diff options
Diffstat (limited to 'fs')
141 files changed, 2584 insertions, 2194 deletions
diff --git a/fs/affs/super.c b/fs/affs/super.c index 6d589f28bf9b..895ac7dc9dbf 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -340,8 +340,6 @@ static int affs_fill_super(struct super_block *sb, void *data, int silent) &blocksize,&sbi->s_prefix, sbi->s_volume, &mount_flags)) { printk(KERN_ERR "AFFS: Error parsing options\n"); - kfree(sbi->s_prefix); - kfree(sbi); return -EINVAL; } /* N.B. after this point s_prefix must be released */ @@ -112,6 +112,11 @@ struct kioctx { struct work_struct free_work; + /* + * signals when all in-flight requests are done + */ + struct completion *requests_done; + struct { /* * This counts the number of available slots in the ringbuffer, @@ -508,6 +513,10 @@ static void free_ioctx_reqs(struct percpu_ref *ref) { struct kioctx *ctx = container_of(ref, struct kioctx, reqs); + /* At this point we know that there are no any in-flight requests */ + if (ctx->requests_done) + complete(ctx->requests_done); + INIT_WORK(&ctx->free_work, free_ioctx); schedule_work(&ctx->free_work); } @@ -718,7 +727,8 @@ err: * when the processes owning a context have all exited to encourage * the rapid destruction of the kioctx. */ -static void kill_ioctx(struct mm_struct *mm, struct kioctx *ctx) +static void kill_ioctx(struct mm_struct *mm, struct kioctx *ctx, + struct completion *requests_done) { if (!atomic_xchg(&ctx->dead, 1)) { struct kioctx_table *table; @@ -747,7 +757,11 @@ static void kill_ioctx(struct mm_struct *mm, struct kioctx *ctx) if (ctx->mmap_size) vm_munmap(ctx->mmap_base, ctx->mmap_size); + ctx->requests_done = requests_done; percpu_ref_kill(&ctx->users); + } else { + if (requests_done) + complete(requests_done); } } @@ -809,7 +823,7 @@ void exit_aio(struct mm_struct *mm) */ ctx->mmap_size = 0; - kill_ioctx(mm, ctx); + kill_ioctx(mm, ctx, NULL); } } @@ -1185,7 +1199,7 @@ SYSCALL_DEFINE2(io_setup, unsigned, nr_events, aio_context_t __user *, ctxp) if (!IS_ERR(ioctx)) { ret = put_user(ioctx->user_id, ctxp); if (ret) - kill_ioctx(current->mm, ioctx); + kill_ioctx(current->mm, ioctx, NULL); percpu_ref_put(&ioctx->users); } @@ -1203,8 +1217,22 @@ SYSCALL_DEFINE1(io_destroy, aio_context_t, ctx) { struct kioctx *ioctx = lookup_ioctx(ctx); if (likely(NULL != ioctx)) { - kill_ioctx(current->mm, ioctx); + struct completion requests_done = + COMPLETION_INITIALIZER_ONSTACK(requests_done); + + /* Pass requests_done to kill_ioctx() where it can be set + * in a thread-safe way. If we try to set it here then we have + * a race condition if two io_destroy() called simultaneously. + */ + kill_ioctx(current->mm, ioctx, &requests_done); percpu_ref_put(&ioctx->users); + + /* Wait until all IO for the context are done. Otherwise kernel + * keep using user-space buffers even if user thinks the context + * is destroyed. + */ + wait_for_completion(&requests_done); + return 0; } pr_debug("EINVAL: io_destroy: invalid context id\n"); @@ -1299,10 +1327,8 @@ rw_common: &iovec, compat) : aio_setup_single_vector(req, rw, buf, &nr_segs, iovec); - if (ret) - return ret; - - ret = rw_verify_area(rw, file, &req->ki_pos, req->ki_nbytes); + if (!ret) + ret = rw_verify_area(rw, file, &req->ki_pos, req->ki_nbytes); if (ret < 0) { if (iovec != &inline_vec) kfree(iovec); diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 2caf36ac3e93..cc87c1abac97 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -179,7 +179,7 @@ static struct dentry *autofs4_lookup_active(struct dentry *dentry) spin_lock(&active->d_lock); /* Already gone? */ - if (!d_count(active)) + if ((int) d_count(active) <= 0) goto next; qstr = &active->d_name; @@ -230,7 +230,7 @@ static struct dentry *autofs4_lookup_expiring(struct dentry *dentry) spin_lock(&expiring->d_lock); - /* Bad luck, we've already been dentry_iput */ + /* We've already been dentry_iput or unlinked */ if (!expiring->d_inode) goto next; diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 4c48df572bd6..ba6b88528dc7 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -2058,6 +2058,20 @@ struct btrfs_ioctl_defrag_range_args { #define btrfs_raw_test_opt(o, opt) ((o) & BTRFS_MOUNT_##opt) #define btrfs_test_opt(root, opt) ((root)->fs_info->mount_opt & \ BTRFS_MOUNT_##opt) +#define btrfs_set_and_info(root, opt, fmt, args...) \ +{ \ + if (!btrfs_test_opt(root, opt)) \ + btrfs_info(root->fs_info, fmt, ##args); \ + btrfs_set_opt(root->fs_info->mount_opt, opt); \ +} + +#define btrfs_clear_and_info(root, opt, fmt, args...) \ +{ \ + if (btrfs_test_opt(root, opt)) \ + btrfs_info(root->fs_info, fmt, ##args); \ + btrfs_clear_opt(root->fs_info->mount_opt, opt); \ +} + /* * Inode flags */ diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 029d46c2e170..983314932af3 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2861,7 +2861,7 @@ retry_root_backup: printk(KERN_ERR "BTRFS: failed to read log tree\n"); free_extent_buffer(log_tree_root->node); kfree(log_tree_root); - goto fail_trans_kthread; + goto fail_qgroup; } /* returns with log_tree_root freed on success */ ret = btrfs_recover_log_trees(log_tree_root); @@ -2870,24 +2870,24 @@ retry_root_backup: "Failed to recover log tree"); free_extent_buffer(log_tree_root->node); kfree(log_tree_root); - goto fail_trans_kthread; + goto fail_qgroup; } if (sb->s_flags & MS_RDONLY) { ret = btrfs_commit_super(tree_root); if (ret) - goto fail_trans_kthread; + goto fail_qgroup; } } ret = btrfs_find_orphan_roots(tree_root); if (ret) - goto fail_trans_kthread; + goto fail_qgroup; if (!(sb->s_flags & MS_RDONLY)) { ret = btrfs_cleanup_fs_roots(fs_info); if (ret) - goto fail_trans_kthread; + goto fail_qgroup; ret = btrfs_recover_relocation(tree_root); if (ret < 0) { diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 1306487c82cf..5590af92094b 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -1542,6 +1542,7 @@ again: ret = 0; } if (ret) { + key.objectid = bytenr; key.type = BTRFS_EXTENT_ITEM_KEY; key.offset = num_bytes; btrfs_release_path(path); @@ -3542,11 +3543,13 @@ static u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags) return extended_to_chunk(flags | tmp); } -static u64 get_alloc_profile(struct btrfs_root *root, u64 flags) +static u64 get_alloc_profile(struct btrfs_root *root, u64 orig_flags) { unsigned seq; + u64 flags; do { + flags = orig_flags; seq = read_seqbegin(&root->fs_info->profiles_lock); if (flags & BTRFS_BLOCK_GROUP_DATA) @@ -5719,6 +5722,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, if (ret > 0 && skinny_metadata) { skinny_metadata = false; + key.objectid = bytenr; key.type = BTRFS_EXTENT_ITEM_KEY; key.offset = num_bytes; btrfs_release_path(path); diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index eb742c07e7a4..ae6af072b635 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -800,7 +800,7 @@ next_slot: if (start > key.offset && end < extent_end) { BUG_ON(del_nr > 0); if (extent_type == BTRFS_FILE_EXTENT_INLINE) { - ret = -EINVAL; + ret = -EOPNOTSUPP; break; } @@ -846,7 +846,7 @@ next_slot: */ if (start <= key.offset && end < extent_end) { if (extent_type == BTRFS_FILE_EXTENT_INLINE) { - ret = -EINVAL; + ret = -EOPNOTSUPP; break; } @@ -872,7 +872,7 @@ next_slot: if (start > key.offset && end >= extent_end) { BUG_ON(del_nr > 0); if (extent_type == BTRFS_FILE_EXTENT_INLINE) { - ret = -EINVAL; + ret = -EOPNOTSUPP; break; } @@ -1777,7 +1777,7 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb, start_pos = round_down(pos, root->sectorsize); if (start_pos > i_size_read(inode)) { /* Expand hole size to cover write data, preventing empty gap */ - end_pos = round_up(pos + iov->iov_len, root->sectorsize); + end_pos = round_up(pos + count, root->sectorsize); err = btrfs_cont_expand(inode, i_size_read(inode), end_pos); if (err) { mutex_unlock(&inode->i_mutex); diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-map.c index cc8ca193d830..86935f5ae291 100644 --- a/fs/btrfs/inode-map.c +++ b/fs/btrfs/inode-map.c @@ -176,7 +176,11 @@ static void start_caching(struct btrfs_root *root) tsk = kthread_run(caching_kthread, root, "btrfs-ino-cache-%llu\n", root->root_key.objectid); - BUG_ON(IS_ERR(tsk)); /* -ENOMEM */ + if (IS_ERR(tsk)) { + btrfs_warn(root->fs_info, "failed to start inode caching task"); + btrfs_clear_and_info(root, CHANGE_INODE_CACHE, + "disabling inode map caching"); + } } int btrfs_find_free_ino(struct btrfs_root *root, u64 *objectid) @@ -205,24 +209,14 @@ again: void btrfs_return_ino(struct btrfs_root *root, u64 objectid) { - struct btrfs_free_space_ctl *ctl = root->free_ino_ctl; struct btrfs_free_space_ctl *pinned = root->free_ino_pinned; if (!btrfs_test_opt(root, INODE_MAP_CACHE)) return; - again: if (root->cached == BTRFS_CACHE_FINISHED) { - __btrfs_add_free_space(ctl, objectid, 1); + __btrfs_add_free_space(pinned, objectid, 1); } else { - /* - * If we are in the process of caching free ino chunks, - * to avoid adding the same inode number to the free_ino - * tree twice due to cross transaction, we'll leave it - * in the pinned tree until a transaction is committed - * or the caching work is done. - */ - down_write(&root->fs_info->commit_root_sem); spin_lock(&root->cache_lock); if (root->cached == BTRFS_CACHE_FINISHED) { @@ -234,11 +228,7 @@ again: start_caching(root); - if (objectid <= root->cache_progress || - objectid >= root->highest_objectid) - __btrfs_add_free_space(ctl, objectid, 1); - else - __btrfs_add_free_space(pinned, objectid, 1); + __btrfs_add_free_space(pinned, objectid, 1); up_write(&root->fs_info->commit_root_sem); } diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index e79ff6b90cb7..2ad7de94efef 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -3066,7 +3066,7 @@ process_slot: new_key.offset + datal, 1); if (ret) { - if (ret != -EINVAL) + if (ret != -EOPNOTSUPP) btrfs_abort_transaction(trans, root, ret); btrfs_end_transaction(trans, root); @@ -3141,7 +3141,7 @@ process_slot: new_key.offset + datal, 1); if (ret) { - if (ret != -EINVAL) + if (ret != -EOPNOTSUPP) btrfs_abort_transaction(trans, root, ret); btrfs_end_transaction(trans, root); diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 1ac3ca98c429..eb6537a08c1b 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -349,6 +349,11 @@ static int fs_path_ensure_buf(struct fs_path *p, int len) if (p->buf_len >= len) return 0; + if (len > PATH_MAX) { + WARN_ON(1); + return -ENOMEM; + } + path_len = p->end - p->start; old_buf_len = p->buf_len; diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 5011aadacab8..9601d25a4607 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -385,20 +385,6 @@ static match_table_t tokens = { {Opt_err, NULL}, }; -#define btrfs_set_and_info(root, opt, fmt, args...) \ -{ \ - if (!btrfs_test_opt(root, opt)) \ - btrfs_info(root->fs_info, fmt, ##args); \ - btrfs_set_opt(root->fs_info->mount_opt, opt); \ -} - -#define btrfs_clear_and_info(root, opt, fmt, args...) \ -{ \ - if (btrfs_test_opt(root, opt)) \ - btrfs_info(root->fs_info, fmt, ##args); \ - btrfs_clear_opt(root->fs_info->mount_opt, opt); \ -} - /* * Regular mount options parser. Everything that is needed only when * reading in a new superblock is parsed here. @@ -1186,7 +1172,6 @@ static struct dentry *mount_subvol(const char *subvol_name, int flags, return ERR_PTR(-ENOMEM); mnt = vfs_kern_mount(&btrfs_fs_type, flags, device_name, newargs); - kfree(newargs); if (PTR_RET(mnt) == -EBUSY) { if (flags & MS_RDONLY) { @@ -1196,17 +1181,22 @@ static struct dentry *mount_subvol(const char *subvol_name, int flags, int r; mnt = vfs_kern_mount(&btrfs_fs_type, flags | MS_RDONLY, device_name, newargs); - if (IS_ERR(mnt)) + if (IS_ERR(mnt)) { + kfree(newargs); return ERR_CAST(mnt); + } r = btrfs_remount(mnt->mnt_sb, &flags, NULL); if (r < 0) { /* FIXME: release vfsmount mnt ??*/ + kfree(newargs); return ERR_PTR(r); } } } + kfree(newargs); + if (IS_ERR(mnt)) return ERR_CAST(mnt); diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 2e5e648eb5c3..c561b628ebce 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -3261,7 +3261,7 @@ int ceph_encode_inode_release(void **p, struct inode *inode, rel->seq = cpu_to_le32(cap->seq); rel->issue_seq = cpu_to_le32(cap->issue_seq), rel->mseq = cpu_to_le32(cap->mseq); - rel->caps = cpu_to_le32(cap->issued); + rel->caps = cpu_to_le32(cap->implemented); rel->wanted = cpu_to_le32(cap->mds_wanted); rel->dname_len = 0; rel->dname_seq = 0; diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 766410a12c2c..c29d6ae68874 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -141,7 +141,7 @@ static int __dcache_readdir(struct file *file, struct dir_context *ctx, /* start at beginning? */ if (ctx->pos == 2 || last == NULL || - ctx->pos < ceph_dentry(last)->offset) { + fpos_cmp(ctx->pos, ceph_dentry(last)->offset) < 0) { if (list_empty(&parent->d_subdirs)) goto out_unlock; p = parent->d_subdirs.prev; @@ -182,9 +182,16 @@ more: spin_unlock(&dentry->d_lock); spin_unlock(&parent->d_lock); + /* make sure a dentry wasn't dropped while we didn't have parent lock */ + if (!ceph_dir_is_complete(dir)) { + dout(" lost dir complete on %p; falling back to mds\n", dir); + dput(dentry); + err = -EAGAIN; + goto out; + } + dout(" %llu (%llu) dentry %p %.