diff options
Diffstat (limited to 'fs')
167 files changed, 3153 insertions, 3322 deletions
diff --git a/fs/9p/vfs_dir.c b/fs/9p/vfs_dir.c index be1e34adc3c6..4d0c2e0be7e5 100644 --- a/fs/9p/vfs_dir.c +++ b/fs/9p/vfs_dir.c @@ -101,16 +101,15 @@ static struct p9_rdir *v9fs_alloc_rdir_buf(struct file *filp, int buflen) } /** - * v9fs_dir_readdir - read a directory - * @filp: opened file structure - * @dirent: directory structure ??? - * @filldir: function to populate directory structure ??? + * v9fs_dir_readdir - iterate through a directory + * @file: opened file structure + * @ctx: actor we feed the entries to * */ -static int v9fs_dir_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int v9fs_dir_readdir(struct file *file, struct dir_context *ctx) { - int over; + bool over; struct p9_wstat st; int err = 0; struct p9_fid *fid; @@ -118,19 +117,19 @@ static int v9fs_dir_readdir(struct file *filp, void *dirent, filldir_t filldir) int reclen = 0; struct p9_rdir *rdir; - p9_debug(P9_DEBUG_VFS, "name %s\n", filp->f_path.dentry->d_name.name); - fid = filp->private_data; + p9_debug(P9_DEBUG_VFS, "name %s\n", file->f_path.dentry->d_name.name); + fid = file->private_data; buflen = fid->clnt->msize - P9_IOHDRSZ; - rdir = v9fs_alloc_rdir_buf(filp, buflen); + rdir = v9fs_alloc_rdir_buf(file, buflen); if (!rdir) return -ENOMEM; while (1) { if (rdir->tail == rdir->head) { - err = v9fs_file_readn(filp, rdir->buf, NULL, - buflen, filp->f_pos); + err = v9fs_file_readn(file, rdir->buf, NULL, + buflen, ctx->pos); if (err <= 0) return err; @@ -148,51 +147,45 @@ static int v9fs_dir_readdir(struct file *filp, void *dirent, filldir_t filldir) } reclen = st.size+2; - over = filldir(dirent, st.name, strlen(st.name), - filp->f_pos, v9fs_qid2ino(&st.qid), dt_type(&st)); - + over = !dir_emit(ctx, st.name, strlen(st.name), + v9fs_qid2ino(&st.qid), dt_type(&st)); p9stat_free(&st); - if (over) return 0; rdir->head += reclen; - filp->f_pos += reclen; + ctx->pos += reclen; } } } /** - * v9fs_dir_readdir_dotl - read a directory - * @filp: opened file structure - * @dirent: buffer to fill dirent structures - * @filldir: function to populate dirent structures + * v9fs_dir_readdir_dotl - iterate through a directory + * @file: opened file structure + * @ctx: actor we feed the entries to * */ -static int v9fs_dir_readdir_dotl(struct file *filp, void *dirent, - filldir_t filldir) +static int v9fs_dir_readdir_dotl(struct file *file, struct dir_context *ctx) { - int over; int err = 0; struct p9_fid *fid; int buflen; struct p9_rdir *rdir; struct p9_dirent curdirent; - u64 oldoffset = 0; - p9_debug(P9_DEBUG_VFS, "name %s\n", filp->f_path.dentry->d_name.name); - fid = filp->private_data; + p9_debug(P9_DEBUG_VFS, "name %s\n", file->f_path.dentry->d_name.name); + fid = file->private_data; buflen = fid->clnt->msize - P9_READDIRHDRSZ; - rdir = v9fs_alloc_rdir_buf(filp, buflen); + rdir = v9fs_alloc_rdir_buf(file, buflen); if (!rdir) return -ENOMEM; while (1) { if (rdir->tail == rdir->head) { err = p9_client_readdir(fid, rdir->buf, buflen, - filp->f_pos); + ctx->pos); if (err <= 0) return err; @@ -210,22 +203,13 @@ static int v9fs_dir_readdir_dotl(struct file *filp, void *dirent, return -EIO; } - /* d_off in dirent structure tracks the offset into - * the next dirent in the dir. However, filldir() - * expects offset into the current dirent. Hence - * while calling filldir send the offset from the - * previous dirent structure. - */ - over = filldir(dirent, curdirent.d_name, - strlen(curdirent.d_name), - oldoffset, v9fs_qid2ino(&curdirent.qid), - curdirent.d_type); - oldoffset = curdirent.d_off; - - if (over) + if (!dir_emit(ctx, curdirent.d_name, + strlen(curdirent.d_name), + v9fs_qid2ino(&curdirent.qid), + curdirent.d_type)) return 0; - filp->f_pos = curdirent.d_off; + ctx->pos = curdirent.d_off; rdir->head += err; } } @@ -254,7 +238,7 @@ int v9fs_dir_release(struct inode *inode, struct file *filp) const struct file_operations v9fs_dir_operations = { .read = generic_read_dir, .llseek = generic_file_llseek, - .readdir = v9fs_dir_readdir, + .iterate = v9fs_dir_readdir, .open = v9fs_file_open, .release = v9fs_dir_release, }; @@ -262,7 +246,7 @@ const struct file_operations v9fs_dir_operations = { const struct file_operations v9fs_dir_operations_dotl = { .read = generic_read_dir, .llseek = generic_file_llseek, - .readdir = v9fs_dir_readdir_dotl, + .iterate = v9fs_dir_readdir_dotl, .open = v9fs_file_open, .release = v9fs_dir_release, .fsync = v9fs_file_fsync_dotl, diff --git a/fs/adfs/dir.c b/fs/adfs/dir.c index 9cf874ce8336..ade28bb058e3 100644 --- a/fs/adfs/dir.c +++ b/fs/adfs/dir.c @@ -17,47 +17,43 @@ static DEFINE_RWLOCK(adfs_dir_lock); static int -adfs_readdir(struct file *filp, void *dirent, filldir_t filldir) +adfs_readdir(struct file *file, struct dir_context *ctx) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; struct adfs_dir_ops *ops = ADFS_SB(sb)->s_dir; struct object_info obj; struct adfs_dir dir; int ret = 0; - if (filp->f_pos >> 32) - goto out; + if (ctx->pos >> 32) + return 0; ret = ops->read(sb, inode->i_ino, inode->i_size, &dir); if (ret) - goto out; + return ret; - switch ((unsigned long)filp->f_pos) { - case 0: - if (filldir(dirent, ".", 1, 0, inode->i_ino, DT_DIR) < 0) + if (ctx->pos == 0) { + if (!dir_emit_dot(file, ctx)) goto free_out; - filp->f_pos += 1; - - case 1: - if (filldir(dirent, "..", 2, 1, dir.parent_id, DT_DIR) < 0) + ctx->pos = 1; + } + if (ctx->pos == 1) { + if (!dir_emit(ctx, "..", 2, dir.parent_id, DT_DIR)) goto free_out; - filp->f_pos += 1; - - default: - break; + ctx->pos = 2; } read_lock(&adfs_dir_lock); - ret = ops->setpos(&dir, filp->f_pos - 2); + ret = ops->setpos(&dir, ctx->pos - 2); if (ret) goto unlock_out; while (ops->getnext(&dir, &obj) == 0) { - if (filldir(dirent, obj.name, obj.name_len, - filp->f_pos, obj.file_id, DT_UNKNOWN) < 0) - goto unlock_out; - filp->f_pos += 1; + if (!dir_emit(ctx, obj.name, obj.name_len, + obj.file_id, DT_UNKNOWN)) + break; + ctx->pos++; } unlock_out: @@ -65,8 +61,6 @@ unlock_out: free_out: ops->free(&dir); - -out: return ret; } @@ -192,7 +186,7 @@ out: const struct file_operations adfs_dir_operations = { .read = generic_read_dir, .llseek = generic_file_llseek, - .readdir = adfs_readdir, + .iterate = adfs_readdir, .fsync = generic_file_fsync, }; diff --git a/fs/affs/dir.c b/fs/affs/dir.c index fd11a6d608ee..f1eba8c3644e 100644 --- a/fs/affs/dir.c +++ b/fs/affs/dir.c @@ -15,12 +15,12 @@ #include "affs.h" -static int affs_readdir(struct file *, void *, filldir_t); +static int affs_readdir(struct file *, struct dir_context *); const struct file_operations affs_dir_operations = { .read = generic_read_dir, .llseek = generic_file_llseek, - .readdir = affs_readdir, + .iterate = affs_readdir, .fsync = affs_file_fsync, }; @@ -40,52 +40,35 @@ const struct inode_operations affs_dir_inode_operations = { }; static int -affs_readdir(struct file *filp, void *dirent, filldir_t filldir) +affs_readdir(struct file *file, struct dir_context *ctx) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; - struct buffer_head *dir_bh; - struct buffer_head *fh_bh; + struct buffer_head *dir_bh = NULL; + struct buffer_head *fh_bh = NULL; unsigned char *name; int namelen; u32 i; int hash_pos; int chain_pos; - u32 f_pos; u32 ino; - int stored; - int res; - pr_debug("AFFS: readdir(ino=%lu,f_pos=%lx)\n",inode->i_ino,(unsigned long)filp->f_pos); + pr_debug("AFFS: readdir(ino=%lu,f_pos=%lx)\n",inode->i_ino,(unsigned long)ctx->pos); - stored = 0; - res = -EIO; - dir_bh = NULL; - fh_bh = NULL; - f_pos = filp->f_pos; - - if (f_pos == 0) { - filp->private_data = (void *)0; - if (filldir(dirent, ".", 1, f_pos, inode->i_ino, DT_DIR) < 0) + if (ctx->pos < 2) { + file->private_data = (void *)0; + if (!dir_emit_dots(file, ctx)) return 0; - filp->f_pos = f_pos = 1; - stored++; - } - if (f_pos == 1) { - if (filldir(dirent, "..", 2, f_pos, parent_ino(filp->f_path.dentry), DT_DIR) < 0) - return stored; - filp->f_pos = f_pos = 2; - stored++; } affs_lock_dir(inode); - chain_pos = (f_pos - 2) & 0xffff; - hash_pos = (f_pos - 2) >> 16; + chain_pos = (ctx->pos - 2) & 0xffff; + hash_pos = (ctx->pos - 2) >> 16; if (chain_pos == 0xffff) { affs_warning(sb, "readdir", "More than 65535 entries in chain"); chain_pos = 0; hash_pos++; - filp->f_pos = ((hash_pos << 16) | chain_pos) + 2; + ctx->pos = ((hash_pos << 16) | chain_pos) + 2; } dir_bh = affs_bread(sb, inode->i_ino); if (!dir_bh) @@ -94,8 +77,8 @@ affs_readdir(struct file *filp, void *dirent, filldir_t filldir) /* If the directory hasn't changed since the last call to readdir(), * we can jump directly to where we left off. */ - ino = (u32)(long)filp->private_data; - if (ino && filp->f_version == inode->i_version) { + ino = (u32)(long)file->private_data; + if (ino && file->f_version == inode->i_version) { pr_debug("AFFS: readdir() left off=%d\n", ino); goto inside; } @@ -105,7 +88,7 @@ affs_readdir(struct file *filp, void *dirent, filldir_t filldir) fh_bh = affs_bread(sb, ino); if (!fh_bh) { affs_error(sb, "readdir","Cannot read block %d", i); - goto readdir_out; + return -EIO; } ino = be32_to_cpu(AFFS_TAIL(sb, fh_bh)->hash_chain); affs_brelse(fh_bh); @@ -119,38 +102,34 @@ affs_readdir(struct file *filp, void *dirent, filldir_t filldir) ino = be32_to_cpu(AFFS_HEAD(dir_bh)->table[hash_pos]); if (!ino) continue; - f_pos = (hash_pos << 16) + 2; + ctx->pos = (hash_pos << 16) + 2; inside: do { fh_bh = affs_bread(sb, ino); if (!fh_bh) { affs_error(sb, "readdir","Cannot read block %d", ino); - goto readdir_done; + break; } namelen = min(AFFS_TAIL(sb, fh_bh)->name[0], (u8)30); name = AFFS_TAIL(sb, fh_bh)->name + 1; pr_debug("AFFS: readdir(): filldir(\"%.*s\", ino=%u), hash=%d, f_pos=%x\n", - namelen, name, ino, hash_pos, f_pos); - if (filldir(dirent, name, namelen, f_pos, ino, DT_UNKNOWN) < 0) + namelen, name, ino, hash_pos, (u32)ctx->pos); + if (!dir_emit(ctx, name, namelen, ino, DT_UNKNOWN)) goto readdir_done; - stored++; - f_pos++; + ctx->pos++; ino = be32_to_cpu(AFFS_TAIL(sb, fh_bh)->hash_chain); affs_brelse(fh_bh); fh_bh = NULL; } while (ino); } readdir_done: - filp->f_pos = f_pos; - filp->f_version = inode->i_version; - filp->private_data = (void *)(long)ino; - res = stored; + file->f_version = inode->i_version; + file->private_data = (void *)(long)ino; readdir_out: affs_brelse(dir_bh); affs_brelse(fh_bh); affs_unlock_dir(inode); - pr_debug("AFFS: readdir()=%d\n", stored); - return res; + return 0; } diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 7a465ed04444..34494fbead0a 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -22,7 +22,7 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags); static int afs_dir_open(struct inode *inode, struct file *file); -static int afs_readdir(struct file *file, void *dirent, filldir_t filldir); +static int afs_readdir(struct file *file, struct dir_context *ctx); static int afs_d_revalidate(struct dentry *dentry, unsigned int flags); static int afs_d_delete(const struct dentry *dentry); static void afs_d_release(struct dentry *dentry); @@ -43,7 +43,7 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, const struct file_operations afs_dir_file_operations = { .open = afs_dir_open, .release = afs_release, - .readdir = afs_readdir, + .iterate = afs_readdir, .lock = afs_lock, .llseek = generic_file_llseek, }; @@ -119,9 +119,9 @@ struct afs_dir_page { }; struct afs_lookup_cookie { + struct dir_context ctx; struct afs_fid fid; - const char *name; - size_t nlen; + struct qstr name; int found; }; @@ -228,20 +228,18 @@ static int afs_dir_open(struct inode *inode, struct file *file) /* * deal with one block in an AFS directory */ -static int afs_dir_iterate_block(unsigned *fpos, +static int afs_dir_iterate_block(struct dir_context *ctx, union afs_dir_block *block, - unsigned blkoff, - void *cookie, - filldir_t filldir) + unsigned blkoff) { union afs_dirent *dire; unsigned offset, next, curr; size_t nlen; - int tmp, ret; + int tmp; - _enter("%u,%x,%p,,",*fpos,blkoff,block); + _enter("%u,%x,%p,,",(unsigned)ctx->pos,blkoff,block); - curr = (*fpos - blkoff) / sizeof(union afs_dirent); + curr = (ctx->pos - blkoff) / sizeof(union afs_dirent); /* walk through the block, an entry at a time */ for (offset = AFS_DIRENT_PER_BLOCK - block->pagehdr.nentries; @@ -256,7 +254,7 @@ static int afs_dir_iterate_block(unsigned *fpos, _debug("ENT[%Zu.%u]: unused", blkoff / sizeof(union afs_dir_block), offset); if (offset >= curr) - *fpos = blkoff + + ctx->pos = blkoff + next * sizeof(union afs_dirent); continue; } @@ -302,19 +300,15 @@ static int afs_dir_iterate_block(unsigned *fpos, continue; /* found the next entry */ - ret = filldir(cookie, - dire->u.name, - nlen, - blkoff + offset * sizeof(union afs_dirent), + if (!dir_emit(ctx, dire->u.name, nlen, ntohl(dire->u.vnode), - filldir == afs_lookup_filldir ? - ntohl(dire->u.unique) : DT_UNKNOWN); - if (ret < 0) { + ctx->actor == afs_lookup_filldir ? + ntohl(dire->u.unique) : DT_UNKNOWN)) { _leave(" = 0 [full]"); return 0; } - *fpos = blkoff + next * sizeof(union afs_dirent); + ctx->pos = blkoff + next * sizeof(union afs_dirent); } _leave(" = 1 [more]"); @@ -324,8 +318,8 @@ static int afs_dir_iterate_block(unsigned *fpos, /* * iterate through the data blob that lists the contents of an AFS directory */ -static int afs_dir_iterate(struct inode *dir, unsigned *fpos, void *cookie, - filldir_t filldir, struct key *key) +static int afs_dir_iterate(struct inode *dir, struct dir_context *ctx, + struct key *key) { union afs_dir_block *dblock; struct afs_dir_page *dbuf; @@ -333,7 +327,7 @@ static int afs_dir_iterate(struct inode *dir, unsigned *fpos, void *cookie, unsigned blkoff, limit; int ret; - _enter("{%lu},%u,,", dir->i_ino, *fpos); + _enter("{%lu},%u,,", dir->i_ino, (unsigned)ctx->pos); if (test_bit(AFS_VNODE_DELETED, &AFS_FS_I(dir)->flags)) { _leave(" = -ESTALE"); @@ -341,13 +335,13 @@ static int afs_dir_iterate(struct inode *dir, unsigned *fpos, void *cookie, } /* round the file position up to the next entry boundary */ - *fpos += sizeof(union afs_dirent) - 1; - *fpos &= ~(sizeof(union afs_dirent) - 1); + ctx->pos += sizeof(union afs_dirent) - 1; + ctx->pos &= ~(sizeof(union afs_dirent) - 1); /* walk through the blocks in sequence */ ret = 0; - while (*fpos < dir->i_size) { - blkoff = *fpos & ~(sizeof(union afs_dir_block) - 1); + while (ctx->pos < dir->i_size) { + blkoff = ctx->pos & ~(sizeof(union afs_dir_block) - 1); /* fetch the appropriate page from the directory */ page = afs_dir_get_page(dir, blkoff / PAGE_SIZE, key); @@ -364,8 +358,7 @@ static int afs_dir_iterate(struct inode *dir, unsigned *fpos, void *cookie, do { dblock = &dbuf->blocks[(blkoff % PAGE_SIZE) / sizeof(union afs_dir_block)]; - ret = afs_dir_iterate_block(fpos, dblock, blkoff, - cookie, filldir); + ret = afs_dir_iterate_block(ctx, dblock, blkoff); if (ret != 1) { afs_dir_put_page(page); goto out; @@ -373,7 +366,7 @@ static int afs_dir_iterate(struct inode *dir, unsigned *fpos, void *cookie, blkoff += sizeof(union afs_dir_block); - } while (*fpos < dir->i_size && blkoff < limit); + } while (ctx->pos < dir->i_size && blkoff < limit); afs_dir_put_page(page); ret = 0; @@ -387,23 +380,10 @@ out: /* * read an AFS directory */ -static int afs_readdir(struct file *file, void *cookie, filldir_t filldir) +static int afs_readdir(struct file *file, struct dir_context *ctx) { - unsigned fpos; - int ret; - - _enter("{%Ld,{%lu}}", - file->f_pos, file_inode(file)->i_ino); - - ASSERT(file->private_data != NULL); - - fpos = file->f_pos; - ret = afs_dir_iterate(file_inode(file), &fpos, - cookie, filldir, file->private_data); - file->f_pos = fpos; - - _leave(" = %d", ret); - return ret; + return afs_dir_iterate(file_inode(file), + ctx, file->private_data); } /* @@ -416,15 +396,16 @@ static int afs_lookup_filldir(void *_cookie, const char *name, int nlen, { struct afs_lookup_cookie *cookie = _cookie; - _enter("{%s,%Zu},%s,%u,,%llu,%u", - cookie->name, cookie->nlen, name, nlen, + _enter("{%s,%u},%s,%u,,%llu,%u", + cookie->name.name, cookie->name.len, name, nlen, (unsigned long long) ino, dtype); /* insanity checks first */ BUILD_BUG_ON(sizeof(union afs_dir_block) != 2048); BUILD_BUG_ON(sizeof(union afs_dirent) != 32); - if (cookie->nlen != nlen || memcmp(cookie->name, name, nlen) != 0) { + if (cookie->name.len != nlen || + memcmp(cookie->name.name, name, nlen) != 0) { _leave(" = 0 [no]"); return 0; } @@ -444,24 +425,18 @@ static int afs_lookup_filldir(void *_cookie, const char *name, int nlen, static int afs_do_lookup(struct inode *dir, struct dentry *dentry, struct afs_fid *fid, struct key *key) { - struct afs_lookup_cookie cookie; - struct afs_super_info *as; - unsigned fpos; + struct afs_super_info *as = dir->i_sb->s_fs_info; + struct afs_lookup_cookie cookie = { + .ctx.actor = afs_lookup_filldir, + .name = dentry->d_name, + .fid.vid = as->volume->vid + }; int ret; _enter("{%lu},%p{%s},", dir->i_ino, dentry, dentry->d_name.name); - as = dir->i_sb->s_fs_info; - /* search the directory */ - cookie.name = dentry->d_name.name; - cookie.nlen = dentry->d_name.len; - cookie.fid.vid = as->volume->vid; - cookie.found = 0; - - fpos = 0; - ret = afs_dir_iterate(dir, &fpos, &cookie, afs_lookup_filldir, - key); + ret = afs_dir_iterate(dir, &cookie.ctx, key); if (ret < 0) { _leave(" = %d [iter]", ret); return ret; @@ -141,9 +141,6 @@ static void aio_free_ring(struct kioctx *ctx) for (i = 0; i < ctx->nr_pages; i++) put_page(ctx->ring_pages[i]); - if (ctx->mmap_size) - vm_munmap(ctx->mmap_base, ctx->mmap_size); - if (ctx->ring_pages && ctx->ring_pages != ctx->internal_pages) kfree(ctx->ring_pages); } @@ -307,7 +304,9 @@ static void free_ioctx(struct kioctx *ctx) kunmap_atomic(ring); while (atomic_read(&ctx->reqs_active) > 0) { - wait_event(ctx->wait, head != ctx->tail); + wait_event(ctx->wait, + head != ctx->tail || + atomic_read(&ctx->reqs_active) <= 0); avail = (head <= ctx->tail ? ctx->tail : ctx->nr_events) - head; @@ -320,11 +319,6 @@ static void free_ioctx(struct kioctx *ctx) aio_free_ring(ctx); - spin_lock(&aio_nr_lock); - BUG_ON(aio_nr - ctx->max_reqs > aio_nr); - aio_nr -= ctx->max_reqs; - spin_unlock(&aio_nr_lock); - pr_debug("freeing %p\n", ctx); /* @@ -433,17 +427,24 @@ static void kill_ioctx(struct kioctx *ctx) { if (!atomic_xchg(&ctx->dead, 1)) { hlist_del_rcu(&ctx->list); - /* Between hlist_del_rcu() and dropping the initial ref */ - synchronize_rcu(); /* - * We can't punt to workqueue here because put_ioctx() -> - * free_ioctx() will unmap the ringbuffer, and that has to be - * done in the original process's context. kill_ioctx_rcu/work() - * exist for exit_aio(), as in that path free_ioctx() won't do - * the unmap. + * It'd be more correct to do this in free_ioctx(), after all + * the outstanding kiocbs have finished - but by then io_destroy + * has already returned, so io_setup() could potentially return + * -EAGAIN with no ioctxs actually in use (as far as userspace + * could tell). */ - kill_ioctx_work(&ctx->rcu_work); + spin_lock(&aio_nr_lock); + BUG_ON(aio_nr - ctx->max_reqs > aio_nr); + aio_nr -= ctx->max_reqs; + spin_unlock(&aio_nr_lock); + + if (ctx->mmap_size) + vm_munmap(ctx->mmap_base, ctx->mmap_size); + + /* Between hlist_del_rcu() and dropping the initial ref */ + call_rcu(&ctx->rcu_head, kill_ioctx_rcu); } } @@ -493,10 +494,7 @@ void exit_aio(struct mm_struct *mm) */ ctx->mmap_size = 0; - if (!atomic_xchg(&ctx->dead, 1)) { - hlist_del_rcu(&ctx->list); - call_rcu(&ctx->rcu_head, kill_ioctx_rcu); - } + kill_ioctx(ctx); } } @@ -1299,8 +1297,7 @@ SYSCALL_DEFINE3(io_cancel, aio_context_t, ctx_id, struct iocb __user *, iocb, * < min_nr if the timeout specified by timeout has elapsed * before sufficient events are available, where timeout == NULL * specifies an infinite timeout. Note that the timeout pointed to by - * timeout is relative and will be updated if not NULL and the - * operation blocks. Will fail with -ENOSYS if not implemented. + * timeout is relative. Will fail with -ENOSYS if not implemented. */ SYSCALL_DEFINE5(io_getevents, aio_context_t, ctx_id, long, min_nr, diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 085da86e07c2..ca8e55548d98 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -41,7 +41,7 @@ const struct file_operations autofs4_root_operations = { .open = dcache_dir_open, .release = dcache_dir_close, .read = generic_read_dir, - .readdir = dcache_readdir, + .iterate = dcache_readdir, .llseek = dcache_dir_lseek, .unlocked_ioctl = autofs4_root_ioctl, #ifdef CONFIG_COMPAT @@ -53,7 +53,7 @@ const struct file_operations autofs4_dir_operations = { .open = autofs4_dir_open, .release = dcache_dir_close, .read = generic_read_dir, - .readdir = dcache_readdir, + .iterate = dcache_readdir, .llseek = dcache_dir_lseek, }; diff --git a/fs/bad_inode.c b/fs/bad_inode.c index 922ad460bff9..7c93953030fb 100644 --- a/fs/bad_inode.c +++ b/fs/bad_inode.c @@ -45,7 +45,7 @@ static ssize_t bad_file_aio_write(struct kiocb *iocb, const struct iovec *iov, return -EIO; } -static int bad_file_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int bad_file_readdir(struct file *file, struct dir_context *ctx) { return -EIO; } @@ -152,7 +152,7 @@ static const struct file_operations bad_file_ops = .write = bad_file_write, .aio_read = bad_file_aio_read, .aio_write = bad_file_aio_write, - .readdir = bad_file_readdir, + .iterate = bad_file_readdir, .poll = bad_file_poll, .unlocked_ioctl = bad_file_unlocked_ioctl, .compat_ioctl = bad_file_compat_ioctl, diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index 8615ee89ab55..e9c75e20db32 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -31,7 +31,7 @@ MODULE_LICENSE("GPL"); /* The units the vfs expects inode->i_blocks to be in */ #define VFS_BLOCK_SIZE 512 -static int befs_readdir(struct file *, void *, filldir_t); +static int befs_readdir(struct file *, struct dir_context *); static int befs_get_block(struct inode *, sector_t, struct buffer_head *, int); static int befs_readpage(struct file *file, struct page *page); static sector_t befs_bmap(struct address_space *mapping, sector_t block); @@ -66,7 +66,7 @@ static struct kmem_cache *befs_inode_cachep; static const struct file_operations befs_dir_operations = { .read = generic_read_dir, - .readdir = befs_readdir, + .iterate = befs_readdir, .llseek = generic_file_llseek, }; @@ -211,9 +211,9 @@ befs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) } static int -befs_readdir(struct file *filp, void *dirent, filldir_t filldir) +befs_readdir(struct file *file, struct dir_context *ctx) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; befs_data_stream *ds = &BEFS_I(inode)->i_data.ds; befs_off_t value; @@ -221,15 +221,14 @@ befs_readdir(struct file *filp, void *dirent, filldir_t filldir) size_t keysize; unsigned char d_type; char keybuf[BEFS_NAME_LEN + 1]; - char *nlsname; - int nlsnamelen; - const char *dirname = filp->f_path.dentry->d_name.name; + const char *dirname = file->f_path.dentry->d_name.name; befs_debug(sb, "---> befs_readdir() " - "name %s, inode %ld, filp->f_pos %Ld", - dirname, inode->i_ino, filp->f_pos); + "name %s, inode %ld, ctx->pos %Ld", + dirname, inode->i_ino, ctx->pos); - result = befs_btree_read(sb, ds, filp->f_pos, BEFS_NAME_LEN + 1, +more: + result = befs_btree_read(sb, ds, ctx->pos, BEFS_NAME_LEN + 1, keybuf, &keysize, &value); if (result == BEFS_ERR) { @@ -251,24 +250,29 @@ befs_readdir(struct file *filp, void *dirent, filldir_t filldir) /* Convert to NLS */ if (BEFS_SB(sb)->nls) { + char *nlsname; + int nlsnamelen; result = befs_utf2nls(sb, keybuf, keysize, &nlsname, &nlsnamelen); if (result < 0) { befs_debug(sb, "<--- befs_readdir() ERROR"); return result; } - result = filldir(dirent, nlsname, nlsnamelen, filp->f_pos, - (ino_t) value, d_type); + if (!dir_emit(ctx, nlsname, nlsnamelen, + (ino_t) value, d_type)) { + kfree(nlsname); + return 0; + } kfree(nlsname); - } else { - result = filldir(dirent, keybuf, keysize, filp->f_pos, - (ino_t) value, d_type); + if (!dir_emit(ctx, keybuf, keysize, + (ino_t) value, d_type)) + return 0; } + ctx->pos++; + goto more; - filp->f_pos++; - - befs_debug(sb, "<--- befs_readdir() filp->f_pos %Ld", filp->f_pos); + befs_debug(sb, "<--- befs_readdir() pos %Ld", ctx->pos); return 0; } diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c index 3f422f6bb5ca..a399e6d9dc74 100644 --- a/fs/bfs/dir.c +++ b/fs/bfs/dir.c @@ -26,58 +26,51 @@ static struct buffer_head *bfs_find_entry(struct inode *dir, const unsigned char *name, int namelen, struct bfs_dirent **res_dir); -static int bfs_readdir(struct file *f, void *dirent, filldir_t filldir) +static int bfs_readdir(struct file *f, struct dir_context *ctx) { struct inode *dir = file_inode(f); struct buffer_head *bh; struct bfs_dirent *de; - struct bfs_sb_info *info = BFS_SB(dir->i_sb); unsigned int offset; int block; - mutex_lock(&info->bfs_lock); - - if (f->f_pos & (BFS_DIRENT_SIZE - 1)) { + if (ctx->pos & (BFS_DIRENT_SIZE - 1)) { printf("Bad f_pos=%08lx for %s:%08lx\n", - (unsigned long)f->f_pos, + (unsigned long)ctx->pos, dir->i_sb->s_id, dir->i_ino); - mutex_unlock(&info->bfs_lock); - return -EBADF; + return -EINVAL; } - while (f->f_pos < dir->i_size) { - offset = f->f_pos & (BFS_BSIZE - 1); - block = BFS_I(dir)->i_sblock + (f->f_pos >> BFS_BSIZE_BITS); + while (ctx->pos < dir->i_size) { + offset = ctx->pos & (BFS_BSIZE - 1); + block = BFS_I(dir)->i_sblock + (ctx->pos >> BFS_BSIZE_BITS); bh = sb_bread(dir->i_sb, block); if (!bh) { - f->f_pos += BFS_BSIZE - offset; + ctx->pos += BFS_BSIZE - offset; continue; } do { de = (struct bfs_dirent *)(bh->b_data + offset); if (de->ino) { int size = strnlen(de->name, BFS_NAMELEN); - if (filldir(dirent, de->name, size, f->f_pos, + if (!dir_emit(ctx, de->name, size, le16_to_cpu(de->ino), - DT_UNKNOWN) < 0) { + DT_UNKNOWN)) { brelse(bh); - mutex_unlock(&info->bfs_lock); return 0; } } offset += BFS_DIRENT_SIZE; - f->f_pos += BFS_DIRENT_SIZE; - } while ((offset < BFS_BSIZE) && (f->f_pos < dir->i_size)); + ctx->pos += BFS_DIRENT_SIZE; + } while ((offset < BFS_BSIZE) && (ctx->pos < dir->i_size)); brelse(bh); } - - mutex_unlock(&info->bfs_lock); - return 0; + return 0; } const struct file_operations bfs_dir_operations = { .read = generic_read_dir, - .readdir = bfs_readdir, + .iterate = bfs_readdir, .fsync = generic_file_fsync, .llseek = generic_file_llseek, }; diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index f26f38ccd194..eb34438ddedb 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -1681,8 +1681,7 @@ int btrfs_should_delete_dir_index(struct list_head *del_list, * btrfs_readdir_delayed_dir_index - read dir info stored in the delayed tree * */ -int btrfs_readdir_delayed_dir_index(struct file *filp, void *dirent, - filldir_t filldir, +int btrfs_readdir_delayed_dir_index(struct dir_context *ctx, struct list_head *ins_list) { struct btrfs_dir_item *di; @@ -1704,13 +1703,13 @@ int btrfs_readdir_delayed_dir_index(struct file *filp, void *dirent, list_for_each_entry_safe(curr, next, ins_list, readdir_list) { list_del(&curr->readdir_list); - if (curr->key.offset < filp->f_pos) { + if (curr->key.offset < ctx->pos) { if (atomic_dec_and_test(&curr->refs)) kfree(curr); continue; } - filp->f_pos = curr->key.offset; + ctx->pos = curr->key.offset; di = (struct btrfs_dir_item *)curr->data; name = (char *)(di + 1); @@ -1719,7 +1718,7 @@ int btrfs_readdir_delayed_dir_index(struct file *filp, void *dirent, d_type = btrfs_filetype_table[di->type]; btrfs_disk_key_to_cpu(&location, &di->location); - over = filldir(dirent, name, name_len, curr->key.offset, + over = !dir_emit(ctx, name, name_len, location.objectid, d_type); if (atomic_dec_and_test(&curr->refs)) diff --git a/fs/btrfs/delayed-inode.h b/fs/btrfs/delayed-inode.h index 1d5c5f7abe3e..a4b38f934d14 100644 --- a/fs/btrfs/delayed-inode.h +++ b/fs/btrfs/delayed-inode.h @@ -139,8 +139,7 @@ void btrfs_put_delayed_items(struct list_head *ins_list, struct list_head *del_list); int btrfs_should_delete_dir_index(struct list_head *del_list, u64 index); -int btrfs_readdir_delayed_dir_index(struct file *filp, void *dirent, - filldir_t filldir, +int btrfs_readdir_delayed_dir_index(struct dir_context *ctx, struct list_head *ins_list); /* for init */ diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 40c7bc300075..b0292b3ead54 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2860,8 +2860,8 @@ fail_qgroup: btrfs_free_qgroup_config(fs_info); fail_trans_kthread: kthread_stop(fs_info->transaction_kthread); - del_fs_roots(fs_info); btrfs_cleanup_transaction(fs_info->tree_root); + del_fs_roots(fs_info); fail_cleaner: kthread_stop(fs_info->cleaner_kthread); @@ -3513,15 +3513,15 @@ int close_ctree(struct btrfs_root *root) percpu_counter_sum(&fs_info->delalloc_bytes)); } - free_root_pointers(fs_info, 1); - btrfs_free_block_groups(fs_info); + btrfs_stop_all_workers(fs_info); + del_fs_roots(fs_info); - iput(fs_info->btree_inode); + free_root_pointers(fs_info, 1); - btrfs_stop_all_workers(fs_info); + iput(fs_info->btree_inode); #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY if (btrfs_test_opt(root, CHECK_INTEGRITY)) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index db57e6384fbb..4f9d16b70d3d 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -5137,10 +5137,9 @@ unsigned char btrfs_filetype_table[] = { DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, DT_FIFO, DT_SOCK, DT_LNK }; -static int btrfs_real_readdir(struct file *filp, void *dirent, - filldir_t filldir) +static int btrfs_real_readdir(struct file *file, struct dir_context *ctx) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_item *item; struct btrfs_dir_item *di; @@ -5161,29 +5160,15 @@ static int btrfs_real_readdir(struct file *filp, void *dirent, char tmp_name[32]; char *name_ptr; int name_len; - int is_curr = 0; /* filp->f_pos points to the current index? */ + int is_curr = 0; /* ctx->pos points to the current index? */ /* FIXME, use a real flag for deciding about the key type */ if (root->fs_info->tree_root == root) key_type = BTRFS_DIR_ITEM_KEY; - /* special case for "." */ - if (filp->f_pos == 0) { - over = filldir(dirent, ".", 1, - filp->f_pos, btrfs_ino(inode), DT_DIR); - if (over) - return 0; - filp->f_pos = 1; - } - /* special case for .., just use the back ref */ - if (filp->f_pos == 1) { - u64 pino = parent_ino(filp->f_path.dentry); - over = filldir(dirent, "..", 2, - filp->f_pos, pino, DT_DIR); - if (over) - return 0; - filp->f_pos = 2; - } + if (!dir_emit_dots(file, ctx)) + return 0; + path = btrfs_alloc_path(); if (!path) return -ENOMEM; @@ -5197,7 +5182,7 @@ static int btrfs_real_readdir(struct file *filp, void *dirent, } btrfs_set_key_type(&key, key_type); - key.offset = filp->f_pos; + key.offset = ctx->pos; key.objectid = btrfs_ino(inode); ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); @@ -5223,14 +5208,14 @@ static int btrfs_real_readdir(struct file *filp, void *dirent, break; if (btrfs_key_type(&found_key) != key_type) break; - if (found_key.offset < filp->f_pos) + if (found_key.offset < ctx->pos) goto next; if (key_type == BTRFS_DIR_INDEX_KEY && btrfs_should_delete_dir_index(&del_list, found_key.offset)) goto next; - filp->f_pos = found_key.offset; + ctx->pos = found_key.offset; is_curr = 1; di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item); @@ -5274,9 +5259,8 @@ static int btrfs_real_readdir(struct file *filp, void *dirent, over = 0; goto skip; } - over = filldir(dirent, name_ptr, name_len, - found_key.offset, location.objectid, - d_type); + over = !dir_emit(ctx, name_ptr, name_len, + location.objectid, d_type); skip: if (name_ptr != tmp_name) @@ -5295,9 +5279,8 @@ next: if (key_type == BTRFS_DIR_INDEX_KEY) { if (is_curr) - filp->f_pos++; - ret = btrfs_readdir_delayed_dir_index(filp, dirent, filldir, - &ins_list); + ctx->pos++; + ret = btrfs_readdir_delayed_dir_index(ctx, &ins_list); if (ret) goto nopos; } @@ -5308,9 +5291,9 @@ next: * 32-bit glibc will use getdents64, but then strtol - * so the last number we can serve is this. */ - filp->f_pos = 0x7fffffff; + ctx->pos = 0x7fffffff; else - filp->f_pos++; + ctx->pos++; nopos: ret = 0; err: @@ -8013,6 +7996,9 @@ int btrfs_drop_inode(struct inode *inode) { struct btrfs_root *root = BTRFS_I(inode)->root; + if (root == NULL) + return 1; + /* the snap/subvol tree is on deleting */ if (btrfs_root_refs(&root->root_item) == 0 && root != root->fs_info->tree_root) @@ -8729,7 +8715,7 @@ static const struct inode_operations btrfs_dir_ro_inode_operations = { static const struct file_operations btrfs_dir_file_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = btrfs_real_readdir, + .iterate = btrfs_real_readdir, .unlocked_ioctl = btrfs_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = btrfs_ioctl, diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 395b82031a42..4febca4fc2de 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -4082,7 +4082,7 @@ out: return inode; } -static struct reloc_control *alloc_reloc_control(void) +static struct reloc_control *alloc_reloc_control(struct btrfs_fs_info *fs_info) { struct reloc_control *rc; @@ -4093,7 +4093,8 @@ static struct reloc_control *alloc_reloc_control(void) INIT_LIST_HEAD(&rc->reloc_roots); backref_cache_init(&rc->backref_cache); mapping_tree_init(&rc->reloc_root_tree); - extent_io_tree_init(&rc->processed_blocks, NULL); + extent_io_tree_init(&rc->processed_blocks, + fs_info->btree_inode->i_mapping); return rc; } @@ -4110,7 +4111,7 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start) int rw = 0; int err = 0; - rc = alloc_reloc_control(); + rc = alloc_reloc_control(fs_info); if (!rc) return -ENOMEM; @@ -4311,7 +4312,7 @@ int btrfs_recover_relocation(struct btrfs_root *root) if (list_empty(&reloc_roots)) goto out; - rc = alloc_reloc_control(); + rc = alloc_reloc_control(root->fs_info); if (!rc) { err = -ENOMEM; goto out; diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index f02d82b7933e..a40ceda47a32 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -111,11 +111,10 @@ static unsigned fpos_off(loff_t p) * defined IFF we hold CEPH_CAP_FILE_SHARED (which will be revoked by * the MDS if/when the directory is modified). */ -static int __dcache_readdir(struct file *filp, - void *dirent, filldir_t filldir) +static int __dcache_readdir(struct file *file, struct dir_context *ctx) { - struct ceph_file_info *fi = filp->private_data; - struct dentry *parent = filp->f_dentry; + struct ceph_file_info *fi = file->private_data; + struct dentry *parent = file->f_dentry; struct inode *dir = parent->d_inode; struct list_head *p; struct dentry *dentry, *last; @@ -126,14 +125,14 @@ static int __dcache_readdir(struct file *filp, last = fi->dentry; fi->dentry = NULL; - dout("__dcache_readdir %p at %llu (last %p)\n", dir, filp->f_pos, + dout("__dcache_readdir %p at %llu (last %p)\n", dir, ctx->pos, last); spin_lock(&parent->d_lock); /* start at beginning? */ - if (filp->f_pos == 2 || last == NULL || - filp->f_pos < ceph_dentry(last)->offset) { + if (ctx->pos == 2 || last == NULL || + ctx->pos < ceph_dentry(last)->offset) { if (list_empty(&parent->d_subdirs)) goto out_unlock; p = parent->d_subdirs.prev; @@ -157,11 +156,11 @@ more: if (!d_unhashed(dentry) && dentry->d_inode && ceph_snap(dentry->d_inode) != CEPH_SNAPDIR && ceph_ino(dentry->d_inode) != CEPH_INO_CEPH && - filp->f_pos <= di->offset) + ctx->pos <= di->offset) break; dout(" skipping %p %.*s at %llu (%llu)%s%s\n", dentry, dentry->d_name.len, dentry->d_name.name, di->offset, - filp->f_pos, d_unhashed(dentry) ? " unhashed" : "", + ctx->pos, d_unhashed(dentry) ? " unhashed" : "", !dentry->d_inode ? " null" : ""); spin_unlock(&dentry->d_lock); p = p->prev; @@ -173,29 +172,27 @@ more: spin_unlock(&dentry->d_lock); spin_unlock(&parent->d_lock); - dout(" %llu (%llu) dentry %p %.*s %p\n", di->offset, filp->f_pos, + dout(" %llu (%llu) dentry %p %.*s %p\n", di->offset, ctx->pos, dentry, dentry->d_name.len, dentry->d_name.name, dentry->d_inode); - filp->f_pos = di->offset; - err = filldir(dirent, dentry->d_name.name, - dentry->d_name.len, di->offset, + ctx->pos = di->offset; + if (!dir_emit(ctx, dentry->d_name.name, + dentry->d_name.len, ceph_translate_ino(dentry->d_sb, dentry->d_inode->i_ino), - dentry->d_inode->i_mode >> 12); - - if (last) { - if (err < 0) { + dentry->d_inode->i_mode >> 12)) { + if (last) { /* remember our position */ fi->dentry = last; fi->next_offset = di->offset; - } else { - dput(last); } + dput(dentry); + return 0; } - last = dentry; - if (err < 0) - goto out; + if (last) + dput(last); + last = dentry; - filp->f_pos++; + ctx->pos++; /* make sure a dentry wasn't dropped while we didn't have parent lock */ if (!ceph_dir_is_complete(dir)) { @@ -235,59 +232,59 @@ static int note_last_dentry(struct ceph_file_info *fi, const char *name, return 0; } -static int ceph_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int ceph_readdir(struct file *file, struct dir_context *ctx) { - struct ceph_file_info *fi = filp->private_data; - struct inode *inode = file_inode(filp); + struct ceph_file_info *fi = file->private_data; + struct inode *inode = file_inode(file); struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_fs_client *fsc = ceph_inode_to_client(inode); struct ceph_mds_client *mdsc = fsc->mdsc; - unsigned frag = fpos_frag(filp->f_pos); - int off = fpos_off(filp->f_pos); + unsigned frag = fpos_frag(ctx->pos); + int off = fpos_off(ctx->pos); int err; u32 ftype; struct ceph_mds_reply_info_parsed *rinfo; const int max_entries = fsc->mount_options->max_readdir; const int max_bytes = fsc->mount_options->max_readdir_bytes; - dout("readdir %p filp %p frag %u off %u\n", inode, filp, frag, off); + dout("readdir %p file %p frag %u off %u\n", inode, file, frag, off); if (fi->flags & CEPH_F_ATEND) return 0; /* always start with . and .. */ - if (filp->f_pos == 0) { + if (ctx->pos == 0) { /* note dir version at start of readdir so we can tell * if any dentries get dropped */ fi->dir_release_count = atomic_read(&ci->i_release_count); dout("readdir off 0 -> '.'\n"); - if (filldir(dirent, ".", 1, ceph_make_fpos(0, 0), + if (!dir_emit(ctx, ".", 1, ceph_translate_ino(inode->i_sb, inode->i_ino), - inode->i_mode >> 12) < 0) + inode->i_mode >> 12)) return 0; - filp->f_pos = 1; + ctx->pos = 1; off = 1; } - if (filp->f_pos == 1) { - ino_t ino = parent_ino(filp->f_dentry); + if (ctx->pos == 1) { + ino_t ino = parent_ino(file->f_dentry); dout("readdir off 1 -> '..'\n"); - if (filldir(dirent, "..", 2, ceph_make_fpos(0, 1), + if (!dir_emit(ctx, "..", 2, ceph_translate_ino(inode->i_sb, ino), - inode->i_mode >> 12) < 0) + inode->i_mode >> 12)) return 0; - filp->f_pos = 2; + ctx->pos = 2; off = 2; } /* can we use the dcache? */ spin_lock(&ci->i_ceph_lock); - if ((filp->f_pos == 2 || fi->dentry) && + if ((ctx->pos == 2 || fi->dentry) && !ceph_test_mount_opt(fsc, NOASYNCREADDIR) && ceph_snap(inode) != CEPH_SNAPDIR && __ceph_dir_is_complete(ci) && __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1)) { spin_unlock(&ci->i_ceph_lock); - err = __dcache_readdir(filp, dirent, filldir); + err = __dcache_readdir(file, ctx); if (err != -EAGAIN) return err; } else { @@ -327,7 +324,7 @@ more: return PTR_ERR(req); req->r_inode = inode; ihold(inode); - req->r_dentry = dget(filp->f_dentry); + req->r_dentry = dget(file->f_dentry); /* hints to request -> mds selection code */ req->r_direct_mode = USE_AUTH_MDS; req->r_direct_hash = ceph_frag_value(frag); @@ -379,15 +376,16 @@ more: rinfo = &fi->last_readdir->r_reply_info; dout("readdir frag %x num %d off %d chunkoff %d\n", frag, rinfo->dir_nr, off, fi->offset); + + ctx->pos = ceph_make_fpos(frag, off); while (off >= fi->offset && off - fi->offset < rinfo->dir_nr) { - u64 pos = ceph_make_fpos(frag, off); struct ceph_mds_reply_inode *in = rinfo->dir_in[off - fi->offset].in; struct ceph_vino vino; ino_t ino; dout("readdir off %d (%d/%d) -> %lld '%.*s' %p\n", - off, off - fi->offset, rinfo->dir_nr, pos, + off, off - fi->offset, rinfo->dir_nr, ctx->pos, rinfo->dir_dname_len[off - fi->offset], rinfo->dir_dname[off - fi->offset], in); BUG_ON(!in); @@ -395,16 +393,15 @@ more: vino.ino = le64_to_cpu(in->ino); vino.snap = le64_to_cpu(in->snapid); ino = ceph_vino_to_ino(vino); - if (filldir(dirent, + if (!dir_emit(ctx, rinfo->dir_dname[off - fi->offset], rinfo->dir_dname_len[off - fi->offset], - pos, - ceph_translate_ino(inode->i_sb, ino), ftype) < 0) { + ceph_translate_ino(inode->i_sb, ino), ftype)) { dout("filldir stopping us...\n"); return 0; } off++; - filp->f_pos = pos + 1; + ctx->pos++; } if (fi->last_name) { @@ -417,7 +414,7 @@ more: if (!ceph_frag_is_rightmost(frag)) { frag = ceph_frag_next(frag); off = 0; - filp->f_pos = ceph_make_fpos(frag, off); + ctx->pos = ceph_make_fpos(frag, off); dout("readdir next frag is %x\n", frag); goto more; } @@ -432,11 +429,11 @@ more: if (atomic_read(&ci->i_release_count) == fi->dir_release_count) { dout(" marking %p complete\n", inode); __ceph_dir_set_complete(ci, fi->dir_release_count); - ci->i_max_offset = filp->f_pos; + ci->i_max_offset = ctx->pos; } spin_unlock(&ci->i_ceph_lock); - dout("readdir %p filp %p done.\n", inode, filp); + dout("readdir %p file %p done.\n", inode, file); return 0; } @@ -1268,7 +1265,7 @@ unsigned ceph_dentry_hash(struct inode *dir, struct dentry *dn) const struct file_operations ceph_dir_fops = { .read = ceph_read_dir, - .readdir = ceph_readdir, + .iterate = ceph_readdir, .llseek = ceph_dir_llseek, .open = ceph_open, .release = ceph_release, diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c index 202dd3d68be0..ebbf680378e2 100644 --- a/fs/ceph/locks.c +++ b/fs/ceph/locks.c @@ -191,27 +191,23 @@ void ceph_count_locks(struct inode *inode, int *fcntl_count, int *flock_count) } /** - * Encode the flock and fcntl locks for the given inode into the pagelist. - * Format is: #fcntl locks, sequential fcntl locks, #flock locks, - * sequential flock locks. - * Must be called with lock_flocks() already held. - * If we encounter more of a specific lock type than expected, - * we return the value 1. + * Encode the flock and fcntl locks for the given inode into the ceph_filelock + * array. Must be called with lock_flocks() already held. + * If we encounter more of a specific lock type than expected, return -ENOSPC. */ -int ceph_encode_locks(struct inode *inode, struct ceph_pagelist *pagelist, - int num_fcntl_locks, int num_flock_locks) +int ceph_encode_locks_to_buffer(struct inode *inode, + struct ceph_filelock *flocks, + int num_fcntl_locks, int num_flock_locks) { struct file_lock *lock; - struct ceph_filelock cephlock; int err = 0; int seen_fcntl = 0; int seen_flock = 0; + int l = 0; dout("encoding %d flock and %d fcntl locks", num_flock_locks, num_fcntl_locks); - err = ceph_pagelist_append(pagelist, &num_fcntl_locks, sizeof(u32)); - if (err) - goto fail; + for (lock = inode->i_flock; lock != NULL; lock = lock->fl_next) { if (lock->fl_flags & FL_POSIX) { ++seen_fcntl; @@ -219,19 +215,12 @@ int ceph_encode_locks(struct inode *inode, struct ceph_pagelist *pagelist, err = -ENOSPC; goto fail; } - err = lock_to_ceph_filelock(lock, &cephlock); + err = lock_to_ceph_filelock(lock, &flocks[l]); if (err) goto fail; - err = ceph_pagelist_append(pagelist, &cephlock, - sizeof(struct ceph_filelock)); + ++l; } - if (err) - goto fail; } - - err = ceph_pagelist_append(pagelist, &num_flock_locks, sizeof(u32)); - if (err) - goto fail; for (lock = inode->i_flock; lock != NULL; lock = lock->fl_next) { if (lock->fl_flags & FL_FLOCK) { ++seen_flock; @@ -239,19 +228,51 @@ int ceph_encode_locks(struct inode *inode, struct ceph_pagelist *pagelist, err = -ENOSPC; goto fail; } - err = lock_to_ceph_filelock(lock, &cephlock); + err = lock_to_ceph_filelock(lock, &flocks[l]); if (err) goto fail; - err = ceph_pagelist_append(pagelist, &cephlock, - sizeof(struct ceph_filelock)); + ++l; } - if (err) - goto fail; } fail: return err; } +/** + * Copy the encoded flock and fcntl locks into the pagelist. + * Format is: #fcntl locks, sequential fcntl locks, #flock locks, + * sequential flock locks. + * Returns zero on success. + */ +int ceph_locks_to_pagelist(struct ceph_filelock *flocks, + struct ceph_pagelist *pagelist, + int num_fcntl_locks, int num_flock_locks) +{ + int err = 0; + __le32 nlocks; + + nlocks = cpu_to_le32(num_fcntl_locks); + err = ceph_pagelist_append(pagelist, &nlocks, sizeof(nlocks)); + if (err) + goto out_fail; + + err = ceph_pagelist_append(pagelist, flocks, + num_fcntl_locks * sizeof(*flocks)); + if (err) + goto out_fail; + + nlocks = cpu_to_le32(num_flock_locks); + err = ceph_pagelist_append(pagelist, &nlocks, sizeof(nlocks)); + if (err) + goto out_fail; + + err = ceph_pagelist_append(pagelist, + &flocks[num_fcntl_locks], + num_flock_locks * sizeof(*flocks)); +out_fail: + return err; +} + /* * Given a pointer to a lock, convert it to a ceph filelock */ diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 4f22671a5bd4..4d2920304be8 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -2478,39 +2478,44 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap, if (recon_state->flock) { int num_fcntl_locks, num_flock_locks; - struct ceph_pagelist_cursor trunc_point; - - ceph_pagelist_set_cursor(pagelist, &trunc_point); - do { - lock_flocks(); - ceph_count_locks(inode, &num_fcntl_locks, - &num_flock_locks); - rec.v2.flock_len = (2*sizeof(u32) + - (num_fcntl_locks+num_flock_locks) * - sizeof(struct ceph_filelock)); - unlock_flocks(); - - /* pre-alloc pagelist */ - ceph_pagelist_truncate(pagelist, &trunc_point); - err = ceph_pagelist_append(pagelist, &rec, reclen); - if (!err) - err = ceph_pagelist_reserve(pagelist, - rec.v2.flock_len); - - /* encode locks */ - if (!err) { - lock_flocks(); - err = ceph_encode_locks(inode, - pagelist, - num_fcntl_locks, - num_flock_locks); - unlock_flocks(); - } - } while (err == -ENOSPC); + struct ceph_filelock *flocks; + +encode_again: + lock_flocks(); + ceph_count_locks(inode, &num_fcntl_locks, &num_flock_locks); + unlock_flocks(); + flocks = kmalloc((num_fcntl_locks+num_flock_locks) * + sizeof(struct ceph_filelock), GFP_NOFS); + if (!flocks) { + err = -ENOMEM; + goto out_free; + } + lock_flocks(); + err = ceph_encode_locks_to_buffer(inode, flocks, + num_fcntl_locks, + num_flock_locks); + unlock_flocks(); + if (err) { + kfree(flocks); + if (err == -ENOSPC) + goto encode_again; + goto out_free; + } + /* + * number of encoded locks is stable, so copy to pagelist + */ + rec.v2.flock_len = cpu_to_le32(2*sizeof(u32) + + (num_fcntl_locks+num_flock_locks) * + sizeof(struct ceph_filelock)); + err = ceph_pagelist_append(pagelist, &rec, reclen); + if (!err) + err = ceph_locks_to_pagelist(flocks, pagelist, + num_fcntl_locks, + num_flock_locks); + kfree(flocks); } else { err = ceph_pagelist_append(pagelist, &rec, reclen); } - out_free: kfree(path); out_dput: diff --git a/fs/ceph/super.h b/fs/ceph/super.h index 8696be2ff679..7ccfdb4aea2e 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -822,8 +822,13 @@ extern const struct export_operations ceph_export_ops; extern int ceph_lock(struct file *file, int cmd, struct file_lock *fl); extern int ceph_flock(struct file *file, int cmd, struct file_lock *fl); extern void ceph_count_locks(struct inode *inode, int *p_num, int *f_num); -extern int ceph_encode_locks(struct inode *i, struct ceph_pagelist *p, - int p_locks, int f_locks); +extern int ceph_encode_locks_to_buffer(struct inode *inode, + struct ceph_filelock *flocks, + int num_fcntl_locks, + int num_flock_locks); +extern int ceph_locks_to_pagelist(struct ceph_filelock *flocks, + struct ceph_pagelist *pagelist, + int num_fcntl_locks, int num_flock_locks); extern int lock_to_ceph_filelock(struct file_lock *fl, struct ceph_filelock *c); /* debugfs.c */ diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c index 8e33ec65847b..58df174deb10 100644 --- a/fs/cifs/cifs_dfs_ref.c +++ b/fs/cifs/cifs_dfs_ref.c @@ -18,6 +18,7 @@ #include <linux/slab.h> #include <linux/vfs.h> #include <linux/fs.h> +#include <linux/inet.h> #include "cifsglob.h" #include "cifsproto.h" #include "cifsfs.h" @@ -48,58 +49,74 @@ void cifs_dfs_release_automount_timer(void) } /** - * cifs_get_share_name - extracts share name from UNC - * @node_name: pointer to UNC string + * cifs_build_devname - build a devicename from a UNC and optional prepath + * @nodename: pointer to UNC string + * @prepath: pointer to prefixpath (or NULL if there isn't one) * - * Extracts sharename form full UNC. - * i.e. strips from UNC trailing path that is not part of share - * name and fixup missing '\' in the beginning of DFS node refferal - * if necessary. - * Returns pointer to share name on success or ERR_PTR on error. - * Caller is responsible for freeing returned string. + * Build a new cifs devicename after chasing a DFS referral. Allocate a buffer + * big enough to hold the final thing. Copy the UNC from the nodename, and + * concatenate the prepath onto the end of it if there is one. + * + * Returns pointer to the built string, or a ERR_PTR. Caller is responsible + * for freeing the returned string. */ -static char *cifs_get_share_name(const char *node_name) +static char * +cifs_build_devname(char *nodename, const char *prepath) { - int len; - char *UNC; - char *pSep; - - len = strlen(node_name); - UNC = kmalloc(len+2 /*for term null and additional \ if it's missed */, - GFP_KERNEL); - if (!UNC) - return ERR_PTR(-ENOMEM); + size_t pplen; + size_t unclen; + char *dev; + char *pos; + + /* skip over any preceding delimiters */ + nodename += strspn(nodename, "\\"); + if (!