (error) - goto out; + goto out_putname; + dentry = lookup_create(&nd, 0); error = PTR_ERR(dentry); if (IS_ERR(dentry)) @@ -2382,7 +2380,6 @@ out_dput: out_unlock: mutex_unlock(&nd.path.dentry->d_inode->i_mutex); path_put(&nd.path); -out: putname(to); out_putname: putname(from); @@ -2449,21 +2446,18 @@ asmlinkage long sys_linkat(int olddfd, const char __user *oldname, struct nameidata nd; struct path old_path; int error; - char * to; + char *to; if ((flags & ~AT_SYMLINK_FOLLOW) != 0) return -EINVAL; - to = getname(newname); - if (IS_ERR(to)) - return PTR_ERR(to); - error = user_path_at(olddfd, oldname, flags & AT_SYMLINK_FOLLOW ? LOOKUP_FOLLOW : 0, &old_path); if (error) - goto exit; - error = do_path_lookup(newdfd, to, LOOKUP_PARENT, &nd); + return error; + + error = user_path_parent(newdfd, newname, &nd, &to); if (error) goto out; error = -EXDEV; @@ -2484,10 +2478,9 @@ out_unlock: mutex_unlock(&nd.path.dentry->d_inode->i_mutex); out_release: path_put(&nd.path); + putname(to); out: path_put(&old_path); -exit: - putname(to); return error; } @@ -2643,20 +2636,22 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry, return error; } -static int do_rename(int olddfd, const char *oldname, - int newdfd, const char *newname) +asmlinkage long sys_renameat(int olddfd, const char __user *oldname, + int newdfd, const char __user *newname) { - int error = 0; - struct dentry * old_dir, * new_dir; - struct dentry * old_dentry, *new_dentry; - struct dentry * trap; + struct dentry *old_dir, *new_dir; + struct dentry *old_dentry, *new_dentry; + struct dentry *trap; struct nameidata oldnd, newnd; + char *from; + char *to; + int error; - error = do_path_lookup(olddfd, oldname, LOOKUP_PARENT, &oldnd); + error = user_path_parent(olddfd, oldname, &oldnd, &from); if (error) goto exit; - error = do_path_lookup(newdfd, newname, LOOKUP_PARENT, &newnd); + error = user_path_parent(newdfd, newname, &newnd, &to); if (error) goto exit1; @@ -2718,29 +2713,11 @@ exit3: unlock_rename(new_dir, old_dir); exit2: path_put(&newnd.path); + putname(to); exit1: path_put(&oldnd.path); -exit: - return error; -} - -asmlinkage long sys_renameat(int olddfd, const char __user *oldname, - int newdfd, const char __user *newname) -{ - int error; - char * from; - char * to; - - from = getname(oldname); - if(IS_ERR(from)) - return PTR_ERR(from); - to = getname(newname); - error = PTR_ERR(to); - if (!IS_ERR(to)) { - error = do_rename(olddfd, from, newdfd, to); - putname(to); - } putname(from); +exit: return error; } -- cgit v1.2.3 From a569c711f63995ad80c23918525111e0cdb0bc73 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 23 Jul 2008 14:42:05 -0400 Subject: [PATCH] don't pass nameidata to gfs2_lookupi() Signed-off-by: Al Viro --- fs/gfs2/inode.c | 6 +++--- fs/gfs2/inode.h | 2 +- fs/gfs2/ops_export.c | 2 +- fs/gfs2/ops_inode.c | 4 ++-- fs/gfs2/super.c | 2 +- 5 files changed, 8 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 6da0ab355b8a..8b0806a32948 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -448,7 +448,7 @@ struct inode *gfs2_lookup_simple(struct inode *dip, const char *name) struct qstr qstr; struct inode *inode; gfs2_str2qstr(&qstr, name); - inode = gfs2_lookupi(dip, &qstr, 1, NULL); + inode = gfs2_lookupi(dip, &qstr, 1); /* gfs2_lookupi has inconsistent callers: vfs * related routines expect NULL for no entry found, * gfs2_lookup_simple callers expect ENOENT @@ -477,7 +477,7 @@ struct inode *gfs2_lookup_simple(struct inode *dip, const char *name) */ struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, - int is_root, struct nameidata *nd) + int is_root) { struct super_block *sb = dir->i_sb; struct gfs2_inode *dip = GFS2_I(dir); @@ -1173,7 +1173,7 @@ int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to) break; } - tmp = gfs2_lookupi(dir, &dotdot, 1, NULL); + tmp = gfs2_lookupi(dir, &dotdot, 1); if (IS_ERR(tmp)) { error = PTR_ERR(tmp); break; diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index 6074c2506f75..58f9607d6a86 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -83,7 +83,7 @@ int gfs2_inode_refresh(struct gfs2_inode *ip); int gfs2_dinode_dealloc(struct gfs2_inode *inode); int gfs2_change_nlink(struct gfs2_inode *ip, int diff); struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, - int is_root, struct nameidata *nd); + int is_root); struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, unsigned int mode, dev_t dev); int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, diff --git a/fs/gfs2/ops_export.c b/fs/gfs2/ops_export.c index 990d9f4bc463..9cda8536530c 100644 --- a/fs/gfs2/ops_export.c +++ b/fs/gfs2/ops_export.c @@ -134,7 +134,7 @@ static struct dentry *gfs2_get_parent(struct dentry *child) struct dentry *dentry; gfs2_str2qstr(&dotdot, ".."); - inode = gfs2_lookupi(child->d_inode, &dotdot, 1, NULL); + inode = gfs2_lookupi(child->d_inode, &dotdot, 1); if (!inode) return ERR_PTR(-ENOENT); diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index 4e982532f085..e2c62f73a778 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -74,7 +74,7 @@ static int gfs2_create(struct inode *dir, struct dentry *dentry, return PTR_ERR(inode); } - inode = gfs2_lookupi(dir, &dentry->d_name, 0, nd); + inode = gfs2_lookupi(dir, &dentry->d_name, 0); if (inode) { if (!IS_ERR(inode)) { gfs2_holder_uninit(ghs); @@ -109,7 +109,7 @@ static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry, dentry->d_op = &gfs2_dops; - inode = gfs2_lookupi(dir, &dentry->d_name, 0, nd); + inode = gfs2_lookupi(dir, &dentry->d_name, 0); if (inode && IS_ERR(inode)) return ERR_CAST(inode); diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 63a8a902d9db..ca831991cbc2 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -389,7 +389,7 @@ int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh) break; INIT_LIST_HEAD(&jd->extent_list); - jd->jd_inode = gfs2_lookupi(sdp->sd_jindex, &name, 1, NULL); + jd->jd_inode = gfs2_lookupi(sdp->sd_jindex, &name, 1); if (!jd->jd_inode || IS_ERR(jd->jd_inode)) { if (!jd->jd_inode) error = -ENOENT; -- cgit v1.2.3 From 58ec42b061bf5dad8fa0370a19966cfd96eaf80c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 23 Jul 2008 14:45:55 -0400 Subject: [PATCH] don't pass nameidata to __ncp_lookup_validate() Signed-off-by: Al Viro --- fs/ncpfs/dir.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/ncpfs/dir.c b/fs/ncpfs/dir.c index 011ef0b6d2d4..07e9715b8658 100644 --- a/fs/ncpfs/dir.c +++ b/fs/ncpfs/dir.c @@ -266,7 +266,7 @@ leave_me:; static int -__ncp_lookup_validate(struct dentry * dentry, struct nameidata *nd) +__ncp_lookup_validate(struct dentry *dentry) { struct ncp_server *server; struct dentry *parent; @@ -340,7 +340,7 @@ ncp_lookup_validate(struct dentry * dentry, struct nameidata *nd) { int res; lock_kernel(); - res = __ncp_lookup_validate(dentry, nd); + res = __ncp_lookup_validate(dentry); unlock_kernel(); return res; } -- cgit v1.2.3 From 3c333937ee3be114b181c4861188cfe8f6a59697 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Fri, 25 Jul 2008 22:32:13 -0400 Subject: [PATCH] dup3 fix Al Viro notice one cornercase that the new dup3() code. The dup2() function, as a special case, handles dup-ing to the same file descriptor. In this case the current dup3() code does nothing at all. I.e., it ingnores the flags parameter. This shouldn't happen, the close-on-exec flag should be set if requested. In case the O_CLOEXEC bit in the flags parameter is not set the dup3() function should behave in this respect identical to dup2(). This means dup3(fd, fd, 0) should not actively reset the c-o-e flag. The patch below implements this minor change. [AV: credits to Artur Grabowski for bringing that up as potential subtle point in dup2() behaviour] Signed-off-by: Ulrich Drepper Signed-off-by: Al Viro --- fs/fcntl.