*s %p\n", di->offset, ctx->pos, dentry, dentry->d_name.len, dentry->d_name.name, dentry->d_inode); - 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), @@ -198,19 +205,12 @@ more: return 0; } + ctx->pos = di->offset + 1; + if (last) dput(last); last = dentry; - ctx->pos++; - - /* make sure a dentry wasn't dropped while we didn't have parent lock */ - if (!ceph_dir_is_complete(dir)) { - dout(" lost dir complete on %p; falling back to mds\n", dir); - err = -EAGAIN; - goto out; - } - spin_lock(&parent->d_lock); p = p->prev; /* advance to next dentry */ goto more; @@ -296,6 +296,8 @@ static int ceph_readdir(struct file *file, struct dir_context *ctx) err = __dcache_readdir(file, ctx, shared_gen); if (err != -EAGAIN) return err; + frag = fpos_frag(ctx->pos); + off = fpos_off(ctx->pos); } else { spin_unlock(&ci->i_ceph_lock); } @@ -446,7 +448,6 @@ 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 = ctx->pos; } spin_unlock(&ci->i_ceph_lock); @@ -935,14 +936,16 @@ static int ceph_rename(struct inode *old_dir, struct dentry *old_dentry, * to do it here. */ - /* d_move screws up d_subdirs order */ - ceph_dir_clear_complete(new_dir); - d_move(old_dentry, new_dentry); /* ensure target dentry is invalidated, despite rehashing bug in vfs_rename_dir */ ceph_invalidate_dentry_lease(new_dentry); + + /* d_move screws up sibling dentries' offsets */ + ceph_dir_clear_complete(old_dir); + ceph_dir_clear_complete(new_dir); + } ceph_mdsc_put_request(req); return err; diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 39da1c2efa50..88a6df4cbe6d 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -1221,9 +1221,6 @@ static long ceph_fallocate(struct file *file, int mode, if (!S_ISREG(inode->i_mode)) return -EOPNOTSUPP; - if (IS_SWAPFILE(inode)) - return -ETXTBSY; - mutex_lock(&inode->i_mutex); if (ceph_snap(inode) != CEPH_NOSNAP) { diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index 0b0728e5be2d..233c6f96910a 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -744,7 +744,6 @@ static int fill_inode(struct inode *inode, !__ceph_dir_is_complete(ci)) { dout(" marking %p complete (empty)\n", inode); __ceph_dir_set_complete(ci, atomic_read(&ci->i_release_count)); - ci->i_max_offset = 2; } no_change: /* only update max_size on auth cap */ @@ -890,41 +889,6 @@ out_unlock: } /* - * Set dentry's directory position based on the current dir's max, and - * order it in d_subdirs, so that dcache_readdir behaves. - * - * Always called under directory's i_mutex. - */ -static void ceph_set_dentry_offset(struct dentry *dn) -{ - struct dentry *dir = dn->d_parent; - struct inode *inode = dir->d_inode; - struct ceph_inode_info *ci; - struct ceph_dentry_info *di; - - BUG_ON(!inode); - - ci = ceph_inode(inode); - di = ceph_dentry(dn); - - spin_lock(&ci->i_ceph_lock); - if (!__ceph_dir_is_complete(ci)) { - spin_unlock(&ci->i_ceph_lock); - return; - } - di->offset = ceph_inode(inode)->i_max_offset++; - spin_unlock(&ci->i_ceph_lock); - - spin_lock(&dir->d_lock); - spin_lock_nested(&dn->d_lock, DENTRY_D_LOCK_NESTED); - list_move(&dn->d_u.d_child, &dir->d_subdirs); - dout("set_dentry_offset %p %lld (%p %p)\n", dn, di->offset, - dn->d_u.d_child.prev, dn->d_u.d_child.next); - spin_unlock(&dn->d_lock); - spin_unlock(&dir->d_lock); -} - -/* * splice a dentry to an inode. * caller must hold directory i_mutex for this to be safe. * @@ -933,7 +897,7 @@ static void ceph_set_dentry_offset(struct dentry *dn) * the caller) if we fail. */ static struct dentry *splice_dentry(struct dentry *dn, struct inode *in, - bool *prehash, bool set_offset) + bool *prehash) { struct dentry *realdn; @@ -965,8 +929,6 @@ static struct dentry *splice_dentry(struct dentry *dn, struct inode *in, } if ((!prehash || *prehash) && d_unhashed(dn)) d_rehash(dn); - if (set_offset) - ceph_set_dentry_offset(dn); out: return dn; } @@ -987,7 +949,6 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req, { struct ceph_mds_reply_info_parsed *rinfo = &req->r_reply_info; struct inode *in = NULL; - struct ceph_mds_reply_inode *ininfo; struct ceph_vino vino; struct ceph_fs_client *fsc = ceph_sb_to_client(sb); int err = 0; @@ -1161,6 +1122,9 @@ retry_lookup: /* rename? */ if (req->r_old_dentry && req->r_op == CEPH_MDS_OP_RENAME) { + struct inode *olddir = req->r_old_dentry_dir; + BUG_ON(!olddir); + dout(" src %p '%.*s' dst %p '%.*s'\n", req->r_old_dentry, req->r_old_dentry->d_name.len, @@ -1180,13 +1144,10 @@ retry_lookup: rehashing bug in vfs_rename_dir */ ceph_invalidate_dentry_lease(dn); - /* - * d_move() puts the renamed dentry at the end of - * d_subdirs. We need to assign it an appropriate - * directory offset so we can behave when dir is - * complete. - */ - ceph_set_dentry_offset(req->r_old_dentry); + /* d_move screws up sibling dentries' offsets */ + ceph_dir_clear_complete(dir); + ceph_dir_clear_complete(olddir); + dout("dn %p gets new offset %lld\n", req->r_old_dentry, ceph_dentry(req->r_old_dentry)->offset); @@ -1213,8 +1174,9 @@ retry_lookup: /* attach proper inode */ if (!dn->d_inode) { + ceph_dir_clear_complete(dir); ihold(in); - dn = splice_dentry(dn, in, &have_lease, true); + dn = splice_dentry(dn, in, &have_lease); if (IS_ERR(dn)) { err = PTR_ERR(dn); goto done; @@ -1235,17 +1197,16 @@ retry_lookup: (req->r_op == CEPH_MDS_OP_LOOKUPSNAP || req->r_op == CEPH_MDS_OP_MKSNAP)) { struct dentry *dn = req->r_dentry; + struct inode *dir = req->r_locked_dir; /* fill out a snapdir LOOKUPSNAP dentry */ BUG_ON(!dn); - BUG_ON(!req->r_locked_dir); - BUG_ON(ceph_snap(req->r_locked_dir) != CEPH_SNAPDIR); - ininfo = rinfo->targeti.in; - vino.ino = le64_to_cpu(ininfo->ino); - vino.snap = le64_to_cpu(ininfo->snapid); + BUG_ON(!dir); + BUG_ON(ceph_snap(dir) != CEPH_SNAPDIR); dout(" linking snapped dir %p to dn %p\n", in, dn); + ceph_dir_clear_complete(dir); ihold(in); - dn = splice_dentry(dn, in, NULL, true); + dn = splice_dentry(dn, in, NULL); if (IS_ERR(dn)) { err = PTR_ERR(dn); goto done; @@ -1407,7 +1368,7 @@ retry_lookup: } if (!dn->d_inode) { - dn = splice_dentry(dn, in, NULL, false); + dn = splice_dentry(dn, in, NULL); if (IS_ERR(dn)) { err = PTR_ERR(dn); dn = NULL; diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c index fdf941b44ff1..a822a6e58290 100644 --- a/fs/ceph/ioctl.c +++ b/fs/ceph/ioctl.c @@ -109,6 +109,8 @@ static long ceph_ioctl_set_layout(struct file *file, void __user *arg) return PTR_ERR(req); req->r_inode = inode; ihold(inode); + req->r_num_caps = 1; + req->r_inode_drop = CEPH_CAP_FILE_SHARED | CEPH_CAP_FILE_EXCL; req->r_args.setlayout.layout.fl_stripe_unit = @@ -153,6 +155,7 @@ static long ceph_ioctl_set_layout_policy (struct file *file, void __user *arg) return PTR_ERR(req); req->r_inode = inode; ihold(inode); + req->r_num_caps = 1; req->r_args.setlayout.layout.fl_stripe_unit = cpu_to_le32(l.stripe_unit); diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c index d94ba0df9f4d..191398852a2e 100644 --- a/fs/ceph/locks.c +++ b/fs/ceph/locks.c @@ -45,6 +45,7 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file, return PTR_ERR(req); req->r_inode = inode; ihold(inode); + req->r_num_caps = 1; /* mds requires start and length rather than start and end */ if (LLONG_MAX == fl->fl_end) diff --git a/fs/ceph/super.h b/fs/ceph/super.h index 7866cd05a6bb..ead05cc1f447 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -266,7 +266,6 @@ struct ceph_inode_info { struct timespec i_rctime; u64 i_rbytes, i_rfiles, i_rsubdirs; u64 i_files, i_subdirs; - u64 i_max_offset; /* largest readdir offset, set with complete dir */ struct rb_root i_fragtree; struct mutex i_fragtree_mutex; diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index df9c9141c099..5be1f997ecde 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -253,6 +253,11 @@ cifs_alloc_inode(struct super_block *sb) cifs_set_oplock_level(cifs_inode, 0); cifs_inode->delete_pending = false; cifs_inode->invalid_mapping = false; + clear_bit(CIFS_INODE_PENDING_OPLOCK_BREAK, &cifs_inode->flags); + clear_bit(CIFS_INODE_PENDING_WRITERS, &cifs_inode->flags); + clear_bit(CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2, &cifs_inode->flags); + spin_lock_init(&cifs_inode->writers_lock); + cifs_inode->writers = 0; cifs_inode->vfs_inode.i_blkbits = 14; /* 2**14 = CIFS_MAX_MSGSIZE */ cifs_inode->server_eof = 0; cifs_inode->uniqueid = 0; @@ -732,19 +737,26 @@ static ssize_t cifs_file_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) { struct inode *inode = file_inode(iocb->ki_filp); + struct cifsInodeInfo *cinode = CIFS_I(inode); ssize_t written; int rc; + written = cifs_get_writer(cinode); + if (written) + return written; + written = generic_file_aio_write(iocb, iov, nr_segs, pos); if (CIFS_CACHE_WRITE(CIFS_I(inode))) - return written; + goto out; rc = filemap_fdatawrite(inode->i_mapping); if (rc) cifs_dbg(FYI, "cifs_file_aio_write: %d rc on %p inode\n", rc, inode); +out: + cifs_put_writer(cinode); return written; } diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index c0f3718b77a8..30f6e9251a4a 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -228,6 +228,8 @@ struct smb_version_operations { /* verify the message */ int (*check_message)(char *, unsigned int); bool (*is_oplock_break)(char *, struct TCP_Server_Info *); + void (*downgrade_oplock)(struct TCP_Server_Info *, + struct cifsInodeInfo *, bool); /* process transaction2 response */ bool (*check_trans2)(struct mid_q_entry *, struct TCP_Server_Info *, char *, int); @@ -1113,6 +1115,12 @@ struct cifsInodeInfo { unsigned int epoch; /* used to track lease state changes */ bool delete_pending; /* DELETE_ON_CLOSE is set */ bool invalid_mapping; /* pagecache is invalid */ + unsigned long flags; +#define CIFS_INODE_PENDING_OPLOCK_BREAK (0) /* oplock break in progress */ +#define CIFS_INODE_PENDING_WRITERS (1) /* Writes in progress */ +#define CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2 (2) /* Downgrade oplock to L2 */ + spinlock_t writers_lock; + unsigned int writers; /* Number of writers on this inode */ unsigned long time; /* jiffies of last update of inode */ u64 server_eof; /* current file size on server -- protected by i_lock */ u64 uniqueid; /* server inode number */ diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index acc4ee8ed075..ca7980a1e303 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -127,6 +127,9 @@ extern u64 cifs_UnixTimeToNT(struct timespec); extern struct timespec cnvrtDosUnixTm(__le16 le_date, __le16 le_time, int offset); extern void cifs_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock); +extern int cifs_get_writer(struct cifsInodeInfo *cinode); +extern void cifs_put_writer(struct cifsInodeInfo *cinode); +extern void cifs_done_oplock_break(struct cifsInodeInfo *cinode); extern int cifs_unlock_range(struct cifsFileInfo *cfile, struct file_lock *flock, const unsigned int xid); extern int cifs_push_mandatory_locks(struct cifsFileInfo *cfile); diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index f3264bd7a83d..6ce4e0954b98 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -6197,6 +6197,9 @@ QAllEAsRetry: cifs_dbg(FYI, "ea length %d\n", list_len); if (list_len <= 8) { cifs_dbg(FYI, "empty EA list returned from server\n"); + /* didn't find the named attribute */ + if (ea_name) + rc = -ENODATA; goto QAllEAsOut; } diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 8add25538a3b..5ed03e0b8b40 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -2599,7 +2599,7 @@ cifs_writev(struct kiocb *iocb, const struct iovec *iov, ssize_t err; err = generic_write_sync(file, iocb->ki_pos - rc, rc); - if (rc < 0) + if (err < 0) rc = err; } } else { @@ -2621,12 +2621,20 @@ cifs_strict_writev(struct kiocb *iocb, const struct iovec *iov, struct cifs_tcon *tcon = tlink_tcon(cfile->tlink); ssize_t written; + written = cifs_get_writer(cinode); + if (written) + return written; + if (CIFS_CACHE_WRITE(cinode)) { if (cap_unix(tcon->ses) && (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) - && ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0)) - return generic_file_aio_write(iocb, iov, nr_segs, pos); - return cifs_writev(iocb, iov, nr_segs, pos); + && ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0)) { + written = generic_file_aio_write( + iocb, iov, nr_segs, pos); + goto out; + } + written = cifs_writev(iocb, iov, nr_segs, pos); + goto out; } /* * For non-oplocked files in strict cache mode we need to write the data @@ -2646,6 +2654,8 @@ cifs_strict_writev(struct kiocb *iocb, const struct iovec *iov, inode); cinode->oplock = 0; } +out: + cifs_put_writer(cinode); return written; } @@ -2872,7 +2882,7 @@ ssize_t cifs_user_readv(struct kiocb *iocb, const struct iovec *iov, cifs_uncached_readv_complete); if (!rdata) { rc = -ENOMEM; - goto error; + break; } rc = cifs_read_allocate_pages(rdata, npages); @@ -3621,6 +3631,13 @@ static int cifs_launder_page(struct page *page) return