*nodename) + return ERR_PTR(-EINVAL); - /* get share name and server name */ - if (node_name[1] != '\\') { - UNC[0] = '\\'; - strncpy(UNC+1, node_name, len); - len++; - UNC[len] = 0; - } else { - strncpy(UNC, node_name, len); - UNC[len] = 0; - } + /* get length of UNC and set pos to last char */ + unclen = strlen(nodename); + pos = nodename + unclen - 1; - /* find server name end */ - pSep = memchr(UNC+2, '\\', len-2); - if (!pSep) { - cifs_dbg(VFS, "%s: no server name end in node name: %s\n", - __func__, node_name); - kfree(UNC); - return ERR_PTR(-EINVAL); + /* trim off any trailing delimiters */ + while (*pos == '\\') { + --pos; + --unclen; } - /* find sharename end */ - pSep++; - pSep = memchr(UNC+(pSep-UNC), '\\', len-(pSep-UNC)); - if (pSep) { - /* trim path up to sharename end - * now we have share name in UNC */ - *pSep = 0; + /* allocate a buffer: + * +2 for preceding "//" + * +1 for delimiter between UNC and prepath + * +1 for trailing NULL + */ + pplen = prepath ? strlen(prepath) : 0; + dev = kmalloc(2 + unclen + 1 + pplen + 1, GFP_KERNEL); + if (!dev) + return ERR_PTR(-ENOMEM); + + pos = dev; + /* add the initial "//" */ + *pos = '/'; + ++pos; + *pos = '/'; + ++pos; + + /* copy in the UNC portion from referral */ + memcpy(pos, nodename, unclen); + pos += unclen; + + /* copy the prefixpath remainder (if there is one) */ + if (pplen) { + *pos = '/'; + ++pos; + memcpy(pos, prepath, pplen); + pos += pplen; } - return UNC; + /* NULL terminator */ + *pos = '\0'; + + convert_delimiter(dev, '/'); + return dev; } @@ -123,6 +140,7 @@ char *cifs_compose_mount_options(const char *sb_mountdata, { int rc; char *mountdata = NULL; + const char *prepath = NULL; int md_len; char *tkn_e; char *srvIP = NULL; @@ -132,7 +150,10 @@ char *cifs_compose_mount_options(const char *sb_mountdata, if (sb_mountdata == NULL) return ERR_PTR(-EINVAL); - *devname = cifs_get_share_name(ref->node_name); + if (strlen(fullpath) - ref->path_consumed) + prepath = fullpath + ref->path_consumed; + + *devname = cifs_build_devname(ref->node_name, prepath); if (IS_ERR(*devname)) { rc = PTR_ERR(*devname); *devname = NULL; @@ -146,12 +167,14 @@ char *cifs_compose_mount_options(const char *sb_mountdata, goto compose_mount_options_err; } - /* md_len = strlen(...) + 12 for 'sep+prefixpath=' - * assuming that we have 'unc=' and 'ip=' in - * the original sb_mountdata + /* + * In most cases, we'll be building a shorter string than the original, + * but we do have to assume that the address in the ip= option may be + * much longer than the original. Add the max length of an address + * string to the length of the original string to allow for worst case. */ - md_len = strlen(sb_mountdata) + rc + strlen(ref->node_name) + 12; - mountdata = kzalloc(md_len+1, GFP_KERNEL); + md_len = strlen(sb_mountdata) + INET6_ADDRSTRLEN; + mountdata = kzalloc(md_len + 1, GFP_KERNEL); if (mountdata == NULL) { rc = -ENOMEM; goto compose_mount_options_err; @@ -195,26 +218,6 @@ char *cifs_compose_mount_options(const char *sb_mountdata, strncat(mountdata, &sep, 1); strcat(mountdata, "ip="); strcat(mountdata, srvIP); - strncat(mountdata, &sep, 1); - strcat(mountdata, "unc="); - strcat(mountdata, *devname); - - /* find & copy prefixpath */ - tkn_e = strchr(ref->node_name + 2, '\\'); - if (tkn_e == NULL) { - /* invalid unc, missing share name*/ - rc = -EINVAL; - goto compose_mount_options_err; - } - - tkn_e = strchr(tkn_e + 1, '\\'); - if (tkn_e || (strlen(fullpath) - ref->path_consumed)) { - strncat(mountdata, &sep, 1); - strcat(mountdata, "prefixpath="); - if (tkn_e) - strcat(mountdata, tkn_e + 1); - strcat(mountdata, fullpath + ref->path_consumed); - } /*cifs_dbg(FYI, "%s: parent mountdata: %s\n", __func__, sb_mountdata);*/ /*cifs_dbg(FYI, "%s: submount mountdata: %s\n", __func__, mountdata );*/ diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 72e4efee1389..540c1ccfcdb2 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -372,9 +372,6 @@ cifs_show_options(struct seq_file *s, struct dentry *root) cifs_show_security(s, tcon->ses->server); cifs_show_cache_flavor(s, cifs_sb); - seq_printf(s, ",unc="); - seq_escape(s, tcon->treeName, " \t\n\\"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER) seq_printf(s, ",multiuser"); else if (tcon->ses->user_name) @@ -971,7 +968,7 @@ const struct file_operations cifs_file_direct_nobrl_ops = { }; const struct file_operations cifs_dir_ops = { - .readdir = cifs_readdir, + .iterate = cifs_readdir, .release = cifs_closedir, .read = generic_read_dir, .unlocked_ioctl = cifs_ioctl, diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 0e32c3446ce9..d05b3028e3b9 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -101,7 +101,7 @@ extern int cifs_file_mmap(struct file * , struct vm_area_struct *); extern int cifs_file_strict_mmap(struct file * , struct vm_area_struct *); extern const struct file_operations cifs_dir_ops; extern int cifs_dir_open(struct inode *inode, struct file *file); -extern int cifs_readdir(struct file *file, void *direntry, filldir_t filldir); +extern int cifs_readdir(struct file *file, struct dir_context *ctx); /* Functions related to dir entries */ extern const struct dentry_operations cifs_dentry_ops; diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 99eeaa17ee00..e3bc39bb9d12 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -1061,6 +1061,7 @@ static int cifs_parse_security_flavors(char *value, #endif case Opt_sec_none: vol->nullauth = 1; + vol->secFlg |= CIFSSEC_MAY_NTLM; break; default: cifs_dbg(VFS, "bad security option: %s\n", value); @@ -1257,14 +1258,18 @@ cifs_parse_mount_options(const char *mountdata, const char *devname, vol->backupuid_specified = false; /* no backup intent for a user */ vol->backupgid_specified = false; /* no backup intent for a group */ - /* - * For now, we ignore -EINVAL errors under the assumption that the - * unc= and prefixpath= options will be usable. - */ - if (cifs_parse_devname(devname, vol) == -ENOMEM) { - printk(KERN_ERR "CIFS: Unable to allocate memory to parse " - "device string.\n"); - goto out_nomem; + switch (cifs_parse_devname(devname, vol)) { + case 0: + break; + case -ENOMEM: + cifs_dbg(VFS, "Unable to allocate memory for devname.\n"); + goto cifs_parse_mount_err; + case -EINVAL: + cifs_dbg(VFS, "Malformed UNC in devname.\n"); + goto cifs_parse_mount_err; + default: + cifs_dbg(VFS, "Unknown error parsing devname.\n"); + goto cifs_parse_mount_err; } while ((data = strsep(&options, separator)) != NULL) { @@ -1826,7 +1831,7 @@ cifs_parse_mount_options(const char *mountdata, const char *devname, } #endif if (!vol->UNC) { - cifs_dbg(VFS, "CIFS mount error: No usable UNC path provided in device string or in unc= option!\n"); + cifs_dbg(VFS, "CIFS mount error: No usable UNC path provided in device string!\n"); goto cifs_parse_mount_err; } @@ -3274,8 +3279,8 @@ build_unc_path_to_root(const struct smb_vol *vol, pos = full_path + unc_len; if (pplen) { - *pos++ = CIFS_DIR_SEP(cifs_sb); - strncpy(pos, vol->prepath, pplen); + *pos = CIFS_DIR_SEP(cifs_sb); + strncpy(pos + 1, vol->prepath, pplen); pos += pplen; } diff --git a/fs/cifs/dns_resolve.c b/fs/cifs/dns_resolve.c index e7512e497611..7ede7306599f 100644 --- a/fs/cifs/dns_resolve.c +++ b/fs/cifs/dns_resolve.c @@ -34,7 +34,7 @@ /** * dns_resolve_server_name_to_ip - Resolve UNC server name to ip address. - * @unc: UNC path specifying the server + * @unc: UNC path specifying the server (with '/' as delimiter) * @ip_addr: Where to return the IP address. * * The IP address will be returned in string form, and the caller is @@ -64,7 +64,7 @@ dns_resolve_server_name_to_ip(const char *unc, char **ip_addr) hostname = unc + 2; /* Search for server name delimiter */ - sep = memchr(hostname, '\\', len); + sep = memchr(hostname, '/', len); if (sep) len = sep - hostname; else diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index fc3025199cb3..20efd81266c6 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -171,7 +171,8 @@ cifs_fattr_to_inode(struct inode *inode, struct cifs_fattr *fattr) if (fattr->cf_flags & CIFS_FATTR_DFS_REFERRAL) inode->i_flags |= S_AUTOMOUNT; - cifs_set_ops(inode); + if (inode->i_state & I_NEW) + cifs_set_ops(inode); } void diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 770d5a9781c1..f1213799de1a 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -537,14 +537,14 @@ static int cifs_save_resume_key(const char *current_entry, * every entry (do not increment for . or .. entry). */ static int -find_cifs_entry(const unsigned int xid, struct cifs_tcon *tcon, +find_cifs_entry(const unsigned int xid, struct cifs_tcon *tcon, loff_t pos, struct file *file, char **current_entry, int *num_to_ret) { __u16 search_flags; int rc = 0; int pos_in_buf = 0; loff_t first_entry_in_buffer; - loff_t index_to_find = file->f_pos; + loff_t index_to_find = pos; struct cifsFileInfo *cfile = file->private_data; struct cifs_sb_info *cifs_sb = CIFS_SB(file->f_path.dentry->d_sb); struct TCP_Server_Info *server = tcon->ses->server; @@ -659,8 +659,9 @@ find_cifs_entry(const unsigned int xid, struct cifs_tcon *tcon, return rc; } -static int cifs_filldir(char *find_entry, struct file *file, filldir_t filldir, - void *dirent, char *scratch_buf, unsigned int max_len) +static int cifs_filldir(char *find_entry, struct file *file, + struct dir_context *ctx, + char *scratch_buf, unsigned int max_len) { struct cifsFileInfo *file_info = file->private_data; struct super_block *sb = file->f_path.dentry->d_sb; @@ -740,13 +741,11 @@ static int cifs_filldir(char *find_entry, struct file *file, filldir_t filldir, cifs_prime_dcache(file->f_dentry, &name, &fattr); ino = cifs_uniqueid_to_ino_t(fattr.cf_uniqueid); - rc = filldir(dirent, name.name, name.len, file->f_pos, ino, - fattr.cf_dtype); - return rc; + return !dir_emit(ctx, name.name, name.len, ino, fattr.cf_dtype); } -int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) +int cifs_readdir(struct file *file, struct dir_context *ctx) { int rc = 0; unsigned int xid; @@ -772,103 +771,86 @@ int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) goto rddir2_exit; } - switch ((int) file->f_pos) { - case 0: - if (filldir(direntry, ".", 1, file->f_pos, - file_inode(file)->i_ino, DT_DIR) < 0) { - cifs_dbg(VFS, "Filldir for current dir failed\n"); - rc = -ENOMEM; - break; - } - file->f_pos++; - case 1: - if (filldir(direntry, "..", 2, file->f_pos, - parent_ino(file->f_path.dentry), DT_DIR) < 0) { - cifs_dbg(VFS, "Filldir for parent dir failed\n"); - rc = -ENOMEM; - break; - } - file->f_pos++; - default: - /* 1) If search is active, - is in current search buffer? - if it before then restart search - if after then keep searching till find it */ - - if (file->private_data == NULL) { - rc = -EINVAL; - free_xid(xid); - return rc; - } - cifsFile = file->private_data; - if (cifsFile->srch_inf.endOfSearch) { - if (cifsFile->srch_inf.emptyDir) { - cifs_dbg(FYI, "End of search, empty dir\n"); - rc = 0; - break; - } - } /* else { - cifsFile->invalidHandle = true; - tcon->ses->server->close(xid, tcon, &cifsFile->fid); - } */ + if (!dir_emit_dots(file, ctx)) + goto rddir2_exit; - tcon = tlink_tcon(cifsFile->tlink); - rc = find_cifs_entry(xid, tcon, file, ¤t_entry, - &num_to_fill); - if (rc) { - cifs_dbg(FYI, "fce error %d\n", rc); - goto rddir2_exit; - } else if (current_entry != NULL) { - cifs_dbg(FYI, "entry %lld found\n", file->f_pos); - } else { - cifs_dbg(FYI, "could not find entry\n"); + /* 1) If search is active, + is in current search buffer? + if it before then restart search + if after then keep searching till find it */ + + if (file->private_data == NULL) { + rc = -EINVAL; + goto rddir2_exit; + } + cifsFile = file->private_data; + if (cifsFile->srch_inf.endOfSearch) { + if (cifsFile->srch_inf.emptyDir) { + cifs_dbg(FYI, "End of search, empty dir\n"); + rc = 0; goto rddir2_exit; } - cifs_dbg(FYI, "loop through %d times filling dir for net buf %p\n", - num_to_fill, cifsFile->srch_inf.ntwrk_buf_start); - max_len = tcon->ses->server->ops->calc_smb_size( - cifsFile->srch_inf.ntwrk_buf_start); - end_of_smb = cifsFile->srch_inf.ntwrk_buf_start + max_len; - - tmp_buf = kmalloc(UNICODE_NAME_MAX, GFP_KERNEL); - if (tmp_buf == NULL) { - rc = -ENOMEM; + } /* else { + cifsFile->invalidHandle = true; + tcon->ses->server->close(xid, tcon, &cifsFile->fid); + } */ + + tcon = tlink_tcon(cifsFile->tlink); + rc = find_cifs_entry(xid, tcon, ctx->pos, file, ¤t_entry, + &num_to_fill); + if (rc) { + cifs_dbg(FYI, "fce error %d\n", rc); + goto rddir2_exit; + } else if (current_entry != NULL) { + cifs_dbg(FYI, "entry %lld found\n", ctx->pos); + } else { + cifs_dbg(FYI, "could not find entry\n"); + goto rddir2_exit; + } + cifs_dbg(FYI, "loop through %d times filling dir for net buf %p\n", + num_to_fill, cifsFile->srch_inf.ntwrk_buf_start); + max_len = tcon->ses->server->ops->calc_smb_size( + cifsFile->srch_inf.ntwrk_buf_start); + end_of_smb = cifsFile->srch_inf.ntwrk_buf_start + max_len; + + tmp_buf = kmalloc(UNICODE_NAME_MAX, GFP_KERNEL); + if (tmp_buf == NULL) { + rc = -ENOMEM; + goto rddir2_exit; + } + + for (i = 0; i < num_to_fill; i++) { + if (current_entry == NULL) { + /* evaluate whether this case is an error */ + cifs_dbg(VFS, "past SMB end, num to fill %d i %d\n", + num_to_fill, i); break; } - - for (i = 0; (i < num_to_fill) && (rc == 0); i++) { - if (current_entry == NULL) { - /* evaluate whether this case is an error */ - cifs_dbg(VFS, "past SMB end, num to fill %d i %d\n", - num_to_fill, i); - break; - } - /* - * if buggy server returns . and .. late do we want to - * check for that here? - */ - rc = cifs_filldir(current_entry, file, filldir, - direntry, tmp_buf, max_len); - if (rc == -EOVERFLOW) { + /* + * if buggy server returns . and .. late do we want to + * check for that here? + */ + rc = cifs_filldir(current_entry, file, ctx, + tmp_buf, max_len); + if (rc) { + if (rc > 0) rc = 0; - break; - } - - file->f_pos++; - if (file->f_pos == - cifsFile->srch_inf.index_of_last_entry) { - cifs_dbg(FYI, "last entry in buf at pos %lld %s\n", - file->f_pos, tmp_buf); - cifs_save_resume_key(current_entry, cifsFile); - break; - } else - current_entry = - nxt_dir_entry(current_entry, end_of_smb, - cifsFile->srch_inf.info_level); + break; } - kfree(tmp_buf); - break; - } /* end switch */ + + ctx->pos++; + if (ctx->pos == + cifsFile->srch_inf.index_of_last_entry) { + cifs_dbg(FYI, "last entry in buf at pos %lld %s\n", + ctx->pos, tmp_buf); + cifs_save_resume_key(current_entry, cifsFile); + break; + } else + current_entry = + nxt_dir_entry(current_entry, end_of_smb, + cifsFile->srch_inf.info_level); + } + kfree(tmp_buf); rddir2_exit: free_xid(xid); diff --git a/fs/coda/dir.c b/fs/coda/dir.c index b7d3a05c062c..87e0ee9f4465 100644 --- a/fs/coda/dir.c +++ b/fs/coda/dir.c @@ -43,15 +43,14 @@ static int coda_rename(struct inode *old_inode, struct dentry *old_dentry, struct inode *new_inode, struct dentry *new_dentry); /* dir file-ops */ -static int coda_readdir(struct file *file, void *buf, filldir_t filldir); +static int coda_readdir(struct file *file, struct dir_context *ctx); /* dentry ops */ static int coda_dentry_revalidate(struct dentry *de, unsigned int flags); static int coda_dentry_delete(const struct dentry *); /* support routines */ -static int coda_venus_readdir(struct file *coda_file, void *buf, - filldir_t filldir); +static int coda_venus_readdir(struct file *, struct dir_context *); /* same as fs/bad_inode.c */ static int coda_return_EIO(void) @@ -85,7 +84,7 @@ const struct inode_operations coda_dir_inode_operations = const struct file_operations coda_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = coda_readdir, + .iterate = coda_readdir, .open = coda_open, .release = coda_release, .fsync = coda_fsync, @@ -378,7 +377,7 @@ static int coda_rename(struct inode *old_dir, struct dentry *old_dentry, /* file operations for directories */ -static int coda_readdir(struct file *coda_file, void *buf, filldir_t filldir) +static int coda_readdir(struct file *coda_file, struct dir_context *ctx) { struct coda_file_info *cfi; struct file *host_file; @@ -391,30 +390,19 @@ static int coda_readdir(struct file *coda_file, void *buf, filldir_t filldir) if (!host_file->f_op) return -ENOTDIR; - if (host_file->f_op->readdir) - { - /* potemkin case: we were handed a directory inode. - * We can't use vfs_readdir because we have to keep the file - * position in sync between the coda_file and the host_file. - * and as such we need grab the inode mutex. */ + if (host_file->f_op->iterate) { struct inode *host_inode = file_inode(host_file); - mutex_lock(&host_inode->i_mutex); - host_file->f_pos = coda_file->f_pos; - ret = -ENOENT; if (!IS_DEADDIR(host_inode)) { - ret = host_file->f_op->readdir(host_file, buf, filldir); + ret = host_file->f_op->iterate(host_file, ctx); file_accessed(host_file); } - - coda_file->f_pos = host_file->f_pos; mutex_unlock(&host_inode->i_mutex); + return ret; } - else /* Venus: we must read Venus dirents from a file */ - ret = coda_venus_readdir(coda_file, buf, filldir); - - return ret; + /* Venus: we must read Venus dirents from a file */ + return coda_venus_readdir(coda_file, ctx); } static inline unsigned int CDT2DT(unsigned char cdt) @@ -437,10 +425,8 @@ static inline unsigned int CDT2DT(unsigned char cdt) } /* support routines */ -static int coda_venus_readdir(struct file *coda_file, void *buf, - filldir_t filldir) +static int coda_venus_readdir(struct file *coda_file, struct dir_context *ctx) { - int result = 0; /* # of entries returned */ struct coda_file_info *cfi; struct coda_inode_info *cii; struct file *host_file; @@ -462,23 +448,12 @@ static int coda_venus_readdir(struct file *coda_file, void *buf, vdir = kmalloc(sizeof(*vdir), GFP_KERNEL); if (!vdir) return -ENOMEM; - if (coda_file->f_pos == 0) { - ret = filldir(buf, ".", 1, 0, de->d_inode->i_ino, DT_DIR); - if (ret < 0) - goto out; - result++; - coda_file->f_pos++; - } - if (coda_file->f_pos == 1) { - ret = filldir(buf, "..", 2, 1, parent_ino(de), DT_DIR); - if (ret < 0) - goto out; - result++; - coda_file->f_pos++; - } + if (!dir_emit_dots(coda_file, ctx)) + goto out; + while (1) { /* read entries from the directory file */ - ret = kernel_read(host_file, coda_file->f_pos - 2, (char *)vdir, + ret = kernel_read(host_file, ctx->pos - 2, (char *)vdir, sizeof(*vdir)); if (ret < 0) { printk(KERN_ERR "coda readdir: read dir %s failed %d\n", @@ -507,7 +482,7 @@ static int coda_venus_readdir(struct file *coda_file, void *buf, /* Make sure we skip '.' and '..', we already got those */ if (name.name[0] == '.' && (name.len == 1 || - (vdir->d_name[1] == '.' && name.len == 2))) + (name.name[1] == '.' && name.len == 2))) vdir->d_fileno = name.len = 0; /* skip null entries */ @@ -520,19 +495,16 @@ static int coda_venus_readdir(struct file *coda_file, void *buf, if (!ino) ino = vdir->d_fileno; type = CDT2DT(vdir->d_type); - ret = filldir(buf, name.name, name.len, - coda_file->f_pos, ino, type); - /* failure means no space for filling in this round */ - if (ret < 0) break; - result++; + if (!dir_emit(ctx, name.name, name.len, ino, type)) + break; } /* we'll always have progress because d_reclen is unsigned and * we've already established it is non-zero. */ - coda_file->f_pos += vdir->d_reclen; + ctx->pos += vdir->d_reclen; } out: kfree(vdir); - return result ? result : ret; + return 0; } /* called when a cache lookup succeeds */ diff --git a/fs/compat.c b/fs/compat.c index fc3b55dce184..6af20de2c1a3 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -832,6 +832,7 @@ struct compat_old_linux_dirent { }; struct compat_readdir_callback { + struct dir_context ctx; struct compat_old_linux_dirent __user *dirent; int result; }; @@ -873,15 +874,15 @@ asmlinkage long compat_sys_old_readdir(unsigned int fd, { int error; struct fd f = fdget(fd); - struct compat_readdir_callback buf; + struct compat_readdir_callback buf = { + .ctx.actor = compat_fillonedir, + .dirent = dirent + }; if (!f.file) return -EBADF; - buf.result = 0; - buf.dirent = dirent; - - error = vfs_readdir(f.file, compat_fillonedir, &buf); + error = iterate_dir(f.file, &buf.ctx); if (buf.result) error = buf.result; @@ -897,6 +898,7 @@ struct compat_linux_dirent { }; struct compat_getdents_callback { + struct dir_context ctx; struct compat_linux_dirent __user *current_dir; struct compat_linux_dirent __user *previous; int count; @@ -951,7 +953,11 @@ asmlinkage long compat_sys_getdents(unsigned int fd, { struct fd f; struct compat_linux_dirent __user * lastdirent; - struct compat_getdents_callback buf; + struct compat_getdents_callback buf = { + .ctx.actor = compat_filldir, + .current_dir = dirent, + .count = count + }; int error; if (!access_ok(VERIFY_WRITE, dirent, count)) @@ -961,17 +967,12 @@ asmlinkage long compat_sys_getdents(unsigned int fd, if (!f.file) return -EBADF; - buf.current_dir = dirent; - buf.previous = NULL; - buf.count = count; - buf.error = 0; - - error = vfs_readdir(f.file, compat_filldir, &buf); + error = iterate_dir(f.file, &buf.ctx); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - if (put_user(f.file->f_pos, &lastdirent->d_off)) + if (put_user(buf.ctx.pos, &lastdirent->d_off)) error = -EFAULT; else error = count - buf.count; @@ -983,6 +984,7 @@ asmlinkage long compat_sys_getdents(unsigned int fd, #ifndef __ARCH_OMIT_COMPAT_SYS_GETDENTS64 struct compat_getdents_callback64 { + struct dir_context ctx; struct linux_dirent64 __user *current_dir; struct linux_dirent64 __user *previous; int count; @@ -1036,7 +1038,11 @@ asmlinkage long compat_sys_getdents64(unsigned int fd, { struct fd f; struct linux_dirent64 __user * lastdirent; - struct compat_getdents_callback64 buf; + struct compat_getdents_callback64 buf = { + .ctx.actor = compat_filldir64, + .current_dir = dirent, + .count = count + }; int error; if (!access_ok(VERIFY_WRITE, dirent, count)) @@ -1046,17 +1052,12 @@ asmlinkage long compat_sys_getdents64(unsigned int fd, if (!f.file) return -EBADF; - buf.current_dir = dirent; - buf.previous = NULL; - buf.count = count; - buf.error = 0; - - error = vfs_readdir(f.file, compat_filldir64, &buf); + error = iterate_dir(f.file, &buf.ctx); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - typeof(lastdirent->d_off) d_off = f.file->f_pos; + typeof(lastdirent->d_off) d_off = buf.ctx.pos; if (__put_user_unaligned(d_off, &lastdirent->d_off)) error = -EFAULT; else diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 996cdc5abb85..5d19acfa7c6c 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -66,7 +66,6 @@ #include <linux/gigaset_dev.h> #ifdef CONFIG_BLOCK -#include <linux/loop.h> #include <linux/cdrom.h> #include <linux/fd.h> #include <scsi/scsi.h> @@ -954,8 +953,6 @@ COMPATIBLE_IOCTL(MTIOCTOP) /* Socket level stuff */ COMPATIBLE_IOCTL(FIOQSIZE) #ifdef CONFIG_BLOCK -/* loop */ -IGNORE_IOCTL(LOOP_CLR_FD) /* md calls this on random blockdevs */ IGNORE_IOCTL(RAID_VERSION) /* qemu/qemu-img might call these two on plain files for probing */ diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c index 7aabc6ad4e9b..64e5323cbbb0 100644 --- a/fs/configfs/dir.c +++ b/fs/configfs/dir.c @@ -1532,84 +1532,66 @@ static inline unsigned char dt_type(struct configfs_dirent *sd) return (sd->s_mode >> 12) & 15; } -static int configfs_readdir(struct file * filp, void * dirent, filldir_t filldir) +static int configfs_readdir(struct file *file, struct dir_context *ctx) { - struct dentry *dentry = filp->f_path.dentry; + struct dentry *dentry = file->f_path.dentry; struct super_block *sb = dentry->d_sb; struct configfs_dirent * parent_sd = dentry->d_fsdata; - struct configfs_dirent *cursor = filp->private_data; + struct configfs_dirent *cursor = file->private_data; struct list_head *p, *q = &cursor->s_sibling; ino_t ino = 0; - int i = filp->f_pos; - switch (i) { - case 0: - ino = dentry->d_inode->i_ino; - if (filldir(dirent, ".", 1, i, ino, DT_DIR) < 0) - break; - filp->f_pos++; - i++; - /* fallthrough */ - case 1: - ino = parent_ino(dentry); - if (filldir(dirent, "..", 2, i, ino, DT_DIR) < 0) - break; - filp->f_pos++; - i++; - /* fallthrough */ - default: - if (filp->f_pos == 2) { - spin_lock(&configfs_dirent_lock); - list_move(q, &parent_sd->s_children); - spin_unlock(&configfs_dirent_lock); - } - for (p=q->next; p!= &parent_sd->s_children; p=p->next) { - struct configfs_dirent *next; - const char * name; - int len; - struct inode *inode = NULL; + if (!dir_emit_dots(file, ctx)) + return 0; + if (ctx->pos == 2) { + spin_lock(&configfs_dirent_lock); + list_move(q, &parent_sd->s_children); + spin_unlock(&configfs_dirent_lock); + } + for (p = q->next; p != &parent_sd->s_children; p = p->next) { + struct configfs_dirent *next; + const char *name; + int len; + struct inode *inode = NULL; + + next = list_entry(p, struct configfs_dirent, s_sibling); + if (!next->s_element) + continue; - next = list_entry(p, struct configfs_dirent, - s_sibling); - if (!next->s_element) - continue; - - name = configfs_get_name(next); - len = strlen(name); - - /* - * We'll have a dentry and an inode for - * PINNED items and for open attribute - * files. We lock here to prevent a race - * with configfs_d_iput() clearing - * s_dentry before calling iput(). - * - * Why do we go to the trouble? If - * someone has an attribute file open, - * the inode number should match until - * they close it. Beyond that, we don't - * care. - */ - spin_lock(&configfs_dirent_lock); - dentry = next->s_dentry; - if (dentry) - inode = dentry->d_inode; - if (inode) - ino = inode->i_ino; - spin_unlock(&configfs_dirent_lock); - if (!inode) - ino = iunique(sb, 2); + name = configfs_get_name(next); + len = strlen(name); + + /* + * We'll have a dentry and an inode for + * PINNED items and for open attribute + * files. We lock here to prevent a race + * with configfs_d_iput() clearing + * s_dentry before calling iput(). + * + * Why do we go to the trouble? If + * someone has an attribute file open, + * the inode number should match until + * they close it. Beyond that, we don't + * care. + */ + spin_lock(&configfs_dirent_lock); + dentry = next->s_dentry; + if (dentry) + inode = dentry->d_inode; + if (inode) + ino = inode->i_ino; + spin_unlock(&configfs_dirent_lock); + if (!inode) + ino = iunique(sb, 2); - if (filldir(dirent, name, len, filp->f_pos, ino, - dt_type(next)) < 0) - return 0; + if (!dir_emit(ctx, name, len, ino, dt_type(next))) + return 0; - spin_lock(&configfs_dirent_lock); - list_move(q, p); - spin_unlock(&configfs_dirent_lock); - p = q; - filp->f_pos++; - } + spin_lock(&configfs_dirent_lock); + list_move(q, p); + spin_unlock(&configfs_dirent_lock); + p = q; + ctx->pos++; } return 0; } @@ -1661,7 +1643,7 @@ const struct file_operations configfs_dir_operations = { .release = configfs_dir_close, .llseek = configfs_dir_lseek, .read = generic_read_dir, - .readdir = configfs_readdir, + .iterate = configfs_readdir, }; int configfs_register_subsystem(struct configfs_subsystem *subsys) diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index 35b1c7bd18b7..e501ac3a49ff 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -349,18 +349,17 @@ static int cramfs_statfs(struct dentry *dentry, struct kstatfs *buf) /* * Read a cramfs directory entry. */ -static int cramfs_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int cramfs_readdir(struct file *file, struct dir_context *ctx) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; char *buf; unsigned int offset; - int copied; /* Offset within the thing. */ - offset = filp->f_pos; - if (offset >= inode->i_size) + if (ctx->pos >= inode->i_size) return 0; + offset = ctx->pos; /* Directory entries are always 4-byte aligned */ if (offset & 3) return -EINVAL; @@ -369,14 +368,13 @@ static int cramfs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (!buf) return -ENOMEM; - copied = 0; while (offset < inode->i_size) { struct cramfs_inode *de; unsigned long nextoffset; char *name; ino_t ino; umode_t mode; - int namelen, error; + int namelen; mutex_lock(&read_mutex); de = cramfs_read(sb, OFFSET(inode) + offset, sizeof(*de)+CRAMFS_MAXPATHLEN); @@ -402,13 +400,10 @@ static int cramfs_readdir(struct file *filp, void *dirent, filldir_t filldir) break; namelen--; } - error = filldir(dirent, buf, namelen, offset, ino, mode >> 12); - if (error) + if (!dir_emit(ctx, buf, namelen, ino, mode >> 12)) break; - offset = nextoffset; - filp->f_pos = offset; - copied++; + ctx->pos = offset = nextoffset; } kfree(buf); return 0; @@ -547,7 +542,7 @@ static const struct address_space_operations cramfs_aops = { static const struct file_operations cramfs_directory_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = cramfs_readdir, + .iterate = cramfs_readdir, }; static const struct inode_operations cramfs_dir_inode_operations = { diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index 201f0a0d6b0a..9aa05e08060b 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -68,9 +68,9 @@ static ssize_t ecryptfs_read_update_atime(struct kiocb *iocb, } struct ecryptfs_getdents_callback { - void *dirent; + struct dir_context ctx; + struct dir_context *caller; struct dentry *dentry; - filldir_t filldir; int filldir_called; int entries_written; }; @@ -96,9 +96,10 @@ ecryptfs_filldir(void *dirent, const char *lower_name, int lower_namelen, rc); goto out; } - rc = buf->filldir(buf->dirent, name, name_size, offset, ino, d_type); + buf->caller->pos = buf->ctx.pos; + rc = !dir_emit(buf->caller, name, name_size, ino, d_type); kfree(name); - if (rc >= 0) + if (!rc) buf->entries_written++; out: return rc; @@ -107,27 +108,23 @@ out: /** * ecryptfs_readdir * @file: The eCryptfs directory file - * @dirent: Directory entry handle - * @filldir: The filldir callback function + * @ctx: The actor to feed the entries to */ -static int ecryptfs_readdir(struct file *file, void *dirent, filldir_t filldir) +static int ecryptfs_readdir(struct file *file, struct dir_context *ctx) { int rc; struct file *lower_file; struct inode *inode; - struct ecryptfs_getdents_callback buf; - + struct ecryptfs_getdents_callback buf = { + .ctx.actor = ecryptfs_filldir, + .caller = ctx, + .dentry = file->f_path.dentry + }; lower_file = ecryptfs_file_to_lower(file); - lower_file->f_pos = file->f_pos; + lower_file->f_pos = ctx->pos; inode = file_inode(file); - memset(&buf, 0, sizeof(buf)); - buf.dirent = dirent; - buf.dentry = file->f_path.dentry; - buf.filldir = filldir; - buf.filldir_called = 0; - buf.entries_written = 0; - rc = vfs_readdir(lower_file, ecryptfs_filldir, (void *)&buf); - file->f_pos = lower_file->f_pos; + rc = iterate_dir(lower_file, &buf.ctx); + ctx->pos = buf.ctx.pos; if (rc < 0) goto out; if (buf.filldir_called && !buf.entries_written) @@ -295,6 +292,12 @@ static int ecryptfs_release(struct inode *inode, struct file *file) static int ecryptfs_fsync(struct file *file, loff_t start, loff_t end, int datasync) { + int rc; + + rc = filemap_write_and_wait(file->f_mapping); + if (rc) + return rc; + return vfs_fsync(ecryptfs_file_to_lower(file), datasync); } @@ -338,7 +341,7 @@ ecryptfs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) #endif const struct file_operations ecryptfs_dir_fops = { - .readdir = ecryptfs_readdir, + .iterate = ecryptfs_readdir, .read = generic_read_dir, .unlocked_ioctl = ecryptfs_unlocked_ioctl, #ifdef CONFIG_COMPAT @@ -359,7 +362,7 @@ const struct file_operations ecryptfs_main_fops = { .aio_read = ecryptfs_read_update_atime, .write = do_sync_write, .aio_write = generic_file_aio_write, - .readdir = ecryptfs_readdir, + .iterate = ecryptfs_readdir, .unlocked_ioctl = ecryptfs_unlocked_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ecryptfs_compat_ioctl, diff --git a/fs/efivarfs/file.c b/fs/efivarfs/file.c index bfb531564319..8dd524f32284 100644 --- a/fs/efivarfs/file.c +++ b/fs/efivarfs/file.c @@ -44,8 +44,11 @@ static ssize_t efivarfs_file_write(struct file *file, bytes = efivar_entry_set_get_size(var, attributes, &datasize, data, &set); - if (!set && bytes) + if (!set && bytes) { + if (bytes == -ENOENT) + bytes = -EIO; goto out; + } if (bytes == -ENOENT) { drop_nlink(inode); @@ -76,7 +79,14 @@ static ssize_t efivarfs_file_read(struct file *file, char __user *userbuf, int err; err = efivar_entry_size(var, &datasize); - if (err) + + /* + * efivarfs represents uncommitted variables with + * zero-length files. Reading them should return EOF. + */ + if (err == -ENOENT) + return 0; + else if (err) return err; data = kmalloc(datasize + sizeof(attributes), GFP_KERNEL); diff --git a/fs/efs/dir.c b/fs/efs/dir.c index 055a9e9ca747..b72307ccdf7a 100644 --- a/fs/efs/dir.c +++ b/fs/efs/dir.c @@ -7,40 +7,38 @@ #include <linux/buffer_head.h> #include "efs.h" -static int efs_readdir(struct file *, void *, filldir_t); +static int efs_readdir(struct file *, struct dir_context *); const struct file_operations efs_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = efs_readdir, + .iterate = efs_readdir, }; const struct inode_operations efs_dir_inode_operations = { .lookup = efs_lookup, }; -static int efs_readdir(struct file *filp, void *dirent, filldir_t filldir) { - struct inode *inode = file_inode(filp); - struct buffer_head *bh; - - struct efs_dir *dirblock; - struct efs_dentry *dirslot; - efs_ino_t inodenum; +static int efs_readdir(struct file *file, struct dir_context *ctx) +{ + struct inode *inode = file_inode(file); efs_block_t block; - int slot, namelen; - char *nameptr; + int slot; if (inode->i_size & (EFS_DIRBSIZE-1)) printk(KERN_WARNING "EFS: WARNING: readdir(): directory size not a multiple of EFS_DIRBSIZE\n"); /* work out where this entry can be found */ - block = filp->f_pos >> EFS_DIRBSIZE_BITS; + block = ctx->pos >> EFS_DIRBSIZE_BITS; /* each block contains at most 256 slots */ - slot = filp->f_pos & 0xff; + slot = ctx->pos & 0xff; /* look at all blocks */ while (block < inode->i_blocks) { + struct efs_dir *dirblock; + struct buffer_head *bh; + /* read the dir block */ bh = sb_bread(inode->i_sb, efs_bmap(inode, block)); @@ -57,11 +55,14 @@ static int efs_readdir(struct file *filp, void *dirent, filldir_t filldir) { break; } - while (slot < dirblock->slots) { - if (dirblock->space[slot] == 0) { - slot++; + for (; slot < dirblock->slots; slot++) { + struct efs_dentry *dirslot; + efs_ino_t inodenum; + const char *nameptr; + int namelen; + + if (dirblock->space[slot] == 0) continue; - } dirslot = (struct efs_dentry *) (((char *) bh->b_data) + EFS_SLOTAT(dirblock, slot)); @@ -72,39 +73,29 @@ static int efs_readdir(struct file *filp, void *dirent, filldir_t filldir) { #ifdef DEBUG printk(KERN_DEBUG "EFS: readdir(): block %d slot %d/%d: inode %u, name \"%s\", namelen %u\n", block, slot, dirblock->slots-1, inodenum, nameptr, namelen); #endif - if (namelen > 0) { - /* found the next entry */ - filp->f_pos = (block << EFS_DIRBSIZE_BITS) | slot; - - /* copy filename and data in dirslot */ - filldir(dirent, nameptr, namelen, filp->f_pos, inodenum, DT_UNKNOWN); - - /* sanity check */ - if (nameptr - (char *) dirblock + namelen > EFS_DIRBSIZE) { - printk(KERN_WARNING "EFS: directory entry %d exceeds directory block\n", slot); - slot++; - continue; - } - - /* store position of next slot */ - if (++slot == dirblock->slots) { - slot = 0; - block++; - } + if (!namelen) + continue; + /* found the next entry */ + ctx->pos = (block << EFS_DIRBSIZE_BITS) | slot; + + /* sanity check */ + if (nameptr - (char *) dirblock + namelen > EFS_DIRBSIZE) { + printk(KERN_WARNING "EFS: directory entry %d exceeds directory block\n", slot); + continue; + } + + /* copy filename and data in dirslot */ + if (!dir_emit(ctx, nameptr, namelen, inodenum, DT_UNKNOWN)) { brelse(bh); - filp->f_pos = (block << EFS_DIRBSIZE_BITS) | slot; - goto out; + return 0; } - slot++; } brelse(bh); slot = 0; block++; } - - filp->f_pos = (block << EFS_DIRBSIZE_BITS) | slot; -out: + ctx->pos = (block << EFS_DIRBSIZE_BITS) | slot; return 0; } diff --git a/fs/exec.c b/fs/exec.c index 643019585574..ffd7a813ad3d 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1135,13 +1135,6 @@ void setup_new_exec(struct linux_binprm * bprm) set_dumpable(current->mm, suid_dumpable); } - /* - * Flush performance counters when crossing a - * security domain: - */ - if (!get_dumpable(current->mm)) - perf_event_exit_task(current); - /* An exec changes our domain. We are no longer part of the thread group */ @@ -1205,6 +1198,15 @@ void install_exec_creds(struct linux_binprm *bprm) commit_creds(bprm->cred); bprm->cred = NULL; + + /* + * Disable monitoring for regular users + * when executing setuid binaries. Must + * wait until new credentials are committed + * by commit_creds() above + */ + if (get_dumpable(current->mm) != SUID_DUMP_USER) + perf_event_exit_task(current); /* * cred_guard_mutex must be held at least to this point to prevent * ptrace_attach() from altering our determination of the task's diff --git a/fs/exofs/dir.c b/fs/exofs/dir.c index 46375896cfc0..49f51ab4caac 100644 --- a/fs/exofs/dir.c +++ b/fs/exofs/dir.c @@ -239,22 +239,19 @@ void exofs_set_de_type(struct exofs_dir_entry *de, struct inode *inode) } static int -exofs_readdir(struct file *filp, void *dirent, filldir_t filldir) +exofs_readdir(struct file *file, struct dir_context *ctx) { - loff_t pos = filp->f_pos; - struct inode *inode = file_inode(filp); + loff_t pos = ctx->pos; + struct inode *inode = file_inode(file); unsigned int offset = pos & ~PAGE_CACHE_MASK; unsigned long n = pos >> PAGE_CACHE_SHIFT; unsigned long npages = dir_pages(inode); unsigned chunk_mask = ~(exofs_chunk_size(inode)-1); - unsigned char *types = NULL; - int need_revalidate = (filp->f_version != inode->i_version); + int need_revalidate = (file->f_version != inode->i_version); if (pos > inode->i_size - EXOFS_DIR_REC_LEN(1)) return 0; - types = exofs_filetype_table; - for ( ; n < npages; n++, offset = 0) { char *kaddr, *limit; struct exofs_dir_entry *de; @@ -263,7 +260,7 @@ exofs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (IS_ERR(page)) { EXOFS_ERR("ERROR: bad page in directory(0x%lx)\n", inode->i_ino); - filp->f_pos += PAGE_CACHE_SIZE - offset; + ctx->pos += PAGE_CACHE_SIZE - offset; return PTR_ERR(page); } kaddr = page_address(page); @@ -271,9 +268,9 @@ exofs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (offset) { offset = exofs_validate_entry(kaddr, offset, chunk_mask); - filp->f_pos = (n<<PAGE_CACHE_SHIFT) + offset; + ctx->pos = (n<<PAGE_CACHE_SHIFT) + offset; } - filp->f_version = inode->i_version; + file->f_version = inode->i_version; need_revalidate = 0; } de = (struct exofs_dir_entry *)(kaddr + offset); @@ -288,27 +285,24 @@ exofs_readdir(struct file *filp, void *dirent, filldir_t filldir) return -EIO; } if (de->inode_no) { - int over; - unsigned char d_type = DT_UNKNOWN; + unsigned char t; - if (types && de->file_type < EXOFS_FT_MAX) - d_type = types[de->file_type]; + if (de->file_type < EXOFS_FT_MAX) + t = exofs_filetype_table[de->file_type]; + else + t = DT_UNKNOWN; - offset = (char *)de - kaddr; - over = filldir(dirent, de->name, de->name_len, - (n<<PAGE_CACHE_SHIFT) | offset, + if (!dir_emit(ctx, de->name, de->name_len, le64_to_cpu(de->inode_no), - d_type); - if (over) { + t)) { exofs_put_page(page); return 0; } } - filp->f_pos += le16_to_cpu(de->rec_len); + ctx->pos += le16_to_cpu(de->rec_len); } exofs_put_page(page); } - return 0; } @@ -669,5 +663,5 @@ not_empty: const struct file_operations exofs_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = exofs_readdir, + .iterate = exofs_readdir, }; diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c index 262fc9940982..293bc2e47a73 100644 --- a/fs/exportfs/expfs.c +++ b/fs/exportfs/expfs.c @@ -212,6 +212,7 @@ reconnect_path(struct vfsmount *mnt, struct dentry *target_dir, char *nbuf) } struct getdents_callback { + struct dir_context ctx; char *name; /* name that was found. It already points to a buffer NAME_MAX+1 is size */ unsigned long ino; /* the inum we are looking for */ @@ -254,7 +255,11 @@ static int get_name(const struct path *path, char *name, struct dentry *child) struct inode *dir = path->dentry->d_inode; int error; struct file *file; - struct getdents_callback buffer; + struct getdents_callback buffer = { + .ctx.actor = filldir_one, + .name = name, + .ino = child->d_inode->i_ino + }; error = -ENOTDIR; if (!dir || !S_ISDIR(dir->i_mode)) @@ -271,17 +276,14 @@ static int get_name(const struct path *path, char *name, struct dentry *child) goto out; error = -EINVAL; - if (!file->f_op->readdir) + if (!file->f_op->iterate) goto out_close; - buffer.name = name; - buffer.ino = child->d_inode->i_ino; - buffer.found = 0; buffer.sequence = 0; while (1) { int old_seq = buffer.sequence; - error = vfs_readdir(file, filldir_one, &buffer); + error = iterate_dir(file, &buffer.ctx); if (buffer.found) { error = 0; break; diff --git a/fs/ext2/dir.c b/fs/ext2/dir.c index 4237722bfd27..6e1d4ab09d72 100644 --- a/fs/ext2/dir.c +++ b/fs/ext2/dir.c @@ -287,17 +287,17 @@ static inline void ext2_set_de_type(ext2_dirent *de, struct inode *inode) } static int -ext2_readdir (struct file * filp, void * dirent, filldir_t filldir) +ext2_readdir(struct file *file, struct dir_context *ctx) { - loff_t pos = filp->f_pos; - struct inode *inode = file_inode(filp); + loff_t pos = ctx->pos; + struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; unsigned int offset = pos & ~PAGE_CACHE_MASK; unsigned long n = pos >> PAGE_CACHE_SHIFT; unsigned long npages = dir_pages(inode); unsigned chunk_mask = ~(ext2_chunk_size(inode)-1); unsigned char *types = NULL; - int need_revalidate = filp->f_version != inode->i_version; + int need_revalidate = file->f_version != inode->i_version; if (pos > inode->i_size - EXT2_DIR_REC_LEN(1)) return 0; @@ -314,16 +314,16 @@ ext2_readdir (struct file * filp, void * dirent, filldir_t filldir) ext2_error(sb, __func__, "bad page in #%lu", inode->i_ino); - filp->f_pos += PAGE_CACHE_SIZE - offset; + ctx->pos += PAGE_CACHE_SIZE - offset; return PTR_ERR(page); } kaddr = page_address(page); if (unlikely(need_revalidate)) { if (offset) { offset = ext2_validate_entry(kaddr, offset, chunk_mask); - filp->f_pos = (n<<PAGE_CACHE_SHIFT) + offset; + ctx->pos = (n<<PAGE_CACHE_SHIFT) + offset; } - filp->f_version = inode->i_version; + file->f_version = inode->i_version; need_revalidate = 0; } de = (ext2_dirent *)(kaddr+offset); @@ -336,22 +336,19 @@ ext2_readdir (struct file * filp, void * dirent, filldir_t filldir) return -EIO; } if (de->inode) { - int over; unsigned char d_type = DT_UNKNOWN; if (types && de->file_type < EXT2_FT_MAX) d_type = types[de->file_type]; - offset = (char *)de - kaddr; - over = filldir(dirent, de->name, de->name_len, - (n<<PAGE_CACHE_SHIFT) | offset, - le32_to_cpu(de->inode), d_type); - if (over) { + if (!dir_emit(ctx, de->name, de->name_len, + le32_to_cpu(de->inode), + d_type)) { ext2_put_page(page); return 0; } } - filp->f_pos += ext2_rec_len_from_disk(de->rec_len); + ctx->pos += ext2_rec_len_from_disk(de->rec_len); } ext2_put_page(page); } @@ -724,7 +721,7 @@ not_empty: const struct file_operations ext2_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = ext2_readdir, + .iterate = ext2_readdir, .unlocked_ioctl = ext2_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ext2_compat_ioctl, diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c index 87eccbbca255..f522425aaa24 100644 --- a/fs/ext3/dir.c +++ b/fs/ext3/dir.c @@ -28,8 +28,7 @@ static unsigned char ext3_filetype_table[] = { DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, DT_FIFO, DT_SOCK, DT_LNK }; -static int ext3_dx_readdir(struct file * filp, - void * dirent, filldir_t filldir); +static int ext3_dx_readdir(struct file *, struct dir_context *); static unsigned char get_dtype(struct super_block *sb, int filetype) { @@ -91,36 +90,30 @@ int ext3_check_dir_entry (const char * function, struct inode * dir, return error_msg == NULL ? 1 : 0; } -static int ext3_readdir(struct file * filp, - void * dirent, filldir_t filldir) +static int ext3_readdir(struct file *file, struct dir_context *ctx) { - int error = 0; unsigned long offset; - int i, stored; + int i; struct ext3_dir_entry_2 *de; int err; - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; - int ret = 0; int dir_has_error = 0; if (is_dx_dir(inode)) { - err = ext3_dx_readdir(filp, dirent, filldir); - if (err != ERR_BAD_DX_DIR) { - ret = err; - goto out; - } + err = ext3_dx_readdir(file, ctx); + if (err != ERR_BAD_DX_DIR) + return err; /* * We don't set the inode dirty flag since it's not * critical that it get flushed back to the disk. */ - EXT3_I(file_inode(filp))->i_flags &= ~EXT3_INDEX_FL; + EXT3_I(inode)->i_flags &= ~EXT3_INDEX_FL; } - stored = 0; - offset = filp->f_pos & (sb->s_blocksize - 1); + offset = ctx->pos & (sb->s_blocksize - 1); - while (!error && !stored && filp->f_pos < inode->i_size) { - unsigned long blk = filp->f_pos >> EXT3_BLOCK_SIZE_BITS(sb); + while (ctx->pos < inode->i_size) { + unsigned long blk = ctx->pos >> EXT3_BLOCK_SIZE_BITS(sb); struct buffer_head map_bh; struct buffer_head *bh = NULL; @@ -129,12 +122,12 @@ static int ext3_readdir(struct file * filp, if (err > 0) { pgoff_t index = map_bh.b_blocknr >> (PAGE_CACHE_SHIFT - inode->i_blkbits); - if (!ra_has_index(&filp->f_ra, index)) + if (!ra_has_index(&file->f_ra, index)) page_cache_sync_readahead( sb->s_bdev->bd_inode->i_mapping, - &filp->f_ra, filp, + &file->f_ra, file, index, 1); - filp->f_ra.prev_pos = (loff_t)index << PAGE_CACHE_SHIFT; + file->f_ra.prev_pos = (loff_t)index << PAGE_CACHE_SHIFT; bh = ext3_bread(NULL, inode, blk, 0, &err); } @@ -146,22 +139,21 @@ static int ext3_readdir(struct file * filp, if (!dir_has_error) { ext3_error(sb, __func__, "directory #%lu " "contains a hole at offset %lld", - inode->i_ino, filp->f_pos); + inode->i_ino, ctx->pos); dir_has_error = 1; } /* corrupt size? Maybe no more blocks to read */ - if (filp->f_pos > inode->i_blocks << 9) + if (ctx->pos > inode->i_blocks << 9) break; - filp->f_pos += sb->s_blocksize - offset; + ctx->pos += sb->s_blocksize - offset; continue; } -revalidate: /* If the dir block has changed since the last call to * readdir(2), then we might be pointing to an invalid * dirent right now. Scan from the start of the block * to make sure. */ - if (filp->f_version != inode->i_version) { + if (offset && file->f_version != inode->i_version) { for (i = 0; i < sb->s_blocksize && i < offset; ) { de = (struct ext3_dir_entry_2 *) (bh->b_data + i); @@ -177,53 +169,40 @@ revalidate: i += ext3_rec_len_from_disk(de->rec_len); } offset = i; - filp->f_pos = (filp->f_pos & ~(sb->s_blocksize - 1)) + ctx->pos = (ctx->pos & ~(sb->s_blocksize - 1)) | offset; - filp->f_version = inode->i_version; + file->f_version = inode->i_version; } - while (!error && filp->f_pos < inode->i_size + while (ctx->pos < inode->i_size && offset < sb->s_blocksize) { de = (struct ext3_dir_entry_2 *) (bh->b_data + offset); if (!ext3_check_dir_entry ("ext3_readdir", inode, de, bh, offset)) { - /* On error, skip the f_pos to the + /* On error, skip the to the next block. */ - filp->f_pos = (filp->f_pos | + ctx->pos = (ctx->pos | (sb->s_blocksize - 1)) + 1; - brelse (bh); - ret = stored; - goto out; + break; } offset += ext3_rec_len_from_disk(de->rec_len); if (le32_to_cpu(de->inode)) { - /* We might block in the next section - * if the data destination is - * currently swapped out. So, use a - * version stamp to detect whether or - * not the directory has been modified - * during the copy operation. - */ - u64 version = filp->f_version; - - error = filldir(dirent, de->name, - de->name_len, - filp->f_pos, - le32_to_cpu(de->inode), - get_dtype(sb, de->file_type)); - if (error) - break; - if (version != filp->f_version) - goto revalidate; - stored ++; + if (!dir_emit(ctx, de->name, de->name_len, + le32_to_cpu(de->inode), + get_dtype(sb, de->file_type))) { + brelse(bh); + return 0; + } } - filp->f_pos += ext3_rec_len_from_disk(de->rec_len); + ctx->pos += ext3_rec_len_from_disk(de->rec_len); } offset = 0; brelse (bh); + if (ctx->pos < inode->i_size) + if (!dir_relax(inode)) + return 0; } -out: - return ret; + return 0; } static inline int is_32bit_api(void) @@ -452,62 +431,54 @@ int ext3_htree_store_dirent(struct file *dir_file, __u32 hash, * for all entres on the fname linked list. (Normally there is only * one entry on the linked list, unless there are 62 bit hash collisions.) */ -static int call_filldir(struct file * filp, void * dirent, - filldir_t filldir, struct fname *fname) +static bool call_filldir(struct file *file, struct dir_context *ctx, + struct fname *fname) { - struct dir_private_info *info = filp->private_data; - loff_t curr_pos; - struct inode *inode = file_inode(filp); - struct super_block * sb; - int error; - - sb = inode->i_sb; + struct dir_private_info *info = file->private_data; + struct inode *inode = file_inode(file); + struct super_block *sb = inode->i_sb; if (!fname) { printk("call_filldir: called with null fname?!?