diff options
Diffstat (limited to 'fs')
58 files changed, 1307 insertions, 1613 deletions
diff --git a/fs/Makefile b/fs/Makefile index f0db9c941a5f..4fe6df3ec28f 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -7,7 +7,7 @@ obj-y := open.o read_write.o file_table.o super.o \ char_dev.o stat.o exec.o pipe.o namei.o fcntl.o \ - ioctl.o readdir.o select.o fifo.o dcache.o inode.o \ + ioctl.o readdir.o select.o dcache.o inode.o \ attr.o bad_inode.o file.o filesystems.o namespace.o \ seq_file.o xattr.o libfs.o fs-writeback.o \ pnode.o splice.o sync.o utimes.o \ diff --git a/fs/afs/proc.c b/fs/afs/proc.c index 096b23f821a1..526e4bbbde59 100644 --- a/fs/afs/proc.c +++ b/fs/afs/proc.c @@ -190,7 +190,7 @@ static int afs_proc_cells_open(struct inode *inode, struct file *file) return ret; m = file->private_data; - m->private = PDE(inode)->data; + m->private = PDE_DATA(inode); return 0; } @@ -448,7 +448,7 @@ static int afs_proc_cell_volumes_open(struct inode *inode, struct file *file) struct seq_file *m; int ret; - cell = PDE(inode)->data; + cell = PDE_DATA(inode); if (!cell) return -ENOENT; @@ -554,7 +554,7 @@ static int afs_proc_cell_vlservers_open(struct inode *inode, struct file *file) struct seq_file *m; int ret; - cell = PDE(inode)->data; + cell = PDE_DATA(inode); if (!cell) return -ENOENT; @@ -659,7 +659,7 @@ static int afs_proc_cell_servers_open(struct inode *inode, struct file *file) struct seq_file *m; int ret; - cell = PDE(inode)->data; + cell = PDE_DATA(inode); if (!cell) return -ENOENT; @@ -1324,6 +1324,8 @@ static ssize_t aio_rw_vect_retry(struct kiocb *iocb) if (iocb->ki_pos < 0) return -EINVAL; + if (opcode == IOCB_CMD_PWRITEV) + file_start_write(file); do { ret = rw_op(iocb, &iocb->ki_iovec[iocb->ki_cur_seg], iocb->ki_nr_segs - iocb->ki_cur_seg, @@ -1336,6 +1338,8 @@ static ssize_t aio_rw_vect_retry(struct kiocb *iocb) } while (ret > 0 && iocb->ki_left > 0 && (opcode == IOCB_CMD_PWRITEV || (!S_ISFIFO(inode->i_mode) && !S_ISSOCK(inode->i_mode)))); + if (opcode == IOCB_CMD_PWRITEV) + file_end_write(file); /* This means we must have transferred all that we could */ /* No need to retry anymore */ diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index 02fe378fc506..bce87694f7b0 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -286,15 +286,12 @@ static int load_aout_binary(struct linux_binprm * bprm) return error; } - error = bprm->file->f_op->read(bprm->file, - (char __user *)text_addr, - ex.a_text+ex.a_data, &pos); + error = read_code(bprm->file, text_addr, pos, + ex.a_text+ex.a_data); if ((signed long)error < 0) { send_sig(SIGKILL, current, 0); return error; } - - flush_icache_range(text_addr, text_addr+ex.a_text+ex.a_data); } else { if ((ex.a_text & 0xfff || ex.a_data & 0xfff) && (N_MAGIC(ex) != NMAGIC) && printk_ratelimit()) @@ -310,14 +307,9 @@ static int load_aout_binary(struct linux_binprm * bprm) } if (!bprm->file->f_op->mmap||((fd_offset & ~PAGE_MASK) != 0)) { - loff_t pos = fd_offset; vm_brk(N_TXTADDR(ex), ex.a_text+ex.a_data); - bprm->file->f_op->read(bprm->file, - (char __user *)N_TXTADDR(ex), - ex.a_text+ex.a_data, &pos); - flush_icache_range((unsigned long) N_TXTADDR(ex), - (unsigned long) N_TXTADDR(ex) + - ex.a_text+ex.a_data); + read_code(bprm->file, N_TXTADDR(ex), fd_offset, + ex.a_text + ex.a_data); goto beyond_if; } @@ -396,8 +388,6 @@ static int load_aout_library(struct file *file) start_addr = ex.a_entry & 0xfffff000; if ((N_TXTOFF(ex) & ~PAGE_MASK) != 0) { - loff_t pos = N_TXTOFF(ex); - if (printk_ratelimit()) { printk(KERN_WARNING @@ -406,11 +396,8 @@ static int load_aout_library(struct file *file) } vm_brk(start_addr, ex.a_text + ex.a_data + ex.a_bss); - file->f_op->read(file, (char __user *)start_addr, - ex.a_text + ex.a_data, &pos); - flush_icache_range((unsigned long) start_addr, - (unsigned long) start_addr + ex.a_text + ex.a_data); - + read_code(file, start_addr, N_TXTOFF(ex), + ex.a_text + ex.a_data); retval = 0; goto out; } diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index c1cc06aed601..9dac212fc6f9 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -926,7 +926,6 @@ static int elf_fdpic_map_file_constdisp_on_uclinux( struct elf32_fdpic_loadseg *seg; struct elf32_phdr *phdr; unsigned long load_addr, base = ULONG_MAX, top = 0, maddr = 0, mflags; - loff_t fpos; int loop, ret; load_addr = params->load_addr; @@ -964,14 +963,12 @@ static int elf_fdpic_map_file_constdisp_on_uclinux( if (params->phdrs[loop].p_type != PT_LOAD) continue; - fpos = phdr->p_offset; - seg->addr = maddr + (phdr->p_vaddr - base); seg->p_vaddr = phdr->p_vaddr; seg->p_memsz = phdr->p_memsz; - ret = file->f_op->read(file, (void *) seg->addr, - phdr->p_filesz, &fpos); + ret = read_code(file, seg->addr, phdr->p_offset, + phdr->p_filesz); if (ret < 0) return ret; diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 2036d21baaef..d50bbe59da1e 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -207,11 +207,12 @@ static int decompress_exec( /* Read in first chunk of data and parse gzip header. */ fpos = offset; - ret = bprm->file->f_op->read(bprm->file, buf, LBUFSIZE, &fpos); + ret = kernel_read(bprm->file, offset, buf, LBUFSIZE); strm.next_in = buf; strm.avail_in = ret; strm.total_in = 0; + fpos += ret; retval = -ENOEXEC; @@ -277,7 +278,7 @@ static int decompress_exec( } while ((ret = zlib_inflate(&strm, Z_NO_FLUSH)) == Z_OK) { - ret = bprm->file->f_op->read(bprm->file, buf, LBUFSIZE, &fpos); + ret = kernel_read(bprm->file, fpos, buf, LBUFSIZE); if (ret <= 0) break; len -= ret; @@ -285,6 +286,7 @@ static int decompress_exec( strm.next_in = buf; strm.avail_in = ret; strm.total_in = 0; + fpos += ret; } if (ret < 0) { @@ -428,6 +430,7 @@ static int load_flat_file(struct linux_binprm * bprm, unsigned long textpos = 0, datapos = 0, result; unsigned long realdatastart = 0; unsigned long text_len, data_len, bss_len, stack_len, flags; + unsigned long full_data; unsigned long len, memp = 0; unsigned long memp_size, extra, rlim; unsigned long *reloc = 0, *rp; @@ -451,6 +454,7 @@ static int load_flat_file(struct linux_binprm * bprm, relocs = ntohl(hdr->reloc_count); flags = ntohl(hdr->flags); rev = ntohl(hdr->rev); + full_data = data_len + relocs * sizeof(unsigned long); if (strncmp(hdr->magic, "bFLT", 4)) { /* @@ -577,12 +581,12 @@ static int load_flat_file(struct linux_binprm * bprm, #ifdef CONFIG_BINFMT_ZFLAT if (flags & FLAT_FLAG_GZDATA) { result = decompress_exec(bprm, fpos, (char *) datapos, - data_len + (relocs * sizeof(unsigned long)), 0); + full_data, 0); } else #endif { - result = bprm->file->f_op->read(bprm->file, (char *) datapos, - data_len + (relocs * sizeof(unsigned long)), &fpos); + result = read_code(bprm->file, datapos, fpos, + full_data); } if (IS_ERR_VALUE(result)) { printk("Unable to read data+bss, errno %d\n", (int)-result); @@ -627,30 +631,25 @@ static int load_flat_file(struct linux_binprm * bprm, if (flags & FLAT_FLAG_GZIP) { result = decompress_exec(bprm, sizeof (struct flat_hdr), (((char *) textpos) + sizeof (struct flat_hdr)), - (text_len + data_len + (relocs * sizeof(unsigned long)) + (text_len + full_data - sizeof (struct flat_hdr)), 0); memmove((void *) datapos, (void *) realdatastart, - data_len + (relocs * sizeof(unsigned long))); + full_data); } else if (flags & FLAT_FLAG_GZDATA) { - fpos = 0; - result = bprm->file->f_op->read(bprm->file, - (char *) textpos, text_len, &fpos); + result = read_code(bprm->file, textpos, 0, text_len); if (!IS_ERR_VALUE(result)) result = decompress_exec(bprm, text_len, (char *) datapos, - data_len + (relocs * sizeof(unsigned long)), 0); + full_data, 0); } else #endif { - fpos = 0; - result = bprm->file->f_op->read(bprm->file, - (char *) textpos, text_len, &fpos); - if (!IS_ERR_VALUE(result)) { - fpos = ntohl(hdr->data_start); - result = bprm->file->f_op->read(bprm->file, (char *) datapos, - data_len + (relocs * sizeof(unsigned long)), &fpos); - } + result = read_code(bprm->file, textpos, 0, text_len); + if (!IS_ERR_VALUE(result)) + result = read_code(bprm->file, datapos, + ntohl(hdr->data_start), + full_data); } if (IS_ERR_VALUE(result)) { printk("Unable to read code+data+bss, errno %d\n",(int)-result); diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index ade03e6f7bd2..bb8b7a0e28a6 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1514,8 +1514,6 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb, size_t count, ocount; bool sync = (file->f_flags & O_DSYNC) || IS_SYNC(file->f_mapping->host); - sb_start_write(inode->i_sb); - mutex_lock(&inode->i_mutex); err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ); @@ -1617,7 +1615,6 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb, if (sync) atomic_dec(&BTRFS_I(inode)->sync_writers); out: - sb_end_write(inode->i_sb); current->backing_dev_info = NULL; return num_written ? num_written : err; } diff --git a/fs/cachefiles/rdwr.c b/fs/cachefiles/rdwr.c index 480992259707..317f9ee9c991 100644 --- a/fs/cachefiles/rdwr.c +++ b/fs/cachefiles/rdwr.c @@ -962,12 +962,14 @@ int cachefiles_write_page(struct fscache_storage *op, struct page *page) } data = kmap(page); + file_start_write(file); old_fs = get_fs(); set_fs(KERNEL_DS); ret = file->f_op->write( file, (const void __user *) data, len, &pos); set_fs(old_fs); kunmap(page); + file_end_write(file); if (ret != len) ret = -EIO; } diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 7a0dd99e4507..2d4a231dd70b 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -2520,8 +2520,6 @@ cifs_writev(struct kiocb *iocb, const struct iovec *iov, BUG_ON(iocb->ki_pos != pos); - sb_start_write(inode->i_sb); - /* * We need to hold the sem to be sure nobody modifies lock list * with a brlock that prevents writing. @@ -2545,7 +2543,6 @@ cifs_writev(struct kiocb *iocb, const struct iovec *iov, } up_read(&cinode->lock_sem); - sb_end_write(inode->i_sb); return rc; } diff --git a/fs/coda/file.c b/fs/coda/file.c index fa4c100bdc7d..380b798f8443 100644 --- a/fs/coda/file.c +++ b/fs/coda/file.c @@ -79,6 +79,7 @@ coda_file_write(struct file *coda_file, const char __user *buf, size_t count, lo return -EINVAL; host_inode = file_inode(host_file); + file_start_write(host_file); mutex_lock(&coda_inode->i_mutex); ret = host_file->f_op->write(host_file, buf, count, ppos); @@ -87,6 +88,7 @@ coda_file_write(struct file *coda_file, const char __user *buf, size_t count, lo coda_inode->i_blocks = (coda_inode->i_size + 511) >> 9; coda_inode->i_mtime = coda_inode->i_ctime = CURRENT_TIME_SEC; mutex_unlock(&coda_inode->i_mutex); + file_end_write(host_file); return ret; } diff --git a/fs/compat.c b/fs/compat.c index 5f83ffa42115..d0560c93973d 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1068,190 +1068,6 @@ asmlinkage long compat_sys_getdents64(unsigned int fd, } #endif /* ! __ARCH_OMIT_COMPAT_SYS_GETDENTS64 */ -static ssize_t compat_do_readv_writev(int type, struct file *file, - const struct compat_iovec __user *uvector, - unsigned long nr_segs, loff_t *pos) -{ - compat_ssize_t tot_len; - struct iovec iovstack[UIO_FASTIOV]; - struct iovec *iov = iovstack; - ssize_t ret; - io_fn_t fn; - iov_fn_t fnv; - - ret = -EINVAL; - if (!file->f_op) - goto out; - - ret = compat_rw_copy_check_uvector(type, uvector, nr_segs, - UIO_FASTIOV, iovstack, &iov); - if (ret <= 0) - goto out; - - tot_len = ret; - ret = rw_verify_area(type, file, pos, tot_len); - if (ret < 0) - goto out; - - fnv = NULL; - if (type == READ) { - fn = file->f_op->read; - fnv = file->f_op->aio_read; - } else { - fn = (io_fn_t)file->f_op->write; - fnv = file->f_op->aio_write; - } - - if (fnv) - ret = do_sync_readv_writev(file, iov, nr_segs, tot_len, - pos, fnv); - else - ret = do_loop_readv_writev(file, iov, nr_segs, pos, fn); - -out: - if (iov != iovstack) - kfree(iov); - if ((ret + (type == READ)) > 0) { - if (type == READ) - fsnotify_access(file); - else - fsnotify_modify(file); - } - return ret; -} - -static size_t compat_readv(struct file *file, - const struct compat_iovec __user *vec, - unsigned long vlen, loff_t *pos) -{ - ssize_t ret = -EBADF; - - if (!(file->f_mode & FMODE_READ)) - goto out; - - ret = -EINVAL; - if (!file->f_op || (!file->f_op->aio_read && !file->f_op->read)) - goto out; - - ret = compat_do_readv_writev(READ, file, vec, vlen, pos); - -out: - if (ret > 0) - add_rchar(current, ret); - inc_syscr(current); - return ret; -} - -asmlinkage ssize_t -compat_sys_readv(unsigned long fd, const struct compat_iovec __user *vec, - unsigned long vlen) -{ - struct fd f = fdget(fd); - ssize_t ret; - loff_t pos; - - if (!f.file) - return -EBADF; - pos = f.file->f_pos; - ret = compat_readv(f.file, vec, vlen, &pos); - f.file->f_pos = pos; - fdput(f); - return ret; -} - -asmlinkage ssize_t -compat_sys_preadv64(unsigned long fd, const struct compat_iovec __user *vec, - unsigned long vlen, loff_t pos) -{ - struct fd f; - ssize_t ret; - - if (pos < 0) - return -EINVAL; - f = fdget(fd); - if (!f.file) - return -EBADF; - ret = -ESPIPE; - if (f.file->f_mode & FMODE_PREAD) - ret = compat_readv(f.file, vec, vlen, &pos); - fdput(f); - return ret; -} - -asmlinkage ssize_t -compat_sys_preadv(unsigned long fd, const struct compat_iovec __user *vec, - unsigned long vlen, u32 pos_low, u32 pos_high) -{ - loff_t pos = ((loff_t)pos_high << 32) | pos_low; - return compat_sys_preadv64(fd, vec, vlen, pos); -} - -static size_t compat_writev(struct file *file, - const struct compat_iovec __user *vec, - unsigned long vlen, loff_t *pos) -{ - ssize_t ret = -EBADF; - - if (!(file->f_mode & FMODE_WRITE)) - goto out; - - ret = -EINVAL; - if (!file->f_op || (!file->f_op->aio_write && !file->f_op->write)) - goto out; - - ret = compat_do_readv_writev(WRITE, file, vec, vlen, pos); - -out: - if (ret > 0) - add_wchar(current, ret); - inc_syscw(current); - return ret; -} - -asmlinkage ssize_t -compat_sys_writev(unsigned long fd, const struct compat_iovec __user *vec, - unsigned long vlen) -{ - struct fd f = fdget(fd); - ssize_t ret; - loff_t pos; - - if (!f.file) - return -EBADF; - pos = f.file->f_pos; - ret = compat_writev(f.file, vec, vlen, &pos); - f.file->f_pos = pos; - fdput(f); - return ret; -} - -asmlinkage ssize_t -compat_sys_pwritev64(unsigned long fd, const struct compat_iovec __user *vec, - unsigned long vlen, loff_t pos) -{ - struct fd f; - ssize_t ret; - - if (pos < 0) - return -EINVAL; - f = fdget(fd); - if (!f.file) - return -EBADF; - ret = -ESPIPE; - if (f.file->f_mode & FMODE_PWRITE) - ret = compat_writev(f.file, vec, vlen, &pos); - fdput(f); - return ret; -} - -asmlinkage ssize_t -compat_sys_pwritev(unsigned long fd, const struct compat_iovec __user *vec, - unsigned long vlen, u32 pos_low, u32 pos_high) -{ - loff_t pos = ((loff_t)pos_high << 32) | pos_low; - return compat_sys_pwritev64(fd, vec, vlen, pos); -} - /* * Exactly like fs/open.c:sys_open(), except that it doesn't set the * O_LARGEFILE flag. diff --git a/fs/coredump.c b/fs/coredump.c index ec306cc9a28a..a9abe313e8d5 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -432,9 +432,7 @@ static bool dump_interrupted(void) static void wait_for_dump_helpers(struct file *file) { - struct pipe_inode_info *pipe; - - pipe = file_inode(file)->i_pipe; + struct pipe_inode_info *pipe = file->private_data; pipe_lock(pipe); pipe->readers++; @@ -656,7 +654,9 @@ void do_coredump(siginfo_t *siginfo) goto close_fail; if (displaced) put_files_struct(displaced); + file_start_write(cprm.file); core_dumped = !dump_interrupted() && binfmt->core_dump(&cprm); + file_end_write(cprm.file); if (ispipe && core_pipe_limit) wait_for_dump_helpers(cprm.file); diff --git a/fs/efivarfs/file.c b/fs/efivarfs/file.c index ede07fc7309f..bfb531564319 100644 --- a/fs/efivarfs/file.c +++ b/fs/efivarfs/file.c @@ -9,6 +9,7 @@ #include <linux/efi.h> #include <linux/fs.h> +#include <linux/slab.h> #include "internal.h" diff --git a/fs/efivarfs/inode.c b/fs/efivarfs/inode.c index 640e289d522e..7e787fb90293 100644 --- a/fs/efivarfs/inode.c +++ b/fs/efivarfs/inode.c @@ -10,6 +10,7 @@ #include <linux/efi.h> #include <linux/fs.h> #include <linux/ctype.h> +#include <linux/slab.h> #include "internal.h" diff --git a/fs/efivarfs/super.c b/fs/efivarfs/super.c index 525a2a1ac16c..141aee31884f 100644 --- a/fs/efivarfs/super.c +++ b/fs/efivarfs/super.c @@ -13,6 +13,8 @@ #include <linux/module.h> #include <linux/pagemap.h> #include <linux/ucs2_string.h> +#include <linux/slab.h> +#include <linux/magic.h> #include "internal.h" diff --git a/fs/exec.c b/fs/exec.c index 963f510a25ab..643019585574 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -802,6 +802,15 @@ int kernel_read(struct file *file, loff_t offset, EXPORT_SYMBOL(kernel_read); +ssize_t read_code(struct file *file, unsigned long addr, loff_t pos, size_t len) +{ + ssize_t res = file->f_op->read(file, (void __user *)addr, len, &pos); + if (res > 