rc; } +static int +cifs_pending_writers_wait(void *unused) +{ + schedule(); + return 0; +} + void cifs_oplock_break(struct work_struct *work) { struct cifsFileInfo *cfile = container_of(work, struct cifsFileInfo, @@ -3628,8 +3645,15 @@ void cifs_oplock_break(struct work_struct *work) struct inode *inode = cfile->dentry->d_inode; struct cifsInodeInfo *cinode = CIFS_I(inode); struct cifs_tcon *tcon = tlink_tcon(cfile->tlink); + struct TCP_Server_Info *server = tcon->ses->server; int rc = 0; + wait_on_bit(&cinode->flags, CIFS_INODE_PENDING_WRITERS, + cifs_pending_writers_wait, TASK_UNINTERRUPTIBLE); + + server->ops->downgrade_oplock(server, cinode, + test_bit(CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2, &cinode->flags)); + if (!CIFS_CACHE_WRITE(cinode) && CIFS_CACHE_READ(cinode) && cifs_has_mand_locks(cinode)) { cifs_dbg(FYI, "Reset oplock to None for inode=%p due to mand locks\n", @@ -3666,6 +3690,7 @@ void cifs_oplock_break(struct work_struct *work) cinode); cifs_dbg(FYI, "Oplock release rc = %d\n", rc); } + cifs_done_oplock_break(cinode); } /* diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 2f9f3790679d..3b0c62e622da 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -466,8 +466,22 @@ is_valid_oplock_break(char *buffer, struct TCP_Server_Info *srv) cifs_dbg(FYI, "file id match, oplock break\n"); pCifsInode = CIFS_I(netfile->dentry->d_inode); - cifs_set_oplock_level(pCifsInode, - pSMB->OplockLevel ? OPLOCK_READ : 0); + set_bit(CIFS_INODE_PENDING_OPLOCK_BREAK, + &pCifsInode->flags); + + /* + * Set flag if the server downgrades the oplock + * to L2 else clear. + */ + if (pSMB->OplockLevel) + set_bit( + CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2, + &pCifsInode->flags); + else + clear_bit( + CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2, + &pCifsInode->flags); + queue_work(cifsiod_wq, &netfile->oplock_break); netfile->oplock_break_cancelled = false; @@ -551,6 +565,62 @@ void cifs_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock) cinode->oplock = 0; } +static int +cifs_oplock_break_wait(void *unused) +{ + schedule(); + return signal_pending(current) ? -ERESTARTSYS : 0; +} + +/* + * We wait for oplock breaks to be processed before we attempt to perform + * writes. + */ +int cifs_get_writer(struct cifsInodeInfo *cinode) +{ + int rc; + +start: + rc = wait_on_bit(&cinode->flags, CIFS_INODE_PENDING_OPLOCK_BREAK, + cifs_oplock_break_wait, TASK_KILLABLE); + if (rc) + return rc; + + spin_lock(&cinode->writers_lock); + if (!cinode->writers) + set_bit(CIFS_INODE_PENDING_WRITERS, &cinode->flags); + cinode->writers++; + /* Check to see if we have started servicing an oplock break */ + if (test_bit(CIFS_INODE_PENDING_OPLOCK_BREAK, &cinode->flags)) { + cinode->writers--; + if (cinode->writers == 0) { + clear_bit(CIFS_INODE_PENDING_WRITERS, &cinode->flags); + wake_up_bit(&cinode->flags, CIFS_INODE_PENDING_WRITERS); + } + spin_unlock(&cinode->writers_lock); + goto start; + } + spin_unlock(&cinode->writers_lock); + return 0; +} + +void cifs_put_writer(struct cifsInodeInfo *cinode) +{ + spin_lock(&cinode->writers_lock); + cinode->writers--; + if (cinode->writers == 0) { + clear_bit(CIFS_INODE_PENDING_WRITERS, &cinode->flags); + wake_up_bit(&cinode->flags, CIFS_INODE_PENDING_WRITERS); + } + spin_unlock(&cinode->writers_lock); +} + +void cifs_done_oplock_break(struct cifsInodeInfo *cinode) +{ + clear_bit(CIFS_INODE_PENDING_OPLOCK_BREAK, &cinode->flags); + wake_up_bit(&cinode->flags, CIFS_INODE_PENDING_OPLOCK_BREAK); +} + bool backup_cred(struct cifs_sb_info *cifs_sb) { diff --git a/fs/cifs/smb1ops.c b/fs/cifs/smb1ops.c index 526fb89f9230..d1fdfa848703 100644 --- a/fs/cifs/smb1ops.c +++ b/fs/cifs/smb1ops.c @@ -372,6 +372,16 @@ coalesce_t2(char *second_buf, struct smb_hdr *target_hdr) return 0; } +static void +cifs_downgrade_oplock(struct TCP_Server_Info *server, + struct cifsInodeInfo *cinode, bool set_level2) +{ + if (set_level2) + cifs_set_oplock_level(cinode, OPLOCK_READ); + else + cifs_set_oplock_level(cinode, 0); +} + static bool cifs_check_trans2(struct mid_q_entry *mid, struct TCP_Server_Info *server, char *buf, int malformed) @@ -1019,6 +1029,7 @@ struct smb_version_operations smb1_operations = { .clear_stats = cifs_clear_stats, .print_stats = cifs_print_stats, .is_oplock_break = is_valid_oplock_break, + .downgrade_oplock = cifs_downgrade_oplock, .check_trans2 = cifs_check_trans2, .need_neg = cifs_need_neg, .negotiate = cifs_negotiate, diff --git a/fs/cifs/smb2misc.c b/fs/cifs/smb2misc.c index fb3966265b6e..b8021fde987d 100644 --- a/fs/cifs/smb2misc.c +++ b/fs/cifs/smb2misc.c @@ -575,9 +575,21 @@ smb2_is_valid_oplock_break(char *buffer, struct TCP_Server_Info *server) else cfile->oplock_break_cancelled = false; - server->ops->set_oplock_level(cinode, - rsp->OplockLevel ? SMB2_OPLOCK_LEVEL_II : 0, - 0, NULL); + set_bit(CIFS_INODE_PENDING_OPLOCK_BREAK, + &cinode->flags); + + /* + * Set flag if the server downgrades the oplock + * to L2 else clear. + */ + if (rsp->OplockLevel) + set_bit( + CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2, + &cinode->flags); + else + clear_bit( + CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2, + &cinode->flags); queue_work(cifsiod_wq, &cfile->oplock_break); diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index 192f51a12cf1..35ddc3ed119d 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -905,6 +905,17 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon, } static void +smb2_downgrade_oplock(struct TCP_Server_Info *server, + struct cifsInodeInfo *cinode, bool set_level2) +{ + if (set_level2) + server->ops->set_oplock_level(cinode, SMB2_OPLOCK_LEVEL_II, + 0, NULL); + else + server->ops->set_oplock_level(cinode, 0, 0, NULL); +} + +static void smb2_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, unsigned int epoch, bool *purge_cache) { @@ -1110,6 +1121,7 @@ struct smb_version_operations smb20_operations = { .clear_stats = smb2_clear_stats, .print_stats = smb2_print_stats, .is_oplock_break = smb2_is_valid_oplock_break, + .downgrade_oplock = smb2_downgrade_oplock, .need_neg = smb2_need_neg, .negotiate = smb2_negotiate, .negotiate_wsize = smb2_negotiate_wsize, @@ -1184,6 +1196,7 @@ struct smb_version_operations smb21_operations = { .clear_stats = smb2_clear_stats, .print_stats = smb2_print_stats, .is_oplock_break = smb2_is_valid_oplock_break, + .downgrade_oplock = smb2_downgrade_oplock, .need_neg = smb2_need_neg, .negotiate = smb2_negotiate, .negotiate_wsize = smb2_negotiate_wsize, @@ -1259,6 +1272,7 @@ struct smb_version_operations smb30_operations = { .print_stats = smb2_print_stats, .dump_share_caps = smb2_dump_share_caps, .is_oplock_break = smb2_is_valid_oplock_break, + .downgrade_oplock = smb2_downgrade_oplock, .need_neg = smb2_need_neg, .negotiate = smb2_negotiate, .negotiate_wsize = smb2_negotiate_wsize, diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index 860344701067..3802f8c94acc 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -1352,7 +1352,6 @@ SMB2_set_compression(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, u64 volatile_fid) { int rc; - char *res_key = NULL; struct compress_ioctl fsctl_input; char *ret_data = NULL; @@ -1365,7 +1364,6 @@ SMB2_set_compression(const unsigned int xid, struct cifs_tcon *tcon, 2 /* in data len */, &ret_data /* out data */, NULL); cifs_dbg(FYI, "set compression rc %d\n", rc); - kfree(res_key); return rc; } diff --git a/fs/compat.c b/fs/compat.c index ca926ad0430c..66d3d3c6b4b2 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -457,9 +457,9 @@ COMPAT_SYSCALL_DEFINE3(fcntl64, unsigned int, fd, unsigned int, cmd, case F_GETLK64: case F_SETLK64: case F_SETLKW64: - case F_GETLKP: - case F_SETLKP: - case F_SETLKPW: + case F_OFD_GETLK: + case F_OFD_SETLK: + case F_OFD_SETLKW: ret = get_compat_flock64(&f, compat_ptr(arg)); if (ret != 0) break; @@ -468,7 +468,7 @@ COMPAT_SYSCALL_DEFINE3(fcntl64, unsigned int, fd, unsigned int, cmd, conv_cmd = convert_fcntl_cmd(cmd); ret = sys_fcntl(fd, conv_cmd, (unsigned long)&f); set_fs(old_fs); - if ((conv_cmd == F_GETLK || conv_cmd == F_GETLKP) && ret == 0) { + if ((conv_cmd == F_GETLK || conv_cmd == F_OFD_GETLK) && ret == 0) { /* need to return lock information - see above for commentary */ if (f.l_start > COMPAT_LOFF_T_MAX) ret = -EOVERFLOW; @@ -493,9 +493,9 @@ COMPAT_SYSCALL_DEFINE3(fcntl, unsigned int, fd, unsigned int, cmd, case F_GETLK64: case F_SETLK64: case F_SETLKW64: - case F_GETLKP: - case F_SETLKP: - case F_SETLKPW: + case F_OFD_GETLK: + case F_OFD_SETLK: + case F_OFD_SETLKW: return -EINVAL; } return compat_sys_fcntl64(fd, cmd, arg); diff --git a/fs/coredump.c b/fs/coredump.c index e3ad709a4232..0b2528fb640e 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -73,10 +73,15 @@ static int expand_corename(struct core_name *cn, int size) static int cn_vprintf(struct core_name *cn, const char *fmt, va_list arg) { int free, need; + va_list arg_copy; again: free = cn->size - cn->used; - need = vsnprintf(cn->corename + cn->used, free, fmt, arg); + + va_copy(arg_copy, arg); + need = vsnprintf(cn->corename + cn->used, free, fmt, arg_copy); + va_end(arg_copy); + if (need < free) { cn->used += need; return 0; diff --git a/fs/dcache.c b/fs/dcache.c index 40707d88a945..42ae01eefc07 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -246,16 +246,8 @@ static void __d_free(struct rcu_head *head) kmem_cache_free(dentry_cache, dentry); } -/* - * no locks, please. - */ -static void d_free(struct dentry *dentry) +static void dentry_free(struct dentry *dentry) { - BUG_ON((int)dentry->d_lockref.count > 0); - this_cpu_dec(nr_dentry); - if (dentry->d_op && dentry->d_op->d_release) - dentry->d_op->d_release(dentry); - /* if dentry was never visible to RCU, immediate free is OK */ if (!(dentry->d_flags & DCACHE_RCUACCESS)) __d_free(&dentry->d_u.d_rcu); @@ -403,56 +395,6 @@ static void dentry_lru_add(struct dentry *dentry) d_lru_add(dentry); } -/* - * Remove a dentry with references from the LRU. - * - * If we are on the shrink list, then we can get to try_prune_one_dentry() and - * lose our last reference through the parent walk. In this case, we need to - * remove ourselves from the shrink list, not the LRU. - */ -static void dentry_lru_del(struct dentry *dentry) -{ - if (dentry->d_flags & DCACHE_LRU_LIST) { - if (dentry->d_flags & DCACHE_SHRINK_LIST) - return d_shrink_del(dentry); - d_lru_del(dentry); - } -} - -/** - * d_kill - kill dentry and return parent - * @dentry: dentry to kill - * @parent: parent dentry - * - * The dentry must already be unhashed and removed from the LRU. - * - * If this is the root of the dentry tree, return NULL. - * - * dentry->d_lock and parent->d_lock must be held by caller, and are dropped by - * d_kill. - */ -static struct dentry *d_kill(struct dentry *dentry, struct dentry *parent) - __releases(dentry->d_lock) - __releases(parent->d_lock) - __releases(dentry->d_inode->i_lock) -{ - list_del(&dentry->d_u.d_child); - /* - * Inform d_walk() that we are no longer attached to the - * dentry tree - */ - dentry->d_flags |= DCACHE_DENTRY_KILLED; - if (parent) - spin_unlock(&parent->d_lock); - dentry_iput(dentry); - /* - * dentry_iput drops the locks, at which point nobody (except - * transient RCU lookups) can reach this dentry. - */ - d_free(dentry); - return parent; -} - /** * d_drop - drop a dentry * @dentry: dentry to drop @@ -510,7 +452,14 @@ dentry_kill(struct dentry *dentry, int unlock_on_failure) __releases(dentry->d_lock) { struct inode *inode; - struct dentry *parent; + struct dentry *parent = NULL; + bool can_free = true; + + if (unlikely(dentry->d_flags & DCACHE_DENTRY_KILLED)) { + can_free = dentry->d_flags & DCACHE_MAY_FREE; + spin_unlock(&dentry->d_lock); + goto out; + } inode = dentry->d_inode; if (inode && !spin_trylock(&inode->i_lock)) { @@ -521,9 +470,7 @@ relock: } return dentry; /* try again with same dentry */ } - if (IS_ROOT(dentry)) - parent = NULL; - else + if (!IS_ROOT(dentry)) parent = dentry->d_parent; if (parent && !spin_trylock(&parent->d_lock)) { if (inode) @@ -543,10 +490,40 @@ relock: if ((dentry->d_flags & DCACHE_OP_PRUNE) && !d_unhashed(dentry)) dentry->d_op->d_prune(dentry); - dentry_lru_del(dentry); + if (dentry->d_flags & DCACHE_LRU_LIST) { + if (!