\n"); - return 0; + return true; } - curr_pos = hash2pos(filp, fname->hash, fname->minor_hash); + ctx->pos = hash2pos(file, fname->hash, fname->minor_hash); while (fname) { - error = filldir(dirent, fname->name, - fname->name_len, curr_pos, + if (!dir_emit(ctx, fname->name, fname->name_len, fname->inode, - get_dtype(sb, fname->file_type)); - if (error) { - filp->f_pos = curr_pos; + get_dtype(sb, fname->file_type))) { info->extra_fname = fname; - return error; + return false; } fname = fname->next; } - return 0; + return true; } -static int ext3_dx_readdir(struct file * filp, - void * dirent, filldir_t filldir) +static int ext3_dx_readdir(struct file *file, struct dir_context *ctx) { - struct dir_private_info *info = filp->private_data; - struct inode *inode = file_inode(filp); + struct dir_private_info *info = file->private_data; + struct inode *inode = file_inode(file); struct fname *fname; int ret; if (!info) { - info = ext3_htree_create_dir_info(filp, filp->f_pos); + info = ext3_htree_create_dir_info(file, ctx->pos); if (!info) return -ENOMEM; - filp->private_data = info; + file->private_data = info; } - if (filp->f_pos == ext3_get_htree_eof(filp)) + if (ctx->pos == ext3_get_htree_eof(file)) return 0; /* EOF */ /* Some one has messed with f_pos; reset the world */ - if (info->last_pos != filp->f_pos) { + if (info->last_pos != ctx->pos) { free_rb_tree_fname(&info->root); info->curr_node = NULL; info->extra_fname = NULL; - info->curr_hash = pos2maj_hash(filp, filp->f_pos); - info->curr_minor_hash = pos2min_hash(filp, filp->f_pos); + info->curr_hash = pos2maj_hash(file, ctx->pos); + info->curr_minor_hash = pos2min_hash(file, ctx->pos); } /* @@ -515,7 +486,7 @@ static int ext3_dx_readdir(struct file * filp, * chain, return them first. */ if (info->extra_fname) { - if (call_filldir(filp, dirent, filldir, info->extra_fname)) + if (!call_filldir(file, ctx, info->extra_fname)) goto finished; info->extra_fname = NULL; goto next_node; @@ -529,17 +500,17 @@ static int ext3_dx_readdir(struct file * filp, * cached entries. */ if ((!info->curr_node) || - (filp->f_version != inode->i_version)) { + (file->f_version != inode->i_version)) { info->curr_node = NULL; free_rb_tree_fname(&info->root); - filp->f_version = inode->i_version; - ret = ext3_htree_fill_tree(filp, info->curr_hash, + file->f_version = inode->i_version; + ret = ext3_htree_fill_tree(file, info->curr_hash, info->curr_minor_hash, &info->next_hash); if (ret < 0) return ret; if (ret == 0) { - filp->f_pos = ext3_get_htree_eof(filp); + ctx->pos = ext3_get_htree_eof(file); break; } info->curr_node = rb_first(&info->root); @@ -548,7 +519,7 @@ static int ext3_dx_readdir(struct file * filp, fname = rb_entry(info->curr_node, struct fname, rb_hash); info->curr_hash = fname->hash; info->curr_minor_hash = fname->minor_hash; - if (call_filldir(filp, dirent, filldir, fname)) + if (!call_filldir(file, ctx, fname)) break; next_node: info->curr_node = rb_next(info->curr_node); @@ -559,7 +530,7 @@ static int ext3_dx_readdir(struct file * filp, info->curr_minor_hash = fname->minor_hash; } else { if (info->next_hash == ~0) { - filp->f_pos = ext3_get_htree_eof(filp); + ctx->pos = ext3_get_htree_eof(file); break; } info->curr_hash = info->next_hash; @@ -567,7 +538,7 @@ static int ext3_dx_readdir(struct file * filp, } } finished: - info->last_pos = filp->f_pos; + info->last_pos = ctx->pos; return 0; } @@ -582,7 +553,7 @@ static int ext3_release_dir (struct inode * inode, struct file * filp) const struct file_operations ext3_dir_operations = { .llseek = ext3_dir_llseek, .read = generic_read_dir, - .readdir = ext3_readdir, + .iterate = ext3_readdir, .unlocked_ioctl = ext3_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ext3_compat_ioctl, diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index f8d56e4254e0..3c7d288ae94c 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -29,8 +29,7 @@ #include "ext4.h" #include "xattr.h" -static int ext4_dx_readdir(struct file *filp, - void *dirent, filldir_t filldir); +static int ext4_dx_readdir(struct file *, struct dir_context *); /** * Check if the given dir-inode refers to an htree-indexed directory @@ -103,60 +102,56 @@ int __ext4_check_dir_entry(const char *function, unsigned int line, return 1; } -static int ext4_readdir(struct file *filp, - void *dirent, filldir_t filldir) +static int ext4_readdir(struct file *file, struct dir_context *ctx) { - int error = 0; unsigned int offset; int i, stored; struct ext4_dir_entry_2 *de; int err; - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; - int ret = 0; int dir_has_error = 0; if (is_dx_dir(inode)) { - err = ext4_dx_readdir(filp, dirent, filldir); + err = ext4_dx_readdir(file, ctx); if (err != ERR_BAD_DX_DIR) { - ret = err; - goto out; + return err; } /* * We don't set the inode dirty flag since it's not * critical that it get flushed back to the disk. */ - ext4_clear_inode_flag(file_inode(filp), + ext4_clear_inode_flag(file_inode(file), EXT4_INODE_INDEX); } if (ext4_has_inline_data(inode)) { int has_inline_data = 1; - ret = ext4_read_inline_dir(filp, dirent, filldir, + int ret = ext4_read_inline_dir(file, ctx, &has_inline_data); if (has_inline_data) return ret; } stored = 0; - offset = filp->f_pos & (sb->s_blocksize - 1); + offset = ctx->pos & (sb->s_blocksize - 1); - while (!error && !stored && filp->f_pos < inode->i_size) { + while (ctx->pos < inode->i_size) { struct ext4_map_blocks map; struct buffer_head *bh = NULL; - map.m_lblk = filp->f_pos >> EXT4_BLOCK_SIZE_BITS(sb); + map.m_lblk = ctx->pos >> EXT4_BLOCK_SIZE_BITS(sb); map.m_len = 1; err = ext4_map_blocks(NULL, inode, &map, 0); if (err > 0) { pgoff_t index = map.m_pblk >> (PAGE_CACHE_SHIFT - inode->i_blkbits); - if (!ra_has_index(&filp->f_ra, index)) + if (!ra_has_index(&file->f_ra, index)) page_cache_sync_readahead( sb->s_bdev->bd_inode->i_mapping, - &filp->f_ra, filp, + &file->f_ra, file, index, 1); - filp->f_ra.prev_pos = (loff_t)index << PAGE_CACHE_SHIFT; + file->f_ra.prev_pos = (loff_t)index << PAGE_CACHE_SHIFT; bh = ext4_bread(NULL, inode, map.m_lblk, 0, &err); } @@ -166,16 +161,16 @@ static int ext4_readdir(struct file *filp, */ if (!bh) { if (!dir_has_error) { - EXT4_ERROR_FILE(filp, 0, + EXT4_ERROR_FILE(file, 0, "directory contains a " "hole at offset %llu", - (unsigned long long) filp->f_pos); + (unsigned long long) ctx->pos); dir_has_error = 1; } /* corrupt size? Maybe no more blocks to read */ - if (filp->f_pos > inode->i_blocks << 9) + if (ctx->pos > inode->i_blocks << 9) break; - filp->f_pos += sb->s_blocksize - offset; + ctx->pos += sb->s_blocksize - offset; continue; } @@ -183,21 +178,20 @@ static int ext4_readdir(struct file *filp, if (!buffer_verified(bh) && !ext4_dirent_csum_verify(inode, (struct ext4_dir_entry *)bh->b_data)) { - EXT4_ERROR_FILE(filp, 0, "directory fails checksum " + EXT4_ERROR_FILE(file, 0, "directory fails checksum " "at offset %llu", - (unsigned long long)filp->f_pos); - filp->f_pos += sb->s_blocksize - offset; + (unsigned long long)ctx->pos); + ctx->pos += sb->s_blocksize - offset; brelse(bh); continue; } set_buffer_verified(bh); -revalidate: /* If the dir block has changed since the last call to * readdir(2), then we might be pointing to an invalid * dirent right now. Scan from the start of the block * to make sure. */ - if (filp->f_version != inode->i_version) { + if (file->f_version != inode->i_version) { for (i = 0; i < sb->s_blocksize && i < offset; ) { de = (struct ext4_dir_entry_2 *) (bh->b_data + i); @@ -214,57 +208,46 @@ revalidate: sb->s_blocksize); } offset = i; - filp->f_pos = (filp->f_pos & ~(sb->s_blocksize - 1)) + ctx->pos = (ctx->pos & ~(sb->s_blocksize - 1)) | offset; - filp->f_version = inode->i_version; + file->f_version = inode->i_version; } - while (!error && filp->f_pos < inode->i_size + while (ctx->pos < inode->i_size && offset < sb->s_blocksize) { de = (struct ext4_dir_entry_2 *) (bh->b_data + offset); - if (ext4_check_dir_entry(inode, filp, de, bh, + if (ext4_check_dir_entry(inode, file, de, bh, bh->b_data, bh->b_size, offset)) { /* - * On error, skip the f_pos to the next block + * On error, skip to the next block */ - filp->f_pos = (filp->f_pos | + ctx->pos = (ctx->pos | (sb->s_blocksize - 1)) + 1; - brelse(bh); - ret = stored; - goto out; + break; } offset += ext4_rec_len_from_disk(de->rec_len, sb->s_blocksize); if (le32_to_cpu(de->inode)) { - /* We might block in the next section - * if the data destination is - * currently swapped out. So, use a - * version stamp to detect whether or - * not the directory has been modified - * during the copy operation. - */ - u64 version = filp->f_version; - - error = filldir(dirent, de->name, + if (!dir_emit(ctx, de->name, de->name_len, - filp->f_pos, le32_to_cpu(de->inode), - get_dtype(sb, de->file_type)); - if (error) - break; - if (version != filp->f_version) - goto revalidate; - stored++; + get_dtype(sb, de->file_type))) { + brelse(bh); + return 0; + } } - filp->f_pos += ext4_rec_len_from_disk(de->rec_len, + ctx->pos += ext4_rec_len_from_disk(de->rec_len, sb->s_blocksize); } offset = 0; brelse(bh); + if (ctx->pos < inode->i_size) { + if (!dir_relax(inode)) + return 0; + } } -out: - return ret; + return 0; } static inline int is_32bit_api(void) @@ -492,16 +475,12 @@ int ext4_htree_store_dirent(struct file *dir_file, __u32 hash, * for all entres on the fname linked list. (Normally there is only * one entry on the linked list, unless there are 62 bit hash collisions.) */ -static int call_filldir(struct file *filp, void *dirent, - filldir_t filldir, struct fname *fname) +static int call_filldir(struct file *file, struct dir_context *ctx, + struct fname *fname) { - struct dir_private_info *info = filp->private_data; - loff_t curr_pos; - struct inode *inode = file_inode(filp); - struct super_block *sb; - int error; - - sb = inode->i_sb; + struct dir_private_info *info = file->private_data; + struct inode *inode = file_inode(file); + struct super_block *sb = inode->i_sb; if (!fname) { ext4_msg(sb, KERN_ERR, "%s:%d: inode #%lu: comm %s: " @@ -509,47 +488,44 @@ static int call_filldir(struct file *filp, void *dirent, inode->i_ino, current->comm); return 0; } - curr_pos = hash2pos(filp, fname->hash, fname->minor_hash); + ctx->pos = hash2pos(file, fname->hash, fname->minor_hash); while (fname) { - error = filldir(dirent, fname->name, - fname->name_len, curr_pos, + if (!dir_emit(ctx, fname->name, + fname->name_len, fname->inode, - get_dtype(sb, fname->file_type)); - if (error) { - filp->f_pos = curr_pos; + get_dtype(sb, fname->file_type))) { info->extra_fname = fname; - return error; + return 1; } fname = fname->next; } return 0; } -static int ext4_dx_readdir(struct file *filp, - void *dirent, filldir_t filldir) +static int ext4_dx_readdir(struct file *file, struct dir_context *ctx) { - struct dir_private_info *info = filp->private_data; - struct inode *inode = file_inode(filp); + struct dir_private_info *info = file->private_data; + struct inode *inode = file_inode(file); struct fname *fname; int ret; if (!info) { - info = ext4_htree_create_dir_info(filp, filp->f_pos); + info = ext4_htree_create_dir_info(file, ctx->pos); if (!info) return -ENOMEM; - filp->private_data = info; + file->private_data = info; } - if (filp->f_pos == ext4_get_htree_eof(filp)) + if (ctx->pos == ext4_get_htree_eof(file)) return 0; /* EOF */ /* Some one has messed with f_pos; reset the world */ - if (info->last_pos != filp->f_pos) { + if (info->last_pos != ctx->pos) { free_rb_tree_fname(&info->root); info->curr_node = NULL; info->extra_fname = NULL; - info->curr_hash = pos2maj_hash(filp, filp->f_pos); - info->curr_minor_hash = pos2min_hash(filp, filp->f_pos); + info->curr_hash = pos2maj_hash(file, ctx->pos); + info->curr_minor_hash = pos2min_hash(file, ctx->pos); } /* @@ -557,7 +533,7 @@ static int ext4_dx_readdir(struct file *filp, * chain, return them first. */ if (info->extra_fname) { - if (call_filldir(filp, dirent, filldir, info->extra_fname)) + if (call_filldir(file, ctx, info->extra_fname)) goto finished; info->extra_fname = NULL; goto next_node; @@ -571,17 +547,17 @@ static int ext4_dx_readdir(struct file *filp, * cached entries. */ if ((!info->curr_node) || - (filp->f_version != inode->i_version)) { + (file->f_version != inode->i_version)) { info->curr_node = NULL; free_rb_tree_fname(&info->root); - filp->f_version = inode->i_version; - ret = ext4_htree_fill_tree(filp, info->curr_hash, + file->f_version = inode->i_version; + ret = ext4_htree_fill_tree(file, info->curr_hash, info->curr_minor_hash, &info->next_hash); if (ret < 0) return ret; if (ret == 0) { - filp->f_pos = ext4_get_htree_eof(filp); + ctx->pos = ext4_get_htree_eof(file); break; } info->curr_node = rb_first(&info->root); @@ -590,7 +566,7 @@ static int ext4_dx_readdir(struct file *filp, fname = rb_entry(info->curr_node, struct fname, rb_hash); info->curr_hash = fname->hash; info->curr_minor_hash = fname->minor_hash; - if (call_filldir(filp, dirent, filldir, fname)) + if (call_filldir(file, ctx, fname)) break; next_node: info->curr_node = rb_next(info->curr_node); @@ -601,7 +577,7 @@ static int ext4_dx_readdir(struct file *filp, info->curr_minor_hash = fname->minor_hash; } else { if (info->next_hash == ~0) { - filp->f_pos = ext4_get_htree_eof(filp); + ctx->pos = ext4_get_htree_eof(file); break; } info->curr_hash = info->next_hash; @@ -609,7 +585,7 @@ static int ext4_dx_readdir(struct file *filp, } } finished: - info->last_pos = filp->f_pos; + info->last_pos = ctx->pos; return 0; } @@ -624,7 +600,7 @@ static int ext4_release_dir(struct inode *inode, struct file *filp) const struct file_operations ext4_dir_operations = { .llseek = ext4_dir_llseek, .read = generic_read_dir, - .readdir = ext4_readdir, + .iterate = ext4_readdir, .unlocked_ioctl = ext4_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ext4_compat_ioctl, diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 6ed348d8d3eb..b577e45425b0 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -2573,7 +2573,7 @@ extern int ext4_try_create_inline_dir(handle_t *handle, struct inode *parent, struct inode *inode); extern int ext4_read_inline_dir(struct file *filp, - void *dirent, filldir_t filldir, + struct dir_context *ctx, int *has_inline_data); extern int htree_inlinedir_to_tree(struct file *dir_file, struct inode *dir, ext4_lblk_t block, diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c index b8a07466144f..d9ecbf1113a7 100644 --- a/fs/ext4/inline.c +++ b/fs/ext4/inline.c @@ -1404,16 +1404,15 @@ out: * offset as if '.' and '..' really take place. * */ -int ext4_read_inline_dir(struct file *filp, - void *dirent, filldir_t filldir, +int ext4_read_inline_dir(struct file *file, + struct dir_context *ctx, int *has_inline_data) { - int error = 0; unsigned int offset, parent_ino; - int i, stored; + int i; struct ext4_dir_entry_2 *de; struct super_block *sb; - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); int ret, inline_size = 0; struct ext4_iloc iloc; void *dir_buf = NULL; @@ -1444,9 +1443,8 @@ int ext4_read_inline_dir(struct file *filp, goto out; sb = inode->i_sb; - stored = 0; parent_ino = le32_to_cpu(((struct ext4_dir_entry_2 *)dir_buf)->inode); - offset = filp->f_pos; + offset = ctx->pos; /* * dotdot_offset and dotdot_size is the real offset and @@ -1460,104 +1458,74 @@ int ext4_read_inline_dir(struct file *filp, extra_offset = dotdot_size - EXT4_INLINE_DOTDOT_SIZE; extra_size = extra_offset + inline_size; - while (!error && !stored && filp->f_pos < extra_size) { -revalidate: - /* - * If the version has changed since the last call to - * readdir(2), then we might be pointing to an invalid - * dirent right now. Scan from the start of the inline - * dir to make sure. - */ - if (filp->f_version != inode->i_version) { - for (i = 0; i < extra_size && i < offset;) { - /* - * "." is with offset 0 and - * ".." is dotdot_offset. - */ - if (!i) { - i = dotdot_offset; - continue; - } else if (i == dotdot_offset) { - i = dotdot_size; - continue; - } - /* for other entry, the real offset in - * the buf has to be tuned accordingly. - */ - de = (struct ext4_dir_entry_2 *) - (dir_buf + i - extra_offset); - /* It's too expensive to do a full - * dirent test each time round this - * loop, but we do have to test at - * least that it is non-zero. A - * failure will be detected in the - * dirent test below. */ - if (ext4_rec_len_from_disk(de->rec_len, - extra_size) < EXT4_DIR_REC_LEN(1)) - break; - i += ext4_rec_len_from_disk(de->rec_len, - extra_size); - } - offset = i; - filp->f_pos = offset; - filp->f_version = inode->i_version; - } - - while (!error && filp->f_pos < extra_size) { - if (filp->f_pos == 0) { - error = filldir(dirent, ".", 1, 0, inode->i_ino, - DT_DIR); - if (error) - break; - stored++; - filp->f_pos = dotdot_offset; + /* + * If the version has changed since the last call to + * readdir(2), then we might be pointing to an invalid + * dirent right now. Scan from the start of the inline + * dir to make sure. + */ + if (file->f_version != inode->i_version) { + for (i = 0; i < extra_size && i < offset;) { + /* + * "." is with offset 0 and + * ".." is dotdot_offset. + */ + if (!i) { + i = dotdot_offset; + continue; + } else if (i == dotdot_offset) { + i = dotdot_size; continue; } + /* for other entry, the real offset in + * the buf has to be tuned accordingly. + */ + de = (struct ext4_dir_entry_2 *) + (dir_buf + i - extra_offset); + /* It's too expensive to do a full + * dirent test each time round this + * loop, but we do have to test at + * least that it is non-zero. A + * failure will be detected in the + * dirent test below. */ + if (ext4_rec_len_from_disk(de->rec_len, extra_size) + < EXT4_DIR_REC_LEN(1)) + break; + i += ext4_rec_len_from_disk(de->rec_len, + extra_size); + } + offset = i; + ctx->pos = offset; + file->f_version = inode->i_version; + } - if (filp->f_pos == dotdot_offset) { - error = filldir(dirent, "..", 2, - dotdot_offset, - parent_ino, DT_DIR); - if (error) - break; - stored++; + while (ctx->pos < extra_size) { + if (ctx->pos == 0) { + if (!dir_emit(ctx, ".", 1, inode->i_ino, DT_DIR)) + goto out; + ctx->pos = dotdot_offset; + continue; + } - filp->f_pos = dotdot_size; - continue; - } + if (ctx->pos == dotdot_offset) { + if (!dir_emit(ctx, "..", 2, parent_ino, DT_DIR)) + goto out; + ctx->pos = dotdot_size; + continue; + } - de = (struct ext4_dir_entry_2 *) - (dir_buf + filp->f_pos - extra_offset); - if (ext4_check_dir_entry(inode, filp, de, - iloc.bh, dir_buf, - extra_size, filp->f_pos)) { - ret = stored; + de = (struct ext4_dir_entry_2 *) + (dir_buf + ctx->pos - extra_offset); + if (ext4_check_dir_entry(inode, file, de, iloc.bh, dir_buf, + extra_size, ctx->pos)) + goto out; + if (le32_to_cpu(de->inode)) { + if (!dir_emit(ctx, de->name, de->name_len, + le32_to_cpu(de->inode), + get_dtype(sb, de->file_type))) goto out; - } - if (le32_to_cpu(de->inode)) { - /* We might block in the next section - * if the data destination is - * currently swapped out. So, use a - * version stamp to detect whether or - * not the directory has been modified - * during the copy operation. - */ - u64 version = filp->f_version; - - error = filldir(dirent, de->name, - de->name_len, - filp->f_pos, - le32_to_cpu(de->inode), - get_dtype(sb, de->file_type)); - if (error) - break; - if (version != filp->f_version) - goto revalidate; - stored++; - } - filp->f_pos += ext4_rec_len_from_disk(de->rec_len, - extra_size); } + ctx->pos += ext4_rec_len_from_disk(de->rec_len, extra_size); } out: kfree(dir_buf); diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c index 1ac6b93036b7..600bb5efe603 100644 --- a/fs/f2fs/dir.c +++ b/fs/f2fs/dir.c @@ -591,24 +591,19 @@ bool f2fs_empty_dir(struct inode *dir) return true; } -static int f2fs_readdir(struct file *file, void *dirent, filldir_t filldir) +static int f2fs_readdir(struct file *file, struct dir_context *ctx) { - unsigned long pos = file->f_pos; struct inode *inode = file_inode(file); unsigned long npages = dir_blocks(inode); - unsigned char *types = NULL; unsigned int bit_pos = 0, start_bit_pos = 0; - int over = 0; struct f2fs_dentry_block *dentry_blk = NULL; struct f2fs_dir_entry *de = NULL; struct page *dentry_page = NULL; - unsigned int n = 0; + unsigned int n = ((unsigned long)ctx->pos / NR_DENTRY_IN_BLOCK); unsigned char d_type = DT_UNKNOWN; int slots; - types = f2fs_filetype_table; - bit_pos = (pos % NR_DENTRY_IN_BLOCK); - n = (pos / NR_DENTRY_IN_BLOCK); + bit_pos = ((unsigned long)ctx->pos % NR_DENTRY_IN_BLOCK); for ( ; n < npages; n++) { dentry_page = get_lock_data_page(inode, n); @@ -618,31 +613,28 @@ static int f2fs_readdir(struct file *file, void *dirent, filldir_t filldir) start_bit_pos = bit_pos; dentry_blk = kmap(dentry_page); while (bit_pos < NR_DENTRY_IN_BLOCK) { - d_type = DT_UNKNOWN; bit_pos = find_next_bit_le(&dentry_blk->dentry_bitmap, NR_DENTRY_IN_BLOCK, bit_pos); if (bit_pos >= NR_DENTRY_IN_BLOCK) break; + ctx->pos += bit_pos - start_bit_pos; de = &dentry_blk->dentry[bit_pos]; - if (types && de->file_type < F2FS_FT_MAX) - d_type = types[de->file_type]; - - over = filldir(dirent, - dentry_blk->filename[bit_pos], - le16_to_cpu(de->name_len), - (n * NR_DENTRY_IN_BLOCK) + bit_pos, - le32_to_cpu(de->ino), d_type); - if (over) { - file->f_pos += bit_pos - start_bit_pos; + if (de->file_type < F2FS_FT_MAX) + d_type = f2fs_filetype_table[de->file_type]; + else + d_type = DT_UNKNOWN; + if (!dir_emit(ctx, + dentry_blk->filename[bit_pos], + le16_to_cpu(de->name_len), + le32_to_cpu(de->ino), d_type)) goto success; - } slots = GET_DENTRY_SLOTS(le16_to_cpu(de->name_len)); bit_pos += slots; } bit_pos = 0; - file->f_pos = (n + 1) * NR_DENTRY_IN_BLOCK; + ctx->pos = (n + 1) * NR_DENTRY_IN_BLOCK; kunmap(dentry_page); f2fs_put_page(dentry_page, 1); dentry_page = NULL; @@ -659,7 +651,7 @@ success: const struct file_operations f2fs_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = f2fs_readdir, + .iterate = f2fs_readdir, .fsync = f2fs_sync_file, .unlocked_ioctl = f2fs_ioctl, }; diff --git a/fs/fat/dir.c b/fs/fat/dir.c index 7a6f02caf286..3963ede84eb0 100644 --- a/fs/fat/dir.c +++ b/fs/fat/dir.c @@ -543,6 +543,7 @@ end_of_dir: EXPORT_SYMBOL_GPL(fat_search_long); struct fat_ioctl_filldir_callback { + struct dir_context ctx; void __user *dirent; int result; /* for dir ioctl */ @@ -552,8 +553,9 @@ struct fat_ioctl_filldir_callback { int short_len; }; -static int __fat_readdir(struct inode *inode, struct file *filp, void *dirent, - filldir_t filldir, int short_only, int both) +static int __fat_readdir(struct inode *inode, struct file *file, + struct dir_context *ctx, int short_only, + struct fat_ioctl_filldir_callback *both) { struct super_block *sb = inode->i_sb; struct msdos_sb_info *sbi = MSDOS_SB(sb); @@ -564,27 +566,20 @@ static int __fat_readdir(struct inode *inode, struct file *filp, void *dirent, unsigned char bufname[FAT_MAX_SHORT_SIZE]; int isvfat = sbi->options.isvfat; const char *fill_name = NULL; - unsigned long inum; - unsigned long lpos, dummy, *furrfu = &lpos; + int fake_offset = 0; loff_t cpos; int short_len = 0, fill_len = 0; int ret = 0; mutex_lock(&sbi->s_lock); - cpos = filp->f_pos; + cpos = ctx->pos; /* Fake . and .. for the root directory. */ if (inode->i_ino == MSDOS_ROOT_INO) { - while (cpos < 2) { - if (filldir(dirent, "..", cpos+1, cpos, - MSDOS_ROOT_INO, DT_DIR) < 0) - goto out; - cpos++; - filp->f_pos++; - } - if (cpos == 2) { - dummy = 2; - furrfu = &dummy; + if (!dir_emit_dots(file, ctx)) + goto out; + if (ctx->pos == 2) { + fake_offset = 1; cpos = 0; } } @@ -619,7 +614,7 @@ parse_record: int status = fat_parse_long(inode, &cpos, &bh, &de, &unicode, &nr_slots); if (status < 0) { - filp->f_pos = cpos; + ctx->pos = cpos; ret = status; goto out; } else if (status == PARSE_INVALID) @@ -639,6 +634,19 @@ parse_record: /* !both && !short_only, so we don't need shortname. */ if (!both) goto start_filldir; + + short_len = fat_parse_short(sb, de, bufname, + sbi->options.dotsOK); + if (short_len == 0) + goto record_end; + /* hack for fat_ioctl_filldir() */ + both->longname = fill_name; + both->long_len = fill_len; + both->shortname = bufname; + both->short_len = short_len; + fill_name = NULL; + fill_len = 0; + goto start_filldir; } } @@ -646,28 +654,21 @@ parse_record: if (short_len == 0) goto record_end; - if (nr_slots) { - /* hack for fat_ioctl_filldir() */ - struct fat_ioctl_filldir_callback *p = dirent; - - p->longname = fill_name; - p->long_len = fill_len; - p->shortname = bufname; - p->short_len = short_len; - fill_name = NULL; - fill_len = 0; - } else { - fill_name = bufname; - fill_len = short_len; - } + fill_name = bufname; + fill_len = short_len; start_filldir: - lpos = cpos - (nr_slots + 1) * sizeof(struct msdos_dir_entry); - if (!memcmp(de->name, MSDOS_DOT, MSDOS_NAME)) - inum = inode->i_ino; - else if (!memcmp(de->name, MSDOS_DOTDOT, MSDOS_NAME)) { - inum = parent_ino(filp->f_path.dentry); + if (!fake_offset) + ctx->pos = cpos - (nr_slots + 1) * sizeof(struct msdos_dir_entry); + + if (!memcmp(de->name, MSDOS_DOT, MSDOS_NAME)) { + if (!dir_emit_dot(file, ctx)) + goto fill_failed; + } else if (!memcmp(de->name, MSDOS_DOTDOT, MSDOS_NAME)) { + if (!dir_emit_dotdot(file, ctx)) + goto fill_failed; } else { + unsigned long inum; loff_t i_pos = fat_make_i_pos(sb, bh, de); struct inode *tmp = fat_iget(sb, i_pos); if (tmp) { @@ -675,18 +676,17 @@ start_filldir: iput(tmp); } else inum = iunique(sb, MSDOS_ROOT_INO); + if (!dir_emit(ctx, fill_name, fill_len, inum, + (de->attr & ATTR_DIR) ? DT_DIR : DT_REG)) + goto fill_failed; } - if (filldir(dirent, fill_name, fill_len, *furrfu, inum, - (de->attr & ATTR_DIR) ? DT_DIR : DT_REG) < 0) - goto fill_failed; - record_end: - furrfu = &lpos; - filp->f_pos = cpos; + fake_offset = 0; + ctx->pos = cpos; goto get_new; end_of_dir: - filp->f_pos = cpos; + ctx->pos = cpos; fill_failed: brelse(bh); if (unicode) @@ -696,10 +696,9 @@ out: return ret; } -static int fat_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int fat_readdir(struct file *file, struct dir_context *ctx) { - struct inode *inode = file_inode(filp); - return __fat_readdir(inode, filp, dirent, filldir, 0, 0); + return __fat_readdir(file_inode(file), file, ctx, 0, NULL); } #define FAT_IOCTL_FILLDIR_FUNC(func, dirent_type) \ @@ -755,20 +754,25 @@ efault: \ FAT_IOCTL_FILLDIR_FUNC(fat_ioctl_filldir, __fat_dirent) -static int fat_ioctl_readdir(struct inode *inode, struct file *filp, +static int fat_ioctl_readdir(struct inode *inode, struct file *file, void __user *dirent, filldir_t filldir, int short_only, int both) { - struct fat_ioctl_filldir_callback buf; + struct fat_ioctl_filldir_callback buf = { + .ctx.actor = filldir, + .dirent = dirent + }; int ret; buf.dirent = dirent; buf.result = 0; mutex_lock(&inode->i_mutex); + buf.ctx.pos = file->f_pos; ret = -ENOENT; if (!IS_DEADDIR(inode)) { - ret = __fat_readdir(inode, filp, &buf, filldir, - short_only, both); + ret = __fat_readdir(inode, file, &buf.ctx, + short_only, both ? &buf : NULL); + file->f_pos = buf.ctx.pos; } mutex_unlock(&inode->i_mutex); if (ret >= 0) @@ -854,7 +858,7 @@ static long fat_compat_dir_ioctl(struct file *filp, unsigned cmd, const struct file_operations fat_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = fat_readdir, + .iterate = fat_readdir, .unlocked_ioctl = fat_dir_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = fat_compat_dir_ioctl, diff --git a/fs/fat/inode.c b/fs/fat/inode.c index dfce656ddb33..5d4513cb1b3c 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -1229,6 +1229,19 @@ static int fat_read_root(struct inode *inode) return 0; } +static unsigned long calc_fat_clusters(struct super_block *sb) +{ + struct msdos_sb_info *sbi = MSDOS_SB(sb); + + /* Divide first to avoid overflow */ + if (sbi->fat_bits != 12) { + unsigned long ent_per_sec = sb->s_blocksize * 8 / sbi->fat_bits; + return ent_per_sec * sbi->fat_length; + } + + return sbi->fat_length * sb->s_blocksize * 8 / sbi->fat_bits; +} + /* * Read the super block of an MS-DOS FS. */ @@ -1434,7 +1447,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, int isvfat, sbi->dirty = b->fat16.state & FAT_STATE_DIRTY; /* check that FAT table does not overflow */ - fat_clusters = sbi->fat_length * sb->s_blocksize * 8 / sbi->fat_bits; + fat_clusters = calc_fat_clusters(sb); total_clusters = min(total_clusters, fat_clusters - FAT_START_ENT); if (total_clusters > MAX_FAT(sb)) { if (!silent) diff --git a/fs/file_table.c b/fs/file_table.c index cd4d87a82951..485dc0eddd67 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -306,17 +306,18 @@ void fput(struct file *file) { if (atomic_long_dec_and_test(&file->f_count)) { struct task_struct *task = current; + unsigned long flags; + file_sb_list_del(file); - if (unlikely(in_interrupt() || task->flags & PF_KTHREAD)) { - unsigned long flags; - spin_lock_irqsave(&delayed_fput_lock, flags); - list_add(&file->f_u.fu_list, &delayed_fput_list); - schedule_work(&delayed_fput_work); - spin_unlock_irqrestore(&delayed_fput_lock, flags); - return; + if (likely(!in_interrupt() && !(task->flags & PF_KTHREAD))) { + init_task_work(&file->f_u.fu_rcuhead, ____fput); + if (!task_work_add(task, &file->f_u.fu_rcuhead, true)) + return; } - init_task_work(&file->f_u.fu_rcuhead, ____fput); - task_work_add(task, &file->f_u.fu_rcuhead, true); + spin_lock_irqsave(&delayed_fput_lock, flags); + list_add(&file->f_u.fu_list, &delayed_fput_list); + schedule_work(&delayed_fput_work); + spin_unlock_irqrestore(&delayed_fput_lock, flags); } } diff --git a/fs/freevxfs/vxfs_lookup.c b/fs/freevxfs/vxfs_lookup.c index 664b07a53870..25d4099a4aea 100644 --- a/fs/freevxfs/vxfs_lookup.c +++ b/fs/freevxfs/vxfs_lookup.c @@ -49,7 +49,7 @@ static struct dentry * vxfs_lookup(struct inode *, struct dentry *, unsigned int); -static int vxfs_readdir(struct file *, void *, filldir_t); +static int vxfs_readdir(struct file *, struct dir_context *); const struct inode_operations vxfs_dir_inode_ops = { .lookup = vxfs_lookup, @@ -58,7 +58,7 @@ const struct inode_operations vxfs_dir_inode_ops = { const struct file_operations vxfs_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = vxfs_readdir, + .iterate = vxfs_readdir, }; @@ -235,7 +235,7 @@ vxfs_lookup(struct inode *dip, struct dentry *dp, unsigned int flags) * Zero. */ static int -vxfs_readdir(struct file *fp, void *retp, filldir_t filler) +vxfs_readdir(struct file *fp, struct dir_context *ctx) { struct inode *ip = file_inode(fp); struct super_block *sbp = ip->i_sb; @@ -243,20 +243,17 @@ vxfs_readdir(struct file *fp, void *retp, filldir_t filler) u_long page, npages, block, pblocks, nblocks, offset; loff_t pos; - switch ((long)fp->f_pos) { - case 0: - if (filler(retp, ".", 1, fp->f_pos, ip->i_ino, DT_DIR) < 0) - goto out; - fp->f_pos++; - /* fallthrough */ - case 1: - if (filler(retp, "..", 2, fp->f_pos, VXFS_INO(ip)->vii_dotdot, DT_DIR) < 0) - goto out; - fp->f_pos++; - /* fallthrough */ + if (ctx->pos == 0) { + if (!dir_emit_dot(fp, ctx)) + return 0; + ctx->pos = 1; } - - pos = fp->f_pos - 2; + if (ctx->pos == 1) { + if (!dir_emit(ctx, "..", 2, VXFS_INO(ip)->vii_dotdot, DT_DIR)) + return 0; + ctx->pos = 2; + } + pos = ctx->pos - 2; if (pos > VXFS_DIRROUND(ip->i_size)) return 0; @@ -270,16 +267,16 @@ vxfs_readdir(struct file *fp, void *retp, filldir_t filler) block = (u_long)(pos >> sbp->s_blocksize_bits) % pblocks; for (; page < npages; page++, block = 0) { - caddr_t kaddr; + char *kaddr; struct page *pp; pp = vxfs_get_page(ip->i_mapping, page); if (IS_ERR(pp)) continue; - kaddr = (caddr_t)page_address(pp); + kaddr = (char *)page_address(pp); for (; block <= nblocks && block <= pblocks; block++) { - caddr_t baddr, limit; + char *baddr, *limit; struct vxfs_dirblk *dbp; struct vxfs_direct *de; @@ -292,21 +289,18 @@ vxfs_readdir(struct file *fp, void *retp, filldir_t filler) (kaddr + offset) : (baddr + VXFS_DIRBLKOV(dbp))); - for (; (caddr_t)de <= limit; de = vxfs_next_entry(de)) { - int over; - + for (; (char *)de <= limit; de = vxfs_next_entry(de)) { if (!de->d_reclen) break; if (!de->d_ino) continue; - offset = (caddr_t)de - kaddr; - over = filler(retp, de->d_name, de->d_namelen, - ((page << PAGE_CACHE_SHIFT) | offset) + 2, - de->d_ino, DT_UNKNOWN); - if (over) { + offset = (char *)de - kaddr; + ctx->pos = ((page << PAGE_CACHE_SHIFT) | offset) + 2; + if (!dir_emit(ctx, de->d_name, de->d_namelen, + de->d_ino, DT_UNKNOWN)) { vxfs_put_page(pp); - goto done; + return 0; } } offset = 0; @@ -314,9 +308,6 @@ vxfs_readdir(struct file *fp, void *retp, filldir_t filler) vxfs_put_page(pp); offset = 0; } - -done: - fp->f_pos = ((page << PAGE_CACHE_SHIFT) | offset) + 2; -out: + ctx->pos = ((page << PAGE_CACHE_SHIFT) | offset) + 2; return 0; } diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 3be57189efd5..a85ac4e33436 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -45,6 +45,7 @@ struct wb_writeback_work { unsigned int for_kupdate:1; unsigned int range_cyclic:1; unsigned int for_background:1; + unsigned int for_sync:1; /* sync(2) WB_SYNC_ALL writeback */ enum wb_reason reason; /* why was writeback initiated? */ struct list_head list; /* pending work list */ @@ -443,9 +444,11 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc) /* * Make sure to wait on the data before writing out the metadata. * This is important for filesystems that modify metadata on data - * I/O completion. + * I/O completion. We don't do it for sync(2) writeback because it has a + * separate, external IO completion path and ->sync_fs for guaranteeing + * inode metadata is written back correctly. */ - if (wbc->sync_mode == WB_SYNC_ALL) { + if (wbc->sync_mode == WB_SYNC_ALL && !wbc->for_sync) { int err = filemap_fdatawait(mapping); if (ret == 0) ret = err; @@ -578,6 +581,7 @@ static long writeback_sb_inodes(struct super_block *sb, .tagged_writepages = work->tagged_writepages, .for_kupdate = work->for_kupdate, .for_background = work->for_background, + .for_sync = work->for_sync, .range_cyclic = work->range_cyclic, .range_start = 0, .range_end = LLONG_MAX, @@ -1362,6 +1366,7 @@ void sync_inodes_sb(struct super_block *sb) .range_cyclic = 0, .done = &done, .reason = WB_REASON_SYNC, + .for_sync = 1, }; /* Nothing to do? */ diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 254df56b847b..0eda52738ec4 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -14,7 +14,7 @@ #include <linux/namei.h> #include <linux/slab.h> -static bool fuse_use_readdirplus(struct inode *dir, struct file *filp) +static bool fuse_use_readdirplus(struct inode *dir, struct dir_context *ctx) { struct fuse_conn *fc = get_fuse_conn(dir); struct fuse_inode *fi = get_fuse_inode(dir); @@ -25,7 +25,7 @@ static bool fuse_use_readdirplus(struct inode *dir, struct file *filp) return true; if (test_and_clear_bit(FUSE_I_ADVISE_RDPLUS, &fi->state)) return true; - if (filp->f_pos == 0) + if (ctx->pos == 0) return true; return false; } @@ -180,6 +180,8 @@ u64 fuse_get_attr_version(struct fuse_conn *fc) static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags) { struct inode *inode; + struct dentry *parent; + struct fuse_conn *fc; inode = ACCESS_ONCE(entry->d_inode); if (inode && is_bad_inode(inode)) @@ -187,10 +189,8 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags) else if (fuse_dentry_time(entry) < get_jiffies_64()) { int err; struct fuse_entry_out outarg; - struct fuse_conn *fc; struct fuse_req *req; struct fuse_forget_link *forget; - struct dentry *parent; u64 attr_version; /* For negative dentries, always do a fresh lookup */ @@ -241,8 +241,14 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags) entry_attr_timeout(&outarg), attr_version); fuse_change_entry_timeout(entry, &outarg); + } else if (inode) { + fc = get_fuse_conn(inode); + if (fc->readdirplus_auto) { + parent = dget_parent(entry); + fuse_advise_use_readdirplus(parent->d_inode); + dput(parent); + } } - fuse_advise_use_readdirplus(inode); return 1; } @@ -1159,25 +1165,23 @@ static int fuse_permission(struct inode *inode, int mask) } static int parse_dirfile(char *buf, size_t nbytes, struct file *file, - void *dstbuf, filldir_t filldir) + struct dir_context *ctx) { while (nbytes >= FUSE_NAME_OFFSET) { struct fuse_dirent *dirent = (struct fuse_dirent *) buf; size_t reclen = FUSE_DIRENT_SIZE(dirent); - int over; if (!dirent->namelen || dirent->namelen > FUSE_NAME_MAX) return -EIO; if (reclen > nbytes) break; - over = filldir(dstbuf, dirent->name, dirent->namelen, - file->f_pos, dirent->ino, dirent->type); - if (over) + if (!dir_emit(ctx, dirent->name, dirent->namelen, + dirent->ino, dirent->type)) break; buf += reclen; nbytes -= reclen; - file->f_pos = dirent->off; + ctx->pos = dirent->off; } return 0; @@ -1278,7 +1282,7 @@ out: } static int parse_dirplusfile(char *buf, size_t nbytes, struct file *file, - void *dstbuf, filldir_t filldir, u64 attr_version) + struct dir_context *ctx, u64 attr_version) { struct fuse_direntplus *direntplus; struct fuse_dirent *dirent; @@ -1303,10 +1307,9 @@ static int parse_dirplusfile(char *buf, size_t nbytes, struct file *file, we need to send a FORGET for each of those which we did not link. */ - over = filldir(dstbuf, dirent->name, dirent->namelen, - file->f_pos, dirent->ino, - dirent->type); - file->f_pos = dirent->off; + over = !dir_emit(ctx, dirent->name, dirent->namelen, + dirent->ino, dirent->type); + ctx->pos = dirent->off; } buf += reclen; @@ -1320,7 +1323,7 @@ static int parse_dirplusfile(char *buf, size_t nbytes, struct file *file, return 0; } -static int fuse_readdir(struct file *file, void *dstbuf, filldir_t filldir) +static int fuse_readdir(struct file *file, struct dir_context *ctx) { int plus, err; size_t nbytes; @@ -1343,17 +1346,17 @@ static int fuse_readdir(struct file *file, void *dstbuf, filldir_t filldir) return -ENOMEM; } - plus = fuse_use_readdirplus(inode, file); + plus = fuse_use_readdirplus(inode, ctx); req->out.argpages = 1; req->num_pages = 1; req->pages[0] = page; req->page_descs[0].length = PAGE_SIZE; if (plus) { attr_version = fuse_get_attr_version(fc); - fuse_read_fill(req, file, file->f_pos, PAGE_SIZE, + fuse_read_fill(req, file, ctx->pos, PAGE_SIZE, FUSE_READDIRPLUS); } else { - fuse_read_fill(req, file, file->f_pos, PAGE_SIZE, + fuse_read_fill(req, file, ctx->pos, PAGE_SIZE, FUSE_READDIR); } fuse_request_send(fc, req); @@ -1363,11 +1366,11 @@ static int fuse_readdir(struct file *file, void *dstbuf, filldir_t filldir) if (!err) { if (plus) { err = parse_dirplusfile(page_address(page), nbytes, - file, dstbuf, filldir, + file, ctx, attr_version); } else { err = parse_dirfile(page_address(page), nbytes, file, - dstbuf, filldir); + ctx); } } @@ -1880,7 +1883,7 @@ static const struct inode_operations fuse_dir_inode_operations = { static const struct file_operations fuse_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = fuse_readdir, + .iterate = fuse_readdir, .open = fuse_dir_open, .release = fuse_dir_release, .fsync = fuse_dir_fsync, diff --git a/fs/fuse/file.c b/fs/fuse/file.c index d1c9b85b3f58..35f281033142 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -16,6 +16,7 @@ #include <linux/compat.h> #include <linux/swap.h> #include <linux/aio.h> +#include <linux/falloc.h> static const struct file_operations fuse_direct_io_file_operations; @@ -1278,7 +1279,10 @@ ssize_t fuse_direct_io(struct fuse_io_priv *io, const struct iovec *iov, iov_iter_init(&ii, iov, nr_segs, count, 0); - req = fuse_get_req(fc, fuse_iter_npages(&ii)); + if (io->async) + req = fuse_get_req_for_background(fc, fuse_iter_npages(&ii)); + else + req = fuse_get_req(fc, fuse_iter_npages(&ii)); if (IS_ERR(req)) return PTR_ERR(req); @@ -1314,7 +1318,11 @@ ssize_t fuse_direct_io(struct fuse_io_priv *io, const struct iovec *iov, break; if (count) { fuse_put_request(fc, req); - req = fuse_get_req(fc, fuse_iter_npages(&ii)); + if (io->async) + req = fuse_get_req_for_background(fc, + fuse_iter_npages(&ii)); + else + req = fuse_get_req(fc, fuse_iter_npages(&ii)); if (IS_ERR(req)) break; } @@ -2365,6 +2373,11 @@ static void fuse_do_truncate(struct file *file) fuse_do_setattr(inode, &attr, file); } +static inline loff_t fuse_round_up(loff_t off) +{ + return round_up(off, FUSE_MAX_PAGES_PER_REQ << PAGE_SHIFT); +} + static ssize_t fuse_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, loff_t offset, unsigned long nr_segs) @@ -2372,6 +2385,7 @@ fuse_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, ssize_t ret = 0; struct file *file = iocb->ki_filp; struct fuse_file *ff = file->private_data; + bool async_dio = ff->fc->async_dio; loff_t pos = 0; struct inode *inode; loff_t i_size; @@ -2383,10 +2397,10 @@ fuse_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, i_size = i_size_read(inode); /* optimization for short read */ - if (rw != WRITE && offset + count > i_size) { + if (async_dio && rw != WRITE && offset + count > i_size) { if (offset >= i_size) return 0; - count = i_size - offset; + count = min_t(loff_t, count, fuse_round_up(i_size - offset)); } io = kmalloc(sizeof(struct fuse_io_priv), GFP_KERNEL); @@ -2404,7 +2418,7 @@ fuse_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, * By default, we want to optimize all I/Os with async request * submission to the client filesystem if supported. */ - io->async = ff->fc->async_dio; + io->async = async_dio; io->iocb = iocb; /* @@ -2412,7 +2426,7 @@ fuse_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, * to wait on real async I/O requests, so we must submit this request * synchronously. */ - if (!is_sync_kiocb(iocb) && (offset + count > i_size)) + if (!is_sync_kiocb(iocb) && (offset + count > i_size) && rw == WRITE) io->async = false; if (rw == WRITE) @@ -2424,7 +2438,7 @@ fuse_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, fuse_aio_complete(io, ret < 0 ? ret : 0, -1); /* we have a non-extending, async request, so return */ - if (ret > 0 && !is_sync_kiocb(iocb)) + if (!is_sync_kiocb(iocb)) return -EIOCBQUEUED; ret = wait_on_sync_kiocb(iocb); @@ -2446,6 +2460,7 @@ static long fuse_file_fallocate(struct file *file, int mode, loff_t offset, loff_t length) { struct fuse_file *ff = file->private_data; + struct inode *inode = file->f_inode; struct fuse_conn *fc = ff->fc; struct fuse_req *req; struct fuse_fallocate_in inarg = { @@ -2455,13 +2470,23 @@ static long fuse_file_fallocate(struct file *file, int mode, loff_t offset, .mode = mode }; int err; + bool lock_inode = !(mode & FALLOC_FL_KEEP_SIZE) || + (mode & FALLOC_FL_PUNCH_HOLE); if (fc->no_fallocate) return -EOPNOTSUPP; + if (lock_inode) { + mutex_lock(&inode->i_mutex); + if (mode & FALLOC_FL_PUNCH_HOLE) + fuse_set_nowrite(inode); + } + req = fuse_get_req_nopages(fc); - if (IS_ERR(req)) - return PTR_ERR(req); + if (IS_ERR(req)) { + err = PTR_ERR(req); + goto out; + } req->in.h.opcode = FUSE_FALLOCATE; req->in.h.nodeid = ff->nodeid; @@ -2476,6 +2501,25 @@ static long fuse_file_fallocate(struct file *file, int mode, loff_t offset, } fuse_put_request(fc, req); + if (err) + goto out; + + /* we could have extended the file */ + if (!(mode & FALLOC_FL_KEEP_SIZE)) + fuse_write_update_size(inode, offset + length); + + if (mode & FALLOC_FL_PUNCH_HOLE) + truncate_pagecache_range(inode, offset, offset + length - 1); + + fuse_invalidate_attr(inode); + +out: + if (lock_inode) { + if (mode & FALLOC_FL_PUNCH_HOLE) + fuse_release_nowrite(inode); + mutex_unlock(&inode->i_mutex); + } + return err; } diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 6201f81e4d3a..9a0cdde14a08 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -867,10 +867,11 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_req *req) fc->dont_mask = 1; if (arg->flags & FUSE_AUTO_INVAL_DATA) fc->auto_inval_data = 1; - if (arg->flags & FUSE_DO_READDIRPLUS) + if (arg->flags & FUSE_DO_READDIRPLUS) { fc->do_readdirplus = 1; - if (arg->flags & FUSE_READDIRPLUS_AUTO) - fc->readdirplus_auto = 1; + if (arg->flags & FUSE_READDIRPLUS_AUTO) + fc->readdirplus_auto = 1; + } if (arg->flags & FUSE_ASYNC_DIO) fc->async_dio = 1; } else { diff --git a/fs/gfs2/Kconfig b/fs/gfs2/Kconfig index eb08c9e43c2a..5a376ab81feb 100644 --- a/fs/gfs2/Kconfig +++ b/fs/gfs2/Kconfig @@ -26,7 +26,7 @@ config GFS2_FS config GFS2_FS_LOCKING_DLM bool "GFS2 DLM locking" depends on (GFS2_FS!=n) && NET && INET && (IPV6 || IPV6=n) && \ - HOTPLUG && DLM && CONFIGFS_FS && SYSFS + HOTPLUG && CONFIGFS_FS && SYSFS && (DLM=y || DLM=GFS2_FS) help Multiple node locking module for GFS2 diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 1dc9a13ce6bb..93b5809c20bb 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -1286,17 +1286,26 @@ int gfs2_setattr_size(struct inode *inode, u64 newsize) if (ret) return ret; + ret = get_write_access(inode); + if (ret) + return ret; + inode_dio_wait(inode); ret = gfs2_rs_alloc(GFS2_I(inode)); if (ret) - return ret; + goto out; oldsize = inode->i_size; - if (newsize >= oldsize) - return do_grow(inode, newsize); + if (newsize >= oldsize) { + ret = do_grow(inode, newsize); + goto out; + } - return do_shrink(inode, oldsize, newsize); + ret = do_shrink(inode, oldsize, newsize); +out: + put_write_access(inode); + return ret; } int gfs2_truncatei_resume(struct gfs2_inode *ip) diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index c3e82bd23179..e0449c10286a 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -354,22 +354,31 @@ static __be64 *gfs2_dir_get_hash_table(struct gfs2_inode *ip) return ERR_PTR(-EIO); } - hc = kmalloc(hsize, GFP_NOFS); - ret = -ENOMEM; + hc = kmalloc(hsize, GFP_NOFS | __GFP_NOWARN); + if (hc == NULL) + hc = __vmalloc(hsize, GFP_NOFS, PAGE_KERNEL); + if (hc == NULL) return ERR_PTR(-ENOMEM); ret = gfs2_dir_read_data(ip, hc, hsize); if (ret < 0) { - kfree(hc); + if (is_vmalloc_addr(hc)) + vfree(hc); + else + kfree(hc); return ERR_PTR(ret); } spin_lock(&inode->i_lock); - if (ip->i_hash_cache) - kfree(hc); - else + if (ip->i_hash_cache) { + if (is_vmalloc_addr(hc)) + vfree(hc); + else + kfree(hc); + } else { ip->i_hash_cache = hc; + } spin_unlock(&inode->i_lock); return ip->i_hash_cache; @@ -385,7 +394,10 @@ void gfs2_dir_hash_inval(struct gfs2_inode *ip) { __be64 *hc = ip->i_hash_cache; ip->i_hash_cache = NULL; - kfree(hc); + if (is_vmalloc_addr(hc)) + vfree(hc); + else + kfree(hc); } static inline int gfs2_dirent_sentinel(const struct gfs2_dirent *dent) @@ -1113,7 +1125,10 @@ static int dir_double_exhash(struct gfs2_inode *dip) if (IS_ERR(hc)) return PTR_ERR(hc); - h = hc2 = kmalloc(hsize_bytes * 2, GFP_NOFS); + h = hc2 = kmalloc(hsize_bytes * 2, GFP_NOFS | __GFP_NOWARN); + if (hc2 == NULL) + hc2 = __vmalloc(hsize_bytes * 2, GFP_NOFS, PAGE_KERNEL); + if (!hc2) return -ENOMEM; @@ -1145,7 +1160,10 @@ fail: gfs2_dinode_out(dip, dibh->b_data); brelse(dibh); out_kfree: - kfree(hc2); + if (is_vmalloc_addr(hc2)) + vfree(hc2); + else + kfree(hc2); return error; } @@ -1194,9 +1212,7 @@ static int compare_dents(const void *a, const void *b) /** * do_filldir_main - read out directory entries * @dip: The GFS2 inode - * @offset: The offset in the file to read from - * @opaque: opaque data to pass to filldir - * @filldir: The function to pass entries to + * @ctx: what to feed the entries to * @darr: an array of struct gfs2_dirent pointers to read * @entries: the number of entries in darr * @copied: pointer to int that's non-zero if a entry has been copied out @@ -1206,11 +1222,10 @@ static int compare_dents(const void *a, const void *b) * the possibility that they will fall into different readdir buffers or * that someone will want to seek to that location. * - * Returns: errno, >0 on exception from filldir + * Returns: errno, >0 if the actor tells you to stop */ -static int do_filldir_main(struct gfs2_inode *dip, u64 *offset, - void *opaque, filldir_t filldir, +static int do_filldir_main(struct gfs2_inode *dip, struct dir_context *ctx, const struct gfs2_dirent **darr, u32 entries, int *copied) { @@ -1218,7 +1233,6 @@ static int do_filldir_main(struct gfs2_inode *dip, u64 *offset, u64 off, off_next; unsigned int x, y; int run = 0; - int error = 0; sort(darr, entries, sizeof(struct gfs2_dirent *), compare_dents, NULL); @@ -1235,9 +1249,9 @@ static int do_filldir_main(struct gfs2_inode *dip, u64 *offset, off_next = be32_to_cpu(dent_next->de_hash); off_next = gfs2_disk_hash2offset(off_next); - if (off < *offset) + if (off < ctx->pos) continue; - *offset = off; + ctx->pos = off; if (off_next == off) { if (*copied && !run) @@ -1246,26 +1260,25 @@ static int do_filldir_main(struct gfs2_inode *dip, u64 *offset, } else run = 0; } else { - if (off < *offset) + if (off < ctx->pos) continue; - *offset = off; + ctx->pos = off; } - error = filldir(opaque, (const char *)(dent + 1), + if (!dir_emit(ctx, (const char *)(dent + 1), be16_to_cpu(dent->de_name_len), - off, be64_to_cpu(dent->de_inum.no_addr), - be16_to_cpu(dent->de_type)); - if (error) + be64_to_cpu(dent->de_inum.no_addr), + be16_to_cpu(dent->de_type))) return 1; *copied = 1; } - /* Increment the *offset by one, so the next time we come into the + /* Increment the ctx->pos by one, so the next time we come into the do_filldir fxn, we get the next entry instead of the last one in the current leaf */ - (*offset)++; + ctx->pos++; return 0; } @@ -1289,8 +1302,8 @@ static void gfs2_free_sort_buffer(void *ptr) kfree(ptr); } -static int gfs2_dir_read_leaf(struct inode *inode, u64 *offset, void *opaque, - filldir_t filldir, int *copied, unsigned *depth, +static int gfs2_dir_read_leaf(struct inode *inode, struct dir_context *ctx, + int *copied, unsigned *depth, u64 leaf_no) { struct gfs2_inode *ip = GFS2_I(inode); @@ -1368,8 +1381,7 @@ static int gfs2_dir_read_leaf(struct inode *inode, u64 *offset, void *opaque, } while(lfn); BUG_ON(entries2 != entries); - error = do_filldir_main(ip, offset, opaque, filldir, darr, - entries, copied); + error = do_filldir_main(ip, ctx, darr, entries, copied); out_free: for(i = 0; i < leaf; i++) brelse(larr[i]); @@ -1428,15 +1440,13 @@ static void gfs2_dir_readahead(struct inode *inode, unsigned hsize, u32 index, /** * dir_e_read - Reads the entries from a directory into a filldir buffer * @dip: dinode pointer - * @offset: the hash of the last entry read shifted to the right once - * @opaque: buffer for the filldir function to fill - * @filldir: points to the filldir function to use + * @ctx: actor to feed the entries to * * Returns: errno */ -static int dir_e_read(struct inode *inode, u64 *offset, void *opaque, - filldir_t filldir, struct file_ra_state *f_ra) +static int dir_e_read(struct inode *inode, struct dir_context *ctx, + struct file_ra_state *f_ra) { struct gfs2_inode *dip = GFS2_I(inode); u32 hsize, len = 0; @@ -1447,7 +1457,7 @@ static int dir_e_read(struct inode *inode, u64 *offset, void *opaque, unsigned depth = 0; hsize = 1 << dip->i_depth; - hash = gfs2_dir_offset2hash(*offset); + hash = gfs2_dir_offset2hash(ctx->pos); index = hash >> (32 - dip->i_depth); if (dip->i_hash_cache == NULL) @@ -1459,7 +1469,7 @@ static int dir_e_read(struct inode *inode, u64 *offset, void *opaque, gfs2_dir_readahead(inode, hsize, index, f_ra); while (index < hsize) { - error = gfs2_dir_read_leaf(inode, offset, opaque, filldir, + error = gfs2_dir_read_leaf(inode, ctx, &copied, &depth, be64_to_cpu(lp[index])); if (error) @@ -1474,8 +1484,8 @@ static int dir_e_read(struct inode *inode, u64 *offset, void *opaque, return error; } -int gfs2_dir_read(struct inode *inode, u64 *offset, void *opaque, - filldir_t filldir, struct file_ra_state *f_ra) +int gfs2_dir_read(struct inode *inode, struct dir_context *ctx, + struct file_ra_state *f_ra) { struct gfs2_inode *dip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); @@ -1489,7 +1499,7 @@ int gfs2_dir_read(struct inode *inode, u64 *offset, void *opaque, return 0; if (dip->i_diskflags & GFS2_DIF_EXHASH) - return dir_e_read(inode, offset, opaque, filldir, f_ra); + return dir_e_read(inode, ctx, f_ra); if (!gfs2_is_stuffed(dip)) { gfs2_consist_inode(dip); @@ -1521,7 +1531,7 @@ int gfs2_dir_read(struct inode *inode, u64 *offset, void *opaque, error = -EIO; goto out; } - error = do_filldir_main(dip, offset, opaque, filldir, darr, + error = do_filldir_main(dip, ctx, darr, dip->i_entries, &copied); out: kfree(darr); @@ -1846,6 +1856,8 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, memset(&rlist, 0, sizeof(struct gfs2_rgrp_list)); ht = kzalloc(size, GFP_NOFS); + if (ht == NULL) + ht = vzalloc(size); if (!ht) return -ENOMEM; @@ -1933,7 +1945,10 @@ out_rlist: gfs2_rlist_free(&rlist); gfs2_quota_unhold(dip); out: - kfree(ht); + if (is_vmalloc_addr(ht)) + vfree(ht); + else + kfree(ht); return error; } diff --git a/fs/gfs2/dir.h b/fs/gfs2/dir.h index 98c960beab35..ba9000bc1397 100644 --- a/fs/gfs2/dir.h +++ b/fs/gfs2/dir.h @@ -24,8 +24,8 @@ extern int gfs2_dir_check(struct inode *dir, const struct qstr *filename, extern int gfs2_dir_add(struct inode *inode, const struct qstr *filename, const struct gfs2_inode *ip); extern int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry); -extern int gfs2_dir_read(struct inode *inode, u64 *offset, void *opaque, - filldir_t filldir, struct file_ra_state *f_ra); +extern int gfs2_dir_read(struct inode *inode, struct dir_context *ctx, + struct file_ra_state *f_ra); extern int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename, const struct gfs2_inode *nip, unsigned int new_type); diff --git a/fs/gfs2/export.c b/fs/gfs2/export.c index 9973df4ff565..8b9b3775e2e7 100644 --- a/fs/gfs2/export.c +++ b/fs/gfs2/export.c @@ -64,6 +64,7 @@ static int gfs2_encode_fh(struct inode *inode, __u32 *p, int *len, } struct get_name_filldir { + struct dir_context ctx; struct gfs2_inum_host inum; char *name; }; @@ -88,9 +89,11 @@ static int gfs2_get_name(struct dentry *parent, char *name, struct inode *dir = parent->d_inode; struct inode *inode = child->d_inode; struct gfs2_inode *dip, *ip; - struct get_name_filldir gnfd; + struct get_name_filldir gnfd = { + .ctx.actor = get_name_filldir, + .name = name + }; struct gfs2_holder gh; - u64 offset = 0; int error; struct file_ra_state f_ra = { .start = 0 }; @@ -106,13 +109,12 @@ static int gfs2_get_name(struct dentry *parent, char *name, *name = 0; gnfd.inum.no_addr = ip->i_no_addr; gnfd.inum.no_formal_ino = ip->i_no_formal_ino; - gnfd.name = name; error = gfs2_glock_nq_init(dip->i_gl, LM_ST_SHARED, 0, &gh); if (error) return error; - error = gfs2_dir_read(dir, &offset, &gnfd, get_name_filldir, &f_ra); + error = gfs2_dir_read(dir, &gnfd.ctx, &f_ra); gfs2_glock_dq_uninit(&gh); diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index acd16764b133..b3333371aebb 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -82,35 +82,28 @@ static loff_t gfs2_llseek(struct file *file, loff_t offset, int whence) } /** - * gfs2_readdir - Read directory entries from a directory + * gfs2_readdir - Iterator for a directory * @file: The directory to read from - * @dirent: Buffer for dirents - * @filldir: Function used to do the copying + * @ctx: What to feed directory entries to * * Returns: errno */ -static int gfs2_readdir(struct file *file, void *dirent, filldir_t filldir) +static int gfs2_readdir(struct file *file, struct dir_context *ctx) { struct inode *dir = file->f_mapping->host; struct gfs2_inode *dip = GFS2_I(dir); struct gfs2_holder d_gh; - u64 offset = file->f_pos; int error; - gfs2_holder_init(dip->i_gl, LM_ST_SHARED, 0, &d_gh); - error = gfs2_glock_nq(&d_gh); - if (error) { - gfs2_holder_uninit(&d_gh); + error = gfs2_glock_nq_init(dip->i_gl, LM_ST_SHARED, 0, &d_gh); + if (error) return error; - } - error = gfs2_dir_read(dir, &offset, dirent, filldir, &file->f_ra); + error = gfs2_dir_read(dir, ctx, &file->f_ra); gfs2_glock_dq_uninit(&d_gh); - file->f_pos = offset; - return error; } @@ -402,16 +395,20 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) /* Update file times before taking page lock */ file_update_time(vma->vm_file); + ret = get_write_access(inode); + if (ret) + goto out; + ret = gfs2_rs_alloc(ip); if (ret) - return ret; + goto out_write_access; gfs2_size_hint(vma->vm_file, pos, PAGE_CACHE_SIZE); gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); ret = gfs2_glock_nq(&gh); if (ret) - goto out; + goto out_uninit; set_bit(GLF_DIRTY, &ip->i_gl->gl_flags); set_bit(GIF_SW_PAGED, &ip->i_flags); @@ -480,12 +477,15 @@ out_quota_unlock: gfs2_quota_unlock(ip); out_unlock: gfs2_glock_dq(&gh); -out: +out_uninit: gfs2_holder_uninit(&gh); if (ret == 0) { set_page_dirty(page); wait_for_stable_page(page); } +out_write_access: + put_write_access(inode); +out: sb_end_pagefault(inode->i_sb); return block_page_mkwrite_return(ret); } @@ -594,10 +594,10 @@ static int gfs2_release(struct inode *inode, struct file *file) kfree(file->private_data); file->private_data = NULL; - if ((file->f_mode & FMODE_WRITE) && - (atomic_read(&inode->i_writecount) == 1)) - gfs2_rs_delete(ip); + if (!