0) + flush_icache_range(addr, addr + len); + return res; +} +EXPORT_SYMBOL(read_code); + static int exec_mmap(struct mm_struct *mm) { struct task_struct *tsk; diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index a11ea4d6164c..b1ed9e07434b 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2260,7 +2260,7 @@ static const struct seq_operations ext4_mb_seq_groups_ops = { static int ext4_mb_seq_groups_open(struct inode *inode, struct file *file) { - struct super_block *sb = PDE(inode)->data; + struct super_block *sb = PDE_DATA(inode); int rc; rc = seq_open(file, &ext4_mb_seq_groups_ops); diff --git a/fs/ext4/super.c b/fs/ext4/super.c index dbc7c090c13a..24a146bde742 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1806,7 +1806,7 @@ static int options_seq_show(struct seq_file *seq, void *offset) static int options_open_fs(struct inode *inode, struct file *file) { - return single_open(file, options_seq_show, PDE(inode)->data); + return single_open(file, options_seq_show, PDE_DATA(inode)); } static const struct file_operations ext4_seq_options_fops = { diff --git a/fs/f2fs/acl.c b/fs/f2fs/acl.c index 137af4255da6..44abc2f286e0 100644 --- a/fs/f2fs/acl.c +++ b/fs/f2fs/acl.c @@ -299,7 +299,7 @@ int f2fs_acl_chmod(struct inode *inode) struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); struct posix_acl *acl; int error; - mode_t mode = get_inode_mode(inode); + umode_t mode = get_inode_mode(inode); if (!test_opt(sbi, POSIX_ACL)) return 0; diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c index a1f38443ecee..1be948768e2f 100644 --- a/fs/f2fs/dir.c +++ b/fs/f2fs/dir.c @@ -60,7 +60,7 @@ static unsigned char f2fs_type_by_mode[S_IFMT >> S_SHIFT] = { static void set_de_type(struct f2fs_dir_entry *de, struct inode *inode) { - mode_t mode = inode->i_mode; + umode_t mode = inode->i_mode; de->file_type = f2fs_type_by_mode[(mode & S_IFMT) >> S_SHIFT]; } diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 958a46da19ae..db626282d424 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -590,7 +590,7 @@ long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { unsigned int oldflags; - ret = mnt_want_write(filp->f_path.mnt); + ret = mnt_want_write_file(filp); if (ret) return ret; @@ -627,7 +627,7 @@ long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) inode->i_ctime = CURRENT_TIME; mark_inode_dirty(inode); out: - mnt_drop_write(filp->f_path.mnt); + mnt_drop_write_file(filp); return ret; } default: diff --git a/fs/fifo.c b/fs/fifo.c deleted file mode 100644 index cf6f4345ceb0..000000000000 --- a/fs/fifo.c +++ /dev/null @@ -1,153 +0,0 @@ -/* - * linux/fs/fifo.c - * - * written by Paul H. Hargrove - * - * Fixes: - * 10-06-1999, AV: fixed OOM handling in fifo_open(), moved - * initialization there, switched to external - * allocation of pipe_inode_info. - */ - -#include <linux/mm.h> -#include <linux/fs.h> -#include <linux/sched.h> -#include <linux/pipe_fs_i.h> - -static int wait_for_partner(struct inode* inode, unsigned int *cnt) -{ - int cur = *cnt; - - while (cur == *cnt) { - pipe_wait(inode->i_pipe); - if (signal_pending(current)) - break; - } - return cur == *cnt ? -ERESTARTSYS : 0; -} - -static void wake_up_partner(struct inode* inode) -{ - wake_up_interruptible(&inode->i_pipe->wait); -} - -static int fifo_open(struct inode *inode, struct file *filp) -{ - struct pipe_inode_info *pipe; - int ret; - - mutex_lock(&inode->i_mutex); - pipe = inode->i_pipe; - if (!pipe) { - ret = -ENOMEM; - pipe = alloc_pipe_info(inode); - if (!pipe) - goto err_nocleanup; - inode->i_pipe = pipe; - } - filp->f_version = 0; - - /* We can only do regular read/write on fifos */ - filp->f_mode &= (FMODE_READ | FMODE_WRITE); - - switch (filp->f_mode) { - case FMODE_READ: - /* - * O_RDONLY - * 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_pipefifo_fops; - pipe->r_counter++; - if (pipe->readers++ == 0) - wake_up_partner(inode); - - if (!pipe->writers) { - if ((filp->f_flags & O_NONBLOCK)) { - /* suppress POLLHUP until we have - * seen a writer */ - filp->f_version = pipe->w_counter; - } else { - if (wait_for_partner(inode, &pipe->w_counter)) - goto err_rd; - } - } - break; - - case FMODE_WRITE: - /* - * O_WRONLY - * POSIX.1 says that O_NONBLOCK means return -1 with - * errno=ENXIO when there is no process reading the FIFO. - */ - ret = -ENXIO; - if ((filp->f_flags & O_NONBLOCK) && !pipe->readers) - goto err; - - filp->f_op = &write_pipefifo_fops; - pipe->w_counter++; - if (!pipe->writers++) - wake_up_partner(inode); - - if (!pipe->readers) { - if (wait_for_partner(inode, &pipe->r_counter)) - goto err_wr; - } - break; - - case FMODE_READ | FMODE_WRITE: - /* - * O_RDWR - * POSIX.1 leaves this case "undefined" when O_NONBLOCK is set. - * This implementation will NEVER block on a O_RDWR open, since - * the process can at least talk to itself. - */ - filp->f_op = &rdwr_pipefifo_fops; - - pipe->readers++; - pipe->writers++; - pipe->r_counter++; - pipe->w_counter++; - if (pipe->readers == 1 || pipe->writers == 1) - wake_up_partner(inode); - break; - - default: - ret = -EINVAL; - goto err; - } - - /* Ok! */ - mutex_unlock(&inode->i_mutex); - return 0; - -err_rd: - if (!--pipe->readers) - wake_up_interruptible(&pipe->wait); - ret = -ERESTARTSYS; - goto err; - -err_wr: - if (!--pipe->writers) - wake_up_interruptible(&pipe->wait); - ret = -ERESTARTSYS; - goto err; - -err: - if (!pipe->readers && !pipe->writers) - free_pipe_info(inode); - -err_nocleanup: - mutex_unlock(&inode->i_mutex); - return ret; -} - -/* - * Dummy default file-operations: the only thing this does - * is contain the open that then fills in the correct operations - * depending on the access mode of the file... - */ -const struct file_operations def_fifo_fops = { - .open = fifo_open, /* will set read_ or write_pipefifo_fops */ - .llseek = noop_llseek, -}; diff --git a/fs/file.c b/fs/file.c index 3906d9577a18..4a78f981557a 100644 --- a/fs/file.c +++ b/fs/file.c @@ -23,24 +23,10 @@ #include <linux/rcupdate.h> #include <linux/workqueue.h> -struct fdtable_defer { - spinlock_t lock; - struct work_struct wq; - struct fdtable *next; -}; - int sysctl_nr_open __read_mostly = 1024*1024; int sysctl_nr_open_min = BITS_PER_LONG; int sysctl_nr_open_max = 1024 * 1024; /* raised later */ -/* - * We use this list to defer free fdtables that have vmalloced - * sets/arrays. By keeping a per-cpu list, we avoid having to embed - * the work_struct in fdtable itself which avoids a 64 byte (i386) increase in - * this per-task structure. - */ -static DEFINE_PER_CPU(struct fdtable_defer, fdtable_defer_list); - static void *alloc_fdmem(size_t size) { /* @@ -67,46 +53,9 @@ static void __free_fdtable(struct fdtable *fdt) kfree(fdt); } -static void free_fdtable_work(struct work_struct *work) -{ - struct fdtable_defer *f = - container_of(work, struct fdtable_defer, wq); - struct fdtable *fdt; - - spin_lock_bh(&f->lock); - fdt = f->next; - f->next = NULL; - spin_unlock_bh(&f->lock); - while(fdt) { - struct fdtable *next = fdt->next; - - __free_fdtable(fdt); - fdt = next; - } -} - static void free_fdtable_rcu(struct rcu_head *rcu) { - struct fdtable *fdt = container_of(rcu, struct fdtable, rcu); - struct fdtable_defer *fddef; - - BUG_ON(!fdt); - BUG_ON(fdt->max_fds <= NR_OPEN_DEFAULT); - - if (!is_vmalloc_addr(fdt->fd) && !is_vmalloc_addr(fdt->open_fds)) { - kfree(fdt->fd); - kfree(fdt->open_fds); - kfree(fdt); - } else { - fddef = &get_cpu_var(fdtable_defer_list); - spin_lock(&fddef->lock); - fdt->next = fddef->next; - fddef->next = fdt; - /* vmallocs are handled from the workqueue context */ - schedule_work(&fddef->wq); - spin_unlock(&fddef->lock); - put_cpu_var(fdtable_defer_list); - } + __free_fdtable(container_of(rcu, struct fdtable, rcu)); } /* @@ -174,7 +123,6 @@ static struct fdtable * alloc_fdtable(unsigned int nr) fdt->open_fds = data; data += nr / BITS_PER_BYTE; fdt->close_on_exec = data; - fdt->next = NULL; return fdt; @@ -221,7 +169,7 @@ static int expand_fdtable(struct files_struct *files, int nr) /* Continue as planned */ copy_fdtable(new_fdt, cur_fdt); rcu_assign_pointer(files->fdt, new_fdt); - if (cur_fdt->max_fds > NR_OPEN_DEFAULT) + if (cur_fdt != &files->fdtab) call_rcu(&cur_fdt->rcu, free_fdtable_rcu); } else { /* Somebody else expanded, so undo our attempt */ @@ -316,7 +264,6 @@ struct files_struct *dup_fd(struct files_struct *oldf, int *errorp) new_fdt->close_on_exec = newf->close_on_exec_init; new_fdt->open_fds = newf->open_fds_init; new_fdt->fd = &newf->fd_array[0]; - new_fdt->next = NULL; spin_lock(&oldf->file_lock); old_fdt = files_fdtable(oldf); @@ -490,19 +437,8 @@ void exit_files(struct task_struct *tsk) } } -static void fdtable_defer_list_init(int cpu) -{ - struct fdtable_defer *fddef = &per_cpu(fdtable_defer_list, cpu); - spin_lock_init(&fddef->lock); - INIT_WORK(&fddef->wq, free_fdtable_work); - fddef->next = NULL; -} - void __init files_defer_init(void) { - int i; - for_each_possible_cpu(i) - fdtable_defer_list_init(i); sysctl_nr_open_max = min((size_t)INT_MAX, ~(size_t)0/sizeof(void *)) & -BITS_PER_LONG; } diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 11dfa0c3fb46..9bfd1a3214e6 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -1319,7 +1319,7 @@ static ssize_t fuse_dev_splice_read(struct file *in, loff_t *ppos, page_nr++; ret += buf->len; - if (pipe->inode) + if (pipe->files) do_wakeup = 1; } diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 34b80ba95bad..d15c6f21c17f 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -971,7 +971,6 @@ static ssize_t fuse_file_aio_write(struct kiocb *iocb, const struct iovec *iov, return err; count = ocount; - sb_start_write(inode->i_sb); mutex_lock(&inode->i_mutex); /* We can write back this queue in page reclaim */ @@ -1030,7 +1029,6 @@ static ssize_t fuse_file_aio_write(struct kiocb *iocb, const struct iovec *iov, out: current->backing_dev_info = NULL; mutex_unlock(&inode->i_mutex); - sb_end_write(inode->i_sb); return written ? written : err; } diff --git a/fs/hpfs/file.c b/fs/hpfs/file.c index 9f9dbeceeee7..3027f4dbbab5 100644 --- a/fs/hpfs/file.c +++ b/fs/hpfs/file.c @@ -131,6 +131,24 @@ static int hpfs_write_begin(struct file *file, struct address_space *mapping, return ret; } +static int hpfs_write_end(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *pagep, void *fsdata) +{ + struct inode *inode = mapping->host; + int err; + err = generic_write_end(file, mapping, pos, len, copied, pagep, fsdata); + if (err < len) + hpfs_write_failed(mapping, pos + len); + if (!(err < 0)) { + /* make sure we write it on close, if not earlier */ + hpfs_lock(inode->i_sb); + hpfs_i(inode)->i_dirty = 1; + hpfs_unlock(inode->i_sb); + } + return err; +} + static sector_t _hpfs_bmap(struct address_space *mapping, sector_t block) { return generic_block_bmap(mapping,block,hpfs_get_block); @@ -140,30 +158,16 @@ const struct address_space_operations hpfs_aops = { .readpage = hpfs_readpage, .writepage = hpfs_writepage, .write_begin = hpfs_write_begin, - .write_end = generic_write_end, + .write_end = hpfs_write_end, .bmap = _hpfs_bmap }; -static ssize_t hpfs_file_write(struct file *file, const char __user *buf, - size_t count, loff_t *ppos) -{ - ssize_t retval; - - retval = do_sync_write(file, buf, count, ppos); - if (retval > 0) { - hpfs_lock(file->f_path.dentry->d_sb); - hpfs_i(file_inode(file))->i_dirty = 1; - hpfs_unlock(file->f_path.dentry->d_sb); - } - return retval; -} - const struct file_operations hpfs_file_ops = { .llseek = generic_file_llseek, .read = do_sync_read, .aio_read = generic_file_aio_read, - .write = hpfs_file_write, + .write = do_sync_write, .aio_write = generic_file_aio_write, .mmap = generic_file_mmap, .release = hpfs_file_release, diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c index 126d3c2e2dee..cd3e38972c86 100644 --- a/fs/hppfs/hppfs.c +++ b/fs/hppfs/hppfs.c @@ -436,7 +436,6 @@ static int hppfs_open(struct inode *inode, struct file *file) path.mnt = inode->i_sb->s_fs_info; path.dentry = HPPFS_I(inode)->proc_dentry; - /* XXX This isn't closed anywhere */ data->proc_file = dentry_open(&path, file_mode(file->f_mode), cred); err = PTR_ERR(data->proc_file); if (IS_ERR(data->proc_file)) @@ -523,12 +522,23 @@ static loff_t hppfs_llseek(struct file *file, loff_t off, int where) return default_llseek(file, off, where); } +static int hppfs_release(struct inode *inode, struct file *file) +{ + struct hppfs_private *data = file->private_data; + struct file *proc_file = data->proc_file; + if (proc_file) + fput(proc_file); + kfree(data); + return 0; +} + static const struct file_operations hppfs_file_fops = { .owner = NULL, .llseek = hppfs_llseek, .read = hppfs_read, .write = hppfs_write, .open = hppfs_open, + .release = hppfs_release, }; struct hppfs_dirent { @@ -570,18 +580,12 @@ static int hppfs_readdir(struct file *file, void *ent, filldir_t filldir) return err; } -static int hppfs_fsync(struct file *file, loff_t start, loff_t end, - int datasync) -{ - return filemap_write_and_wait_range(file->f_mapping, start, end); -} - static const struct file_operations hppfs_dir_fops = { .owner = NULL, .readdir = hppfs_readdir, .open = hppfs_dir_open, - .fsync = hppfs_fsync, .llseek = default_llseek, + .release = hppfs_release, }; static int hppfs_statfs(struct dentry *dentry, struct kstatfs *sf) diff --git a/fs/inode.c b/fs/inode.c index a898b3d43ccf..00d5fc3b86e1 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1803,7 +1803,7 @@ void init_special_inode(struct inode *inode, umode_t mode, dev_t rdev) inode->i_fop = &def_blk_fops; inode->i_rdev = rdev; } else if (S_ISFIFO(mode)) - inode->i_fop = &def_fifo_fops; + inode->i_fop = &pipefifo_fops; else if (S_ISSOCK(mode)) inode->i_fop = &bad_sock_fops; else diff --git a/fs/internal.h b/fs/internal.h index 4be78237d896..eaa75f75b625 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -130,3 +130,8 @@ extern struct dentry *__d_alloc(struct super_block *, const struct qstr *); * read_write.c */ extern ssize_t __kernel_write(struct file *, const char *, size_t, loff_t *); + +/* + * pipe.c + */ +extern const struct file_operations pipefifo_fops; diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index f6c5ba027f4f..95457576e434 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -979,7 +979,7 @@ static const struct seq_operations jbd2_seq_info_ops = { static int jbd2_seq_info_open(struct inode *inode, struct file *file) { - journal_t *journal = PDE(inode)->data; + journal_t *journal = PDE_DATA(inode); struct jbd2_stats_proc_session *s; int rc, size; diff --git a/fs/mount.h b/fs/mount.h index cd5007980400..64a858143ff9 100644 --- a/fs/mount.h +++ b/fs/mount.h @@ -18,6 +18,12 @@ struct mnt_pcp { int mnt_writers; }; +struct mountpoint { + struct list_head m_hash; + struct dentry *m_dentry; + int m_count; +}; + struct mount { struct list_head mnt_hash; struct mount *mnt_parent; @@ -40,6 +46,7 @@ struct mount { struct list_head mnt_slave; /* slave list entry */ struct mount *mnt_master; /* slave is on master->mnt_slave_list */ struct mnt_namespace *mnt_ns; /* containing namespace */ + struct mountpoint *mnt_mp; /* where is it mounted */ #ifdef CONFIG_FSNOTIFY struct hlist_head mnt_fsnotify_marks; __u32 mnt_fsnotify_mask; diff --git a/fs/namespace.c b/fs/namespace.c index 341d3f564082..b4f96a5230a3 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -21,7 +21,8 @@ #include <linux/fs_struct.h> /* get_fs_root et.al. */ #include <linux/fsnotify.h> /* fsnotify_vfsmount_delete */ #include <linux/uaccess.h> -#include <linux/proc_fs.h> +#include <linux/proc_ns.h> +#include <linux/magic.h> #include "pnode.h" #include "internal.h" @@ -36,6 +37,7 @@ static int mnt_id_start = 0; static int mnt_group_start = 1; static struct list_head *mount_hashtable __read_mostly; +static struct list_head *mountpoint_hashtable __read_mostly; static struct kmem_cache *mnt_cache __read_mostly; static struct rw_semaphore namespace_sem; @@ -605,6 +607,51 @@ struct vfsmount *lookup_mnt(struct path *path) } } +static struct mountpoint *new_mountpoint(struct dentry *dentry) +{ + struct list_head *chain = mountpoint_hashtable + hash(NULL, dentry); + struct mountpoint *mp; + + list_for_each_entry(mp, chain, m_hash) { + if (mp->m_dentry == dentry) { + /* might be worth a WARN_ON() */ + if (d_unlinked(dentry)) + return ERR_PTR(-ENOENT); + mp->m_count++; + return mp; + } + } + + mp = kmalloc(sizeof(struct mountpoint), GFP_KERNEL); + if (!mp) + return ERR_PTR(-ENOMEM); + + spin_lock(&dentry->d_lock); + if (d_unlinked(dentry)) { + spin_unlock(&dentry->d_lock); + kfree(mp); + return ERR_PTR(-ENOENT); + } + dentry->d_flags |= DCACHE_MOUNTED; + spin_unlock(&dentry->d_lock); + mp->m_dentry = dentry; + mp->m_count = 1; + list_add(&mp->m_hash, chain); + return mp; +} + +static void put_mountpoint(struct mountpoint *mp) +{ + if (!