(dentry->d_flags & DCACHE_SHRINK_LIST)) + d_lru_del(dentry); + } /* if it was on the hash then remove it */ __d_drop(dentry); - return d_kill(dentry, parent); + list_del(&dentry->d_u.d_child); + /* + * Inform d_walk() that we are no longer attached to the + * dentry tree + */ + dentry->d_flags |= DCACHE_DENTRY_KILLED; + if (parent) + spin_unlock(&parent->d_lock); + dentry_iput(dentry); + /* + * dentry_iput drops the locks, at which point nobody (except + * transient RCU lookups) can reach this dentry. + */ + BUG_ON((int)dentry->d_lockref.count > 0); + this_cpu_dec(nr_dentry); + if (dentry->d_op && dentry->d_op->d_release) + dentry->d_op->d_release(dentry); + + spin_lock(&dentry->d_lock); + if (dentry->d_flags & DCACHE_SHRINK_LIST) { + dentry->d_flags |= DCACHE_MAY_FREE; + can_free = false; + } + spin_unlock(&dentry->d_lock); +out: + if (likely(can_free)) + dentry_free(dentry); + return parent; } /* @@ -815,65 +792,13 @@ restart: } EXPORT_SYMBOL(d_prune_aliases); -/* - * Try to throw away a dentry - free the inode, dput the parent. - * Requires dentry->d_lock is held, and dentry->d_count == 0. - * Releases dentry->d_lock. - * - * This may fail if locks cannot be acquired no problem, just try again. - */ -static struct dentry * try_prune_one_dentry(struct dentry *dentry) - __releases(dentry->d_lock) -{ - struct dentry *parent; - - parent = dentry_kill(dentry, 0); - /* - * If dentry_kill returns NULL, we have nothing more to do. - * if it returns the same dentry, trylocks failed. In either - * case, just loop again. - * - * Otherwise, we need to prune ancestors too. This is necessary - * to prevent quadratic behavior of shrink_dcache_parent(), but - * is also expected to be beneficial in reducing dentry cache - * fragmentation. - */ - if (!parent) - return NULL; - if (parent == dentry) - return dentry; - - /* Prune ancestors. */ - dentry = parent; - while (dentry) { - if (lockref_put_or_lock(&dentry->d_lockref)) - return NULL; - dentry = dentry_kill(dentry, 1); - } - return NULL; -} - static void shrink_dentry_list(struct list_head *list) { - struct dentry *dentry; + struct dentry *dentry, *parent; - rcu_read_lock(); - for (;;) { - dentry = list_entry_rcu(list->prev, struct dentry, d_lru); - if (&dentry->d_lru == list) - break; /* empty */ - - /* - * Get the dentry lock, and re-verify that the dentry is - * this on the shrinking list. If it is, we know that - * DCACHE_SHRINK_LIST and DCACHE_LRU_LIST are set. - */ + while (!list_empty(list)) { + dentry = list_entry(list->prev, struct dentry, d_lru); spin_lock(&dentry->d_lock); - if (dentry != list_entry(list->prev, struct dentry, d_lru)) { - spin_unlock(&dentry->d_lock); - continue; - } - /* * The dispose list is isolated and dentries are not accounted * to the LRU here, so we can simply remove it from the list @@ -885,30 +810,38 @@ static void shrink_dentry_list(struct list_head *list) * We found an inuse dentry which was not removed from * the LRU because of laziness during lookup. Do not free it. */ - if (dentry->d_lockref.count) { + if ((int)dentry->d_lockref.count > 0) { spin_unlock(&dentry->d_lock); continue; } - rcu_read_unlock(); + parent = dentry_kill(dentry, 0); /* - * If 'try_to_prune()' returns a dentry, it will - * be the same one we passed in, and d_lock will - * have been held the whole time, so it will not - * have been added to any other lists. We failed - * to get the inode lock. - * - * We just add it back to the shrink list. + * If dentry_kill returns NULL, we have nothing more to do. */ - dentry = try_prune_one_dentry(dentry); + if (!parent) + continue; - rcu_read_lock(); - if (dentry) { + if (unlikely(parent == dentry)) { + /* + * trylocks have failed and d_lock has been held the + * whole time, so it could not have been added to any + * other lists. Just add it back to the shrink list. + */ d_shrink_add(dentry, list); spin_unlock(&dentry->d_lock); + continue; } + /* + * We need to prune ancestors too. This is necessary to prevent + * quadratic behavior of shrink_dcache_parent(), but is also + * expected to be beneficial in reducing dentry cache + * fragmentation. + */ + dentry = parent; + while (dentry && !lockref_put_or_lock(&dentry->d_lockref)) + dentry = dentry_kill(dentry, 1); } - rcu_read_unlock(); } static enum lru_status @@ -1261,34 +1194,23 @@ static enum d_walk_ret select_collect(void *_data, struct dentry *dentry) if (data->start == dentry) goto out; - /* - * move only zero ref count dentries to the dispose list. - * - * Those which are presently on the shrink list, being processed - * by shrink_dentry_list(), shouldn't be moved. Otherwise the - * loop in shrink_dcache_parent() might not make any progress - * and loop forever. - */ - if (dentry->d_lockref.count) { - dentry_lru_del(dentry); - } else if (!(dentry->d_flags & DCACHE_SHRINK_LIST)) { - /* - * We can't use d_lru_shrink_move() because we - * need to get the global LRU lock and do the - * LRU accounting. - */ - d_lru_del(dentry); - d_shrink_add(dentry, &data->dispose); + if (dentry->d_flags & DCACHE_SHRINK_LIST) { data->found++; - ret = D_WALK_NORETRY; + } else { + if (dentry->d_flags & DCACHE_LRU_LIST) + d_lru_del(dentry); + if (!dentry->d_lockref.count) { + d_shrink_add(dentry, &data->dispose); + data->found++; + } } /* * We can return to the caller if we have found some (this * ensures forward progress). We'll be coming back to find * the rest. */ - if (data->found && need_resched()) - ret = D_WALK_QUIT; + if (!list_empty(&data->dispose)) + ret = need_resched() ? D_WALK_QUIT : D_WALK_NORETRY; out: return ret; } @@ -1318,45 +1240,35 @@ void shrink_dcache_parent(struct dentry *parent) } EXPORT_SYMBOL(shrink_dcache_parent); -static enum d_walk_ret umount_collect(void *_data, struct dentry *dentry) +static enum d_walk_ret umount_check(void *_data, struct dentry *dentry) { - struct select_data *data = _data; - enum d_walk_ret ret = D_WALK_CONTINUE; + /* it has busy descendents; complain about those instead */ + if (!list_empty(&dentry->d_subdirs)) + return D_WALK_CONTINUE; - if (dentry->d_lockref.count) { - dentry_lru_del(dentry); - if (likely(!list_empty(&dentry->d_subdirs))) - goto out; - if (dentry == data->start && dentry->d_lockref.count == 1) - goto out; - printk(KERN_ERR - "BUG: Dentry %p{i=%lx,n=%s}" - " still in use (%d)" - " [unmount of %s %s]\n", + /* root with refcount 1 is fine */ + if (dentry == _data && dentry->d_lockref.count == 1) + return D_WALK_CONTINUE; + + printk(KERN_ERR "BUG: Dentry %p{i=%lx,n=%pd} " + " still in use (%d) [unmount of %s %s]\n", dentry, dentry->d_inode ? dentry->d_inode->i_ino : 0UL, - dentry->d_name.name, + dentry, dentry->d_lockref.count, dentry->d_sb->s_type->name, dentry->d_sb->s_id); - BUG(); - } else if (!(dentry->d_flags & DCACHE_SHRINK_LIST)) { - /* - * We can't use d_lru_shrink_move() because we - * need to get the global LRU lock and do the - * LRU accounting. - */ - if (dentry->d_flags & DCACHE_LRU_LIST) - d_lru_del(dentry); - d_shrink_add(dentry, &data->dispose); - data->found++; - ret = D_WALK_NORETRY; - } -out: - if (data->found && need_resched()) - ret = D_WALK_QUIT; - return ret; + WARN_ON(1); + return D_WALK_CONTINUE; +} + +static void do_one_tree(struct dentry *dentry) +{ + shrink_dcache_parent(dentry); + d_walk(dentry, dentry, umount_check, NULL); + d_drop(dentry); + dput(dentry); } /* @@ -1366,40 +1278,15 @@ void shrink_dcache_for_umount(struct super_block *sb) { struct dentry *dentry; - if (down_read_trylock(&sb->s_umount)) - BUG(); + WARN(down_read_trylock(&sb->s_umount), "s_umount should've been locked"); dentry = sb->s_root; sb->s_root = NULL; - for (;;) { - struct select_data data; - - INIT_LIST_HEAD(&data.dispose); - data.start = dentry; - data.found = 0; - - d_walk(dentry, &data, umount_collect, NULL); - if (!data.found) - break; - - shrink_dentry_list(&data.dispose); - cond_resched(); - } - d_drop(dentry); - dput(dentry); + do_one_tree(dentry); while (!hlist_bl_empty(&sb->s_anon)) { - struct select_data data; - dentry = hlist_bl_entry(hlist_bl_first(&sb->s_anon), struct dentry, d_hash); - - INIT_LIST_HEAD(&data.dispose); - data.start = NULL; - data.found = 0; - - d_walk(dentry, &data, umount_collect, NULL); - if (data.found) - shrink_dentry_list(&data.dispose); - cond_resched(); + dentry = dget(hlist_bl_entry(hlist_bl_first(&sb->s_anon), struct dentry, d_hash)); + do_one_tree(dentry); } } @@ -1647,8 +1534,7 @@ static void __d_instantiate(struct dentry *dentry, struct inode *inode) unsigned add_flags = d_flags_for_inode(inode); spin_lock(&dentry->d_lock); - dentry->d_flags &= ~DCACHE_ENTRY_TYPE; - dentry->d_flags |= add_flags; + __d_set_type(dentry, add_flags); if (inode) hlist_add_head(&dentry->d_alias, &inode->i_dentry); dentry->d_inode = inode; diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 6ea7b1436bbc..5c56785007e0 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -667,7 +667,7 @@ ext4_fsblk_t ext4_count_free_clusters(struct super_block *sb) continue; x = ext4_count_free(bitmap_bh->b_data, - EXT4_BLOCKS_PER_GROUP(sb) / 8); + EXT4_CLUSTERS_PER_GROUP(sb) / 8); printk(KERN_DEBUG "group %u: stored = %d, counted = %u\n", i, ext4_free_group_clusters(sb, gdp), x); bitmap_count += x; diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index f1c65dc7cc0a..66946aa62127 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -2466,23 +2466,6 @@ static inline void ext4_update_i_disksize(struct inode *inode, loff_t newsize) up_write(&EXT4_I(inode)->i_data_sem); } -/* - * Update i_disksize after writeback has been started. Races with truncate - * are avoided by checking i_size under i_data_sem. - */ -static inline void ext4_wb_update_i_disksize(struct inode *inode, loff_t newsize) -{ - loff_t i_size; - - down_write(&EXT4_I(inode)->i_data_sem); - i_size = i_size_read(inode); - if (newsize > i_size) - newsize = i_size; - if (newsize > EXT4_I(inode)->i_disksize) - EXT4_I(inode)->i_disksize = newsize; - up_write(&EXT4_I(inode)->i_data_sem); -} - struct ext4_group_info { unsigned long bb_state; struct rb_root bb_free_root; diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 82df3ce9874a..01b0c208f625 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -3313,6 +3313,11 @@ static int ext4_split_extent(handle_t *handle, return PTR_ERR(path); depth = ext_depth(inode); ex = path[depth].p_ext; + if (!ex) { + EXT4_ERROR_INODE(inode, "unexpected hole at %lu", + (unsigned long) map->m_lblk); + return -EIO; + } uninitialized = ext4_ext_is_uninitialized(ex); split_flag1 = 0; @@ -3694,6 +3699,12 @@ static int ext4_convert_initialized_extents(handle_t *handle, } depth = ext_depth(inode); ex = path[depth].p_ext; + if (!ex) { + EXT4_ERROR_INODE(inode, "unexpected hole at %lu", + (unsigned long) map->m_lblk); + err = -EIO; + goto out; + } } err = ext4_ext_get_access(handle, inode, path + depth); @@ -4730,6 +4741,9 @@ static long ext4_zero_range(struct file *file, loff_t offset, trace_ext4_zero_range(inode, offset, len, mode); + if (!S_ISREG(inode->i_mode)) + return -EINVAL; + /* * Write out all dirty pages to avoid race conditions * Then release them. @@ -4878,9 +4892,6 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len) if (mode & FALLOC_FL_PUNCH_HOLE) return ext4_punch_hole(inode, offset, len); - if (mode & FALLOC_FL_COLLAPSE_RANGE) - return ext4_collapse_range(inode, offset, len); - ret = ext4_convert_inline_data(inode); if (ret) return ret; @@ -4892,6 +4903,9 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len) if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) return -EOPNOTSUPP; + if (mode & FALLOC_FL_COLLAPSE_RANGE) + return ext4_collapse_range(inode, offset, len); + if (mode & FALLOC_FL_ZERO_RANGE) return ext4_zero_range(file, offset, len, mode); @@ -5229,18 +5243,19 @@ ext4_ext_shift_path_extents(struct ext4_ext_path *path, ext4_lblk_t shift, if (ex_start == EXT_FIRST_EXTENT(path[depth].p_hdr)) update = 1; - *start = ex_last->ee_block + + *start = le32_to_cpu(ex_last->ee_block) + ext4_ext_get_actual_len(ex_last); while (ex_start <= ex_last) { - ex_start->ee_block -= shift; - if (ex_start > - EXT_FIRST_EXTENT(path[depth].p_hdr)) { - if (ext4_ext_try_to_merge_right(inode, - path, ex_start - 1)) - ex_last--; - } - ex_start++; + le32_add_cpu(&ex_start->ee_block, -shift); + /* Try to merge to the left. */ + if ((ex_start > + EXT_FIRST_EXTENT(path[depth].p_hdr)) && + ext4_ext_try_to_merge_right(inode, + path, ex_start - 1)) + ex_last--; + else + ex_start++; } err = ext4_ext_dirty(handle, inode, path + depth); if (err) @@ -5255,7 +5270,7 @@ ext4_ext_shift_path_extents(struct ext4_ext_path *path, ext4_lblk_t shift, if (err) goto out; - path[depth].p_idx->ei_block -= shift; + le32_add_cpu(&path[depth].p_idx->ei_block, -shift); err = ext4_ext_dirty(handle, inode, path + depth); if (err) goto out; @@ -5300,7 +5315,8 @@ ext4_ext_shift_extents(struct inode *inode, handle_t *handle, return ret; } - stop_block = extent->ee_block + ext4_ext_get_actual_len(extent); + stop_block = le32_to_cpu(extent->ee_block) + + ext4_ext_get_actual_len(extent); ext4_ext_drop_refs(path); kfree(path); @@ -5313,10 +5329,18 @@ ext4_ext_shift_extents(struct inode *inode, handle_t *handle, * enough to accomodate the shift. */ path = ext4_ext_find_extent(inode, start - 1, NULL, 0); + if (IS_ERR(path)) + return PTR_ERR(path); depth = path->p_depth; extent = path[depth].p_ext; - ex_start = extent->ee_block; - ex_end = extent->ee_block + ext4_ext_get_actual_len(extent); + if (extent) { + ex_start = le32_to_cpu(extent->ee_block); + ex_end = le32_to_cpu(extent->ee_block) + + ext4_ext_get_actual_len(extent); + } else { + ex_start = 0; + ex_end = 0; + } ext4_ext_drop_refs(path); kfree(path); @@ -5331,7 +5355,13 @@ ext4_ext_shift_extents(struct inode *inode, handle_t *handle, return PTR_ERR(path); depth = path->p_depth; extent = path[depth].p_ext; - current_block = extent->ee_block; + if (!extent) { + EXT4_ERROR_INODE(inode, "unexpected hole at %lu", + (unsigned long) start); + return -EIO; + } + + current_block = le32_to_cpu(extent->ee_block); if (start > current_block) { /* Hole, move to the next extent */ ret = mext_next_extent(inode, path, &extent); @@ -5365,17 +5395,18 @@ int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len) ext4_lblk_t punch_start, punch_stop; handle_t *handle; unsigned int credits; - loff_t new_size; + loff_t new_size, ioffset; int ret; - BUG_ON(offset + len > i_size_read(inode)); - /* Collapse range works only on fs block size aligned offsets. */ if (offset & (EXT4_BLOCK_SIZE(sb) - 1) || len & (EXT4_BLOCK_SIZE(sb) - 1)) return -EINVAL; if (!S_ISREG(inode->i_mode)) + return -EINVAL; + + if (EXT4_SB(inode->i_sb)->s_cluster_ratio > 1) return -EOPNOTSUPP; trace_ext4_collapse_range(inode, offset, len); @@ -5383,22 +5414,34 @@ int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len) punch_start = offset >> EXT4_BLOCK_SIZE_BITS(sb); punch_stop = (offset + len) >> EXT4_BLOCK_SIZE_BITS(sb); + /* Call ext4_force_commit to flush all data in case of data=journal. */ + if (ext4_should_journal_data(inode)) { + ret = ext4_force_commit(inode->i_sb); + if (ret) + return ret; + } + + /* + * Need to round down offset to be aligned with page size boundary + * for page size > block size. + */ + ioffset = round_down(offset, PAGE_SIZE); + /* Write out all dirty pages */ - ret = filemap_write_and_wait_range(inode->i_mapping, offset, -1); + ret = filemap_write_and_wait_range(inode->i_mapping, ioffset, + LLONG_MAX); if (ret) return ret; /* Take mutex lock */ mutex_lock(&inode->i_mutex); - /* It's not possible punch hole on append only file */ - if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) { - ret = -EPERM; - goto out_mutex; - } - - if (IS_SWAPFILE(inode)) { - ret = -ETXTBSY; + /* + * There is no need to overlap collapse range with EOF, in which case + * it is effectively a truncate operation + */ + if (offset + len >= i_size_read(inode)) { + ret = -EINVAL; goto out_mutex; } @@ -5408,7 +5451,7 @@ int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len) goto out_mutex; } - truncate_pagecache_range(inode, offset, -1); + truncate_pagecache(inode, ioffset); /* Wait for existing dio to complete */ ext4_inode_block_unlocked_dio(inode); @@ -5425,7 +5468,7 @@ int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len) ext4_discard_preallocations(inode); ret = ext4_es_remove_extent(inode, punch_start, - EXT_MAX_BLOCKS - punch_start - 1); + EXT_MAX_BLOCKS - punch_start); if (ret) { up_write(&EXT4_I(inode)->i_data_sem); goto out_stop; @@ -5436,6 +5479,7 @@ int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len) up_write(&EXT4_I(inode)->i_data_sem); goto out_stop; } + ext4_discard_preallocations(inode); ret = ext4_ext_shift_extents(inode, handle, punch_stop, punch_stop - punch_start); @@ -5445,10 +5489,9 @@ int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len) } new_size = i_size_read(inode) - len; - truncate_setsize(inode, new_size); + i_size_write(inode, new_size); EXT4_I(inode)->i_disksize = new_size; - ext4_discard_preallocations(inode); up_write(&EXT4_I(inode)->i_data_sem); if (IS_SYNC(inode)) ext4_handle_sync(handle); diff --git a/fs/ext4/extents_status.c b/fs/ext4/extents_status.c index 0a014a7194b2..0ebc21204b51 100644 --- a/fs/ext4/extents_status.c +++ b/fs/ext4/extents_status.c @@ -810,7 +810,7 @@ retry: newes.es_lblk = end + 1; newes.es_len = len2; - block = 0x7FDEADBEEF; + block = 0x7FDEADBEEFULL; if (ext4_es_is_written(&orig_es) || ext4_es_is_unwritten(&orig_es)) block = ext4_es_pblock(&orig_es) + diff --git a/fs/ext4/file.c b/fs/ext4/file.c index ca7502d89fde..063fc1538355 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -82,7 +82,7 @@ ext4_unaligned_aio(struct inode *inode, const struct iovec *iov, size_t count = iov_length(iov, nr_segs); loff_t final_size = pos + count; - if (pos >= inode->i_size) + if (pos >= i_size_read(inode)) return 0; if ((pos & blockmask) || (final_size & blockmask)) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 5b0d2c7d5408..d7b7462a0e13 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -522,6 +522,10 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode, if (unlikely(map->m_len > INT_MAX)) map->m_len = INT_MAX; + /* We can handle the block number less than EXT_MAX_BLOCKS */ + if (unlikely(map->m_lblk >= EXT_MAX_BLOCKS)) + return -EIO; + /* Lookup extent status tree firstly */ if (ext4_es_lookup_extent(inode, map->m_lblk, &es)) { ext4_es_lru_add(inode); @@ -2243,13 +2247,23 @@ static int mpage_map_and_submit_extent(handle_t *handle, return err; } while (map->m_len); - /* Update on-disk size after IO is submitted */ + /* + * Update on-disk size after IO is submitted. Races with + * truncate are avoided by checking i_size under i_data_sem. + */ disksize = ((loff_t)mpd->first_page) << PAGE_CACHE_SHIFT; if (disksize > EXT4_I(inode)->i_disksize) { int err2; - - ext4_wb_update_i_disksize(inode, disksize); + loff_t i_size; + + down_write(&EXT4_I(inode)->i_data_sem); + i_size = i_size_read(inode); + if (disksize > i_size) + disksize = i_size; + if (disksize > EXT4_I(inode)->i_disksize) + EXT4_I(inode)->i_disksize = disksize; err2 = ext4_mark_inode_dirty(handle, inode); + up_write(&EXT4_I(inode)->i_data_sem); if (err2) ext4_error(inode->i_sb, "Failed to mark inode %lu dirty", @@ -3527,15 +3541,6 @@ int ext4_punch_hole(struct inode *inode, loff_t offset, loff_t length) } mutex_lock(&inode->i_mutex); - /* It's not possible punch hole on append only file */ - if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) { - ret = -EPERM; - goto out_mutex; - } - if (IS_SWAPFILE(inode)) { - ret = -ETXTBSY; - goto out_mutex; - } /* No need to punch hole beyond i_size */ if (offset >= inode->i_size) @@ -3616,7 +3621,6 @@ int ext4_punch_hole(struct inode *inode, loff_t offset, loff_t length) ret = ext4_free_hole_blocks(handle, inode, first_block, stop_block); - ext4_discard_preallocations(inode); up_write(&EXT4_I(inode)->i_data_sem); if (IS_SYNC(inode)) ext4_handle_sync(handle); @@ -4423,21 +4427,20 @@ out_brelse: * * We are called from a few places: * - * - Within generic_file_write() for O_SYNC files. + * - Within generic_file_aio_write() -> generic_write_sync() for O_SYNC files. * Here, there will be no transaction running. We wait for any running * transaction to commit. * - * - Within sys_sync(), kupdate and such. - * We wait on commit, if tol to. + * - Within flush work (sys_sync(), kupdate and such). + * We wait on commit, if told to. * - * - Within prune_icache() (PF_MEMALLOC == true) - * Here we simply return. We can't afford to block kswapd on the - * journal commit. + * - Within iput_final() -> write_inode_now() + * We wait on commit, if told to. * * In all cases it is actually safe for us to return without doing anything, * because the inode has been copied into a raw inode buffer in - * ext4_mark_inode_dirty(). This is a correctness thing for O_SYNC and for - * knfsd. + * ext4_mark_inode_dirty(). This is a correctness thing for WB_SYNC_ALL + * writeback. * * Note that we are absolutely dependent upon all inode dirtiers doing the * right thing: they *must* call mark_inode_dirty() after dirtying info in @@ -4449,15 +4452,15 @@ out_brelse: * stuff(); * inode->i_size = expr; * - * is in error because a kswapd-driven write_inode() could occur while - * `stuff()' is running, and the new i_size will be lost. Plus the inode - * will no longer be on the superblock's dirty inode list. + * is in error because write_inode() could occur while `stuff()' is running, + * and the new i_size will be lost. Plus the inode will no longer be on the + * superblock's dirty inode list. */ int ext4_write_inode(struct inode *inode, struct writeback_control *wbc) { int err; - if (current->flags & PF_MEMALLOC) + if (WARN_ON_ONCE(current->flags & PF_MEMALLOC)) return 0; if (EXT4_SB(inode->i_sb)->s_journal) { diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index a888cac76e9c..c8238a26818c 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -989,7 +989,7 @@ static int ext4_mb_get_buddy_page_lock(struct super_block *sb, poff = block % blocks_per_page; page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS); if (!page) - return -EIO; + return -ENOMEM; BUG_ON(page->mapping != inode->i_mapping); e4b->bd_bitmap_page = page; e4b->bd_bitmap = page_address(page) + (poff * sb->s_blocksize); @@ -1003,7 +1003,7 @@ static int ext4_mb_get_buddy_page_lock(struct super_block *sb, pnum = block / blocks_per_page; page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS); if (!page) - return -EIO; + return -ENOMEM; BUG_ON(page->mapping != inode->i_mapping); e4b->bd_buddy_page = page; return 0; @@ -1168,7 +1168,11 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group, unlock_page(page); } } - if (page == NULL || !PageUptodate(page)) { + if (page == NULL) { + ret = -ENOMEM; + goto err; + } + if (!PageUptodate(page)) { ret = -EIO; goto err; } @@ -1197,7 +1201,11 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group, unlock_page(page); } } - if (page == NULL || !PageUptodate(page)) { + if (page == NULL) { + ret = -ENOMEM; + goto err; + } + if (!PageUptodate(page)) { ret = -EIO; goto err; } @@ -5008,6 +5016,8 @@ error_return: */ static int ext4_trim_extent(struct super_block *sb, int start, int count, ext4_group_t group, struct ext4_buddy *e4b) +__releases(bitlock) +__acquires(bitlock) { struct ext4_free_extent ex; int ret = 0; diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index ab95508e3d40..c18d95b50540 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -308,13 +308,14 @@ static void ext4_end_bio(struct bio *bio, int error) if (error) { struct inode *inode = io_end->inode; - ext4_warning(inode->i_sb, "I/O error writing to inode %lu " + ext4_warning(inode->i_sb, "I/O error %d writing to inode %lu " "(offset %llu size %ld starting block %llu)", - inode->i_ino, + error, inode->i_ino, (unsigned long long) io_end->offset, (long) io_end->size, (unsigned long long) bi_sector >> (inode->i_blkbits - 9)); + mapping_set_error(inode->i_mapping, error); } if (io_end->flag & EXT4_IO_END_UNWRITTEN) { diff --git a/fs/ext4/super.c b/fs/ext4/super.c index f3c667091618..6f9e6fadac04 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -3869,19 +3869,38 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) goto failed_mount2; } } + + /* + * set up enough so that it can read an inode, + * and create new inode for buddy allocator + */ + sbi->s_gdb_count = db_count; + if (!test_opt(sb, NOLOAD) && + EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL)) + sb->s_op = &ext4_sops; + else + sb->s_op = &ext4_nojournal_sops; + + ext4_ext_init(sb); + err = ext4_mb_init(sb); + if (err) { + ext4_msg(sb, KERN_ERR, "failed to initialize mballoc (%d)", + err); + goto failed_mount2; + } + if (!ext4_check_descriptors(sb, &first_not_zeroed)) { ext4_msg(sb, KERN_ERR, "group descriptors corrupted!"); - goto failed_mount2; + goto failed_mount2a; } if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG)) if (!ext4_fill_flex_info(sb)) { ext4_msg(sb, KERN_ERR, "unable to initialize " "flex_bg meta info!"); - goto failed_mount2; + goto failed_mount2a; } - sbi->s_gdb_count = db_count; get_random_bytes(&sbi->s_next_generation, sizeof(u32)); spin_lock_init(&sbi->s_next_gen_lock); @@ -3916,14 +3935,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) sbi->s_stripe = ext4_get_stripe_size(sbi); sbi->s_extent_max_zeroout_kb = 32; - /* - * set up enough so that it can read an inode - */ - if (!test_opt(sb, NOLOAD) && - EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL)) - sb->s_op = &ext4_sops; - else - sb->s_op = &ext4_nojournal_sops; sb->s_export_op = &ext4_export_ops; sb->s_xattr = ext4_xattr_handlers; #ifdef CONFIG_QUOTA @@ -4113,21 +4124,13 @@ no_journal: if (err) { ext4_msg(sb, KERN_ERR, "failed to reserve %llu clusters for " "reserved pool", ext4_calculate_resv_clusters(sb)); - goto failed_mount4a; + goto failed_mount5; } err = ext4_setup_system_zone(sb); if (err) { ext4_msg(sb, KERN_ERR, "failed to initialize system " "zone (%d)", err); - goto failed_mount4a; - } - - ext4_ext_init(sb); - err = ext4_mb_init(sb); - if (err) { - ext4_msg(sb, KERN_ERR, "failed to initialize mballoc (%d)", - err); goto failed_mount5; } @@ -4204,11 +4207,8 @@ failed_mount8: failed_mount7: ext4_unregister_li_request(sb); failed_mount6: - ext4_mb_release(sb); -failed_mount5: - ext4_ext_release(sb); ext4_release_system_zone(sb); -failed_mount4a: +failed_mount5: dput(sb->s_root); sb->s_root = NULL; failed_mount4: @@ -4232,11 +4232,14 @@ failed_mount3: percpu_counter_destroy(&sbi->s_extent_cache_cnt); if (sbi->s_mmp_tsk) kthread_stop(sbi->s_mmp_tsk); +failed_mount2a: + ext4_mb_release(sb); failed_mount2: for (i = 0; i < db_count; i++) brelse(sbi->s_group_desc[i]); ext4_kvfree(sbi->s_group_desc); failed_mount: + ext4_ext_release(sb); if (sbi->s_chksum_driver) crypto_free_shash(sbi->s_chksum_driver); if (sbi->s_proc) { diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 1f5cf5880718..4eec399ec807 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -520,8 +520,8 @@ static void ext4_xattr_update_super_block(handle_t *handle, } /* - * Release the xattr block BH: If the reference count is > 1, decrement - * it; otherwise free the block. + * Release the xattr block BH: If the reference count is > 1, decrement it; + * otherwise free the block. */ static void ext4_xattr_release_block(handle_t *handle, struct inode *inode, @@ -542,16 +542,31 @@ ext4_xattr_release_block(handle_t *handle, struct inode *inode, if (ce) mb_cache_entry_free(ce); get_bh(bh); + unlock_buffer(bh); ext4_free_blocks(handle, inode, bh, 0, 1, EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET); - unlock_buffer(bh); } else { le32_add_cpu(&BHDR(bh)->h_refcount, -1); if (ce) mb_cache_entry_release(ce); + /* + * Beware of this ugliness: Releasing of xattr block references + * from different inodes can race and so we have to protect + * from a race where someone else frees the block (and releases + * its journal_head) before we are done dirtying the buffer. In + * nojournal mode this race is harmless and we actually cannot + * call ext4_handle_dirty_xattr_block() with locked buffer as + * that function can call sync_dirty_buffer() so for that case + * we handle the dirtying after unlocking the buffer. + */ + if (ext4_handle_valid(handle)) + error = ext4_handle_dirty_xattr_block(handle, inode, + bh); unlock_buffer(bh); - error = ext4_handle_dirty_xattr_block(handle, inode, bh); + if (!ext4_handle_valid(handle)) + error = ext4_handle_dirty_xattr_block(handle, inode, + bh); if (IS_SYNC(inode)) ext4_handle_sync(handle); dquot_free_block(inode, EXT4_C2B(EXT4_SB(inode->i_sb), 1)); diff --git a/fs/fcntl.c b/fs/fcntl.c index 9ead1596399a..72c82f69b01b 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -274,15 +274,15 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg, break; #if BITS_PER_LONG != 32 /* 32-bit arches must use fcntl64() */ - case F_GETLKP: + case F_OFD_GETLK: #endif case F_GETLK: err = fcntl_getlk(filp, cmd, (struct flock __user *) arg); break; #if BITS_PER_LONG != 32 /* 32-bit arches must use fcntl64() */ - case F_SETLKP: - case F_SETLKPW: + case F_OFD_SETLK: + case F_OFD_SETLKW: #endif /* Fallthrough */ case F_SETLK: @@ -399,13 +399,13 @@ SYSCALL_DEFINE3(fcntl64, unsigned int, fd, unsigned int, cmd, switch (cmd) { case F_GETLK64: - case F_GETLKP: + case F_OFD_GETLK: err = fcntl_getlk64(f.file, cmd, (struct flock64 __user *) arg); break; case F_SETLK64: case F_SETLKW64: - case F_SETLKP: - case F_SETLKPW: + case F_OFD_SETLK: + case F_OFD_SETLKW: err = fcntl_setlk64(fd, f.file, cmd, (struct flock64 __user *) arg); break; diff --git a/fs/fuse/control.c b/fs/fuse/control.c index a0b0855d00a9..205e0d5d5307 100644 --- a/fs/fuse/control.c +++ b/fs/fuse/control.c @@ -348,7 +348,7 @@ int __init fuse_ctl_init(void) return register_filesystem(&fuse_ctl_fs_type); } -void fuse_ctl_cleanup(void) +void __exit fuse_ctl_cleanup(void) { unregister_filesystem(&fuse_ctl_fs_type); } diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 5b4e035b364c..42198359fa1b 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -679,6 +679,14 @@ static int fuse_symlink(struct inode *dir, struct dentry *entry, return create_new_entry(fc, req, dir, entry, S_IFLNK); } +static inline void fuse_update_ctime(struct inode *inode) +{ + if (!IS_NOCMTIME(inode)) { + inode->i_ctime = current_fs_time(inode->i_sb); + mark_inode_dirty_sync(inode); + } +} + static int fuse_unlink(struct inode *dir, struct dentry *entry) { int err; @@ -713,6 +721,7 @@ static int fuse_unlink(struct inode *dir, struct dentry *entry) fuse_invalidate_attr(inode); fuse_invalidate_attr(dir); fuse_invalidate_entry_cache(entry); + fuse_update_ctime(inode); } else if (err == -EINTR) fuse_invalidate_entry(entry); return err; @@ -743,23 +752,26 @@ static int fuse_rmdir(struct inode *dir, struct dentry *entry) return err; } -static int fuse_rename(struct inode *olddir, struct dentry *oldent, - struct inode *newdir, struct dentry *newent) +static int fuse_rename_common(struct inode *olddir, struct dentry *oldent, + struct inode *newdir, struct dentry *newent, + unsigned int flags, int opcode, size_t argsize) { int err; - struct fuse_rename_in inarg; + struct fuse_rename2_in inarg; struct fuse_conn *fc = get_fuse_conn(olddir); - struct fuse_req *req = fuse_get_req_nopages(fc); + struct fuse_req *req; + req = fuse_get_req_nopages(fc); if (IS_ERR(req)) return PTR_ERR(req); - memset(&inarg, 0, sizeof(inarg)); + memset(&inarg, 0, argsize); inarg.newdir = get_node_id(newdir); - req->in.h.opcode = FUSE_RENAME; + inarg.flags = flags; + req->in.h.opcode = opcode; req->in.h.nodeid = get_node_id(olddir); req->in.numargs = 3; - req->in.args[0].size = sizeof(inarg); + req->in.args[0].size = argsize; req->in.args[0].value = &inarg; req->in.args[1].size = oldent->d_name.len + 1; req->in.args[1].value = oldent->d_name.name; @@ -771,15 +783,22 @@ static int fuse_rename(struct inode *olddir, struct dentry *oldent, if (!err) { /* ctime changes */ fuse_invalidate_attr(oldent->d_inode); + fuse_update_ctime(oldent->d_inode); + + if (flags & RENAME_EXCHANGE) { + fuse_invalidate_attr(newent->d_inode); + fuse_update_ctime(newent->d_inode); + } fuse_invalidate_attr(olddir); if (olddir != newdir) fuse_invalidate_attr(newdir); /* newent will end up negative */ - if (newent->d_inode) { + if (!(flags & RENAME_EXCHANGE) && newent->d_inode) { fuse_invalidate_attr(newent->d_inode); fuse_invalidate_entry_cache(newent); + fuse_update_ctime(newent->d_inode); } } else if (err == -EINTR) { /* If request was interrupted, DEITY only knows if the @@ -795,6 +814,36 @@ static int fuse_rename(struct inode *olddir, struct dentry *oldent, return err; } +static int fuse_rename(struct inode *olddir, struct dentry *oldent, + struct inode *newdir, struct dentry *newent) +{ + return fuse_rename_common(olddir, oldent, newdir, newent, 0, + FUSE_RENAME, sizeof(struct fuse_rename_in)); +} + +static int fuse_rename2(struct inode *olddir, struct dentry *oldent, + struct inode *newdir, struct dentry *newent, + unsigned int flags) +{ + struct fuse_conn *fc = get_fuse_conn(olddir); + int err; + + if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE)) + return -EINVAL; + + if (fc->no_rename2 || fc->minor < 23) + return -EINVAL; + + err = fuse_rename_common(olddir, oldent, newdir, newent, flags, + FUSE_RENAME2, sizeof(struct fuse_rename2_in)); + if (err == -ENOSYS) { + fc->no_rename2 = 1; + err = -EINVAL; + } + return err; + +} + static int fuse_link(struct dentry *entry, struct inode *newdir, struct dentry *newent) { @@ -829,6 +878,7 @@ static int fuse_link(struct dentry *entry, struct inode *newdir, inc_nlink(inode); spin_unlock(&fc->lock); fuse_invalidate_attr(inode); + fuse_update_ctime(inode); } else if (err == -EINTR) { fuse_invalidate_attr(inode); } @@ -846,6 +896,8 @@ static void fuse_fillattr(struct inode *inode, struct fuse_attr *attr, attr->size = i_size_read(inode); attr->mtime = inode->i_mtime.tv_sec; attr->mtimensec = inode->i_mtime.tv_nsec; + attr->ctime = inode->i_ctime.tv_sec; + attr->ctimensec = inode->i_ctime.tv_nsec; } stat->dev = inode->i_sb->s_dev; @@ -1504,7 +1556,7 @@ static bool update_mtime(unsigned ivalid, bool trust_local_mtime) } static void iattr_to_fattr(struct iattr *iattr, struct fuse_setattr_in *arg, - bool trust_local_mtime) + bool trust_local_cmtime) { unsigned ivalid = iattr->ia_valid; @@ -1523,13 +1575,18 @@ static void iattr_to_fattr(struct iattr *iattr, struct fuse_setattr_in *arg, if (!(ivalid & ATTR_ATIME_SET)) arg->valid |= FATTR_ATIME_NOW; } - if ((ivalid & ATTR_MTIME) && update_mtime(ivalid, trust_local_mtime)) { + if ((ivalid & ATTR_MTIME) && update_mtime(ivalid, trust_local_cmtime)) { arg->valid |= FATTR_MTIME; arg->mtime = iattr->ia_mtime.tv_sec; arg->mtimensec = iattr->ia_mtime.tv_nsec; - if (!(ivalid & ATTR_MTIME_SET) && !trust_local_mtime) + if (!(ivalid & ATTR_MTIME_SET) && !trust_local_cmtime) arg->valid |= FATTR_MTIME_NOW; } + if ((ivalid & ATTR_CTIME) && trust_local_cmtime) { + arg->valid |= FATTR_CTIME; + arg->ctime = iattr->ia_ctime.tv_sec; + arg->ctimensec = iattr->ia_ctime.tv_nsec; + } } /* @@ -1597,39 +1654,38 @@ static void fuse_setattr_fill(struct fuse_conn *fc, struct fuse_req *req, /* * Flush inode->i_mtime to the server */ -int fuse_flush_mtime(struct file *file, bool nofail) +int fuse_flush_times(struct inode *inode, struct fuse_file *ff) { - struct inode *inode = file->f_mapping->host; - struct fuse_inode *fi = get_fuse_inode(inode); struct fuse_conn *fc = get_fuse_conn(inode); - struct fuse_req *req = NULL; + struct fuse_req *req; struct fuse_setattr_in inarg; struct fuse_attr_out outarg; int err; - if (nofail) { - req = fuse_get_req_nofail_nopages(fc, file); - } else { - req = fuse_get_req_nopages(fc); - if (IS_ERR(req)) - return PTR_ERR(req); - } + req = fuse_get_req_nopages(fc); + if (IS_ERR(req)) + return PTR_ERR(req); memset(&inarg, 0, sizeof(inarg)); memset(&outarg, 0, sizeof(outarg)); - inarg.valid |= FATTR_MTIME; + inarg.valid = FATTR_MTIME; inarg.mtime = inode->i_mtime.tv_sec; inarg.mtimensec = inode->i_mtime.tv_nsec; - + if (fc->minor >= 23) { + inarg.valid |= FATTR_CTIME; + inarg.ctime = inode->i_ctime.tv_sec; + inarg.ctimensec = inode->i_ctime.tv_nsec; + } + if (ff) { + inarg.valid |= FATTR_FH; + inarg.fh = ff->fh; + } fuse_setattr_fill(fc, req, inode, &inarg, &outarg); fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); - if (!err) - clear_bit(FUSE_I_MTIME_DIRTY, &fi->state); - return err; } @@ -1653,7 +1709,7 @@ int fuse_do_setattr(struct inode *inode, struct iattr *attr, bool is_wb = fc->writeback_cache; loff_t oldsize; int err; - bool trust_local_mtime = is_wb && S_ISREG(inode->i_mode); + bool trust_local_cmtime = is_wb && S_ISREG(inode->i_mode); if (!(fc->flags & FUSE_DEFAULT_PERMISSIONS)) attr->ia_valid |= ATTR_FORCE; @@ -1678,11 +1734,13 @@ int fuse_do_setattr(struct inode *inode, struct iattr *attr, if (is_truncate) { fuse_set_nowrite(inode); set_bit(FUSE_I_SIZE_UNSTABLE, &fi->state); + if (trust_local_cmtime && attr->ia_size != inode->i_size) + attr->ia_valid |= ATTR_MTIME | ATTR_CTIME; } memset(&inarg, 0, sizeof(inarg)); memset(&outarg, 0, sizeof(outarg)); - iattr_to_fattr(attr, &inarg, trust_local_mtime); + iattr_to_fattr(attr, &inarg, trust_local_cmtime); if (file) { struct fuse_file *ff = file->private_data; inarg.valid |= FATTR_FH; @@ -1711,9 +1769,12 @@ int fuse_do_setattr(struct inode *inode, struct iattr *attr, spin_lock(&fc->lock); /* the kernel maintains i_mtime locally */ - if (trust_local_mtime && (attr->ia_valid & ATTR_MTIME)) { - inode->i_mtime = attr->ia_mtime; - clear_bit(FUSE_I_MTIME_DIRTY, &fi->state); + if (trust_local_cmtime) { + if (attr->ia_valid & ATTR_MTIME) + inode->i_mtime = attr->ia_mtime; + if (attr->ia_valid & ATTR_CTIME) + inode->i_ctime = attr->ia_ctime; + /* FIXME: clear I_DIRTY_SYNC? */ } fuse_change_attributes_common(inode, &outarg.attr, @@ -1810,8 +1871,10 @@ static int fuse_setxattr(struct dentry *entry, const char *name, fc->no_setxattr = 1; err = -EOPNOTSUPP; } - if (!err) + if (!err) { fuse_invalidate_attr(inode); + fuse_update_ctime(inode); + } return err; } @@ -1941,20 +2004,11 @@ static int fuse_removexattr(struct dentry *entry, const char *name) fc->no_removexattr = 1; err = -EOPNOTSUPP; } - if (!err) + if (!err) { fuse_invalidate_attr(inode); - return err; -} - -static int fuse_update_time(struct inode *inode, struct timespec *now, - int flags) -{ - if (flags & S_MTIME) { - inode->i_mtime = *now; - set_bit(FUSE_I_MTIME_DIRTY, &get_fuse_inode(inode)->state); - BUG_ON(!S_ISREG(inode->i_mode)); + fuse_update_ctime(inode); } - return 0; + return err; } static const struct inode_operations fuse_dir_inode_operations = { @@ -1964,6 +2018,7 @@ static const struct inode_operations fuse_dir_inode_operations = { .unlink = fuse_unlink, .rmdir = fuse_rmdir, .rename = fuse_rename, + .rename2 = fuse_rename2, .link = fuse_link, .setattr = fuse_setattr, .create = fuse_create, @@ -1996,7 +2051,6 @@ static const struct inode_operations fuse_common_inode_operations = { .getxattr = fuse_getxattr, .listxattr = fuse_listxattr, .removexattr = fuse_removexattr, - .update_time = fuse_update_time, }; static const struct inode_operations fuse_symlink_inode_operations = { diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 13f8bdec5110..96d513e01a5d 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -223,6 +223,8 @@ void fuse_finish_open(struct inode *inode, struct file *file) i_size_write(inode, 0); spin_unlock(&fc->lock); fuse_invalidate_attr(inode); + if (fc->writeback_cache) + file_update_time(file); } if ((file->f_mode & FMODE_WRITE) && fc->writeback_cache) fuse_link_write_file(file); @@ -232,18 +234,26 @@ int fuse_open_common(struct inode *inode, struct file *file, bool isdir) { struct fuse_conn *fc = get_fuse_conn(inode); int err; + bool lock_inode = (file->f_flags & O_TRUNC) && + fc->atomic_o_trunc && + fc->writeback_cache; err = generic_file_open(inode, file); if (err) return err; + if (lock_inode) + mutex_lock(&inode->i_mutex); + err = fuse_do_open(fc, get_node_id(inode), file, isdir); - if (err) - return err; - fuse_finish_open(inode, file); + if (!err) + fuse_finish_open(inode, file); - return 0; + if (lock_inode) + mutex_unlock(&inode->i_mutex); + + return err; } static void fuse_prepare_release(struct fuse_file *ff, int flags, int opcode) @@ -314,10 +324,7 @@ static int fuse_release(struct inode *inode, struct file *file) /* see fuse_vma_close() for !writeback_cache case */ if (fc->writeback_cache) - filemap_write_and_wait(file->f_mapping); - - if (test_bit(FUSE_I_MTIME_DIRTY, &get_fuse_inode(inode)->state)) - fuse_flush_mtime(file, true); + write_inode_now(inode, 1); fuse_release_common(file, FUSE_RELEASE); @@ -439,7 +446,7 @@ static int fuse_flush(struct file *file, fl_owner_t id) if (fc->no_flush) return 0; - err = filemap_write_and_wait(file->f_mapping); + err = write_inode_now(inode, 1); if (err) return err; @@ -480,13 +487,6 @@ int fuse_fsync_common(struct file *file, loff_t start, loff_t end, if (is_bad_inode(inode)) return -EIO; - err = filemap_write_and_wait_range(inode->i_mapping, start, end); - if (err) - return err; - - if ((!isdir && fc->no_fsync) || (isdir && fc->no_fsyncdir)) - return 0; - mutex_lock(&inode->i_mutex); /* @@ -494,17 +494,17 @@ int fuse_fsync_common(struct file *file, loff_t start, loff_t end, * wait for all outstanding writes, before sending the FSYNC * request. */ - err = write_inode_now(inode, 0); + err = filemap_write_and_wait_range(inode->i_mapping, start, end); if (err) goto out; fuse_sync_writes(inode); + err = sync_inode_metadata(inode, 1); + if (err) + goto out; - if (test_bit(FUSE_I_MTIME_DIRTY, &get_fuse_inode(inode)->state)) { - int err = fuse_flush_mtime(file, false); - if (err) - goto out; - } + if ((!isdir && fc->no_fsync) || (isdir && fc->no_fsyncdir)) + goto out; req = fuse_get_req_nopages(fc); if (IS_ERR(req)) { @@ -1659,13 +1659,13 @@ static void fuse_writepage_end(struct fuse_conn *fc, struct fuse_req *req) fuse_writepage_free(fc, req); } -static struct fuse_file *fuse_write_file_get(struct fuse_conn *fc, - struct fuse_inode *fi) +static struct fuse_file *__fuse_write_file_get(struct fuse_conn *fc, + struct fuse_inode *fi) { struct fuse_file *ff = NULL; spin_lock(&fc->lock); - if (!WARN_ON(list_empty(&fi->write_files))) { + if (!list_empty(&fi->write_files)) { ff = list_entry(fi->write_files.next, struct fuse_file, write_entry); fuse_file_get(ff); @@ -1675,6 +1675,29 @@ static struct fuse_file *fuse_write_file_get(struct fuse_conn *fc, return ff; } +static struct fuse_file *fuse_write_file_get(struct fuse_conn *fc, + struct fuse_inode *fi) +{ + struct fuse_file *ff = __fuse_write_file_get(fc, fi); + WARN_ON(!ff); + return ff; +} + +int fuse_write_inode(struct inode *inode, struct writeback_control *wbc) +{ + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_inode *fi = get_fuse_inode(inode); + struct fuse_file *ff; + int err; + + ff = __fuse_write_file_get(fc, fi); + err = fuse_flush_times(inode, ff); + if (ff) + fuse_file_put(ff, 0); + + return err; +} + static int fuse_writepage_locked(struct page *page) { struct address_space *mapping = page->mapping; @@ -2972,6 +2995,9 @@ static long fuse_file_fallocate(struct file *file, int mode, loff_t offset, bool lock_inode = !(mode & FALLOC_FL_KEEP_SIZE) || (mode & FALLOC_FL_PUNCH_HOLE); + if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE)) + return -EOPNOTSUPP; + if (fc->no_fallocate) return -EOPNOTSUPP; @@ -3017,12 +3043,8 @@ static long fuse_file_fallocate(struct file *file, int mode, loff_t offset, if (!(mode & FALLOC_FL_KEEP_SIZE)) { bool changed = fuse_write_update_size(inode, offset + length); - if (changed && fc->writeback_cache) { - struct fuse_inode *fi = get_fuse_inode(inode); - - inode->i_mtime = current_fs_time(inode->i_sb); - set_bit(FUSE_I_MTIME_DIRTY, &fi->state); - } + if (changed && fc->writeback_cache) + file_update_time(file); } if (mode & FALLOC_FL_PUNCH_HOLE) diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index a257ed8ebee6..7aa5c75e0de1 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -119,8 +119,6 @@ enum { FUSE_I_INIT_RDPLUS, /** An operation changing file size is in progress */ FUSE_I_SIZE_UNSTABLE, - /** i_mtime has been updated locally; a flush to userspace needed */ - FUSE_I_MTIME_DIRTY, }; struct fuse_conn; @@ -544,6 +542,9 @@ struct fuse_conn { /** Is fallocate not implemented by fs? */ unsigned no_fallocate:1; + /** Is rename with flags implemented by fs? */ + unsigned no_rename2:1; + /** Use enhanced/automatic page cache invalidation. */ unsigned auto_inval_data:1; @@ -725,7 +726,7 @@ int fuse_dev_init(void); void fuse_dev_cleanup(void); int fuse_ctl_init(void); -void fuse_ctl_cleanup(void); +void __exit fuse_ctl_cleanup(void); /** * Allocate a request @@ -891,7 +892,8 @@ int fuse_dev_release(struct inode *inode, struct file *file); bool fuse_write_update_size(struct inode *inode, loff_t pos); -int fuse_flush_mtime(struct file *file, bool nofail); +int fuse_flush_times(struct inode *inode, struct fuse_file *ff); +int fuse_write_inode(struct inode *inode, struct writeback_control *wbc); int fuse_do_setattr(struct inode *inode, struct iattr *attr, struct file *file); diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 8d611696fcad..754dcf23de8a 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -175,9 +175,9 @@ void fuse_change_attributes_common(struct inode *inode, struct fuse_attr *attr, if (!fc->writeback_cache || !S_ISREG(inode->i_mode)) { inode->i_mtime.tv_sec = attr->mtime; inode->i_mtime.tv_nsec = attr->mtimensec; + inode->i_ctime.tv_sec = attr->ctime; + inode->i_ctime.tv_nsec = attr->ctimensec; } - inode->i_ctime.tv_sec = attr->ctime; - inode->i_ctime.tv_nsec = attr->ctimensec; if (attr->blksize != 0) inode->i_blkbits = ilog2(attr->blksize); @@ -256,6 +256,8 @@ static void fuse_init_inode(struct inode *inode, struct fuse_attr *attr) inode->i_size = attr->size; inode->i_mtime.tv_sec = attr->mtime; inode->i_mtime.tv_nsec = attr->mtimensec; + inode->i_ctime.tv_sec = attr->ctime; + inode->i_ctime.tv_nsec = attr->ctimensec; if (S_ISREG(inode->i_mode)) { fuse_init_common(inode); fuse_init_file_inode(inode); @@ -303,7 +305,7 @@ struct inode *fuse_iget(struct super_block *sb, u64 nodeid, if ((inode->i_state & I_NEW)) { inode->i_flags |= S_NOATIME; - if (!fc->writeback_cache || !S_ISREG(inode->i_mode)) + if (!fc->writeback_cache || !S_ISREG(attr->mode)) inode->i_flags |= S_NOCMTIME; inode->i_generation = generation; inode->i_data.backing_dev_info = &fc->bdi; @@ -788,6 +790,7 @@ static const struct super_operations fuse_super_operations = { .alloc_inode = fuse_alloc_inode, .destroy_inode = fuse_destroy_inode, .evict_inode = fuse_evict_inode, + .write_inode = fuse_write_inode, .drop_inode = generic_delete_inode, .remount_fs = fuse_remount_fs, .put_super = fuse_put_super, @@ -890,6 +893,11 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_req *req) fc->async_dio = 1; if (arg->flags & FUSE_WRITEBACK_CACHE) fc->writeback_cache = 1; + if (arg->time_gran && arg->time_gran <= 1000000000) + fc->sb->s_time_gran = arg->time_gran; + else + fc->sb->s_time_gran = 1000000000; + } else { ra_pages = fc->max_read / PAGE_CACHE_SIZE; fc->no_lock = 1; @@ -996,7 +1004,7 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) if (sb->s_flags & MS_MANDLOCK) goto err; - sb->s_flags &= ~MS_NOSEC; + sb->s_flags &= ~(MS_NOSEC | MS_I_VERSION); if (!parse_fuse_opt((char *) data, &d, is_bdev)) goto err; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 204027520937..e19d4c0cacae 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -1030,6 +1030,11 @@ static int __init init_hugetlbfs_fs(void) int error; int i; + if (!hugepages_supported()) { + pr_info("hugetlbfs: disabling because there are no supported hugepage sizes\n"); + return -ENOTSUPP; + } + error = bdi_init(&hugetlbfs_backing_dev_info); if (error) return error; diff --git a/fs/kernfs/dir.c b/fs/kernfs/dir.c index 78f3403300af..ac127cd008bf 100644 --- a/fs/kernfs/dir.c +++ b/fs/kernfs/dir.c @@ -232,9 +232,6 @@ static int kernfs_link_sibling(struct kernfs_node *kn) struct rb_node **node = &kn->parent->dir.children.rb_node; struct rb_node *parent = NULL; - if (kernfs_type(kn) == KERNFS_DIR) - kn->parent->dir.subdirs++; - while (*node) { struct kernfs_node *pos; int result; @@ -249,9 +246,15 @@ static int kernfs_link_sibling(struct kernfs_node *kn) else return -EEXIST; } + /* add new node and rebalance the tree */ rb_link_node(&kn->rb, parent, node); rb_insert_color(&kn->rb, &kn->parent->dir.children); + + /* successfully added, account subdir number */ + if (kernfs_type(kn) == KERNFS_DIR) + kn->parent->dir.subdirs++; + return 0; } diff --git a/fs/kernfs/file.c b/fs/kernfs/file.c index 8034706a7af8..e01ea4a14a01 100644 --- a/fs/kernfs/file.c +++ b/fs/kernfs/file.c @@ -484,6 +484,8 @@ static int kernfs_fop_mmap(struct file *file, struct vm_area_struct *vma) ops = kernfs_ops(of->kn); rc = ops->mmap(of, vma); + if (rc) + goto out_put; /* * PowerPC's pci_mmap of legacy_mem uses shmem_zero_setup() diff --git a/fs/kernfs/inode.c b/fs/kernfs/inode.c index abb0f1f53d93..985217626e66 100644 --- a/fs/kernfs/inode.c +++ b/fs/kernfs/inode.c @@ -48,14 +48,18 @@ void __init kernfs_inode_init(void) static struct kernfs_iattrs *kernfs_iattrs(struct kernfs_node *kn) { + static DEFINE_MUTEX(iattr_mutex); + struct kernfs_iattrs *ret; struct iattr *iattrs; + mutex_lock(&iattr_mutex); + if (kn->iattr) - return kn->iattr; + goto out_unlock; kn->iattr = kzalloc(sizeof(struct kernfs_iattrs), GFP_KERNEL); if (!kn->iattr) - return NULL; + goto out_unlock; iattrs = &kn->iattr->ia_iattr; /* assign default attributes */ @@ -65,8 +69,10 @@ static struct kernfs_iattrs *kernfs_iattrs(struct kernfs_node *kn) iattrs->ia_atime = iattrs->ia_mtime = iattrs->ia_ctime = CURRENT_TIME; simple_xattrs_init(&kn->iattr->xattrs); - - return kn->iattr; +out_unlock: + ret = kn->iattr; + mutex_unlock(&iattr_mutex); + return ret; } static int __kernfs_setattr(struct kernfs_node *kn, const struct iattr *iattr) diff --git a/fs/locks.c b/fs/locks.c index 13fc7a6d380a..e663aeac579e 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -135,7 +135,7 @@ #define IS_POSIX(fl) (fl->fl_flags & FL_POSIX) #define IS_FLOCK(fl) (fl->fl_flags & FL_FLOCK) #define IS_LEASE(fl) (fl->fl_flags & (FL_LEASE|FL_DELEG)) -#define IS_FILE_PVT(fl) (fl->fl_flags & FL_FILE_PVT) +#define IS_OFDLCK(fl) (fl->fl_flags & FL_OFDLCK) static bool lease_breaking(struct file_lock *fl) { @@ -564,7 +564,7 @@ static void __locks_insert_block(struct file_lock *blocker, BUG_ON(!list_empty(&waiter->fl_block)); waiter->fl_next = blocker; list_add_tail(&waiter->fl_block, &blocker->fl_block); - if (IS_POSIX(blocker) && !IS_FILE_PVT(blocker)) + if (IS_POSIX(blocker) && !IS_OFDLCK(blocker)) locks_insert_global_blocked(waiter); } @@ -759,12 +759,12 @@ EXPORT_SYMBOL(posix_test_lock); * of tasks (such as posix threads) sharing the same open file table. * To handle those cases, we just bail out after a few iterations. * - * For FL_FILE_PVT locks, the owner is the filp, not the files_struct. + * For FL_OFDLCK locks, the owner is the filp, not the files_struct. * Because the owner is not even nominally tied to a thread of * execution, the deadlock detection below can't reasonably work well. Just * skip it for those. * - * In principle, we could do a more limited deadlock detection on FL_FILE_PVT + * In principle, we could do a more limited deadlock detection on FL_OFDLCK * locks that just checks for the case where two tasks are attempting to * upgrade from read to write locks on the same inode. */ @@ -791,9 +791,9 @@ static int posix_locks_deadlock(struct file_lock *caller_fl, /* * This deadlock detector can't reasonably detect deadlocks with - * FL_FILE_PVT locks, since they aren't owned by a process, per-se. + * FL_OFDLCK locks, since they aren't owned by a process, per-se. */ - if (IS_FILE_PVT(caller_fl)) + if (IS_OFDLCK(caller_fl)) return 0; while ((block_fl = what_owner_is_waiting_for(block_fl))) { @@ -1391,11 +1391,10 @@ int __break_lease(struct inode *inode, unsigned int mode, unsigned int type) restart: break_time = flock->fl_break_time; - if (break_time != 0) { + if (break_time != 0) break_time -= jiffies; - if (break_time == 0) - break_time++; - } + if (break_time == 0) + break_time++; locks_insert_block(flock, new_fl); spin_unlock(&inode->i_lock); error = wait_event_interruptible_timeout(new_fl->fl_wait, @@ -1891,7 +1890,7 @@ EXPORT_SYMBOL_GPL(vfs_test_lock); static int posix_lock_to_flock(struct flock *flock, struct file_lock *fl) { - flock->l_pid = IS_FILE_PVT(fl) ? -1 : fl->fl_pid; + flock->l_pid = IS_OFDLCK(fl) ? -1 : fl->fl_pid; #if BITS_PER_LONG == 32 /* * Make sure we can represent the posix lock via @@ -1913,7 +1912,7 @@ static int posix_lock_to_flock(struct flock *flock, struct file_lock *fl) #if BITS_PER_LONG == 32 static void posix_lock_to_flock64(struct flock64 *flock, struct file_lock *fl) { - flock->l_pid = IS_FILE_PVT(fl) ? -1 : fl->fl_pid; + flock->l_pid = IS_OFDLCK(fl) ? -1 : fl->fl_pid; flock->l_start = fl->fl_start; flock->l_len = fl->fl_end == OFFSET_MAX ? 