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/fcntl.c b/fs/fcntl.c index 9679fcbdeaa0..ce12a6885115 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -139,8 +139,13 @@ asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags) if (!(file = fcheck(oldfd))) goto out_unlock; err = newfd; - if (newfd == oldfd) + if (unlikely(newfd == oldfd)) { + if (flags & O_CLOEXEC) { + fdt = files_fdtable(files); + FD_SET(newfd, fdt->close_on_exec); + } goto out_unlock; + } err = -EBADF; if (newfd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) goto out_unlock; -- cgit v1.2.3 From 516e0cc5646f377ab80fcc2ee639892eccb99853 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 26 Jul 2008 00:39:17 -0400 Subject: [PATCH] f_count may wrap around make it atomic_long_t; while we are at it, get rid of useless checks in affs, hfs and hpfs - ->open() always has it equal to 1, ->release() - to 0. Signed-off-by: Al Viro --- drivers/net/ppp_generic.c | 6 +++--- fs/affs/file.c | 4 ---- fs/aio.c | 6 +++--- fs/file_table.c | 10 +++++----- fs/hfs/inode.c | 4 ---- fs/hfsplus/inode.c | 4 ---- include/linux/fs.h | 6 +++--- include/net/af_unix.h | 2 +- net/sched/sch_atm.c | 4 ++-- net/unix/af_unix.c | 2 +- net/unix/garbage.c | 18 +++++++++--------- 11 files changed, 27 insertions(+), 39 deletions(-) (limited to 'fs') diff --git a/drivers/net/ppp_generic.c b/drivers/net/ppp_generic.c index 739b3ab7bccc..ddccc074a76a 100644 --- a/drivers/net/ppp_generic.c +++ b/drivers/net/ppp_generic.c @@ -581,12 +581,12 @@ static long ppp_ioctl(struct file *file, unsigned int cmd, unsigned long arg) if (file == ppp->owner) ppp_shutdown_interface(ppp); } - if (atomic_read(&file->f_count) <= 2) { + if (atomic_long_read(&file->f_count) <= 2) { ppp_release(NULL, file); err = 0; } else - printk(KERN_DEBUG "PPPIOCDETACH file->f_count=%d\n", - atomic_read(&file->f_count)); + printk(KERN_DEBUG "PPPIOCDETACH file->f_count=%ld\n", + atomic_long_read(&file->f_count)); unlock_kernel(); return err; } diff --git a/fs/affs/file.c b/fs/affs/file.c index 6eac7bdeec94..1377b1240b6e 100644 --- a/fs/affs/file.c +++ b/fs/affs/file.c @@ -46,8 +46,6 @@ const struct inode_operations affs_file_inode_operations = { static int affs_file_open(struct inode *inode, struct file *filp) { - if (atomic_read(&filp->f_count) != 1) - return 0; pr_debug("AFFS: open(%lu,%d)\n", inode->i_ino, atomic_read(&AFFS_I(inode)->i_opencnt)); atomic_inc(&AFFS_I(inode)->i_opencnt); @@ -57,8 +55,6 @@ affs_file_open(struct inode *inode, struct file *filp) static int affs_file_release(struct inode *inode, struct file *filp) { - if (atomic_read(&filp->f_count) != 0) - return 0; pr_debug("AFFS: release(%lu, %d)\n", inode->i_ino, atomic_read(&AFFS_I(inode)->i_opencnt)); diff --git a/fs/aio.c b/fs/aio.c index 0051fd94b44e..f658441d5666 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -512,8 +512,8 @@ static void aio_fput_routine(struct work_struct *data) */ static int __aio_put_req(struct kioctx *ctx, struct kiocb *req) { - dprintk(KERN_DEBUG "aio_put(%p): f_count=%d\n", - req, atomic_read(&req->ki_filp->f_count)); + dprintk(KERN_DEBUG "aio_put(%p): f_count=%ld\n", + req, atomic_long_read(&req->ki_filp->f_count)); assert_spin_locked(&ctx->ctx_lock); @@ -528,7 +528,7 @@ static int __aio_put_req(struct kioctx *ctx, struct kiocb *req) /* Must be done under the lock to serialise against cancellation. * Call this aio_fput as it duplicates fput via the fput_work. */ - if (unlikely(atomic_dec_and_test(&req->ki_filp->f_count))) { + if (unlikely(atomic_long_dec_and_test(&req->ki_filp->f_count))) { get_ioctx(ctx); spin_lock(&fput_lock); list_add(&req->ki_list, &fput_head); diff --git a/fs/file_table.c b/fs/file_table.c index 83084225b4c3..f45a4493f9e7 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -120,7 +120,7 @@ struct file *get_empty_filp(void) tsk = current; INIT_LIST_HEAD(&f->f_u.fu_list); - atomic_set(&f->f_count, 1); + atomic_long_set(&f->f_count, 1); rwlock_init(&f->f_owner.lock); f->f_uid = tsk->fsuid; f->f_gid = tsk->fsgid; @@ -219,7 +219,7 @@ EXPORT_SYMBOL(init_file); void fput(struct file *file) { - if (atomic_dec_and_test(&file->f_count)) + if (atomic_long_dec_and_test(&file->f_count)) __fput(file); } @@ -294,7 +294,7 @@ struct file *fget(unsigned int fd) rcu_read_lock(); file = fcheck_files(files, fd); if (file) { - if (!atomic_inc_not_zero(&file->f_count)) { + if (!atomic_long_inc_not_zero(&file->f_count)) { /* File object ref couldn't be taken */ rcu_read_unlock(); return NULL; @@ -326,7 +326,7 @@ struct file *fget_light(unsigned int fd, int *fput_needed) rcu_read_lock(); file = fcheck_files(files, fd); if (file) { - if (atomic_inc_not_zero(&file->f_count)) + if (atomic_long_inc_not_zero(&file->f_count)) *fput_needed = 1; else /* Didn't get the reference, someone's freed */ @@ -341,7 +341,7 @@ struct file *fget_light(unsigned int fd, int *fput_needed) void put_filp(struct file *file) { - if (atomic_dec_and_test(&file->f_count)) { + if (atomic_long_dec_and_test(&file->f_count)) { security_file_free(file); file_kill(file); file_free(file); diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index aa73f3fd5dd9..7e19835efa2e 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -522,8 +522,6 @@ static int hfs_file_open(struct inode *inode, struct file *file) { if (HFS_IS_RSRC(inode)) inode = HFS_I(inode)->rsrc_inode; - if (atomic_read(&file->f_count) != 1) - return 0; atomic_inc(&HFS_I(inode)->opencnt); return 0; } @@ -534,8 +532,6 @@ static int hfs_file_release(struct inode *inode, struct file *file) if (HFS_IS_RSRC(inode)) inode = HFS_I(inode)->rsrc_inode; - if (atomic_read(&file->f_count) != 0) - return 0; if (atomic_dec_and_test(&HFS_I(inode)->opencnt)) { mutex_lock(&inode->i_mutex); hfs_file_truncate(inode); diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index d4014e3044d2..b085d64a2b67 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -254,8 +254,6 @@ static int hfsplus_file_open(struct inode *inode, struct file *file) { if (HFSPLUS_IS_RSRC(inode)) inode = HFSPLUS_I(inode).rsrc_inode; - if (atomic_read(&file->f_count) != 1) - return 0; atomic_inc(&HFSPLUS_I(inode).opencnt); return 0; } @@ -266,8 +264,6 @@ static int hfsplus_file_release(struct inode *inode, struct file *file) if (HFSPLUS_IS_RSRC(inode)) inode = HFSPLUS_I(inode).rsrc_inode; - if (atomic_read(&file->f_count) != 0) - return 0; if (atomic_dec_and_test(&HFSPLUS_I(inode).opencnt)) { mutex_lock(&inode->i_mutex); hfsplus_file_truncate(inode); diff --git a/include/linux/fs.h b/include/linux/fs.h index 9d2de4cadabd..7676fa1c20ae 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -795,7 +795,7 @@ struct file { #define f_dentry f_path.dentry #define f_vfsmnt f_path.mnt const struct file_operations *f_op; - atomic_t f_count; + atomic_long_t f_count; unsigned int f_flags; mode_t f_mode; loff_t f_pos; @@ -824,8 +824,8 @@ extern spinlock_t files_lock; #define file_list_lock() spin_lock(&files_lock); #define file_list_unlock() spin_unlock(&files_lock); -#define get_file(x) atomic_inc(&(x)->f_count) -#define file_count(x) atomic_read(&(x)->f_count) +#define get_file(x) atomic_long_inc(&(x)->f_count) +#define file_count(x) atomic_long_read(&(x)->f_count) #ifdef CONFIG_DEBUG_WRITECOUNT static inline void file_take_write(struct file *f) diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 2dfa96b0575e..7dd29b7e461d 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -51,7 +51,7 @@ struct unix_sock { struct sock *peer; struct sock *other; struct list_head link; - atomic_t inflight; + atomic_long_t inflight; spinlock_t lock; unsigned int gc_candidate : 1; wait_queue_head_t peer_wait; diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index 04faa835be17..6b517b9dac5b 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -162,7 +162,7 @@ static void atm_tc_put(struct Qdisc *sch, unsigned long cl) qdisc_destroy(flow->q); tcf_destroy_chain(&flow->filter_list); if (flow->sock) { - pr_debug("atm_tc_put: f_count %d\n", + pr_debug("atm_tc_put: f_count %ld\n", file_count(flow->sock->file)); flow->vcc->pop = flow->old_pop; sockfd_put(flow->sock); @@ -259,7 +259,7 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent, sock = sockfd_lookup(fd, &error); if (!sock) return error; /* f_count++ */ - pr_debug("atm_tc_change: f_count %d\n", file_count(sock->file)); + pr_debug("atm_tc_change: f_count %ld\n", file_count(sock->file)); if (sock->ops->family != PF_ATMSVC && sock->ops->family != PF_ATMPVC) { error = -EPROTOTYPE; goto err_out; diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 70ceb1604ad8..6e7fec74bdb3 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -603,7 +603,7 @@ static struct sock * unix_create1(struct net *net, struct socket *sock) u->dentry = NULL; u->mnt = NULL; spin_lock_init(&u->lock); - atomic_set(&u->inflight, 0); + atomic_long_set(&u->inflight, 0); INIT_LIST_HEAD(&u->link); mutex_init(&u->readlock); /* single task reading lock */ init_waitqueue_head(&u->peer_wait); diff --git a/net/unix/garbage.c b/net/unix/garbage.c index ebdff3d877a1..2a27b84f740b 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -127,7 +127,7 @@ void unix_inflight(struct file *fp) if(s) { struct unix_sock *u = unix_sk(s); spin_lock(&unix_gc_lock); - if (atomic_inc_return(&u->inflight) == 1) { + if (atomic_long_inc_return(&u->inflight) == 1) { BUG_ON(!list_empty(&u->link)); list_add_tail(&u->link, &gc_inflight_list); } else { @@ -145,7 +145,7 @@ void unix_notinflight(struct file *fp) struct unix_sock *u = unix_sk(s); spin_lock(&unix_gc_lock); BUG_ON(list_empty(&u->link)); - if (atomic_dec_and_test(&u->inflight)) + if (atomic_long_dec_and_test(&u->inflight)) list_del_init(&u->link); unix_tot_inflight--; spin_unlock(&unix_gc_lock); @@ -237,17 +237,17 @@ static void scan_children(struct sock *x, void (*func)(struct unix_sock *), static void dec_inflight(struct unix_sock *usk) { - atomic_dec(&usk->inflight); + atomic_long_dec(&usk->inflight); } static void inc_inflight(struct unix_sock *usk) { - atomic_inc(&usk->inflight); + atomic_long_inc(&usk->inflight); } static void inc_inflight_move_tail(struct unix_sock *u) { - atomic_inc(&u->inflight); + atomic_long_inc(&u->inflight); /* * If this is still a candidate, move it to the end of the * list, so that it's checked even if it was already passed @@ -288,11 +288,11 @@ void unix_gc(void) * before the detach without atomicity guarantees. */ list_for_each_entry_safe(u, next, &gc_inflight_list, link) { - int total_refs; - int inflight_refs; + long total_refs; + long inflight_refs; total_refs = file_count(u->sk.sk_socket->file); - inflight_refs = atomic_read(&u->inflight); + inflight_refs = atomic_long_read(&u->inflight); BUG_ON(inflight_refs < 1); BUG_ON(total_refs < inflight_refs); @@ -324,7 +324,7 @@ void unix_gc(void) /* Move cursor to after the current position. */ list_move(&cursor, &u->link); - if (atomic_read(&u->inflight) > 0) { + if (atomic_long_read(&u->inflight) > 0) { list_move_tail(&u->link, &gc_inflight_list); u->gc_candidate = 0; scan_children(&u->sk, inc_inflight_move_tail, NULL); -- cgit v1.2.3 From 964bd183624c03680796b63b4ab97ee3905a806a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 26 Jul 2008 03:33:14 -0400 Subject: [PATCH] get rid of __user_path_lookup_open Signed-off-by: Al Viro --- fs/exec.c | 14 ++++++++++---- fs/namei.c | 13 ------------- include/linux/namei.h | 1 - 3 files changed, 10 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/exec.c b/fs/exec.c index eca58c29eded..9696bbf0f0b1 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -106,11 +106,17 @@ static inline void put_binfmt(struct linux_binfmt * fmt) */ asmlinkage long sys_uselib(const char __user * library) { - struct file * file; + struct file *file; struct nameidata nd; - int error; - - error = __user_path_lookup_open(library, LOOKUP_FOLLOW, &nd, FMODE_READ|FMODE_EXEC); + char *tmp = getname(library); + int error = PTR_ERR(tmp); + + if (!IS_ERR(tmp)) { + error = path_lookup_open(AT_FDCWD, tmp, + LOOKUP_FOLLOW, &nd, + FMODE_READ|FMODE_EXEC); + putname(tmp); + } if (error) goto out; diff --git a/fs/namei.c b/fs/namei.c index 38ceb6e06eba..a7b0a0b80128 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1193,19 +1193,6 @@ static int path_lookup_create(int dfd, const char *name, nd, open_flags, create_mode); } -int __user_path_lookup_open(const char __user *name, unsigned int lookup_flags, - struct nameidata *nd, int open_flags) -{ - char *tmp = getname(name); - int err = PTR_ERR(tmp); - - if (!IS_ERR(tmp)) { - err = __path_lookup_intent_open(AT_FDCWD, tmp, lookup_flags, nd, open_flags, 0); - putname(tmp); - } - return err; -} - static struct dentry *__lookup_hash(struct qstr *name, struct dentry *base, struct nameidata *nd) { diff --git a/include/linux/namei.h b/include/linux/namei.h index 00888ff69504..68f8c3203c89 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -65,7 +65,6 @@ extern int path_lookup(const char *, unsigned, struct nameidata *); extern int vfs_path_lookup(struct dentry *, struct vfsmount *, const char *, unsigned int, struct nameidata *); -extern int __user_path_lookup_open(const char __user *, unsigned lookup_flags, struct nameidata *nd, int open_flags); extern int path_lookup_open(int dfd, const char *name, unsigned lookup_flags, struct nameidata *, int open_flags); extern struct file *lookup_instantiate_filp(struct nameidata *nd, struct dentry *dentry, int (*open)(struct inode *, struct file *)); -- cgit v1.2.3 From 3f8206d496e9e9495afb1d4e70d29712b4d403c9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 26 Jul 2008 03:46:43 -0400 Subject: [PATCH] get rid of indirect users of namei.h fs.h needs path.h, not namei.h; nfs_fs.h doesn't need it at all. Several places in the tree needed direct include. Signed-off-by: Al Viro --- fs/nfsd/nfsctl.c | 1 + fs/ubifs/file.c | 1 + include/linux/fs.h | 2 +- include/linux/nfs_fs.h | 1 - kernel/cgroup.c | 1 + 5 files changed, 4 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 1955a2702e60..c53e65f8f3a2 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c index 005a3b854d96..8565e586e533 100644 --- a/fs/ubifs/file.c +++ b/fs/ubifs/file.c @@ -53,6 +53,7 @@ #include "ubifs.h" #include +#include static int read_block(struct inode *inode, void *addr, unsigned int block, struct ubifs_data_node *dn) diff --git a/include/linux/fs.h b/include/linux/fs.h index 7676fa1c20ae..8252b045e624 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -279,7 +279,7 @@ extern int dir_notify_enable; #include #include #include -#include +#include #include #include #include diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index f08f9ca602af..78a5922a2f11 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -42,7 +42,6 @@ #include #include #include -#include #include #include #include diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 89bd6fb7894f..657f8f8d93a5 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -45,6 +45,7 @@ #include #include #include +#include #include -- cgit v1.2.3 From 6c5d0512a091480c9f981162227fdb1c9d70e555 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 26 Jul 2008 13:38:19 -0400 Subject: [PATCH] get rid of corner case in dup3() entirely Since Ulrich is OK with getting rid of dup3(fd, fd, flags) completely, to hell the damn thing goes. Corner case for dup2() is handled in sys_dup2() (complete with -EBADF if dup2(fd, fd) is called with fd that is not open), the rest is done in dup3(). Signed-off-by: Al Viro --- fs/fcntl.