--mp->m_count) { + struct dentry *dentry = mp->m_dentry; + spin_lock(&dentry->d_lock); + dentry->d_flags &= ~DCACHE_MOUNTED; + spin_unlock(&dentry->d_lock); + list_del(&mp->m_hash); + kfree(mp); + } +} + static inline int check_mnt(struct mount *mnt) { return mnt->mnt_ns == current->nsproxy->mnt_ns; @@ -633,27 +680,6 @@ static void __touch_mnt_namespace(struct mnt_namespace *ns) } /* - * Clear dentry's mounted state if it has no remaining mounts. - * vfsmount_lock must be held for write. - */ -static void dentry_reset_mounted(struct dentry *dentry) -{ - unsigned u; - - for (u = 0; u < HASH_SIZE; u++) { - struct mount *p; - - list_for_each_entry(p, &mount_hashtable[u], mnt_hash) { - if (p->mnt_mountpoint == dentry) - return; - } - } - spin_lock(&dentry->d_lock); - dentry->d_flags &= ~DCACHE_MOUNTED; - spin_unlock(&dentry->d_lock); -} - -/* * vfsmount lock must be held for write */ static void detach_mnt(struct mount *mnt, struct path *old_path) @@ -664,32 +690,35 @@ static void detach_mnt(struct mount *mnt, struct path *old_path) mnt->mnt_mountpoint = mnt->mnt.mnt_root; list_del_init(&mnt->mnt_child); list_del_init(&mnt->mnt_hash); - dentry_reset_mounted(old_path->dentry); + put_mountpoint(mnt->mnt_mp); + mnt->mnt_mp = NULL; } /* * vfsmount lock must be held for write */ -void mnt_set_mountpoint(struct mount *mnt, struct dentry *dentry, +void mnt_set_mountpoint(struct mount *mnt, + struct mountpoint *mp, struct mount *child_mnt) { + mp->m_count++; mnt_add_count(mnt, 1); /* essentially, that's mntget */ - child_mnt->mnt_mountpoint = dget(dentry); + child_mnt->mnt_mountpoint = dget(mp->m_dentry); child_mnt->mnt_parent = mnt; - spin_lock(&dentry->d_lock); - dentry->d_flags |= DCACHE_MOUNTED; - spin_unlock(&dentry->d_lock); + child_mnt->mnt_mp = mp; } /* * vfsmount lock must be held for write */ -static void attach_mnt(struct mount *mnt, struct path *path) +static void attach_mnt(struct mount *mnt, + struct mount *parent, + struct mountpoint *mp) { - mnt_set_mountpoint(real_mount(path->mnt), path->dentry, mnt); + mnt_set_mountpoint(parent, mp, mnt); list_add_tail(&mnt->mnt_hash, mount_hashtable + - hash(path->mnt, path->dentry)); - list_add_tail(&mnt->mnt_child, &real_mount(path->mnt)->mnt_mounts); + hash(&parent->mnt, mp->m_dentry)); + list_add_tail(&mnt->mnt_child, &parent->mnt_mounts); } /* @@ -1095,11 +1124,23 @@ int may_umount(struct vfsmount *mnt) EXPORT_SYMBOL(may_umount); -void release_mounts(struct list_head *head) +static LIST_HEAD(unmounted); /* protected by namespace_sem */ + +static void namespace_unlock(void) { struct mount *mnt; - while (!list_empty(head)) { - mnt = list_first_entry(head, struct mount, mnt_hash); + LIST_HEAD(head); + + if (likely(list_empty(&unmounted))) { + up_write(&namespace_sem); + return; + } + + list_splice_init(&unmounted, &head); + up_write(&namespace_sem); + + while (!list_empty(&head)) { + mnt = list_first_entry(&head, struct mount, mnt_hash); list_del_init(&mnt->mnt_hash); if (mnt_has_parent(mnt)) { struct dentry *dentry; @@ -1119,11 +1160,16 @@ void release_mounts(struct list_head *head) } } +static inline void namespace_lock(void) +{ + down_write(&namespace_sem); +} + /* * vfsmount lock must be held for write * namespace_sem must be held for write */ -void umount_tree(struct mount *mnt, int propagate, struct list_head *kill) +void umount_tree(struct mount *mnt, int propagate) { LIST_HEAD(tmp_list); struct mount *p; @@ -1142,20 +1188,20 @@ void umount_tree(struct mount *mnt, int propagate, struct list_head *kill) list_del_init(&p->mnt_child); if (mnt_has_parent(p)) { p->mnt_parent->mnt_ghosts++; - dentry_reset_mounted(p->mnt_mountpoint); + put_mountpoint(p->mnt_mp); + p->mnt_mp = NULL; } change_mnt_propagation(p, MS_PRIVATE); } - list_splice(&tmp_list, kill); + list_splice(&tmp_list, &unmounted); } -static void shrink_submounts(struct mount *mnt, struct list_head *umounts); +static void shrink_submounts(struct mount *mnt); static int do_umount(struct mount *mnt, int flags) { struct super_block *sb = mnt->mnt.mnt_sb; int retval; - LIST_HEAD(umount_list); retval = security_sb_umount(&mnt->mnt, flags); if (retval) @@ -1222,22 +1268,21 @@ static int do_umount(struct mount *mnt, int flags) return retval; } - down_write(&namespace_sem); + namespace_lock(); br_write_lock(&vfsmount_lock); event++; if (!(flags & MNT_DETACH)) - shrink_submounts(mnt, &umount_list); + shrink_submounts(mnt); retval = -EBUSY; if (flags & MNT_DETACH || !propagate_mount_busy(mnt, 2)) { if (!list_empty(&mnt->mnt_list)) - umount_tree(mnt, 1, &umount_list); + umount_tree(mnt, 1); retval = 0; } br_write_unlock(&vfsmount_lock); - up_write(&namespace_sem); - release_mounts(&umount_list); + namespace_unlock(); return retval; } @@ -1310,13 +1355,13 @@ static bool mnt_ns_loop(struct path *path) * mount namespace loop? */ struct inode *inode = path->dentry->d_inode; - struct proc_inode *ei; + struct proc_ns *ei; struct mnt_namespace *mnt_ns; if (!proc_ns_inode(inode)) return false; - ei = PROC_I(inode); + ei = get_proc_ns(inode); if (ei->ns_ops != &mntns_operations) return false; @@ -1327,8 +1372,7 @@ static bool mnt_ns_loop(struct path *path) struct mount *copy_tree(struct mount *mnt, struct dentry *dentry, int flag) { - struct mount *res, *p, *q, *r; - struct path path; + struct mount *res, *p, *q, *r, *parent; if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(mnt)) return ERR_PTR(-EINVAL); @@ -1355,25 +1399,22 @@ struct mount *copy_tree(struct mount *mnt, struct dentry *dentry, q = q->mnt_parent; } p = s; - path.mnt = &q->mnt; - path.dentry = p->mnt_mountpoint; + parent = q; q = clone_mnt(p, p->mnt.mnt_root, flag); if (IS_ERR(q)) goto out; br_write_lock(&vfsmount_lock); list_add_tail(&q->mnt_list, &res->mnt_list); - attach_mnt(q, &path); + attach_mnt(q, parent, p->mnt_mp); br_write_unlock(&vfsmount_lock); } } return res; out: if (res) { - LIST_HEAD(umount_list); br_write_lock(&vfsmount_lock); - umount_tree(res, 0, &umount_list); + umount_tree(res, 0); br_write_unlock(&vfsmount_lock); - release_mounts(&umount_list); } return q; } @@ -1383,10 +1424,10 @@ out: struct vfsmount *collect_mounts(struct path *path) { struct mount *tree; - down_write(&namespace_sem); + namespace_lock(); tree = copy_tree(real_mount(path->mnt), path->dentry, CL_COPY_ALL | CL_PRIVATE); - up_write(&namespace_sem); + namespace_unlock(); if (IS_ERR(tree)) return NULL; return &tree->mnt; @@ -1394,13 +1435,11 @@ struct vfsmount *collect_mounts(struct path *path) void drop_collected_mounts(struct vfsmount *mnt) { - LIST_HEAD(umount_list); - down_write(&namespace_sem); + namespace_lock(); br_write_lock(&vfsmount_lock); - umount_tree(real_mount(mnt), 0, &umount_list); + umount_tree(real_mount(mnt), 0); br_write_unlock(&vfsmount_lock); - up_write(&namespace_sem); - release_mounts(&umount_list); + namespace_unlock(); } int iterate_mounts(int (*f)(struct vfsmount *, void *), void *arg, @@ -1509,11 +1548,11 @@ static int invent_group_ids(struct mount *mnt, bool recurse) * in allocations. */ static int attach_recursive_mnt(struct mount *source_mnt, - struct path *path, struct path *parent_path) + struct mount *dest_mnt, + struct mountpoint *dest_mp, + struct path *parent_path) { LIST_HEAD(tree_list); - struct mount *dest_mnt = real_mount(path->mnt); - struct dentry *dest_dentry = path->dentry; struct mount *child, *p; int err; @@ -1522,7 +1561,7 @@ static int attach_recursive_mnt(struct mount *source_mnt, if (err) goto out; } - err = propagate_mnt(dest_mnt, dest_dentry, source_mnt, &tree_list); + err = propagate_mnt(dest_mnt, dest_mp, source_mnt, &tree_list); if (err) goto out_cleanup_ids; @@ -1534,10 +1573,10 @@ static int attach_recursive_mnt(struct mount *source_mnt, } if (parent_path) { detach_mnt(source_mnt, parent_path); - attach_mnt(source_mnt, path); + attach_mnt(source_mnt, dest_mnt, dest_mp); touch_mnt_namespace(source_mnt->mnt_ns); } else { - mnt_set_mountpoint(dest_mnt, dest_dentry, source_mnt); + mnt_set_mountpoint(dest_mnt, dest_mp, source_mnt); commit_tree(source_mnt); } @@ -1556,46 +1595,53 @@ static int attach_recursive_mnt(struct mount *source_mnt, return err; } -static int lock_mount(struct path *path) +static struct mountpoint *lock_mount(struct path *path) { struct vfsmount *mnt; + struct dentry *dentry = path->dentry; retry: - mutex_lock(&path->dentry->d_inode->i_mutex); - if (unlikely(cant_mount(path->dentry))) { - mutex_unlock(&path->dentry->d_inode->i_mutex); - return -ENOENT; + mutex_lock(&dentry->d_inode->i_mutex); + if (unlikely(cant_mount(dentry))) { + mutex_unlock(&dentry->d_inode->i_mutex); + return ERR_PTR(-ENOENT); } - down_write(&namespace_sem); + namespace_lock(); mnt = lookup_mnt(path); - if (likely(!mnt)) - return 0; - up_write(&namespace_sem); + if (likely(!mnt)) { + struct mountpoint *mp = new_mountpoint(dentry); + if (IS_ERR(mp)) { + namespace_unlock(); + mutex_unlock(&dentry->d_inode->i_mutex); + return mp; + } + return mp; + } + namespace_unlock(); mutex_unlock(&path->dentry->d_inode->i_mutex); path_put(path); path->mnt = mnt; - path->dentry = dget(mnt->mnt_root); + dentry = path->dentry = dget(mnt->mnt_root); goto retry; } -static void unlock_mount(struct path *path) +static void unlock_mount(struct mountpoint *where) { - up_write(&namespace_sem); - mutex_unlock(&path->dentry->d_inode->i_mutex); + struct dentry *dentry = where->m_dentry; + put_mountpoint(where); + namespace_unlock(); + mutex_unlock(&dentry->d_inode->i_mutex); } -static int graft_tree(struct mount *mnt, struct path *path) +static int graft_tree(struct mount *mnt, struct mount *p, struct mountpoint *mp) { if (mnt->mnt.mnt_sb->s_flags & MS_NOUSER) return -EINVAL; - if (S_ISDIR(path->dentry->d_inode->i_mode) != + if (S_ISDIR(mp->m_dentry->d_inode->i_mode) != S_ISDIR(mnt->mnt.mnt_root->d_inode->i_mode)) return -ENOTDIR; - if (d_unlinked(path->dentry)) - return -ENOENT; - - return attach_recursive_mnt(mnt, path, NULL); + return attach_recursive_mnt(mnt, p, mp, NULL); } /* @@ -1633,7 +1679,7 @@ static int do_change_type(struct path *path, int flag) if (!type) return -EINVAL; - down_write(&namespace_sem); + namespace_lock(); if (type == MS_SHARED) { err = invent_group_ids(mnt, recurse); if (err) @@ -1646,7 +1692,7 @@ static int do_change_type(struct path *path, int flag) br_write_unlock(&vfsmount_lock); out_unlock: - up_write(&namespace_sem); + namespace_unlock(); return err; } @@ -1656,9 +1702,9 @@ static int do_change_type(struct path *path, int flag) static int do_loopback(struct path *path, const char *old_name, int recurse) { - LIST_HEAD(umount_list); struct path old_path; - struct mount *mnt = NULL, *old; + struct mount *mnt = NULL, *old, *parent; + struct mountpoint *mp; int err; if (!old_name || !*old_name) return -EINVAL; @@ -1670,17 +1716,19 @@ static int do_loopback(struct path *path, const char *old_name, if (mnt_ns_loop(&old_path)) goto out; - err = lock_mount(path); - if (err) + mp = lock_mount(path); + err = PTR_ERR(mp); + if (IS_ERR(mp)) goto out; old = real_mount(old_path.mnt); + parent = real_mount(path->mnt); err = -EINVAL; if (IS_MNT_UNBINDABLE(old)) goto out2; - if (!check_mnt(real_mount(path->mnt)) || !check_mnt(old)) + if (!check_mnt(parent) || !check_mnt(old)) goto out2; if (recurse) @@ -1693,15 +1741,14 @@ static int do_loopback(struct path *path, const char *old_name, goto out2; } - err = graft_tree(mnt, path); + err = graft_tree(mnt, parent, mp); if (err) { br_write_lock(&vfsmount_lock); - umount_tree(mnt, 0, &umount_list); + umount_tree(mnt, 0); br_write_unlock(&vfsmount_lock); } out2: - unlock_mount(path); - release_mounts(&umount_list); + unlock_mount(mp); out: path_put(&old_path); return err; @@ -1786,6 +1833,7 @@ static int do_move_mount(struct path *path, const char *old_name) struct path old_path, parent_path; struct mount *p; struct mount *old; + struct mountpoint *mp; int err; if (!old_name || !*old_name) return -EINVAL; @@ -1793,8 +1841,9 @@ static int do_move_mount(struct path *path, const char *old_name) if (err) return err; - err = lock_mount(path); - if (err < 0) + mp = lock_mount(path); + err = PTR_ERR(mp); + if (IS_ERR(mp)) goto out; old = real_mount(old_path.mnt); @@ -1804,9 +1853,6 @@ static int do_move_mount(struct path *path, const char *old_name) if (!check_mnt(p) || !check_mnt(old)) goto out1; - if (d_unlinked(path->dentry)) - goto out1; - err = -EINVAL; if (old_path.dentry != old_path.mnt->mnt_root) goto out1; @@ -1833,7 +1879,7 @@ static int do_move_mount(struct path *path, const char *old_name) if (p == old) goto out1; - err = attach_recursive_mnt(old, path, &parent_path); + err = attach_recursive_mnt(old, real_mount(path->mnt), mp, &parent_path); if (err) goto out1; @@ -1841,7 +1887,7 @@ static int do_move_mount(struct path *path, const char *old_name) * automatically */ list_del_init(&old->mnt_expire); out1: - unlock_mount(path); + unlock_mount(mp); out: if (!err) path_put(&parent_path); @@ -1877,21 +1923,24 @@ static struct vfsmount *fs_set_subtype(struct vfsmount *mnt, const char *fstype) */ static int do_add_mount(struct mount *newmnt, struct path *path, int mnt_flags) { + struct mountpoint *mp; + struct mount *parent; int err; mnt_flags &= ~(MNT_SHARED | MNT_WRITE_HOLD | MNT_INTERNAL); - err = lock_mount(path); - if (err) - return err; + mp = lock_mount(path); + if (IS_ERR(mp)) + return PTR_ERR(mp); + parent = real_mount(path->mnt); err = -EINVAL; - if (unlikely(!check_mnt(real_mount(path->mnt)))) { + if (unlikely(!check_mnt(parent))) { /* that's acceptable only for automounts done in private ns */ if (!(mnt_flags & MNT_SHRINKABLE)) goto unlock; /* ... and for those we'd better have mountpoint still alive */ - if (!real_mount(path->mnt)->mnt_ns) + if (!parent->mnt_ns) goto unlock; } @@ -1906,10 +1955,10 @@ static int do_add_mount(struct mount *newmnt, struct path *path, int mnt_flags) goto unlock; newmnt->mnt.mnt_flags = mnt_flags; - err = graft_tree(newmnt, path); + err = graft_tree(newmnt, parent, mp); unlock: - unlock_mount(path); + unlock_mount(mp); return err; } @@ -1982,11 +2031,11 @@ int finish_automount(struct vfsmount *m, struct path *path) fail: /* remove m from any expiration list it may be on */ if (!list_empty(&mnt->mnt_expire)) { - down_write(&namespace_sem); + namespace_lock(); br_write_lock(&vfsmount_lock); list_del_init(&mnt->mnt_expire); br_write_unlock(&vfsmount_lock); - up_write(&namespace_sem); + namespace_unlock(); } mntput(m); mntput(m); @@ -2000,13 +2049,13 @@ fail: */ void mnt_set_expiry(struct vfsmount *mnt, struct list_head *expiry_list) { - down_write(&namespace_sem); + namespace_lock(); br_write_lock(&vfsmount_lock); list_add_tail(&real_mount(mnt)->mnt_expire, expiry_list); br_write_unlock(&vfsmount_lock); - up_write(&namespace_sem); + namespace_unlock(); } EXPORT_SYMBOL(mnt_set_expiry); @@ -2019,12 +2068,11 @@ void mark_mounts_for_expiry(struct list_head *mounts) { struct mount *mnt, *next; LIST_HEAD(graveyard); - LIST_HEAD(umounts); if (list_empty(mounts)) return; - down_write(&namespace_sem); + namespace_lock(); br_write_lock(&vfsmount_lock); /* extract from the expiration list every vfsmount that matches the @@ -2042,12 +2090,10 @@ void mark_mounts_for_expiry(struct list_head *mounts) while (!list_empty(&graveyard)) { mnt = list_first_entry(&graveyard, struct mount, mnt_expire); touch_mnt_namespace(mnt->mnt_ns); - umount_tree(mnt, 1, &umounts); + umount_tree(mnt, 1); } br_write_unlock(&vfsmount_lock); - up_write(&namespace_sem); - - release_mounts(&umounts); + namespace_unlock(); } EXPORT_SYMBOL_GPL(mark_mounts_for_expiry); @@ -2104,7 +2150,7 @@ resume: * * vfsmount_lock must be held for write */ -static void shrink_submounts(struct mount *mnt, struct list_head *umounts) +static void shrink_submounts(struct mount *mnt) { LIST_HEAD(graveyard); struct mount *m; @@ -2115,7 +2161,7 @@ static void shrink_submounts(struct mount *mnt, struct list_head *umounts) m = list_first_entry(&graveyard, struct mount, mnt_expire); touch_mnt_namespace(m->mnt_ns); - umount_tree(m, 1, umounts); + umount_tree(m, 1); } } } @@ -2342,14 +2388,14 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, if (IS_ERR(new_ns)) return new_ns; - down_write(&namespace_sem); + namespace_lock(); /* First pass: copy the tree topology */ copy_flags = CL_COPY_ALL | CL_EXPIRE; if (user_ns != mnt_ns->user_ns) copy_flags |= CL_SHARED_TO_SLAVE | CL_UNPRIVILEGED; new = copy_tree(old, old->mnt.mnt_root, copy_flags); if (IS_ERR(new)) { - up_write(&namespace_sem); + namespace_unlock(); free_mnt_ns(new_ns); return ERR_CAST(new); } @@ -2380,7 +2426,7 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, p = next_mnt(p, old); q = next_mnt(q, new); } - up_write(&namespace_sem); + namespace_unlock(); if (rootmnt) mntput(rootmnt); @@ -2550,7 +2596,8 @@ SYSCALL_DEFINE2(pivot_root, const