(file->f_mode & FMODE_WRITE)) + return 0; + gfs2_rs_delete(ip); return 0; } @@ -1041,7 +1041,7 @@ const struct file_operations gfs2_file_fops = { }; const struct file_operations gfs2_dir_fops = { - .readdir = gfs2_readdir, + .iterate = gfs2_readdir, .unlocked_ioctl = gfs2_ioctl, .open = gfs2_open, .release = gfs2_release, @@ -1071,7 +1071,7 @@ const struct file_operations gfs2_file_fops_nolock = { }; const struct file_operations gfs2_dir_fops_nolock = { - .readdir = gfs2_readdir, + .iterate = gfs2_readdir, .unlocked_ioctl = gfs2_ioctl, .open = gfs2_open, .release = gfs2_release, diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 8833a4f264e3..62b484e4a9e4 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -189,6 +189,7 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type, return inode; fail_refresh: + ip->i_iopen_gh.gh_flags |= GL_NOCACHE; ip->i_iopen_gh.gh_gl->gl_object = NULL; gfs2_glock_dq_uninit(&ip->i_iopen_gh); fail_iopen: diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index c5fa758fd844..6c33d7b6e0c4 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -212,7 +212,7 @@ static void gfs2_end_log_write(struct bio *bio, int error) fs_err(sdp, "Error %d writing to log\n", error); } - bio_for_each_segment(bvec, bio, i) { + bio_for_each_segment_all(bvec, bio, i) { page = bvec->bv_page; if (page_has_buffers(page)) gfs2_end_log_write_bh(sdp, bvec, error); @@ -419,7 +419,9 @@ static void gfs2_before_commit(struct gfs2_sbd *sdp, unsigned int limit, if (total > limit) num = limit; gfs2_log_unlock(sdp); - page = gfs2_get_log_desc(sdp, GFS2_LOG_DESC_METADATA, num + 1, num); + page = gfs2_get_log_desc(sdp, + is_databuf ? GFS2_LOG_DESC_JDATA : + GFS2_LOG_DESC_METADATA, num + 1, num); ld = page_address(page); gfs2_log_lock(sdp); ptr = (__be64 *)(ld + 1); diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index c7c840e916f8..c253b13722e8 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -121,7 +121,7 @@ static u64 qd2index(struct gfs2_quota_data *qd) { struct kqid qid = qd->qd_id; return (2 * (u64)from_kqid(&init_user_ns, qid)) + - (qid.type == USRQUOTA) ? 0 : 1; + ((qid.type == USRQUOTA) ? 0 : 1); } static u64 qd2offset(struct gfs2_quota_data *qd) @@ -721,7 +721,7 @@ get_a_page: goto unlock_out; } - gfs2_trans_add_meta(ip->i_gl, bh); + gfs2_trans_add_data(ip->i_gl, bh); kaddr = kmap_atomic(page); if (offset + sizeof(struct gfs2_quota) > PAGE_CACHE_SIZE) diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 0c5a575b513e..9809156e3d04 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -638,8 +638,10 @@ void gfs2_rs_deltree(struct gfs2_blkreserv *rs) */ void gfs2_rs_delete(struct gfs2_inode *ip) { + struct inode *inode = &ip->i_inode; + down_write(&ip->i_rw_mutex); - if (ip->i_res) { + if (ip->i_res && atomic_read(&inode->i_writecount) <= 1) { gfs2_rs_deltree(ip->i_res); BUG_ON(ip->i_res->rs_free); kmem_cache_free(gfs2_rsrv_cachep, ip->i_res); @@ -1401,9 +1403,14 @@ static void rg_mblk_search(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip, u32 extlen; u32 free_blocks = rgd->rd_free_clone - rgd->rd_reserved; int ret; + struct inode *inode = &ip->i_inode; - extlen = max_t(u32, atomic_read(&rs->rs_sizehint), requested); - extlen = clamp(extlen, RGRP_RSRV_MINBLKS, free_blocks); + if (S_ISDIR(inode->i_mode)) + extlen = 1; + else { + extlen = max_t(u32, atomic_read(&rs->rs_sizehint), requested); + extlen = clamp(extlen, RGRP_RSRV_MINBLKS, free_blocks); + } if ((rgd->rd_free_clone < rgd->rd_reserved) || (free_blocks < extlen)) return; diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 917c8e1eb4ae..e5639dec66c4 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -1444,6 +1444,7 @@ static void gfs2_evict_inode(struct inode *inode) /* Must not read inode block until block type has been verified */ error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_SKIP, &gh); if (unlikely(error)) { + ip->i_iopen_gh.gh_flags |= GL_NOCACHE; gfs2_glock_dq_uninit(&ip->i_iopen_gh); goto out; } @@ -1514,8 +1515,10 @@ out_unlock: if (gfs2_rs_active(ip->i_res)) gfs2_rs_deltree(ip->i_res); - if (test_bit(HIF_HOLDER, &ip->i_iopen_gh.gh_iflags)) + if (test_bit(HIF_HOLDER, &ip->i_iopen_gh.gh_iflags)) { + ip->i_iopen_gh.gh_flags |= GL_NOCACHE; gfs2_glock_dq(&ip->i_iopen_gh); + } gfs2_holder_uninit(&ip->i_iopen_gh); gfs2_glock_dq_uninit(&gh); if (error && error != GLR_TRYFAILED && error != -EROFS) @@ -1534,6 +1537,7 @@ out: ip->i_gl = NULL; if (ip->i_iopen_gh.gh_gl) { ip->i_iopen_gh.gh_gl->gl_object = NULL; + ip->i_iopen_gh.gh_flags |= GL_NOCACHE; gfs2_glock_dq_uninit(&ip->i_iopen_gh); } } diff --git a/fs/hfs/bnode.c b/fs/hfs/bnode.c index f3b1a15ccd59..d3fa6bd9503e 100644 --- a/fs/hfs/bnode.c +++ b/fs/hfs/bnode.c @@ -415,7 +415,11 @@ struct hfs_bnode *hfs_bnode_create(struct hfs_btree *tree, u32 num) spin_lock(&tree->hash_lock); node = hfs_bnode_findhash(tree, num); spin_unlock(&tree->hash_lock); - BUG_ON(node); + if (node) { + pr_crit("new node %u already hashed?\n", num); + WARN_ON(1); + return node; + } node = __hfs_bnode_create(tree, num); if (!node) return ERR_PTR(-ENOMEM); diff --git a/fs/hfs/dir.c b/fs/hfs/dir.c index e0101b6fb0d7..145566851e7a 100644 --- a/fs/hfs/dir.c +++ b/fs/hfs/dir.c @@ -51,9 +51,9 @@ done: /* * hfs_readdir */ -static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int hfs_readdir(struct file *file, struct dir_context *ctx) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; int len, err; char strbuf[HFS_MAX_NAMELEN]; @@ -62,7 +62,7 @@ static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir) struct hfs_readdir_data *rd; u16 type; - if (filp->f_pos >= inode->i_size) + if (ctx->pos >= inode->i_size) return 0; err = hfs_find_init(HFS_SB(sb)->cat_tree, &fd); @@ -73,14 +73,13 @@ static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (err) goto out; - switch ((u32)filp->f_pos) { - case 0: + if (ctx->pos == 0) { /* This is completely artificial... */ - if (filldir(dirent, ".", 1, 0, inode->i_ino, DT_DIR)) + if (!dir_emit_dot(file, ctx)) goto out; - filp->f_pos++; - /* fall through */ - case 1: + ctx->pos = 1; + } + if (ctx->pos == 1) { if (fd.entrylength > sizeof(entry) || fd.entrylength < 0) { err = -EIO; goto out; @@ -97,18 +96,16 @@ static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir) // err = -EIO; // goto out; //} - if (filldir(dirent, "..", 2, 1, + if (!dir_emit(ctx, "..", 2, be32_to_cpu(entry.thread.ParID), DT_DIR)) goto out; - filp->f_pos++; - /* fall through */ - default: - if (filp->f_pos >= inode->i_size) - goto out; - err = hfs_brec_goto(&fd, filp->f_pos - 1); - if (err) - goto out; + ctx->pos = 2; } + if (ctx->pos >= inode->i_size) + goto out; + err = hfs_brec_goto(&fd, ctx->pos - 1); + if (err) + goto out; for (;;) { if (be32_to_cpu(fd.key->cat.ParID) != inode->i_ino) { @@ -131,7 +128,7 @@ static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir) err = -EIO; goto out; } - if (filldir(dirent, strbuf, len, filp->f_pos, + if (!dir_emit(ctx, strbuf, len, be32_to_cpu(entry.dir.DirID), DT_DIR)) break; } else if (type == HFS_CDR_FIL) { @@ -140,7 +137,7 @@ static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir) err = -EIO; goto out; } - if (filldir(dirent, strbuf, len, filp->f_pos, + if (!dir_emit(ctx, strbuf, len, be32_to_cpu(entry.file.FlNum), DT_REG)) break; } else { @@ -148,22 +145,22 @@ static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir) err = -EIO; goto out; } - filp->f_pos++; - if (filp->f_pos >= inode->i_size) + ctx->pos++; + if (ctx->pos >= inode->i_size) goto out; err = hfs_brec_goto(&fd, 1); if (err) goto out; } - rd = filp->private_data; + rd = file->private_data; if (!rd) { rd = kmalloc(sizeof(struct hfs_readdir_data), GFP_KERNEL); if (!rd) { err = -ENOMEM; goto out; } - filp->private_data = rd; - rd->file = filp; + file->private_data = rd; + rd->file = file; list_add(&rd->list, &HFS_I(inode)->open_dir_list); } memcpy(&rd->key, &fd.key, sizeof(struct hfs_cat_key)); @@ -306,7 +303,7 @@ static int hfs_rename(struct inode *old_dir, struct dentry *old_dentry, const struct file_operations hfs_dir_operations = { .read = generic_read_dir, - .readdir = hfs_readdir, + .iterate = hfs_readdir, .llseek = generic_file_llseek, .release = hfs_dir_release, }; diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c index a37ac934732f..d8ce4bd17fc5 100644 --- a/fs/hfsplus/dir.c +++ b/fs/hfsplus/dir.c @@ -121,9 +121,9 @@ fail: return ERR_PTR(err); } -static int hfsplus_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int hfsplus_readdir(struct file *file, struct dir_context *ctx) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; int len, err; char strbuf[HFSPLUS_MAX_STRLEN + 1]; @@ -132,7 +132,7 @@ static int hfsplus_readdir(struct file *filp, void *dirent, filldir_t filldir) struct hfsplus_readdir_data *rd; u16 type; - if (filp->f_pos >= inode->i_size) + if (file->f_pos >= inode->i_size) return 0; err = hfs_find_init(HFSPLUS_SB(sb)->cat_tree, &fd); @@ -143,14 +143,13 @@ static int hfsplus_readdir(struct file *filp, void *dirent, filldir_t filldir) if (err) goto out; - switch ((u32)filp->f_pos) { - case 0: + if (ctx->pos == 0) { /* This is completely artificial... */ - if (filldir(dirent, ".", 1, 0, inode->i_ino, DT_DIR)) + if (!dir_emit_dot(file, ctx)) goto out; - filp->f_pos++; - /* fall through */ - case 1: + ctx->pos = 1; + } + if (ctx->pos == 1) { if (fd.entrylength > sizeof(entry) || fd.entrylength < 0) { err = -EIO; goto out; @@ -168,19 +167,16 @@ static int hfsplus_readdir(struct file *filp, void *dirent, filldir_t filldir) err = -EIO; goto out; } - if (filldir(dirent, "..", 2, 1, + if (!dir_emit(ctx, "..", 2, be32_to_cpu(entry.thread.parentID), DT_DIR)) goto out; - filp->f_pos++; - /* fall through */ - default: - if (filp->f_pos >= inode->i_size) - goto out; - err = hfs_brec_goto(&fd, filp->f_pos - 1); - if (err) - goto out; + ctx->pos = 2; } - + if (ctx->pos >= inode->i_size) + goto out; + err = hfs_brec_goto(&fd, ctx->pos - 1); + if (err) + goto out; for (;;) { if (be32_to_cpu(fd.key->cat.parent) != inode->i_ino) { pr_err("walked past end of dir\n"); @@ -211,7 +207,7 @@ static int hfsplus_readdir(struct file *filp, void *dirent, filldir_t filldir) HFSPLUS_SB(sb)->hidden_dir->i_ino == be32_to_cpu(entry.folder.id)) goto next; - if (filldir(dirent, strbuf, len, filp->f_pos, + if (!dir_emit(ctx, strbuf, len, be32_to_cpu(entry.folder.id), DT_DIR)) break; } else if (type == HFSPLUS_FILE) { @@ -220,7 +216,7 @@ static int hfsplus_readdir(struct file *filp, void *dirent, filldir_t filldir) err = -EIO; goto out; } - if (filldir(dirent, strbuf, len, filp->f_pos, + if (!dir_emit(ctx, strbuf, len, be32_to_cpu(entry.file.id), DT_REG)) break; } else { @@ -229,22 +225,22 @@ static int hfsplus_readdir(struct file *filp, void *dirent, filldir_t filldir) goto out; } next: - filp->f_pos++; - if (filp->f_pos >= inode->i_size) + ctx->pos++; + if (ctx->pos >= inode->i_size) goto out; err = hfs_brec_goto(&fd, 1); if (err) goto out; } - rd = filp->private_data; + rd = file->private_data; if (!rd) { rd = kmalloc(sizeof(struct hfsplus_readdir_data), GFP_KERNEL); if (!rd) { err = -ENOMEM; goto out; } - filp->private_data = rd; - rd->file = filp; + file->private_data = rd; + rd->file = file; list_add(&rd->list, &HFSPLUS_I(inode)->open_dir_list); } memcpy(&rd->key, fd.key, sizeof(struct hfsplus_cat_key)); @@ -538,7 +534,7 @@ const struct inode_operations hfsplus_dir_inode_operations = { const struct file_operations hfsplus_dir_operations = { .fsync = hfsplus_file_fsync, .read = generic_read_dir, - .readdir = hfsplus_readdir, + .iterate = hfsplus_readdir, .unlocked_ioctl = hfsplus_ioctl, .llseek = generic_file_llseek, .release = hfsplus_dir_release, diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index 32f35f187989..cddb05217512 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -277,7 +277,7 @@ static const struct super_operations hostfs_sbops = { .show_options = hostfs_show_options, }; -int hostfs_readdir(struct file *file, void *ent, filldir_t filldir) +int hostfs_readdir(struct file *file, struct dir_context *ctx) { void *dir; char *name; @@ -292,12 +292,11 @@ int hostfs_readdir(struct file *file, void *ent, filldir_t filldir) __putname(name); if (dir == NULL) return -error; - next = file->f_pos; + next = ctx->pos; while ((name = read_dir(dir, &next, &ino, &len, &type)) != NULL) { - error = (*filldir)(ent, name, len, file->f_pos, - ino, type); - if (error) break; - file->f_pos = next; + if (!dir_emit(ctx, name, len, ino, type)) + break; + ctx->pos = next; } close_dir(dir); return 0; @@ -393,7 +392,7 @@ static const struct file_operations hostfs_file_fops = { static const struct file_operations hostfs_dir_fops = { .llseek = generic_file_llseek, - .readdir = hostfs_readdir, + .iterate = hostfs_readdir, .read = generic_read_dir, }; diff --git a/fs/hpfs/dir.c b/fs/hpfs/dir.c index 546f6d39713a..292b1acb9b81 100644 --- a/fs/hpfs/dir.c +++ b/fs/hpfs/dir.c @@ -33,36 +33,38 @@ static loff_t hpfs_dir_lseek(struct file *filp, loff_t off, int whence) if (whence == SEEK_DATA || whence == SEEK_HOLE) return -EINVAL; + mutex_lock(&i->i_mutex); hpfs_lock(s); /*printk("dir lseek\n");*/ if (new_off == 0 || new_off == 1 || new_off == 11 || new_off == 12 || new_off == 13) goto ok; - mutex_lock(&i->i_mutex); pos = ((loff_t) hpfs_de_as_down_as_possible(s, hpfs_inode->i_dno) << 4) + 1; while (pos != new_off) { if (map_pos_dirent(i, &pos, &qbh)) hpfs_brelse4(&qbh); else goto fail; if (pos == 12) goto fail; } - mutex_unlock(&i->i_mutex); + hpfs_add_pos(i, &filp->f_pos); ok: + filp->f_pos = new_off; hpfs_unlock(s); - return filp->f_pos = new_off; -fail: mutex_unlock(&i->i_mutex); + return new_off; +fail: /*printk("illegal lseek: %016llx\n", new_off);*/ hpfs_unlock(s); + mutex_unlock(&i->i_mutex); return -ESPIPE; } -static int hpfs_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int hpfs_readdir(struct file *file, struct dir_context *ctx) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct hpfs_inode_info *hpfs_inode = hpfs_i(inode); struct quad_buffer_head qbh; struct hpfs_dirent *de; int lc; - long old_pos; + loff_t next_pos; unsigned char *tempname; int c1, c2 = 0; int ret = 0; @@ -103,11 +105,11 @@ static int hpfs_readdir(struct file *filp, void *dirent, filldir_t filldir) } } lc = hpfs_sb(inode->i_sb)->sb_lowercase; - if (filp->f_pos == 12) { /* diff -r requires this (note, that diff -r */ - filp->f_pos = 13; /* also fails on msdos filesystem in 2.0) */ + if (ctx->pos == 12) { /* diff -r requires this (note, that diff -r */ + ctx->pos = 13; /* also fails on msdos filesystem in 2.0) */ goto out; } - if (filp->f_pos == 13) { + if (ctx->pos == 13) { ret = -ENOENT; goto out; } @@ -118,33 +120,34 @@ static int hpfs_readdir(struct file *filp, void *dirent, filldir_t filldir) accepted by filldir, but what can I do? maybe killall -9 ls helps */ if (hpfs_sb(inode->i_sb)->sb_chk) - if (hpfs_stop_cycles(inode->i_sb, filp->f_pos, &c1, &c2, "hpfs_readdir")) { + if (hpfs_stop_cycles(inode->i_sb, ctx->pos, &c1, &c2, "hpfs_readdir")) { ret = -EFSERROR; goto out; } - if (filp->f_pos == 12) + if (ctx->pos == 12) goto out; - if (filp->f_pos == 3 || filp->f_pos == 4 || filp->f_pos == 5) { - printk("HPFS: warning: pos==%d\n",(int)filp->f_pos); + if (ctx->pos == 3 || ctx->pos == 4 || ctx->pos == 5) { + printk("HPFS: warning: pos==%d\n",(int)ctx->pos); goto out; } - if (filp->f_pos == 0) { - if (filldir(dirent, ".", 1, filp->f_pos, inode->i_ino, DT_DIR) < 0) + if (ctx->pos == 0) { + if (!dir_emit_dot(file, ctx)) goto out; - filp->f_pos = 11; + ctx->pos = 11; } - if (filp->f_pos == 11) { - if (filldir(dirent, "..", 2, filp->f_pos, hpfs_inode->i_parent_dir, DT_DIR) < 0) + if (ctx->pos == 11) { + if (!dir_emit(ctx, "..", 2, hpfs_inode->i_parent_dir, DT_DIR)) goto out; - filp->f_pos = 1; + ctx->pos = 1; } - if (filp->f_pos == 1) { - filp->f_pos = ((loff_t) hpfs_de_as_down_as_possible(inode->i_sb, hpfs_inode->i_dno) << 4) + 1; - hpfs_add_pos(inode, &filp->f_pos); - filp->f_version = inode->i_version; + if (ctx->pos == 1) { + ctx->pos = ((loff_t) hpfs_de_as_down_as_possible(inode->i_sb, hpfs_inode->i_dno) << 4) + 1; + hpfs_add_pos(inode, &file->f_pos); + file->f_version = inode->i_version; } - old_pos = filp->f_pos; - if (!(de = map_pos_dirent(inode, &filp->f_pos, &qbh))) { + next_pos = ctx->pos; + if (!(de = map_pos_dirent(inode, &next_pos, &qbh))) { + ctx->pos = next_pos; ret = -EIOERROR; goto out; } @@ -152,20 +155,21 @@ static int hpfs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (hpfs_sb(inode->i_sb)->sb_chk) { if (de->first && !de->last && (de->namelen != 2 || de ->name[0] != 1 || de->name[1] != 1)) - hpfs_error(inode->i_sb, "hpfs_readdir: bad ^A^A entry; pos = %08lx", old_pos); + hpfs_error(inode->i_sb, "hpfs_readdir: bad ^A^A entry; pos = %08lx", (unsigned long)ctx->pos); if (de->last && (de->namelen != 1 || de ->name[0] != 255)) - hpfs_error(inode->i_sb, "hpfs_readdir: bad \\377 entry; pos = %08lx", old_pos); + hpfs_error(inode->i_sb, "hpfs_readdir: bad \\377 entry; pos = %08lx", (unsigned long)ctx->pos); } hpfs_brelse4(&qbh); + ctx->pos = next_pos; goto again; } tempname = hpfs_translate_name(inode->i_sb, de->name, de->namelen, lc, de->not_8x3); - if (filldir(dirent, tempname, de->namelen, old_pos, le32_to_cpu(de->fnode), DT_UNKNOWN) < 0) { - filp->f_pos = old_pos; + if (!dir_emit(ctx, tempname, de->namelen, le32_to_cpu(de->fnode), DT_UNKNOWN)) { if (tempname != de->name) kfree(tempname); hpfs_brelse4(&qbh); goto out; } + ctx->pos = next_pos; if (tempname != de->name) kfree(tempname); hpfs_brelse4(&qbh); } @@ -320,7 +324,7 @@ const struct file_operations hpfs_dir_ops = { .llseek = hpfs_dir_lseek, .read = generic_read_dir, - .readdir = hpfs_readdir, + .iterate = hpfs_readdir, .release = hpfs_dir_release, .fsync = hpfs_file_fsync, }; diff --git a/fs/hpfs/file.c b/fs/hpfs/file.c index 3027f4dbbab5..e4ba5fe4c3b5 100644 --- a/fs/hpfs/file.c +++ b/fs/hpfs/file.c @@ -109,10 +109,14 @@ static void hpfs_write_failed(struct address_space *mapping, loff_t to) { struct inode *inode = mapping->host; + hpfs_lock(inode->i_sb); + if (to > inode->i_size) { truncate_pagecache(inode, to, inode->i_size); hpfs_truncate(inode); } + + hpfs_unlock(inode->i_sb); } static int hpfs_write_begin(struct file *file, struct address_space *mapping, diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c index cd3e38972c86..fc90ab11c340 100644 --- a/fs/hppfs/hppfs.c +++ b/fs/hppfs/hppfs.c @@ -542,8 +542,8 @@ static const struct file_operations hppfs_file_fops = { }; struct hppfs_dirent { - void *vfs_dirent; - filldir_t filldir; + struct dir_context ctx; + struct dir_context *caller; struct dentry *dentry; }; @@ -555,34 +555,29 @@ static int hppfs_filldir(void *d, const char *name, int size, if (file_removed(dirent->dentry, name)) return 0; - return (*dirent->filldir)(dirent->vfs_dirent, name, size, offset, - inode, type); + dirent->caller->pos = dirent->ctx.pos; + return !dir_emit(dirent->caller, name, size, inode, type); } -static int hppfs_readdir(struct file *file, void *ent, filldir_t filldir) +static int hppfs_readdir(struct file *file, struct dir_context *ctx) { struct hppfs_private *data = file->private_data; struct file *proc_file = data->proc_file; - int (*readdir)(struct file *, void *, filldir_t); - struct hppfs_dirent dirent = ((struct hppfs_dirent) - { .vfs_dirent = ent, - .filldir = filldir, - .dentry = file->f_path.dentry - }); + struct hppfs_dirent d = { + .ctx.actor = hppfs_filldir, + .caller = ctx, + .dentry = file->f_path.dentry + }; int err; - - readdir = file_inode(proc_file)->i_fop->readdir; - - proc_file->f_pos = file->f_pos; - err = (*readdir)(proc_file, &dirent, hppfs_filldir); - file->f_pos = proc_file->f_pos; - + proc_file->f_pos = ctx->pos; + err = iterate_dir(proc_file, &d.ctx); + ctx->pos = d.ctx.pos; return err; } static const struct file_operations hppfs_dir_fops = { .owner = NULL, - .readdir = hppfs_readdir, + .iterate = hppfs_readdir, .open = hppfs_dir_open, .llseek = default_llseek, .release = hppfs_release, diff --git a/fs/internal.h b/fs/internal.h index eaa75f75b625..68121584ae37 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -132,6 +132,12 @@ extern struct dentry *__d_alloc(struct super_block *, const struct qstr *); extern ssize_t __kernel_write(struct file *, const char *, size_t, loff_t *); /* + * splice.c + */ +extern long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, + loff_t *opos, size_t len, unsigned int flags); + +/* * pipe.c */ extern const struct file_operations pipefifo_fops; diff --git a/fs/isofs/dir.c b/fs/isofs/dir.c index a7d5c3c3d4e6..b943cbd963bb 100644 --- a/fs/isofs/dir.c +++ b/fs/isofs/dir.c @@ -78,8 +78,8 @@ int get_acorn_filename(struct iso_directory_record *de, /* * This should _really_ be cleaned up some day.. */ -static int do_isofs_readdir(struct inode *inode, struct file *filp, - void *dirent, filldir_t filldir, +static int do_isofs_readdir(struct inode *inode, struct file *file, + struct dir_context *ctx, char *tmpname, struct iso_directory_record *tmpde) { unsigned long bufsize = ISOFS_BUFFER_SIZE(inode); @@ -94,10 +94,10 @@ static int do_isofs_readdir(struct inode *inode, struct file *filp, struct iso_directory_record *de; struct isofs_sb_info *sbi = ISOFS_SB(inode->i_sb); - offset = filp->f_pos & (bufsize - 1); - block = filp->f_pos >> bufbits; + offset = ctx->pos & (bufsize - 1); + block = ctx->pos >> bufbits; - while (filp->f_pos < inode->i_size) { + while (ctx->pos < inode->i_size) { int de_len; if (!bh) { @@ -108,7 +108,7 @@ static int do_isofs_readdir(struct inode *inode, struct file *filp, de = (struct iso_directory_record *) (bh->b_data + offset); - de_len = *(unsigned char *) de; + de_len = *(unsigned char *)de; /* * If the length byte is zero, we should move on to the next @@ -119,8 +119,8 @@ static int do_isofs_readdir(struct inode *inode, struct file *filp, if (de_len == 0) { brelse(bh); bh = NULL; - filp->f_pos = (filp->f_pos + ISOFS_BLOCK_SIZE) & ~(ISOFS_BLOCK_SIZE - 1); - block = filp->f_pos >> bufbits; + ctx->pos = (ctx->pos + ISOFS_BLOCK_SIZE) & ~(ISOFS_BLOCK_SIZE - 1); + block = ctx->pos >> bufbits; offset = 0; continue; } @@ -164,16 +164,16 @@ static int do_isofs_readdir(struct inode *inode, struct file *filp, if (de->flags[-sbi->s_high_sierra] & 0x80) { first_de = 0; - filp->f_pos += de_len; + ctx->pos += de_len; continue; } first_de = 1; /* Handle the case of the '.' directory */ if (de->name_len[0] == 1 && de->name[0] == 0) { - if (filldir(dirent, ".", 1, filp->f_pos, inode->i_ino, DT_DIR) < 0) + if (!dir_emit_dot(file, ctx)) break; - filp->f_pos += de_len; + ctx->pos += de_len; continue; } @@ -181,10 +181,9 @@ static int do_isofs_readdir(struct inode *inode, struct file *filp, /* Handle the case of the '..' directory */ if (de->name_len[0] == 1 && de->name[0] == 1) { - inode_number = parent_ino(filp->f_path.dentry); - if (filldir(dirent, "..", 2, filp->f_pos, inode_number, DT_DIR) < 0) + if (!dir_emit_dotdot(file, ctx)) break; - filp->f_pos += de_len; + ctx->pos += de_len; continue; } @@ -198,7 +197,7 @@ static int do_isofs_readdir(struct inode *inode, struct file *filp, if ((sbi->s_hide && (de->flags[-sbi->s_high_sierra] & 1)) || (!sbi->s_showassoc && (de->flags[-sbi->s_high_sierra] & 4))) { - filp->f_pos += de_len; + ctx->pos += de_len; continue; } @@ -230,10 +229,10 @@ static int do_isofs_readdir(struct inode *inode, struct file *filp, } } if (len > 0) { - if (filldir(dirent, p, len, filp->f_pos, inode_number, DT_UNKNOWN) < 0) + if (!dir_emit(ctx, p, len, inode_number, DT_UNKNOWN)) break; } - filp->f_pos += de_len; + ctx->pos += de_len; continue; } @@ -247,13 +246,12 @@ static int do_isofs_readdir(struct inode *inode, struct file *filp, * handling split directory entries.. The real work is done by * "do_isofs_readdir()". */ -static int isofs_readdir(struct file *filp, - void *dirent, filldir_t filldir) +static int isofs_readdir(struct file *file, struct dir_context *ctx) { int result; char *tmpname; struct iso_directory_record *tmpde; - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); tmpname = (char *)__get_free_page(GFP_KERNEL); if (tmpname == NULL) @@ -261,7 +259,7 @@ static int isofs_readdir(struct file *filp, tmpde = (struct iso_directory_record *) (tmpname+1024); - result = do_isofs_readdir(inode, filp, dirent, filldir, tmpname, tmpde); + result = do_isofs_readdir(inode, file, ctx, tmpname, tmpde); free_page((unsigned long) tmpname); return result; @@ -271,7 +269,7 @@ const struct file_operations isofs_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = isofs_readdir, + .iterate = isofs_readdir, }; /* diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index acd46a4160cb..e3aac222472e 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -22,7 +22,7 @@ #include <linux/time.h> #include "nodelist.h" -static int jffs2_readdir (struct file *, void *, filldir_t); +static int jffs2_readdir (struct file *, struct dir_context *); static int jffs2_create (struct inode *,struct dentry *,umode_t, bool); @@ -40,7 +40,7 @@ static int jffs2_rename (struct inode *, struct dentry *, const struct file_operations jffs2_dir_operations = { .read = generic_read_dir, - .readdir = jffs2_readdir, + .iterate = jffs2_readdir, .unlocked_ioctl=jffs2_ioctl, .fsync = jffs2_fsync, .llseek = generic_file_llseek, @@ -114,60 +114,40 @@ static struct dentry *jffs2_lookup(struct inode *dir_i, struct dentry *target, /***********************************************************************/ -static int jffs2_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int jffs2_readdir(struct file *file, struct dir_context *ctx) { - struct jffs2_inode_info *f; - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); + struct jffs2_inode_info *f = JFFS2_INODE_INFO(inode); struct jffs2_full_dirent *fd; - unsigned long offset, curofs; + unsigned long curofs = 1; - jffs2_dbg(1, "jffs2_readdir() for dir_i #%lu\n", - file_inode(filp)->i_ino); + jffs2_dbg(1, "jffs2_readdir() for dir_i #%lu\n", inode->i_ino); - f = JFFS2_INODE_INFO(inode); - - offset = filp->f_pos; - - if (offset == 0) { - jffs2_dbg(1, "Dirent 0: \".\", ino #%lu\n", inode->i_ino); - if (filldir(dirent, ".", 1, 0, inode->i_ino, DT_DIR) < 0) - goto out; - offset++; - } - if (offset == 1) { - unsigned long pino = parent_ino(filp->f_path.dentry); - jffs2_dbg(1, "Dirent 1: \"..\", ino #%lu\n", pino); - if (filldir(dirent, "..", 2, 1, pino, DT_DIR) < 0) - goto out; - offset++; - } + if (!dir_emit_dots(file, ctx)) + return 0; - curofs=1; mutex_lock(&f->sem); for (fd = f->dents; fd; fd = fd->next) { - curofs++; - /* First loop: curofs = 2; offset = 2 */ - if (curofs < offset) { + /* First loop: curofs = 2; pos = 2 */ + if (curofs < ctx->pos) { jffs2_dbg(2, "Skipping dirent: \"%s\", ino #%u, type %d, because curofs %ld < offset %ld\n", - fd->name, fd->ino, fd->type, curofs, offset); + fd->name, fd->ino, fd->type, curofs, (unsigned long)ctx->pos); continue; } if (!fd->ino) { jffs2_dbg(2, "Skipping deletion dirent \"%s\"\n", fd->name); - offset++; + ctx->pos++; continue; } jffs2_dbg(2, "Dirent %ld: \"%s\", ino #%u, type %d\n", - offset, fd->name, fd->ino, fd->type); - if (filldir(dirent, fd->name, strlen(fd->name), offset, fd->ino, fd->type) < 0) + (unsigned long)ctx->pos, fd->name, fd->ino, fd->type); + if (!dir_emit(ctx, fd->name, strlen(fd->name), fd->ino, fd->type)) break; - offset++; + ctx->pos++; } mutex_unlock(&f->sem); - out: - filp->f_pos = offset; return 0; } diff --git a/fs/jfs/jfs_dtree.c b/fs/jfs/jfs_dtree.c index 0ddbeceafc62..9f4ed13d9f15 100644 --- a/fs/jfs/jfs_dtree.c +++ b/fs/jfs/jfs_dtree.c @@ -3002,9 +3002,9 @@ static inline struct jfs_dirent *next_jfs_dirent(struct jfs_dirent *dirent) * return: offset = (pn, index) of start entry * of next jfs_readdir()/dtRead() */ -int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir) +int jfs_readdir(struct file *file, struct dir_context *ctx) { - struct inode *ip = file_inode(filp); + struct inode *ip = file_inode(file); struct nls_table *codepage = JFS_SBI(ip->i_sb)->nls_tab; int rc = 0; loff_t dtpos; /* legacy OS/2 style position */ @@ -3033,7 +3033,7 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir) int overflow, fix_page, page_fixed = 0; static int unique_pos = 2; /* If we can't fix broken index */ - if (filp->f_pos == DIREND) + if (ctx->pos == DIREND) return 0; if (DO_INDEX(ip)) { @@ -3045,7 +3045,7 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir) */ do_index = 1; - dir_index = (u32) filp->f_pos; + dir_index = (u32) ctx->pos; if (dir_index > 1) { struct dir_table_slot dirtab_slot; @@ -3053,25 +3053,25 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (dtEmpty(ip) || (dir_index >= JFS_IP(ip)->next_index)) { /* Stale position. Directory has shrunk */ - filp->f_pos = DIREND; + ctx->pos = DIREND; return 0; } repeat: rc = read_index(ip, dir_index, &dirtab_slot); if (rc) { - filp->f_pos = DIREND; + ctx->pos = DIREND; return rc; } if (dirtab_slot.flag == DIR_INDEX_FREE) { if (loop_count++ > JFS_IP(ip)->next_index) { jfs_err("jfs_readdir detected " "infinite loop!"); - filp->f_pos = DIREND; + ctx->pos = DIREND; return 0; } dir_index = le32_to_cpu(dirtab_slot.addr2); if (dir_index == -1) { - filp->f_pos = DIREND; + ctx->pos = DIREND; return 0; } goto repeat; @@ -3080,13 +3080,13 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir) index = dirtab_slot.slot; DT_GETPAGE(ip, bn, mp, PSIZE, p, rc); if (rc) { - filp->f_pos = DIREND; + ctx->pos = DIREND; return 0; } if (p->header.flag & BT_INTERNAL) { jfs_err("jfs_readdir: bad index table"); DT_PUTPAGE(mp); - filp->f_pos = -1; + ctx->pos = -1; return 0; } } else { @@ -3094,23 +3094,22 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir) /* * self "." */ - filp->f_pos = 0; - if (filldir(dirent, ".", 1, 0, ip->i_ino, - DT_DIR)) + ctx->pos = 0; + if (!dir_emit(ctx, ".", 1, ip->i_ino, DT_DIR)) return 0; } /* * parent ".." */ - filp->f_pos = 1; - if (filldir(dirent, "..", 2, 1, PARENT(ip), DT_DIR)) + ctx->pos = 1; + if (!dir_emit(ctx, "..", 2, PARENT(ip), DT_DIR)) return 0; /* * Find first entry of left-most leaf */ if (dtEmpty(ip)) { - filp->f_pos = DIREND; + ctx->pos = DIREND; return 0; } @@ -3128,23 +3127,19 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir) * pn > 0: Real entries, pn=1 -> leftmost page * pn = index = -1: No more entries */ - dtpos = filp->f_pos; + dtpos = ctx->pos; if (dtpos == 0) { /* build "." entry */ - - if (filldir(dirent, ".", 1, filp->f_pos, ip->i_ino, - DT_DIR)) + if (!dir_emit(ctx, ".", 1, ip->i_ino, DT_DIR)) return 0; dtoffset->index = 1; - filp->f_pos = dtpos; + ctx->pos = dtpos; } if (dtoffset->pn == 0) { if (dtoffset->index == 1) { /* build ".." entry */ - - if (filldir(dirent, "..", 2, filp->f_pos, - PARENT(ip), DT_DIR)) + if (!dir_emit(ctx, "..", 2, PARENT(ip), DT_DIR)) return 0; } else { jfs_err("jfs_readdir called with " @@ -3152,18 +3147,18 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir) } dtoffset->pn = 1; dtoffset->index = 0; - filp->f_pos = dtpos; + ctx->pos = dtpos; } if (dtEmpty(ip)) { - filp->f_pos = DIREND; + ctx->pos = DIREND; return 0; } - if ((rc = dtReadNext(ip, &filp->f_pos, &btstack))) { + if ((rc = dtReadNext(ip, &ctx->pos, &btstack))) { jfs_err("jfs_readdir: unexpected rc = %d " "from dtReadNext", rc); - filp->f_pos = DIREND; + ctx->pos = DIREND; return 0; } /* get start leaf page and index */ @@ -3171,7 +3166,7 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir) /* offset beyond directory eof ? */ if (bn < 0) { - filp->f_pos = DIREND; + ctx->pos = DIREND; return 0; } } @@ -3180,7 +3175,7 @@ int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (dirent_buf == 0) { DT_PUTPAGE(mp); jfs_warn("jfs_readdir: __get_free_page failed!"); - filp->f_pos = DIREND; + ctx->pos = DIREND; return -ENOMEM; } @@ -3295,9 +3290,9 @@ skip_one: jfs_dirent = (struct jfs_dirent *) dirent_buf; while (jfs_dirents--) { - filp->f_pos = jfs_dirent->position; - if (filldir(dirent, jfs_dirent->name, - jfs_dirent->name_len, filp->f_pos, + ctx->pos = jfs_dirent->position; + if (!dir_emit(ctx, jfs_dirent->name, + jfs_dirent->name_len, jfs_dirent->ino, DT_UNKNOWN)) goto out; jfs_dirent = next_jfs_dirent(jfs_dirent); @@ -3309,7 +3304,7 @@ skip_one: } if (!overflow && (bn == 0)) { - filp->f_pos = DIREND; + ctx->pos = DIREND; break; } diff --git a/fs/jfs/jfs_dtree.h b/fs/jfs/jfs_dtree.h index 2545bb317235..fd4169e6e698 100644 --- a/fs/jfs/jfs_dtree.h +++ b/fs/jfs/jfs_dtree.h @@ -265,5 +265,5 @@ extern int dtDelete(tid_t tid, struct inode *ip, struct component_name * key, extern int dtModify(tid_t tid, struct inode *ip, struct component_name * key, ino_t * orig_ino, ino_t new_ino, int flag); -extern int jfs_readdir(struct file *filp, void *dirent, filldir_t filldir); +extern int jfs_readdir(struct file *file, struct dir_context *ctx); #endif /* !_H_JFS_DTREE */ diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c index c57499dca89c..360d27c48887 100644 --- a/fs/jfs/jfs_logmgr.c +++ b/fs/jfs/jfs_logmgr.c @@ -2009,7 +2009,13 @@ static int lbmRead(struct jfs_log * log, int pn, struct lbuf ** bpp) bio->bi_end_io = lbmIODone; bio->bi_private = bp; - submit_bio(READ_SYNC, bio); + /*check if journaling to disk has been disabled*/ + if (log->no_integrity) { + bio->bi_size = 0; + lbmIODone(bio, 0); + } else { + submit_bio(READ_SYNC, bio); + } wait_event(bp->l_ioevent, (bp->l_flag != lbmREAD)); diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c index 3b91a7ad6086..89186b7b9002 100644 --- a/fs/jfs/namei.c +++ b/fs/jfs/namei.c @@ -1529,7 +1529,7 @@ const struct inode_operations jfs_dir_inode_operations = { const struct file_operations jfs_dir_operations = { .read = generic_read_dir, - .readdir = jfs_readdir, + .iterate = jfs_readdir, .fsync = jfs_fsync, .unlocked_ioctl = jfs_ioctl, #ifdef CONFIG_COMPAT diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 2003e830ed1c..788e0a9c1fb0 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -611,11 +611,28 @@ static int jfs_freeze(struct super_block *sb) { struct jfs_sb_info *sbi = JFS_SBI(sb); struct jfs_log *log = sbi->log; + int rc = 0; if (!(sb->s_flags & MS_RDONLY)) { txQuiesce(sb); - lmLogShutdown(log); - updateSuper(sb, FM_CLEAN); + rc = lmLogShutdown(log); + if (rc) { + jfs_error(sb, "jfs_freeze: lmLogShutdown failed"); + + /* let operations fail rather than hang */ + txResume(sb); + + return rc; + } + rc = updateSuper(sb, FM_CLEAN); + if (rc) { + jfs_err("jfs_freeze: updateSuper failed\n"); + /* + * Don't fail here. Everything succeeded except + * marking the superblock clean, so there's really + * no harm in leaving it frozen for now. + */ + } } return 0; } @@ -627,13 +644,18 @@ static int jfs_unfreeze(struct super_block *sb) int rc = 0; if (!(sb->s_flags & MS_RDONLY)) { - updateSuper(sb, FM_MOUNT); - if ((rc = lmLogInit(log))) - jfs_err("jfs_unlock failed with return code %d", rc); - else - txResume(sb); + rc = updateSuper(sb, FM_MOUNT); + if (rc) { + jfs_error(sb, "jfs_unfreeze: updateSuper failed"); + goto out; + } + rc = lmLogInit(log); + if (rc) + jfs_error(sb, "jfs_unfreeze: lmLogInit failed"); +out: + txResume(sb); } - return 0; + return rc; } static struct dentry *jfs_do_mount(struct file_system_type *fs_type, diff --git a/fs/libfs.c b/fs/libfs.c index 916da8c4158b..c3a0837fb861 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -135,60 +135,40 @@ static inline unsigned char dt_type(struct inode *inode) * both impossible due to the lock on directory. */ -int dcache_readdir(struct file * filp, void * dirent, filldir_t filldir) +int dcache_readdir(struct file *file, struct dir_context *ctx) { - struct dentry *dentry = filp->f_path.dentry; - struct dentry *cursor = filp->private_data; + struct dentry *dentry = file->f_path.dentry; + struct dentry *cursor = file->private_data; struct list_head *p, *q = &cursor->d_u.d_child; - ino_t ino; - int i = filp->f_pos; - switch (i) { - case 0: - ino = dentry->d_inode->i_ino; - if (filldir(dirent, ".", 1, i, ino, DT_DIR) < 0) - break; - filp->f_pos++; - i++; - /* fallthrough */ - case 1: - ino = parent_ino(dentry); - if (filldir(dirent, "..", 2, i, ino, DT_DIR) < 0) - break; - filp->f_pos++; - i++; - /* fallthrough */ - default: - spin_lock(&dentry->d_lock); - if (filp->f_pos == 2) - list_move(q, &dentry->d_subdirs); - - for (p=q->next; p != &dentry->d_subdirs; p=p->next) { - struct dentry *next; - next = list_entry(p, struct dentry, d_u.d_child); - spin_lock_nested(&next->d_lock, DENTRY_D_LOCK_NESTED); - if (!simple_positive(next)) { - spin_unlock(&next->d_lock); - continue; - } + if (!dir_emit_dots(file, ctx)) + return 0; + spin_lock(&dentry->d_lock); + if (ctx->pos == 2) + list_move(q, &dentry->d_subdirs); + + for (p = q->next; p != &dentry->d_subdirs; p = p->next) { + struct dentry *next = list_entry(p, struct dentry, d_u.d_child); + spin_lock_nested(&next->d_lock, DENTRY_D_LOCK_NESTED); + if (!simple_positive(next)) { + spin_unlock(&next->d_lock); + continue; + } - spin_unlock(&next->d_lock); - spin_unlock(&dentry->d_lock); - if (filldir(dirent, next->d_name.name, - next->d_name.len, filp->f_pos, - next->d_inode->i_ino, - dt_type(next->d_inode)) < 0) - return 0; - spin_lock(&dentry->d_lock); - spin_lock_nested(&next->d_lock, DENTRY_D_LOCK_NESTED); - /* next is still alive */ - list_move(q, p); - spin_unlock(&next->d_lock); - p = q; - filp->f_pos++; - } - spin_unlock(&dentry->d_lock); + spin_unlock(&next->d_lock); + spin_unlock(&dentry->d_lock); + if (!dir_emit(ctx, next->d_name.name, next->d_name.len, + next->d_inode->i_ino, dt_type(next->d_inode))) + return 0; + spin_lock(&dentry->d_lock); + spin_lock_nested(&next->d_lock, DENTRY_D_LOCK_NESTED); + /* next is still alive */ + list_move(q, p); + spin_unlock(&next->d_lock); + p = q; + ctx->pos++; } + spin_unlock(&dentry->d_lock); return 0; } @@ -202,7 +182,7 @@ const struct file_operations simple_dir_operations = { .release = dcache_dir_close, .llseek = dcache_dir_lseek, .read = generic_read_dir, - .readdir = dcache_readdir, + .iterate = dcache_readdir, .fsync = noop_fsync, }; diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c index b82751082112..6bdc347008f5 100644 --- a/fs/logfs/dir.c +++ b/fs/logfs/dir.c @@ -281,17 +281,23 @@ static int logfs_rmdir(struct inode *dir, struct dentry *dentry) /* FIXME: readdir currently has it's own dir_walk code. I don't see a good * way to combine the two copies */ -#define IMPLICIT_NODES 2 -static int __logfs_readdir(struct file *file, void *buf, filldir_t filldir) +static int logfs_readdir(struct file *file, struct dir_context *ctx) { struct inode *dir = file_inode(file); - loff_t pos = file->f_pos - IMPLICIT_NODES; + loff_t pos; struct page *page; struct logfs_disk_dentry *dd; - int full; + if (ctx->pos < 0) + return -EINVAL; + + if (!dir_emit_dots(file, ctx)) + return 0; + + pos = ctx->pos - 2; BUG_ON(pos < 0); - for (;; pos++) { + for (;; pos++, ctx->pos++) { + bool full; if (beyond_eof(dir, pos)) break; if (!logfs_exist_block(dir, pos)) { @@ -306,42 +312,17 @@ static int __logfs_readdir(struct file *file, void *buf, filldir_t filldir) dd = kmap(page); BUG_ON(dd->namelen == 0); - full = filldir(buf, (char *)dd->name, be16_to_cpu(dd->namelen), - pos, be64_to_cpu(dd->ino), dd->type); + full = !dir_emit(ctx, (char *)dd->name, + be16_to_cpu(dd->namelen), + be64_to_cpu(dd->ino), dd->type); kunmap(page); page_cache_release(page); if (full) break; } - - file->f_pos = pos + IMPLICIT_NODES; return 0; } -static int logfs_readdir(struct file *file, void *buf, filldir_t filldir) -{ - struct inode *inode = file_inode(file); - ino_t pino = parent_ino(file->f_dentry); - int err; - - if (file->f_pos < 0) - return -EINVAL; - - if (file->f_pos == 0) { - if (filldir(buf, ".", 1, 1, inode->i_ino, DT_DIR) < 0) - return 0; - file->f_pos++; - } - if (file->f_pos == 1) { - if (filldir(buf, "..", 2, 2, pino, DT_DIR) < 0) - return 0; - file->f_pos++; - } - - err = __logfs_readdir(file, buf, filldir); - return err; -} - static void logfs_set_name(struct logfs_disk_dentry *dd, struct qstr *name) { dd->namelen = cpu_to_be16(name->len); @@ -814,7 +795,7 @@ const struct inode_operations logfs_dir_iops = { const struct file_operations logfs_dir_fops = { .fsync = logfs_fsync, .unlocked_ioctl = logfs_ioctl, - .readdir = logfs_readdir, + .iterate = logfs_readdir, .read = generic_read_dir, .llseek = default_llseek, }; diff --git a/fs/minix/dir.c b/fs/minix/dir.c index a9ed6f36e6ea..08c442902fcd 100644 --- a/fs/minix/dir.c +++ b/fs/minix/dir.c @@ -16,12 +16,12 @@ typedef struct minix_dir_entry minix_dirent; typedef struct minix3_dir_entry minix3_dirent; -static int minix_readdir(struct file *, void *, filldir_t); +static int minix_readdir(struct file *, struct dir_context *); const struct file_operations minix_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = minix_readdir, + .iterate = minix_readdir, .fsync = generic_file_fsync, }; @@ -82,22 +82,23 @@ static inline void *minix_next_entry(void *de, struct minix_sb_info *sbi) return (void*)((char*)de + sbi->s_dirsize); } -static int minix_readdir(struct file * filp, void * dirent, filldir_t filldir) +static int minix_readdir(struct file *file, struct dir_context *ctx) { - unsigned long pos = filp->f_pos; - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; - unsigned offset = pos & ~PAGE_CACHE_MASK; - unsigned long n = pos >> PAGE_CACHE_SHIFT; - unsigned long npages = dir_pages(inode); struct minix_sb_info *sbi = minix_sb(sb); unsigned chunk_size = sbi->s_dirsize; - char *name; - __u32 inumber; + unsigned long npages = dir_pages(inode); + unsigned long pos = ctx->pos; + unsigned offset; + unsigned long n; - pos = (pos + chunk_size-1) & ~(chunk_size-1); + ctx->pos = pos = (pos + chunk_size-1) & ~(chunk_size-1); if (pos >= inode->i_size) - goto done; + return 0; + + offset = pos & ~PAGE_CACHE_MASK; + n = pos >> PAGE_CACHE_SHIFT; for ( ; n < npages; n++, offset = 0) { char *p, *kaddr, *limit; @@ -109,6 +110,8 @@ static int minix_readdir(struct file * filp, void * dirent, filldir_t filldir) p = kaddr+offset; limit = kaddr + minix_last_byte(inode, n) - chunk_size; for ( ; p <= limit; p = minix_next_entry(p, sbi)) { + const char *name; + __u32 inumber; if (sbi->s_version == MINIX_V3) { minix3_dirent *de3 = (minix3_dirent *)p; name = de3->name; @@ -119,24 +122,17 @@ static int minix_readdir(struct file * filp, void * dirent, filldir_t filldir) inumber = de->inode; } if (inumber) { - int over; - unsigned l = strnlen(name, sbi->s_namelen); - offset = p - kaddr; - over = filldir(dirent, name, l, - (n << PAGE_CACHE_SHIFT) | offset, - inumber, DT_UNKNOWN); - if (over) { + if (!dir_emit(ctx, name, l, + inumber, DT_UNKNOWN)) { dir_put_page(page); - goto done; + return 0; } } + ctx->pos += chunk_size; } dir_put_page(page); } - -done: - filp->f_pos = (n << PAGE_CACHE_SHIFT) | offset; return 0; } diff --git a/fs/namei.c b/fs/namei.c index 85e40d1c0a8f..9ed9361223c0 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1976,7 +1976,7 @@ static int path_lookupat(int dfd, const char *name, err = complete_walk(nd); if (!err && nd->flags & LOOKUP_DIRECTORY) { - if (!nd->inode->i_op->lookup) { + if (!can_lookup(nd->inode)) { path_put(&nd->path); err = -ENOTDIR; } @@ -2850,7 +2850,7 @@ finish_lookup: if ((open_flag & O_CREAT) && S_ISDIR(nd->inode->i_mode)) goto out; error = -ENOTDIR; - if ((nd->flags & LOOKUP_DIRECTORY) && !nd->inode->i_op->lookup) + if ((nd->flags & LOOKUP_DIRECTORY) && !can_lookup(nd->inode)) goto