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/fcntl.c b/fs/fcntl.c index ce12a6885115..3deec9887089 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -135,18 +135,12 @@ asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags) if ((flags & ~O_CLOEXEC) != 0) return -EINVAL; + if (unlikely(oldfd == newfd)) + return -EINVAL; + spin_lock(&files->file_lock); if (!(file = fcheck(oldfd))) goto out_unlock; - err = newfd; - if (unlikely(newfd == oldfd)) { - if (flags & O_CLOEXEC) { - fdt = files_fdtable(files); - FD_SET(newfd, fdt->close_on_exec); - } - goto out_unlock; - } - err = -EBADF; if (newfd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) goto out_unlock; get_file(file); /* We are now finished with oldfd */ @@ -194,6 +188,14 @@ out_fput: asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd) { + if (unlikely(newfd == oldfd)) { /* corner case */ + struct files_struct *files = current->files; + rcu_read_lock(); + if (!fcheck_files(files, oldfd)) + oldfd = -EBADF; + rcu_read_unlock(); + return oldfd; + } return sys_dup3(oldfd, newfd, 0); } -- cgit v1.2.3 From 4e1e018ecc6f7bfd10fc75b3ff9715cc8164e0a2 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 26 Jul 2008 16:01:20 -0400 Subject: [PATCH] fix RLIM_NOFILE handling * dup2() should return -EBADF on exceeded sysctl_nr_open * dup() should *not* return -EINVAL even if you have rlimit set to 0; it should get -EMFILE instead. Check for orig_start exceeding rlimit taken to sys_fcntl(). Failing expand_files() in dup{2,3}() now gets -EMFILE remapped to -EBADF. Consequently, remaining checks for rlimit are taken to expand_files(). Signed-off-by: Al Viro --- fs/fcntl.c | 18 ++++++------------ fs/file.c | 9 +++++++++ fs/open.c | 9 --------- 3 files changed, 15 insertions(+), 21 deletions(-) (limited to 'fs') diff --git a/fs/fcntl.c b/fs/fcntl.c index 3deec9887089..61d625136813 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -64,11 +64,6 @@ static int locate_fd(unsigned int orig_start, int cloexec) struct fdtable *fdt; spin_lock(&files->file_lock); - - error = -EINVAL; - if (orig_start >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) - goto out; - repeat: fdt = files_fdtable(files); /* @@ -83,10 +78,6 @@ repeat: if (start < fdt->max_fds) newfd = find_next_zero_bit(fdt->open_fds->fds_bits, fdt->max_fds, start); - - error = -EMFILE; - if (newfd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) - goto out; error = expand_files(files, newfd); if (error < 0) @@ -141,13 +132,14 @@ asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags) spin_lock(&files->file_lock); if (!(file = fcheck(oldfd))) goto out_unlock; - if (newfd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) - goto out_unlock; get_file(file); /* We are now finished with oldfd */ err = expand_files(files, newfd); - if (err < 0) + if (unlikely(err < 0)) { + if (err == -EMFILE) + err = -EBADF; goto out_fput; + } /* To avoid races with open() and dup(), we will mark the fd as * in-use in the open-file bitmap throughout the entire dup2() @@ -328,6 +320,8 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg, switch (cmd) { case F_DUPFD: case F_DUPFD_CLOEXEC: + if (arg >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) + break; get_file(filp); err = dupfd(filp, arg, cmd == F_DUPFD_CLOEXEC); break; diff --git a/fs/file.c b/fs/file.c index 7b3887e054d0..d8773b19fe47 100644 --- a/fs/file.c +++ b/fs/file.c @@ -250,9 +250,18 @@ int expand_files(struct files_struct *files, int nr) struct fdtable *fdt; fdt = files_fdtable(files); + + /* + * N.B. For clone tasks sharing a files structure, this test + * will limit the total number of files that can be opened. + */ + if (nr >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) + return -EMFILE; + /* Do we need to expand? */ if (nr < fdt->max_fds) return 0; + /* Can we expand? */ if (nr >= sysctl_nr_open) return -EMFILE; diff --git a/fs/open.c b/fs/open.c index 3fe1a6857c75..52647be277a2 100644 --- a/fs/open.c +++ b/fs/open.c @@ -972,7 +972,6 @@ int get_unused_fd_flags(int flags) int fd, error; struct fdtable *fdt; - error = -EMFILE; spin_lock(&files->file_lock); repeat: @@ -980,13 +979,6 @@ repeat: fd = find_next_zero_bit(fdt->open_fds->fds_bits, fdt->max_fds, files->next_fd); - /* - * N.B. For clone tasks sharing a files structure, this test - * will limit the total number of files that can be opened. - */ - if (fd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) - goto out; - /* Do we need to expand the fd array or fd set? */ error = expand_files(files, fd); if (error < 0) @@ -997,7 +989,6 @@ repeat: * If we needed to expand the fs array we * might have blocked - try again. */ - error = -EMFILE; goto repeat; } -- cgit v1.2.3 From b2d002dba5a8a4c0c3ec96fd1ff3c9def6bd71a1 Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Sat, 26 Jul 2008 15:22:27 -0700 Subject: task IO accounting: correctly account threads IO statistics Oleg Nesterov points out that we should check that the task is still alive before we iterate over the threads. This patch includes a fixup for this. Also simplify do_io_accounting() implementation. Signed-off-by: Andrea Righi Signed-off-by: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/base.c | 56 ++++++++++++++++++++++---------------------------------- 1 file changed, 22 insertions(+), 34 deletions(-) (limited to 'fs') diff --git a/fs/proc/base.c b/fs/proc/base.c index 81bce6791bfc..d744aa3c9f74 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -2406,35 +2406,18 @@ static int do_io_accounting(struct task_struct *task, char *buffer, int whole) u64 rchar, wchar, syscr, syscw; struct task_io_accounting ioac; - if (!whole) { - rchar = task->rchar; - wchar = task->wchar; - syscr = task->syscr; - syscw = task->syscw; - memcpy(&ioac, &task->ioac, sizeof(ioac)); - } else { - unsigned long flags; - struct task_struct *t = task; - rchar = wchar = syscr = syscw = 0; - memset(&ioac, 0, sizeof(ioac)); + rchar = task->rchar; + wchar = task->wchar; + syscr = task->syscr; + syscw = task->syscw; + memcpy(&ioac, &task->ioac, sizeof(ioac)); - rcu_read_lock(); - do { - rchar += t->rchar; - wchar += t->wchar; - syscr += t->syscr; - syscw += t->syscw; - - ioac.read_bytes += t->ioac.read_bytes; - ioac.write_bytes += t->ioac.write_bytes; - ioac.cancelled_write_bytes += - t->ioac.cancelled_write_bytes; - t = next_thread(t); - } while (t != task); - rcu_read_unlock(); + if (whole) { + unsigned long flags; if (lock_task_sighand(task, &flags)) { struct signal_struct *sig = task->signal; + struct task_struct *t = task; rchar += sig->rchar; wchar += sig->wchar; @@ -2445,11 +2428,20 @@ static int do_io_accounting(struct task_struct *task, char *buffer, int whole) ioac.write_bytes += sig->ioac.write_bytes; ioac.cancelled_write_bytes += sig->ioac.cancelled_write_bytes; - + while_each_thread(task, t) { + rchar += t->rchar; + wchar += t->wchar; + syscr += t->syscr; + syscw += t->syscw; + + ioac.read_bytes += t->ioac.read_bytes; + ioac.write_bytes += t->ioac.write_bytes; + ioac.cancelled_write_bytes += + t->ioac.cancelled_write_bytes; + } unlock_task_sighand(task, &flags); } } - return sprintf(buffer, "rchar: %llu\n" "wchar: %llu\n" @@ -2458,13 +2450,9 @@ static int do_io_accounting(struct task_struct *task, char *buffer, int whole) "read_bytes: %llu\n" "write_bytes: %llu\n" "cancelled_write_bytes: %llu\n", - (unsigned long long)rchar, - (unsigned long long)wchar, - (unsigned long long)syscr, - (unsigned long long)syscw, - (unsigned long long)ioac.read_bytes, - (unsigned long long)ioac.write_bytes, - (unsigned long long)ioac.cancelled_write_bytes); + rchar, wchar, syscr, syscw, + ioac.read_bytes, ioac.write_bytes, + ioac.cancelled_write_bytes); } static int proc_tid_io_accounting(struct task_struct *task, char *buffer) -- cgit v1.2.3 From 5995477ab7f3522c497c9c4a1c55373e9d655574 Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Sun, 27 Jul 2008 17:29:15 +0200 Subject: task IO accounting: improve code readability Put all i/o statistics in struct proc_io_accounting and use inline functions to initialize and increment statistics, removing a lot of single variable assignments. This also reduces the kernel size as following (with CONFIG_TASK_XACCT=y and CONFIG_TASK_IO_ACCOUNTING=y). text data bss dec hex filename 11651 0 0 11651 2d83 kernel/exit.o.before 11619 0 0 11619 2d63 kernel/exit.o.after 10886 132 136 11154 2b92 kernel/fork.o.before 10758 132 136 11026 2b12 kernel/fork.o.after 3082029 807968 4818600 8708597 84e1f5 vmlinux.o.before 3081869 807968 4818600 8708437 84e155 vmlinux.o.after Signed-off-by: Andrea Righi Acked-by: Oleg Nesterov Signed-off-by: Linus Torvalds --- fs/proc/base.c | 57 ++++++++++------------------------ include/linux/sched.h | 19 ++++-------- include/linux/task_io_accounting.h | 27 ++++++++++++++-- include/linux/task_io_accounting_ops.h | 56 +++++++++++++++++++++++++++------ kernel/exit.c | 30 ++---------------- kernel/fork.c | 15 ++------- kernel/tsacct.c | 14 ++++----- 7 files changed, 104 insertions(+), 114 deletions(-) (limited to 'fs') diff --git a/fs/proc/base.c b/fs/proc/base.c index e74308bdabd3..3d94906c7aa8 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -53,6 +53,7 @@ #include #include #include +#include #include #include #include @@ -2402,44 +2403,17 @@ static int proc_base_fill_cache(struct file *filp, void *dirent, #ifdef CONFIG_TASK_IO_ACCOUNTING static int do_io_accounting(struct task_struct *task, char *buffer, int whole) { - u64 rchar, wchar, syscr, syscw; - struct task_io_accounting ioac; - - rchar = task->rchar; - wchar = task->wchar; - syscr = task->syscr; - syscw = task->syscw; - memcpy(&ioac, &task->ioac, sizeof(ioac)); - - if (whole) { - unsigned long flags; - - if (lock_task_sighand(task, &flags)) { - struct signal_struct *sig = task->signal; - struct task_struct *t = task; - - rchar += sig->rchar; - wchar += sig->wchar; - syscr += sig->syscr; - syscw += sig->syscw; - - ioac.read_bytes += sig->ioac.read_bytes; - ioac.write_bytes += sig->ioac.write_bytes; - ioac.cancelled_write_bytes += - sig->ioac.cancelled_write_bytes; - while_each_thread(task, t) { - rchar += t->rchar; - wchar += t->wchar; - syscr += t->syscr; - syscw += t->syscw; - - ioac.read_bytes += t->ioac.read_bytes; - ioac.write_bytes += t->ioac.write_bytes; - ioac.cancelled_write_bytes += - t->ioac.cancelled_write_bytes; - } - unlock_task_sighand(task, &flags); - } + struct proc_io_accounting acct = task->ioac; + unsigned long flags; + + if (whole && lock_task_sighand(task, &flags)) { + struct task_struct *t = task; + + task_io_accounting_add(&acct, &task->signal->ioac); + while_each_thread(task, t) + task_io_accounting_add(&acct, &t->ioac); + + unlock_task_sighand(task, &flags); } return sprintf(buffer, "rchar: %llu\n" @@ -2449,9 +2423,10 @@ static int do_io_accounting(struct task_struct *task, char *buffer, int whole) "read_bytes: %llu\n" "write_bytes: %llu\n" "cancelled_write_bytes: %llu\n", - rchar, wchar, syscr, syscw, - ioac.read_bytes, ioac.write_bytes, - ioac.cancelled_write_bytes); + acct.chr.rchar, acct.chr.wchar, + acct.chr.syscr, acct.chr.syscw, + acct.blk.read_bytes, acct.blk.write_bytes, + acct.blk.cancelled_write_bytes); } static int proc_tid_io_accounting(struct task_struct *task, char *buffer) diff --git a/include/linux/sched.h b/include/linux/sched.h index f59318a0099b..034c1ca6b332 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -505,10 +505,7 @@ struct signal_struct { unsigned long nvcsw, nivcsw, cnvcsw, cnivcsw; unsigned long min_flt, maj_flt, cmin_flt, cmaj_flt; unsigned long inblock, oublock, cinblock, coublock; -#ifdef CONFIG_TASK_XACCT - u64 rchar, wchar, syscr, syscw; -#endif - struct task_io_accounting ioac; + struct proc_io_accounting ioac; /* * Cumulative ns of scheduled CPU time for dead threads in the @@ -1256,11 +1253,7 @@ struct task_struct { unsigned long ptrace_message; siginfo_t *last_siginfo; /* For ptrace use. */ -#ifdef CONFIG_TASK_XACCT -/* i/o counters(bytes read/written, #syscalls */ - u64 rchar, wchar, syscr, syscw; -#endif - struct task_io_accounting ioac; + struct proc_io_accounting ioac; #if defined(CONFIG_TASK_XACCT) u64 acct_rss_mem1; /* accumulated rss usage */ u64 acct_vm_mem1; /* accumulated virtual memory usage */ @@ -2190,22 +2183,22 @@ extern long sched_group_rt_period(struct task_group *tg); #ifdef CONFIG_TASK_XACCT static inline void add_rchar(struct task_struct *tsk, ssize_t amt) { - tsk->rchar += amt; + tsk->ioac.chr.rchar += amt; } static inline void add_wchar(struct task_struct *tsk, ssize_t amt) { - tsk->wchar += amt; + tsk->ioac.chr.wchar += amt; } static inline void inc_syscr(struct task_struct *tsk) { - tsk->syscr++; + tsk->ioac.chr.syscr++; } static inline void inc_syscw(struct task_struct *tsk) { - tsk->syscw++; + tsk->ioac.chr.syscw++; } #else static inline void add_rchar(struct task_struct *tsk, ssize_t amt) diff --git a/include/linux/task_io_accounting.h b/include/linux/task_io_accounting.h index 44d00e9cceea..165390f8b936 100644 --- a/include/linux/task_io_accounting.h +++ b/include/linux/task_io_accounting.h @@ -1,5 +1,5 @@ /* - * task_io_accounting: a structure which is used for recording a single task's + * proc_io_accounting: a structure which is used for recording a single task's * IO statistics. * * Don't include this header file directly - it is designed to be dragged in via @@ -8,6 +8,22 @@ * Blame akpm@osdl.org for all this. */ +#ifdef CONFIG_TASK_XACCT +struct task_chr_io_accounting { + /* bytes read */ + u64 rchar; + /* bytes written */ + u64 wchar; + /* # of read syscalls */ + u64 syscr; + /* # of write syscalls */ + u64 syscw; +}; +#else /* CONFIG_TASK_XACCT */ +struct task_chr_io_accounting { +}; +#endif /* CONFIG_TASK_XACCT */ + #ifdef CONFIG_TASK_IO_ACCOUNTING struct task_io_accounting { /* @@ -31,7 +47,12 @@ struct task_io_accounting { */ u64 cancelled_write_bytes; }; -#else +#else /* CONFIG_TASK_IO_ACCOUNTING */ struct task_io_accounting { }; -#endif +#endif /* CONFIG_TASK_IO_ACCOUNTING */ + +struct proc_io_accounting { + struct task_chr_io_accounting chr; + struct task_io_accounting blk; +}; diff --git a/include/linux/task_io_accounting_ops.h b/include/linux/task_io_accounting_ops.h index ff46c6fad79d..e6f958ebe97f 100644 --- a/include/linux/task_io_accounting_ops.h +++ b/include/linux/task_io_accounting_ops.h @@ -9,7 +9,7 @@ #ifdef CONFIG_TASK_IO_ACCOUNTING static inline void task_io_account_read(size_t bytes) { - current->ioac.read_bytes += bytes; + current->ioac.blk.read_bytes += bytes; } /* @@ -18,12 +18,12 @@ static inline void task_io_account_read(size_t bytes) */ static inline unsigned long task_io_get_inblock(const struct task_struct *p) { - return p->ioac.read_bytes >> 9; + return p->ioac.blk.read_bytes >> 9; } static inline void task_io_account_write(size_t bytes) { - current->ioac.write_bytes += bytes; + current->ioac.blk.write_bytes += bytes; } /* @@ -32,17 +32,25 @@ static inline void task_io_account_write(size_t bytes) */ static inline unsigned long task_io_get_oublock(const struct task_struct *p) { - return p->ioac.write_bytes >> 9; + return p->ioac.blk.write_bytes >> 9; } static inline void task_io_account_cancelled_write(size_t bytes) { - current->ioac.cancelled_write_bytes += bytes; + current->ioac.blk.cancelled_write_bytes += bytes; } -static inline void task_io_accounting_init(struct task_struct *tsk) +static inline void task_io_accounting_init(struct proc_io_accounting *ioac) { - memset(&tsk->ioac, 0, sizeof(tsk->ioac)); + memset(ioac, 0, sizeof(*ioac)); +} + +static inline void task_blk_io_accounting_add(struct proc_io_accounting *dst, + struct proc_io_accounting *src) +{ + dst->blk.read_bytes += src->blk.read_bytes; + dst->blk.write_bytes += src->blk.write_bytes; + dst->blk.cancelled_write_bytes += src->blk.cancelled_write_bytes; } #else @@ -69,9 +77,37 @@ static inline void task_io_account_cancelled_write(size_t bytes) { } -static inline void task_io_accounting_init(struct task_struct *tsk) +static inline void task_io_accounting_init(struct proc_io_accounting *ioac) +{ +} + +static inline void task_blk_io_accounting_add(struct proc_io_accounting *dst, + struct proc_io_accounting *src) { } -#endif /* CONFIG_TASK_IO_ACCOUNTING */ -#endif /* __TASK_IO_ACCOUNTING_OPS_INCLUDED */ +#endif /* CONFIG_TASK_IO_ACCOUNTING */ + +#ifdef CONFIG_TASK_XACCT +static inline void task_chr_io_accounting_add(struct proc_io_accounting *dst, + struct proc_io_accounting *src) +{ + dst->chr.rchar += src->chr.rchar; + dst->chr.wchar += src->chr.wchar; + dst->chr.syscr += src->chr.syscr; + dst->chr.syscw += src->chr.syscw; +} +#else +static inline void task_chr_io_accounting_add(struct proc_io_accounting *dst, + struct proc_io_accounting *src) +{ +} +#endif /* CONFIG_TASK_XACCT */ + +static inline void task_io_accounting_add(struct proc_io_accounting *dst, + struct proc_io_accounting *src) +{ + task_chr_io_accounting_add(dst, src); + task_blk_io_accounting_add(dst, src); +} +#endif /* __TASK_IO_ACCOUNTING_OPS_INCLUDED */ diff --git a/kernel/exit.c b/kernel/exit.c index 0caf590548a0..eb4d6470d1d0 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -121,18 +121,7 @@ static void __exit_signal(struct task_struct *tsk) sig->nivcsw += tsk->nivcsw; sig->inblock += task_io_get_inblock(tsk); sig->oublock += task_io_get_oublock(tsk); -#ifdef CONFIG_TASK_XACCT - sig->rchar += tsk->rchar; - sig->wchar += tsk->wchar; - sig->syscr += tsk->syscr; - sig->syscw += tsk->syscw; -#endif /* CONFIG_TASK_XACCT */ -#ifdef CONFIG_TASK_IO_ACCOUNTING - sig->ioac.read_bytes += tsk->ioac.read_bytes; - sig->ioac.write_bytes += tsk->ioac.write_bytes; - sig->ioac.cancelled_write_bytes += - tsk->ioac.cancelled_write_bytes; -#endif /* CONFIG_TASK_IO_ACCOUNTING */ + task_io_accounting_add(&sig->ioac, &tsk->ioac); sig->sum_sched_runtime += tsk->se.sum_exec_runtime; sig = NULL; /* Marker for below. */ } @@ -1363,21 +1352,8 @@ static int wait_task_zombie(struct task_struct *p, int options, psig->coublock += task_io_get_oublock(p) + sig->oublock + sig->coublock; -#ifdef CONFIG_TASK_XACCT - psig->rchar += p->rchar + sig->rchar; - psig->wchar += p->wchar + sig->wchar; - psig->syscr += p->syscr + sig->syscr; - psig->syscw += p->syscw + sig->syscw; -#endif /* CONFIG_TASK_XACCT */ -#ifdef CONFIG_TASK_IO_ACCOUNTING - psig->ioac.read_bytes += - p->ioac.read_bytes + sig->ioac.read_bytes; - psig->ioac.write_bytes += - p->ioac.write_bytes + sig->ioac.write_bytes; - psig->ioac.cancelled_write_bytes += - p->ioac.cancelled_write_bytes + - sig->ioac.cancelled_write_bytes; -#endif /* CONFIG_TASK_IO_ACCOUNTING */ + task_io_accounting_add(&psig->ioac, &p->ioac); + task_io_accounting_add(&psig->ioac, &sig->ioac); spin_unlock_irq(&p->parent->sighand->siglock); } diff --git a/kernel/fork.c b/kernel/fork.c index 5e050c1317c4..8214ba7c8bb1 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -806,12 +806,7 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) sig->nvcsw = sig->nivcsw = sig->cnvcsw = sig->cnivcsw = 0; sig->min_flt = sig->maj_flt = sig->cmin_flt = sig->cmaj_flt = 0; sig->inblock = sig->oublock = sig->cinblock = sig->coublock = 0; -#ifdef CONFIG_TASK_XACCT - sig->rchar = sig->wchar = sig->syscr = sig->syscw = 0; -#endif -#ifdef CONFIG_TASK_IO_ACCOUNTING - memset(&sig->ioac, 0, sizeof(sig->ioac)); -#endif + task_io_accounting_init(&sig->ioac); sig->sum_sched_runtime = 0; INIT_LIST_HEAD(&sig->cpu_timers[0]); INIT_LIST_HEAD(&sig->cpu_timers[1]); @@ -994,13 +989,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->last_switch_timestamp = 0; #endif -#ifdef CONFIG_TASK_XACCT - p->rchar = 0; /* I/O counter: bytes read */ - p->wchar = 0; /* I/O counter: bytes written */ - p->syscr = 0; /* I/O counter: read syscalls */ - p->syscw = 0; /* I/O counter: write syscalls */ -#endif - task_io_accounting_init(p); + task_io_accounting_init(&p->ioac); acct_clear_integrals(p); p->it_virt_expires = cputime_zero; diff --git a/kernel/tsacct.c b/kernel/tsacct.c index 3da47ccdc5e5..f9cd2561689c 100644 --- a/kernel/tsacct.c +++ b/kernel/tsacct.c @@ -94,14 +94,14 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p) stats->hiwater_vm = mm->hiwater_vm * PAGE_SIZE / KB; mmput(mm); } - stats->read_char = p->rchar; - stats->write_char = p->wchar; - stats->read_syscalls = p->syscr; - stats->write_syscalls = p->syscw; + stats->read_char = p->ioac.chr.rchar; + stats->write_char = p->ioac.chr.wchar; + stats->read_syscalls = p->ioac.chr.syscr; + stats->write_syscalls = p->ioac.chr.syscw; #ifdef CONFIG_TASK_IO_ACCOUNTING - stats->read_bytes = p->ioac.read_bytes; - stats->write_bytes = p->ioac.write_bytes; - stats->cancelled_write_bytes = p->ioac.cancelled_write_bytes; + stats->read_bytes = p->ioac.blk.read_bytes; + stats->write_bytes = p->ioac.blk.write_bytes; + stats->cancelled_write_bytes = p->ioac.blk.cancelled_write_bytes; #else stats->read_bytes = 0; stats->write_bytes = 0; -- cgit v1.2.3 From 31c9446993f412ecb7875e30bba4bc7f216ae016 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Thu, 17 Jul 2008 13:21:55 +0200 Subject: nfs_remount oops when rebooting + possible fix Jeff, Trond, The commit 48b605f83c920d8daa50e43fc2c7f718e04c7bfa (NFS: implement option checking when remounting NFS filesystems (resend)) generate an Oops on my platform when rebooting while its root FS on an NFS share (NFSv3, TCP) : Unmounting local filesystems...done. Unable to handle kernel NULL pointer dereference at virtual address 00000000 pgd = c3d00000 [00000000] *pgd=a3d72031, *pte=00000000, *ppte=00000000 Internal error: Oops: 17 [#1] Modules linked in: cpufreq_powersave cpufreq_ondemand cpufreq_userspace cpufreq_conservative ext3 jbd sd_mod pata_pcmcia libata scsi_mod pcmcia loop firmware_class pxafb cfbcopyarea cfbimgblt cfbfillrect pxa2xx_cs pxa2xx_core pcmcia_core snd_pxa2xx_ac97 snd_ac97_codec ac97_bus snd_pxa2xx_pcm snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd isp116x_hcd soundcore rtc_sa1100 snd_page_alloc pxa25x_udc usbcore rtc_ds1307 rtc_core CPU: 0 Not tainted (2.6.26-03414-g33af79d-dirty #15) PC is at nfs_remount+0x40/0x264 LR is at do_remount_sb+0x158/0x194 pc : [] lr : [] psr: 60000013 sp : c2dd1e70 ip : c2dd1e98 fp : c2dd1e94 r10: 00000040 r9 : c3d17000 r8 : c3c3fc40 r7 : 00000000 r6 : 00000000 r5 : c3d2b200 r4 : 00000000 r3 : 00000003 r2 : 00000000 r1 : c2dd1e9c r0 : c3c3fc00 Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user Control: 0000397f Table: a3d00000 DAC: 00000015 Process mount (pid: 1462, stack limit = 0xc2dd0270) Stack: (0xc2dd1e70 to 0xc2dd2000) 1e60: 00000000 c3c3fc00 00000000 00000000 1e80: c3c3fc40 c3d17000 c2dd1ebc c2dd1e98 c0076c40 c00bbf20 c01c61e4 00000001 1ea0: c2dd1ebc 00000001 c3c3fc00 c2dd1ef0 c2dd1ee4 c2dd1ec0 c008c6d8 c0076af4 1ec0: 00000021 00000040 c2dd1ef0 c3d77000 c3eaa000 00000000 c2dd1f6c c2dd1ee8 1ee0: c008d1bc c008c5f8 00000000 c2dd0000 c3c0c320 c3805b38 c002064c 0001f820 1f00: 0001f810 00000001 00000001 00000000 c2dd0000 00000000 c2dd1f34 c2dd1f28 1f20: c005ead8 c005e6f8 c2dd1f44 c2dd1f38 c005eaf8 c005ead0 c2dd1f6c c2dd1f48 1f40: c008ae3c 00000000 c3d77000 0001f810 c0ed0021 c0020ca8 c2dd0000 00000000 1f60: c2dd1fa4 c2dd1f70 c008d2d4 c008d0bc 00000000 0001f810 c2dd1f9c c3eaa000 1f80: c3d17000 00000000 00000000 be8b6aa8 be8b6ad0 00000015 00000000 c2dd1fa8 1fa0: c0020b00 c008d254 00000000 be8b6aa8 0001f810 0001f820 0001f830 c0ed0021 1fc0: 00000000 be8b6aa8 be8b6ad0 00000015 00000000 be8b6ad0 0001f810 be8b6aa8 1fe0: 0001f810 be8b6964 0000aab8 40125124 60000010 0001f810 00000000 00000000 Backtrace: [] (nfs_remount+0x0/0x264) from [] (do_remount_sb+0x158/0x194) r9:c3d17000 r8:c3c3fc40 r7:00000000 r6:00000000 r5:c3c3fc00 r4:00000000 [] (do_remount_sb+0x0/0x194) from [] (do_remount+0xec/0x118) r6:c2dd1ef0 r5:c3c3fc00 r4:00000001 [] (do_remount+0x0/0x118) from [] (do_mount+0x10c/0x198) [] (do_mount+0x0/0x198) from [] (sys_mount+0x8c/0xd4) [] (sys_mount+0x0/0xd4) from [] (ret_fast_syscall+0x0/0x2c) r7:00000015 r6:be8b6ad0 r5:be8b6aa8 r4:00000000 Code: 0a000086 ea000006 e3530003 8a000004 (e5923000) ---[ end trace 55e1b689cf8c8a6a ]--- ------------[ cut here ]------------ WARNING: at kernel/exit.c:966 do_exit+0x3c/0x628() Modules linked in: cpufreq_powersave cpufreq_ondemand cpufreq_userspace cpufreq_conservative ext3 jbd sd_mod pata_pcmcia libata scsi_mod pcmcia loop firmware_class pxafb cfbcopyarea cfbimgblt cfbfillrect pxa2xx_cs pxa2xx_core pcmcia_core snd_pxa2xx_ac97 snd_ac97_codec ac97_bus snd_pxa2xx_pcm snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd isp116x_hcd soundcore rtc_sa1100 snd_page_alloc pxa25x_udc usbcore rtc_ds1307 rtc_core [] (dump_stack+0x0/0x14) from [] (warn_on_slowpath+0x4c/0x68) [] (warn_on_slowpath+0x0/0x68) from [] (do_exit+0x3c/0x628) r6:0000000b r5:c3c3dc80 r4:c2dd0000 [] (do_exit+0x0/0x628) from [] (die+0x2b0/0x30c) [] (die+0x0/0x30c) from [] (__do_kernel_fault+0x6c/0x80) [] (__do_kernel_fault+0x0/0x80) from [] (do_page_fault+0x210/0x230) r7:c3fa7118 r6:c3c3dc80 r5:c3d166a8 r4:00010000 [] (do_page_fault+0x0/0x230) from [] (do_DataAbort+0x3c/0xa0) [] (do_DataAbort+0x0/0xa0) from [] (__dabt_svc+0x4c/0x60) Exception stack(0xc2dd1e28 to 0xc2dd1e70) 1e20: c3c3fc00 c2dd1e9c 00000000 00000003 00000000 c3d2b200 1e40: 00000000 00000000 c3c3fc40 c3d17000 00000040 c2dd1e94 c2dd1e98 c2dd1e70 1e60: c0076c40 c00bbf54 60000013 ffffffff r8:c3c3fc40 r7:00000000 r6:00000000 r5:c2dd1e5c r4:ffffffff [] (nfs_remount+0x0/0x264) from [] (do_remount_sb+0x158/0x194) r9:c3d17000 r8:c3c3fc40 r7:00000000 r6:00000000 r5:c3c3fc00 r4:00000000 [] (do_remount_sb+0x0/0x194) from [] (do_remount+0xec/0x118) r6:c2dd1ef0 r5:c3c3fc00 r4:00000001 [] (do_remount+0x0/0x118) from [] (do_mount+0x10c/0x198) [] (do_mount+0x0/0x198) from [] (sys_mount+0x8c/0xd4) [] (sys_mount+0x0/0xd4) from [] (ret_fast_syscall+0x0/0x2c) r7:00000015 r6:be8b6ad0 r5:be8b6aa8 r4:00000000 ---[ end trace 55e1b689cf8c8a6a ]--- /etc/rc6.d/S60umountroot: line 17: 1462 Segmentation fault mount $MOUNT_FORCE_OPT -n -o remount,ro -t dummytype dummydev / 2> /dev/null The new super.c:nfs_remount function doesn't check the validity of the options/options4 pointers. Unfortunately, this seems to happend. The obvious patch seems to check the pointers, and not to do anything if the happend to be NULL. Tested on an XScale PXA255 system, latest git. Regards, M. Signed-off-by: Marc Zyngier Signed-off-by: Trond Myklebust --- fs/nfs/super.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 1b94e3650f5c..9abcd2b329f7 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -1718,9 +1718,9 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data) * ones were explicitly specified. Fall back to legacy behavior and * just return success. */ - if ((nfsvers == 4 && options4->version == 1) || - (nfsvers <= 3 && options->version >= 1 && - options->version <= 6)) + if ((nfsvers == 4 && (!options4 || options4->version == 1)) || + (nfsvers <= 3 && (!options || (options->version >= 1 && + options->version <= 6)))) return 0; data = kzalloc(sizeof(*data), GFP_KERNEL); -- cgit v1.2.3 From 744d18dbfae07482ea461701b0aaec3a75ec9224 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 27 Jul 2008 18:03:19 -0400 Subject: NFS: Ensure we call nfs_sb_deactive() after releasing the directory inode In order to avoid the "Busy inodes after unmount" error message, we need to ensure that nfs_async_unlink_release() releases the super block after the call to nfs_free_unlinkdata(). Signed-off-by: Trond Myklebust --- fs/nfs/unlink.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index 3adf8b266461..f089e5839d7d 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -95,10 +95,11 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata) static void nfs_async_unlink_release(void *calldata) { struct nfs_unlinkdata *data = calldata; + struct super_block *sb = data->dir->i_sb; nfs_dec_sillycount(data->dir); - nfs_sb_deactive(NFS_SERVER(data->dir)); nfs_free_unlinkdata(data); + nfs_sb_deactive(NFS_SB(sb)); } static const struct rpc_call_ops nfs_unlink_ops = { -- cgit v1.2.3 From 940389b8afad6495211614c13eb91ef7001773ec Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Mon, 28 Jul 2008 00:48:12 +0200 Subject: task IO accounting: move all IO statistics in struct task_io_accounting Simplify the code of include/linux/task_io_accounting.h. It is also more reasonable to have all the task i/o-related statistics in a single struct (task_io_accounting). Signed-off-by: Andrea Righi Signed-off-by: Oleg Nesterov Signed-off-by: Linus Torvalds --- fs/proc/base.c | 10 +++---- include/linux/sched.h | 12 ++++----- include/linux/task_io_accounting.h | 17 ++---------- include/linux/task_io_accounting_ops.h | 48 +++++++++++++++++----------------- kernel/tsacct.c | 14 +++++----- 5 files changed, 44 insertions(+), 57 deletions(-) (limited to 'fs') diff --git a/fs/proc/base.c b/fs/proc/base.c index 3d94906c7aa8..01ed610f9b87 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -2403,7 +2403,7 @@ static int proc_base_fill_cache(struct file *filp, void *dirent, #ifdef CONFIG_TASK_IO_ACCOUNTING static int do_io_accounting(struct task_struct *task, char *buffer, int whole) { - struct proc_io_accounting acct = task->ioac; + struct task_io_accounting acct = task->ioac; unsigned long flags; if (whole && lock_task_sighand(task, &flags)) { @@ -2423,10 +2423,10 @@ static int do_io_accounting(struct task_struct *task, char *buffer, int whole) "read_bytes: %llu\n" "write_bytes: %llu\n" "cancelled_write_bytes: %llu\n", - acct.chr.rchar, acct.chr.wchar, - acct.chr.syscr, acct.chr.syscw, - acct.blk.read_bytes, acct.blk.write_bytes, - acct.blk.cancelled_write_bytes); + acct.rchar, acct.wchar, + acct.syscr, acct.syscw, + acct.read_bytes, acct.write_bytes, + acct.cancelled_write_bytes); } static int proc_tid_io_accounting(struct task_struct *task, char *buffer) diff --git a/include/linux/sched.h b/include/linux/sched.h index 034c1ca6b332..5270d449ff9d 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -505,7 +505,7 @@ struct signal_struct { unsigned long nvcsw, nivcsw, cnvcsw, cnivcsw; unsigned long min_flt, maj_flt, cmin_flt, cmaj_flt; unsigned long inblock, oublock, cinblock, coublock; - struct proc_io_accounting ioac; + struct task_io_accounting ioac; /* * Cumulative ns of scheduled CPU time for dead threads in the @@ -1253,7 +1253,7 @@ struct task_struct { unsigned long ptrace_message; siginfo_t *last_siginfo; /* For ptrace use. */ - struct proc_io_accounting ioac; + struct task_io_accounting ioac; #if defined(CONFIG_TASK_XACCT) u64 acct_rss_mem1; /* accumulated rss usage */ u64 acct_vm_mem1; /* accumulated virtual memory usage */ @@ -2183,22 +2183,22 @@ extern long sched_group_rt_period(struct task_group *tg); #ifdef CONFIG_TASK_XACCT static inline void add_rchar(struct task_struct *tsk, ssize_t amt) { - tsk->ioac.chr.rchar += amt; + tsk->ioac.rchar += amt; } static inline void add_wchar(struct task_struct *tsk, ssize_t amt) { - tsk->ioac.chr.wchar += amt; + tsk->ioac.wchar += amt; } static inline void inc_syscr(struct task_struct *tsk) { - tsk->ioac.chr.syscr++; + tsk->ioac.syscr++; } static inline void inc_syscw(struct task_struct *tsk) { - tsk->ioac.chr.syscw++; + tsk->ioac.syscw++; } #else static inline void add_rchar(struct task_struct *tsk, ssize_t amt) diff --git a/include/linux/task_io_accounting.h b/include/linux/task_io_accounting.h index 165390f8b936..5e88afc9a2fb 100644 --- a/include/linux/task_io_accounting.h +++ b/include/linux/task_io_accounting.h @@ -1,5 +1,5 @@ /* - * proc_io_accounting: a structure which is used for recording a single task's + * task_io_accounting: a structure which is used for recording a single task's * IO statistics. * * Don't include this header file directly - it is designed to be dragged in via @@ -8,8 +8,8 @@ * Blame akpm@osdl.org for all this. */ +struct task_io_accounting { #ifdef CONFIG_TASK_XACCT -struct task_chr_io_accounting { /* bytes read */ u64 rchar; /* bytes written */ @@ -18,14 +18,9 @@ struct task_chr_io_accounting { u64 syscr; /* # of write syscalls */ u64 syscw; -}; -#else /* CONFIG_TASK_XACCT */ -struct task_chr_io_accounting { -}; #endif /* CONFIG_TASK_XACCT */ #ifdef CONFIG_TASK_IO_ACCOUNTING -struct task_io_accounting { /* * The number of bytes which this task has caused to be read from * storage. @@ -46,13 +41,5 @@ struct task_io_accounting { * information loss in doing that. */ u64 cancelled_write_bytes; -}; -#else /* CONFIG_TASK_IO_ACCOUNTING */ -struct task_io_accounting { -}; #endif /* CONFIG_TASK_IO_ACCOUNTING */ - -struct proc_io_accounting { - struct task_chr_io_accounting chr; - struct task_io_accounting blk; }; diff --git a/include/linux/task_io_accounting_ops.h b/include/linux/task_io_accounting_ops.h index e6f958ebe97f..4d090f9ee608 100644 --- a/include/linux/task_io_accounting_ops.h +++ b/include/linux/task_io_accounting_ops.h @@ -9,7 +9,7 @@ #ifdef CONFIG_TASK_IO_ACCOUNTING static inline void task_io_account_read(size_t bytes) { - current->ioac.blk.read_bytes += bytes; + current->ioac.read_bytes += bytes; } /* @@ -18,12 +18,12 @@ static inline void task_io_account_read(size_t bytes) */ static inline unsigned long task_io_get_inblock(const struct task_struct *p) { - return p->ioac.blk.read_bytes >> 9; + return p->ioac.read_bytes >> 9; } static inline void task_io_account_write(size_t bytes) { - current->ioac.blk.write_bytes += bytes; + current->ioac.write_bytes += bytes; } /* @@ -32,25 +32,25 @@ static inline void task_io_account_write(size_t bytes) */ static inline unsigned long task_io_get_oublock(const struct task_struct *p) { - return p->ioac.blk.write_bytes >> 9; + return p->ioac.write_bytes >> 9; } static inline void task_io_account_cancelled_write(size_t bytes) { - current->ioac.blk.cancelled_write_bytes += bytes; + current->ioac.cancelled_write_bytes += bytes; } -static inline void task_io_accounting_init(struct proc_io_accounting *ioac) +static inline void task_io_accounting_init(struct task_io_accounting *ioac) { memset(ioac, 0, sizeof(*ioac)); } -static inline void task_blk_io_accounting_add(struct proc_io_accounting *dst, - struct proc_io_accounting *src) +static inline void task_blk_io_accounting_add(struct task_io_accounting *dst, + struct task_io_accounting *src) { - dst->blk.read_bytes += src->blk.read_bytes; - dst->blk.write_bytes += src->blk.write_bytes; - dst->blk.cancelled_write_bytes += src->blk.cancelled_write_bytes; + dst->read_bytes += src->read_bytes; + dst->write_bytes += src->write_bytes; + dst->cancelled_write_bytes += src->cancelled_write_bytes; } #else @@ -77,35 +77,35 @@ static inline void task_io_account_cancelled_write(size_t bytes) { } -static inline void task_io_accounting_init(struct proc_io_accounting *ioac) +static inline void task_io_accounting_init(struct task_io_accounting *ioac) { } -static inline void task_blk_io_accounting_add(struct proc_io_accounting *dst, - struct proc_io_accounting *src) +static inline void task_blk_io_accounting_add(struct task_io_accounting *dst, + struct task_io_accounting *src) { } #endif /* CONFIG_TASK_IO_ACCOUNTING */ #ifdef CONFIG_TASK_XACCT -static inline void task_chr_io_accounting_add(struct proc_io_accounting *dst, - struct proc_io_accounting *src) +static inline void task_chr_io_accounting_add(struct task_io_accounting *dst, + struct task_io_accounting *src) { - dst->chr.rchar += src->chr.rchar; - dst->chr.wchar += src->chr.wchar; - dst->chr.syscr += src->chr.syscr; - dst->chr.syscw += src->chr.syscw; + dst->rchar += src->rchar; + dst->wchar += src->wchar; + dst->syscr += src->syscr; + dst->syscw += src->syscw; } #else -static inline void task_chr_io_accounting_add(struct proc_io_accounting *dst, - struct proc_io_accounting *src) +static inline void task_chr_io_accounting_add(struct task_io_accounting *dst, + struct task_io_accounting *src) { } #endif /* CONFIG_TASK_XACCT */ -static inline void task_io_accounting_add(struct proc_io_accounting *dst, - struct proc_io_accounting *src) +static inline void task_io_accounting_add(struct task_io_accounting *dst, + struct task_io_accounting *src) { task_chr_io_accounting_add(dst, src); task_blk_io_accounting_add(dst, src); diff --git a/kernel/tsacct.c b/kernel/tsacct.c index f9cd2561689c..8ebcd8532dfb 100644 --- a/kernel/tsacct.c +++ b/kernel/tsacct.c @@ -94,14 +94,14 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p) stats->hiwater_vm = mm->hiwater_vm * PAGE_SIZE / KB; mmput(mm); } - stats->read_char = p->ioac.chr.rchar; - stats->write_char = p->ioac.chr.wchar; - stats->read_syscalls = p->ioac.chr.syscr; - stats->write_syscalls = p->ioac.chr.syscw; + stats->read_char = p->ioac.rchar; + stats->write_char = p->ioac.wchar; + stats->read_syscalls = p->ioac.syscr; + stats->write_syscalls = p->ioac.syscw; #ifdef CONFIG_TASK_IO_ACCOUNTING - stats->read_bytes = p->ioac.blk.read_bytes; - stats->write_bytes = p->ioac.blk.write_bytes; - stats->cancelled_write_bytes = p->ioac.blk.cancelled_write_bytes; + stats->read_bytes = p->ioac.read_bytes; + stats->write_bytes = p->ioac.write_bytes; + stats->cancelled_write_bytes = p->ioac.cancelled_write_bytes; #else stats->read_bytes = 0; stats->write_bytes = 0; -- cgit v1.2.3