char __user *, new_root, const char __user *, put_old) { struct path new, old, parent_path, root_parent, root; - struct mount *new_mnt, *root_mnt; + struct mount *new_mnt, *root_mnt, *old_mnt; + struct mountpoint *old_mp, *root_mp; int error; if (!may_mount()) @@ -2569,14 +2616,16 @@ SYSCALL_DEFINE2(pivot_root, const char __user *, new_root, goto out2; get_fs_root(current->fs, &root); - error = lock_mount(&old); - if (error) + old_mp = lock_mount(&old); + error = PTR_ERR(old_mp); + if (IS_ERR(old_mp)) goto out3; error = -EINVAL; new_mnt = real_mount(new.mnt); root_mnt = real_mount(root.mnt); - if (IS_MNT_SHARED(real_mount(old.mnt)) || + old_mnt = real_mount(old.mnt); + if (IS_MNT_SHARED(old_mnt) || IS_MNT_SHARED(new_mnt->mnt_parent) || IS_MNT_SHARED(root_mnt->mnt_parent)) goto out4; @@ -2585,37 +2634,37 @@ SYSCALL_DEFINE2(pivot_root, const char __user *, new_root, error = -ENOENT; if (d_unlinked(new.dentry)) goto out4; - if (d_unlinked(old.dentry)) - goto out4; error = -EBUSY; - if (new.mnt == root.mnt || - old.mnt == root.mnt) + if (new_mnt == root_mnt || old_mnt == root_mnt) goto out4; /* loop, on the same file system */ error = -EINVAL; if (root.mnt->mnt_root != root.dentry) goto out4; /* not a mountpoint */ if (!mnt_has_parent(root_mnt)) goto out4; /* not attached */ + root_mp = root_mnt->mnt_mp; if (new.mnt->mnt_root != new.dentry) goto out4; /* not a mountpoint */ if (!mnt_has_parent(new_mnt)) goto out4; /* not attached */ /* make sure we can reach put_old from new_root */ - if (!is_path_reachable(real_mount(old.mnt), old.dentry, &new)) + if (!is_path_reachable(old_mnt, old.dentry, &new)) goto out4; + root_mp->m_count++; /* pin it so it won't go away */ br_write_lock(&vfsmount_lock); detach_mnt(new_mnt, &parent_path); detach_mnt(root_mnt, &root_parent); /* mount old root on put_old */ - attach_mnt(root_mnt, &old); + attach_mnt(root_mnt, old_mnt, old_mp); /* mount new_root on / */ - attach_mnt(new_mnt, &root_parent); + attach_mnt(new_mnt, real_mount(root_parent.mnt), root_mp); touch_mnt_namespace(current->nsproxy->mnt_ns); br_write_unlock(&vfsmount_lock); chroot_fs_refs(&root, &new); + put_mountpoint(root_mp); error = 0; out4: - unlock_mount(&old); + unlock_mount(old_mp); if (!error) { path_put(&root_parent); path_put(&parent_path); @@ -2670,14 +2719,17 @@ void __init mnt_init(void) 0, SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL); mount_hashtable = (struct list_head *)__get_free_page(GFP_ATOMIC); + mountpoint_hashtable = (struct list_head *)__get_free_page(GFP_ATOMIC); - if (!mount_hashtable) + if (!mount_hashtable || !mountpoint_hashtable) panic("Failed to allocate mount hash table\n"); printk(KERN_INFO "Mount-cache hash table entries: %lu\n", HASH_SIZE); for (u = 0; u < HASH_SIZE; u++) INIT_LIST_HEAD(&mount_hashtable[u]); + for (u = 0; u < HASH_SIZE; u++) + INIT_LIST_HEAD(&mountpoint_hashtable[u]); br_lock_init(&vfsmount_lock); @@ -2694,16 +2746,13 @@ void __init mnt_init(void) void put_mnt_ns(struct mnt_namespace *ns) { - LIST_HEAD(umount_list); - if (!atomic_dec_and_test(&ns->count)) return; - down_write(&namespace_sem); + namespace_lock(); br_write_lock(&vfsmount_lock); - umount_tree(ns->root, 0, &umount_list); + umount_tree(ns->root, 0); br_write_unlock(&vfsmount_lock); - up_write(&namespace_sem); - release_mounts(&umount_list); + namespace_unlock(); free_mnt_ns(ns); } diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index f33455b4d957..5bee0313dffd 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -177,7 +177,7 @@ static int export_features_open(struct inode *inode, struct file *file) return single_open(file, export_features_show, NULL); } -static struct file_operations export_features_operations = { +static const struct file_operations export_features_operations = { .open = export_features_open, .read = seq_read, .llseek = seq_lseek, @@ -196,7 +196,7 @@ static int supported_enctypes_open(struct inode *inode, struct file *file) return single_open(file, supported_enctypes_show, NULL); } -static struct file_operations supported_enctypes_ops = { +static const struct file_operations supported_enctypes_ops = { .open = supported_enctypes_open, .read = seq_read, .llseek = seq_lseek, diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c index c616a70e8cf9..959815c1e017 100644 --- a/fs/notify/inotify/inotify_user.c +++ b/fs/notify/inotify/inotify_user.c @@ -287,9 +287,6 @@ static int inotify_release(struct inode *ignored, struct file *file) pr_debug("%s: group=%p\n", __func__, group); - if (file->f_flags & FASYNC) - fsnotify_fasync(-1, file, 0); - /* free this group, matching get was inotify_init->fsnotify_obtain_group */ fsnotify_destroy_group(group); diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c index 5b2d4f0853ac..1da4b81e6f76 100644 --- a/fs/ntfs/file.c +++ b/fs/ntfs/file.c @@ -2129,7 +2129,6 @@ static ssize_t ntfs_file_aio_write(struct kiocb *iocb, const struct iovec *iov, BUG_ON(iocb->ki_pos != pos); - sb_start_write(inode->i_sb); mutex_lock(&inode->i_mutex); ret = ntfs_file_aio_write_nolock(iocb, iov, nr_segs, &iocb->ki_pos); mutex_unlock(&inode->i_mutex); @@ -2138,7 +2137,6 @@ static ssize_t ntfs_file_aio_write(struct kiocb *iocb, const struct iovec *iov, if (err < 0) ret = err; } - sb_end_write(inode->i_sb); return ret; } diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 6474cb44004d..8a7509f9e6f5 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -2248,8 +2248,6 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb, if (iocb->ki_left == 0) return 0; - sb_start_write(inode->i_sb); - appending = file->f_flags & O_APPEND ? 1 : 0; direct_io = file->f_flags & O_DIRECT ? 1 : 0; @@ -2423,7 +2421,6 @@ out_sems: ocfs2_iocb_clear_sem_locked(iocb); mutex_unlock(&inode->i_mutex); - sb_end_write(inode->i_sb); if (written) ret = written; @@ -2468,8 +2465,7 @@ static ssize_t ocfs2_file_splice_write(struct pipe_inode_info *pipe, out->f_path.dentry->d_name.len, out->f_path.dentry->d_name.name, len); - if (pipe->inode) - mutex_lock_nested(&pipe->inode->i_mutex, I_MUTEX_PARENT); + pipe_lock(pipe); splice_from_pipe_begin(&sd); do { @@ -2489,8 +2485,7 @@ static ssize_t ocfs2_file_splice_write(struct pipe_inode_info *pipe, } while (ret > 0); splice_from_pipe_end(pipe, &sd); - if (pipe->inode) - mutex_unlock(&pipe->inode->i_mutex); + pipe_unlock(pipe); if (sd.num_spliced) ret = sd.num_spliced; diff --git a/fs/pipe.c b/fs/pipe.c index 2234f3f61f8d..a029a14bacf1 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -25,6 +25,8 @@ #include <asm/uaccess.h> #include <asm/ioctls.h> +#include "internal.h" + /* * The max size that a non-root user is allowed to grow the pipe. Can * be set by root in /proc/sys/fs/pipe-max-size @@ -53,8 +55,8 @@ unsigned int pipe_min_size = PAGE_SIZE; static void pipe_lock_nested(struct pipe_inode_info *pipe, int subclass) { - if (pipe->inode) - mutex_lock_nested(&pipe->inode->i_mutex, subclass); + if (pipe->files) + mutex_lock_nested(&pipe->mutex, subclass); } void pipe_lock(struct pipe_inode_info *pipe) @@ -68,11 +70,21 @@ EXPORT_SYMBOL(pipe_lock); void pipe_unlock(struct pipe_inode_info *pipe) { - if (pipe->inode) - mutex_unlock(&pipe->inode->i_mutex); + if (pipe->files) + mutex_unlock(&pipe->mutex); } EXPORT_SYMBOL(pipe_unlock); +static inline void __pipe_lock(struct pipe_inode_info *pipe) +{ + mutex_lock_nested(&pipe->mutex, I_MUTEX_PARENT); +} + +static inline void __pipe_unlock(struct pipe_inode_info *pipe) +{ + mutex_unlock(&pipe->mutex); +} + void pipe_double_lock(struct pipe_inode_info *pipe1, struct pipe_inode_info *pipe2) { @@ -361,8 +373,7 @@ pipe_read(struct kiocb *iocb, const struct iovec *_iov, unsigned long nr_segs, loff_t pos) { struct file *filp = iocb->ki_filp; - struct inode *inode = file_inode(filp); - struct pipe_inode_info *pipe; + struct pipe_inode_info *pipe = filp->private_data; int do_wakeup; ssize_t ret; struct iovec *iov = (struct iovec *)_iov; @@ -375,8 +386,7 @@ pipe_read(struct kiocb *iocb, const struct iovec *_iov, do_wakeup = 0; ret = 0; - mutex_lock(&inode->i_mutex); - pipe = inode->i_pipe; + __pipe_lock(pipe); for (;;) { int bufs = pipe->nrbufs; if (bufs) { @@ -464,7 +474,7 @@ redo: } pipe_wait(pipe); } - mutex_unlock(&inode->i_mutex); + __pipe_unlock(pipe); /* Signal writers asynchronously that there is more room. */ if (do_wakeup) { @@ -486,8 +496,7 @@ pipe_write(struct kiocb *iocb, const struct iovec *_iov, unsigned long nr_segs, loff_t ppos) { struct file *filp = iocb->ki_filp; - struct inode *inode = file_inode(filp); - struct pipe_inode_info *pipe; + struct pipe_inode_info *pipe = filp->private_data; ssize_t ret; int do_wakeup; struct iovec *iov = (struct iovec *)_iov; @@ -501,8 +510,7 @@ pipe_write(struct kiocb *iocb, const struct iovec *_iov, do_wakeup = 0; ret = 0; - mutex_lock(&inode->i_mutex); - pipe = inode->i_pipe; + __pipe_lock(pipe); if (!pipe->readers) { send_sig(SIGPIPE, current, 0); @@ -649,7 +657,7 @@ redo2: pipe->waiting_writers--; } out: - mutex_unlock(&inode->i_mutex); + __pipe_unlock(pipe); if (do_wakeup) { wake_up_interruptible_sync_poll(&pipe->wait, POLLIN | POLLRDNORM); kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); @@ -662,29 +670,14 @@ out: return ret; } -static ssize_t -bad_pipe_r(struct file *filp, char __user *buf, size_t count, loff_t *ppos) -{ - return -EBADF; -} - -static ssize_t -bad_pipe_w(struct file *filp, const char __user *buf, size_t count, - loff_t *ppos) -{ - return -EBADF; -} - static long pipe_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { - struct inode *inode = file_inode(filp); - struct pipe_inode_info *pipe; + struct pipe_inode_info *pipe = filp->private_data; int count, buf, nrbufs; switch (cmd) { case FIONREAD: - mutex_lock(&inode->i_mutex); - pipe = inode->i_pipe; + __pipe_lock(pipe); count = 0; buf = pipe->curbuf; nrbufs = pipe->nrbufs; @@ -692,7 +685,7 @@ static long pipe_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) count += pipe->bufs[buf].len; buf = (buf+1) & (pipe->buffers - 1); } - mutex_unlock(&inode->i_mutex); + __pipe_unlock(pipe); return put_user(count, (int __user *)arg); default: @@ -705,8 +698,7 @@ static unsigned int pipe_poll(struct file *filp, poll_table *wait) { unsigned int mask; - struct inode *inode = file_inode(filp); - struct pipe_inode_info *pipe = inode->i_pipe; + struct pipe_inode_info *pipe = filp->private_data; int nrbufs; poll_wait(filp, &pipe->wait, wait); @@ -734,197 +726,56 @@ pipe_poll(struct file *filp, poll_table *wait) } static int -pipe_release(struct inode *inode, int decr, int decw) +pipe_release(struct inode *inode, struct file *file) { - struct pipe_inode_info *pipe; + struct pipe_inode_info *pipe = inode->i_pipe; + int kill = 0; - mutex_lock(&inode->i_mutex); - pipe = inode->i_pipe; - pipe->readers -= decr; - pipe->writers -= decw; + __pipe_lock(pipe); + if (file->f_mode & FMODE_READ) + pipe->readers--; + if (file->f_mode & FMODE_WRITE) + pipe->writers--; - if (!pipe->readers && !pipe->writers) { - free_pipe_info(inode); - } else { + if (pipe->readers || pipe->writers) { wake_up_interruptible_sync_poll(&pipe->wait, POLLIN | POLLOUT | POLLRDNORM | POLLWRNORM | POLLERR | POLLHUP); kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT); } - mutex_unlock(&inode->i_mutex); - - return 0; -} - -static int -pipe_read_fasync(int fd, struct file *filp, int on) -{ - struct inode *inode = file_inode(filp); - int retval; - - mutex_lock(&inode->i_mutex); - retval = fasync_helper(fd, filp, on, &inode->i_pipe->fasync_readers); - mutex_unlock(&inode->i_mutex); - - return retval; -} - - -static int -pipe_write_fasync(int fd, struct file *filp, int on) -{ - struct inode *inode = file_inode(filp); - int retval; + spin_lock(&inode->i_lock); + if (!--pipe->files) { + inode->i_pipe = NULL; + kill = 1; + } + spin_unlock(&inode->i_lock); + __pipe_unlock(pipe); - mutex_lock(&inode->i_mutex); - retval = fasync_helper(fd, filp, on, &inode->i_pipe->fasync_writers); - mutex_unlock(&inode->i_mutex); + if (kill) + free_pipe_info(pipe); - return retval; + return 0; } - static int -pipe_rdwr_fasync(int fd, struct file *filp, int on) +pipe_fasync(int fd, struct file *filp, int on) { - struct inode *inode = file_inode(filp); - struct pipe_inode_info *pipe = inode->i_pipe; - int retval; + struct pipe_inode_info *pipe = filp->private_data; + int retval = 0; - mutex_lock(&inode->i_mutex); - retval = fasync_helper(fd, filp, on, &pipe->fasync_readers); - if (retval >= 0) { + __pipe_lock(pipe); + if (filp->f_mode & FMODE_READ) + retval = fasync_helper(fd, filp, on, &pipe->fasync_readers); + if ((filp->f_mode & FMODE_WRITE) && retval >= 0) { retval = fasync_helper(fd, filp, on, &pipe->fasync_writers); - if (retval < 0) /* this can happen only if on == T */ + if (retval < 0 && (filp->f_mode & FMODE_READ)) + /* this can happen only if on == T */ fasync_helper(-1, filp, 0, &pipe->fasync_readers); } - mutex_unlock(&inode->i_mutex); + __pipe_unlock(pipe); return retval; } - -static int -pipe_read_release(struct inode *inode, struct file *filp) -{ - return pipe_release(inode, 1, 0); -} - -static int -pipe_write_release(struct inode *inode, struct file *filp) -{ - return pipe_release(inode, 0, 1); -} - -static int -pipe_rdwr_release(struct inode *inode, struct file *filp) -{ - int decr, decw; - - decr = (filp->f_mode & FMODE_READ) != 0; - decw = (filp->f_mode & FMODE_WRITE) != 0; - return pipe_release(inode, decr, decw); -} - -static int -pipe_read_open(struct inode *inode, struct file *filp) -{ - int ret = -ENOENT; - - mutex_lock(&inode->i_mutex); - - if (inode->i_pipe) { - ret = 0; - inode->i_pipe->readers++; - } - - mutex_unlock(&inode->i_mutex); - - return ret; -} - -static int -pipe_write_open(struct inode *inode, struct file *filp) -{ - int ret = -ENOENT; - - mutex_lock(&inode->i_mutex); - - if (inode->i_pipe) { - ret = 0; - inode->i_pipe->writers++; - } - - mutex_unlock(&inode->i_mutex); - - return ret; -} - -static int -pipe_rdwr_open(struct inode *inode, struct file *filp) -{ - int ret = -ENOENT; - - if (!(filp->f_mode & (FMODE_READ|FMODE_WRITE))) - return -EINVAL; - - mutex_lock(&inode->i_mutex); - - if (inode->i_pipe) { - ret = 0; - if (filp->f_mode & FMODE_READ) - inode->i_pipe->readers++; - if (filp->f_mode & FMODE_WRITE) - inode->i_pipe->writers++; - } - - mutex_unlock(&inode->i_mutex); - - return ret; -} - -/* - * 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_pipefifo_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_pipefifo_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_pipefifo_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, -}; - -struct pipe_inode_info * alloc_pipe_info(struct inode *inode) +struct pipe_inode_info *alloc_pipe_info(void) { struct pipe_inode_info *pipe; @@ -934,8 +785,8 @@ struct pipe_inode_info * alloc_pipe_info(struct inode *inode) if (pipe->bufs) { init_waitqueue_head(&pipe->wait); pipe->r_counter = pipe->w_counter = 1; - pipe->inode = inode; pipe->buffers = PIPE_DEF_BUFFERS; + mutex_init(&pipe->mutex); return pipe; } kfree(pipe); @@ -944,7 +795,7 @@ struct pipe_inode_info * alloc_pipe_info(struct inode *inode) return NULL; } -void __free_pipe_info(struct pipe_inode_info *pipe) +void free_pipe_info(struct pipe_inode_info *pipe) { int i; @@ -959,12 +810,6 @@ void __free_pipe_info(struct pipe_inode_info *pipe) kfree(pipe); } -void free_pipe_info(struct inode *inode) -{ - __free_pipe_info(inode->i_pipe); - inode->i_pipe = NULL; -} - static struct vfsmount *pipe_mnt __read_mostly; /* @@ -990,13 +835,14 @@ static struct inode * get_pipe_inode(void) inode->i_ino = get_next_ino(); - pipe = alloc_pipe_info(inode); + pipe = alloc_pipe_info(); if (!pipe) goto fail_iput; - inode->i_pipe = pipe; + inode->i_pipe = pipe; + pipe->files = 2; pipe->readers = pipe->writers = 1; - inode->i_fop = &rdwr_pipefifo_fops; + inode->i_fop = &pipefifo_fops; /* * Mark the inode dirty from the very beginning, @@ -1039,17 +885,19 @@ int create_pipe_files(struct file **res, int flags) d_instantiate(path.dentry, inode); err = -ENFILE; - f = alloc_file(&path, FMODE_WRITE, &write_pipefifo_fops); + f = alloc_file(&path, FMODE_WRITE, &pipefifo_fops); if (IS_ERR(f)) goto err_dentry; f->f_flags = O_WRONLY | (flags & (O_NONBLOCK | O_DIRECT)); + f->private_data = inode->i_pipe; - res[0] = alloc_file(&path, FMODE_READ, &read_pipefifo_fops); + res[0] = alloc_file(&path, FMODE_READ, &pipefifo_fops); if (IS_ERR(res[0])) goto err_file; path_get(&path); + res[0]->private_data = inode->i_pipe; res[0]->f_flags = O_RDONLY | (flags & O_NONBLOCK); res[1] = f; return 0; @@ -1057,12 +905,12 @@ int create_pipe_files(struct file **res, int flags) err_file: put_filp(f); err_dentry: - free_pipe_info(inode); + free_pipe_info(inode->i_pipe); path_put(&path); return err; err_inode: - free_pipe_info(inode); + free_pipe_info(inode->i_pipe); iput(inode); return