out; audit_inode(name, nd->path.dentry, 0); finish_open: diff --git a/fs/ncpfs/dir.c b/fs/ncpfs/dir.c index 816326093656..0e7f00298213 100644 --- a/fs/ncpfs/dir.c +++ b/fs/ncpfs/dir.c @@ -23,12 +23,12 @@ #include "ncp_fs.h" -static void ncp_read_volume_list(struct file *, void *, filldir_t, +static void ncp_read_volume_list(struct file *, struct dir_context *, struct ncp_cache_control *); -static void ncp_do_readdir(struct file *, void *, filldir_t, +static void ncp_do_readdir(struct file *, struct dir_context *, struct ncp_cache_control *); -static int ncp_readdir(struct file *, void *, filldir_t); +static int ncp_readdir(struct file *, struct dir_context *); static int ncp_create(struct inode *, struct dentry *, umode_t, bool); static struct dentry *ncp_lookup(struct inode *, struct dentry *, unsigned int); @@ -49,7 +49,7 @@ const struct file_operations ncp_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = ncp_readdir, + .iterate = ncp_readdir, .unlocked_ioctl = ncp_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ncp_compat_ioctl, @@ -424,9 +424,9 @@ static time_t ncp_obtain_mtime(struct dentry *dentry) return ncp_date_dos2unix(i.modifyTime, i.modifyDate); } -static int ncp_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int ncp_readdir(struct file *file, struct dir_context *ctx) { - struct dentry *dentry = filp->f_path.dentry; + struct dentry *dentry = file->f_path.dentry; struct inode *inode = dentry->d_inode; struct page *page = NULL; struct ncp_server *server = NCP_SERVER(inode); @@ -440,7 +440,7 @@ static int ncp_readdir(struct file *filp, void *dirent, filldir_t filldir) DDPRINTK("ncp_readdir: reading %s/%s, pos=%d\n", dentry->d_parent->d_name.name, dentry->d_name.name, - (int) filp->f_pos); + (int) ctx->pos); result = -EIO; /* Do not generate '.' and '..' when server is dead. */ @@ -448,16 +448,8 @@ static int ncp_readdir(struct file *filp, void *dirent, filldir_t filldir) goto out; result = 0; - if (filp->f_pos == 0) { - if (filldir(dirent, ".", 1, 0, inode->i_ino, DT_DIR)) - goto out; - filp->f_pos = 1; - } - if (filp->f_pos == 1) { - if (filldir(dirent, "..", 2, 1, parent_ino(dentry), DT_DIR)) - goto out; - filp->f_pos = 2; - } + if (!dir_emit_dots(file, ctx)) + goto out; page = grab_cache_page(&inode->i_data, 0); if (!page) @@ -469,7 +461,7 @@ static int ncp_readdir(struct file *filp, void *dirent, filldir_t filldir) if (!PageUptodate(page) || !ctl.head.eof) goto init_cache; - if (filp->f_pos == 2) { + if (ctx->pos == 2) { if (jiffies - ctl.head.time >= NCP_MAX_AGE(server)) goto init_cache; @@ -479,10 +471,10 @@ static int ncp_readdir(struct file *filp, void *dirent, filldir_t filldir) goto init_cache; } - if (filp->f_pos > ctl.head.end) + if (ctx->pos > ctl.head.end) goto finished; - ctl.fpos = filp->f_pos + (NCP_DIRCACHE_START - 2); + ctl.fpos = ctx->pos + (NCP_DIRCACHE_START - 2); ctl.ofs = ctl.fpos / NCP_DIRCACHE_SIZE; ctl.idx = ctl.fpos % NCP_DIRCACHE_SIZE; @@ -497,21 +489,21 @@ static int ncp_readdir(struct file *filp, void *dirent, filldir_t filldir) } while (ctl.idx < NCP_DIRCACHE_SIZE) { struct dentry *dent; - int res; + bool over; dent = ncp_dget_fpos(ctl.cache->dentry[ctl.idx], - dentry, filp->f_pos); + dentry, ctx->pos); if (!dent) goto invalid_cache; - res = filldir(dirent, dent->d_name.name, - dent->d_name.len, filp->f_pos, + over = !dir_emit(ctx, dent->d_name.name, + dent->d_name.len, dent->d_inode->i_ino, DT_UNKNOWN); dput(dent); - if (res) + if (over) goto finished; - filp->f_pos += 1; + ctx->pos += 1; ctl.idx += 1; - if (filp->f_pos > ctl.head.end) + if (ctx->pos > ctl.head.end) goto finished; } if (ctl.page) { @@ -548,9 +540,9 @@ init_cache: ctl.valid = 1; read_really: if (ncp_is_server_root(inode)) { - ncp_read_volume_list(filp, dirent, filldir, &ctl); + ncp_read_volume_list(file, ctx, &ctl); } else { - ncp_do_readdir(filp, dirent, filldir, &ctl); + ncp_do_readdir(file, ctx, &ctl); } ctl.head.end = ctl.fpos - 1; ctl.head.eof = ctl.valid; @@ -573,11 +565,11 @@ out: } static int -ncp_fill_cache(struct file *filp, void *dirent, filldir_t filldir, +ncp_fill_cache(struct file *file, struct dir_context *ctx, struct ncp_cache_control *ctrl, struct ncp_entry_info *entry, int inval_childs) { - struct dentry *newdent, *dentry = filp->f_path.dentry; + struct dentry *newdent, *dentry = file->f_path.dentry; struct inode *dir = dentry->d_inode; struct ncp_cache_control ctl = *ctrl; struct qstr qname; @@ -666,15 +658,15 @@ ncp_fill_cache(struct file *filp, void *dirent, filldir_t filldir, end_advance: if (!valid) ctl.valid = 0; - if (!ctl.filled && (ctl.fpos == filp->f_pos)) { + if (!ctl.filled && (ctl.fpos == ctx->pos)) { if (!ino) ino = find_inode_number(dentry, &qname); if (!ino) ino = iunique(dir->i_sb, 2); - ctl.filled = filldir(dirent, qname.name, qname.len, - filp->f_pos, ino, DT_UNKNOWN); + ctl.filled = !dir_emit(ctx, qname.name, qname.len, + ino, DT_UNKNOWN); if (!ctl.filled) - filp->f_pos += 1; + ctx->pos += 1; } ctl.fpos += 1; ctl.idx += 1; @@ -683,10 +675,10 @@ end_advance: } static void -ncp_read_volume_list(struct file *filp, void *dirent, filldir_t filldir, +ncp_read_volume_list(struct file *file, struct dir_context *ctx, struct ncp_cache_control *ctl) { - struct dentry *dentry = filp->f_path.dentry; + struct dentry *dentry = file->f_path.dentry; struct inode *inode = dentry->d_inode; struct ncp_server *server = NCP_SERVER(inode); struct ncp_volume_info info; @@ -694,7 +686,7 @@ ncp_read_volume_list(struct file *filp, void *dirent, filldir_t filldir, int i; DPRINTK("ncp_read_volume_list: pos=%ld\n", - (unsigned long) filp->f_pos); + (unsigned long) ctx->pos); for (i = 0; i < NCP_NUMBER_OF_VOLUMES; i++) { int inval_dentry; @@ -715,16 +707,16 @@ ncp_read_volume_list(struct file *filp, void *dirent, filldir_t filldir, } inval_dentry = ncp_update_known_namespace(server, entry.i.volNumber, NULL); entry.volume = entry.i.volNumber; - if (!ncp_fill_cache(filp, dirent, filldir, ctl, &entry, inval_dentry)) + if (!ncp_fill_cache(file, ctx, ctl, &entry, inval_dentry)) return; } } static void -ncp_do_readdir(struct file *filp, void *dirent, filldir_t filldir, +ncp_do_readdir(struct file *file, struct dir_context *ctx, struct ncp_cache_control *ctl) { - struct dentry *dentry = filp->f_path.dentry; + struct dentry *dentry = file->f_path.dentry; struct inode *dir = dentry->d_inode; struct ncp_server *server = NCP_SERVER(dir); struct nw_search_sequence seq; @@ -736,7 +728,7 @@ ncp_do_readdir(struct file *filp, void *dirent, filldir_t filldir, DPRINTK("ncp_do_readdir: %s/%s, fpos=%ld\n", dentry->d_parent->d_name.name, dentry->d_name.name, - (unsigned long) filp->f_pos); + (unsigned long) ctx->pos); PPRINTK("ncp_do_readdir: init %s, volnum=%d, dirent=%u\n", dentry->d_name.name, NCP_FINFO(dir)->volNumber, NCP_FINFO(dir)->dirEntNum); @@ -778,7 +770,7 @@ ncp_do_readdir(struct file *filp, void *dirent, filldir_t filldir, rpl += onerpl; rpls -= onerpl; entry.volume = entry.i.volNumber; - if (!ncp_fill_cache(filp, dirent, filldir, ctl, &entry, 0)) + if (!ncp_fill_cache(file, ctx, ctl, &entry, 0)) break; } } while (more); @@ -1029,15 +1021,6 @@ static int ncp_rmdir(struct inode *dir, struct dentry *dentry) DPRINTK("ncp_rmdir: removing %s/%s\n", dentry->d_parent->d_name.name, dentry->d_name.name); - /* - * fail with EBUSY if there are still references to this - * directory. - */ - dentry_unhash(dentry); - error = -EBUSY; - if (!d_unhashed(dentry)) - goto out; - len = sizeof(__name); error = ncp_io2vol(server, __name, &len, dentry->d_name.name, dentry->d_name.len, !ncp_preserve_case(dir)); diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index a13d26ede254..0bc27684ebfa 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -414,7 +414,7 @@ __be32 nfs4_callback_sequence(struct cb_sequenceargs *args, spin_lock(&tbl->slot_tbl_lock); /* state manager is resetting the session */ - if (test_bit(NFS4_SESSION_DRAINING, &clp->cl_session->session_state)) { + if (test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state)) { spin_unlock(&tbl->slot_tbl_lock); status = htonl(NFS4ERR_DELAY); /* Return NFS4ERR_BADSESSION if we're draining the session diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 59461c957d9d..a35582c9d444 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -763,7 +763,7 @@ static void nfs4_callback_free_slot(struct nfs4_session *session) * A single slot, so highest used slotid is either 0 or -1 */ tbl->highest_used_slotid = NFS4_NO_SLOT; - nfs4_session_drain_complete(session, tbl); + nfs4_slot_tbl_drain_complete(tbl); spin_unlock(&tbl->slot_tbl_lock); } diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index e093e73178b7..5d051419527b 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -46,7 +46,7 @@ static int nfs_opendir(struct inode *, struct file *); static int nfs_closedir(struct inode *, struct file *); -static int nfs_readdir(struct file *, void *, filldir_t); +static int nfs_readdir(struct file *, struct dir_context *); static int nfs_fsync_dir(struct file *, loff_t, loff_t, int); static loff_t nfs_llseek_dir(struct file *, loff_t, int); static void nfs_readdir_clear_array(struct page*); @@ -54,7 +54,7 @@ static void nfs_readdir_clear_array(struct page*); const struct file_operations nfs_dir_operations = { .llseek = nfs_llseek_dir, .read = generic_read_dir, - .readdir = nfs_readdir, + .iterate = nfs_readdir, .open = nfs_opendir, .release = nfs_closedir, .fsync = nfs_fsync_dir, @@ -147,6 +147,7 @@ typedef int (*decode_dirent_t)(struct xdr_stream *, struct nfs_entry *, int); typedef struct { struct file *file; struct page *page; + struct dir_context *ctx; unsigned long page_index; u64 *dir_cookie; u64 last_cookie; @@ -252,7 +253,7 @@ out: static int nfs_readdir_search_for_pos(struct nfs_cache_array *array, nfs_readdir_descriptor_t *desc) { - loff_t diff = desc->file->f_pos - desc->current_index; + loff_t diff = desc->ctx->pos - desc->current_index; unsigned int index; if (diff < 0) @@ -289,7 +290,7 @@ int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, nfs_readdir_des || (nfsi->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA))) { ctx->duped = 0; ctx->attr_gencount = nfsi->attr_gencount; - } else if (new_pos < desc->file->f_pos) { + } else if (new_pos < desc->ctx->pos) { if (ctx->duped > 0 && ctx->dup_cookie == *desc->dir_cookie) { if (printk_ratelimit()) { @@ -307,7 +308,7 @@ int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, nfs_readdir_des ctx->dup_cookie = *desc->dir_cookie; ctx->duped = -1; } - desc->file->f_pos = new_pos; + desc->ctx->pos = new_pos; desc->cache_entry_index = i; return 0; } @@ -405,13 +406,13 @@ different: } static -bool nfs_use_readdirplus(struct inode *dir, struct file *filp) +bool nfs_use_readdirplus(struct inode *dir, struct dir_context *ctx) { if (!nfs_server_capable(dir, NFS_CAP_READDIRPLUS)) return false; if (test_and_clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(dir)->flags)) return true; - if (filp->f_pos == 0) + if (ctx->pos == 0) return true; return false; } @@ -702,8 +703,7 @@ int readdir_search_pagecache(nfs_readdir_descriptor_t *desc) * Once we've found the start of the dirent within a page: fill 'er up... */ static -int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent, - filldir_t filldir) +int nfs_do_filldir(nfs_readdir_descriptor_t *desc) { struct file *file = desc->file; int i = 0; @@ -721,13 +721,12 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent, struct nfs_cache_array_entry *ent; ent = &array->array[i]; - if (filldir(dirent, ent->string.name, ent->string.len, - file->f_pos, nfs_compat_user_ino64(ent->ino), - ent->d_type) < 0) { + if (!dir_emit(desc->ctx, ent->string.name, ent->string.len, + nfs_compat_user_ino64(ent->ino), ent->d_type)) { desc->eof = 1; break; } - file->f_pos++; + desc->ctx->pos++; if (i < (array->size-1)) *desc->dir_cookie = array->array[i+1].cookie; else @@ -759,8 +758,7 @@ out: * directory in the page cache by the time we get here. */ static inline -int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent, - filldir_t filldir) +int uncached_readdir(nfs_readdir_descriptor_t *desc) { struct page *page = NULL; int status; @@ -785,7 +783,7 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent, if (status < 0) goto out_release; - status = nfs_do_filldir(desc, dirent, filldir); + status = nfs_do_filldir(desc); out: dfprintk(DIRCACHE, "NFS: %s: returns %d\n", @@ -800,35 +798,36 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent, last cookie cache takes care of the common case of reading the whole directory. */ -static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int nfs_readdir(struct file *file, struct dir_context *ctx) { - struct dentry *dentry = filp->f_path.dentry; + struct dentry *dentry = file->f_path.dentry; struct inode *inode = dentry->d_inode; nfs_readdir_descriptor_t my_desc, *desc = &my_desc; - struct nfs_open_dir_context *dir_ctx = filp->private_data; + struct nfs_open_dir_context *dir_ctx = file->private_data; int res; dfprintk(FILE, "NFS: readdir(%s/%s) starting at cookie %llu\n", dentry->d_parent->d_name.name, dentry->d_name.name, - (long long)filp->f_pos); + (long long)ctx->pos); nfs_inc_stats(inode, NFSIOS_VFSGETDENTS); /* - * filp->f_pos points to the dirent entry number. + * ctx->pos points to the dirent entry number. * *desc->dir_cookie has the cookie for the next entry. We have * to either find the entry with the appropriate number or * revalidate the cookie. */ memset(desc, 0, sizeof(*desc)); - desc->file = filp; + desc->file = file; + desc->ctx = ctx; desc->dir_cookie = &dir_ctx->dir_cookie; desc->decode = NFS_PROTO(inode)->decode_dirent; - desc->plus = nfs_use_readdirplus(inode, filp) ? 1 : 0; + desc->plus = nfs_use_readdirplus(inode, ctx) ? 1 : 0; nfs_block_sillyrename(dentry); - res = nfs_revalidate_mapping(inode, filp->f_mapping); + res = nfs_revalidate_mapping(inode, file->f_mapping); if (res < 0) goto out; @@ -840,7 +839,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) /* This means either end of directory */ if (*desc->dir_cookie && desc->eof == 0) { /* Or that the server has 'lost' a cookie */ - res = uncached_readdir(desc, dirent, filldir); + res = uncached_readdir(desc); if (res == 0) continue; } @@ -857,7 +856,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (res < 0) break; - res = nfs_do_filldir(desc, dirent, filldir); + res = nfs_do_filldir(desc); if (res < 0) break; } while (!desc->eof); diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c index 947b0c908aa9..4cbad5d6b276 100644 --- a/fs/nfs/nfs4client.c +++ b/fs/nfs/nfs4client.c @@ -203,7 +203,7 @@ struct nfs_client *nfs4_init_client(struct nfs_client *clp, __set_bit(NFS_CS_DISCRTRY, &clp->cl_flags); error = nfs_create_rpc_client(clp, timeparms, RPC_AUTH_GSS_KRB5I); if (error == -EINVAL) - error = nfs_create_rpc_client(clp, timeparms, RPC_AUTH_NULL); + error = nfs_create_rpc_client(clp, timeparms, RPC_AUTH_UNIX); if (error < 0) goto error; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 8fbc10054115..d7ba5616989c 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -572,7 +572,7 @@ int nfs41_setup_sequence(struct nfs4_session *session, task->tk_timeout = 0; spin_lock(&tbl->slot_tbl_lock); - if (test_bit(NFS4_SESSION_DRAINING, &session->session_state) && + if (test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state) && !args->sa_privileged) { /* The state manager will wait until the slot table is empty */ dprintk("%s session is draining\n", __func__); @@ -1078,7 +1078,7 @@ static struct nfs4_state *nfs4_try_open_cached(struct nfs4_opendata *opendata) struct nfs4_state *state = opendata->state; struct nfs_inode *nfsi = NFS_I(state->inode); struct nfs_delegation *delegation; - int open_mode = opendata->o_arg.open_flags & (O_EXCL|O_TRUNC); + int open_mode = opendata->o_arg.open_flags; fmode_t fmode = opendata->o_arg.fmode; nfs4_stateid stateid; int ret = -EAGAIN; diff --git a/fs/nfs/nfs4session.c b/fs/nfs/nfs4session.c index ebda5f4a031b..c4e225e4a9af 100644 --- a/fs/nfs/nfs4session.c +++ b/fs/nfs/nfs4session.c @@ -73,7 +73,7 @@ void nfs4_free_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot) tbl->highest_used_slotid = new_max; else { tbl->highest_used_slotid = NFS4_NO_SLOT; - nfs4_session_drain_complete(tbl->session, tbl); + nfs4_slot_tbl_drain_complete(tbl); } } dprintk("%s: slotid %u highest_used_slotid %d\n", __func__, @@ -226,7 +226,7 @@ static bool nfs41_assign_slot(struct rpc_task *task, void *pslot) struct nfs4_slot *slot = pslot; struct nfs4_slot_table *tbl = slot->table; - if (nfs4_session_draining(tbl->session) && !args->sa_privileged) + if (nfs4_slot_tbl_draining(tbl) && !args->sa_privileged) return false; slot->generation = tbl->generation; args->sa_slot = slot; diff --git a/fs/nfs/nfs4session.h b/fs/nfs/nfs4session.h index 6f3cb39386d4..ff7d9f0f8a65 100644 --- a/fs/nfs/nfs4session.h +++ b/fs/nfs/nfs4session.h @@ -25,6 +25,10 @@ struct nfs4_slot { }; /* Sessions */ +enum nfs4_slot_tbl_state { + NFS4_SLOT_TBL_DRAINING, +}; + #define SLOT_TABLE_SZ DIV_ROUND_UP(NFS4_MAX_SLOT_TABLE, 8*sizeof(long)) struct nfs4_slot_table { struct nfs4_session *session; /* Parent session */ @@ -43,6 +47,7 @@ struct nfs4_slot_table { unsigned long generation; /* Generation counter for target_highest_slotid */ struct completion complete; + unsigned long slot_tbl_state; }; /* @@ -68,7 +73,6 @@ struct nfs4_session { enum nfs4_session_state { NFS4_SESSION_INITING, - NFS4_SESSION_DRAINING, }; #if defined(CONFIG_NFS_V4_1) @@ -88,12 +92,11 @@ extern void nfs4_destroy_session(struct nfs4_session *session); extern int nfs4_init_session(struct nfs_server *server); extern int nfs4_init_ds_session(struct nfs_client *, unsigned long); -extern void nfs4_session_drain_complete(struct nfs4_session *session, - struct nfs4_slot_table *tbl); +extern void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl); -static inline bool nfs4_session_draining(struct nfs4_session *session) +static inline bool nfs4_slot_tbl_draining(struct nfs4_slot_table *tbl) { - return !!test_bit(NFS4_SESSION_DRAINING, &session->session_state); + return !!test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state); } bool nfs41_wake_and_assign_slot(struct nfs4_slot_table *tbl, diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 300d17d85c0e..1fab140764c4 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -241,7 +241,7 @@ static void nfs4_end_drain_session(struct nfs_client *clp) if (ses == NULL) return; tbl = &ses->fc_slot_table; - if (test_and_clear_bit(NFS4_SESSION_DRAINING, &ses->session_state)) { + if (test_and_clear_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state)) { spin_lock(&tbl->slot_tbl_lock); nfs41_wake_slot_table(tbl); spin_unlock(&tbl->slot_tbl_lock); @@ -251,15 +251,15 @@ static void nfs4_end_drain_session(struct nfs_client *clp) /* * Signal state manager thread if session fore channel is drained */ -void nfs4_session_drain_complete(struct nfs4_session *session, - struct nfs4_slot_table *tbl) +void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl) { - if (nfs4_session_draining(session)) + if (nfs4_slot_tbl_draining(tbl)) complete(&tbl->complete); } -static int nfs4_wait_on_slot_tbl(struct nfs4_slot_table *tbl) +static int nfs4_drain_slot_tbl(struct nfs4_slot_table *tbl) { + set_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state); spin_lock(&tbl->slot_tbl_lock); if (tbl->highest_used_slotid != NFS4_NO_SLOT) { INIT_COMPLETION(tbl->complete); @@ -275,13 +275,12 @@ static int nfs4_begin_drain_session(struct nfs_client *clp) struct nfs4_session *ses = clp->cl_session; int ret = 0; - set_bit(NFS4_SESSION_DRAINING, &ses->session_state); /* back channel */ - ret = nfs4_wait_on_slot_tbl(&ses->bc_slot_table); + ret = nfs4_drain_slot_tbl(&ses->bc_slot_table); if (ret) return ret; /* fore channel */ - return nfs4_wait_on_slot_tbl(&ses->fc_slot_table); + return nfs4_drain_slot_tbl(&ses->fc_slot_table); } static void nfs41_finish_session_reset(struct nfs_client *clp) diff --git a/fs/nfs/super.c b/fs/nfs/super.c index a366107a7331..2d7525fbcf25 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -1942,6 +1942,7 @@ static int nfs23_validate_mount_data(void *options, args->namlen = data->namlen; args->bsize = data->bsize; + args->auth_flavors[0] = RPC_AUTH_UNIX; if (data->flags & NFS_MOUNT_SECFLAVOUR) args->auth_flavors[0] = data->pseudoflavor; if (!args->nfs_server.hostname) @@ -2637,6 +2638,7 @@ static int nfs4_validate_mount_data(void *options, goto out_no_address; args->nfs_server.port = ntohs(((struct sockaddr_in *)sap)->sin_port); + args->auth_flavors[0] = RPC_AUTH_UNIX; if (data->auth_flavourlen) { if (data->auth_flavourlen > 1) goto out_inval_auth; diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 4e9a21db867a..105a3b080d12 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -240,11 +240,16 @@ struct name_list { struct list_head list; }; +struct nfs4_dir_ctx { + struct dir_context ctx; + struct list_head names; +}; + static int nfsd4_build_namelist(void *arg, const char *name, int namlen, loff_t offset, u64 ino, unsigned int d_type) { - struct list_head *names = arg; + struct nfs4_dir_ctx *ctx = arg; struct name_list *entry; if (namlen != HEXDIR_LEN - 1) @@ -254,7 +259,7 @@ nfsd4_build_namelist(void *arg, const char *name, int namlen, return -ENOMEM; memcpy(entry->name, name, HEXDIR_LEN - 1); entry->name[HEXDIR_LEN - 1] = '\0'; - list_add(&entry->list, names); + list_add(&entry->list, &ctx->names); return 0; } @@ -263,7 +268,10 @@ nfsd4_list_rec_dir(recdir_func *f, struct nfsd_net *nn) { const struct cred *original_cred; struct dentry *dir = nn->rec_file->f_path.dentry; - LIST_HEAD(names); + struct nfs4_dir_ctx ctx = { + .ctx.actor = nfsd4_build_namelist, + .names = LIST_HEAD_INIT(ctx.names) + }; int status; status = nfs4_save_creds(&original_cred); @@ -276,11 +284,11 @@ nfsd4_list_rec_dir(recdir_func *f, struct nfsd_net *nn) return status; } - status = vfs_readdir(nn->rec_file, nfsd4_build_namelist, &names); + status = iterate_dir(nn->rec_file, &ctx.ctx); mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT); - while (!list_empty(&names)) { + while (!list_empty(&ctx.names)) { struct name_list *entry; - entry = list_entry(names.next, struct name_list, list); + entry = list_entry(ctx.names.next, struct name_list, list); if (!status) { struct dentry *dentry; dentry = lookup_one_len(entry->name, dir, HEXDIR_LEN-1); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 84ce601d8063..a6bc8a7423db 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1912,6 +1912,7 @@ struct buffered_dirent { }; struct readdir_data { + struct dir_context ctx; char *dirent; size_t used; int full; @@ -1943,13 +1944,15 @@ static int nfsd_buffered_filldir(void *__buf, const char *name, int namlen, static __be32 nfsd_buffered_readdir(struct file *file, filldir_t func, struct readdir_cd *cdp, loff_t *offsetp) { - struct readdir_data buf; struct buffered_dirent *de; int host_err; int size; loff_t offset; + struct readdir_data buf = { + .ctx.actor = nfsd_buffered_filldir, + .dirent = (void *)__get_free_page(GFP_KERNEL) + }; - buf.dirent = (void *)__get_free_page(GFP_KERNEL); if (!buf.dirent) return nfserrno(-ENOMEM); @@ -1963,7 +1966,7 @@ static __be32 nfsd_buffered_readdir(struct file *file, filldir_t func, buf.used = 0; buf.full = 0; - host_err = vfs_readdir(file, nfsd_buffered_filldir, &buf); + host_err = iterate_dir(file, &buf.ctx); if (buf.full) host_err = 0; diff --git a/fs/nilfs2/dir.c b/fs/nilfs2/dir.c index f30b017740a7..197a63e9d102 100644 --- a/fs/nilfs2/dir.c +++ b/fs/nilfs2/dir.c @@ -256,22 +256,18 @@ static void nilfs_set_de_type(struct nilfs_dir_entry *de, struct inode *inode) de->file_type = nilfs_type_by_mode[(mode & S_IFMT)>>S_SHIFT]; } -static int nilfs_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int nilfs_readdir(struct file *file, struct dir_context *ctx) { - loff_t pos = filp->f_pos; - struct inode *inode = file_inode(filp); + loff_t pos = ctx->pos; + struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; unsigned int offset = pos & ~PAGE_CACHE_MASK; unsigned long n = pos >> PAGE_CACHE_SHIFT; unsigned long npages = dir_pages(inode); /* unsigned chunk_mask = ~(nilfs_chunk_size(inode)-1); */ - unsigned char *types = NULL; - int ret; if (pos > inode->i_size - NILFS_DIR_REC_LEN(1)) - goto success; - - types = nilfs_filetype_table; + return 0; for ( ; n < npages; n++, offset = 0) { char *kaddr, *limit; @@ -281,9 +277,8 @@ static int nilfs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (IS_ERR(page)) { nilfs_error(sb, __func__, "bad page in #%lu", inode->i_ino); - filp->f_pos += PAGE_CACHE_SIZE - offset; - ret = -EIO; - goto done; + ctx->pos += PAGE_CACHE_SIZE - offset; + return -EIO; } kaddr = page_address(page); de = (struct nilfs_dir_entry *)(kaddr + offset); @@ -293,35 +288,28 @@ static int nilfs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (de->rec_len == 0) { nilfs_error(sb, __func__, "zero-length directory entry"); - ret = -EIO; nilfs_put_page(page); - goto done; + return -EIO; } if (de->inode) { - int over; - unsigned char d_type = DT_UNKNOWN; + unsigned char t; - if (types && de->file_type < NILFS_FT_MAX) - d_type = types[de->file_type]; + if (de->file_type < NILFS_FT_MAX) + t = nilfs_filetype_table[de->file_type]; + else + t = DT_UNKNOWN; - offset = (char *)de - kaddr; - over = filldir(dirent, de->name, de->name_len, - (n<<PAGE_CACHE_SHIFT) | offset, - le64_to_cpu(de->inode), d_type); - if (over) { + if (!dir_emit(ctx, de->name, de->name_len, + le64_to_cpu(de->inode), t)) { nilfs_put_page(page); - goto success; + return 0; } } - filp->f_pos += nilfs_rec_len_from_disk(de->rec_len); + ctx->pos += nilfs_rec_len_from_disk(de->rec_len); } nilfs_put_page(page); } - -success: - ret = 0; -done: - return ret; + return 0; } /* @@ -678,7 +666,7 @@ not_empty: const struct file_operations nilfs_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = nilfs_readdir, + .iterate = nilfs_readdir, .unlocked_ioctl = nilfs_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = nilfs_compat_ioctl, diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c index 689fb608648e..bccfec8343c5 100644 --- a/fs/nilfs2/inode.c +++ b/fs/nilfs2/inode.c @@ -219,13 +219,32 @@ static int nilfs_writepage(struct page *page, struct writeback_control *wbc) static int nilfs_set_page_dirty(struct page *page) { - int ret = __set_page_dirty_buffers(page); + int ret = __set_page_dirty_nobuffers(page); - if (ret) { + if (page_has_buffers(page)) { struct inode *inode = page->mapping->host; - unsigned nr_dirty = 1 << (PAGE_SHIFT - inode->i_blkbits); + unsigned nr_dirty = 0; + struct buffer_head *bh, *head; - nilfs_set_file_dirty(inode, nr_dirty); + /* + * This page is locked by callers, and no other thread + * concurrently marks its buffers dirty since they are + * only dirtied through routines in fs/buffer.c in + * which call sites of mark_buffer_dirty are protected + * by page lock. + */ + bh = head = page_buffers(page); + do { + /* Do not mark hole blocks dirty */ + if (buffer_dirty(bh) || !buffer_mapped(bh)) + continue; + + set_buffer_dirty(bh); + nr_dirty++; + } while (bh = bh->b_this_page, bh != head); + + if (nr_dirty) + nilfs_set_file_dirty(inode, nr_dirty); } return ret; } diff --git a/fs/ntfs/dir.c b/fs/ntfs/dir.c index aa411c3f20e9..9e38dafa3bc7 100644 --- a/fs/ntfs/dir.c +++ b/fs/ntfs/dir.c @@ -1004,13 +1004,11 @@ dir_err_out: /** * ntfs_filldir - ntfs specific filldir method * @vol: current ntfs volume - * @fpos: position in the directory * @ndir: ntfs inode of current directory * @ia_page: page in which the index allocation buffer @ie is in resides * @ie: current index entry * @name: buffer to use for the converted name - * @dirent: vfs filldir callback context - * @filldir: vfs filldir callback + * @actor: what to feed the entries to * * Convert the Unicode @name to the loaded NLS and pass it to the @filldir * callback. @@ -1024,12 +1022,12 @@ dir_err_out: * retake the lock if we are returning a non-zero value as ntfs_readdir() * would need to drop the lock immediately anyway. */ -static inline int ntfs_filldir(ntfs_volume *vol, loff_t fpos, +static inline int ntfs_filldir(ntfs_volume *vol, ntfs_inode *ndir, struct page *ia_page, INDEX_ENTRY *ie, - u8 *name, void *dirent, filldir_t filldir) + u8 *name, struct dir_context *actor) { unsigned long mref; - int name_len, rc; + int name_len; unsigned dt_type; FILE_NAME_TYPE_FLAGS name_type; @@ -1068,13 +1066,14 @@ static inline int ntfs_filldir(ntfs_volume *vol, loff_t fpos, if (ia_page) unlock_page(ia_page); ntfs_debug("Calling filldir for %s with len %i, fpos 0x%llx, inode " - "0x%lx, DT_%s.", name, name_len, fpos, mref, + "0x%lx, DT_%s.", name, name_len, actor->pos, mref, dt_type == DT_DIR ? "DIR" : "REG"); - rc = filldir(dirent, name, name_len, fpos, mref, dt_type); + if (!dir_emit(actor, name, name_len, mref, dt_type)) + return 1; /* Relock the page but not if we are aborting ->readdir. */ - if (!rc && ia_page) + if (ia_page) lock_page(ia_page); - return rc; + return 0; } /* @@ -1097,11 +1096,11 @@ static inline int ntfs_filldir(ntfs_volume *vol, loff_t fpos, * removes them again after the write is complete after which it * unlocks the page. */ -static int ntfs_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int ntfs_readdir(struct file *file, struct dir_context *actor) { s64 ia_pos, ia_start, prev_ia_pos, bmp_pos; - loff_t fpos, i_size; - struct inode *bmp_vi, *vdir = file_inode(filp); + loff_t i_size; + struct inode *bmp_vi, *vdir = file_inode(file); struct super_block *sb = vdir->i_sb; ntfs_inode *ndir = NTFS_I(vdir); ntfs_volume *vol = NTFS_SB(sb); @@ -1116,33 +1115,16 @@ static int ntfs_readdir(struct file *filp, void *dirent, filldir_t filldir) u8 *kaddr, *bmp, *index_end; ntfs_attr_search_ctx *ctx; - fpos = filp->f_pos; ntfs_debug("Entering for inode 0x%lx, fpos 0x%llx.", - vdir->i_ino, fpos); + vdir->i_ino, actor->pos); rc = err = 0; /* Are we at end of dir yet? */ i_size = i_size_read(vdir); - if (fpos >= i_size + vol->mft_record_size) - goto done; + if (actor->pos >= i_size + vol->mft_record_size) + return 0; /* Emulate . and .. for all directories. */ - if (!fpos) { - ntfs_debug("Calling filldir for . with len 1, fpos 0x0, " - "inode 0x%lx, DT_DIR.", vdir->i_ino); - rc = filldir(dirent, ".", 1, fpos, vdir->i_ino, DT_DIR); - if (rc) - goto done; - fpos++; - } - if (fpos == 1) { - ntfs_debug("Calling filldir for .. with len 2, fpos 0x1, " - "inode 0x%lx, DT_DIR.", - (unsigned long)parent_ino(filp->f_path.dentry)); - rc = filldir(dirent, "..", 2, fpos, - parent_ino(filp->f_path.dentry), DT_DIR); - if (rc) - goto done; - fpos++; - } + if (!dir_emit_dots(file, actor)) + return 0; m = NULL; ctx = NULL; /* @@ -1155,7 +1137,7 @@ static int ntfs_readdir(struct file *filp, void *dirent, filldir_t filldir) goto err_out; } /* Are we jumping straight into the index allocation attribute? */ - if (fpos >= vol->mft_record_size) + if (actor->pos >= vol->mft_record_size) goto skip_index_root; /* Get hold of the mft record for the directory. */ m = map_mft_record(ndir); @@ -1170,7 +1152,7 @@ static int ntfs_readdir(struct file *filp, void *dirent, filldir_t filldir) goto err_out; } /* Get the offset into the index root attribute. */ - ir_pos = (s64)fpos; + ir_pos = (s64)actor->pos; /* Find the index root attribute in the mft record. */ err = ntfs_attr_lookup(AT_INDEX_ROOT, I30, 4, CASE_SENSITIVE, 0, NULL, 0, ctx); @@ -1226,10 +1208,9 @@ static int ntfs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (ir_pos > (u8*)ie - (u8*)ir) continue; /* Advance the position even if going to skip the entry. */ - fpos = (u8*)ie - (u8*)ir; + actor->pos = (u8*)ie - (u8*)ir; /* Submit the name to the filldir callback. */ - rc = ntfs_filldir(vol, fpos, ndir, NULL, ie, name, dirent, - filldir); + rc = ntfs_filldir(vol, ndir, NULL, ie, name, actor); if (rc) { kfree(ir); goto abort; @@ -1242,12 +1223,12 @@ static int ntfs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (!NInoIndexAllocPresent(ndir)) goto EOD; /* Advance fpos to the beginning of the index allocation. */ - fpos = vol->mft_record_size; + actor->pos = vol->mft_record_size; skip_index_root: kaddr = NULL; prev_ia_pos = -1LL; /* Get the offset into the index allocation attribute. */ - ia_pos = (s64)fpos - vol->mft_record_size; + ia_pos = (s64)actor->pos - vol->mft_record_size; ia_mapping = vdir->i_mapping; ntfs_debug("Inode 0x%lx, getting index bitmap.", vdir->i_ino); bmp_vi = ntfs_attr_iget(vdir, AT_BITMAP, I30, 4); @@ -1409,7 +1390,7 @@ find_next_index_buffer: if (ia_pos - ia_start > (u8*)ie - (u8*)ia) continue; /* Advance the position even if going to skip the entry. */ - fpos = (u8*)ie - (u8*)ia + + actor->pos = (u8*)ie - (u8*)ia + (sle64_to_cpu(ia->index_block_vcn) << ndir->itype.index.vcn_size_bits) + vol->mft_record_size; @@ -1419,8 +1400,7 @@ find_next_index_buffer: * before returning, unless a non-zero value is returned in * which case the page is left unlocked. */ - rc = ntfs_filldir(vol, fpos, ndir, ia_page, ie, name, dirent, - filldir); + rc = ntfs_filldir(vol, ndir, ia_page, ie, name, actor); if (rc) { /* @ia_page is already unlocked in this case. */ ntfs_unmap_page(ia_page); @@ -1439,18 +1419,9 @@ unm_EOD: iput(bmp_vi); EOD: /* We are finished, set fpos to EOD. */ - fpos = i_size + vol->mft_record_size; + actor->pos = i_size + vol->mft_record_size; abort: kfree(name); -done: -#ifdef DEBUG - if (!rc) - ntfs_debug("EOD, fpos 0x%llx, returning 0.", fpos); - else - ntfs_debug("filldir returned %i, fpos 0x%llx, returning 0.", - rc, fpos); -#endif - filp->f_pos = fpos; return 0; err_out: if (bmp_page) { @@ -1471,7 +1442,6 @@ iput_err_out: if (!err) err = -EIO; ntfs_debug("Failed. Returning error code %i.", -err); - filp->f_pos = fpos; return err; } @@ -1571,7 +1541,7 @@ static int ntfs_dir_fsync(struct file *filp, loff_t start, loff_t end, const struct file_operations ntfs_dir_ops = { .llseek = generic_file_llseek, /* Seek inside directory. */ .read = generic_read_dir, /* Return -EISDIR. */ - .readdir = ntfs_readdir, /* Read directory contents. */ + .iterate = ntfs_readdir, /* Read directory contents. */ #ifdef NTFS_RW .fsync = ntfs_dir_fsync, /* Sync a directory to disk. */ /*.aio_fsync = ,*/ /* Sync all outstanding async diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index f1e1aed8f638..eb760d8acd50 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -1761,11 +1761,10 @@ bail: static int ocfs2_dir_foreach_blk_id(struct inode *inode, u64 *f_version, - loff_t *f_pos, void *priv, - filldir_t filldir, int *filldir_err) + struct dir_context *ctx) { - int ret, i, filldir_ret; - unsigned long offset = *f_pos; + int ret, i; + unsigned long offset = ctx->pos; struct buffer_head *di_bh = NULL; struct ocfs2_dinode *di; struct ocfs2_inline_data *data; @@ -1781,8 +1780,7 @@ static int ocfs2_dir_foreach_blk_id(struct inode *inode, di = (struct ocfs2_dinode *)di_bh->b_data; data = &di->id2.i_data; - while (*f_pos < i_size_read(inode)) { -revalidate: + while (ctx->pos < i_size_read(inode)) { /* If the dir block has changed since the last call to * readdir(2), then we might be pointing to an invalid * dirent right now. Scan from the start of the block @@ -1802,50 +1800,31 @@ revalidate: break; i += le16_to_cpu(de->rec_len); } - *f_pos = offset = i; + ctx->pos = offset = i; *f_version = inode->i_version; } - de = (struct ocfs2_dir_entry *) (data->id_data + *f_pos); - if (!ocfs2_check_dir_entry(inode, de, di_bh, *f_pos)) { + de = (struct ocfs2_dir_entry *) (data->id_data + ctx->pos); + if (!ocfs2_check_dir_entry(inode, de, di_bh, ctx->pos)) { /* On error, skip the f_pos to the end. */ - *f_pos = i_size_read(inode); - goto out; + ctx->pos = i_size_read(inode); + break; } offset += le16_to_cpu(de->rec_len); if (le64_to_cpu(de->inode)) { - /* We might block in the next section - * if the data destination is - * currently swapped out. So, use a - * version stamp to detect whether or - * not the directory has been modified - * during the copy operation. - */ - u64 version = *f_version; unsigned char d_type = DT_UNKNOWN; if (de->file_type < OCFS2_FT_MAX) d_type = ocfs2_filetype_table[de->file_type]; - filldir_ret = filldir(priv, de->name, - de->name_len, - *f_pos, - le64_to_cpu(de->inode), - d_type); - if (filldir_ret) { - if (filldir_err) - *filldir_err = filldir_ret; - break; - } - if (version != *f_version) - goto revalidate; + if (!dir_emit(ctx, de->name, de->name_len, + le64_to_cpu(de->inode), d_type)) + goto out; } - *f_pos += le16_to_cpu(de->rec_len); + ctx->pos += le16_to_cpu(de->rec_len); } - out: brelse(di_bh); - return 0; } @@ -1855,27 +1834,26 @@ out: */ static int ocfs2_dir_foreach_blk_el(struct inode *inode, u64 *f_version, - loff_t *f_pos, void *priv, - filldir_t filldir, int *filldir_err) + struct dir_context *ctx, + bool persist) { - int error = 0; unsigned long offset, blk, last_ra_blk = 0; - int i, stored; + int i; struct buffer_head * bh, * tmp; struct ocfs2_dir_entry * de; struct super_block * sb = inode->i_sb; unsigned int ra_sectors = 16; + int stored = 0; - stored = 0; bh = NULL; - offset = (*f_pos) & (sb->s_blocksize - 1); + offset = ctx->pos & (sb->s_blocksize - 1); - while (!error && !stored && *f_pos < i_size_read(inode)) { - blk = (*f_pos) >> sb->s_blocksize_bits; + while (ctx->pos < i_size_read(inode)) { + blk = ctx->pos >> sb->s_blocksize_bits; if (ocfs2_read_dir_block(inode, blk, &bh, 0)) { /* Skip the corrupt dirblock and keep trying */ - *f_pos += sb->s_blocksize - offset; + ctx->pos += sb->s_blocksize - offset; continue; } @@ -1897,7 +1875,6 @@ static int ocfs2_dir_foreach_blk_el(struct inode *inode, ra_sectors = 8; } -revalidate: /* If the dir block has changed since the last call to * readdir(2), then we might be pointing to an invalid * dirent right now. Scan from the start of the block @@ -1917,93 +1894,64 @@ revalidate: i += le16_to_cpu(de->rec_len); } offset = i; - *f_pos = ((*f_pos) & ~(sb->s_blocksize - 1)) + ctx->pos = (ctx->pos & ~(sb->s_blocksize - 1)) | offset; *f_version = inode->i_version; } - while (!error && *f_pos < i_size_read(inode) + while (ctx->pos < i_size_read(inode) && offset < sb->s_blocksize) { de = (struct ocfs2_dir_entry *) (bh->b_data + offset); if (!ocfs2_check_dir_entry(inode, de, bh, offset)) { /* On error, skip the f_pos to the next block. */ - *f_pos = ((*f_pos) | (sb->s_blocksize - 1)) + 1; + ctx->pos = (ctx->pos | (sb->s_blocksize - 1)) + 1; brelse(bh); - goto out; + continue; } - offset += le16_to_cpu(de->rec_len); if (le64_to_cpu(de->inode)) { - /* We might block in the next section - * if the data destination is - * currently swapped out. So, use a - * version stamp to detect whether or - * not the directory has been modified - * during the copy operation. - */ - unsigned long version = *f_version; unsigned char d_type = DT_UNKNOWN; if (de->file_type < OCFS2_FT_MAX) d_type = ocfs2_filetype_table[de->file_type]; - error = filldir(priv, de->name, + if (!dir_emit(ctx, de->name, de->name_len, - *f_pos, le64_to_cpu(de->inode), - d_type); - if (error) { - if (filldir_err) - *filldir_err = error; - break; + d_type)) { + brelse(bh); + return 0; } - if (version != *f_version) - goto revalidate; - stored ++; + stored++; } - *f_pos += le16_to_cpu(de->rec_len); + offset += le16_to_cpu(de->rec_len); + ctx->pos += le16_to_cpu(de->rec_len); } offset = 0; brelse(bh); bh = NULL; + if (!persist && stored) + break; } - - stored = 0; -out: - return stored; + return 0; } static int ocfs2_dir_foreach_blk(struct inode *inode, u64 *f_version, - loff_t *f_pos, void *priv, filldir_t filldir, - int *filldir_err) + struct dir_context *ctx, + bool persist) { if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) - return ocfs2_dir_foreach_blk_id(inode, f_version, f_pos, priv, - filldir, filldir_err); - - return ocfs2_dir_foreach_blk_el(inode, f_version, f_pos, priv, filldir, - filldir_err); + return ocfs2_dir_foreach_blk_id(inode, f_version, ctx); + return ocfs2_dir_foreach_blk_el(inode, f_version, ctx, persist); } /* * This is intended to be called from inside other kernel functions, * so we fake some arguments. */ -int ocfs2_dir_foreach(struct inode *inode, loff_t *f_pos, void *priv, - filldir_t filldir) +int ocfs2_dir_foreach(struct inode *inode, struct dir_context *ctx) { - int ret = 0, filldir_err = 0; u64 version = inode->i_version; - - while (*f_pos < i_size_read(inode)) { - ret = ocfs2_dir_foreach_blk(inode, &version, f_pos, priv, - filldir, &filldir_err); - if (ret || filldir_err) - break; - } - - if (ret > 0) - ret = -EIO; - + ocfs2_dir_foreach_blk(inode, &version, ctx, true); return 0; } @@ -2011,15 +1959,15 @@ int ocfs2_dir_foreach(struct inode *inode, loff_t *f_pos, void *priv, * ocfs2_readdir() * */ -int ocfs2_readdir(struct file * filp, void * dirent, filldir_t filldir) +int ocfs2_readdir(struct file *file, struct dir_context *ctx) { int error = 0; - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); int lock_level = 0; trace_ocfs2_readdir((unsigned long long)OCFS2_I(inode)->ip_blkno); - error = ocfs2_inode_lock_atime(inode, filp->f_path.mnt, &lock_level); + error = ocfs2_inode_lock_atime(inode, file->f_path.mnt, &lock_level); if (lock_level && error >= 0) { /* We release EX lock which used to update atime * and get PR lock again to reduce contention @@ -2035,8 +1983,7 @@ int ocfs2_readdir(struct file * filp, void * dirent, filldir_t filldir) goto bail_nolock; } - error = ocfs2_dir_foreach_blk(inode, &filp->f_version, &filp->f_pos, - dirent, filldir, NULL); + error = ocfs2_dir_foreach_blk(inode, &file->f_version, ctx, false); ocfs2_inode_unlock(inode, lock_level); if (error) @@ -2120,6 +2067,7 @@ bail: } struct ocfs2_empty_dir_priv { + struct dir_context ctx; unsigned seen_dot; unsigned seen_dot_dot; unsigned seen_other; @@ -2204,8 +2152,9 @@ out: int ocfs2_empty_dir(struct inode *inode) { int ret; - loff_t start = 0; - struct ocfs2_empty_dir_priv priv; + struct ocfs2_empty_dir_priv priv = { + .ctx.actor = ocfs2_empty_dir_filldir + }; memset(&priv, 0, sizeof(priv)); @@ -2219,7 +2168,7 @@ int ocfs2_empty_dir(struct inode *inode) */ } - ret = ocfs2_dir_foreach(inode, &start, &priv, ocfs2_empty_dir_filldir); + ret = ocfs2_dir_foreach(inode, &priv.ctx); if (ret) mlog_errno(ret); diff --git a/fs/ocfs2/dir.h b/fs/ocfs2/dir.h index e683f3deb645..f0344b75b14d 100644 --- a/fs/ocfs2/dir.h +++ b/fs/ocfs2/dir.h @@ -92,9 +92,8 @@ int ocfs2_find_files_on_disk(const char *name, struct ocfs2_dir_lookup_result *res); int ocfs2_lookup_ino_from_name(struct inode *dir, const char *name, int namelen, u64 *blkno); -int ocfs2_readdir(struct file *filp, void *dirent, filldir_t filldir); -int ocfs2_dir_foreach(struct inode *inode, loff_t *f_pos, void *priv, - filldir_t filldir); +int ocfs2_readdir(struct file *file, struct dir_context *ctx); +int ocfs2_dir_foreach(struct inode *inode, struct dir_context *ctx); int ocfs2_prepare_dir_for_insert(struct ocfs2_super *osb, struct inode *dir, struct buffer_head *parent_fe_bh, diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index b3fdd1a323d6..e68588e6b1e8 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -1408,6 +1408,7 @@ int dlm_mig_lockres_handler(struct o2net_msg *msg, u32 len, void *data, mres->lockname_len, mres->lockname); ret = -EFAULT; spin_unlock(&res->spinlock); + dlm_lockres_put(res); goto leave; } res->state |= DLM_LOCK_RES_MIGRATING; diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c index 1c39efb71bab..2487116d0d33 100644 --- a/fs/ocfs2/extent_map.c +++ b/fs/ocfs2/extent_map.c @@ -790,7 +790,7 @@ int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, &hole_size, &rec, &is_last); if (ret) { mlog_errno(ret); - goto out; + goto out_unlock; } if (rec.e_blkno == 0ULL) { diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 8a7509f9e6f5..8a38714f1d92 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -2288,7 +2288,7 @@ relock: ret = ocfs2_inode_lock(inode, NULL, 1); if (ret < 0) { mlog_errno(ret); - goto out_sems; + goto out; } ocfs2_inode_unlock(inode, 1); @@ -2712,7 +2712,7 @@ const struct file_operations ocfs2_fops = { const struct file_operations ocfs2_dops = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = ocfs2_readdir, + .iterate = ocfs2_readdir, .fsync = ocfs2_sync_file, .release = ocfs2_dir_release, .open = ocfs2_dir_open, @@ -2759,7 +2759,7 @@ const struct file_operations ocfs2_fops_no_plocks = { const struct file_operations ocfs2_dops_no_plocks = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = ocfs2_readdir, + .iterate = ocfs2_readdir, .fsync = ocfs2_sync_file, .release = ocfs2_dir_release, .open = ocfs2_dir_open, diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index 8eccfabcd12e..242170d83971 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -1941,6 +1941,7 @@ void ocfs2_orphan_scan_start(struct ocfs2_super *osb) } struct ocfs2_orphan_filldir_priv { + struct dir_context ctx; struct inode *head; struct ocfs2_super *osb; }; @@ -1977,11 +1978,11 @@ static int ocfs2_queue_orphans(struct ocfs2_super *osb, { int status; struct inode *orphan_dir_inode = NULL; - struct ocfs2_orphan_filldir_priv priv; - loff_t pos = 0; - - priv.osb = osb; - priv.head = *head; + struct ocfs2_orphan_filldir_priv priv = { + .ctx.actor = ocfs2_orphan_filldir, + .osb = osb, + .head = *head + }; orphan_dir_inode = ocfs2_get_system_file_inode(osb, ORPHAN_DIR_SYSTEM_INODE, @@ -1999,8 +2000,7 @@ static int ocfs2_queue_orphans(struct ocfs2_super *osb, goto out; } - status = ocfs2_dir_foreach(orphan_dir_inode, &pos, &priv, - ocfs2_orphan_filldir); + status = ocfs2_dir_foreach(orphan_dir_inode, &priv.ctx); if (status) { mlog_errno(status); goto out_cluster; diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index 04ee1b57c243..b4a5cdf9dbc5 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -947,7 +947,7 @@ leave: ocfs2_free_dir_lookup_result(&orphan_insert); ocfs2_free_dir_lookup_result(&lookup); - if (status) + if (status && (status != -ENOTEMPTY)) mlog_errno(status); return status; @@ -2216,7 +2216,7 @@ out: brelse(orphan_dir_bh); - return 0; + return ret; } int ocfs2_create_inode_in_orphan(struct inode *dir, diff --git a/fs/omfs/dir.c b/fs/omfs/dir.c index acbaebcad3a8..1b8e9e8405b2 100644 --- a/fs/omfs/dir.c +++ b/fs/omfs/dir.c @@ -327,26 +327,23 @@ int omfs_is_bad(struct omfs_sb_info *sbi, struct omfs_header *header, return is_bad; } -static int omfs_fill_chain(struct file *filp, void *dirent, filldir_t filldir, +static bool omfs_fill_chain(struct inode *dir, struct dir_context *ctx, u64 fsblock, int hindex) { - struct inode *dir = file_inode(filp); - struct buffer_head *bh; - struct omfs_inode *oi; - u64 self; - int res = 0; - unsigned char d_type; - /* follow chain in this bucket */ while (fsblock != ~0) { - bh = omfs_bread(dir->i_sb, fsblock); + struct buffer_head *bh = omfs_bread(dir->i_sb, fsblock); + struct omfs_inode *oi; + u64 self; + unsigned char d_type; + if (!bh) - goto out; + return true; oi = (struct omfs_inode *) bh->b_data; if (omfs_is_bad(OMFS_SB(dir->i_sb), &oi->i_head, fsblock)) { brelse(bh); - goto out; + return true; } self = fsblock; @@ -361,15 +358,16 @@ static int omfs_fill_chain(struct file *filp, void *dirent, filldir_t filldir, d_type = (oi->i_type == OMFS_DIR) ? DT_DIR : DT_REG; - res = filldir(dirent, oi->i_name, strnlen(oi->i_name, - OMFS_NAMELEN), filp->f_pos, self, d_type); + if (!dir_emit(ctx, oi->i_name, + strnlen(oi->i_name, OMFS_NAMELEN), + self, d_type)) { + brelse(bh); + return false; + } brelse(bh); - if (res < 0) - break; - filp->f_pos++; + ctx->pos++; } -out: - return res; + return true; } static int omfs_rename(struct inode *old_dir, struct dentry *old_dentry, @@ -403,60 +401,44 @@ out: return err; } -static int omfs_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int omfs_readdir(struct file *file, struct dir_context *ctx) { - struct inode *dir = file_inode(filp); + struct inode *dir = file_inode(file); struct buffer_head *bh; - loff_t offset, res; + __be64 *p; unsigned int hchain, hindex; int nbuckets; - u64 fsblock; - int ret = -EINVAL; - - if (filp->f_pos >> 32) - goto success; - - switch ((unsigned long) filp->f_pos) { - case 0: - if (filldir(dirent, ".", 1, 0, dir->i_ino, DT_DIR) < 0) - goto success; - filp->f_pos++; - /* fall through */ - case 1: - if (filldir(dirent, "..", 2, 1, - parent_ino(filp->f_dentry), DT_DIR) < 0) - goto success; - filp->f_pos = 1 << 20; - /* fall through */ + + if (ctx->pos >> 32) + return -EINVAL; + + if (ctx->pos < 1 << 20) { + if (!dir_emit_dots(file, ctx)) + return 0; + ctx->pos = 1 << 20; } nbuckets = (dir->i_size - OMFS_DIR_START) / 8; /* high 12 bits store bucket + 1 and low 20 bits store hash index */ - hchain = (filp->f_pos >> 20) - 1; - hindex = filp->f_pos & 0xfffff; + hchain = (ctx->pos >> 20) - 1; + hindex = ctx->pos & 0xfffff; bh = omfs_bread(dir->i_sb, dir->i_ino); if (!bh) - goto out; + return -EINVAL; - offset = OMFS_DIR_START + hchain * 8; + p = (__be64 *)(bh->b_data + OMFS_DIR_START) + hchain; - for (; hchain < nbuckets; hchain++, offset += 8) { - fsblock = be64_to_cpu(*((__be64 *) &bh->b_data[offset])); - - res = omfs_fill_chain(filp, dirent, filldir, fsblock, hindex); - hindex = 0; - if (res < 0) + for (; hchain < nbuckets; hchain++) { + __u64 fsblock = be64_to_cpu(*p++); + if (!omfs_fill_chain(dir, ctx, fsblock, hindex)) break; - - filp->f_pos = (hchain+2) << 20; + hindex = 0; + ctx->pos = (hchain+2) << 20; } brelse(bh); -success: - ret = 0; -out: - return ret; + return 0; } const struct inode_operations omfs_dir_inops = { @@ -470,6 +452,6 @@ const struct inode_operations omfs_dir_inops = { const struct file_operations omfs_dir_operations = { .read = generic_read_dir, - .readdir = omfs_readdir, + .iterate = omfs_readdir, .llseek = generic_file_llseek, }; diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index 75885ffde44e..8c0ceb8dd1f7 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -162,11 +162,11 @@ static const struct file_operations openpromfs_prop_ops = { .release = seq_release, }; -static int openpromfs_readdir(struct file *, void *, filldir_t); +static int openpromfs_readdir(struct file *, struct dir_context *); static const struct file_operations openprom_operations = { .read = generic_read_dir, - .readdir = openpromfs_readdir, + .iterate = openpromfs_readdir, .llseek = generic_file_llseek, }; @@ -260,71 +260,64 @@ found: return NULL; } -static int openpromfs_readdir(struct file * filp, void * dirent, filldir_t filldir) +static int openpromfs_readdir(struct file *file, struct dir_context *ctx) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct op_inode_info *oi = OP_I(inode); struct device_node *dp = oi->u.node; struct device_node *child; struct property *prop; - unsigned int ino; int i; mutex_lock(&op_mutex); - ino = inode->i_ino; - i = filp->f_pos; - switch (i) { - case 0: - if (filldir(dirent, ".", 1, i, ino, DT_DIR) < 0) + if (ctx->pos == 0) { + if (!dir_emit(ctx, ".", 1, inode->i_ino, DT_DIR)) goto out; - i++; - filp->f_pos++; - /* fall thru */ - case 1: - if (filldir(dirent, "..", 2, i, + ctx->pos = 1; + } + if (ctx->pos == 1) { + if (!dir_emit(ctx, "..", 2, (dp->parent == NULL ? OPENPROM_ROOT_INO : - dp->parent->unique_id), DT_DIR) < 0) + dp->parent->unique_id), DT_DIR)) goto out; - i++; - filp->f_pos++; - /* fall thru */ - default: - i -= 2; - - /* First, the children nodes as directories. */ - child = dp->child; - while (i && child) { - child = child->sibling; - i--; - } - while (child) { - if (filldir(dirent, - child->path_component_name, - strlen(child->path_component_name), - filp->f_pos, child->unique_id, DT_DIR) < 0) - goto out; - - filp->f_pos++; - child = child->sibling; - } + ctx->pos = 2; + } + i = ctx->pos - 2; - /* Next, the properties as files. */ - prop = dp->properties; - while (i && prop) { - prop = prop->next; - i--; - } - while (prop) { - if (filldir(dirent, prop->name, strlen(prop->name), - filp->f_pos, prop->unique_id, DT_REG) < 0) - goto out; + /* First, the children nodes as directories. */ + child = dp->child; + while (i && child) { + child = child->sibling; + i--; + } + while (child) { + if (!dir_emit(ctx, + child->path_component_name, + strlen(child->path_component_name), + child->unique_id, DT_DIR)) + goto out; - filp->f_pos++; - prop = prop->next; - } + ctx->pos++; + child = child->sibling; + } + + /* Next, the properties as files. */ + prop = dp->properties; + while (i && prop) { + prop = prop->next; + i--; } + while (prop) { + if (!dir_emit(ctx, prop->name, strlen(prop->name), + prop->unique_id, DT_REG)) + goto out; + + ctx->pos++; + prop = prop->next; + } + out: mutex_unlock(&op_mutex); return 0; diff --git a/fs/pnode.c b/fs/pnode.c index 3d2a7141b87a..9af0df15256e 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -83,7 +83,8 @@ static int do_make_slave(struct mount *mnt) if (peer_mnt == mnt) peer_mnt = NULL; } - if (IS_MNT_SHARED(mnt) && list_empty(&mnt->mnt_share)) + if (mnt->mnt_group_id && IS_MNT_SHARED(mnt) && + list_empty(&mnt->mnt_share)) mnt_release_group_id(mnt); list_del_init(&mnt->mnt_share); diff --git a/fs/proc/base.c b/fs/proc/base.c index dd51e50001fe..0016350ad95e 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1681,11 +1681,11 @@ const struct dentry_operations pid_dentry_operations = * reported by readdir in sync with the inode numbers reported * by stat. */ -int proc_fill_cache(struct file *filp, void *dirent, filldir_t filldir, +bool proc_fill_cache(struct file *file, struct dir_context *ctx, const char *name, int len, instantiate_t instantiate, struct task_struct *task, const void *ptr) { - struct dentry *child, *dir = filp->f_path.dentry; + struct dentry *child, *dir = file->f_path.dentry; struct inode *inode; struct qstr qname; ino_t ino = 0; @@ -1720,7 +1720,7 @@ end_instantiate: ino = find_inode_number(dir, &qname); if (!ino) ino = 1; - return filldir(dirent, name, len, filp->f_pos, ino, type); + return dir_emit(ctx, name, len, ino, type); } #ifdef CONFIG_CHECKPOINT_RESTORE @@ -1931,14 +1931,15 @@ static const struct inode_operations proc_map_files_inode_operations = { }; static int -proc_map_files_readdir(struct file *filp, void *dirent, filldir_t filldir) +proc_map_files_readdir(struct file *file, struct dir_context *ctx) { - struct dentry *dentry = filp->f_path.dentry; - struct inode *inode = dentry->d_inode; struct vm_area_struct *vma; struct task_struct *task; struct mm_struct *mm; - ino_t ino; + unsigned long nr_files, pos, i; + struct flex_array *fa = NULL; + struct map_files_info info; + struct map_files_info *p; int ret; ret = -EPERM; @@ -1946,7 +1947,7 @@ proc_map_files_readdir(struct file *filp, void *dirent, filldir_t filldir) goto out; ret = -ENOENT; - task = get_proc_task(inode); + task = get_proc_task(file_inode(file)); if (!task) goto out; @@ -1955,91 +1956,73 @@ proc_map_files_readdir(struct file *filp, void *dirent, filldir_t filldir) goto out_put_task; ret = 0; - switch (filp->f_pos) { - case 0: - ino = inode->i_ino; - if (filldir(dirent, ".", 1, 0, ino, DT_DIR) < 0) - goto out_put_task; - filp->f_pos++; - case 1: - ino = parent_ino(dentry); - if (filldir(dirent, "..", 2, 1, ino, DT_DIR) < 0) - goto out_put_task; - filp->f_pos++; - default: - { - unsigned long nr_files, pos, i; - struct flex_array *fa = NULL; - struct map_files_info info; - struct map_files_info *p; - - mm = get_task_mm(task); - if (!mm) - goto out_put_task; - down_read(&mm->mmap_sem); + if (!dir_emit_dots(file, ctx)) + goto out_put_task; - nr_files = 0; + mm = get_task_mm(task); + if (!mm) + goto out_put_task; + down_read(&mm->mmap_sem); - /* - * We need two passes here: - * - * 1) Collect vmas of mapped files with mmap_sem taken - * 2) Release mmap_sem and instantiate entries - * - * otherwise we get lockdep complained, since filldir() - * routine might require mmap_sem taken in might_fault(). - */ + nr_files = 0; - for (vma = mm->mmap, pos = 2; vma; vma = vma->vm_next) { - if (vma->vm_file && ++pos > filp->f_pos) - nr_files++; - } + /* + * We need two passes here: + * + * 1) Collect vmas of mapped files with mmap_sem taken + * 2) Release mmap_sem and instantiate entries + * + * otherwise we get lockdep complained, since filldir() + * routine might require mmap_sem taken in might_fault(). + */ - if (nr_files) { - fa = flex_array_alloc(sizeof(info), nr_files, - GFP_KERNEL); - if (!fa || flex_array_prealloc(fa, 0, nr_files, - GFP_KERNEL)) { - ret = -ENOMEM; - if (fa) - flex_array_free(fa); - up_read(&mm->mmap_sem); - mmput(mm); - goto out_put_task; - } - for (i = 0, vma = mm->mmap, pos = 2; vma; - vma = vma->vm_next) { - if (!vma->vm_file) - continue; - if (++pos <= filp->f_pos) - continue; - - info.mode = vma->vm_file->f_mode; - info.len = snprintf(info.name, - sizeof(info.name), "%lx-%lx", - vma->vm_start, vma->vm_end); - if (flex_array_put(fa, i++, &info, GFP_KERNEL)) - BUG(); - } + for (vma = mm->mmap, pos = 2; vma; vma = vma->vm_next) { + if (vma->vm_file && ++pos > ctx->pos) + nr_files++; + } + + if (nr_files) { + fa = flex_array_alloc(sizeof(info), nr_files, + GFP_KERNEL); + if (!fa || flex_array_prealloc(fa, 0, nr_files, + GFP_KERNEL)) { + ret = -ENOMEM; + if (fa) + flex_array_free(fa); + up_read(&mm->mmap_sem); + mmput(mm); + goto out_put_task; } - up_read(&mm->mmap_sem); - - for (i = 0; i < nr_files; i++) { - p = flex_array_get(fa, i); - ret = proc_fill_cache(filp, dirent, filldir, - p->name, p->len, - proc_map_files_instantiate, - task, - (void *)(unsigned long)p->mode); - if (ret) - break; - filp->f_pos++; + for (i = 0, vma = mm->mmap, pos = 2; vma; + vma = vma->vm_next) { + if (!vma->vm_file) + continue; + if (++pos <= ctx->pos) + continue; + + info.mode = vma->vm_file->f_mode; + info.len = snprintf(info.name, + sizeof(info.name), "%lx-%lx", + vma->vm_start, vma->vm_end); + if (flex_array_put(fa, i++, &info, GFP_KERNEL)) + BUG(); } - if (fa) - flex_array_free(fa); - mmput(mm); } + up_read(&mm->mmap_sem); + + for (i = 0; i < nr_files; i++) { + p = flex_array_get(fa, i); + if (!proc_fill_cache(file, ctx, + p->name, p->len, + proc_map_files_instantiate, + task, + (void *)(unsigned long)p->mode)) + break; + ctx->pos++; } + if (fa) + flex_array_free(fa); + mmput(mm); out_put_task: put_task_struct(task); @@ -2049,7 +2032,7 @@ out: static const struct file_operations proc_map_files_operations = { .read = generic_read_dir, - .readdir = proc_map_files_readdir, + .iterate = proc_map_files_readdir, .llseek = default_llseek, }; @@ -2118,6 +2101,7 @@ static int show_timer(struct seq_file *m, void *v) nstr[notify & ~SIGEV_THREAD_ID], (notify & SIGEV_THREAD_ID) ? "tid" : "pid", pid_nr_ns(timer->it_pid, tp->ns)); + seq_printf(m, "ClockID: %d\n", timer->it_clock); return 0; } @@ -2216,67 +2200,30 @@ out_no_task: return error; } -static int proc_pident_fill_cache(struct file *filp, void *dirent, - filldir_t filldir, struct task_struct *task, const struct pid_entry *p) -{ - return proc_fill_cache(filp, dirent, filldir, p->name, p->len, - proc_pident_instantiate, task, p); -} - -static int proc_pident_readdir(struct file *filp, - void *dirent, filldir_t filldir, +static int proc_pident_readdir(struct file *file, struct dir_context *ctx, const struct pid_entry *ents, unsigned int nents) { - int i; - struct dentry *dentry = filp->f_path.dentry; - struct inode *inode = dentry->d_inode; - struct task_struct *task = get_proc_task(inode); - const struct pid_entry *p, *last; - ino_t ino; - int ret; + struct task_struct *task = get_proc_task(file_inode(file)); + const struct pid_entry *p; - ret = -ENOENT; if (!task) - goto out_no_task; + return -ENOENT; - ret = 0; - i = filp->f_pos; - switch (i) { - case 0: - ino = inode->i_ino; - if (filldir(dirent, ".", 1, i, ino, DT_DIR) < 0) - goto out; - i++; - filp->f_pos++; - /* fall through */ - case 1: - ino = parent_ino(dentry); - if (filldir(dirent, "..", 2, i, ino, DT_DIR) < 0) - goto out; - i++; - filp->f_pos++; - /* fall through */ - default: - i -= 2; - if (i >= nents) { - ret = 1; - goto out; - } - p = ents + i; - last = &ents[nents - 1]; - while (p <= last) { - if (proc_pident_fill_cache(filp, dirent, filldir, task, p) < 0) - goto out; - filp->f_pos++; - p++; - } - } + if (!dir_emit_dots(file, ctx)) + goto out; + + if (ctx->pos >= nents + 2) + goto out; - ret = 1; + for (p = ents + (ctx->pos - 2); p <= ents + nents - 1; p++) { + if (!proc_fill_cache(file, ctx, p->name, p->len, + proc_pident_instantiate, task, p)) + break; + ctx->pos++; + } out: put_task_struct(task); -out_no_task: - return ret; + return 0; } #ifdef CONFIG_SECURITY @@ -2361,16 +2308,15 @@ static const struct pid_entry attr_dir_stuff[] = { REG("sockcreate", S_IRUGO|S_IWUGO, proc_pid_attr_operations), }; -static int proc_attr_dir_readdir(struct file * filp, - void * dirent, filldir_t filldir) +static int proc_attr_dir_readdir(struct file *file, struct dir_context *ctx) { - return proc_pident_readdir(filp,dirent,filldir, - attr_dir_stuff,ARRAY_SIZE(attr_dir_stuff)); + return proc_pident_readdir(file, ctx, + attr_dir_stuff, ARRAY_SIZE(attr_dir_stuff)); } static const struct file_operations proc_attr_dir_operations = { .read = generic_read_dir, - .readdir = proc_attr_dir_readdir, + .iterate = proc_attr_dir_readdir, .llseek = default_llseek, }; @@ -2724,16 +2670,15 @@ static const struct pid_entry tgid_base_stuff[] = { #endif }; -static int proc_tgid_base_readdir(struct file * filp, - void * dirent, filldir_t filldir) +static int proc_tgid_base_readdir(struct file *file, struct dir_context *ctx) { - return proc_pident_readdir(filp,dirent,filldir, - tgid_base_stuff,ARRAY_SIZE(tgid_base_stuff)); + return proc_pident_readdir(file, ctx, + tgid_base_stuff, ARRAY_SIZE(tgid_base_stuff)); } static const struct file_operations proc_tgid_base_operations = { .read = generic_read_dir, - .readdir = proc_tgid_base_readdir, + .iterate = proc_tgid_base_readdir, .llseek = default_llseek, }; @@ -2935,58 +2880,42 @@ retry: #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) -{ - char name[PROC_NUMBUF]; - int len = snprintf(name, sizeof(name), "%d", iter.tgid); - return proc_fill_cache(filp, dirent, filldir, name, len, - proc_pid_instantiate, iter.task, NULL); -} - -static int fake_filldir(void *buf, const char *name, int namelen, - loff_t offset, u64 ino, unsigned d_type) -{ - return 0; -} - /* for the /proc/ directory itself, after non-process stuff has been done */ -int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir) +int proc_pid_readdir(struct file *file, struct dir_context *ctx) { struct tgid_iter iter; struct pid_namespace *ns; - filldir_t __filldir; - loff_t pos = filp->f_pos; + loff_t pos = ctx->pos; if (pos >= PID_MAX_LIMIT + TGID_OFFSET) - goto out; + return 0; if (pos == TGID_OFFSET - 1) { - if (proc_fill_cache(filp, dirent, filldir, "self", 4, - NULL, NULL, NULL) < 0) - goto out; + if (!proc_fill_cache(file, ctx, "self", 4, NULL, NULL, NULL)) + return 0; iter.tgid = 0; } else { iter.tgid = pos - TGID_OFFSET; } iter.task = NULL; - ns = filp->f_dentry->d_sb->s_fs_info; + ns = file->f_dentry->d_sb->s_fs_info; for (iter = next_tgid(ns, iter); iter.task; iter.tgid += 1, iter = next_tgid(ns, iter)) { - if (has_pid_permissions(ns, iter.task, 2)) - __filldir = filldir; - else - __filldir = fake_filldir; + char name[PROC_NUMBUF]; + int len; + if (!has_pid_permissions(ns, iter.task, 2)) + continue; - filp->f_pos = iter.tgid + TGID_OFFSET; - if (proc_pid_fill_cache(filp, dirent, __filldir, iter) < 0) { + len = snprintf(name, sizeof(name), "%d", iter.tgid); + ctx->pos = iter.tgid + TGID_OFFSET; + if (!proc_fill_cache(file, ctx, name, len, + proc_pid_instantiate, iter.task, NULL)) { put_task_struct(iter.task); - goto out; + return 0; } } - filp->f_pos = PID_MAX_LIMIT + TGID_OFFSET; -out: + ctx->pos = PID_MAX_LIMIT + TGID_OFFSET; return 0; } @@ -3074,11 +3003,10 @@ static const struct pid_entry tid_base_stuff[] = { #endif }; -static int proc_tid_base_readdir(struct file * filp, - void * dirent, filldir_t filldir) +static int proc_tid_base_readdir(struct file *file, struct dir_context *ctx) { - return proc_pident_readdir(filp,dirent,filldir, - tid_base_stuff,ARRAY_SIZE(tid_base_stuff)); + return proc_pident_readdir(file, ctx, + tid_base_stuff, ARRAY_SIZE(tid_base_stuff)); } static struct dentry *proc_tid_base_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) @@ -3089,7 +3017,7 @@ static struct dentry *proc_tid_base_lookup(struct inode *dir, struct dentry *den static const struct file_operations proc_tid_base_operations = { .read = generic_read_dir, - .readdir = proc_tid_base_readdir, + .iterate = proc_tid_base_readdir, .llseek = default_llseek, }; @@ -3230,30 +3158,16 @@ static struct task_struct *next_tid(struct task_struct *start) return pos; } -static int proc_task_fill_cache(struct file *filp, void *dirent, filldir_t filldir, - struct task_struct *task, int tid) -{ - char name[PROC_NUMBUF]; - int len = snprintf(name, sizeof(name), "%d", tid); - return proc_fill_cache(filp, dirent, filldir, name, len, - proc_task_instantiate, task, NULL); -} - /* for the /proc/TGID/task/ directories */ -static int proc_task_readdir(struct file * filp, void * dirent, filldir_t filldir) +static int proc_task_readdir(struct file *file, struct dir_context *ctx) { - struct dentry *dentry = filp->f_path.dentry; - struct inode *inode = dentry->d_inode; struct task_struct *leader = NULL; - struct task_struct *task; - int retval = -ENOENT; - ino_t ino; - int tid; + struct task_struct *task = get_proc_task(file_inode(file)); struct pid_namespace *ns; + int tid; - task = get_proc_task(inode); if (!task) - goto out_no_task; + return -ENOENT; rcu_read_lock(); if (pid_alive(task)) { leader = task->group_leader; @@ -3262,46 +3176,36 @@ static int proc_task_readdir(struct file * filp, void * dirent, filldir_t filldi rcu_read_unlock(); put_task_struct(task); if (!leader) - goto out_no_task; - retval = 0; + return -ENOENT; - switch ((unsigned long)filp->f_pos) { - case 0: - ino = inode->i_ino; - if (filldir(dirent, ".", 1, filp->f_pos, ino, DT_DIR) < 0) - goto out; - filp->f_pos++; - /* fall through */ - case 1: - ino = parent_ino(dentry); - if (filldir(dirent, "..", 2, filp->f_pos, ino, DT_DIR) < 0) - goto out; - filp->f_pos++; - /* fall through */ - } + if (!dir_emit_dots(file, ctx)) + goto out; /* f_version caches the tgid value that the last readdir call couldn't * return. lseek aka telldir automagically resets f_version to 0. */ - ns = filp->f_dentry->d_sb->s_fs_info; - tid = (int)filp->f_version; - filp->f_version = 0; - for (task = first_tid(leader, tid, filp->f_pos - 2, ns); + ns = file->f_dentry->d_sb->s_fs_info; + tid = (int)file->f_version; + file->f_version = 0; + for (task = first_tid(leader, tid, ctx->pos - 2, ns); task; - task = next_tid(task), filp->f_pos++) { + task = next_tid(task), ctx->pos++) { + char name[PROC_NUMBUF]; + int len; tid = task_pid_nr_ns(task, ns); - if (proc_task_fill_cache(filp, dirent, filldir, task, tid) < 0) { + len = snprintf(name, sizeof(name), "%d", tid); + if (!proc_fill_cache(file, ctx, name, len, + proc_task_instantiate, task, NULL)) { /* returning this tgid failed, save it as the first * pid for the next readir call */ - filp->f_version = (u64)tid; + file->f_version = (u64)tid; put_task_struct(task); break; } } out: put_task_struct(leader); -out_no_task: - return retval; + return 0; } static int proc_task_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) @@ -3327,6 +3231,6 @@ static const struct inode_operations proc_task_inode_operations = { static const struct file_operations proc_task_operations = { .read = generic_read_dir, - .readdir = proc_task_readdir, + .iterate = proc_task_readdir, .llseek = default_llseek, }; diff --git a/fs/proc/fd.c b/fs/proc/fd.c index d7a4a28ef630..1441f143c43b 100644 --- a/fs/proc/fd.c +++ b/fs/proc/fd.c @@ -219,74 +219,58 @@ out_no_task: return result; } -static int proc_readfd_common(struct file * filp, void * dirent, - filldir_t filldir, instantiate_t instantiate) +static int proc_readfd_common(struct file *file, struct dir_context *ctx, + instantiate_t instantiate) { - struct dentry *dentry = filp->f_path.dentry; - struct inode *inode = dentry->d_inode; - struct task_struct *p = get_proc_task(inode); + struct task_struct *p = get_proc_task(file_inode(file)); struct files_struct *files; - unsigned int fd, ino; - int retval; + unsigned int fd; - retval = -ENOENT; if (!p) - goto out_no_task; - retval = 0; - - fd = filp->f_pos; - switch (fd) { - case 0: - if (filldir(dirent, ".", 1, 0, inode->i_ino, DT_DIR) < 0) - goto out; - filp->f_pos++; - case 1: - ino = parent_ino(dentry); - if (filldir(dirent, "..", 2, 1, ino, DT_DIR) < 0) - goto out; - filp->f_pos++; - default: - files = get_files_struct(p); - if (!files) - goto out; - rcu_read_lock(); - for (fd = filp->f_pos - 2; - fd < files_fdtable(files)->max_fds; - fd++, filp->f_pos++) { - char name[PROC_NUMBUF]; - int len; - int rv; - - if (!fcheck_files(files, fd)) - continue; - rcu_read_unlock(); + return -ENOENT; - len = snprintf(name, sizeof(name), "%d", fd); - rv = proc_fill_cache(filp, dirent, filldir, - name, len, instantiate, p, - (void *)(unsigned long)fd); - if (rv < 0) - goto out_fd_loop; - rcu_read_lock(); - } - rcu_read_unlock(); -out_fd_loop: - put_files_struct(files); + if (!dir_emit_dots(file, ctx)) + goto out; + if (!dir_emit_dots(file, ctx)) + goto out; + files = get_files_struct(p); + if (!files) + goto out; + + rcu_read_lock(); + for (fd = ctx->pos - 2; + fd < files_fdtable(files)->max_fds; + fd++, ctx->pos++) { + char name[PROC_NUMBUF]; + int len; + + if (!fcheck_files(files, fd)) + continue; + rcu_read_unlock(); + + len = snprintf(name, sizeof(name), "%d", fd); + if (!proc_fill_cache(file, ctx, + name, len, instantiate, p, + (void *)(unsigned long)fd)) + goto out_fd_loop; + rcu_read_lock(); } + rcu_read_unlock(); +out_fd_loop: + put_files_struct(files); out: put_task_struct(p); -out_no_task: - return retval; + return 0; } -static int proc_readfd(struct file *filp, void *dirent, filldir_t filldir) +static int proc_readfd(struct file *file, struct dir_context *ctx) { - return proc_readfd_common(filp, dirent, filldir, proc_fd_instantiate); + return proc_readfd_common(file, ctx, proc_fd_instantiate); } const struct file_operations proc_fd_operations = { .read = generic_read_dir, - .readdir = proc_readfd, + .iterate = proc_readfd, .llseek = default_llseek, }; @@ -351,9 +335,9 @@ proc_lookupfdinfo(struct inode *dir, struct dentry *dentry, unsigned int flags) return proc_lookupfd_common(dir, dentry, proc_fdinfo_instantiate); } -static int proc_readfdinfo(struct file *filp, void *dirent, filldir_t filldir) +static int proc_readfdinfo(struct file *file, struct dir_context *ctx) { - return proc_readfd_common(filp, dirent, filldir, + return proc_readfd_common(file, ctx, proc_fdinfo_instantiate); } @@ -364,6 +348,6 @@ const struct inode_operations proc_fdinfo_inode_operations = { const struct file_operations proc_fdinfo_operations = { .read = generic_read_dir, - .readdir = proc_readfdinfo, + .iterate = proc_readfdinfo, .llseek = default_llseek, }; diff --git a/fs/proc/generic.c b/fs/proc/generic.c index a2596afffae6..94441a407337 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -233,76 +233,52 @@ struct dentry *proc_lookup(struct inode *dir, struct dentry *dentry, * value of the readdir() call, as long as it's non-negative * for success.. */ -int proc_readdir_de(struct proc_dir_entry *de, struct file *filp, void *dirent, - filldir_t filldir) +int proc_readdir_de(struct proc_dir_entry *de, struct file *file, + struct dir_context *ctx) { - unsigned int ino; int i; - struct inode *inode = file_inode(filp); - int ret = 0; - - ino = inode->i_ino; - i = filp->f_pos; - switch (i) { - case 0: - if (filldir(dirent, ".", 1, i, ino, DT_DIR) < 0) - goto out; - i++; - filp->f_pos++; - /* fall through */ - case 1: - if (filldir(dirent, "..", 2, i, - parent_ino(filp->f_path.dentry), - DT_DIR) < 0) - goto out; - i++; - filp->f_pos++; - /* fall through */ - default: - spin_lock(&proc_subdir_lock); - de = de->subdir; - i -= 2; - for (;;) { - if (!de) { - ret = 1; - spin_unlock(&proc_subdir_lock); - goto out; - } - if (!i) - break; - de = de->next; - i--; - } - do { - struct proc_dir_entry *next; - - /* filldir passes info to user space */ - pde_get(de); - spin_unlock(&proc_subdir_lock); - if (filldir(dirent, de->name, de->namelen, filp->f_pos, - de->low_ino, de->mode >> 12) < 0) { - pde_put(de); - goto out; - } - spin_lock(&proc_subdir_lock); - filp->f_pos++; - next = de->next; - pde_put(de); - de = next; - } while (de); + if (!dir_emit_dots(file, ctx)) + return 0; + + spin_lock(&proc_subdir_lock); + de = de->subdir; + i = ctx->pos - 2; + for (;;) { + if (!de) { spin_unlock(&proc_subdir_lock); + return 0; + } + if (!i) + break; + de = de->next; + i--; } - ret = 1; -out: - return ret; + + do { + struct proc_dir_entry *next; + pde_get(de); + spin_unlock(&proc_subdir_lock); + if (!dir_emit(ctx, de->name, de->namelen, + de->low_ino, de->mode >> 12)) { + pde_put(de); + return 0; + } + spin_lock(&proc_subdir_lock); + ctx->pos++; + next = de->next; + pde_put(de); + de = next; + } while (de); + spin_unlock(&proc_subdir_lock); + return 0; } -int proc_readdir(struct file *filp, void *dirent, filldir_t filldir) +int proc_readdir(struct file *file, struct dir_context *ctx) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); - return proc_readdir_de(PDE(inode), filp, dirent, filldir); + return proc_readdir_de(PDE(inode), file, ctx); } /* @@ -313,7 +289,7 @@ int proc_readdir(struct file *filp, void *dirent, filldir_t filldir) static const struct file_operations proc_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = proc_readdir, + .iterate = proc_readdir, }; /* diff --git a/fs/proc/internal.h b/fs/proc/internal.h index d600fb098b6a..4eae2e149f31 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -165,14 +165,14 @@ 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 int proc_pid_readdir(struct file *, struct dir_context *); 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, +extern bool proc_fill_cache(struct file *, struct dir_context *, const char *, int, instantiate_t, struct task_struct *, const void *); /* @@ -183,8 +183,8 @@ extern spinlock_t proc_subdir_lock; 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); +extern int proc_readdir(struct file *, struct dir_context *); +extern int proc_readdir_de(struct proc_dir_entry *, struct file *, struct dir_context *); static inline struct proc_dir_entry *pde_get(struct proc_dir_entry *pde) { diff --git a/fs/proc/kmsg.c b/fs/proc/kmsg.c index bd4b5a740ff1..bdfabdaefdce 100644 --- a/fs/proc/kmsg.c +++ b/fs/proc/kmsg.c @@ -21,12 +21,12 @@ extern wait_queue_head_t log_wait; static int kmsg_open(struct inode * inode, struct file * file) { - return do_syslog(SYSLOG_ACTION_OPEN, NULL, 0, SYSLOG_FROM_FILE); + return do_syslog(SYSLOG_ACTION_OPEN, NULL, 0, SYSLOG_FROM_PROC); } static int kmsg_release(struct inode * inode, struct file * file) { - (void) do_syslog(SYSLOG_ACTION_CLOSE, NULL, 0, SYSLOG_FROM_FILE); + (void) do_syslog(SYSLOG_ACTION_CLOSE, NULL, 0, SYSLOG_FROM_PROC); return 0; } @@ -34,15 +34,15 @@ static ssize_t kmsg_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { if ((file->f_flags & O_NONBLOCK) && - !do_syslog(SYSLOG_ACTION_SIZE_UNREAD, NULL, 0, SYSLOG_FROM_FILE)) + !do_syslog(SYSLOG_ACTION_SIZE_UNREAD, NULL, 0, SYSLOG_FROM_PROC)) return -EAGAIN; - return do_syslog(SYSLOG_ACTION_READ, buf, count, SYSLOG_FROM_FILE); + return do_syslog(SYSLOG_ACTION_READ, buf, count, SYSLOG_FROM_PROC); } static unsigned int kmsg_poll(struct file *file, poll_table *wait) { poll_wait(file, &log_wait, wait); - if (do_syslog(SYSLOG_ACTION_SIZE_UNREAD, NULL, 0, SYSLOG_FROM_FILE)) + if (do_syslog(SYSLOG_ACTION_SIZE_UNREAD, NULL, 0, SYSLOG_FROM_PROC)) return POLLIN | POLLRDNORM; return 0; } diff --git a/fs/proc/namespaces.c b/fs/proc/namespaces.c index 54bdc6701e9f..f6abbbbfad8a 100644 --- a/fs/proc/namespaces.c +++ b/fs/proc/namespaces.c @@ -213,74 +213,36 @@ out: return error; } -static int proc_ns_fill_cache(struct file *filp, void *dirent, - filldir_t filldir, struct task_struct *task, - const struct proc_ns_operations *ops) +static int proc_ns_dir_readdir(struct file *file, struct dir_context *ctx) { - return proc_fill_cache(filp, dirent, filldir, - ops->name, strlen(ops->name), - proc_ns_instantiate, task, ops); -} - -static int proc_ns_dir_readdir(struct file *filp, void *dirent, - filldir_t filldir) -{ - int i; - struct dentry *dentry = filp->f_path.dentry; - struct inode *inode = dentry->d_inode; - struct task_struct *task = get_proc_task(inode); + struct task_struct *task = get_proc_task(file_inode(file)); const struct proc_ns_operations **entry, **last; - ino_t ino; - int ret; - ret = -ENOENT; if (!task) - goto out_no_task; + return -ENOENT; - ret = 0; - i = filp->f_pos; - switch (i) { - case 0: - ino = inode->i_ino; - if (filldir(dirent, ".", 1, i, ino, DT_DIR) < 0) - goto out; - i++; - filp->f_pos++; - /* fall through */ - case 1: - ino = parent_ino(dentry); - if (filldir(dirent, "..", 2, i, ino, DT_DIR) < 0) - goto out; - i++; - filp->f_pos++; - /* fall through */ - default: - i -= 2; - if (i >= ARRAY_SIZE(ns_entries)) { - ret = 1; - goto out; - } - entry = ns_entries + i; - last = &ns_entries[ARRAY_SIZE(ns_entries) - 1]; - while (entry <= last) { - if (proc_ns_fill_cache(filp, dirent, filldir, - task, *entry) < 0) - goto out; - filp->f_pos++; - entry++; - } + if (!dir_emit_dots(file, ctx)) + goto out; + if (ctx->pos >= 2 + ARRAY_SIZE(ns_entries)) + goto out; + entry = ns_entries + (ctx->pos - 2); + last = &ns_entries[ARRAY_SIZE(ns_entries) - 1]; + while (entry <= last) { + const struct proc_ns_operations *ops = *entry; + if (!proc_fill_cache(file, ctx, ops->name, strlen(ops->name), + proc_ns_instantiate, task, ops)) + break; + ctx->pos++; + entry++; } - - ret = 1; out: put_task_struct(task); -out_no_task: - return ret; + return 0; } const struct file_operations proc_ns_dir_operations = { .read = generic_read_dir, - .readdir = proc_ns_dir_readdir, + .iterate = proc_ns_dir_readdir, }; static struct dentry *proc_ns_dir_lookup(struct inode *dir, diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index 986e83220d56..4677bb7dc7c2 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -160,16 +160,15 @@ const struct inode_operations proc_net_inode_operations = { .getattr = proc_tgid_net_getattr, }; -static int proc_tgid_net_readdir(struct file *filp, void *dirent, - filldir_t filldir) +static int proc_tgid_net_readdir(struct file *file, struct dir_context *ctx) { int ret; struct net *net; ret = -EINVAL; - net = get_proc_task_net(file_inode(filp)); + net = get_proc_task_net(file_inode(file)); if (net != NULL) { - ret = proc_readdir_de(net->proc_net, filp, dirent, filldir); + ret = proc_readdir_de(net->proc_net, file, ctx); put_net(net); } return ret; @@ -178,7 +177,7 @@ static int proc_tgid_net_readdir(struct file *filp, void *dirent, const struct file_operations proc_net_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = proc_tgid_net_readdir, + .iterate = proc_tgid_net_readdir, }; static __net_init int proc_net_ns_init(struct net *net) diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index ac05f33a0dde..f3a570e7c257 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -573,12 +573,12 @@ out: return ret; } -static int proc_sys_fill_cache(struct file *filp, void *dirent, - filldir_t filldir, +static bool proc_sys_fill_cache(struct file *file, + struct dir_context *ctx, struct ctl_table_header *head, struct ctl_table *table) { - struct dentry *child, *dir = filp->f_path.dentry; + struct dentry *child, *dir = file->f_path.dentry; struct inode *inode; struct qstr qname; ino_t ino = 0; @@ -595,38 +595,38 @@ static int proc_sys_fill_cache(struct file *filp, void *dirent, inode = proc_sys_make_inode(dir->d_sb, head, table); if (!inode) { dput(child); - return -ENOMEM; + return false; } else { d_set_d_op(child, &proc_sys_dentry_operations); d_add(child, inode); } } else { - return -ENOMEM; + return false; } } inode = child->d_inode; ino = inode->i_ino; type = inode->i_mode >> 12; dput(child); - return !!filldir(dirent, qname.name, qname.len, filp->f_pos, ino, type); + return dir_emit(ctx, qname.name, qname.len, ino, type); } -static int proc_sys_link_fill_cache(struct file *filp, void *dirent, - filldir_t filldir, +static bool proc_sys_link_fill_cache(struct file *file, + struct dir_context *ctx, struct ctl_table_header *head, struct ctl_table *table) { - int err, ret = 0; + bool ret = true; head = sysctl_head_grab(head); if (S_ISLNK(table->mode)) { /* It is not an error if we can not follow the link ignore it */ - err = sysctl_follow_link(&head, &table, current->nsproxy); + int err = sysctl_follow_link(&head, &table, current->nsproxy); if (err) goto out; } - ret = proc_sys_fill_cache(filp, dirent, filldir, head, table); + ret = proc_sys_fill_cache(file, ctx, head, table); out: sysctl_head_finish(head); return ret; @@ -634,67 +634,50 @@ out: static int scan(struct ctl_table_header *head, ctl_table *table, unsigned long *pos, struct file *file, - void *dirent, filldir_t filldir) + struct dir_context *ctx) { - int res; + bool res; - if ((*pos)++ < file->f_pos) - return 0; + if ((*pos)++ < ctx->pos) + return true; if (unlikely(S_ISLNK(table->mode))) - res = proc_sys_link_fill_cache(file, dirent, filldir, head, table); + res = proc_sys_link_fill_cache(file, ctx, head, table); else - res = proc_sys_fill_cache(file, dirent, filldir, head, table); + res = proc_sys_fill_cache(file, ctx, head, table); - if (res == 0) - file->f_pos = *pos; + if (res) + ctx->pos = *pos; return res; } -static int proc_sys_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int proc_sys_readdir(struct file *file, struct dir_context *ctx) { - struct dentry *dentry = filp->f_path.dentry; - struct inode *inode = dentry->d_inode; - struct ctl_table_header *head = grab_header(inode); + struct ctl_table_header *head = grab_header(file_inode(file)); struct ctl_table_header *h = NULL; struct ctl_table *entry; struct ctl_dir *ctl_dir; unsigned long pos; - int ret = -EINVAL; if (IS_ERR(head)) return PTR_ERR(head); ctl_dir = container_of(head, struct ctl_dir, header); - ret = 0; - /* Avoid a switch here: arm builds fail with missing __cmpdi2 */ - if (filp->f_pos == 0) { - if (filldir(dirent, ".", 1, filp->f_pos, - inode->i_ino, DT_DIR) < 0) - goto out; - filp->f_pos++; - } - if (filp->f_pos == 1) { - if (filldir(dirent, "..", 2, filp->f_pos, - parent_ino(dentry), DT_DIR) < 0) - goto out; - filp->f_pos++; - } + if (!dir_emit_dots(file, ctx)) + return 0; + pos = 2; for (first_entry(ctl_dir, &h, &entry); h; next_entry(&h, &entry)) { - ret = scan(h, entry, &pos, filp, dirent, filldir); - if (ret) { + if (!scan(h, entry, &pos, file, ctx)) { sysctl_head_finish(h); break; } } - ret = 1; -out: sysctl_head_finish(head); - return ret; + return 0; } static int proc_sys_permission(struct inode *inode, int mask) @@ -769,7 +752,7 @@ static const struct file_operations proc_sys_file_operations = { static const struct file_operations proc_sys_dir_file_operations = { .read = generic_read_dir, - .readdir = proc_sys_readdir, + .iterate = proc_sys_readdir, .llseek = generic_file_llseek, }; diff --git a/fs/proc/root.c b/fs/proc/root.c index 41a6ea93f486..229e366598da 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -202,21 +202,14 @@ static struct dentry *proc_root_lookup(struct inode * dir, struct dentry * dentr return proc_pid_lookup(dir, dentry, flags); } -static int proc_root_readdir(struct file * filp, - void * dirent, filldir_t filldir) +static int proc_root_readdir(struct file *file, struct dir_context *ctx) { - unsigned int nr = filp->f_pos; - int ret; - - if (nr < FIRST_PROCESS_ENTRY) { - int error = proc_readdir(filp, dirent, filldir); - if (error <= 0) - return error; - filp->f_pos = FIRST_PROCESS_ENTRY; + if (ctx->pos < FIRST_PROCESS_ENTRY) { + proc_readdir(file, ctx); + ctx->pos = FIRST_PROCESS_ENTRY; } - ret = proc_pid_readdir(filp, dirent, filldir); - return ret; + return proc_pid_readdir(file, ctx); } /* @@ -226,7 +219,7 @@ static int proc_root_readdir(struct file * filp, */ static const struct file_operations proc_root_operations = { .read = generic_read_dir, - .readdir = proc_root_readdir, + .iterate = proc_root_readdir, .llseek = default_llseek, }; diff --git a/fs/qnx4/dir.c b/fs/qnx4/dir.c index 28ce014b3cef..b218f965817b 100644 --- a/fs/qnx4/dir.c +++ b/fs/qnx4/dir.c @@ -14,9 +14,9 @@ #include <linux/buffer_head.h> #include "qnx4.h" -static int qnx4_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int qnx4_readdir(struct file *file, struct dir_context *ctx) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); unsigned int offset; struct buffer_head *bh; struct qnx4_inode_entry *de; @@ -26,48 +26,44 @@ static int qnx4_readdir(struct file *filp, void *dirent, filldir_t filldir) int size; QNX4DEBUG((KERN_INFO "qnx4_readdir:i_size = %ld\n", (long) inode->i_size)); - QNX4DEBUG((KERN_INFO "filp->f_pos = %ld\n", (long) filp->f_pos)); + QNX4DEBUG((KERN_INFO "pos = %ld\n", (long) ctx->pos)); - while (filp->f_pos < inode->i_size) { - blknum = qnx4_block_map( inode, filp->f_pos >> QNX4_BLOCK_SIZE_BITS ); + while (ctx->pos < inode->i_size) { + blknum = qnx4_block_map(inode, ctx->pos >> QNX4_BLOCK_SIZE_BITS); bh = sb_bread(inode->i_sb, blknum); - if(bh==NULL) { + if (bh == NULL) { printk(KERN_ERR "qnx4_readdir: bread failed (%ld)\n", blknum); - break; + return 0; } - ix = (int)(filp->f_pos >> QNX4_DIR_ENTRY_SIZE_BITS) % QNX4_INODES_PER_BLOCK; - while (ix < QNX4_INODES_PER_BLOCK) { + ix = (ctx->pos >> QNX4_DIR_ENTRY_SIZE_BITS) % QNX4_INODES_PER_BLOCK; + for (; ix < QNX4_INODES_PER_BLOCK; ix++, ctx->pos += QNX4_DIR_ENTRY_SIZE) { offset = ix * QNX4_DIR_ENTRY_SIZE; de = (struct qnx4_inode_entry *) (bh->b_data + offset); - size = strlen(de->di_fname); - if (size) { - if ( !( de->di_status & QNX4_FILE_LINK ) && size > QNX4_SHORT_NAME_MAX ) - size = QNX4_SHORT_NAME_MAX; - else if ( size > QNX4_NAME_MAX ) - size = QNX4_NAME_MAX; - - if ( ( de->di_status & (QNX4_FILE_USED|QNX4_FILE_LINK) ) != 0 ) { - QNX4DEBUG((KERN_INFO "qnx4_readdir:%.*s\n", size, de->di_fname)); - if ( ( de->di_status & QNX4_FILE_LINK ) == 0 ) - ino = blknum * QNX4_INODES_PER_BLOCK + ix - 1; - else { - le = (struct qnx4_link_info*)de; - ino = ( le32_to_cpu(le->dl_inode_blk) - 1 ) * - QNX4_INODES_PER_BLOCK + - le->dl_inode_ndx; - } - if (filldir(dirent, de->di_fname, size, filp->f_pos, ino, DT_UNKNOWN) < 0) { - brelse(bh); - goto out; - } - } + if (!de->di_fname[0]) + continue; + if (!(de->di_status & (QNX4_FILE_USED|QNX4_FILE_LINK))) + continue; + if (!(de->di_status & QNX4_FILE_LINK)) + size = QNX4_SHORT_NAME_MAX; + else + size = QNX4_NAME_MAX; + size = strnlen(de->di_fname, size); + QNX4DEBUG((KERN_INFO "qnx4_readdir:%.*s\n", size, de->di_fname)); + if (!(de->di_status & QNX4_FILE_LINK)) + ino = blknum * QNX4_INODES_PER_BLOCK + ix - 1; + else { + le = (struct qnx4_link_info*)de; + ino = ( le32_to_cpu(le->dl_inode_blk) - 1 ) * + QNX4_INODES_PER_BLOCK + + le->dl_inode_ndx; + } + if (!dir_emit(ctx, de->di_fname, size, ino, DT_UNKNOWN)) { + brelse(bh); + return 0; } - ix++; - filp->f_pos += QNX4_DIR_ENTRY_SIZE; } brelse(bh); } -out: return 0; } @@ -75,7 +71,7 @@ const struct file_operations qnx4_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = qnx4_readdir, + .iterate = qnx4_readdir, .fsync = generic_file_fsync, }; diff --git a/fs/qnx6/dir.c b/fs/qnx6/dir.c index 8798d065e400..15b7d92ed60d 100644 --- a/fs/qnx6/dir.c +++ b/fs/qnx6/dir.c @@ -65,8 +65,8 @@ static struct qnx6_long_filename *qnx6_longname(struct super_block *sb, static int qnx6_dir_longfilename(struct inode *inode, struct qnx6_long_dir_entry *de, - void *dirent, loff_t pos, - unsigned de_inode, filldir_t filldir) + struct dir_context *ctx, + unsigned de_inode) { struct qnx6_long_filename *lf; struct super_block *s = inode->i_sb; @@ -104,8 +104,7 @@ static int qnx6_dir_longfilename(struct inode *inode, QNX6DEBUG((KERN_INFO "qnx6_readdir:%.*s inode:%u\n", lf_size, lf->lf_fname, de_inode)); - if (filldir(dirent, lf->lf_fname, lf_size, pos, de_inode, - DT_UNKNOWN) < 0) { + if (!dir_emit(ctx, lf->lf_fname, lf_size, de_inode, DT_UNKNOWN)) { qnx6_put_page(page); return 0; } @@ -115,18 +114,19 @@ static int qnx6_dir_longfilename(struct inode *inode, return 1; } -static int qnx6_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int qnx6_readdir(struct file *file, struct dir_context *ctx) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct super_block *s = inode->i_sb; struct qnx6_sb_info *sbi = QNX6_SB(s); - loff_t pos = filp->f_pos & (QNX6_DIR_ENTRY_SIZE - 1); + loff_t pos = ctx->pos & ~(QNX6_DIR_ENTRY_SIZE - 1); unsigned long npages = dir_pages(inode); unsigned long n = pos >> PAGE_CACHE_SHIFT; unsigned start = (pos & ~PAGE_CACHE_MASK) / QNX6_DIR_ENTRY_SIZE; bool done = false; - if (filp->f_pos >= inode->i_size) + ctx->pos = pos; + if (ctx->pos >= inode->i_size) return 0; for ( ; !done && n < npages; n++, start = 0) { @@ -137,11 +137,11 @@ static int qnx6_readdir(struct file *filp, void *dirent, filldir_t filldir) if (IS_ERR(page)) { printk(KERN_ERR "qnx6_readdir: read failed\n"); - filp->f_pos = (n + 1) << PAGE_CACHE_SHIFT; + ctx->pos = (n + 1) << PAGE_CACHE_SHIFT; return PTR_ERR(page); } de = ((struct qnx6_dir_entry *)page_address(page)) + start; - for (; i < limit; i++, de++, pos += QNX6_DIR_ENTRY_SIZE) { + for (; i < limit; i++, de++, ctx->pos += QNX6_DIR_ENTRY_SIZE) { int size = de->de_size; u32 no_inode = fs32_to_cpu(sbi, de->de_inode); @@ -154,8 +154,7 @@ static int qnx6_readdir(struct file *filp, void *dirent, filldir_t filldir) structure / block */ if (!qnx6_dir_longfilename(inode, (struct qnx6_long_dir_entry *)de, - dirent, pos, no_inode, - filldir)) { + ctx, no_inode)) { done = true; break; } @@ -163,9 +162,8 @@ static int qnx6_readdir(struct file *filp, void *dirent, filldir_t filldir) QNX6DEBUG((KERN_INFO "qnx6_readdir:%.*s" " inode:%u\n", size, de->de_fname, no_inode)); - if (filldir(dirent, de->de_fname, size, - pos, no_inode, DT_UNKNOWN) - < 0) { + if (!dir_emit(ctx, de->de_fname, size, + no_inode, DT_UNKNOWN)) { done = true; break; } @@ -173,7 +171,6 @@ static int qnx6_readdir(struct file *filp, void *dirent, filldir_t filldir) } qnx6_put_page(page); } - filp->f_pos = pos; return 0; } @@ -282,7 +279,7 @@ found: const struct file_operations qnx6_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = qnx6_readdir, + .iterate = qnx6_readdir, .fsync = generic_file_fsync, }; diff --git a/fs/read_write.c b/fs/read_write.c index 03430008704e..2cefa417be34 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -1064,6 +1064,7 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, struct fd in, out; struct inode *in_inode, *out_inode; loff_t pos; + loff_t out_pos; ssize_t retval; int fl; @@ -1077,12 +1078,14 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, if (!(in.file->f_mode & FMODE_READ)) goto fput_in; retval = -ESPIPE; - if (!ppos) - ppos = &in.file->f_pos; - else + if (!ppos) { + pos = in.file->f_pos; + } else { + pos = *ppos; if (!(in.file->f_mode & FMODE_PREAD)) goto fput_in; - retval = rw_verify_area(READ, in.file, ppos, count); + } + retval = rw_verify_area(READ, in.file, &pos, count); if (retval < 0) goto fput_in; count = retval; @@ -1099,7 +1102,8 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, retval = -EINVAL; in_inode = file_inode(in.file); out_inode = file_inode(out.file); - retval = rw_verify_area(WRITE, out.file, &out.file->f_pos, count); + out_pos = out.file->f_pos; + retval = rw_verify_area(WRITE, out.file, &out_pos, count); if (retval < 0) goto fput_out; count = retval; @@ -1107,7 +1111,6 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, if (!max) max = min(in_inode->i_sb->s_maxbytes, out_inode->i_sb->s_maxbytes); - pos = *ppos; if (unlikely(pos + count > max)) { retval = -EOVERFLOW; if (pos >= max) @@ -1126,18 +1129,23 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, if (in.file->f_flags & O_NONBLOCK) fl = SPLICE_F_NONBLOCK; #endif - retval = do_splice_direct(in.file, ppos, out.file, count, fl); + retval = do_splice_direct(in.file, &pos, out.file, &out_pos, count, fl); if (retval > 0) { add_rchar(current, retval); add_wchar(current, retval); fsnotify_access(in.file); fsnotify_modify(out.file); + out.file->f_pos = out_pos; + if (ppos) + *ppos = pos; + else + in.file->f_pos = pos; } inc_syscr(current); inc_syscw(current); - if (*ppos > max) + if (pos > max) retval = -EOVERFLOW; fput_out: diff --git a/fs/readdir.c b/fs/readdir.c index fee38e04fae4..93d71e574310 100644 --- a/fs/readdir.c +++ b/fs/readdir.c @@ -20,11 +20,11 @@ #include <asm/uaccess.h> -int vfs_readdir(struct file *file, filldir_t filler, void *buf) +int iterate_dir(struct file *file, struct dir_context *ctx) { struct inode *inode = file_inode(file); int res = -ENOTDIR; - if (!file->f_op || !file->f_op->readdir) + if (!file->f_op || !file->f_op->iterate) goto out; res = security_file_permission(file, MAY_READ); @@ -37,15 +37,16 @@ int vfs_readdir(struct file *file, filldir_t filler, void *buf) res = -ENOENT; if (!IS_DEADDIR(inode)) { - res = file->f_op->readdir(file, buf, filler); + ctx->pos = file->f_pos; + res = file->f_op->iterate(file, ctx); + file->f_pos = ctx->pos; file_accessed(file); } mutex_unlock(&inode->i_mutex); out: return res; } - -EXPORT_SYMBOL(vfs_readdir); +EXPORT_SYMBOL(iterate_dir); /* * Traditional linux readdir() handling.. @@ -66,6 +67,7 @@ struct old_linux_dirent { }; struct readdir_callback { + struct dir_context ctx; struct old_linux_dirent __user * dirent; int result; }; @@ -73,7 +75,7 @@ struct readdir_callback { static int fillonedir(void * __buf, const char * name, int namlen, loff_t offset, u64 ino, unsigned int d_type) { - struct readdir_callback * buf = (struct readdir_callback *) __buf; + struct readdir_callback *buf = (struct readdir_callback *) __buf; struct old_linux_dirent __user * dirent; unsigned long d_ino; @@ -107,15 +109,15 @@ SYSCALL_DEFINE3(old_readdir, unsigned int, fd, { int error; struct fd f = fdget(fd); - struct readdir_callback buf; + struct readdir_callback buf = { + .ctx.actor = fillonedir, + .dirent = dirent + }; if (!f.file) return -EBADF; - buf.result = 0; - buf.dirent = dirent; - - error = vfs_readdir(f.file, fillonedir, &buf); + error = iterate_dir(f.file, &buf.ctx); if (buf.result) error = buf.result; @@ -137,6 +139,7 @@ struct linux_dirent { }; struct getdents_callback { + struct dir_context ctx; struct linux_dirent __user * current_dir; struct linux_dirent __user * previous; int count; @@ -191,7 +194,11 @@ SYSCALL_DEFINE3(getdents, unsigned int, fd, { struct fd f; struct linux_dirent __user * lastdirent; - struct getdents_callback buf; + struct getdents_callback buf = { + .ctx.actor = filldir, + .count = count, + .current_dir = dirent + }; int error; if (!access_ok(VERIFY_WRITE, dirent, count)) @@ -201,17 +208,12 @@ SYSCALL_DEFINE3(getdents, unsigned int, fd, if (!f.file) return -EBADF; - buf.current_dir = dirent; - buf.previous = NULL; - buf.count = count; - buf.error = 0; - - error = vfs_readdir(f.file, filldir, &buf); + error = iterate_dir(f.file, &buf.ctx); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - if (put_user(f.file->f_pos, &lastdirent->d_off)) + if (put_user(buf.ctx.pos, &lastdirent->d_off)) error = -EFAULT; else error = count - buf.count; @@ -221,6 +223,7 @@ SYSCALL_DEFINE3(getdents, unsigned int, fd, } struct getdents_callback64 { + struct dir_context ctx; struct linux_dirent64 __user * current_dir; struct linux_dirent64 __user * previous; int count; @@ -271,7 +274,11 @@ SYSCALL_DEFINE3(getdents64, unsigned int, fd, { struct fd f; struct linux_dirent64 __user * lastdirent; - struct getdents_callback64 buf; + struct getdents_callback64 buf = { + .ctx.actor = filldir64, + .count = count, + .current_dir = dirent + }; int error; if (!access_ok(VERIFY_WRITE, dirent, count)) @@ -281,17 +288,12 @@ SYSCALL_DEFINE3(getdents64, unsigned int, fd, if (!f.file) return -EBADF; - buf.current_dir = dirent; - buf.previous = NULL; - buf.count = count; - buf.error = 0; - - error = vfs_readdir(f.file, filldir64, &buf); + error = iterate_dir(f.file, &buf.ctx); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - typeof(lastdirent->d_off) d_off = f.file->f_pos; + typeof(lastdirent->d_off) d_off = buf.ctx.pos; if (__put_user(d_off, &lastdirent->d_off)) error = -EFAULT; else diff --git a/fs/reiserfs/dir.c b/fs/reiserfs/dir.c index 66c53b642a88..03e4ca5624d6 100644 --- a/fs/reiserfs/dir.c +++ b/fs/reiserfs/dir.c @@ -13,14 +13,14 @@ extern const struct reiserfs_key MIN_KEY; -static int reiserfs_readdir(struct file *, void *, filldir_t); +static int reiserfs_readdir(struct file *, struct dir_context *); static int reiserfs_dir_fsync(struct file *filp, loff_t start, loff_t end, int datasync); const struct file_operations reiserfs_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = reiserfs_readdir, + .iterate = reiserfs_readdir, .fsync = reiserfs_dir_fsync, .unlocked_ioctl = reiserfs_ioctl, #ifdef CONFIG_COMPAT @@ -50,18 +50,15 @@ static int reiserfs_dir_fsync(struct file *filp, loff_t start, loff_t end, #define store_ih(where,what) copy_item_head (where, what) -static inline bool is_privroot_deh(struct dentry *dir, - struct reiserfs_de_head *deh) +static inline bool is_privroot_deh(struct inode *dir, struct reiserfs_de_head *deh) { - struct dentry *privroot = REISERFS_SB(dir->d_sb)->priv_root; - return (dir == dir->d_parent && privroot->d_inode && + struct dentry *privroot = REISERFS_SB(dir->i_sb)->priv_root; + return (privroot->d_inode && deh->deh_objectid == INODE_PKEY(privroot->d_inode)->k_objectid); } -int reiserfs_readdir_dentry(struct dentry *dentry, void *dirent, - filldir_t filldir, loff_t *pos) +int reiserfs_readdir_inode(struct inode *inode, struct dir_context *ctx) { - struct inode *inode = dentry->d_inode; struct cpu_key pos_key; /* key of current position in the directory (key of directory entry) */ INITIALIZE_PATH(path_to_entry); struct buffer_head *bh; @@ -81,7 +78,7 @@ int reiserfs_readdir_dentry(struct dentry *dentry, void *dirent, /* form key for search the next directory entry using f_pos field of file structure */ - make_cpu_key(&pos_key, inode, *pos ?: DOT_OFFSET, TYPE_DIRENTRY, 3); + make_cpu_key(&pos_key, inode, ctx->pos ?: DOT_OFFSET, TYPE_DIRENTRY, 3); next_pos = cpu_key_k_offset(&pos_key); path_to_entry.reada = PATH_READA; @@ -126,7 +123,6 @@ int reiserfs_readdir_dentry(struct dentry *dentry, void *dirent, entry_num++, deh++) { int d_reclen; char *d_name; - off_t d_off; ino_t d_ino; if (!de_visible(deh)) @@ -155,11 +151,10 @@ int reiserfs_readdir_dentry(struct dentry *dentry, void *dirent, } /* Ignore the .reiserfs_priv entry */ - if (is_privroot_deh(dentry, deh)) + if (is_privroot_deh(inode, deh)) continue; - d_off = deh_offset(deh); - *pos = d_off; + ctx->pos = deh_offset(deh); d_ino = deh_objectid(deh); if (d_reclen <= 32) { local_buf = small_buf; @@ -187,9 +182,9 @@ int reiserfs_readdir_dentry(struct dentry *dentry, void *dirent, * the write lock here for other waiters */ reiserfs_write_unlock(inode->i_sb); - if (filldir - (dirent, local_buf, d_reclen, d_off, d_ino, - DT_UNKNOWN) < 0) { + if (!dir_emit + (ctx, local_buf, d_reclen, d_ino, + DT_UNKNOWN)) { reiserfs_write_lock(inode->i_sb); if (local_buf != small_buf) { kfree(local_buf); @@ -204,6 +199,8 @@ int reiserfs_readdir_dentry(struct dentry *dentry, void *dirent, next_pos = deh_offset(deh) + 1; if (item_moved(&tmp_ih, &path_to_entry)) { + set_cpu_key_k_offset(&pos_key, + next_pos); goto research; } } /* for */ @@ -235,7 +232,7 @@ int reiserfs_readdir_dentry(struct dentry *dentry, void *dirent, } /* while */ end: - *pos = next_pos; + ctx->pos = next_pos; pathrelse(&path_to_entry); reiserfs_check_path(&path_to_entry); out: @@ -243,10 +240,9 @@ out: return ret; } -static int reiserfs_readdir(struct file *file, void *dirent, filldir_t filldir) +static int reiserfs_readdir(struct file *file, struct dir_context *ctx) { - struct dentry *dentry = file->f_path.dentry; - return reiserfs_readdir_dentry(dentry, dirent, filldir, &file->f_pos); + return reiserfs_readdir_inode(file_inode(file), ctx); } /* compose directory item containing "." and ".." entries (entries are diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 76c8faddf137..0048cc16a6a8 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -1811,11 +1811,16 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th, TYPE_STAT_DATA, SD_SIZE, MAX_US_INT); memcpy(INODE_PKEY(inode), &(ih.ih_key), KEY_SIZE); args.dirid = le32_to_cpu(ih.ih_key.k_dir_id); - if (insert_inode_locked4(inode, args.objectid, - reiserfs_find_actor, &args) < 0) { + + reiserfs_write_unlock(inode->i_sb); + err = insert_inode_locked4(inode, args.objectid, + reiserfs_find_actor, &args); + reiserfs_write_lock(inode->i_sb); + if (err) { err = -EINVAL; goto out_bad_inode; } + if (old_format_only(sb)) /* not a perfect generation count, as object ids can be reused, but ** this is as good as reiserfs can do right now. diff --git a/fs/reiserfs/reiserfs.h b/fs/reiserfs/reiserfs.h index 157e474ab303..3df5ce6c724d 100644 --- a/fs/reiserfs/reiserfs.h +++ b/fs/reiserfs/reiserfs.h @@ -2709,7 +2709,7 @@ extern const struct inode_operations reiserfs_dir_inode_operations; extern const struct inode_operations reiserfs_symlink_inode_operations; extern const struct inode_operations reiserfs_special_inode_operations; extern const struct file_operations reiserfs_dir_operations; -int reiserfs_readdir_dentry(struct dentry *, void *, filldir_t, loff_t *); +int reiserfs_readdir_inode(struct inode *, struct dir_context *); /* tail_conversion.c */ int direct2indirect(struct reiserfs_transaction_handle *, struct inode *, diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 4cce1d9552fb..c69cdd749f09 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -171,6 +171,7 @@ static struct dentry *open_xa_dir(const struct inode *inode, int flags) * modifying extended attributes. This includes operations such as permissions * or ownership changes, object deletions, etc. */ struct reiserfs_dentry_buf { + struct dir_context ctx; struct dentry *xadir; int count; struct dentry *dentries[8]; @@ -223,9 +224,8 @@ static int reiserfs_for_each_xattr(struct inode *inode, { struct dentry *dir; int i, err = 0; - loff_t pos = 0; struct reiserfs_dentry_buf buf = { - .count = 0, + .ctx.actor = fill_with_dentries, }; /* Skip out, an xattr has no xattrs associated with it */ @@ -249,29 +249,27 @@ static int reiserfs_for_each_xattr(struct inode *inode, reiserfs_write_lock(inode->i_sb); buf.xadir = dir; - err = reiserfs_readdir_dentry(dir, &buf, fill_with_dentries, &pos); - while ((err == 0 || err == -ENOSPC) && buf.count) { - err = 0; - - for (i = 0; i < buf.count && buf.dentries[i]; i++) { - int lerr = 0; + while (1) { + err = reiserfs_readdir_inode(dir->d_inode, &buf.ctx); + if (err) + break; + if (!buf.count) + break; + for (i = 0; !err && i < buf.count && buf.dentries[i]; i++) { struct dentry *dentry = buf.dentries[i]; - if (err == 0 && !S_ISDIR(dentry->d_inode->i_mode)) - lerr = action(dentry, data); + if (!S_ISDIR(dentry->d_inode->i_mode)) + err = action(dentry, data); dput(dentry); buf.dentries[i] = NULL; - err = lerr ?: err; } + if (err) + break; buf.count = 0; - if (!err) - err = reiserfs_readdir_dentry(dir, &buf, - fill_with_dentries, &pos); } mutex_unlock(&dir->d_inode->i_mutex); - /* Clean up after a failed readdir */ cleanup_dentry_buf(&buf); if (!err) { @@ -318,7 +316,19 @@ static int delete_one_xattr(struct dentry *dentry, void *data) static int chown_one_xattr(struct dentry *dentry, void *data) { struct iattr *attrs = data; - return reiserfs_setattr(dentry, attrs); + int ia_valid = attrs->ia_valid; + int err; + + /* + * We only want the ownership bits. Otherwise, we'll do + * things like change a directory to a regular file if + * ATTR_MODE is set. + */ + attrs->ia_valid &= (ATTR_UID|ATTR_GID); + err = reiserfs_setattr(dentry, attrs); + attrs->ia_valid = ia_valid; + + return err; } /* No i_mutex, but the inode is unconnected. */ @@ -788,6 +798,7 @@ int reiserfs_removexattr(struct dentry *dentry, const char *name) } struct listxattr_buf { + struct dir_context ctx; size_t size; size_t pos; char *buf; @@ -833,8 +844,8 @@ ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size) { struct dentry *dir; int err = 0; - loff_t pos = 0; struct listxattr_buf buf = { + .ctx.actor = listxattr_filler, .dentry = dentry, .buf = buffer, .size = buffer ? size : 0, @@ -856,7 +867,7 @@ ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size) } mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_XATTR); - err = reiserfs_readdir_dentry(dir, &buf, listxattr_filler, &pos); + err = reiserfs_readdir_inode(dir->d_inode, &buf.ctx); mutex_unlock(&dir->d_inode->i_mutex); if (!err) diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c index d7c01ef64eda..6c8767fdfc6a 100644 --- a/fs/reiserfs/xattr_acl.c +++ b/fs/reiserfs/xattr_acl.c @@ -443,6 +443,9 @@ int reiserfs_acl_chmod(struct inode *inode) int depth; int error; + if (IS_PRIVATE(inode)) + return 0; + if (S_ISLNK(inode->i_mode)) return -EOPNOTSUPP; diff --git a/fs/romfs/super.c b/fs/romfs/super.c index 15cbc41ee365..ff1d3d42e72a 100644 --- a/fs/romfs/super.c +++ b/fs/romfs/super.c @@ -145,19 +145,18 @@ static const struct address_space_operations romfs_aops = { /* * read the entries from a directory */ -static int romfs_readdir(struct file *filp, void *dirent, filldir_t filldir) +static int romfs_readdir(struct file *file, struct dir_context *ctx) { - struct inode *i = file_inode(filp); + struct inode *i = file_inode(file); struct romfs_inode ri; unsigned long offset, maxoff; int j, ino, nextfh; - int stored = 0; char fsname[ROMFS_MAXFN]; /* XXX dynamic? */ int ret; maxoff = romfs_maxsize(i->i_sb); - offset = filp->f_pos; + offset = ctx->pos; if (!offset) { offset = i->i_ino & ROMFH_MASK; ret = romfs_dev_read(i->i_sb, offset, &ri, ROMFH_SIZE); @@ -170,10 +169,10 @@ static int romfs_readdir(struct file *filp, void *dirent, filldir_t filldir) for (;;) { if (!offset || offset >= maxoff) { offset = maxoff; - filp->f_pos = offset; + ctx->pos = offset; goto out; } - filp->f_pos = offset; + ctx->pos = offset; /* Fetch inode info */ ret = romfs_dev_read(i->i_sb, offset, &ri, ROMFH_SIZE); @@ -194,16 +193,14 @@ static int romfs_readdir(struct file *filp, void *dirent, filldir_t filldir) nextfh = be32_to_cpu(ri.next); if ((nextfh & ROMFH_TYPE) == ROMFH_HRD) ino = be32_to_cpu(ri.spec); - if (filldir(dirent, fsname, j, offset, ino, - romfs_dtype_table[nextfh & ROMFH_TYPE]) < 0) + if (!dir_emit(ctx, fsname, j, ino, + romfs_dtype_table[nextfh & ROMFH_TYPE])) goto out; - stored++; offset = nextfh & ROMFH_MASK; } - out: - return stored; + return 0; } /* @@ -281,7 +278,7 @@ error: static const struct file_operations romfs_dir_operations = { .read = generic_read_dir, - .readdir = romfs_readdir, + .iterate = romfs_readdir, .llseek = default_llseek, }; diff --git a/fs/splice.c b/fs/splice.c index e6b25598c8c4..d37431dd60a1 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -1274,7 +1274,7 @@ static int direct_splice_actor(struct pipe_inode_info *pipe, { struct file *file = sd->u.file; - return do_splice_from(pipe, file, &file->f_pos, sd->total_len, + return do_splice_from(pipe, file, sd->opos, sd->total_len, sd->flags); } @@ -1283,6 +1283,7 @@ static int direct_splice_actor(struct pipe_inode_info *pipe, * @in: file to splice from * @ppos: input file offset * @out: file to splice to + * @opos: output file offset * @len: number of bytes to splice * @flags: splice modifier flags * @@ -1294,7 +1295,7 @@ static int direct_splice_actor(struct pipe_inode_info *pipe, * */ long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, - size_t len, unsigned int flags) + loff_t *opos, size_t len, unsigned int flags) { struct splice_desc sd = { .len = len, @@ -1302,6 +1303,7 @@ long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, .flags = flags, .pos = *ppos, .u.file = out, + .opos = opos, }; long ret; @@ -1325,7 +1327,7 @@ static long do_splice(struct file *in, loff_t __user *off_in, { struct pipe_inode_info *ipipe; struct pipe_inode_info *opipe; - loff_t offset, *off; + loff_t offset; long ret; ipipe = get_pipe_info(in); @@ -1356,13 +1358,15 @@ static long do_splice(struct file *in, loff_t __user *off_in, return -EINVAL; if (copy_from_user(&offset, off_out, sizeof(loff_t))) return -EFAULT; - off = &offset; - } else - off = &out->f_pos; + } else { + offset = out->f_pos; + } - ret = do_splice_from(ipipe, out, off, len, flags); + ret = do_splice_from(ipipe, out, &offset, len, flags); - if (off_out && copy_to_user(off_out, off, sizeof(loff_t))) + if (!off_out) + out->f_pos = offset; + else if (copy_to_user(off_out, &offset, sizeof(loff_t))) ret = -EFAULT; return ret; @@ -1376,13 +1380,15 @@ static long do_splice(struct file *in, loff_t __user *off_in, return -EINVAL; if (copy_from_user(&offset, off_in, sizeof(loff_t))) return -EFAULT; - off = &offset; - } else - off = &in->f_pos; + } else { + offset = in->f_pos; + } - ret = do_splice_to(in, off, opipe, len, flags); + ret = do_splice_to(in, &offset, opipe, len, flags); - if (off_in && copy_to_user(off_in, off, sizeof(loff_t))) + if (!off_in) + in->f_pos = offset; + else if (copy_to_user(off_in, &offset, sizeof(loff_t))) ret = -EFAULT; return ret; diff --git a/fs/squashfs/dir.c b/fs/squashfs/dir.c index 57dc70ebbb19..f7f527bf8c10 100644 --- a/fs/squashfs/dir.c +++ b/fs/squashfs/dir.c @@ -100,7 +100,7 @@ static int get_dir_index_using_offset(struct super_block *sb, } -static int squashfs_readdir(struct file *file, void *dirent, filldir_t filldir) +static int squashfs_readdir(struct file *file, struct dir_context *ctx) { struct inode *inode = file_inode(file); struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info; @@ -127,11 +127,11 @@ static int squashfs_readdir(struct file *file, void *dirent, filldir_t filldir) * It also means that the external f_pos is offset by 3 from the * on-disk directory f_pos. */ - while (file->f_pos < 3) { + while (ctx->pos < 3) { char *name; int i_ino; - if (file->f_pos == 0) { + if (ctx->pos == 0) { name = "."; size = 1; i_ino = inode->i_ino; @@ -141,24 +141,18 @@ static int squashfs_readdir(struct file *file, void *dirent, filldir_t filldir) i_ino = squashfs_i(inode)->parent; } - TRACE("Calling filldir(%p, %s, %d, %lld, %d, %d)\n", - dirent, name, size, file->f_pos, i_ino, - squashfs_filetype_table[1]); - - if (filldir(dirent, name, size, file->f_pos, i_ino, - squashfs_filetype_table[1]) < 0) { - TRACE("Filldir returned less than 0\n"); + if (!dir_emit(ctx, name, size, i_ino, + squashfs_filetype_table[1])) goto finish; - } - file->f_pos += size; + ctx->pos += size; } length = get_dir_index_using_offset(inode->i_sb, &block, &offset, squashfs_i(inode)->dir_idx_start, squashfs_i(inode)->dir_idx_offset, squashfs_i(inode)->dir_idx_cnt, - file->f_pos); + ctx->pos); while (length < i_size_read(inode)) { /* @@ -198,7 +192,7 @@ static int squashfs_readdir(struct file *file, void *dirent, filldir_t filldir) length += sizeof(*dire) + size; - if (file->f_pos >= length) + if (ctx->pos >= length) continue; dire->name[size] = '\0'; @@ -206,22 +200,12 @@ static int squashfs_readdir(struct file *file, void *dirent, filldir_t filldir) ((short) le16_to_cpu(dire->inode_number)); type = le16_to_cpu(dire->type); - TRACE("Calling filldir(%p, %s, %d, %lld, %x:%x, %d, %d)" - "\n", dirent, dire->name, size, - file->f_pos, - le32_to_cpu(dirh.start_block), - le16_to_cpu(dire->offset), - inode_number, - squashfs_filetype_table[type]); - - if (filldir(dirent, dire->name, size, file->f_pos, + if (!dir_emit(ctx, dire->name, size, inode_number, - squashfs_filetype_table[type]) < 0) { - TRACE("Filldir returned less than 0\n"); + squashfs_filetype_table[type])) goto finish; - } - file->f_pos = length; + ctx->pos = length; } } @@ -238,6 +222,6 @@ failed_read: const struct file_operations squashfs_dir_ops = { .read = generic_read_dir, - .readdir = squashfs_readdir, + .iterate = squashfs_readdir, .llseek = default_llseek, }; diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index e8e0e71b29d5..4cfd742d260d 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -998,68 +998,38 @@ static struct sysfs_dirent *sysfs_dir_next_pos(const void *ns, return pos; } -static int sysfs_readdir(struct file * filp, void * dirent, filldir_t filldir) +static int sysfs_readdir(struct file *file, struct dir_context *ctx) { - struct dentry *dentry = filp->f_path.dentry; + struct dentry *dentry = file->f_path.dentry; struct sysfs_dirent * parent_sd = dentry->d_fsdata; - struct sysfs_dirent *pos = filp->private_data; + struct sysfs_dirent *pos = file->private_data; enum kobj_ns_type type; const void *ns; - ino_t ino; - loff_t off; type = sysfs_ns_type(parent_sd); ns = sysfs_info(dentry->d_sb)->ns[type]; - if (filp->f_pos == 0) { - ino = parent_sd->s_ino; - if (filldir(dirent, ".", 1, filp->f_pos, ino, DT_DIR) == 0) - filp->f_pos++; - else - return 0; - } - if (filp->f_pos == 1) { - if (parent_sd->s_parent) - ino = parent_sd->s_parent->s_ino; - else - ino = parent_sd->s_ino; - if (filldir(dirent, "..", 2, filp->f_pos, ino, DT_DIR) == 0) - filp->f_pos++; - else - return 0; - } + if (!dir_emit_dots(file, ctx)) + return 0; mutex_lock(&sysfs_mutex); - off = filp->f_pos; - for (pos = sysfs_dir_pos(ns, parent_sd, filp->f_pos, pos); + for (pos = sysfs_dir_pos(ns, parent_sd, ctx->pos, pos); pos; - pos = sysfs_dir_next_pos(ns, parent_sd, filp->f_pos, pos)) { - const char * name; - unsigned int type; - int len, ret; - - name = pos->s_name; - len = strlen(name); - ino = pos->s_ino; - type = dt_type(pos); - off = filp->f_pos = pos->s_hash; - filp->private_data = sysfs_get(pos); + pos = sysfs_dir_next_pos(ns, parent_sd, ctx->pos, pos)) { + const char *name = pos->s_name; + unsigned int type = dt_type(pos); + int len = strlen(name); + ino_t ino = pos->s_ino; + ctx->pos = pos->s_hash; + file->private_data = sysfs_get(pos); mutex_unlock(&sysfs_mutex); - ret = filldir(dirent, name, len, off, ino, type); + if (!dir_emit(ctx, name, len, ino, type)) + return 0; mutex_lock(&sysfs_mutex); - if (ret < 0) - break; } mutex_unlock(&sysfs_mutex); - - /* don't reference last entry if its refcount is dropped */ - if (!pos) { - filp->private_data = NULL; - - /* EOF and not changed as 0 or 1 in read/write path */ - if (off == filp->f_pos && off > 1) - filp->f_pos = INT_MAX; - } + file->private_data = NULL; + ctx->pos = INT_MAX; return 0; } @@ -1077,7 +1047,7 @@ static loff_t sysfs_dir_llseek(struct file *file, loff_t offset, int whence) const struct file_operations sysfs_dir_operations = { .read = generic_read_dir, - .readdir = sysfs_readdir, + .iterate = sysfs_readdir, .release = sysfs_dir_release, .llseek = sysfs_dir_llseek, }; diff --git a/fs/sysv/dir.c b/fs/sysv/dir.c index 3799e8dac3eb..d42291d08215 100644 --- a/fs/sysv/dir.c +++ b/fs/sysv/dir.c @@ -18,12 +18,12 @@ #include <linux/swap.h> #include "sysv.h" -static int sysv_readdir(struct file *, void *, filldir_t); +static int sysv_readdir(struct file *, struct dir_context *); const struct file_operations sysv_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = sysv_readdir, + .iterate = sysv_readdir, .fsync = generic_file_fsync, }; @@ -65,18 +65,21 @@ static struct page * dir_get_page(struct inode *dir, unsigned long n) return page; } -static int sysv_readdir(struct file * filp, void * dirent, filldir_t filldir) +static int sysv_readdir(struct file *file, struct dir_context *ctx) { - unsigned long pos = filp->f_pos; - struct inode *inode = file_inode(filp); + unsigned long pos = ctx->pos; + struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; - unsigned offset = pos & ~PAGE_CACHE_MASK; - unsigned long n = pos >> PAGE_CACHE_SHIFT; unsigned long npages = dir_pages(inode); + unsigned offset; + unsigned long n; - pos = (pos + SYSV_DIRSIZE-1) & ~(SYSV_DIRSIZE-1); + ctx->pos = pos = (pos + SYSV_DIRSIZE-1) & ~(SYSV_DIRSIZE-1); if (pos >= inode->i_size) - goto done; + return 0; + + offset = pos & ~PAGE_CACHE_MASK; + n = pos >> PAGE_CACHE_SHIFT; for ( ; n < npages; n++, offset = 0) { char *kaddr, *limit; @@ -88,29 +91,21 @@ static int sysv_readdir(struct file * filp, void * dirent, filldir_t filldir) kaddr = (char *)page_address(page); de = (struct sysv_dir_entry *)(kaddr+offset); limit = kaddr + PAGE_CACHE_SIZE - SYSV_DIRSIZE; - for ( ;(char*)de <= limit; de++) { + for ( ;(char*)de <= limit; de++, ctx->pos += sizeof(*de)) { char *name = de->name; - int over; if (!de->inode) continue; - offset = (char *)de - kaddr; - - over = filldir(dirent, name, strnlen(name,SYSV_NAMELEN), - ((loff_t)n<<PAGE_CACHE_SHIFT) | offset, + if (!dir_emit(ctx, name, strnlen(name,SYSV_NAMELEN), fs16_to_cpu(SYSV_SB(sb), de->inode), - DT_UNKNOWN); - if (over) { + DT_UNKNOWN)) { dir_put_page(page); - goto done; + return 0; } } dir_put_page(page); } - -done: - filp->f_pos = ((loff_t)n << PAGE_CACHE_SHIFT) | offset; return 0; } diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c index de08c92f2e23..6b4947f75af7 100644 --- a/fs/ubifs/dir.c +++ b/fs/ubifs/dir.c @@ -346,38 +346,46 @@ static unsigned int vfs_dent_type(uint8_t type) * This means that UBIFS cannot support NFS which requires full * 'seekdir()'/'telldir()' support. */ -static int ubifs_readdir(struct file *file, void *dirent, filldir_t filldir) +static int ubifs_readdir(struct file *file, struct dir_context *ctx) { - int err, over = 0; + int err; struct qstr nm; union ubifs_key key; struct ubifs_dent_node *dent; struct inode *dir = file_inode(file); struct ubifs_info *c = dir->i_sb->s_fs_info; - dbg_gen("dir ino %lu, f_pos %#llx", dir->i_ino, file->f_pos); + dbg_gen("dir ino %lu, f_pos %#llx", dir->i_ino, ctx->pos); - if (file->f_pos > UBIFS_S_KEY_HASH_MASK || file->f_pos == 2) + if (ctx->pos > UBIFS_S_KEY_HASH_MASK || ctx->pos == 2) /* * The directory was seek'ed to a senseless position or there * are no more entries. */ return 0; - /* File positions 0 and 1 correspond to "." and ".." */ - if (file->f_pos == 0) { - ubifs_assert(!file->private_data); - over = filldir(dirent, ".", 1, 0, dir->i_ino, DT_DIR); - if (over) - return 0; - file->f_pos = 1; + if (file->f_version == 0) { + /* + * The file was seek'ed, which means that @file->private_data + * is now invalid. This may also be just the first + * 'ubifs_readdir()' invocation, in which case + * @file->private_data is NULL, and the below code is + * basically a no-op. + */ + kfree(file->private_data); + file->private_data = NULL; } - if (file->f_pos == 1) { + /* + * 'generic_file_llseek()' unconditionally sets @file->f_version to + * zero, and we use this for detecting whether the file was seek'ed. + */ + file->f_version = 1; + + /* File positions 0 and 1 correspond to "." and ".." */ + if (ctx->pos < 2) { ubifs_assert(!file->private_data); - over = filldir(dirent, "..", 2, 1, - parent_ino(file->f_path.dentry), DT_DIR); - if (over) + if (!dir_emit_dots(file, ctx)) return 0; /* Find the first entry in TNC and save it */ @@ -389,7 +397,7 @@ static int ubifs_readdir(struct file *file, void *dirent, filldir_t filldir) goto out; } - file->f_pos = key_hash_flash(c, &dent->key); + ctx->pos = key_hash_flash(c, &dent->key); file->private_data = dent; } @@ -397,17 +405,16 @@ static int ubifs_readdir(struct file *file, void *dirent, filldir_t filldir) if (!dent) { /* * The directory was seek'ed to and is now readdir'ed. - * Find the entry corresponding to @file->f_pos or the - * closest one. + * Find the entry corresponding to @ctx->pos or the closest one. */ - dent_key_init_hash(c, &key, dir->i_ino, file->f_pos); + dent_key_init_hash(c, &key, dir->i_ino, ctx->pos); nm.name = NULL; dent = ubifs_tnc_next_ent(c, &key, &nm); if (IS_ERR(dent)) { err = PTR_ERR(dent); goto out; } - file->f_pos = key_hash_flash(c, &dent->key); + ctx->pos = key_hash_flash(c, &dent->key); file->private_data = dent; } @@ -419,10 +426,9 @@ static int ubifs_readdir(struct file *file, void *dirent, filldir_t filldir) ubifs_inode(dir)->creat_sqnum); nm.len = le16_to_cpu(dent->nlen); - over = filldir(dirent, dent->name, nm.len, file->f_pos, + if (!dir_emit(ctx, dent->name, nm.len, le64_to_cpu(dent->inum), - vfs_dent_type(dent->type)); - if (over) + vfs_dent_type(dent->type))) return 0; /* Switch to the next entry */ @@ -435,7 +441,7 @@ static int ubifs_readdir(struct file *file, void *dirent, filldir_t filldir) } kfree(file->private_data); - file->f_pos = key_hash_flash(c, &dent->key); + ctx->pos = key_hash_flash(c, &dent->key); file->private_data = dent; cond_resched(); } @@ -448,18 +454,11 @@ out: kfree(file->private_data); file->private_data = NULL; - file->f_pos = 2; + /* 2 is a special value indicating that there are no more direntries */ + ctx->pos = 2; return 0; } -/* If a directory is seeked, we have to free saved readdir() state */ -static loff_t ubifs_dir_llseek(struct file *file, loff_t offset, int whence) -{ - kfree(file->private_data); - file->private_data = NULL; - return generic_file_llseek(file, offset, whence); -} - /* Free saved readdir() state when the directory is closed */ static int ubifs_dir_release(struct inode *dir, struct file *file) { @@ -1177,10 +1176,10 @@ const struct inode_operations ubifs_dir_inode_operations = { }; const struct file_operations ubifs_dir_operations = { - .llseek = ubifs_dir_llseek, + .llseek = generic_file_llseek, .release = ubifs_dir_release, .read = generic_read_dir, - .readdir = ubifs_readdir, + .iterate = ubifs_readdir, .fsync = ubifs_fsync, .unlocked_ioctl = ubifs_ioctl, #ifdef CONFIG_COMPAT diff --git a/fs/udf/dir.c b/fs/udf/dir.c index b3e93f5e17c3..a012c51caffd 100644 --- a/fs/udf/dir.c +++ b/fs/udf/dir.c @@ -35,14 +35,16 @@ #include "udf_i.h" #include "udf_sb.h" -static int do_udf_readdir(struct inode *dir, struct file *filp, - filldir_t filldir, void *dirent) + +static int udf_readdir(struct file *file, struct dir_context *ctx) { + struct inode *dir = file_inode(file); + struct udf_inode_info *iinfo = UDF_I(dir); struct udf_fileident_bh fibh = { .sbh = NULL, .ebh = NULL}; struct fileIdentDesc *fi = NULL; struct fileIdentDesc cfi; int block, iblock; - loff_t nf_pos = (filp->f_pos - 1) << 2; + loff_t nf_pos; int flen; unsigned char *fname = NULL; unsigned char *nameptr; @@ -54,10 +56,14 @@ static int do_udf_readdir(struct inode *dir, struct file *filp, uint32_t elen; sector_t offset; int i, num, ret = 0; - unsigned int dt_type; struct extent_position epos = { NULL, 0, {0, 0} }; - struct udf_inode_info *iinfo; + if (ctx->pos == 0) { + if (!dir_emit_dot(file, ctx)) + return 0; + ctx->pos = 1; + } + nf_pos = (ctx->pos - 1) << 2; if (nf_pos >= size) goto out; @@ -71,7 +77,6 @@ static int do_udf_readdir(struct inode *dir, struct file *filp, nf_pos = udf_ext0_offset(dir); fibh.soffset = fibh.eoffset = nf_pos & (dir->i_sb->s_blocksize - 1); - iinfo = UDF_I(dir); if (iinfo->i_alloc_type != ICBTAG_FLAG_AD_IN_ICB) { if (inode_bmap(dir, nf_pos >> dir->i_sb->s_blocksize_bits, &epos, &eloc, &elen, &offset) @@ -116,7 +121,9 @@ static int do_udf_readdir(struct inode *dir, struct file *filp, } while (nf_pos < size) { - filp->f_pos = (nf_pos >> 2) + 1; + struct kernel_lb_addr tloc; + + ctx->pos = (nf_pos >> 2) + 1; fi = udf_fileident_read(dir, &nf_pos, &fibh, &cfi, &epos, &eloc, &elen, &offset); @@ -155,24 +162,22 @@ static int do_udf_readdir(struct inode *dir, struct file *filp, } if (cfi.fileCharacteristics & FID_FILE_CHAR_PARENT) { - iblock = parent_ino(filp->f_path.dentry); - flen = 2; - memcpy(fname, "..", flen); - dt_type = DT_DIR; - } else { - struct kernel_lb_addr tloc = lelb_to_cpu(cfi.icb.extLocation); - - iblock = udf_get_lb_pblock(dir->i_sb, &tloc, 0); - flen = udf_get_filename(dir->i_sb, nameptr, fname, lfi); - dt_type = DT_UNKNOWN; + if (!dir_emit_dotdot(file, ctx)) + goto out; + continue; } - if (flen && filldir(dirent, fname, flen, filp->f_pos, - iblock, dt_type) < 0) + flen = udf_get_filename(dir->i_sb, nameptr, fname, lfi); + if (!flen) + continue; + + tloc = lelb_to_cpu(cfi.icb.extLocation); + iblock = udf_get_lb_pblock(dir->i_sb, &tloc, 0); + if (!dir_emit(ctx, fname, flen, iblock, DT_UNKNOWN)) goto out; } /* end while */ - filp->f_pos = (nf_pos >> 2) + 1; + ctx->pos = (nf_pos >> 2) + 1; out: if (fibh.sbh != fibh.ebh) @@ -184,27 +189,11 @@ out: return ret; } -static int udf_readdir(struct file *filp, void *dirent, filldir_t filldir) -{ - struct inode *dir = file_inode(filp); - int result; - - if (filp->f_pos == 0) { - if (filldir(dirent, ".", 1, filp->f_pos, dir->i_ino, DT_DIR) < 0) { - return 0; - } - filp->f_pos++; - } - - result = do_udf_readdir(dir, filp, filldir, dirent); - return result; -} - /* readdir and lookup functions */ const struct file_operations udf_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = udf_readdir, + .iterate = udf_readdir, .unlocked_ioctl = udf_ioctl, .fsync = generic_file_fsync, }; diff --git a/fs/ufs/dir.c b/fs/ufs/dir.c index 3a75ca09c506..0ecc2cebed8f 100644 --- a/fs/ufs/dir.c +++ b/fs/ufs/dir.c @@ -430,16 +430,16 @@ ufs_validate_entry(struct super_block *sb, char *base, * This is blatantly stolen from ext2fs */ static int -ufs_readdir(struct file *filp, void *dirent, filldir_t filldir) +ufs_readdir(struct file *file, struct dir_context *ctx) { - loff_t pos = filp->f_pos; - struct inode *inode = file_inode(filp); + loff_t pos = ctx->pos; + struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; unsigned int offset = pos & ~PAGE_CACHE_MASK; unsigned long n = pos >> PAGE_CACHE_SHIFT; unsigned long npages = ufs_dir_pages(inode); unsigned chunk_mask = ~(UFS_SB(sb)->s_uspi->s_dirblksize - 1); - int need_revalidate = filp->f_version != inode->i_version; + int need_revalidate = file->f_version != inode->i_version; unsigned flags = UFS_SB(sb)->s_flags; UFSD("BEGIN\n"); @@ -457,16 +457,16 @@ ufs_readdir(struct file *filp, void *dirent, filldir_t filldir) ufs_error(sb, __func__, "bad page in #%lu", inode->i_ino); - filp->f_pos += PAGE_CACHE_SIZE - offset; + ctx->pos += PAGE_CACHE_SIZE - offset; return -EIO; } kaddr = page_address(page); if (unlikely(need_revalidate)) { if (offset) { offset = ufs_validate_entry(sb, kaddr, offset, chunk_mask); - filp->f_pos = (n<<PAGE_CACHE_SHIFT) + offset; + ctx->pos = (n<<PAGE_CACHE_SHIFT) + offset; } - filp->f_version = inode->i_version; + file->f_version = inode->i_version; need_revalidate = 0; } de = (struct ufs_dir_entry *)(kaddr+offset); @@ -479,11 +479,8 @@ ufs_readdir(struct file *filp, void *dirent, filldir_t filldir) return -EIO; } if (de->d_ino) { - int over; unsigned char d_type = DT_UNKNOWN; - offset = (char *)de - kaddr; - UFSD("filldir(%s,%u)\n", de->d_name, fs32_to_cpu(sb, de->d_ino)); UFSD("namlen %u\n", ufs_get_de_namlen(sb, de)); @@ -491,16 +488,15 @@ ufs_readdir(struct file *filp, void *dirent, filldir_t filldir) if ((flags & UFS_DE_MASK) == UFS_DE_44BSD) d_type = de->d_u.d_44.d_type; - over = filldir(dirent, de->d_name, + if (!dir_emit(ctx, de->d_name, ufs_get_de_namlen(sb, de), - (n<<PAGE_CACHE_SHIFT) | offset, - fs32_to_cpu(sb, de->d_ino), d_type); - if (over) { + fs32_to_cpu(sb, de->d_ino), + d_type)) { ufs_put_page(page); return 0; } } - filp->f_pos += fs16_to_cpu(sb, de->d_reclen); + ctx->pos += fs16_to_cpu(sb, de->d_reclen); } ufs_put_page(page); } @@ -660,7 +656,7 @@ not_empty: const struct file_operations ufs_dir_operations = { .read = generic_read_dir, - .readdir = ufs_readdir, + .iterate = ufs_readdir, .fsync = generic_file_fsync, .llseek = generic_file_llseek, }; diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c index 1d32f1d52763..306d883d89bc 100644 --- a/fs/xfs/xfs_acl.c +++ b/fs/xfs/xfs_acl.c @@ -21,6 +21,8 @@ #include "xfs_bmap_btree.h" #include "xfs_inode.h" #include "xfs_vnodeops.h" +#include "xfs_sb.h" +#include "xfs_mount.h" #include "xfs_trace.h" #include <linux/slab.h> #include <linux/xattr.h> @@ -34,7 +36,9 @@ */ STATIC struct posix_acl * -xfs_acl_from_disk(struct xfs_acl *aclp) +xfs_acl_from_disk( + struct xfs_acl *aclp, + int max_entries) { struct posix_acl_entry *acl_e; struct posix_acl *acl; @@ -42,7 +46,7 @@ xfs_acl_from_disk(struct xfs_acl *aclp) unsigned int count, i; count = be32_to_cpu(aclp->acl_cnt); - if (count > XFS_ACL_MAX_ENTRIES) + if (count > max_entries) return ERR_PTR(-EFSCORRUPTED); acl = posix_acl_alloc(count, GFP_KERNEL); @@ -108,9 +112,9 @@ xfs_get_acl(struct inode *inode, int type) struct xfs_inode *ip = XFS_I(inode); struct posix_acl *acl; struct xfs_acl *xfs_acl; - int len = sizeof(struct xfs_acl); unsigned char *ea_name; int error; + int len; acl = get_cached_acl(inode, type); if (acl != ACL_NOT_CACHED) @@ -133,8 +137,8 @@ xfs_get_acl(struct inode *inode, int type) * If we have a cached ACLs value just return it, not need to * go out to the disk. */ - - xfs_acl = kzalloc(sizeof(struct xfs_acl), GFP_KERNEL); + len = XFS_ACL_MAX_SIZE(ip->i_mount); + xfs_acl = kzalloc(len, GFP_KERNEL); if (!xfs_acl) return ERR_PTR(-ENOMEM); @@ -153,7 +157,7 @@ xfs_get_acl(struct inode *inode, int type) goto out; } - acl = xfs_acl_from_disk(xfs_acl); + acl = xfs_acl_from_disk(xfs_acl, XFS_ACL_MAX_ENTRIES(ip->i_mount)); if (IS_ERR(acl)) goto out; @@ -189,16 +193,17 @@ xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl) if (acl) { struct xfs_acl *xfs_acl; - int len; + int len = XFS_ACL_MAX_SIZE(ip->i_mount); - xfs_acl = kzalloc(sizeof(struct xfs_acl), GFP_KERNEL); + xfs_acl = kzalloc(len, GFP_KERNEL); if (!xfs_acl) return -ENOMEM; xfs_acl_to_disk(xfs_acl, acl); - len = sizeof(struct xfs_acl) - - (sizeof(struct xfs_acl_entry) * - (XFS_ACL_MAX_ENTRIES - acl->a_count)); + + /* subtract away the unused acl entries */ + len -= sizeof(struct xfs_acl_entry) * + (XFS_ACL_MAX_ENTRIES(ip->i_mount) - acl->a_count); error = -xfs_attr_set(ip, ea_name, (unsigned char *)xfs_acl, len, ATTR_ROOT); @@ -243,7 +248,7 @@ xfs_set_mode(struct inode *inode, umode_t mode) static int xfs_acl_exists(struct inode *inode, unsigned char *name) { - int len = sizeof(struct xfs_acl); + int len = XFS_ACL_MAX_SIZE(XFS_M(inode->i_sb)); return (xfs_attr_get(XFS_I(inode), name, NULL, &len, ATTR_ROOT|ATTR_KERNOVAL) == 0); @@ -379,7 +384,7 @@ xfs_xattr_acl_set(struct dentry *dentry, const char *name, goto out_release; error = -EINVAL; - if (acl->a_count > XFS_ACL_MAX_ENTRIES) + if (acl->a_count > XFS_ACL_MAX_ENTRIES(XFS_M(inode->i_sb))) goto out_release; if (type == ACL_TYPE_ACCESS) { diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h index 39632d941354..4016a567b83c 100644 --- a/fs/xfs/xfs_acl.h +++ b/fs/xfs/xfs_acl.h @@ -22,19 +22,36 @@ struct inode; struct posix_acl; struct xfs_inode; -#define XFS_ACL_MAX_ENTRIES 25 #define XFS_ACL_NOT_PRESENT (-1) /* On-disk XFS access control list structure */ +struct xfs_acl_entry { + __be32 ae_tag; + __be32 ae_id; + __be16 ae_perm; + __be16 ae_pad; /* fill the implicit hole in the structure */ +}; + struct xfs_acl { - __be32 acl_cnt; - struct xfs_acl_entry { - __be32 ae_tag; - __be32 ae_id; - __be16 ae_perm; - } acl_entry[XFS_ACL_MAX_ENTRIES]; + __be32 acl_cnt; + struct xfs_acl_entry acl_entry[0]; }; +/* + * The number of ACL entries allowed is defined by the on-disk format. + * For v4 superblocks, that is limited to 25 entries. For v5 superblocks, it is + * limited only by the maximum size of the xattr that stores the information. + */ +#define XFS_ACL_MAX_ENTRIES(mp) \ + (xfs_sb_version_hascrc(&mp->m_sb) \ + ? (XATTR_SIZE_MAX - sizeof(struct xfs_acl)) / \ + sizeof(struct xfs_acl_entry) \ + : 25) + +#define XFS_ACL_MAX_SIZE(mp) \ + (sizeof(struct xfs_acl) + \ + sizeof(struct xfs_acl_entry) * XFS_ACL_MAX_ENTRIES((mp))) + /* On-disk XFS extended attribute names */ #define SGI_ACL_FILE (unsigned char *)"SGI_ACL_FILE" #define SGI_ACL_DEFAULT (unsigned char *)"SGI_ACL_DEFAULT" diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 4ef1340d98f2..596ec71da00e 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -725,6 +725,25 @@ xfs_convert_page( (xfs_off_t)(page->index + 1) << PAGE_CACHE_SHIFT, i_size_read(inode)); + /* + * If the current map does not span the entire page we are about to try + * to write, then give up. The only way we can write a page that spans + * multiple mappings in a single writeback iteration is via the + * xfs_vm_writepage() function. Data integrity writeback requires the + * entire page to be written in a single attempt, otherwise the part of + * the page we don't write here doesn't get written as part of the data + * integrity sync. + * + * For normal writeback, we also don't attempt to write partial pages + * here as it simply means that write_cache_pages() will see it under + * writeback and ignore the page until some point in the future, at + * which time this will be the only page in the file that needs + * writeback. Hence for more optimal IO patterns, we should always + * avoid partial page writeback due to multiple mappings on a page here. + */ + if (!xfs_imap_valid(inode, imap, end_offset)) + goto fail_unlock_page; + len = 1 << inode->i_blkbits; p_offset = min_t(unsigned long, end_offset & (PAGE_CACHE_SIZE - 1), PAGE_CACHE_SIZE); diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index 08d5457c948e..31d3cd129269 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -931,20 +931,22 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context) */ int xfs_attr_shortform_allfit( - struct xfs_buf *bp, - struct xfs_inode *dp) + struct xfs_buf *bp, + struct xfs_inode *dp) { - xfs_attr_leafblock_t *leaf; - xfs_attr_leaf_entry_t *entry; + struct xfs_attr_leafblock *leaf; + struct xfs_attr_leaf_entry *entry; xfs_attr_leaf_name_local_t *name_loc; - int bytes, i; + struct xfs_attr3_icleaf_hdr leafhdr; + int bytes; + int i; leaf = bp->b_addr; - ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC)); + xfs_attr3_leaf_hdr_from_disk(&leafhdr, leaf); + entry = xfs_attr3_leaf_entryp(leaf); - entry = &leaf->entries[0]; bytes = sizeof(struct xfs_attr_sf_hdr); - for (i = 0; i < be16_to_cpu(leaf->hdr.count); entry++, i++) { + for (i = 0; i < leafhdr.count; entry++, i++) { if (entry->flags & XFS_ATTR_INCOMPLETE) continue; /* don't copy partial entries */ if (!(entry->flags & XFS_ATTR_LOCAL)) @@ -954,15 +956,15 @@ xfs_attr_shortform_allfit( return(0); if (be16_to_cpu(name_loc->valuelen) >= XFS_ATTR_SF_ENTSIZE_MAX) return(0); - bytes += sizeof(struct xfs_attr_sf_entry)-1 + bytes += sizeof(struct xfs_attr_sf_entry) - 1 + name_loc->namelen + be16_to_cpu(name_loc->valuelen); } if ((dp->i_mount->m_flags & XFS_MOUNT_ATTR2) && (dp->i_d.di_format != XFS_DINODE_FMT_BTREE) && (bytes == sizeof(struct xfs_attr_sf_hdr))) - return(-1); - return(xfs_attr_shortform_bytesfit(dp, bytes)); + return -1; + return xfs_attr_shortform_bytesfit(dp, bytes); } /* @@ -1410,7 +1412,7 @@ xfs_attr3_leaf_add_work( name_rmt->valuelen = 0; name_rmt->valueblk = 0; args->rmtblkno = 1; - args->rmtblkcnt = XFS_B_TO_FSB(mp, args->valuelen); + args->rmtblkcnt = xfs_attr3_rmt_blocks(mp, args->valuelen); } xfs_trans_log_buf(args->trans, bp, XFS_DA_LOGRANGE(leaf, xfs_attr3_leaf_name(leaf, args->index), @@ -1443,11 +1445,12 @@ xfs_attr3_leaf_add_work( STATIC void xfs_attr3_leaf_compact( struct xfs_da_args *args, - struct xfs_attr3_icleaf_hdr *ichdr_d, + struct xfs_attr3_icleaf_hdr *ichdr_dst, struct xfs_buf *bp) { - xfs_attr_leafblock_t *leaf_s, *leaf_d; - struct xfs_attr3_icleaf_hdr ichdr_s; + struct xfs_attr_leafblock *leaf_src; + struct xfs_attr_leafblock *leaf_dst; + struct xfs_attr3_icleaf_hdr ichdr_src; struct xfs_trans *trans = args->trans; struct xfs_mount *mp = trans->t_mountp; char *tmpbuffer; @@ -1455,29 +1458,38 @@ xfs_attr3_leaf_compact( trace_xfs_attr_leaf_compact(args); tmpbuffer = kmem_alloc(XFS_LBSIZE(mp), KM_SLEEP); - ASSERT(tmpbuffer != NULL); memcpy(tmpbuffer, bp->b_addr, XFS_LBSIZE(mp)); memset(bp->b_addr, 0, XFS_LBSIZE(mp)); + leaf_src = (xfs_attr_leafblock_t *)tmpbuffer; + leaf_dst = bp->b_addr; /* - * Copy basic information + * Copy the on-disk header back into the destination buffer to ensure + * all the information in the header that is not part of the incore + * header structure is preserved. */ - leaf_s = (xfs_attr_leafblock_t *)tmpbuffer; - leaf_d = bp->b_addr; - ichdr_s = *ichdr_d; /* struct copy */ - ichdr_d->firstused = XFS_LBSIZE(mp); - ichdr_d->usedbytes = 0; - ichdr_d->count = 0; - ichdr_d->holes = 0; - ichdr_d->freemap[0].base = xfs_attr3_leaf_hdr_size(leaf_s); - ichdr_d->freemap[0].size = ichdr_d->firstused - ichdr_d->freemap[0].base; + memcpy(bp->b_addr, tmpbuffer, xfs_attr3_leaf_hdr_size(leaf_src)); + + /* Initialise the incore headers */ + ichdr_src = *ichdr_dst; /* struct copy */ + ichdr_dst->firstused = XFS_LBSIZE(mp); + ichdr_dst->usedbytes = 0; + ichdr_dst->count = 0; + ichdr_dst->holes = 0; + ichdr_dst->freemap[0].base = xfs_attr3_leaf_hdr_size(leaf_src); + ichdr_dst->freemap[0].size = ichdr_dst->firstused - + ichdr_dst->freemap[0].base; + + + /* write the header back to initialise the underlying buffer */ + xfs_attr3_leaf_hdr_to_disk(leaf_dst, ichdr_dst); /* * Copy all entry's in the same (sorted) order, * but allocate name/value pairs packed and in sequence. */ - xfs_attr3_leaf_moveents(leaf_s, &ichdr_s, 0, leaf_d, ichdr_d, 0, - ichdr_s.count, mp); + xfs_attr3_leaf_moveents(leaf_src, &ichdr_src, 0, leaf_dst, ichdr_dst, 0, + ichdr_src.count, mp); /* * this logs the entire buffer, but the caller must write the header * back to the buffer when it is finished modifying it. @@ -2179,14 +2191,24 @@ xfs_attr3_leaf_unbalance( struct xfs_attr_leafblock *tmp_leaf; struct xfs_attr3_icleaf_hdr tmphdr; - tmp_leaf = kmem_alloc(state->blocksize, KM_SLEEP); - memset(tmp_leaf, 0, state->blocksize); - memset(&tmphdr, 0, sizeof(tmphdr)); + tmp_leaf = kmem_zalloc(state->blocksize, KM_SLEEP); + + /* + * Copy the header into the temp leaf so that all the stuff + * not in the incore header is present and gets copied back in + * once we've moved all the entries. + */ + memcpy(tmp_leaf, save_leaf, xfs_attr3_leaf_hdr_size(save_leaf)); + memset(&tmphdr, 0, sizeof(tmphdr)); tmphdr.magic = savehdr.magic; tmphdr.forw = savehdr.forw; tmphdr.back = savehdr.back; tmphdr.firstused = state->blocksize; + + /* write the header to the temp buffer to initialise it */ + xfs_attr3_leaf_hdr_to_disk(tmp_leaf, &tmphdr); + if (xfs_attr3_leaf_order(save_blk->bp, &savehdr, drop_blk->bp, &drophdr)) { xfs_attr3_leaf_moveents(drop_leaf, &drophdr, 0, @@ -2330,9 +2352,11 @@ xfs_attr3_leaf_lookup_int( if (!xfs_attr_namesp_match(args->flags, entry->flags)) continue; args->index = probe; + args->valuelen = be32_to_cpu(name_rmt->valuelen); args->rmtblkno = be32_to_cpu(name_rmt->valueblk); - args->rmtblkcnt = XFS_B_TO_FSB(args->dp->i_mount, - be32_to_cpu(name_rmt->valuelen)); + args->rmtblkcnt = xfs_attr3_rmt_blocks( + args->dp->i_mount, + args->valuelen); return XFS_ERROR(EEXIST); } } @@ -2383,7 +2407,8 @@ xfs_attr3_leaf_getvalue( ASSERT(memcmp(args->name, name_rmt->name, args->namelen) == 0); valuelen = be32_to_cpu(name_rmt->valuelen); args->rmtblkno = be32_to_cpu(name_rmt->valueblk); - args->rmtblkcnt = XFS_B_TO_FSB(args->dp->i_mount, valuelen); + args->rmtblkcnt = xfs_attr3_rmt_blocks(args->dp->i_mount, + valuelen); if (args->flags & ATTR_KERNOVAL) { args->valuelen = valuelen; return 0; @@ -2709,7 +2734,8 @@ xfs_attr3_leaf_list_int( args.valuelen = valuelen; args.value = kmem_alloc(valuelen, KM_SLEEP | KM_NOFS); args.rmtblkno = be32_to_cpu(name_rmt->valueblk); - args.rmtblkcnt = XFS_B_TO_FSB(args.dp->i_mount, valuelen); + args.rmtblkcnt = xfs_attr3_rmt_blocks( + args.dp->i_mount, valuelen); retval = xfs_attr_rmtval_get(&args); if (retval) return retval; @@ -3232,7 +3258,7 @@ xfs_attr3_leaf_inactive( name_rmt = xfs_attr3_leaf_name_remote(leaf, i); if (name_rmt->valueblk) { lp->valueblk = be32_to_cpu(name_rmt->valueblk); - lp->valuelen = XFS_B_TO_FSB(dp->i_mount, + lp->valuelen = xfs_attr3_rmt_blocks(dp->i_mount, be32_to_cpu(name_rmt->valuelen)); lp++; } diff --git a/fs/xfs/xfs_attr_leaf.h b/fs/xfs/xfs_attr_leaf.h index f9d7846097e2..444a7704596c 100644 --- a/fs/xfs/xfs_attr_leaf.h +++ b/fs/xfs/xfs_attr_leaf.h @@ -128,6 +128,7 @@ struct xfs_attr3_leaf_hdr { __u8 holes; __u8 pad1; struct xfs_attr_leaf_map freemap[XFS_ATTR_LEAF_MAPSIZE]; + __be32 pad2; /* 64 bit alignment */ }; #define XFS_ATTR3_LEAF_CRC_OFF (offsetof(struct xfs_attr3_leaf_hdr, info.crc)) diff --git a/fs/xfs/xfs_attr_remote.c b/fs/xfs/xfs_attr_remote.c index dee84466dcc9..ef6b0c124528 100644 --- a/fs/xfs/xfs_attr_remote.c +++ b/fs/xfs/xfs_attr_remote.c @@ -47,22 +47,55 @@ * Each contiguous block has a header, so it is not just a simple attribute * length to FSB conversion. */ -static int +int xfs_attr3_rmt_blocks( struct xfs_mount *mp, int attrlen) { - int buflen = XFS_ATTR3_RMT_BUF_SPACE(mp, - mp->m_sb.sb_blocksize); - return (attrlen + buflen - 1) / buflen; + if (xfs_sb_version_hascrc(&mp->m_sb)) { + int buflen = XFS_ATTR3_RMT_BUF_SPACE(mp, mp->m_sb.sb_blocksize); + return (attrlen + buflen - 1) / buflen; + } + return XFS_B_TO_FSB(mp, attrlen); +} + +/* + * Checking of the remote attribute header is split into two parts. The verifier + * does CRC, location and bounds checking, the unpacking function checks the + * attribute parameters and owner. + */ +static bool +xfs_attr3_rmt_hdr_ok( + struct xfs_mount *mp, + void *ptr, + xfs_ino_t ino, + uint32_t offset, + uint32_t size, + xfs_daddr_t bno) +{ + struct xfs_attr3_rmt_hdr *rmt = ptr; + + if (bno != be64_to_cpu(rmt->rm_blkno)) + return false; + if (offset != be32_to_cpu(rmt->rm_offset)) + return false; + if (size != be32_to_cpu(rmt->rm_bytes)) + return false; + if (ino != be64_to_cpu(rmt->rm_owner)) + return false; + + /* ok */ + return true; } static bool xfs_attr3_rmt_verify( - struct xfs_buf *bp) + struct xfs_mount *mp, + void *ptr, + int fsbsize, + xfs_daddr_t bno) { - struct xfs_mount *mp = bp->b_target->bt_mount; - struct xfs_attr3_rmt_hdr *rmt = bp->b_addr; + struct xfs_attr3_rmt_hdr *rmt = ptr; if (!xfs_sb_version_hascrc(&mp->m_sb)) return false; @@ -70,7 +103,9 @@ xfs_attr3_rmt_verify( return false; if (!uuid_equal(&rmt->rm_uuid, &mp->m_sb.sb_uuid)) return false; - if (bp->b_bn != be64_to_cpu(rmt->rm_blkno)) + if (be64_to_cpu(rmt->rm_blkno) != bno) + return false; + if (be32_to_cpu(rmt->rm_bytes) > fsbsize - sizeof(*rmt)) return false; if (be32_to_cpu(rmt->rm_offset) + be32_to_cpu(rmt->rm_bytes) >= XATTR_SIZE_MAX) @@ -86,17 +121,40 @@ xfs_attr3_rmt_read_verify( struct xfs_buf *bp) { struct xfs_mount *mp = bp->b_target->bt_mount; + char *ptr; + int len; + bool corrupt = false; + xfs_daddr_t bno; /* no verification of non-crc buffers */ if (!xfs_sb_version_hascrc(&mp->m_sb)) return; - if (!xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length), - XFS_ATTR3_RMT_CRC_OFF) || - !xfs_attr3_rmt_verify(bp)) { + ptr = bp->b_addr; + bno = bp->b_bn; + len = BBTOB(bp->b_length); + ASSERT(len >= XFS_LBSIZE(mp)); + + while (len > 0) { + if (!xfs_verify_cksum(ptr, XFS_LBSIZE(mp), + XFS_ATTR3_RMT_CRC_OFF)) { + corrupt = true; + break; + } + if (!xfs_attr3_rmt_verify(mp, ptr, XFS_LBSIZE(mp), bno)) { + corrupt = true; + break; + } + len -= XFS_LBSIZE(mp); + ptr += XFS_LBSIZE(mp); + bno += mp->m_bsize; + } + + if (corrupt) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr); xfs_buf_ioerror(bp, EFSCORRUPTED); - } + } else + ASSERT(len == 0); } static void @@ -105,23 +163,39 @@ xfs_attr3_rmt_write_verify( { struct xfs_mount *mp = bp->b_target->bt_mount; struct xfs_buf_log_item *bip = bp->b_fspriv; + char *ptr; + int len; + xfs_daddr_t bno; /* no verification of non-crc buffers */ if (!xfs_sb_version_hascrc(&mp->m_sb)) return; - if (!xfs_attr3_rmt_verify(bp)) { - XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr); - xfs_buf_ioerror(bp, EFSCORRUPTED); - return; - } + ptr = bp->b_addr; + bno = bp->b_bn; + len = BBTOB(bp->b_length); + ASSERT(len >= XFS_LBSIZE(mp)); + + while (len > 0) { + if (!xfs_attr3_rmt_verify(mp, ptr, XFS_LBSIZE(mp), bno)) { + XFS_CORRUPTION_ERROR(__func__, + XFS_ERRLEVEL_LOW, mp, bp->b_addr); + xfs_buf_ioerror(bp, EFSCORRUPTED); + return; + } + if (bip) { + struct xfs_attr3_rmt_hdr *rmt; + + rmt = (struct xfs_attr3_rmt_hdr *)ptr; + rmt->rm_lsn = cpu_to_be64(bip->bli_item.li_lsn); + } + xfs_update_cksum(ptr, XFS_LBSIZE(mp), XFS_ATTR3_RMT_CRC_OFF); - if (bip) { - struct xfs_attr3_rmt_hdr *rmt = bp->b_addr; - rmt->rm_lsn = cpu_to_be64(bip->bli_item.li_lsn); + len -= XFS_LBSIZE(mp); + ptr += XFS_LBSIZE(mp); + bno += mp->m_bsize; } - xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), - XFS_ATTR3_RMT_CRC_OFF); + ASSERT(len == 0); } const struct xfs_buf_ops xfs_attr3_rmt_buf_ops = { @@ -129,15 +203,16 @@ const struct xfs_buf_ops xfs_attr3_rmt_buf_ops = { .verify_write = xfs_attr3_rmt_write_verify, }; -static int +STATIC int xfs_attr3_rmt_hdr_set( struct xfs_mount *mp, + void *ptr, xfs_ino_t ino, uint32_t offset, uint32_t size, - struct xfs_buf *bp) + xfs_daddr_t bno) { - struct xfs_attr3_rmt_hdr *rmt = bp->b_addr; + struct xfs_attr3_rmt_hdr *rmt = ptr; if (!xfs_sb_version_hascrc(&mp->m_sb)) return 0; @@ -147,36 +222,107 @@ xfs_attr3_rmt_hdr_set( rmt->rm_bytes = cpu_to_be32(size); uuid_copy(&rmt->rm_uuid, &mp->m_sb.sb_uuid); rmt->rm_owner = cpu_to_be64(ino); - rmt->rm_blkno = cpu_to_be64(bp->b_bn); - bp->b_ops = &xfs_attr3_rmt_buf_ops; + rmt->rm_blkno = cpu_to_be64(bno); return sizeof(struct xfs_attr3_rmt_hdr); } /* - * Checking of the remote attribute header is split into two parts. the verifier - * does CRC, location and bounds checking, the unpacking function checks the - * attribute parameters and owner. + * Helper functions to copy attribute data in and out of the one disk extents */ -static bool -xfs_attr3_rmt_hdr_ok( - struct xfs_mount *mp, - xfs_ino_t ino, - uint32_t offset, - uint32_t size, - struct xfs_buf *bp) +STATIC int +xfs_attr_rmtval_copyout( + struct xfs_mount *mp, + struct xfs_buf *bp, + xfs_ino_t ino, + int *offset, + int *valuelen, + char **dst) { - struct xfs_attr3_rmt_hdr *rmt = bp->b_addr; + char *src = bp->b_addr; + xfs_daddr_t bno = bp->b_bn; + int len = BBTOB(bp->b_length); - if (offset != be32_to_cpu(rmt->rm_offset)) - return false; - if (size != be32_to_cpu(rmt->rm_bytes)) - return false; - if (ino != be64_to_cpu(rmt->rm_owner)) - return false; + ASSERT(len >= XFS_LBSIZE(mp)); - /* ok */ - return true; + while (len > 0 && *valuelen > 0) { + int hdr_size = 0; + int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, XFS_LBSIZE(mp)); + + byte_cnt = min_t(int, *valuelen, byte_cnt); + + if (xfs_sb_version_hascrc(&mp->m_sb)) { + if (!xfs_attr3_rmt_hdr_ok(mp, src, ino, *offset, + byte_cnt, bno)) { + xfs_alert(mp, +"remote attribute header mismatch bno/off/len/owner (0x%llx/0x%x/Ox%x/0x%llx)", + bno, *offset, byte_cnt, ino); + return EFSCORRUPTED; + } + hdr_size = sizeof(struct xfs_attr3_rmt_hdr); + } + + memcpy(*dst, src + hdr_size, byte_cnt); + + /* roll buffer forwards */ + len -= XFS_LBSIZE(mp); + src += XFS_LBSIZE(mp); + bno += mp->m_bsize; + + /* roll attribute data forwards */ + *valuelen -= byte_cnt; + *dst += byte_cnt; + *offset += byte_cnt; + } + return 0; +} + +STATIC void +xfs_attr_rmtval_copyin( + struct xfs_mount *mp, + struct xfs_buf *bp, + xfs_ino_t ino, + int *offset, + int *valuelen, + char **src) +{ + char *dst = bp->b_addr; + xfs_daddr_t bno = bp->b_bn; + int len = BBTOB(bp->b_length); + + ASSERT(len >= XFS_LBSIZE(mp)); + + while (len > 0 && *valuelen > 0) { + int hdr_size; + int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, XFS_LBSIZE(mp)); + + byte_cnt = min(*valuelen, byte_cnt); + hdr_size = xfs_attr3_rmt_hdr_set(mp, dst, ino, *offset, + byte_cnt, bno); + + memcpy(dst + hdr_size, *src, byte_cnt); + + /* + * If this is the last block, zero the remainder of it. + * Check that we are actually the last block, too. + */ + if (byte_cnt + hdr_size < XFS_LBSIZE(mp)) { + ASSERT(*valuelen - byte_cnt == 0); + ASSERT(len == XFS_LBSIZE(mp)); + memset(dst + hdr_size + byte_cnt, 0, + XFS_LBSIZE(mp) - hdr_size - byte_cnt); + } + + /* roll buffer forwards */ + len -= XFS_LBSIZE(mp); + dst += XFS_LBSIZE(mp); + bno += mp->m_bsize; + + /* roll attribute data forwards */ + *valuelen -= byte_cnt; + *src += byte_cnt; + *offset += byte_cnt; + } } /* @@ -190,13 +336,12 @@ xfs_attr_rmtval_get( struct xfs_bmbt_irec map[ATTR_RMTVALUE_MAPSIZE]; struct xfs_mount *mp = args->dp->i_mount; struct xfs_buf *bp; - xfs_daddr_t dblkno; xfs_dablk_t lblkno = args->rmtblkno; - void *dst = args->value; + char *dst = args->value; int valuelen = args->valuelen; int nmap; int error; - int blkcnt; + int blkcnt = args->rmtblkcnt; int i; int offset = 0; @@ -207,52 +352,36 @@ xfs_attr_rmtval_get( while (valuelen > 0) { nmap = ATTR_RMTVALUE_MAPSIZE; error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno, - args->rmtblkcnt, map, &nmap, + blkcnt, map, &nmap, XFS_BMAPI_ATTRFORK); if (error) return error; ASSERT(nmap >= 1); for (i = 0; (i < nmap) && (valuelen > 0); i++) { - int byte_cnt; - char *src; + xfs_daddr_t