err; } @@ -1144,6 +992,168 @@ SYSCALL_DEFINE1(pipe, int __user *, fildes) return sys_pipe2(fildes, 0); } +static int wait_for_partner(struct pipe_inode_info *pipe, unsigned int *cnt) +{ + int cur = *cnt; + + while (cur == *cnt) { + pipe_wait(pipe); + if (signal_pending(current)) + break; + } + return cur == *cnt ? -ERESTARTSYS : 0; +} + +static void wake_up_partner(struct pipe_inode_info *pipe) +{ + wake_up_interruptible(&pipe->wait); +} + +static int fifo_open(struct inode *inode, struct file *filp) +{ + struct pipe_inode_info *pipe; + bool is_pipe = inode->i_sb->s_magic == PIPEFS_MAGIC; + int kill = 0; + int ret; + + filp->f_version = 0; + + spin_lock(&inode->i_lock); + if (inode->i_pipe) { + pipe = inode->i_pipe; + pipe->files++; + spin_unlock(&inode->i_lock); + } else { + spin_unlock(&inode->i_lock); + pipe = alloc_pipe_info(); + if (!pipe) + return -ENOMEM; + pipe->files = 1; + spin_lock(&inode->i_lock); + if (unlikely(inode->i_pipe)) { + inode->i_pipe->files++; + spin_unlock(&inode->i_lock); + free_pipe_info(pipe); + pipe = inode->i_pipe; + } else { + inode->i_pipe = pipe; + spin_unlock(&inode->i_lock); + } + } + filp->private_data = pipe; + /* OK, we have a pipe and it's pinned down */ + + __pipe_lock(pipe); + + /* We can only do regular read/write on fifos */ + filp->f_mode &= (FMODE_READ | FMODE_WRITE); + + switch (filp->f_mode) { + case FMODE_READ: + /* + * O_RDONLY + * POSIX.1 says that O_NONBLOCK means return with the FIFO + * opened, even when there is no process writing the FIFO. + */ + pipe->r_counter++; + if (pipe->readers++ == 0) + wake_up_partner(pipe); + + if (!is_pipe && !pipe->writers) { + if ((filp->f_flags & O_NONBLOCK)) { + /* suppress POLLHUP until we have + * seen a writer */ + filp->f_version = pipe->w_counter; + } else { + if (wait_for_partner(pipe, &pipe->w_counter)) + goto err_rd; + } + } + break; + + case FMODE_WRITE: + /* + * O_WRONLY + * POSIX.1 says that O_NONBLOCK means return -1 with + * errno=ENXIO when there is no process reading the FIFO. + */ + ret = -ENXIO; + if (!is_pipe && (filp->f_flags & O_NONBLOCK) && !pipe->readers) + goto err; + + pipe->w_counter++; + if (!pipe->writers++) + wake_up_partner(pipe); + + if (!is_pipe && !pipe->readers) { + if (wait_for_partner(pipe, &pipe->r_counter)) + goto err_wr; + } + break; + + case FMODE_READ | FMODE_WRITE: + /* + * O_RDWR + * POSIX.1 leaves this case "undefined" when O_NONBLOCK is set. + * This implementation will NEVER block on a O_RDWR open, since + * the process can at least talk to itself. + */ + + pipe->readers++; + pipe->writers++; + pipe->r_counter++; + pipe->w_counter++; + if (pipe->readers == 1 || pipe->writers == 1) + wake_up_partner(pipe); + break; + + default: + ret = -EINVAL; + goto err; + } + + /* Ok! */ + __pipe_unlock(pipe); + return 0; + +err_rd: + if (!--pipe->readers) + wake_up_interruptible(&pipe->wait); + ret = -ERESTARTSYS; + goto err; + +err_wr: + if (!--pipe->writers) + wake_up_interruptible(&pipe->wait); + ret = -ERESTARTSYS; + goto err; + +err: + spin_lock(&inode->i_lock); + if (!--pipe->files) { + inode->i_pipe = NULL; + kill = 1; + } + spin_unlock(&inode->i_lock); + __pipe_unlock(pipe); + if (kill) + free_pipe_info(pipe); + return ret; +} + +const struct file_operations pipefifo_fops = { + .open = fifo_open, + .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, + .release = pipe_release, + .fasync = pipe_fasync, +}; + /* * Allocate a new array of pipe buffers and copy the info over. Returns the * pipe size if successful, or return -ERROR on error. @@ -1229,9 +1239,7 @@ int pipe_proc_fn(struct ctl_table *table, int write, void __user *buf, */ struct pipe_inode_info *get_pipe_info(struct file *file) { - struct inode *i = file_inode(file); - - return S_ISFIFO(i->i_mode) ? i->i_pipe : NULL; + return file->f_op == &pipefifo_fops ? file->private_data : NULL; } long pipe_fcntl(struct file *file, unsigned int cmd, unsigned long arg) @@ -1243,7 +1251,7 @@ long pipe_fcntl(struct file *file, unsigned int cmd, unsigned long arg) if (!pipe) return -EBADF; - mutex_lock(&pipe->inode->i_mutex); + __pipe_lock(pipe); switch (cmd) { case F_SETPIPE_SZ: { @@ -1272,7 +1280,7 @@ long pipe_fcntl(struct file *file, unsigned int cmd, unsigned long arg) } out: - mutex_unlock(&pipe->inode->i_mutex); + __pipe_unlock(pipe); return ret; } diff --git a/fs/pnode.c b/fs/pnode.c index 8b29d2164da6..3d2a7141b87a 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -218,7 +218,7 @@ static struct mount *get_source(struct mount *dest, * @source_mnt: source mount. * @tree_list : list of heads of trees to be attached. */ -int propagate_mnt(struct mount *dest_mnt, struct dentry *dest_dentry, +int propagate_mnt(struct mount *dest_mnt, struct mountpoint *dest_mp, struct mount *source_mnt, struct list_head *tree_list) { struct user_namespace *user_ns = current->nsproxy->mnt_ns->user_ns; @@ -227,7 +227,6 @@ int propagate_mnt(struct mount *dest_mnt, struct dentry *dest_dentry, struct mount *prev_dest_mnt = dest_mnt; struct mount *prev_src_mnt = source_mnt; LIST_HEAD(tmp_list); - LIST_HEAD(umount_list); for (m = propagation_next(dest_mnt, dest_mnt); m; m = propagation_next(m, dest_mnt)) { @@ -250,8 +249,8 @@ int propagate_mnt(struct mount *dest_mnt, struct dentry *dest_dentry, goto out; } - if (is_subdir(dest_dentry, m->mnt.mnt_root)) { - mnt_set_mountpoint(m, dest_dentry, child); + if (is_subdir(dest_mp->m_dentry, m->mnt.mnt_root)) { + mnt_set_mountpoint(m, dest_mp, child); list_add_tail(&child->mnt_hash, tree_list); } else { /* @@ -267,10 +266,9 @@ out: br_write_lock(&vfsmount_lock); while (!list_empty(&tmp_list)) { child = list_first_entry(&tmp_list, struct mount, mnt_hash); - umount_tree(child, 0, &umount_list); + umount_tree(child, 0); } br_write_unlock(&vfsmount_lock); - release_mounts(&umount_list); return ret; } diff --git a/fs/pnode.h b/fs/pnode.h index a0493d5ebfbf..b091445c1c4a 100644 --- a/fs/pnode.h +++ b/fs/pnode.h @@ -32,17 +32,16 @@ static inline void set_mnt_shared(struct mount *mnt) } void change_mnt_propagation(struct mount *, int); -int propagate_mnt(struct mount *, struct dentry *, struct mount *, +int propagate_mnt(struct mount *, struct mountpoint *, struct mount *, struct list_head *); int propagate_umount(struct list_head *); int propagate_mount_busy(struct mount *, int); void mnt_release_group_id(struct mount *); int get_dominating_id(struct mount *mnt, const struct path *root); unsigned int mnt_get_count(struct mount *mnt); -void mnt_set_mountpoint(struct mount *, struct dentry *, +void mnt_set_mountpoint(struct mount *, struct mountpoint *, struct mount *); -void release_mounts(struct list_head *); -void umount_tree(struct mount *, int, struct list_head *); +void umount_tree(struct mount *, int); struct mount *copy_tree(struct mount *, struct dentry *, int); bool is_path_reachable(struct mount *, struct dentry *, const struct path *root); diff --git a/fs/proc/base.c b/fs/proc/base.c index 3861bcec41ff..dd51e50001fe 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -405,6 +405,37 @@ static const struct file_operations proc_lstats_operations = { #endif +#ifdef CONFIG_CGROUPS +static int cgroup_open(struct inode *inode, struct file *file) +{ + struct pid *pid = PROC_I(inode)->pid; + return single_open(file, proc_cgroup_show, pid); +} + +static const struct file_operations proc_cgroup_operations = { + .open = cgroup_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; +#endif + +#ifdef CONFIG_PROC_PID_CPUSET + +static int cpuset_open(struct inode *inode, struct file *file) +{ + struct pid *pid = PROC_I(inode)->pid; + return single_open(file, proc_cpuset_show, pid); +} + +static const struct file_operations proc_cpuset_operations = { + .open = cpuset_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; +#endif + static int proc_oom_score(struct task_struct *task, char *buffer) { unsigned long totalpages = totalram_pages + total_swap_pages; @@ -1621,6 +1652,15 @@ int pid_revalidate(struct dentry *dentry, unsigned int flags) return 0; } +int pid_delete_dentry(const struct dentry *dentry) +{ + /* Is the task we represent dead? + * If so, then don't put the dentry on the lru list, + * kill it immediately. + */ + return !proc_pid(dentry->d_inode)->tasks[PIDTYPE_PID].first; +} + const struct dentry_operations pid_dentry_operations = { .d_revalidate = pid_revalidate, @@ -2893,7 +2933,7 @@ retry: return iter; } -#define TGID_OFFSET (FIRST_PROCESS_ENTRY) +#define TGID_OFFSET (FIRST_PROCESS_ENTRY + 1) static int proc_pid_fill_cache(struct file *filp, void *dirent, filldir_t filldir, struct tgid_iter iter) @@ -2916,13 +2956,21 @@ int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir) struct tgid_iter iter; struct pid_namespace *ns; filldir_t __filldir; + loff_t pos = filp->f_pos; - if (filp->f_pos >= PID_MAX_LIMIT + TGID_OFFSET) + if (pos >= PID_MAX_LIMIT + TGID_OFFSET) goto out; - ns = filp->f_dentry->d_sb->s_fs_info; + if (pos == TGID_OFFSET - 1) { + if (proc_fill_cache(filp, dirent, filldir, "self", 4, + NULL, NULL, NULL) < 0) + goto out; + iter.tgid = 0; + } else { + iter.tgid = pos - TGID_OFFSET; + } iter.task = NULL; - iter.tgid = filp->f_pos - TGID_OFFSET; + ns = filp->f_dentry->d_sb->s_fs_info; for (iter = next_tgid(ns, iter); iter.task; iter.tgid += 1, iter = next_tgid(ns, iter)) { diff --git a/fs/proc/fd.h b/fs/proc/fd.h index cbb1d47deda8..7c047f256ae2 100644 --- a/fs/proc/fd.h +++ b/fs/proc/fd.h @@ -11,4 +11,9 @@ extern const struct inode_operations proc_fdinfo_inode_operations; extern int proc_fd_permission(struct inode *inode, int mask); +static inline int proc_fd(struct inode *inode) +{ + return PROC_I(inode)->fd; +} + #endif /* __PROCFS_FD_H__ */ diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 21e1a8f1659d..a2596afffae6 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -36,212 +36,6 @@ static int proc_match(unsigned int len, const char *name, struct proc_dir_entry return !memcmp(name, de->name, len); } -/* buffer size is one page but our output routines use some slack for overruns */ -#define PROC_BLOCK_SIZE (PAGE_SIZE - 1024) - -static ssize_t -__proc_file_read(struct file *file, char __user *buf, size_t nbytes, - loff_t *ppos) -{ - struct inode * inode = file_inode(file); - char *page; - ssize_t retval=0; - int eof=0; - ssize_t n, count; - char *start; - struct proc_dir_entry * dp; - unsigned long long pos; - - /* - * Gaah, please just use "seq_file" instead. The legacy /proc - * interfaces cut loff_t down to off_t for reads, and ignore - * the offset entirely for writes.. - */ - pos = *ppos; - if (pos > MAX_NON_LFS) - return 0; - if (nbytes > MAX_NON_LFS - pos) - nbytes = MAX_NON_LFS - pos; - - dp = PDE(inode); - if (!(page = (char*) __get_free_page(GFP_TEMPORARY))) - return -ENOMEM; - - while ((nbytes > 0) && !eof) { - count = min_t(size_t, PROC_BLOCK_SIZE, nbytes); - - start = NULL; - if (dp->read_proc) { - /* - * How to be a proc read function - * ------------------------------ - * Prototype: - * int f(char *buffer, char **start, off_t offset, - * int count, int *peof, void *dat) - * - * Assume that the buffer is "count" bytes in size. - * - * If you know you have supplied all the data you - * have, set *peof. - * - * You have three ways to return data: - * 0) Leave *start = NULL. (This is the default.) - * Put the data of the requested offset at that - * offset within the buffer. Return the number (n) - * of bytes there are from the beginning of the - * buffer up to the last byte of data. If the - * number of supplied bytes (= n - offset) is - * greater than zero and you didn't signal eof - * and the reader is prepared to take more data - * you will be called again with the requested - * offset advanced by the number of bytes - * absorbed. This interface is useful for files - * no larger than the buffer. - * 1) Set *start = an unsigned long value less than - * the buffer address but greater than zero. - * Put the data of the requested offset at the - * beginning of the buffer. Return the number of - * bytes of data placed there. If this number is - * greater than zero and you didn't signal eof - * and the reader is prepared to take more data - * you will be called again with the requested - * offset advanced by *start. This interface is - * useful when you have a large file consisting - * of a series of blocks which you want to count - * and return as wholes. - * (Hack by Paul.Russell@rustcorp.com.au) - * 2) Set *start = an address within the buffer. - * Put the data of the requested offset at *start. - * Return the number of bytes of data placed there. - * If this number is greater than zero and you - * didn't signal eof and the reader is prepared to - * take more data you will be called again with the - * requested offset advanced by the number of bytes - * absorbed. - */ - n = dp->read_proc(page, &start, *ppos, - count, &eof, dp->data); - } else - break; - - if (n == 0) /* end of file */ - break; - if (n < 0) { /* error */ - if (retval == 0) - retval = n; - break; - } - - if (start == NULL) { - if (n > PAGE_SIZE) /* Apparent buffer overflow */ - n = PAGE_SIZE; - n -= *ppos; - if (n <= 0) - break; - if (n > count) - n = count; - start = page + *ppos; - } else if (start < page) { - if (n > PAGE_SIZE) /* Apparent buffer overflow */ - n = PAGE_SIZE; - if (n > count) { - /* - * Don't reduce n because doing so might - * cut off part of a data block. - */ - pr_warn("proc_file_read: count exceeded\n"); - } - } else /* start >= page */ { - unsigned long startoff = (unsigned long)(start - page); - if (n > (PAGE_SIZE - startoff)) /* buffer overflow? */ - n = PAGE_SIZE - startoff; - if (n > count) - n = count; - } - - n -= copy_to_user(buf, start < page ? page : start, n); - if (n == 0) { - if (retval == 0) - retval = -EFAULT; - break; - } - - *ppos += start < page ? (unsigned long)start : n; - nbytes -= n; - buf += n; - retval += n; - } - free_page((unsigned long) page); - return retval; -} - -static ssize_t -proc_file_read(struct file *file, char __user *buf, size_t nbytes, - loff_t *ppos) -{ - struct proc_dir_entry *pde = PDE(file_inode(file)); - ssize_t rv = -EIO; - - spin_lock(&pde->pde_unload_lock); - if (!pde->proc_fops) { - spin_unlock(&pde->pde_unload_lock); - return rv; - } - pde->pde_users++; - spin_unlock(&pde->pde_unload_lock); - - rv = __proc_file_read(file, buf, nbytes, ppos); - - pde_users_dec(pde); - return rv; -} - -static ssize_t -proc_file_write(struct file *file, const char __user *buffer, - size_t count, loff_t *ppos) -{ - struct proc_dir_entry *pde = PDE(file_inode(file)); - ssize_t rv = -EIO; - - if (pde->write_proc) { - spin_lock(&pde->pde_unload_lock); - if (!pde->proc_fops) { - spin_unlock(&pde->pde_unload_lock); - return rv; - } - pde->pde_users++; - spin_unlock(&pde->pde_unload_lock); - - /* FIXME: does this routine need ppos? probably... */ - rv = pde->write_proc(file, buffer, count, pde->data); - pde_users_dec(pde); - } - return rv; -} - - -static loff_t -proc_file_lseek(struct file *file, loff_t offset, int orig) -{ - loff_t retval = -EINVAL; - switch (orig) { - case 1: - offset += file->f_pos; - /* fallthrough */ - case 0: - if (offset < 0 || offset > MAX_NON_LFS) - break; - file->f_pos = retval = offset; - } - return retval; -} - -static const struct file_operations proc_file_operations = { - .llseek = proc_file_lseek, - .read = proc_file_read, - .write = proc_file_write, -}; - static int proc_notify_change(struct dentry *dentry, struct iattr *iattr) { struct inode *inode = dentry->d_inode; @@ -371,7 +165,7 @@ void proc_free_inum(unsigned int inum) static void *proc_follow_link(struct dentry *dentry, struct nameidata *nd) { - nd_set_link(nd, PDE(dentry->d_inode)->data); + nd_set_link(nd, __PDE_DATA(dentry->d_inode)); return NULL; } @@ -541,19 +335,17 @@ static int proc_register(struct proc_dir_entry * dir, struct proc_dir_entry * dp return ret; if (S_ISDIR(dp->mode)) { - if (dp->proc_iops == NULL) { - dp->proc_fops = &proc_dir_operations; - dp->proc_iops = &proc_dir_inode_operations; - } + dp->proc_fops = &proc_dir_operations; + dp->proc_iops = &proc_dir_inode_operations; dir->nlink++; } else if (S_ISLNK(dp->mode)) { - if (dp->proc_iops == NULL) - dp->proc_iops = &proc_link_inode_operations; + dp->proc_iops = &proc_link_inode_operations; } else if (S_ISREG(dp->mode)) { - if (dp->proc_fops == NULL) - dp->proc_fops = &proc_file_operations; - if (dp->proc_iops == NULL) - dp->proc_iops = &proc_file_inode_operations; + BUG_ON(dp->proc_fops == NULL); + dp->proc_iops = &proc_file_inode_operations; + } else { + WARN_ON(1); + return -EINVAL; } spin_lock(&proc_subdir_lock); @@ -636,13 +428,17 @@ struct proc_dir_entry *proc_symlink(const char *name, } EXPORT_SYMBOL(proc_symlink); -struct proc_dir_entry *proc_mkdir_mode(const char *name, umode_t mode, - struct proc_dir_entry *parent) +struct proc_dir_entry *proc_mkdir_data(const char *name, umode_t mode, + struct proc_dir_entry *parent, void *data) { struct proc_dir_entry *ent; + if (mode == 0) + mode = S_IRUGO | S_IXUGO; + ent = __proc_create(&parent, name, S_IFDIR | mode, 2); if (ent) { + ent->data = data; if (proc_register(parent, ent) < 0) { kfree(ent); ent = NULL; @@ -650,82 +446,39 @@ struct proc_dir_entry *proc_mkdir_mode(const char *name, umode_t mode, } return ent; } -EXPORT_SYMBOL(proc_mkdir_mode); +EXPORT_SYMBOL_GPL(proc_mkdir_data); -struct proc_dir_entry *proc_net_mkdir(struct net *net, const char *name, - struct proc_dir_entry *parent) +struct proc_dir_entry *proc_mkdir_mode(const char *name, umode_t mode, + struct proc_dir_entry *parent) { - struct proc_dir_entry *ent; - - ent = __proc_create(&parent, name, S_IFDIR | S_IRUGO | S_IXUGO, 2); - if (ent) { - ent->data = net; - if (proc_register(parent, ent) < 0) { - kfree(ent); - ent = NULL; - } - } - return ent; + return proc_mkdir_data(name, mode, parent, NULL); } -EXPORT_SYMBOL_GPL(proc_net_mkdir); +EXPORT_SYMBOL(proc_mkdir_mode); struct proc_dir_entry *proc_mkdir(const char *name, struct proc_dir_entry *parent) { - return proc_mkdir_mode(name, S_IRUGO | S_IXUGO, parent); + return proc_mkdir_data(name, 0, parent, NULL); } EXPORT_SYMBOL(proc_mkdir); -struct proc_dir_entry *create_proc_entry(const char *name, umode_t mode, - struct proc_dir_entry *parent) -{ - struct proc_dir_entry *ent; - nlink_t nlink; - - if (S_ISDIR(mode)) { - if ((mode & S_IALLUGO) == 0) - mode |= S_IRUGO | S_IXUGO; - nlink = 2; - } else { - if ((mode & S_IFMT) == 0) - mode |= S_IFREG; - if ((mode & S_IALLUGO) == 0) - mode |= S_IRUGO; - nlink = 1; - } - - ent = __proc_create(&parent, name, mode, nlink); - if (ent) { - if (proc_register(parent, ent) < 0) { - kfree(ent); - ent = NULL; - } - } - return ent; -} -EXPORT_SYMBOL(create_proc_entry); - struct proc_dir_entry *proc_create_data(const char *name, umode_t mode, struct proc_dir_entry *parent, const struct file_operations *proc_fops, void *data) { struct proc_dir_entry *pde; - nlink_t nlink; + if ((mode & S_IFMT) == 0) + mode |= S_IFREG; - if (S_ISDIR(mode)) { - if ((mode & S_IALLUGO) == 0) - mode |= S_IRUGO | S_IXUGO; - nlink = 2; - } else { - if ((mode & S_IFMT) == 0) - mode |= S_IFREG; - if ((mode & S_IALLUGO) == 0) - mode |= S_IRUGO; - nlink = 1; + if (!S_ISREG(mode)) { + WARN_ON(1); /* use proc_mkdir() */ + return NULL; } - pde = __proc_create(&parent, name, mode, nlink); + if ((mode & S_IALLUGO) == 0) + mode |= S_IRUGO; + pde = __proc_create(&parent, name, mode, 1); if (!pde) goto out; pde->proc_fops = proc_fops; @@ -739,6 +492,19 @@ out: return NULL; } EXPORT_SYMBOL(proc_create_data); + +void proc_set_size(struct proc_dir_entry *de, loff_t size) +{ + de->size = size; +} +EXPORT_SYMBOL(proc_set_size); + +void proc_set_user(struct proc_dir_entry *de, kuid_t uid, kgid_t gid) +{ + de->uid = uid; + de->gid = gid; +} +EXPORT_SYMBOL(proc_set_user); static void free_proc_entry(struct proc_dir_entry *de) { @@ -755,41 +521,6 @@ void pde_put(struct proc_dir_entry *pde) free_proc_entry(pde); } -static void entry_rundown(struct proc_dir_entry *de) -{ - spin_lock(&de->pde_unload_lock); - /* - * Stop accepting new callers into module. If you're - * dynamically allocating ->proc_fops, save a pointer somewhere. - */ - de->proc_fops = NULL; - /* Wait until all existing callers into module are done. */ - if (de->pde_users > 0) { - DECLARE_COMPLETION_ONSTACK(c); - - if (!de->pde_unload_completion) - de->pde_unload_completion = &c; - - spin_unlock(&de->pde_unload_lock); - - wait_for_completion(de->pde_unload_completion); - - spin_lock(&de->pde_unload_lock); - } - - while (!list_empty(&de->pde_openers)) { - struct pde_opener *pdeo; - - pdeo = list_first_entry(&de->pde_openers, struct pde_opener, lh); - list_del(&pdeo->lh); - spin_unlock(&de->pde_unload_lock); - pdeo->release(pdeo->inode, pdeo->file); - kfree(pdeo); - spin_lock(&de->pde_unload_lock); - } - spin_unlock(&de->pde_unload_lock); -} - /* * Remove a /proc entry and free it if it's not currently in use. */ @@ -821,7 +552,7 @@ void remove_proc_entry(const char *name, struct proc_dir_entry *parent) return; } - entry_rundown(de); + proc_entry_rundown(de); if (S_ISDIR(de->mode)) parent->nlink--; @@ -870,7 +601,7 @@ int remove_proc_subtree(const char *name, struct proc_dir_entry *parent) } spin_unlock(&proc_subdir_lock); - entry_rundown(de); + proc_entry_rundown(de); next = de->parent; if (S_ISDIR(de->mode)) next->nlink--; @@ -886,3 +617,23 @@ int remove_proc_subtree(const char *name, struct proc_dir_entry *parent) return 0; } EXPORT_SYMBOL(remove_proc_subtree); + +void *proc_get_parent_data(const struct inode *inode) +{ + struct proc_dir_entry *de = PDE(inode); + return de->parent->data; +} +EXPORT_SYMBOL_GPL(proc_get_parent_data); + +void proc_remove(struct proc_dir_entry *de) +{ + if (de) + remove_proc_subtree(de->name, de->parent); +} +EXPORT_SYMBOL(proc_remove); + +void *PDE_DATA(const struct inode *inode) +{ + return __PDE_DATA(inode); +} +EXPORT_SYMBOL(PDE_DATA); diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 869116c2afbe..073aea60cf8f 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -22,6 +22,7 @@ #include <linux/seq_file.h> #include <linux/slab.h> #include <linux/mount.h> +#include <linux/magic.h> #include <asm/uaccess.h> @@ -50,8 +51,8 @@ static void proc_evict_inode(struct inode *inode) sysctl_head_put(head); } /* Release any associated namespace */ - ns_ops = PROC_I(inode)->ns_ops; - ns = PROC_I(inode)->ns; + ns_ops = PROC_I(inode)->ns.ns_ops; + ns = PROC_I(inode)->ns.ns; if (ns_ops && ns) ns_ops->put(ns); } @@ -72,8 +73,8 @@ static struct inode *proc_alloc_inode(struct super_block *sb) ei->pde = NULL; ei->sysctl = NULL; ei->sysctl_entry = NULL; - ei->ns = NULL; - ei->ns_ops = NULL; + ei->ns.ns = NULL; + ei->ns.ns_ops = NULL; inode = &ei->vfs_inode; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; return inode; @@ -129,96 +130,100 @@ static const struct super_operations proc_sops = { .show_options = proc_show_options, }; -static void __pde_users_dec(struct proc_dir_entry *pde) +enum {BIAS = -1U<<31}; + +static inline int use_pde(struct proc_dir_entry *pde) +{ + return atomic_inc_unless_negative(&pde->in_use); +} + +static void unuse_pde(struct proc_dir_entry *pde) { - pde->pde_users--; - if (pde->pde_unload_completion && pde->pde_users == 0) + if (atomic_dec_return(&pde->in_use) == BIAS) complete(pde->pde_unload_completion); } -void pde_users_dec(struct proc_dir_entry *pde) +/* pde is locked */ +static void close_pdeo(struct proc_dir_entry *pde, struct pde_opener *pdeo) { - spin_lock(&pde->pde_unload_lock); - __pde_users_dec(pde); - spin_unlock(&pde->pde_unload_lock); + if (pdeo->closing) { + /* somebody else is doing that, just wait */ + DECLARE_COMPLETION_ONSTACK(c); + pdeo->c = &c; + spin_unlock(&pde->pde_unload_lock); + wait_for_completion(&c); + spin_lock(&pde->pde_unload_lock); + } else { + struct file *file; + pdeo->closing = 1; + spin_unlock(&pde->pde_unload_lock); + file = pdeo->file; + pde->proc_fops->release(file_inode(file), file); + spin_lock(&pde->pde_unload_lock); + list_del_init(&pdeo->lh); + if (pdeo->c) + complete(pdeo->c); + kfree(pdeo); + } +} + +void proc_entry_rundown(struct proc_dir_entry *de) +{ + DECLARE_COMPLETION_ONSTACK(c); + /* Wait until all existing callers into module are done. */ + de->pde_unload_completion = &c; + if (atomic_add_return(BIAS, &de->in_use) != BIAS) + wait_for_completion(&c); + + spin_lock(&de->pde_unload_lock); + while (!list_empty(&de->pde_openers)) { + struct pde_opener *pdeo; + pdeo = list_first_entry(&de->pde_openers, struct pde_opener, lh); + close_pdeo(de, pdeo); + } + spin_unlock(&de->pde_unload_lock); } static loff_t proc_reg_llseek(struct file *file, loff_t offset, int whence) { struct proc_dir_entry *pde = PDE(file_inode(file)); loff_t rv = -EINVAL; - loff_t (*llseek)(struct file *, loff_t, int); - - spin_lock(&pde->pde_unload_lock); - /* - * remove_proc_entry() is going to delete PDE (as part of module - * cleanup sequence). No new callers into module allowed. - */ - if (!pde->proc_fops) { - spin_unlock(&pde->pde_unload_lock); - return rv; + if (use_pde(pde)) { + loff_t (*llseek)(struct file *, loff_t, int); + llseek = pde->proc_fops->llseek; + if (!llseek) + llseek = default_llseek; + rv = llseek(file, offset, whence); + unuse_pde(pde); } - /* - * Bump refcount so that remove_proc_entry will wail for ->llseek to - * complete. - */ - pde->pde_users++; - /* - * Save function pointer under lock, to protect against ->proc_fops - * NULL'ifying right after ->pde_unload_lock is dropped. - */ - llseek = pde->proc_fops->llseek; - spin_unlock(&pde->pde_unload_lock); - - if (!llseek) - llseek = default_llseek; - rv = llseek(file, offset, whence); - - pde_users_dec(pde); return rv; } static ssize_t proc_reg_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { + ssize_t (*read)(struct file *, char __user *, size_t, loff_t *); struct proc_dir_entry *pde = PDE(file_inode(file)); ssize_t rv = -EIO; - ssize_t (*read)(struct file *, char __user *, size_t, loff_t *); - - spin_lock(&pde->pde_unload_lock); - if (!pde->proc_fops) { - spin_unlock(&pde->pde_unload_lock); - return rv; + if (use_pde(pde)) { + read = pde->proc_fops->read; + if (read) + rv = read(file, buf, count, ppos); + unuse_pde(pde); } - pde->pde_users++; - read = pde->proc_fops->read; - spin_unlock(&pde->pde_unload_lock); - - if (read) - rv = read(file, buf, count, ppos); - - pde_users_dec(pde); return rv; } static ssize_t proc_reg_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos) { + ssize_t (*write)(struct file *, const char __user *, size_t, loff_t *); struct proc_dir_entry *pde = PDE(file_inode(file)); ssize_t rv = -EIO; - ssize_t (*write)(struct file *, const char __user *, size_t, loff_t *); - - spin_lock(&pde->pde_unload_lock); - if (!pde->proc_fops) { - spin_unlock(&pde->pde_unload_lock); - return rv; + if (use_pde(pde)) { + write = pde->proc_fops->write; + if (write) + rv = write(file, buf, count, ppos); + unuse_pde(pde); } - pde->pde_users++; - write = pde->proc_fops->write; - spin_unlock(&pde->pde_unload_lock); - - if (write) - rv = write(file, buf, count, ppos); - - pde_users_dec(pde); return rv; } @@ -227,20 +232,12 @@ static unsigned int proc_reg_poll(struct file *file, struct poll_table_struct *p struct proc_dir_entry *pde = PDE(file_inode(file)); unsigned int rv = DEFAULT_POLLMASK; unsigned int (*poll)(struct file *, struct poll_table_struct *); - - spin_lock(&pde->pde_unload_lock); - if (!pde->proc_fops) { - spin_unlock(&pde->pde_unload_lock); - return rv; + if (use_pde(pde)) { + poll = pde->proc_fops->poll; + if (poll) + rv = poll(file, pts); + unuse_pde(pde); } - pde->pde_users++; - poll = pde->proc_fops->poll; - spin_unlock(&pde->pde_unload_lock); - - if (poll) - rv = poll(file, pts); - - pde_users_dec(pde); return rv; } @@ -249,20 +246,12 @@ static long proc_reg_unlocked_ioctl(struct file *file, unsigned int cmd, unsigne struct proc_dir_entry *pde = PDE(file_inode(file)); long rv = -ENOTTY; long (*ioctl)(struct file *, unsigned int, unsigned long); - - spin_lock(&pde->pde_unload_lock); - if (!pde->proc_fops) { - spin_unlock(&pde->pde_unload_lock); - return rv; + if (use_pde(pde)) { + ioctl = pde->proc_fops->unlocked_ioctl; + if (ioctl) + rv = ioctl(file, cmd, arg); + unuse_pde(pde); } - pde->pde_users++; - ioctl = pde->proc_fops->unlocked_ioctl; - spin_unlock(&pde->pde_unload_lock); - - if (ioctl) - rv = ioctl(file, cmd, arg); - - pde_users_dec(pde); return rv; } @@ -272,20 +261,12 @@ static long proc_reg_compat_ioctl(struct file *file, unsigned int cmd, unsigned struct proc_dir_entry *pde = PDE(file_inode(file)); long rv = -ENOTTY; long (*compat_ioctl)(struct file *, unsigned int, unsigned long); - - spin_lock(&pde->pde_unload_lock); - if (!pde->proc_fops) { - spin_unlock(&pde->pde_unload_lock); - return rv; + if (use_pde(pde)) { + compat_ioctl = pde->proc_fops->compat_ioctl; + if (compat_ioctl) + rv = compat_ioctl(file, cmd, arg); + unuse_pde(pde); } - pde->pde_users++; - compat_ioctl = pde->proc_fops->compat_ioctl; - spin_unlock(&pde->pde_unload_lock); - - if (compat_ioctl) - rv = compat_ioctl(file, cmd, arg); - - pde_users_dec(pde); return rv; } #endif @@ -295,20 +276,12 @@ static int proc_reg_mmap(struct file *file, struct vm_area_struct *vma) struct proc_dir_entry *pde = PDE(file_inode(file)); int rv = -EIO; int (*mmap)(struct file *, struct vm_area_struct *); - - spin_lock(&pde->pde_unload_lock); - if (!pde->proc_fops) { - spin_unlock(&pde->pde_unload_lock); - return rv; + if (use_pde(pde)) { + mmap = pde->proc_fops->mmap; + if (mmap) + rv = mmap(file, vma); + unuse_pde(pde); } - pde->pde_users++; - mmap = pde->proc_fops->mmap; - spin_unlock(&pde->pde_unload_lock); - - if (mmap) - rv = mmap(file, vma); - - pde_users_dec(pde); return rv; } @@ -330,91 +303,47 @@ static int proc_reg_open(struct inode *inode, struct file *file) * by hand in remove_proc_entry(). For this, save opener's credentials * for later. */ - pdeo = kmalloc(sizeof(struct pde_opener), GFP_KERNEL); + pdeo = kzalloc(sizeof(struct pde_opener), GFP_KERNEL); if (!pdeo) return -ENOMEM; - spin_lock(&pde->pde_unload_lock); - if (!pde->proc_fops) { - spin_unlock(&pde->pde_unload_lock); + if (!use_pde(pde)) { kfree(pdeo); return -ENOENT; } - pde->pde_users++; open = pde->proc_fops->open; release = pde->proc_fops->release; - spin_unlock(&pde->pde_unload_lock); if (open) rv = open(inode, file); - spin_lock(&pde->pde_unload_lock); if (rv == 0 && release) { /* To know what to release. */ - pdeo->inode = inode; pdeo->file = file; /* Strictly for "too late" ->release in proc_reg_release(). */ - pdeo->release = release; + spin_lock(&pde->pde_unload_lock); list_add(&pdeo->lh, &pde->pde_openers); + spin_unlock(&pde->pde_unload_lock); } else kfree(pdeo); - __pde_users_dec(pde); - spin_unlock(&pde->pde_unload_lock); - return rv; -} - -static struct pde_opener *find_pde_opener(struct proc_dir_entry *pde, - struct inode *inode, struct file *file) -{ - struct pde_opener *pdeo; - list_for_each_entry(pdeo, &pde->pde_openers, lh) { - if (pdeo->inode == inode && pdeo->file == file) - return pdeo; - } - return NULL; + unuse_pde(pde); + return rv; } static int proc_reg_release(struct inode *inode, struct file *file) { struct proc_dir_entry *pde = PDE(inode); - int rv = 0; - int (*release)(struct inode *, struct file *); struct pde_opener *pdeo; - spin_lock(&pde->pde_unload_lock); - pdeo = find_pde_opener(pde, inode, file); - if (!pde->proc_fops) { - /* - * Can't simply exit, __fput() will think that everything is OK, - * and move on to freeing struct file. remove_proc_entry() will - * find slacker in opener's list and will try to do non-trivial - * things with struct file. Therefore, remove opener from list. - * - * But if opener is removed from list, who will ->release it? - */ - if (pdeo) { - list_del(&pdeo->lh); - spin_unlock(&pde->pde_unload_lock); - rv = pdeo->release(inode, file); - kfree(pdeo); - } else - spin_unlock(&pde->pde_unload_lock); - return rv; - } - pde->pde_users++; - release = pde->proc_fops->release; - if (pdeo) { - list_del(&pdeo->lh); - kfree(pdeo); + list_for_each_entry(pdeo, &pde->pde_openers, lh) { + if (pdeo->file == file) { + close_pdeo(pde, pdeo); + break; + } } spin_unlock(&pde->pde_unload_lock); - - if (release) - rv = release(inode, file); - - pde_users_dec(pde); - return rv; + return 0; } static const struct file_operations proc_reg_file_ops = { @@ -462,8 +391,8 @@ struct inode *proc_get_inode(struct super_block *sb, struct proc_dir_entry *de) inode->i_size = de->size; if (de->nlink) set_nlink(inode, de->nlink); - if (de->proc_iops) - inode->i_op = de->proc_iops; + WARN_ON(!de->proc_iops); + inode->i_op = de->proc_iops; if (de->proc_fops) { if (S_ISREG(inode->i_mode)) { #ifdef CONFIG_COMPAT @@ -506,5 +435,5 @@ int proc_fill_super(struct super_block *s) return -ENOMEM; } - return 0; + return proc_setup_self(s); } diff --git a/fs/proc/internal.h b/fs/proc/internal.h index 75710357a517..d600fb098b6a 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -1,4 +1,4 @@ -/* internal.h: internal procfs definitions +/* Internal procfs definitions * * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) @@ -9,62 +9,83 @@ * 2 of the License, or (at your option) any later version. */ -#include <linux/sched.h> #include <linux/proc_fs.h> +#include <linux/proc_ns.h> +#include <linux/spinlock.h> +#include <linux/atomic.h> #include <linux/binfmts.h> -struct ctl_table_header; -struct mempolicy; -extern struct proc_dir_entry proc_root; -extern void proc_self_init(void); -#ifdef CONFIG_PROC_SYSCTL -extern int proc_sys_init(void); -extern