dblkno; + int dblkcnt; ASSERT((map[i].br_startblock != DELAYSTARTBLOCK) && (map[i].br_startblock != HOLESTARTBLOCK)); dblkno = XFS_FSB_TO_DADDR(mp, map[i].br_startblock); - blkcnt = XFS_FSB_TO_BB(mp, map[i].br_blockcount); + dblkcnt = XFS_FSB_TO_BB(mp, map[i].br_blockcount); error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, - dblkno, blkcnt, 0, &bp, + dblkno, dblkcnt, 0, &bp, &xfs_attr3_rmt_buf_ops); if (error) return error; - byte_cnt = min_t(int, valuelen, BBTOB(bp->b_length)); - byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, byte_cnt); - - src = bp->b_addr; - if (xfs_sb_version_hascrc(&mp->m_sb)) { - if (!xfs_attr3_rmt_hdr_ok(mp, args->dp->i_ino, - offset, byte_cnt, bp)) { - xfs_alert(mp, -"remote attribute header does not match required off/len/owner (0x%x/Ox%x,0x%llx)", - offset, byte_cnt, args->dp->i_ino); - xfs_buf_relse(bp); - return EFSCORRUPTED; - - } - - src += sizeof(struct xfs_attr3_rmt_hdr); - } - - memcpy(dst, src, byte_cnt); + error = xfs_attr_rmtval_copyout(mp, bp, args->dp->i_ino, + &offset, &valuelen, + &dst); xfs_buf_relse(bp); + if (error) + return error; - offset += byte_cnt; - dst += byte_cnt; - valuelen -= byte_cnt; - + /* roll attribute extent map forwards */ lblkno += map[i].br_blockcount; + blkcnt -= map[i].br_blockcount; } } ASSERT(valuelen == 0); @@ -270,17 +399,13 @@ xfs_attr_rmtval_set( struct xfs_inode *dp = args->dp; struct xfs_mount *mp = dp->i_mount; struct xfs_bmbt_irec map; - struct xfs_buf *bp; - xfs_daddr_t dblkno; xfs_dablk_t lblkno; xfs_fileoff_t lfileoff = 0; - void *src = args->value; + char *src = args->value; int blkcnt; int valuelen; int nmap; int error; - int hdrcnt = 0; - bool crcs = xfs_sb_version_hascrc(&mp->m_sb); int offset = 0; trace_xfs_attr_rmtval_set(args); @@ -289,24 +414,14 @@ xfs_attr_rmtval_set( * Find a "hole" in the attribute address space large enough for * us to drop the new attribute's value into. Because CRC enable * attributes have headers, we can't just do a straight byte to FSB - * conversion. We calculate the worst case block count in this case - * and we may not need that many, so we have to handle this when - * allocating the blocks below. + * conversion and have to take the header space into account. */ - if (!crcs) - blkcnt = XFS_B_TO_FSB(mp, args->valuelen); - else - blkcnt = xfs_attr3_rmt_blocks(mp, args->valuelen); - + blkcnt = xfs_attr3_rmt_blocks(mp, args->valuelen); error = xfs_bmap_first_unused(args->trans, args->dp, blkcnt, &lfileoff, XFS_ATTR_FORK); if (error) return error; - /* Start with the attribute data. We'll allocate the rest afterwards. */ - if (crcs) - blkcnt = XFS_B_TO_FSB(mp, args->valuelen); - args->rmtblkno = lblkno = (xfs_dablk_t)lfileoff; args->rmtblkcnt = blkcnt; @@ -349,26 +464,6 @@ xfs_attr_rmtval_set( (map.br_startblock != HOLESTARTBLOCK)); lblkno += map.br_blockcount; blkcnt -= map.br_blockcount; - hdrcnt++; - - /* - * If we have enough blocks for the attribute data, calculate - * how many extra blocks we need for headers. We might run - * through this multiple times in the case that the additional - * headers in the blocks needed for the data fragments spills - * into requiring more blocks. e.g. for 512 byte blocks, we'll - * spill for another block every 9 headers we require in this - * loop. - */ - if (crcs && blkcnt == 0) { - int total_len; - - total_len = args->valuelen + - hdrcnt * sizeof(struct xfs_attr3_rmt_hdr); - blkcnt = XFS_B_TO_FSB(mp, total_len); - blkcnt -= args->rmtblkcnt; - args->rmtblkcnt += blkcnt; - } /* * Start the next trans in the chain. @@ -385,18 +480,19 @@ xfs_attr_rmtval_set( * the INCOMPLETE flag. */ lblkno = args->rmtblkno; + blkcnt = args->rmtblkcnt; valuelen = args->valuelen; while (valuelen > 0) { - int byte_cnt; - char *buf; + struct xfs_buf *bp; + xfs_daddr_t dblkno; + int dblkcnt; + + ASSERT(blkcnt > 0); - /* - * Try to remember where we decided to put the value. - */ xfs_bmap_init(args->flist, args->firstblock); nmap = 1; error = xfs_bmapi_read(dp, (xfs_fileoff_t)lblkno, - args->rmtblkcnt, &map, &nmap, + blkcnt, &map, &nmap, XFS_BMAPI_ATTRFORK); if (error) return(error); @@ -405,41 +501,27 @@ xfs_attr_rmtval_set( (map.br_startblock != HOLESTARTBLOCK)); dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock), - blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount); + dblkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount); - bp = xfs_buf_get(mp->m_ddev_targp, dblkno, blkcnt, 0); + bp = xfs_buf_get(mp->m_ddev_targp, dblkno, dblkcnt, 0); if (!bp) return ENOMEM; bp->b_ops = &xfs_attr3_rmt_buf_ops; - byte_cnt = BBTOB(bp->b_length); - byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, byte_cnt); - if (valuelen < byte_cnt) - byte_cnt = valuelen; - - buf = bp->b_addr; - buf += xfs_attr3_rmt_hdr_set(mp, dp->i_ino, offset, - byte_cnt, bp); - memcpy(buf, src, byte_cnt); - - if (byte_cnt < BBTOB(bp->b_length)) - xfs_buf_zero(bp, byte_cnt, - BBTOB(bp->b_length) - byte_cnt); + xfs_attr_rmtval_copyin(mp, bp, args->dp->i_ino, &offset, + &valuelen, &src); error = xfs_bwrite(bp); /* GROT: NOTE: synchronous write */ xfs_buf_relse(bp); if (error) return error; - src += byte_cnt; - valuelen -= byte_cnt; - offset += byte_cnt; - hdrcnt--; + /* roll attribute extent map forwards */ lblkno += map.br_blockcount; + blkcnt -= map.br_blockcount; } ASSERT(valuelen == 0); - ASSERT(hdrcnt == 0); return 0; } @@ -448,33 +530,40 @@ xfs_attr_rmtval_set( * out-of-line buffer that it is stored on. */ int -xfs_attr_rmtval_remove(xfs_da_args_t *args) +xfs_attr_rmtval_remove( + struct xfs_da_args *args) { - xfs_mount_t *mp; - xfs_bmbt_irec_t map; - xfs_buf_t *bp; - xfs_daddr_t dblkno; - xfs_dablk_t lblkno; - int valuelen, blkcnt, nmap, error, done, committed; + struct xfs_mount *mp = args->dp->i_mount; + xfs_dablk_t lblkno; + int blkcnt; + int error; + int done; trace_xfs_attr_rmtval_remove(args); - mp = args->dp->i_mount; - /* - * Roll through the "value", invalidating the attribute value's - * blocks. + * Roll through the "value", invalidating the attribute value's blocks. + * Note that args->rmtblkcnt is the minimum number of data blocks we'll + * see for a CRC enabled remote attribute. Each extent will have a + * header, and so we may have more blocks than we realise here. If we + * fail to map the blocks correctly, we'll have problems with the buffer + * lookups. */ lblkno = args->rmtblkno; - valuelen = args->rmtblkcnt; - while (valuelen > 0) { + blkcnt = args->rmtblkcnt; + while (blkcnt > 0) { + struct xfs_bmbt_irec map; + struct xfs_buf *bp; + xfs_daddr_t dblkno; + int dblkcnt; + int nmap; + /* * Try to remember where we decided to put the value. */ nmap = 1; error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno, - args->rmtblkcnt, &map, &nmap, - XFS_BMAPI_ATTRFORK); + blkcnt, &map, &nmap, XFS_BMAPI_ATTRFORK); if (error) return(error); ASSERT(nmap == 1); @@ -482,21 +571,20 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args) (map.br_startblock != HOLESTARTBLOCK)); dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock), - blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount); + dblkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount); /* * If the "remote" value is in the cache, remove it. */ - bp = xfs_incore(mp->m_ddev_targp, dblkno, blkcnt, XBF_TRYLOCK); + bp = xfs_incore(mp->m_ddev_targp, dblkno, dblkcnt, XBF_TRYLOCK); if (bp) { xfs_buf_stale(bp); xfs_buf_relse(bp); bp = NULL; } - valuelen -= map.br_blockcount; - lblkno += map.br_blockcount; + blkcnt -= map.br_blockcount; } /* @@ -506,6 +594,8 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args) blkcnt = args->rmtblkcnt; done = 0; while (!done) { + int committed; + xfs_bmap_init(args->flist, args->firstblock); error = xfs_bunmapi(args->trans, args->dp, lblkno, blkcnt, XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA, diff --git a/fs/xfs/xfs_attr_remote.h b/fs/xfs/xfs_attr_remote.h index c7cca60a062a..92a8fd7977cc 100644 --- a/fs/xfs/xfs_attr_remote.h +++ b/fs/xfs/xfs_attr_remote.h @@ -20,6 +20,14 @@ #define XFS_ATTR3_RMT_MAGIC 0x5841524d /* XARM */ +/* + * There is one of these headers per filesystem block in a remote attribute. + * This is done to ensure there is a 1:1 mapping between the attribute value + * length and the number of blocks needed to store the attribute. This makes the + * verification of a buffer a little more complex, but greatly simplifies the + * allocation, reading and writing of these attributes as we don't have to guess + * the number of blocks needed to store the attribute data. + */ struct xfs_attr3_rmt_hdr { __be32 rm_magic; __be32 rm_offset; @@ -39,6 +47,8 @@ struct xfs_attr3_rmt_hdr { extern const struct xfs_buf_ops xfs_attr3_rmt_buf_ops; +int xfs_attr3_rmt_blocks(struct xfs_mount *mp, int attrlen); + int xfs_attr_rmtval_get(struct xfs_da_args *args); int xfs_attr_rmtval_set(struct xfs_da_args *args); int xfs_attr_rmtval_remove(struct xfs_da_args *args); diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c index 8804b8a3c310..0903960410a2 100644 --- a/fs/xfs/xfs_btree.c +++ b/fs/xfs/xfs_btree.c @@ -2544,7 +2544,17 @@ xfs_btree_new_iroot( if (error) goto error0; + /* + * we can't just memcpy() the root in for CRC enabled btree blocks. + * In that case have to also ensure the blkno remains correct + */ memcpy(cblock, block, xfs_btree_block_len(cur)); + if (cur->bc_flags & XFS_BTREE_CRC_BLOCKS) { + if (cur->bc_flags & XFS_BTREE_LONG_PTRS) + cblock->bb_u.l.bb_blkno = cpu_to_be64(cbp->b_bn); + else + cblock->bb_u.s.bb_blkno = cpu_to_be64(cbp->b_bn); + } be16_add_cpu(&block->bb_level, 1); xfs_btree_set_numrecs(block, 1); diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index 82b70bda9f47..1b2472a46e46 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -513,6 +513,7 @@ _xfs_buf_find( xfs_alert(btp->bt_mount, "%s: Block out of range: block 0x%llx, EOFS 0x%llx ", __func__, blkno, eofs); + WARN_ON(1); return NULL; } @@ -1649,7 +1650,7 @@ xfs_alloc_buftarg( { xfs_buftarg_t *btp; - btp = kmem_zalloc(sizeof(*btp), KM_SLEEP); + btp = kmem_zalloc(sizeof(*btp), KM_SLEEP | KM_NOFS); btp->bt_mount = mp; btp->bt_dev = bdev->bd_dev; diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index cf263476d6b4..4ec431777048 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -262,12 +262,7 @@ xfs_buf_item_format_segment( vecp->i_addr = xfs_buf_offset(bp, buffer_offset); vecp->i_len = nbits * XFS_BLF_CHUNK; vecp->i_type = XLOG_REG_TYPE_BCHUNK; -/* - * You would think we need to bump the nvecs here too, but we do not - * this number is used by recovery, and it gets confused by the boundary - * split here - * nvecs++; - */ + nvecs++; vecp++; first_bit = next_bit; last_bit = next_bit; diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c index 9b26a99ebfe9..0b8b2a13cd24 100644 --- a/fs/xfs/xfs_da_btree.c +++ b/fs/xfs/xfs_da_btree.c @@ -270,6 +270,7 @@ xfs_da3_node_read_verify( break; return; case XFS_ATTR_LEAF_MAGIC: + case XFS_ATTR3_LEAF_MAGIC: bp->b_ops = &xfs_attr3_leaf_buf_ops; bp->b_ops->verify_read(bp); return; @@ -2464,7 +2465,8 @@ xfs_buf_map_from_irec( ASSERT(nirecs >= 1); if (nirecs > 1) { - map = kmem_zalloc(nirecs * sizeof(struct xfs_buf_map), KM_SLEEP); + map = kmem_zalloc(nirecs * sizeof(struct xfs_buf_map), + KM_SLEEP | KM_NOFS); if (!map) return ENOMEM; *mapp = map; @@ -2520,7 +2522,8 @@ xfs_dabuf_map( * Optimize the one-block case. */ if (nfsb != 1) - irecs = kmem_zalloc(sizeof(irec) * nfsb, KM_SLEEP); + irecs = kmem_zalloc(sizeof(irec) * nfsb, + KM_SLEEP | KM_NOFS); nirecs = nfsb; error = xfs_bmapi_read(dp, (xfs_fileoff_t)bno, nfsb, irecs, diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c index f852b082a084..c407e1ccff43 100644 --- a/fs/xfs/xfs_dfrag.c +++ b/fs/xfs/xfs_dfrag.c @@ -219,6 +219,14 @@ xfs_swap_extents( int taforkblks = 0; __uint64_t tmp; + /* + * We have no way of updating owner information in the BMBT blocks for + * each inode on CRC enabled filesystems, so to avoid corrupting the + * this metadata we simply don't allow extent swaps to occur. + */ + if (xfs_sb_version_hascrc(&mp->m_sb)) + return XFS_ERROR(EINVAL); + tempifp = kmem_alloc(sizeof(xfs_ifork_t), KM_MAYFAIL); if (!tempifp) { error = XFS_ERROR(ENOMEM); diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c index b26a50f9921d..8f023dee404d 100644 --- a/fs/xfs/xfs_dir2.c +++ b/fs/xfs/xfs_dir2.c @@ -368,10 +368,8 @@ xfs_dir_removename( int xfs_readdir( xfs_inode_t *dp, - void *dirent, - size_t bufsize, - xfs_off_t *offset, - filldir_t filldir) + struct dir_context *ctx, + size_t bufsize) { int rval; /* return value */ int v; /* type-checking value */ @@ -385,14 +383,13 @@ xfs_readdir( XFS_STATS_INC(xs_dir_getdents); if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) - rval = xfs_dir2_sf_getdents(dp, dirent, offset, filldir); + rval = xfs_dir2_sf_getdents(dp, ctx); else if ((rval = xfs_dir2_isblock(NULL, dp, &v))) ; else if (v) - rval = xfs_dir2_block_getdents(dp, dirent, offset, filldir); + rval = xfs_dir2_block_getdents(dp, ctx); else - rval = xfs_dir2_leaf_getdents(dp, dirent, bufsize, offset, - filldir); + rval = xfs_dir2_leaf_getdents(dp, ctx, bufsize); return rval; } diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c index e59f5fc816fe..09aea0247d96 100644 --- a/fs/xfs/xfs_dir2_block.c +++ b/fs/xfs/xfs_dir2_block.c @@ -569,9 +569,7 @@ xfs_dir2_block_addname( int /* error */ xfs_dir2_block_getdents( xfs_inode_t *dp, /* incore inode */ - void *dirent, - xfs_off_t *offset, - filldir_t filldir) + struct dir_context *ctx) { xfs_dir2_data_hdr_t *hdr; /* block header */ struct xfs_buf *bp; /* buffer for block */ @@ -589,7 +587,7 @@ xfs_dir2_block_getdents( /* * If the block number in the offset is out of range, we're done. */ - if (xfs_dir2_dataptr_to_db(mp, *offset) > mp->m_dirdatablk) + if (xfs_dir2_dataptr_to_db(mp, ctx->pos) > mp->m_dirdatablk) return 0; error = xfs_dir3_block_read(NULL, dp, &bp); @@ -600,7 +598,7 @@ xfs_dir2_block_getdents( * Extract the byte offset we start at from the seek pointer. * We'll skip entries before this. */ - wantoff = xfs_dir2_dataptr_to_off(mp, *offset); + wantoff = xfs_dir2_dataptr_to_off(mp, ctx->pos); hdr = bp->b_addr; xfs_dir3_data_check(dp, bp); /* @@ -639,13 +637,12 @@ xfs_dir2_block_getdents( cook = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk, (char *)dep - (char *)hdr); + ctx->pos = cook & 0x7fffffff; /* * If it didn't fit, set the final offset to here & return. */ - if (filldir(dirent, (char *)dep->name, dep->namelen, - cook & 0x7fffffff, be64_to_cpu(dep->inumber), - DT_UNKNOWN)) { - *offset = cook & 0x7fffffff; + if (!dir_emit(ctx, (char *)dep->name, dep->namelen, + be64_to_cpu(dep->inumber), DT_UNKNOWN)) { xfs_trans_brelse(NULL, bp); return 0; } @@ -655,7 +652,7 @@ xfs_dir2_block_getdents( * Reached the end of the block. * Set the offset to a non-existent block 1 and return. */ - *offset = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk + 1, 0) & + ctx->pos = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk + 1, 0) & 0x7fffffff; xfs_trans_brelse(NULL, bp); return 0; diff --git a/fs/xfs/xfs_dir2_format.h b/fs/xfs/xfs_dir2_format.h index a3b1bd841a80..7826782b8d78 100644 --- a/fs/xfs/xfs_dir2_format.h +++ b/fs/xfs/xfs_dir2_format.h @@ -266,6 +266,7 @@ struct xfs_dir3_blk_hdr { struct xfs_dir3_data_hdr { struct xfs_dir3_blk_hdr hdr; xfs_dir2_data_free_t best_free[XFS_DIR2_DATA_FD_COUNT]; + __be32 pad; /* 64 bit alignment */ }; #define XFS_DIR3_DATA_CRC_OFF offsetof(struct xfs_dir3_data_hdr, hdr.crc) @@ -477,7 +478,7 @@ struct xfs_dir3_leaf_hdr { struct xfs_da3_blkinfo info; /* header for da routines */ __be16 count; /* count of entries */ __be16 stale; /* count of stale entries */ - __be32 pad; + __be32 pad; /* 64 bit alignment */ }; struct xfs_dir3_icleaf_hdr { @@ -715,6 +716,7 @@ struct xfs_dir3_free_hdr { __be32 firstdb; /* db of first entry */ __be32 nvalid; /* count of valid entries */ __be32 nused; /* count of used entries */ + __be32 pad; /* 64 bit alignment */ }; struct xfs_dir3_free { diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c index 721ba2fe8e54..e0cc1243a8aa 100644 --- a/fs/xfs/xfs_dir2_leaf.c +++ b/fs/xfs/xfs_dir2_leaf.c @@ -1300,10 +1300,8 @@ out: int /* error */ xfs_dir2_leaf_getdents( xfs_inode_t *dp, /* incore directory inode */ - void *dirent, - size_t bufsize, - xfs_off_t *offset, - filldir_t filldir) + struct dir_context *ctx, + size_t bufsize) { struct xfs_buf *bp = NULL; /* data block buffer */ xfs_dir2_data_hdr_t *hdr; /* data block header */ @@ -1322,7 +1320,7 @@ xfs_dir2_leaf_getdents( * If the offset is at or past the largest allowed value, * give up right away. */ - if (*offset >= XFS_DIR2_MAX_DATAPTR) + if (ctx->pos >= XFS_DIR2_MAX_DATAPTR) return 0; mp = dp->i_mount; @@ -1336,14 +1334,14 @@ xfs_dir2_leaf_getdents( mp->m_sb.sb_blocksize); map_info = kmem_zalloc(offsetof(struct xfs_dir2_leaf_map_info, map) + (length * sizeof(struct xfs_bmbt_irec)), - KM_SLEEP); + KM_SLEEP | KM_NOFS); map_info->map_size = length; /* * Inside the loop we keep the main offset value as a byte offset * in the directory file. */ - curoff = xfs_dir2_dataptr_to_byte(mp, *offset); + curoff = xfs_dir2_dataptr_to_byte(mp, ctx->pos); /* * Force this conversion through db so we truncate the offset @@ -1444,8 +1442,8 @@ xfs_dir2_leaf_getdents( dep = (xfs_dir2_data_entry_t *)ptr; length = xfs_dir2_data_entsize(dep->namelen); - if (filldir(dirent, (char *)dep->name, dep->namelen, - xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff, + ctx->pos = xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff; + if (!dir_emit(ctx, (char *)dep->name, dep->namelen, be64_to_cpu(dep->inumber), DT_UNKNOWN)) break; @@ -1462,9 +1460,9 @@ xfs_dir2_leaf_getdents( * All done. Set output offset value to current offset. */ if (curoff > xfs_dir2_dataptr_to_byte(mp, XFS_DIR2_MAX_DATAPTR)) - *offset = XFS_DIR2_MAX_DATAPTR & 0x7fffffff; + ctx->pos = XFS_DIR2_MAX_DATAPTR & 0x7fffffff; else - *offset = xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff; + ctx->pos = xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff; kmem_free(map_info); if (bp) xfs_trans_brelse(NULL, bp); diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c index 5246de4912d4..2226a00acd15 100644 --- a/fs/xfs/xfs_dir2_node.c +++ b/fs/xfs/xfs_dir2_node.c @@ -263,18 +263,19 @@ xfs_dir3_free_get_buf( * Initialize the new block to be empty, and remember * its first slot as our empty slot. */ - hdr.magic = XFS_DIR2_FREE_MAGIC; - hdr.firstdb = 0; - hdr.nused = 0; - hdr.nvalid = 0; + memset(bp->b_addr, 0, sizeof(struct xfs_dir3_free_hdr)); + memset(&hdr, 0, sizeof(hdr)); + if (xfs_sb_version_hascrc(&mp->m_sb)) { struct xfs_dir3_free_hdr *hdr3 = bp->b_addr; hdr.magic = XFS_DIR3_FREE_MAGIC; + hdr3->hdr.blkno = cpu_to_be64(bp->b_bn); hdr3->hdr.owner = cpu_to_be64(dp->i_ino); uuid_copy(&hdr3->hdr.uuid, &mp->m_sb.sb_uuid); - } + } else + hdr.magic = XFS_DIR2_FREE_MAGIC; xfs_dir3_free_hdr_to_disk(bp->b_addr, &hdr); *bpp = bp; return 0; @@ -1921,8 +1922,6 @@ xfs_dir2_node_addname_int( */ freehdr.firstdb = (fbno - XFS_DIR2_FREE_FIRSTDB(mp)) * xfs_dir3_free_max_bests(mp); - free->hdr.nvalid = 0; - free->hdr.nused = 0; } else { free = fbp->b_addr; bests = xfs_dir3_free_bests_p(mp, free); diff --git a/fs/xfs/xfs_dir2_priv.h b/fs/xfs/xfs_dir2_priv.h index 7cf573c88aad..0511cda4a712 100644 --- a/fs/xfs/xfs_dir2_priv.h +++ b/fs/xfs/xfs_dir2_priv.h @@ -33,8 +33,8 @@ extern int xfs_dir_cilookup_result(struct xfs_da_args *args, extern const struct xfs_buf_ops xfs_dir3_block_buf_ops; extern int xfs_dir2_block_addname(struct xfs_da_args *args); -extern int xfs_dir2_block_getdents(struct xfs_inode *dp, void *dirent, - xfs_off_t *offset, filldir_t filldir); +extern int xfs_dir2_block_getdents(struct xfs_inode *dp, + struct dir_context *ctx); extern int xfs_dir2_block_lookup(struct xfs_da_args *args); extern int xfs_dir2_block_removename(struct xfs_da_args *args); extern int xfs_dir2_block_replace(struct xfs_da_args *args); @@ -91,8 +91,8 @@ extern void xfs_dir3_leaf_compact(struct xfs_da_args *args, extern void xfs_dir3_leaf_compact_x1(struct xfs_dir3_icleaf_hdr *leafhdr, struct xfs_dir2_leaf_entry *ents, int *indexp, int *lowstalep, int *highstalep, int *lowlogp, int *highlogp); -extern int xfs_dir2_leaf_getdents(struct xfs_inode *dp, void *dirent, - size_t bufsize, xfs_off_t *offset, filldir_t filldir); +extern int xfs_dir2_leaf_getdents(struct xfs_inode *dp, struct dir_context *ctx, + size_t bufsize); extern int xfs_dir3_leaf_get_buf(struct xfs_da_args *args, xfs_dir2_db_t bno, struct xfs_buf **bpp, __uint16_t magic); extern void xfs_dir3_leaf_log_ents(struct xfs_trans *tp, struct xfs_buf *bp, @@ -153,8 +153,7 @@ extern int xfs_dir2_block_to_sf(struct xfs_da_args *args, struct xfs_buf *bp, int size, xfs_dir2_sf_hdr_t *sfhp); extern int xfs_dir2_sf_addname(struct xfs_da_args *args); extern int xfs_dir2_sf_create(struct xfs_da_args *args, xfs_ino_t pino); -extern int xfs_dir2_sf_getdents(struct xfs_inode *dp, void *dirent, - xfs_off_t *offset, filldir_t filldir); +extern int xfs_dir2_sf_getdents(struct xfs_inode *dp, struct dir_context *ctx); extern int xfs_dir2_sf_lookup(struct xfs_da_args *args); extern int xfs_dir2_sf_removename(struct xfs_da_args *args); extern int xfs_dir2_sf_replace(struct xfs_da_args *args); diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/xfs_dir2_sf.c index 6157424dbf8f..97676a347da1 100644 --- a/fs/xfs/xfs_dir2_sf.c +++ b/fs/xfs/xfs_dir2_sf.c @@ -768,9 +768,7 @@ xfs_dir2_sf_create( int /* error */ xfs_dir2_sf_getdents( xfs_inode_t *dp, /* incore directory inode */ - void *dirent, - xfs_off_t *offset, - filldir_t filldir) + struct dir_context *ctx) { int i; /* shortform entry number */ xfs_mount_t *mp; /* filesystem mount point */ @@ -802,7 +800,7 @@ xfs_dir2_sf_getdents( /* * If the block number in the offset is out of range, we're done. */ - if (xfs_dir2_dataptr_to_db(mp, *offset) > mp->m_dirdatablk) + if (xfs_dir2_dataptr_to_db(mp, ctx->pos) > mp->m_dirdatablk) return 0; /* @@ -819,22 +817,20 @@ xfs_dir2_sf_getdents( /* * Put . entry unless we're starting past it. */ - if (*offset <= dot_offset) { - if (filldir(dirent, ".", 1, dot_offset & 0x7fffffff, dp->i_ino, DT_DIR)) { - *offset = dot_offset & 0x7fffffff; + if (ctx->pos <= dot_offset) { + ctx->pos = dot_offset & 0x7fffffff; + if (!dir_emit(ctx, ".", 1, dp->i_ino, DT_DIR)) return 0; - } } /* * Put .. entry unless we're starting past it. */ - if (*offset <= dotdot_offset) { + if (ctx->pos <= dotdot_offset) { ino = xfs_dir2_sf_get_parent_ino(sfp); - if (filldir(dirent, "..", 2, dotdot_offset & 0x7fffffff, ino, DT_DIR)) { - *offset = dotdot_offset & 0x7fffffff; + ctx->pos = dotdot_offset & 0x7fffffff; + if (!dir_emit(ctx, "..", 2, ino, DT_DIR)) return 0; - } } /* @@ -845,21 +841,20 @@ xfs_dir2_sf_getdents( off = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk, xfs_dir2_sf_get_offset(sfep)); - if (*offset > off) { + if (ctx->pos > off) { sfep = xfs_dir2_sf_nextentry(sfp, sfep); continue; } ino = xfs_dir2_sfe_get_ino(sfp, sfep); - if (filldir(dirent, (char *)sfep->name, sfep->namelen, - off & 0x7fffffff, ino, DT_UNKNOWN)) { - *offset = off & 0x7fffffff; + ctx->pos = off & 0x7fffffff; + if (!dir_emit(ctx, (char *)sfep->name, sfep->namelen, + ino, DT_UNKNOWN)) return 0; - } sfep = xfs_dir2_sf_nextentry(sfp, sfep); } - *offset = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk + 1, 0) & + ctx->pos = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk + 1, 0) & 0x7fffffff; return 0; } diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index a41f8bf1da37..044e97a33c8d 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -249,8 +249,11 @@ xfs_qm_init_dquot_blk( d->dd_diskdq.d_version = XFS_DQUOT_VERSION; d->dd_diskdq.d_id = cpu_to_be32(curid); d->dd_diskdq.d_flags = type; - if (xfs_sb_version_hascrc(&mp->m_sb)) + if (xfs_sb_version_hascrc(&mp->m_sb)) { uuid_copy(&d->dd_uuid, &mp->m_sb.sb_uuid); + xfs_update_cksum((char *)d, sizeof(struct xfs_dqblk), + XFS_DQUOT_CRC_OFF); + } } xfs_trans_dquot_buf(tp, bp, @@ -286,23 +289,6 @@ xfs_dquot_set_prealloc_limits(struct xfs_dquot *dqp) dqp->q_low_space[XFS_QLOWSP_5_PCNT] = space * 5; } -STATIC void -xfs_dquot_buf_calc_crc( - struct xfs_mount *mp, - struct xfs_buf *bp) -{ - struct xfs_dqblk *d = (struct xfs_dqblk *)bp->b_addr; - int i; - - if (!xfs_sb_version_hascrc(&mp->m_sb)) - return; - - for (i = 0; i < mp->m_quotainfo->qi_dqperchunk; i++, d++) { - xfs_update_cksum((char *)d, sizeof(struct xfs_dqblk), - offsetof(struct xfs_dqblk, dd_crc)); - } -} - STATIC bool xfs_dquot_buf_verify_crc( struct xfs_mount *mp, @@ -328,12 +314,11 @@ xfs_dquot_buf_verify_crc( for (i = 0; i < ndquots; i++, d++) { if (!xfs_verify_cksum((char *)d, sizeof(struct xfs_dqblk), - offsetof(struct xfs_dqblk, dd_crc))) + XFS_DQUOT_CRC_OFF)) return false; if (!uuid_equal(&d->dd_uuid, &mp->m_sb.sb_uuid)) return false; } - return true; } @@ -393,6 +378,11 @@ xfs_dquot_buf_read_verify( } } +/* + * we don't calculate the CRC here as that is done when the dquot is flushed to + * the buffer after the update is done. This ensures that the dquot in the + * buffer always has an up-to-date CRC value. + */ void xfs_dquot_buf_write_verify( struct xfs_buf *bp) @@ -404,7 +394,6 @@ xfs_dquot_buf_write_verify( xfs_buf_ioerror(bp, EFSCORRUPTED); return; } - xfs_dquot_buf_calc_crc(mp, bp); } const struct xfs_buf_ops xfs_dquot_buf_ops = { @@ -1151,11 +1140,17 @@ xfs_qm_dqflush( * copy the lsn into the on-disk dquot now while we have the in memory * dquot here. This can't be done later in the write verifier as we * can't get access to the log item at that point in time. + * + * We also calculate the CRC here so that the on-disk dquot in the + * buffer always has a valid CRC. This ensures there is no possibility + * of a dquot without an up-to-date CRC getting to disk. */ if (xfs_sb_version_hascrc(&mp->m_sb)) { struct xfs_dqblk *dqb = (struct xfs_dqblk *)ddqp; dqb->dd_lsn = cpu_to_be64(dqp->q_logitem.qli_item.li_lsn); + xfs_update_cksum((char *)dqb, sizeof(struct xfs_dqblk), + XFS_DQUOT_CRC_OFF); } /* diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c index c0f375087efc..452920a3f03f 100644 --- a/fs/xfs/xfs_extfree_item.c +++ b/fs/xfs/xfs_extfree_item.c @@ -305,11 +305,12 @@ xfs_efi_release(xfs_efi_log_item_t *efip, { ASSERT(atomic_read(&efip->efi_next_extent) >= nextents); if (atomic_sub_and_test(nextents, &efip->efi_next_extent)) { - __xfs_efi_release(efip); - /* recovery needs us to drop the EFI reference, too */ if (test_bit(XFS_EFI_RECOVERED, &efip->efi_flags)) __xfs_efi_release(efip); + + __xfs_efi_release(efip); + /* efip may now have been freed, do not reference it again. */ } } diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index a5f2042aec8b..0ad2b95fca12 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -906,11 +906,10 @@ xfs_file_release( STATIC int xfs_file_readdir( - struct file *filp, - void *dirent, - filldir_t filldir) + struct file *file, + struct dir_context *ctx) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); xfs_inode_t *ip = XFS_I(inode); int error; size_t bufsize; @@ -929,8 +928,7 @@ xfs_file_readdir( */ bufsize = (size_t)min_t(loff_t, 32768, ip->i_d.di_size); - error = xfs_readdir(ip, dirent, bufsize, - (xfs_off_t *)&filp->f_pos, filldir); + error = xfs_readdir(ip, ctx, bufsize); if (error) return -error; return 0; @@ -1432,7 +1430,7 @@ const struct file_operations xfs_file_operations = { const struct file_operations xfs_dir_file_operations = { .open = xfs_dir_open, .read = generic_read_dir, - .readdir = xfs_file_readdir, + .iterate = xfs_file_readdir, .llseek = generic_file_llseek, .unlocked_ioctl = xfs_file_ioctl, #ifdef CONFIG_COMPAT diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h index 6dda3f949b04..d04695545397 100644 --- a/fs/xfs/xfs_fs.h +++ b/fs/xfs/xfs_fs.h @@ -236,6 +236,7 @@ typedef struct xfs_fsop_resblks { #define XFS_FSOP_GEOM_FLAGS_PROJID32 0x0800 /* 32-bit project IDs */ #define XFS_FSOP_GEOM_FLAGS_DIRV2CI 0x1000 /* ASCII only CI names */ #define XFS_FSOP_GEOM_FLAGS_LAZYSB 0x4000 /* lazy superblock counters */ +#define XFS_FSOP_GEOM_FLAGS_V5SB 0x8000 /* version 5 superblock */ /* diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index 87595b211da1..3c3644ea825b 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -99,7 +99,9 @@ xfs_fs_geometry( (xfs_sb_version_hasattr2(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_ATTR2 : 0) | (xfs_sb_version_hasprojid32bit(&mp->m_sb) ? - XFS_FSOP_GEOM_FLAGS_PROJID32 : 0); + XFS_FSOP_GEOM_FLAGS_PROJID32 : 0) | + (xfs_sb_version_hascrc(&mp->m_sb) ? + XFS_FSOP_GEOM_FLAGS_V5SB : 0); geo->logsectsize = xfs_sb_version_hassector(&mp->m_sb) ? mp->m_sb.sb_logsectsize : BBSIZE; geo->rtsectsize = mp->m_sb.sb_blocksize; diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index efbe1accb6ca..7f7be5f98f52 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1638,6 +1638,10 @@ xfs_iunlink( dip->di_next_unlinked = agi->agi_unlinked[bucket_index]; offset = ip->i_imap.im_boffset + offsetof(xfs_dinode_t, di_next_unlinked); + + /* need to recalc the inode CRC if appropriate */ + xfs_dinode_calc_crc(mp, dip); + xfs_trans_inode_buf(tp, ibp); xfs_trans_log_buf(tp, ibp, offset, (offset + sizeof(xfs_agino_t) - 1)); @@ -1723,6 +1727,10 @@ xfs_iunlink_remove( dip->di_next_unlinked = cpu_to_be32(NULLAGINO); offset = ip->i_imap.im_boffset + offsetof(xfs_dinode_t, di_next_unlinked); + + /* need to recalc the inode CRC if appropriate */ + xfs_dinode_calc_crc(mp, dip); + xfs_trans_inode_buf(tp, ibp); xfs_trans_log_buf(tp, ibp, offset, (offset + sizeof(xfs_agino_t) - 1)); @@ -1796,6 +1804,10 @@ xfs_iunlink_remove( dip->di_next_unlinked = cpu_to_be32(NULLAGINO); offset = ip->i_imap.im_boffset + offsetof(xfs_dinode_t, di_next_unlinked); + + /* need to recalc the inode CRC if appropriate */ + xfs_dinode_calc_crc(mp, dip); + xfs_trans_inode_buf(tp, ibp); xfs_trans_log_buf(tp, ibp, offset, (offset + sizeof(xfs_agino_t) - 1)); @@ -1809,6 +1821,10 @@ xfs_iunlink_remove( last_dip->di_next_unlinked = cpu_to_be32(next_agino); ASSERT(next_agino != 0); offset = last_offset + offsetof(xfs_dinode_t, di_next_unlinked); + + /* need to recalc the inode CRC if appropriate */ + xfs_dinode_calc_crc(mp, last_dip); + xfs_trans_inode_buf(tp, last_ibp); xfs_trans_log_buf(tp, last_ibp, offset, (offset + sizeof(xfs_agino_t) - 1)); diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index d82efaa2ac73..ca9ecaa81112 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -455,6 +455,28 @@ xfs_vn_getattr( return 0; } +static void +xfs_setattr_mode( + struct xfs_trans *tp, + struct xfs_inode *ip, + struct iattr *iattr) +{ + struct inode *inode = VFS_I(ip); + umode_t mode = iattr->ia_mode; + + ASSERT(tp); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + + if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID)) + mode &= ~S_ISGID; + + ip->i_d.di_mode &= S_IFMT; + ip->i_d.di_mode |= mode & ~S_IFMT; + + inode->i_mode &= S_IFMT; + inode->i_mode |= mode & ~S_IFMT; +} + int xfs_setattr_nonsize( struct xfs_inode *ip, @@ -606,18 +628,8 @@ xfs_setattr_nonsize( /* * Change file access modes. */ - if (mask & ATTR_MODE) { - umode_t mode = iattr->ia_mode; - - if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID)) - mode &= ~S_ISGID; - - ip->i_d.di_mode &= S_IFMT; - ip->i_d.di_mode |= mode & ~S_IFMT; - - inode->i_mode &= S_IFMT; - inode->i_mode |= mode & ~S_IFMT; - } + if (mask & ATTR_MODE) + xfs_setattr_mode(tp, ip, iattr); /* * Change file access or modified times. @@ -714,9 +726,8 @@ xfs_setattr_size( return XFS_ERROR(error); ASSERT(S_ISREG(ip->i_d.di_mode)); - ASSERT((mask & (ATTR_MODE|ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET| - ATTR_MTIME_SET|ATTR_KILL_SUID|ATTR_KILL_SGID| - ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0); + ASSERT((mask & (ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET| + ATTR_MTIME_SET|ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0); if (!(flags & XFS_ATTR_NOLOCK)) { lock_flags |= XFS_IOLOCK_EXCL; @@ -860,6 +871,12 @@ xfs_setattr_size( xfs_inode_clear_eofblocks_tag(ip); } + /* + * Change file access modes. + */ + if (mask & ATTR_MODE) + xfs_setattr_mode(tp, ip, iattr); + if (mask & ATTR_CTIME) { inode->i_ctime = iattr->ia_ctime; ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec; diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c index e3d0b85d852b..d0833b54e55d 100644 --- a/fs/xfs/xfs_log_cil.c +++ b/fs/xfs/xfs_log_cil.c @@ -139,7 +139,7 @@ xlog_cil_prepare_log_vecs( new_lv = kmem_zalloc(sizeof(*new_lv) + niovecs * sizeof(struct xfs_log_iovec), - KM_SLEEP); + KM_SLEEP|KM_NOFS); /* The allocated iovec region lies beyond the log vector. */ new_lv->lv_iovecp = (struct xfs_log_iovec *)&new_lv[1]; diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 93f03ec17eec..7cf5e4eafe28 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -1599,10 +1599,43 @@ xlog_recover_add_to_trans( } /* - * Sort the log items in the transaction. Cancelled buffers need - * to be put first so they are processed before any items that might - * modify the buffers. If they are cancelled, then the modifications - * don't need to be replayed. + * Sort the log items in the transaction. + * + * The ordering constraints are defined by the inode allocation and unlink + * behaviour. The rules are: + * + * 1. Every item is only logged once in a given transaction. Hence it + * represents the last logged state of the item. Hence ordering is + * dependent on the order in which operations need to be performed so + * required initial conditions are always met. + * + * 2. Cancelled buffers are recorded in pass 1 in a separate table and + * there's nothing to replay from them so we can simply cull them + * from the transaction. However, we can't do that until after we've + * replayed all the other items because they may be dependent on the + * cancelled buffer and replaying the cancelled buffer can remove it + * form the cancelled buffer table. Hence they have tobe done last. + * + * 3. Inode allocation buffers must be replayed before inode items that + * read the buffer and replay changes into it. + * + * 4. Inode unlink buffers must be replayed after inode items are replayed. + * This ensures that inodes are completely flushed to the inode buffer + * in a "free" state before we remove the unlinked inode list pointer. + * + * Hence the ordering needs to be inode allocation buffers first, inode items + * second, inode unlink buffers third and cancelled buffers last. + * + * But there's a problem with that - we can't tell an inode allocation buffer + * apart from a regular buffer, so we can't separate them. We can, however, + * tell an inode unlink buffer from the others, and so we can separate them out + * from all the other buffers and move them to last. + * + * Hence, 4 lists, in order from head to tail: + * - buffer_list for all buffers except cancelled/inode unlink buffers + * - item_list for all non-buffer items + * - inode_buffer_list for inode unlink buffers + * - cancel_list for the cancelled buffers */ STATIC int xlog_recover_reorder_trans( @@ -1612,6 +1645,10 @@ xlog_recover_reorder_trans( { xlog_recover_item_t *item, *n; LIST_HEAD(sort_list); + LIST_HEAD(cancel_list); + LIST_HEAD(buffer_list); + LIST_HEAD(inode_buffer_list); + LIST_HEAD(inode_list); list_splice_init(&trans->r_itemq, &sort_list); list_for_each_entry_safe(item, n, &sort_list, ri_list) { @@ -1619,12 +1656,18 @@ xlog_recover_reorder_trans( switch (ITEM_TYPE(item)) { case XFS_LI_BUF: - if (!(buf_f->blf_flags & XFS_BLF_CANCEL)) { + if (buf_f->blf_flags & XFS_BLF_CANCEL) { trace_xfs_log_recover_item_reorder_head(log, trans, item, pass); - list_move(&item->ri_list, &trans->r_itemq); + list_move(&item->ri_list, &cancel_list); + break; + } + if (buf_f->blf_flags & XFS_BLF_INODE_BUF) { + list_move(&item->ri_list, &inode_buffer_list); break; } + list_move_tail(&item->ri_list, &buffer_list); + break; case XFS_LI_INODE: case XFS_LI_DQUOT: case XFS_LI_QUOTAOFF: @@ -1632,7 +1675,7 @@ xlog_recover_reorder_trans( case XFS_LI_EFI: trace_xfs_log_recover_item_reorder_tail(log, trans, item, pass); - list_move_tail(&item->ri_list, &trans->r_itemq); + list_move_tail(&item->ri_list, &inode_list); break; default: xfs_warn(log->l_mp, @@ -1643,6 +1686,14 @@ xlog_recover_reorder_trans( } } ASSERT(list_empty(&sort_list)); + if (!list_empty(&buffer_list)) + list_splice(&buffer_list, &trans->r_itemq); + if (!list_empty(&inode_list)) + list_splice_tail(&inode_list, &trans->r_itemq); + if (!list_empty(&inode_buffer_list)) + list_splice_tail(&inode_buffer_list, &trans->r_itemq); + if (!list_empty(&cancel_list)) + list_splice_tail(&cancel_list, &trans->r_itemq); return 0; } @@ -1794,7 +1845,13 @@ xlog_recover_do_inode_buffer( xfs_agino_t *buffer_nextp; trace_xfs_log_recover_buf_inode_buf(mp->m_log, buf_f); - bp->b_ops = &xfs_inode_buf_ops; + + /* + * Post recovery validation only works properly on CRC enabled + * filesystems. + */ + if (xfs_sb_version_hascrc(&mp->m_sb)) + bp->b_ops = &xfs_inode_buf_ops; inodes_per_buf = BBTOB(bp->b_io_length) >> mp->m_sb.sb_inodelog; for (i = 0; i < inodes_per_buf; i++) { @@ -1861,6 +1918,15 @@ xlog_recover_do_inode_buffer( buffer_nextp = (xfs_agino_t *)xfs_buf_offset(bp, next_unlinked_offset); *buffer_nextp = *logged_nextp; + + /* + * If necessary, recalculate the CRC in the on-disk inode. We + * have to leave the inode in a consistent state for whoever + * reads it next.... + */ + xfs_dinode_calc_crc(mp, (struct xfs_dinode *) + xfs_buf_offset(bp, i * mp->m_sb.sb_inodesize)); + } return 0; @@ -2097,6 +2163,17 @@ xlog_recover_do_reg_buffer( ((uint)bit << XFS_BLF_SHIFT) + (nbits << XFS_BLF_SHIFT)); /* + * The dirty regions logged in the buffer, even though + * contiguous, may span multiple chunks. This is because the + * dirty region may span a physical page boundary in a buffer + * and hence be split into two separate vectors for writing into + * the log. Hence we need to trim nbits back to the length of + * the current region being copied out of the log. + */ + if (item->ri_buf[i].i_len < (nbits << XFS_BLF_SHIFT)) + nbits = item->ri_buf[i].i_len >> XFS_BLF_SHIFT; + + /* * Do a sanity check if this is a dquot buffer. Just checking * the first dquot in the buffer should do. XXXThis is * probably a good thing to do for other buf types also. @@ -2134,7 +2211,16 @@ xlog_recover_do_reg_buffer( /* Shouldn't be any more regions */ ASSERT(i == item->ri_total); - xlog_recovery_validate_buf_type(mp, bp, buf_f); + /* + * We can only do post recovery validation on items on CRC enabled + * fielsystems as we need to know when the buffer was written to be able + * to determine if we should have replayed the item. If we replay old + * metadata over a newer buffer, then it will enter a temporarily + * inconsistent state resulting in verification failures. Hence for now + * just avoid the verification stage for non-crc filesystems + */ + if (xfs_sb_version_hascrc(&mp->m_sb)) + xlog_recovery_validate_buf_type(mp, bp, buf_f); } /* @@ -2255,6 +2341,12 @@ xfs_qm_dqcheck( d->dd_diskdq.d_flags = type; d->dd_diskdq.d_id = cpu_to_be32(id); + if (xfs_sb_version_hascrc(&mp->m_sb)) { + uuid_copy(&d->dd_uuid, &mp->m_sb.sb_uuid); + xfs_update_cksum((char *)d, sizeof(struct xfs_dqblk), + XFS_DQUOT_CRC_OFF); + } + return errs; } @@ -2782,6 +2874,10 @@ xlog_recover_dquot_pass2( } memcpy(ddq, recddq, item->ri_buf[1].i_len); + if (xfs_sb_version_hascrc(&mp->m_sb)) { + xfs_update_cksum((char *)ddq, sizeof(struct xfs_dqblk), + XFS_DQUOT_CRC_OFF); + } ASSERT(dq_f->qlf_size == 2); ASSERT(bp->b_target->bt_mount == mp); diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index f6bfbd734669..e8e310c05097 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -314,7 +314,8 @@ STATIC int xfs_mount_validate_sb( xfs_mount_t *mp, xfs_sb_t *sbp, - bool check_inprogress) + bool check_inprogress, + bool check_version) { /* @@ -337,9 +338,10 @@ xfs_mount_validate_sb( /* * Version 5 superblock feature mask validation. Reject combinations the - * kernel cannot support up front before checking anything else. + * kernel cannot support up front before checking anything else. For + * write validation, we don't need to check feature masks. */ - if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) { + if (check_version && XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) { xfs_alert(mp, "Version 5 superblock detected. This kernel has EXPERIMENTAL support enabled!\n" "Use of these features in this kernel is at your own risk!"); @@ -675,7 +677,8 @@ xfs_sb_to_disk( static int xfs_sb_verify( - struct xfs_buf *bp) + struct xfs_buf *bp, + bool check_version) { struct xfs_mount *mp = bp->b_target->bt_mount; struct xfs_sb sb; @@ -686,7 +689,8 @@ xfs_sb_verify( * Only check the in progress field for the primary superblock as * mkfs.xfs doesn't clear it from secondary superblocks. */ - return xfs_mount_validate_sb(mp, &sb, bp->b_bn == XFS_SB_DADDR); + return xfs_mount_validate_sb(mp, &sb, bp->b_bn == XFS_SB_DADDR, + check_version); } /* @@ -719,7 +723,7 @@ xfs_sb_read_verify( goto out_error; } } - error = xfs_sb_verify(bp); + error = xfs_sb_verify(bp, true); out_error: if (error) { @@ -758,7 +762,7 @@ xfs_sb_write_verify( struct xfs_buf_log_item *bip = bp->b_fspriv; int error; - error = xfs_sb_verify(bp); + error = xfs_sb_verify(bp, false); if (error) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr); xfs_buf_ioerror(bp, error); diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c index f41702b43003..b75c9bb6e71e 100644 --- a/fs/xfs/xfs_qm.c +++ b/fs/xfs/xfs_qm.c @@ -41,6 +41,7 @@ #include "xfs_qm.h" #include "xfs_trace.h" #include "xfs_icache.h" +#include "xfs_cksum.h" /* * The global quota manager. There is only one of these for the entire @@ -839,7 +840,7 @@ xfs_qm_reset_dqcounts( xfs_dqid_t id, uint type) { - xfs_disk_dquot_t *ddq; + struct xfs_dqblk *dqb; int j; trace_xfs_reset_dqcounts(bp, _RET_IP_); @@ -853,8 +854,12 @@ xfs_qm_reset_dqcounts( do_div(j, sizeof(xfs_dqblk_t)); ASSERT(mp->m_quotainfo->qi_dqperchunk == j); #endif - ddq = bp->b_addr; + dqb = bp->b_addr; for (j = 0; j < mp->m_quotainfo->qi_dqperchunk; j++) { + struct xfs_disk_dquot *ddq; + + ddq = (struct xfs_disk_dquot *)&dqb[j]; + /* * Do a sanity check, and if needed, repair the dqblk. Don't * output any warnings because it's perfectly possible to @@ -871,7 +876,12 @@ xfs_qm_reset_dqcounts( ddq->d_bwarns = 0; ddq->d_iwarns = 0; ddq->d_rtbwarns = 0; - ddq = (xfs_disk_dquot_t *) ((xfs_dqblk_t *)ddq + 1); + + if (xfs_sb_version_hascrc(&mp->m_sb)) { + xfs_update_cksum((char *)&dqb[j], + sizeof(struct xfs_dqblk), + XFS_DQUOT_CRC_OFF); + } } } @@ -907,19 +917,29 @@ xfs_qm_dqiter_bufs( XFS_FSB_TO_DADDR(mp, bno), mp->m_quotainfo->qi_dqchunklen, 0, &bp, &xfs_dquot_buf_ops); - if (error) - break; /* - * XXX(hch): need to figure out if it makes sense to validate - * the CRC here. + * CRC and validation errors will return a EFSCORRUPTED here. If + * this occurs, re-read without CRC validation so that we can + * repair the damage via xfs_qm_reset_dqcounts(). This process + * will leave a trace in the log indicating corruption has + * been detected. */ + if (error == EFSCORRUPTED) { + error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, + XFS_FSB_TO_DADDR(mp, bno), + mp->m_quotainfo->qi_dqchunklen, 0, &bp, + NULL); + } + + if (error) + break; + xfs_qm_reset_dqcounts(mp, bp, firstid, type); xfs_buf_delwri_queue(bp, buffer_list); xfs_buf_relse(bp); - /* - * goto the next block. - */ + + /* goto the next block. */ bno++; firstid += mp->m_quotainfo->qi_dqperchunk; } diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c index c41190cad6e9..6cdf6ffc36a1 100644 --- a/fs/xfs/xfs_qm_syscalls.c +++ b/fs/xfs/xfs_qm_syscalls.c @@ -489,31 +489,36 @@ xfs_qm_scall_setqlim( if ((newlim->d_fieldmask & XFS_DQ_MASK) == 0) return 0; - tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SETQLIM); - error = xfs_trans_reserve(tp, 0, XFS_QM_SETQLIM_LOG_RES(mp), - 0, 0, XFS_DEFAULT_LOG_COUNT); - if (error) { - xfs_trans_cancel(tp, 0); - return (error); - } - /* * We don't want to race with a quotaoff so take the quotaoff lock. - * (We don't hold an inode lock, so there's nothing else to stop - * a quotaoff from happening). (XXXThis doesn't currently happen - * because we take the vfslock before calling xfs_qm_sysent). + * We don't hold an inode lock, so there's nothing else to stop + * a quotaoff from happening. */ mutex_lock(&q->qi_quotaofflock); /* - * Get the dquot (locked), and join it to the transaction. - * Allocate the dquot if this doesn't exist. + * Get the dquot (locked) before we start, as we need to do a + * transaction to allocate it if it doesn't exist. Once we have the + * dquot, unlock it so we can start the next transaction safely. We hold + * a reference to the dquot, so it's safe to do this unlock/lock without + * it being reclaimed in the mean time. */ - if ((error = xfs_qm_dqget(mp, NULL, id, type, XFS_QMOPT_DQALLOC, &dqp))) { - xfs_trans_cancel(tp, XFS_TRANS_ABORT); + error = xfs_qm_dqget(mp, NULL, id, type, XFS_QMOPT_DQALLOC, &dqp); + if (error) { ASSERT(error != ENOENT); goto out_unlock; } + xfs_dqunlock(dqp); + + tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SETQLIM); + error = xfs_trans_reserve(tp, 0, XFS_QM_SETQLIM_LOG_RES(mp), + 0, 0, XFS_DEFAULT_LOG_COUNT); + if (error) { + xfs_trans_cancel(tp, 0); + goto out_rele; + } + + xfs_dqlock(dqp); xfs_trans_dqjoin(tp, dqp); ddq = &dqp->q_core; @@ -621,9 +626,10 @@ xfs_qm_scall_setqlim( xfs_trans_log_dquot(tp, dqp); error = xfs_trans_commit(tp, 0); - xfs_qm_dqrele(dqp); - out_unlock: +out_rele: + xfs_qm_dqrele(dqp); +out_unlock: mutex_unlock(&q->qi_quotaofflock); return error; } diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h index c61e31c7d997..c38068f26c55 100644 --- a/fs/xfs/xfs_quota.h +++ b/fs/xfs/xfs_quota.h @@ -87,6 +87,8 @@ typedef struct xfs_dqblk { uuid_t dd_uuid; /* location information */ } xfs_dqblk_t; +#define XFS_DQUOT_CRC_OFF offsetof(struct xfs_dqblk, dd_crc) + /* * flags for q_flags field in the dquot. */ diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index ea341cea68cb..3033ba5e9762 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1373,6 +1373,17 @@ xfs_finish_flags( } /* + * V5 filesystems always use attr2 format for attributes. + */ + if (xfs_sb_version_hascrc(&mp->m_sb) && + (mp->m_flags & XFS_MOUNT_NOATTR2)) { + xfs_warn(mp, +"Cannot mount a V5 filesystem as %s. %s is always enabled for V5 filesystems.", + MNTOPT_NOATTR2, MNTOPT_ATTR2); + return XFS_ERROR(EINVAL); + } + + /* * mkfs'ed attr2 will turn on attr2 mount unless explicitly * told by noattr2 to turn it off */ diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c index 5f234389327c..195a403e1522 100644 --- a/fs/xfs/xfs_symlink.c +++ b/fs/xfs/xfs_symlink.c @@ -56,16 +56,9 @@ xfs_symlink_blocks( struct xfs_mount *mp, int pathlen) { - int fsblocks = 0; - int len = pathlen; + int buflen = XFS_SYMLINK_BUF_SPACE(mp, mp->m_sb.sb_blocksize); - do { - fsblocks++; - len -= XFS_SYMLINK_BUF_SPACE(mp, mp->m_sb.sb_blocksize); - } while (len > 0); - - ASSERT(fsblocks <= XFS_SYMLINK_MAPS); - return fsblocks; + return (pathlen + buflen - 1) / buflen; } static int @@ -405,7 +398,7 @@ xfs_symlink( if (pathlen <= XFS_LITINO(mp, dp->i_d.di_version)) fs_blocks = 0; else - fs_blocks = XFS_B_TO_FSB(mp, pathlen); + fs_blocks = xfs_symlink_blocks(mp, pathlen); resblks = XFS_SYMLINK_SPACE_RES(mp, link_name->len, fs_blocks); error = xfs_trans_reserve(tp, resblks, XFS_SYMLINK_LOG_RES(mp), 0, XFS_TRANS_PERM_LOG_RES, XFS_SYMLINK_LOG_COUNT); @@ -512,7 +505,7 @@ xfs_symlink( cur_chunk = target_path; offset = 0; for (n = 0; n < nmaps; n++) { - char *buf; + char *buf; d = XFS_FSB_TO_DADDR(mp, mval[n].br_startblock); byte_cnt = XFS_FSB_TO_B(mp, mval[n].br_blockcount); @@ -525,9 +518,7 @@ xfs_symlink( bp->b_ops = &xfs_symlink_buf_ops; byte_cnt = XFS_SYMLINK_BUF_SPACE(mp, byte_cnt); - if (pathlen < byte_cnt) { - byte_cnt = pathlen; - } + byte_cnt = min(byte_cnt, pathlen); buf = bp->b_addr; buf += xfs_symlink_hdr_set(mp, ip->i_ino, offset, @@ -542,6 +533,7 @@ xfs_symlink( xfs_trans_log_buf(tp, bp, 0, (buf + byte_cnt - 1) - (char *)bp->b_addr); } + ASSERT(pathlen == 0); } /* diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 1501f4fa51a6..0176bb21f09a 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -1453,7 +1453,7 @@ xfs_free_file_space( xfs_mount_t *mp; int nimap; uint resblks; - uint rounding; + xfs_off_t rounding; int rt; xfs_fileoff_t startoffset_fsb; xfs_trans_t *tp; @@ -1482,7 +1482,7 @@ xfs_free_file_space( inode_dio_wait(VFS_I(ip)); } - rounding = max_t(uint, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE); + rounding = max_t(xfs_off_t, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE); ioffset = offset & ~(rounding - 1); error = -filemap_write_and_wait_range(VFS_I(ip)->i_mapping, ioffset, -1); diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h index 5163022d9808..38c67c34d73f 100644 --- a/fs/xfs/xfs_vnodeops.h +++ b/fs/xfs/xfs_vnodeops.h @@ -31,8 +31,7 @@ int xfs_remove(struct xfs_inode *dp, struct xfs_name *name, struct xfs_inode *ip); int xfs_link(struct xfs_inode *tdp, struct xfs_inode *sip, struct xfs_name *target_name); -int xfs_readdir(struct xfs_inode *dp, void *dirent, size_t bufsize, - xfs_off_t *offset, filldir_t filldir); +int xfs_readdir(struct xfs_inode *dp, struct dir_context *ctx, size_t bufsize); int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name, const char *target_path, umode_t mode, struct xfs_inode **ipp); int xfs_set_dmattrs(struct xfs_inode *ip, u_int evmask, u_int16_t state); |