void sysctl_head_put(struct ctl_table_header *head); -#else -static inline void proc_sys_init(void) { } -static inline void sysctl_head_put(struct ctl_table_header *head) { } -#endif -#ifdef CONFIG_NET -extern int proc_net_init(void); -#else -static inline int proc_net_init(void) { return 0; } -#endif +struct ctl_table_header; +struct mempolicy; -extern int proc_tid_stat(struct seq_file *m, struct pid_namespace *ns, - struct pid *pid, struct task_struct *task); -extern int proc_tgid_stat(struct seq_file *m, struct pid_namespace *ns, - struct pid *pid, struct task_struct *task); -extern int proc_pid_status(struct seq_file *m, struct pid_namespace *ns, - struct pid *pid, struct task_struct *task); -extern int proc_pid_statm(struct seq_file *m, struct pid_namespace *ns, - struct pid *pid, struct task_struct *task); -extern loff_t mem_lseek(struct file *file, loff_t offset, int orig); +/* + * This is not completely implemented yet. The idea is to + * create an in-memory tree (like the actual /proc filesystem + * tree) of these proc_dir_entries, so that we can dynamically + * add new files to /proc. + * + * The "next" pointer creates a linked list of one /proc directory, + * while parent/subdir create the directory structure (every + * /proc file has a parent, but "subdir" is NULL for all + * non-directory entries). + */ +struct proc_dir_entry { + unsigned int low_ino; + umode_t mode; + nlink_t nlink; + kuid_t uid; + kgid_t gid; + loff_t size; + const struct inode_operations *proc_iops; + const struct file_operations *proc_fops; + struct proc_dir_entry *next, *parent, *subdir; + void *data; + atomic_t count; /* use count */ + atomic_t in_use; /* number of callers into module in progress; */ + /* negative -> it's going away RSN */ + struct completion *pde_unload_completion; + struct list_head pde_openers; /* who did ->open, but not ->release */ + spinlock_t pde_unload_lock; /* proc_fops checks and pde_users bumps */ + u8 namelen; + char name[]; +}; -extern const struct file_operations proc_tid_children_operations; -extern const struct file_operations proc_pid_maps_operations; -extern const struct file_operations proc_tid_maps_operations; -extern const struct file_operations proc_pid_numa_maps_operations; -extern const struct file_operations proc_tid_numa_maps_operations; -extern const struct file_operations proc_pid_smaps_operations; -extern const struct file_operations proc_tid_smaps_operations; -extern const struct file_operations proc_clear_refs_operations; -extern const struct file_operations proc_pagemap_operations; -extern const struct file_operations proc_net_operations; -extern const struct inode_operations proc_net_inode_operations; -extern const struct inode_operations proc_pid_link_inode_operations; +union proc_op { + int (*proc_get_link)(struct dentry *, struct path *); + int (*proc_read)(struct task_struct *task, char *page); + int (*proc_show)(struct seq_file *m, + struct pid_namespace *ns, struct pid *pid, + struct task_struct *task); +}; -struct proc_maps_private { +struct proc_inode { struct pid *pid; - struct task_struct *task; -#ifdef CONFIG_MMU - struct vm_area_struct *tail_vma; -#endif -#ifdef CONFIG_NUMA - struct mempolicy *task_mempolicy; -#endif + int fd; + union proc_op op; + struct proc_dir_entry *pde; + struct ctl_table_header *sysctl; + struct ctl_table *sysctl_entry; + struct proc_ns ns; + struct inode vfs_inode; }; -void proc_init_inodecache(void); +/* + * General functions + */ +static inline struct proc_inode *PROC_I(const struct inode *inode) +{ + return container_of(inode, struct proc_inode, vfs_inode); +} + +static inline struct proc_dir_entry *PDE(const struct inode *inode) +{ + return PROC_I(inode)->pde; +} + +static inline void *__PDE_DATA(const struct inode *inode) +{ + return PDE(inode)->data; +} static inline struct pid *proc_pid(struct inode *inode) { @@ -76,11 +97,6 @@ static inline struct task_struct *get_proc_task(struct inode *inode) return get_pid_task(proc_pid(inode), PIDTYPE_PID); } -static inline int proc_fd(struct inode *inode) -{ - return PROC_I(inode)->fd; -} - static inline int task_dumpable(struct task_struct *task) { int dumpable = 0; @@ -96,15 +112,6 @@ static inline int task_dumpable(struct task_struct *task) return 0; } -static inline int pid_delete_dentry(const struct dentry * dentry) -{ - /* Is the task we represent dead? - * If so, then don't put the dentry on the lru list, - * kill it immediately. - */ - return !proc_pid(dentry->d_inode)->tasks[PIDTYPE_PID].first; -} - static inline unsigned name_to_int(struct dentry *dentry) { const char *name = dentry->d_name.name; @@ -127,63 +134,165 @@ out: return ~0U; } -struct dentry *proc_lookup_de(struct proc_dir_entry *de, struct inode *ino, - struct dentry *dentry); -int proc_readdir_de(struct proc_dir_entry *de, struct file *filp, void *dirent, - filldir_t filldir); +/* + * Offset of the first process in the /proc root directory.. + */ +#define FIRST_PROCESS_ENTRY 256 + +/* Worst case buffer size needed for holding an integer. */ +#define PROC_NUMBUF 13 -struct pde_opener { - struct inode *inode; - struct file *file; - int (*release)(struct inode *, struct file *); - struct list_head lh; -}; -void pde_users_dec(struct proc_dir_entry *pde); +/* + * array.c + */ +extern const struct file_operations proc_tid_children_operations; +extern int proc_tid_stat(struct seq_file *, struct pid_namespace *, + struct pid *, struct task_struct *); +extern int proc_tgid_stat(struct seq_file *, struct pid_namespace *, + struct pid *, struct task_struct *); +extern int proc_pid_status(struct seq_file *, struct pid_namespace *, + struct pid *, struct task_struct *); +extern int proc_pid_statm(struct seq_file *, struct pid_namespace *, + struct pid *, struct task_struct *); + +/* + * base.c + */ +extern const struct dentry_operations pid_dentry_operations; +extern int pid_getattr(struct vfsmount *, struct dentry *, struct kstat *); +extern int proc_setattr(struct dentry *, struct iattr *); +extern struct inode *proc_pid_make_inode(struct super_block *, struct task_struct *); +extern int pid_revalidate(struct dentry *, unsigned int); +extern int pid_delete_dentry(const struct dentry *); +extern int proc_pid_readdir(struct file *, void *, filldir_t); +extern struct dentry *proc_pid_lookup(struct inode *, struct dentry *, unsigned int); +extern loff_t mem_lseek(struct file *, loff_t, int); + +/* Lookups */ +typedef struct dentry *instantiate_t(struct inode *, struct dentry *, + struct task_struct *, const void *); +extern int proc_fill_cache(struct file *, void *, filldir_t, const char *, int, + instantiate_t, struct task_struct *, const void *); + +/* + * generic.c + */ extern spinlock_t proc_subdir_lock; -struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, unsigned int); -int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir); -unsigned long task_vsize(struct mm_struct *); -unsigned long task_statm(struct mm_struct *, - unsigned long *, unsigned long *, unsigned long *, unsigned long *); -void task_mem(struct seq_file *, struct mm_struct *); +extern struct dentry *proc_lookup(struct inode *, struct dentry *, unsigned int); +extern struct dentry *proc_lookup_de(struct proc_dir_entry *, struct inode *, + struct dentry *); +extern int proc_readdir(struct file *, void *, filldir_t); +extern int proc_readdir_de(struct proc_dir_entry *, struct file *, void *, filldir_t); static inline struct proc_dir_entry *pde_get(struct proc_dir_entry *pde) { atomic_inc(&pde->count); return pde; } -void pde_put(struct proc_dir_entry *pde); - -int proc_fill_super(struct super_block *); -struct inode *proc_get_inode(struct super_block *, struct proc_dir_entry *); -int proc_remount(struct super_block *sb, int *flags, char *data); +extern void pde_put(struct proc_dir_entry *); /* - * These are generic /proc routines that use the internal - * "struct proc_dir_entry" tree to traverse the filesystem. - * - * The /proc root directory has extended versions to take care - * of the /proc/<pid> subdirectories. + * inode.c */ -int proc_readdir(struct file *, void *, filldir_t); -struct dentry *proc_lookup(struct inode *, struct dentry *, unsigned int); +struct pde_opener { + struct file *file; + struct list_head lh; + int closing; + struct completion *c; +}; +extern const struct inode_operations proc_pid_link_inode_operations; +extern void proc_init_inodecache(void); +extern struct inode *proc_get_inode(struct super_block *, struct proc_dir_entry *); +extern int proc_fill_super(struct super_block *); +extern void proc_entry_rundown(struct proc_dir_entry *); -/* Lookups */ -typedef struct dentry *instantiate_t(struct inode *, struct dentry *, - struct task_struct *, const void *); -int proc_fill_cache(struct file *filp, void *dirent, filldir_t filldir, - const char *name, int len, - instantiate_t instantiate, struct task_struct *task, const void *ptr); -int pid_revalidate(struct dentry *dentry, unsigned int flags); -struct inode *proc_pid_make_inode(struct super_block * sb, struct task_struct *task); -extern const struct dentry_operations pid_dentry_operations; -int pid_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); -int proc_setattr(struct dentry *dentry, struct iattr *attr); +/* + * proc_devtree.c + */ +#ifdef CONFIG_PROC_DEVICETREE +extern void proc_device_tree_init(void); +#endif +/* + * proc_namespaces.c + */ extern const struct inode_operations proc_ns_dir_inode_operations; extern const struct file_operations proc_ns_dir_operations; +/* + * proc_net.c + */ +extern const struct file_operations proc_net_operations; +extern const struct inode_operations proc_net_inode_operations; + +#ifdef CONFIG_NET +extern int proc_net_init(void); +#else +static inline int proc_net_init(void) { return 0; } +#endif + +/* + * proc_self.c + */ +extern int proc_setup_self(struct super_block *); + +/* + * proc_sysctl.c + */ +#ifdef CONFIG_PROC_SYSCTL +extern int proc_sys_init(void); +extern void sysctl_head_put(struct ctl_table_header *); +#else +static inline void proc_sys_init(void) { } +static inline void sysctl_head_put(struct ctl_table_header *head) { } +#endif + +/* + * proc_tty.c + */ +#ifdef CONFIG_TTY +extern void proc_tty_init(void); +#else +static inline void proc_tty_init(void) {} +#endif + +/* + * root.c + */ +extern struct proc_dir_entry proc_root; + +extern void proc_self_init(void); +extern int proc_remount(struct super_block *, int *, char *); + +/* + * task_[no]mmu.c + */ +struct proc_maps_private { + struct pid *pid; + struct task_struct *task; +#ifdef CONFIG_MMU + struct vm_area_struct *tail_vma; +#endif +#ifdef CONFIG_NUMA + struct mempolicy *task_mempolicy; +#endif +}; + +extern const struct file_operations proc_pid_maps_operations; +extern const struct file_operations proc_tid_maps_operations; +extern const struct file_operations proc_pid_numa_maps_operations; +extern const struct file_operations proc_tid_numa_maps_operations; +extern const struct file_operations proc_pid_smaps_operations; +extern const struct file_operations proc_tid_smaps_operations; +extern const struct file_operations proc_clear_refs_operations; +extern const struct file_operations proc_pagemap_operations; + +extern unsigned long task_vsize(struct mm_struct *); +extern unsigned long task_statm(struct mm_struct *, + unsigned long *, unsigned long *, + unsigned long *, unsigned long *); +extern void task_mem(struct seq_file *, struct mm_struct *); diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c index f6a13f489e30..0a22194e5d58 100644 --- a/fs/proc/kcore.c +++ b/fs/proc/kcore.c @@ -11,6 +11,7 @@ #include <linux/mm.h> #include <linux/proc_fs.h> +#include <linux/kcore.h> #include <linux/user.h> #include <linux/capability.h> #include <linux/elf.h> @@ -28,6 +29,7 @@ #include <linux/ioport.h> #include <linux/memory.h> #include <asm/sections.h> +#include "internal.h" #define CORE_STR "CORE" diff --git a/fs/proc/namespaces.c b/fs/proc/namespaces.c index 66b51c0383da..54bdc6701e9f 100644 --- a/fs/proc/namespaces.c +++ b/fs/proc/namespaces.c @@ -51,7 +51,7 @@ static int ns_delete_dentry(const struct dentry *dentry) static char *ns_dname(struct dentry *dentry, char *buffer, int buflen) { struct inode *inode = dentry->d_inode; - const struct proc_ns_operations *ns_ops = PROC_I(inode)->ns_ops; + const struct proc_ns_operations *ns_ops = PROC_I(inode)->ns.ns_ops; return dynamic_dname(dentry, buffer, buflen, "%s:[%lu]", ns_ops->name, inode->i_ino); @@ -95,8 +95,8 @@ static struct dentry *proc_ns_get_dentry(struct super_block *sb, inode->i_op = &ns_inode_operations; inode->i_mode = S_IFREG | S_IRUGO; inode->i_fop = &ns_file_operations; - ei->ns_ops = ns_ops; - ei->ns = ns; + ei->ns.ns_ops = ns_ops; + ei->ns.ns = ns; unlock_new_inode(inode); } else { ns_ops->put(ns); @@ -128,7 +128,7 @@ static void *proc_ns_follow_link(struct dentry *dentry, struct nameidata *nd) if (!ptrace_may_access(task, PTRACE_MODE_READ)) goto out_put_task; - ns_path.dentry = proc_ns_get_dentry(sb, task, ei->ns_ops); + ns_path.dentry = proc_ns_get_dentry(sb, task, ei->ns.ns_ops); if (IS_ERR(ns_path.dentry)) { error = ERR_CAST(ns_path.dentry); goto out_put_task; @@ -148,7 +148,7 @@ static int proc_ns_readlink(struct dentry *dentry, char __user *buffer, int bufl { struct inode *inode = dentry->d_inode; struct proc_inode *ei = PROC_I(inode); - const struct proc_ns_operations *ns_ops = ei->ns_ops; + const struct proc_ns_operations *ns_ops = ei->ns.ns_ops; struct task_struct *task; void *ns; char name[50]; @@ -202,7 +202,7 @@ static struct dentry *proc_ns_instantiate(struct inode *dir, ei = PROC_I(inode); inode->i_mode = S_IFLNK|S_IRWXUGO; inode->i_op = &proc_ns_link_inode_operations; - ei->ns_ops = ns_ops; + ei->ns.ns_ops = ns_ops; d_set_d_op(dentry, &pid_dentry_operations); d_add(dentry, inode); @@ -337,6 +337,11 @@ out_invalid: return ERR_PTR(-EINVAL); } +struct proc_ns *get_proc_ns(struct inode *inode) +{ + return &PROC_I(inode)->ns; +} + bool proc_ns_inode(struct inode *inode) { return inode->i_fop == &ns_file_operations; diff --git a/fs/proc/proc_devtree.c b/fs/proc/proc_devtree.c index 30b590f5bd35..505afc950e0a 100644 --- a/fs/proc/proc_devtree.c +++ b/fs/proc/proc_devtree.c @@ -41,7 +41,7 @@ static int property_proc_show(struct seq_file *m, void *v) static int property_proc_open(struct inode *inode, struct file *file) { - return single_open(file, property_proc_show, PDE(inode)->data); + return single_open(file, property_proc_show, __PDE_DATA(inode)); } static const struct file_operations property_proc_fops = { diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index b4ac6572474f..986e83220d56 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -26,6 +26,10 @@ #include "internal.h" +static inline struct net *PDE_NET(struct proc_dir_entry *pde) +{ + return pde->parent->data; +} static struct net *get_proc_net(const struct inode *inode) { diff --git a/fs/proc/root.c b/fs/proc/root.c index 9c7fab1d23f0..41a6ea93f486 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -141,6 +141,8 @@ static void proc_kill_sb(struct super_block *sb) struct pid_namespace *ns; ns = (struct pid_namespace *)sb->s_fs_info; + if (ns->proc_self) + dput(ns->proc_self); kill_anon_super(sb); put_pid_ns(ns); } diff --git a/fs/proc/self.c b/fs/proc/self.c index aa5cc3bff140..6b6a993b5c25 100644 --- a/fs/proc/self.c +++ b/fs/proc/self.c @@ -1,6 +1,8 @@ -#include <linux/proc_fs.h> #include <linux/sched.h> #include <linux/namei.h> +#include <linux/slab.h> +#include <linux/pid_namespace.h> +#include "internal.h" /* * /proc/self: @@ -48,12 +50,43 @@ static const struct inode_operations proc_self_inode_operations = { .put_link = proc_self_put_link, }; -void __init proc_self_init(void) +static unsigned self_inum; + +int proc_setup_self(struct super_block *s) { - struct proc_dir_entry *proc_self_symlink; - mode_t mode; + struct inode *root_inode = s->s_root->d_inode; + struct pid_namespace *ns = s->s_fs_info; + struct dentry *self; + + mutex_lock(&root_inode->i_mutex); + self = d_alloc_name(s->s_root, "self"); + if (self) { + struct inode *inode = new_inode_pseudo(s); + if (inode) { + inode->i_ino = self_inum; + inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; + inode->i_mode = S_IFLNK | S_IRWXUGO; + inode->i_uid = GLOBAL_ROOT_UID; + inode->i_gid = GLOBAL_ROOT_GID; + inode->i_op = &proc_self_inode_operations; + d_add(self, inode); + } else { + dput(self); + self = ERR_PTR(-ENOMEM); + } + } else { + self = ERR_PTR(-ENOMEM); + } + mutex_unlock(&root_inode->i_mutex); + if (IS_ERR(self)) { + pr_err("proc_fill_super: can't allocate /proc/self\n"); + return PTR_ERR(self); + } + ns->proc_self = self; + return 0; +} - mode = S_IFLNK | S_IRWXUGO; - proc_self_symlink = proc_create("self", mode, NULL, NULL ); - proc_self_symlink->proc_iops = &proc_self_inode_operations; +void __init proc_self_init(void) +{ + proc_alloc_inum(&self_inum); } diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c index b870f740ab5a..17f7e080d7ff 100644 --- a/fs/proc/vmcore.c +++ b/fs/proc/vmcore.c @@ -8,7 +8,7 @@ */ #include <linux/mm.h> -#include <linux/proc_fs.h> +#include <linux/kcore.h> #include <linux/user.h> #include <linux/elf.h> #include <linux/elfcore.h> @@ -22,6 +22,7 @@ #include <linux/list.h> #include <asm/uaccess.h> #include <asm/io.h> +#include "internal.h" /* List representing chunks of contiguous memory areas and their offsets in * vmcore file. @@ -698,7 +699,7 @@ void vmcore_cleanup(void) struct list_head *pos, *next; if (proc_vmcore) { - remove_proc_entry(proc_vmcore->name, proc_vmcore->parent); + proc_remove(proc_vmcore); proc_vmcore = NULL; } diff --git a/fs/read_write.c b/fs/read_write.c index 8274a794253b..605dbbcb1973 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -459,6 +459,7 @@ ssize_t vfs_write(struct file *file, const char __user *buf, size_t count, loff_ ret = rw_verify_area(WRITE, file, pos, count); if (ret >= 0) { count = ret; + file_start_write(file); if (file->f_op->write) ret = file->f_op->write(file, buf, count, pos); else @@ -468,6 +469,7 @@ ssize_t vfs_write(struct file *file, const char __user *buf, size_t count, loff_ add_wchar(current, ret); } inc_syscw(current); + file_end_write(file); } return ret; @@ -576,7 +578,7 @@ unsigned long iov_shorten(struct iovec *iov, unsigned long nr_segs, size_t to) } EXPORT_SYMBOL(iov_shorten); -ssize_t do_sync_readv_writev(struct file *filp, const struct iovec *iov, +static ssize_t do_sync_readv_writev(struct file *filp, const struct iovec *iov, unsigned long nr_segs, size_t len, loff_t *ppos, iov_fn_t fn) { struct kiocb kiocb; @@ -601,7 +603,7 @@ ssize_t do_sync_readv_writev(struct file *filp, const struct iovec *iov, } /* Do it by hand, with file-ops */ -ssize_t do_loop_readv_writev(struct file *filp, struct iovec *iov, +static ssize_t do_loop_readv_writev(struct file *filp, struct iovec *iov, unsigned long nr_segs, loff_t *ppos, io_fn_t fn) { struct iovec *vector = iov; @@ -743,6 +745,7 @@ static ssize_t do_readv_writev(int type, struct file *file, } else { fn = (io_fn_t)file->f_op->write; fnv = file->f_op->aio_write; + file_start_write(file); } if (fnv) @@ -751,6 +754,9 @@ static ssize_t do_readv_writev(int type, struct file *file, else ret = do_loop_readv_writev(file, iov, nr_segs, pos, fn); + if (type != READ) + file_end_write(file); + out: if (iov != iovstack) kfree(iov); @@ -881,6 +887,201 @@ SYSCALL_DEFINE5(pwritev, unsigned long, fd, const struct iovec __user *, vec, return ret; } +#ifdef CONFIG_COMPAT + +static ssize_t compat_do_readv_writev(int type, struct file *file, + const struct compat_iovec __user *uvector, + unsigned long nr_segs, loff_t *pos) +{ + compat_ssize_t tot_len; + struct iovec iovstack[UIO_FASTIOV]; + struct iovec *iov = iovstack; + ssize_t ret; + io_fn_t fn; + iov_fn_t fnv; + + ret = -EINVAL; + if (!file->f_op) + goto out; + + ret = -EFAULT; + if (!access_ok(VERIFY_READ, uvector, nr_segs*sizeof(*uvector))) + goto out; + + ret = compat_rw_copy_check_uvector(type, uvector, nr_segs, + UIO_FASTIOV, iovstack, &iov); + if (ret <= 0) + goto out; + + tot_len = ret; + ret = rw_verify_area(type, file, pos, tot_len); + if (ret < 0) + goto out; + + fnv = NULL; + if (type == READ) { + fn = file->f_op->read; + fnv = file->f_op->aio_read; + } else { + fn = (io_fn_t)file->f_op->write; + fnv = file->f_op->aio_write; + file_start_write(file); + } + + if (fnv) + ret = do_sync_readv_writev(file, iov, nr_segs, tot_len, + pos, fnv); + else + ret = do_loop_readv_writev(file, iov, nr_segs, pos, fn); + + if (type != READ) + file_end_write(file); + +out: + if (iov != iovstack) + kfree(iov); + if ((ret + (type == READ)) > 0) { + if (type == READ) + fsnotify_access(file); + else + fsnotify_modify(file); + } + return ret; +} + +static size_t compat_readv(struct file *file, + const struct compat_iovec __user *vec, + unsigned long vlen, loff_t *pos) +{ + ssize_t ret = -EBADF; + + if (!(file->f_mode & FMODE_READ)) + goto out; + + ret = -EINVAL; + if (!file->f_op || (!file->f_op->aio_read && !file->f_op->read)) + goto out; + + ret = compat_do_readv_writev(READ, file, vec, vlen, pos); + +out: + if (ret > 0) + add_rchar(current, ret); + inc_syscr(current); + return ret; +} + +COMPAT_SYSCALL_DEFINE3(readv, unsigned long, fd, + const struct compat_iovec __user *,vec, + unsigned long, vlen) +{ + struct fd f = fdget(fd); + ssize_t ret; + loff_t pos; + + if (!f.file) + return -EBADF; + pos = f.file->f_pos; + ret = compat_readv(f.file, vec, vlen, &pos); + f.file->f_pos = pos; + fdput(f); + return ret; +} + +COMPAT_SYSCALL_DEFINE4(preadv64, unsigned long, fd, + const struct compat_iovec __user *,vec, + unsigned long, vlen, loff_t, pos) +{ + struct fd f; + ssize_t ret; + + if (pos < 0) + return -EINVAL; + f = fdget(fd); + if (!f.file) + return -EBADF; + ret = -ESPIPE; + if (f.file->f_mode & FMODE_PREAD) + ret = compat_readv(f.file, vec, vlen, &pos); + fdput(f); + return ret; +} + +COMPAT_SYSCALL_DEFINE5(preadv, unsigned long, fd, + const struct compat_iovec __user *,vec, + unsigned long, vlen, u32, pos_low, u32, pos_high) +{ + loff_t pos = ((loff_t)pos_high << 32) | pos_low; + return compat_sys_preadv64(fd, vec, vlen, pos); +} + +static size_t compat_writev(struct file *file, + const struct compat_iovec __user *vec, + unsigned long vlen, loff_t *pos) +{ + ssize_t ret = -EBADF; + + if (!(file->f_mode & FMODE_WRITE)) + goto out; + + ret = -EINVAL; + if (!file->f_op || (!file->f_op->aio_write && !file->f_op->write)) + goto out; + + ret = compat_do_readv_writev(WRITE, file, vec, vlen, pos); + +out: + if (ret > 0) + add_wchar(current, ret); + inc_syscw(current); + return ret; +} + +COMPAT_SYSCALL_DEFINE3(writev, unsigned long, fd, + const struct compat_iovec __user *, vec, + unsigned long, vlen) +{ + struct fd f = fdget(fd); + ssize_t ret; + loff_t pos; + + if (!f.file) + return -EBADF; + pos = f.file->f_pos; + ret = compat_writev(f.file, vec, vlen, &pos); + f.file->f_pos = pos; + fdput(f); + return ret; +} + +COMPAT_SYSCALL_DEFINE4(pwritev64, unsigned long, fd, + const struct compat_iovec __user *,vec, + unsigned long, vlen, loff_t, pos) +{ + struct fd f; + ssize_t ret; + + if (pos < 0) + return -EINVAL; + f = fdget(fd); + if (!f.file) + return -EBADF; + ret = -ESPIPE; + if (f.file->f_mode & FMODE_PWRITE) + ret = compat_writev(f.file, vec, vlen, &pos); + fdput(f); + return ret; +} + +COMPAT_SYSCALL_DEFINE5(pwritev, unsigned long, fd, + const struct compat_iovec __user *,vec, + unsigned long, vlen, u32, pos_low, u32, pos_high) +{ + loff_t pos = ((loff_t)pos_high << 32) | pos_low; + return compat_sys_pwritev64(fd, vec, vlen, pos); +} +#endif + static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, size_t count, loff_t max) { diff --git a/fs/read_write.h b/fs/read_write.h index d07b954c6e0c..0ec530d9305b 100644 --- a/fs/read_write.h +++ b/fs/read_write.h @@ -7,8 +7,3 @@ typedef ssize_t (*io_fn_t)(struct file *, char __user *, size_t, loff_t *); typedef ssize_t (*iov_fn_t)(struct kiocb *, const struct iovec *, unsigned long, loff_t); - -ssize_t do_sync_readv_writev(struct file *filp, const struct iovec *iov, - unsigned long nr_segs, size_t len, loff_t *ppos, iov_fn_t fn); -ssize_t do_loop_readv_writev(struct file *filp, struct iovec *iov, - unsigned long nr_segs, loff_t *ppos, io_fn_t fn); diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index 6165bd4784f6..dcaafcfc23b0 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -234,68 +234,9 @@ int reiserfs_commit_page(struct inode *inode, struct page *page, return ret; } -/* Write @count bytes at position @ppos in a file indicated by @file - from the buffer @buf. - - generic_file_write() is only appropriate for filesystems that are not seeking to optimize performance and want - something simple that works. It is not for serious use by general purpose filesystems, excepting the one that it was - written for (ext2/3). This is for several reasons: - - * It has no understanding of any filesystem specific optimizations. - - * It enters the filesystem repeatedly for each page that is written. - - * It depends on reiserfs_get_block() function which if implemented by reiserfs performs costly search_by_key - * operation for each page it is supplied with. By contrast reiserfs_file_write() feeds as much as possible at a time - * to reiserfs which allows for fewer tree traversals. - - * Each indirect pointer insertion takes a lot of cpu, because it involves memory moves inside of blocks. - - * Asking the block allocation code for blocks one at a time is slightly less efficient. - - All of these reasons for not using only generic file write were understood back when reiserfs was first miscoded to - use it, but we were in a hurry to make code freeze, and so it couldn't be revised then. This new code should make - things right finally. - - Future Features: providing search_by_key with hints. - -*/ -static ssize_t reiserfs_file_write(struct file *file, /* the file we are going to write into */ - const char __user * buf, /* pointer to user supplied data - (in userspace) */ - size_t count, /* amount of bytes to write */ - loff_t * ppos /* pointer to position in file that we start writing at. Should be updated to - * new current position before returning. */ - ) -{ - struct inode *inode = file_inode(file); // Inode of the file that we are writing to. - /* To simplify coding at this time, we store - locked pages in array for now */ - struct reiserfs_transaction_handle th; - th.t_trans_id = 0; - - /* If a filesystem is converted from 3.5 to 3.6, we'll have v3.5 items - * lying around (most of the disk, in fact). Despite the filesystem - * now being a v3.6 format, the old items still can't support large - * file sizes. Catch this case here, as the rest of the VFS layer is - * oblivious to the different limitations between old and new items. - * reiserfs_setattr catches this for truncates. This chunk is lifted - * from generic_write_checks. */ - if (get_inode_item_key_version (inode) == KEY_FORMAT_3_5 && - *ppos + count > MAX_NON_LFS) { - if (*ppos >= MAX_NON_LFS) { - return -EFBIG; - } - if (count > MAX_NON_LFS - (unsigned long)*ppos) - count = MAX_NON_LFS - (unsigned long)*ppos; - } - - return do_sync_write(file, buf, count, ppos); -} - const struct file_operations reiserfs_file_operations = { .read = do_sync_read, - .write = reiserfs_file_write, + .write = do_sync_write, .unlocked_ioctl = reiserfs_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = reiserfs_compat_ioctl, diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c index 9cc0740adffa..33532f79b4f7 100644 --- a/fs/reiserfs/procfs.c +++ b/fs/reiserfs/procfs.c @@ -394,20 +394,24 @@ static int set_sb(struct super_block *sb, void *data) return -ENOENT; } +struct reiserfs_seq_private { + struct super_block *sb; + int (*show) (struct seq_file *, struct super_block *); +}; + static void *r_start(struct seq_file *m, loff_t * pos) { - struct proc_dir_entry *de = m->private; - struct super_block *s = de->parent->data; + struct reiserfs_seq_private *priv = m->private; loff_t l = *pos; if (l) return NULL; - if (IS_ERR(sget(&reiserfs_fs_type, test_sb, set_sb, 0, s))) + if (IS_ERR(sget(&reiserfs_fs_type, test_sb, set_sb, 0, priv->sb))) return NULL; - up_write(&s->s_umount); - return s; + up_write(&priv->sb->s_umount); + return priv->sb; } static void *r_next(struct seq_file *m, void *v, loff_t * pos) @@ -426,9 +430,8 @@ static void r_stop(struct seq_file *m, void *v) static int r_show(struct seq_file *m, void *v) { - struct proc_dir_entry *de = m->private; - int (*show) (struct seq_file *, struct super_block *) = de->data; - return show(m, v); + struct reiserfs_seq_private *priv = m->private; + return priv->show(m, v); } static const struct seq_operations r_ops = { @@ -440,11 +443,15 @@ static const struct seq_operations r_ops = { static int r_open(struct inode *inode, struct file *file) { - int ret = seq_open(file, &r_ops); + struct reiserfs_seq_private *priv; + int ret = seq_open_private(file, &r_ops, + sizeof(struct reiserfs_seq_private)); if (!ret) { struct seq_file *m = file->private_data; - m->private = PDE(inode); + priv = m->private; + priv->sb = proc_get_parent_data(inode); + priv->show = PDE_DATA(inode); } return ret; } @@ -453,7 +460,7 @@ static const struct file_operations r_file_operations = { .open = r_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release, + .release = seq_release_private, .owner = THIS_MODULE, }; @@ -479,9 +486,8 @@ int reiserfs_proc_info_init(struct super_block *sb) *s = '!'; spin_lock_init(&__PINFO(sb).lock); - REISERFS_SB(sb)->procdir = proc_mkdir(b, proc_info_root); + REISERFS_SB(sb)->procdir = proc_mkdir_data(b, 0, proc_info_root, sb); if (REISERFS_SB(sb)->procdir) { - REISERFS_SB(sb)->procdir->data = sb; add_file(sb, "version", show_version); add_file(sb, "super", show_super); add_file(sb, "per-level", show_per_level); @@ -499,29 +505,17 @@ int reiserfs_proc_info_init(struct super_block *sb) int reiserfs_proc_info_done(struct super_block *sb) { struct proc_dir_entry *de = REISERFS_SB(sb)->procdir; - char b[BDEVNAME_SIZE]; - char *s; + if (de) { + char b[BDEVNAME_SIZE]; + char *s; - /* Some block devices use /'s */ - strlcpy(b, reiserfs_bdevname(sb), BDEVNAME_SIZE); - s = strchr(b, '/'); - if (s) - *s = '!'; + /* Some block devices use /'s */ + strlcpy(b, reiserfs_bdevname(sb), BDEVNAME_SIZE); + s = strchr(b, '/'); + if (s) + *s = '!'; - if (de) { - remove_proc_entry("journal", de); - remove_proc_entry("oidmap", de); - remove_proc_entry("on-disk-super", de); - remove_proc_entry("bitmap", de); - remove_proc_entry("per-level", de); - remove_proc_entry("super", de); - remove_proc_entry("version", de); - } - spin_lock(&__PINFO(sb).lock); - __PINFO(sb).exiting = 1; - spin_unlock(&__PINFO(sb).lock); - if (proc_info_root) { - remove_proc_entry(b, proc_info_root); + remove_proc_subtree(b, proc_info_root); REISERFS_SB(sb)->procdir = NULL; } return 0; diff --git a/fs/seq_file.c b/fs/seq_file.c index 38bb59f3f2ad..774c1eb7f1c9 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -599,6 +599,24 @@ int single_open(struct file *file, int (*show)(struct seq_file *, void *), } EXPORT_SYMBOL(single_open); +int single_open_size(struct file *file, int (*show)(struct seq_file *, void *), + void *data, size_t size) +{ + char *buf = kmalloc(size, GFP_KERNEL); + int ret; + if (!buf) + return -ENOMEM; + ret = single_open(file, show, data); + if (ret) { + kfree(buf); + return ret; + } + ((struct seq_file *)file->private_data)->buf = buf; + ((struct seq_file *)file->private_data)->size = size; + return 0; +} +EXPORT_SYMBOL(single_open_size); + int single_release(struct inode *inode, struct file *file) { const struct seq_operations *op = ((struct seq_file *)file->private_data)->op; diff --git a/fs/splice.c b/fs/splice.c index 6b485b8753bd..e6b25598c8c4 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -219,7 +219,7 @@ ssize_t splice_to_pipe(struct pipe_inode_info *pipe, page_nr++; ret += buf->len; - if (pipe->inode) + if (pipe->files) do_wakeup = 1; if (!--spd->nr_pages) @@ -829,7 +829,7 @@ int splice_from_pipe_feed(struct pipe_inode_info *pipe, struct splice_desc *sd, ops->release(pipe, buf); pipe->curbuf = (pipe->curbuf + 1) & (pipe->buffers - 1); pipe->nrbufs--; - if (pipe->inode) + if (pipe->files) sd->need_wakeup = true; } @@ -1001,8 +1001,6 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, }; ssize_t ret; - sb_start_write(inode->i_sb); - pipe_lock(pipe); splice_from_pipe_begin(&sd); @@ -1038,7 +1036,6 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, *ppos += ret; balance_dirty_pages_ratelimited(mapping); } - sb_end_write(inode->i_sb); return ret; } @@ -1118,7 +1115,10 @@ static long do_splice_from(struct pipe_inode_info *pipe, struct file *out, else splice_write = default_file_splice_write; - return splice_write(pipe, out, ppos, len, flags); + file_start_write(out); + ret = splice_write(pipe, out, ppos, len, flags); + file_end_write(out); + return ret; } /* @@ -1184,7 +1184,7 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd, */ pipe = current->splice_pipe; if (unlikely(!pipe)) { - pipe = alloc_pipe_info(NULL); + pipe = alloc_pipe_info(); if (!pipe) return -ENOMEM; diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index f03bf1a456fb..3800128d2171 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -775,8 +775,6 @@ xfs_file_aio_write( if (ocount == 0) return 0; - sb_start_write(inode->i_sb); - if (XFS_FORCED_SHUTDOWN(ip->i_mount)) { ret = -EIO; goto out; @@ -800,7 +798,6 @@ xfs_file_aio_write( } out: - sb_end_write(inode->i_sb); return ret; } |