0 : fl->fl_end - fl->fl_start + 1; @@ -1942,13 +1941,13 @@ int fcntl_getlk(struct file *filp, unsigned int cmd, struct flock __user *l) if (error) goto out; - if (cmd == F_GETLKP) { + if (cmd == F_OFD_GETLK) { error = -EINVAL; if (flock.l_pid != 0) goto out; cmd = F_GETLK; - file_lock.fl_flags |= FL_FILE_PVT; + file_lock.fl_flags |= FL_OFDLCK; file_lock.fl_owner = (fl_owner_t)filp; } @@ -2074,25 +2073,25 @@ again: /* * If the cmd is requesting file-private locks, then set the - * FL_FILE_PVT flag and override the owner. + * FL_OFDLCK flag and override the owner. */ switch (cmd) { - case F_SETLKP: + case F_OFD_SETLK: error = -EINVAL; if (flock.l_pid != 0) goto out; cmd = F_SETLK; - file_lock->fl_flags |= FL_FILE_PVT; + file_lock->fl_flags |= FL_OFDLCK; file_lock->fl_owner = (fl_owner_t)filp; break; - case F_SETLKPW: + case F_OFD_SETLKW: error = -EINVAL; if (flock.l_pid != 0) goto out; cmd = F_SETLKW; - file_lock->fl_flags |= FL_FILE_PVT; + file_lock->fl_flags |= FL_OFDLCK; file_lock->fl_owner = (fl_owner_t)filp; /* Fallthrough */ case F_SETLKW: @@ -2144,13 +2143,13 @@ int fcntl_getlk64(struct file *filp, unsigned int cmd, struct flock64 __user *l) if (error) goto out; - if (cmd == F_GETLKP) { + if (cmd == F_OFD_GETLK) { error = -EINVAL; if (flock.l_pid != 0) goto out; cmd = F_GETLK64; - file_lock.fl_flags |= FL_FILE_PVT; + file_lock.fl_flags |= FL_OFDLCK; file_lock.fl_owner = (fl_owner_t)filp; } @@ -2209,25 +2208,25 @@ again: /* * If the cmd is requesting file-private locks, then set the - * FL_FILE_PVT flag and override the owner. + * FL_OFDLCK flag and override the owner. */ switch (cmd) { - case F_SETLKP: + case F_OFD_SETLK: error = -EINVAL; if (flock.l_pid != 0) goto out; cmd = F_SETLK64; - file_lock->fl_flags |= FL_FILE_PVT; + file_lock->fl_flags |= FL_OFDLCK; file_lock->fl_owner = (fl_owner_t)filp; break; - case F_SETLKPW: + case F_OFD_SETLKW: error = -EINVAL; if (flock.l_pid != 0) goto out; cmd = F_SETLKW64; - file_lock->fl_flags |= FL_FILE_PVT; + file_lock->fl_flags |= FL_OFDLCK; file_lock->fl_owner = (fl_owner_t)filp; /* Fallthrough */ case F_SETLKW64: @@ -2413,8 +2412,8 @@ static void lock_get_status(struct seq_file *f, struct file_lock *fl, if (IS_POSIX(fl)) { if (fl->fl_flags & FL_ACCESS) seq_printf(f, "ACCESS"); - else if (IS_FILE_PVT(fl)) - seq_printf(f, "FLPVT "); + else if (IS_OFDLCK(fl)) + seq_printf(f, "OFDLCK"); else seq_printf(f, "POSIX "); diff --git a/fs/namei.c b/fs/namei.c index c6157c894fce..80168273396b 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1542,7 +1542,7 @@ static inline int walk_component(struct nameidata *nd, struct path *path, inode = path->dentry->d_inode; } err = -ENOENT; - if (!inode) + if (!inode || d_is_negative(path->dentry)) goto out_path_put; if (should_follow_link(path->dentry, follow)) { @@ -2249,7 +2249,7 @@ mountpoint_last(struct nameidata *nd, struct path *path) mutex_unlock(&dir->d_inode->i_mutex); done: - if (!dentry->d_inode) { + if (!dentry->d_inode || d_is_negative(dentry)) { error = -ENOENT; dput(dentry); goto out; @@ -2994,7 +2994,7 @@ retry_lookup: finish_lookup: /* we _can_ be in RCU mode here */ error = -ENOENT; - if (d_is_negative(path->dentry)) { + if (!inode || d_is_negative(path->dentry)) { path_to_nameidata(path, nd); goto out; } diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 39c8ef875f91..2c73cae9899d 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -654,9 +654,11 @@ static struct rpc_clnt *create_backchannel_client(struct rpc_create_args *args) static int setup_callback_client(struct nfs4_client *clp, struct nfs4_cb_conn *conn, struct nfsd4_session *ses) { + int maxtime = max_cb_time(clp->net); struct rpc_timeout timeparms = { - .to_initval = max_cb_time(clp->net), + .to_initval = maxtime, .to_retries = 0, + .to_maxval = maxtime, }; struct rpc_create_args args = { .net = clp->net, diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 2723c1badd01..18881f34737a 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -3627,14 +3627,6 @@ nfsd4_encode_operation(struct nfsd4_compoundres *resp, struct nfsd4_op *op) /* nfsd4_check_resp_size guarantees enough room for error status */ if (!op->status) op->status = nfsd4_check_resp_size(resp, 0); - if (op->status == nfserr_resource && nfsd4_has_session(&resp->cstate)) { - struct nfsd4_slot *slot = resp->cstate.slot; - - if (slot->sl_flags & NFSD4_SLOT_CACHETHIS) - op->status = nfserr_rep_too_big_to_cache; - else - op->status = nfserr_rep_too_big; - } if (so) { so->so_replay.rp_status = op->status; so->so_replay.rp_buflen = (char *)resp->p - (char *)(statp+1); diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index 4e565c814309..732648b270dc 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -698,6 +698,8 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags) } group->overflow_event = &oevent->fse; + if (force_o_largefile()) + event_f_flags |= O_LARGEFILE; group->fanotify_data.f_flags = event_f_flags; #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS spin_lock_init(&group->fanotify_data.access_lock); diff --git a/fs/open.c b/fs/open.c index 3d30eb1fc95e..9d64679cec73 100644 --- a/fs/open.c +++ b/fs/open.c @@ -254,17 +254,22 @@ int do_fallocate(struct file *file, int mode, loff_t offset, loff_t len) return -EBADF; /* - * It's not possible to punch hole or perform collapse range - * on append only file + * We can only allow pure fallocate on append only files */ - if (mode & (FALLOC_FL_PUNCH_HOLE | FALLOC_FL_COLLAPSE_RANGE) - && IS_APPEND(inode)) + if ((mode & ~FALLOC_FL_KEEP_SIZE) && IS_APPEND(inode)) return -EPERM; if (IS_IMMUTABLE(inode)) return -EPERM; /* + * We can not allow to do any fallocate operation on an active + * swapfile + */ + if (IS_SWAPFILE(inode)) + ret = -ETXTBSY; + + /* * Revalidate the write permissions, in case security policy has * changed since the files were opened. */ @@ -286,14 +291,6 @@ int do_fallocate(struct file *file, int mode, loff_t offset, loff_t len) if (((offset + len) > inode->i_sb->s_maxbytes) || ((offset + len) < 0)) return -EFBIG; - /* - * There is no need to overlap collapse range with EOF, in which case - * it is effectively a truncate operation - */ - if ((mode & FALLOC_FL_COLLAPSE_RANGE) && - (offset + len >= i_size_read(inode))) - return -EINVAL; - if (!file->f_op->fallocate) return -EOPNOTSUPP; diff --git a/fs/posix_acl.c b/fs/posix_acl.c index 9e363e41dacc..0855f772cd41 100644 --- a/fs/posix_acl.c +++ b/fs/posix_acl.c @@ -246,6 +246,12 @@ posix_acl_equiv_mode(const struct posix_acl *acl, umode_t *mode_p) umode_t mode = 0; int not_equiv = 0; + /* + * A null ACL can always be presented as mode bits. + */ + if (!acl) + return 0; + FOREACH_ACL_ENTRY(pa, acl, pe) { switch (pa->e_tag) { case ACL_USER_OBJ: diff --git a/fs/quota/quota.c b/fs/quota/quota.c index 2b363e23f36e..ff3f0b3cfdb3 100644 --- a/fs/quota/quota.c +++ b/fs/quota/quota.c @@ -278,6 +278,17 @@ static int quota_getxquota(struct super_block *sb, int type, qid_t id, return ret; } +static int quota_rmxquota(struct super_block *sb, void __user *addr) +{ + __u32 flags; + + if (copy_from_user(&flags, addr, sizeof(flags))) + return -EFAULT; + if (!sb->s_qcop->rm_xquota) + return -ENOSYS; + return sb->s_qcop->rm_xquota(sb, flags); +} + /* Copy parameters and call proper function */ static int do_quotactl(struct super_block *sb, int type, int cmd, qid_t id, void __user *addr, struct path *path) @@ -316,8 +327,9 @@ static int do_quotactl(struct super_block *sb, int type, int cmd, qid_t id, return sb->s_qcop->quota_sync(sb, type); case Q_XQUOTAON: case Q_XQUOTAOFF: - case Q_XQUOTARM: return quota_setxstate(sb, cmd, addr); + case Q_XQUOTARM: + return quota_rmxquota(sb, addr); case Q_XGETQSTAT: return quota_getxstate(sb, addr); case Q_XGETQSTATV: diff --git a/fs/super.c b/fs/super.c index e9dc3c3fe159..48377f7463c0 100644 --- a/fs/super.c +++ b/fs/super.c @@ -800,7 +800,10 @@ void emergency_remount(void) static DEFINE_IDA(unnamed_dev_ida); static DEFINE_SPINLOCK(unnamed_dev_lock);/* protects the above */ -static int unnamed_dev_start = 0; /* don't bother trying below it */ +/* Many userspace utilities consider an FSID of 0 invalid. + * Always return at least 1 from get_anon_bdev. + */ +static int unnamed_dev_start = 1; int get_anon_bdev(dev_t *p) { diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 1b8b91b67fdb..28cc1acd5439 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -453,95 +453,3 @@ void sysfs_remove_bin_file(struct kobject *kobj, kernfs_remove_by_name(kobj->sd, attr->attr.name); } EXPORT_SYMBOL_GPL(sysfs_remove_bin_file); - -struct sysfs_schedule_callback_struct { - struct list_head workq_list; - struct kobject *kobj; - void (*func)(void *); - void *data; - struct module *owner; - struct work_struct work; -}; - -static struct workqueue_struct *sysfs_workqueue; -static DEFINE_MUTEX(sysfs_workq_mutex); -static LIST_HEAD(sysfs_workq); -static void sysfs_schedule_callback_work(struct work_struct *work) -{ - struct sysfs_schedule_callback_struct *ss = container_of(work, - struct sysfs_schedule_callback_struct, work); - - (ss->func)(ss->data); - kobject_put(ss->kobj); - module_put(ss->owner); - mutex_lock(&sysfs_workq_mutex); - list_del(&ss->workq_list); - mutex_unlock(&sysfs_workq_mutex); - kfree(ss); -} - -/** - * sysfs_schedule_callback - helper to schedule a callback for a kobject - * @kobj: object we're acting for. - * @func: callback function to invoke later. - * @data: argument to pass to @func. - * @owner: module owning the callback code - * - * sysfs attribute methods must not unregister themselves or their parent - * kobject (which would amount to the same thing). Attempts to do so will - * deadlock, since unregistration is mutually exclusive with driver - * callbacks. - * - * Instead methods can call this routine, which will attempt to allocate - * and schedule a workqueue request to call back @func with @data as its - * argument in the workqueue's process context. @kobj will be pinned - * until @func returns. - * - * Returns 0 if the request was submitted, -ENOMEM if storage could not - * be allocated, -ENODEV if a reference to @owner isn't available, - * -EAGAIN if a callback has already been scheduled for @kobj. - */ -int sysfs_schedule_callback(struct kobject *kobj, void (*func)(void *), - void *data, struct module *owner) -{ - struct sysfs_schedule_callback_struct *ss, *tmp; - - if (!try_module_get(owner)) - return -ENODEV; - - mutex_lock(&sysfs_workq_mutex); - list_for_each_entry_safe(ss, tmp, &sysfs_workq, workq_list) - if (ss->kobj == kobj) { - module_put(owner); - mutex_unlock(&sysfs_workq_mutex); - return -EAGAIN; - } - mutex_unlock(&sysfs_workq_mutex); - - if (sysfs_workqueue == NULL) { - sysfs_workqueue = create_singlethread_workqueue("sysfsd"); - if (sysfs_workqueue == NULL) { - module_put(owner); - return -ENOMEM; - } - } - - ss = kmalloc(sizeof(*ss), GFP_KERNEL); - if (!ss) { - module_put(owner); - return -ENOMEM; - } - kobject_get(kobj); - ss->kobj = kobj; - ss->func = func; - ss->data = data; - ss->owner = owner; - INIT_WORK(&ss->work, sysfs_schedule_callback_work); - INIT_LIST_HEAD(&ss->workq_list); - mutex_lock(&sysfs_workq_mutex); - list_add_tail(&ss->workq_list, &sysfs_workq); - mutex_unlock(&sysfs_workq_mutex); - queue_work(sysfs_workqueue, &ss->work); - return 0; -} -EXPORT_SYMBOL_GPL(sysfs_schedule_callback); diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index a1266089eca1..a81c7b556896 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -1556,7 +1556,7 @@ static int ubifs_remount_rw(struct ubifs_info *c) if (c->space_fixup) { err = ubifs_fixup_free_space(c); if (err) - return err; + goto out; } err = check_free_space(c); diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h index 0fdd4109c624..6e247a99f5db 100644 --- a/fs/xfs/xfs_ag.h +++ b/fs/xfs/xfs_ag.h @@ -160,30 +160,38 @@ typedef struct xfs_agi { * still being referenced. */ __be32 agi_unlinked[XFS_AGI_UNLINKED_BUCKETS]; - + /* + * This marks the end of logging region 1 and start of logging region 2. + */ uuid_t agi_uuid; /* uuid of filesystem */ __be32 agi_crc; /* crc of agi sector */ __be32 agi_pad32; __be64 agi_lsn; /* last write sequence */ + __be32 agi_free_root; /* root of the free inode btree */ + __be32 agi_free_level;/* levels in free inode btree */ + /* structure must be padded to 64 bit alignment */ } xfs_agi_t; #define XFS_AGI_CRC_OFF offsetof(struct xfs_agi, agi_crc) -#define XFS_AGI_MAGICNUM 0x00000001 -#define XFS_AGI_VERSIONNUM 0x00000002 -#define XFS_AGI_SEQNO 0x00000004 -#define XFS_AGI_LENGTH 0x00000008 -#define XFS_AGI_COUNT 0x00000010 -#define XFS_AGI_ROOT 0x00000020 -#define XFS_AGI_LEVEL 0x00000040 -#define XFS_AGI_FREECOUNT 0x00000080 -#define XFS_AGI_NEWINO 0x00000100 -#define XFS_AGI_DIRINO 0x00000200 -#define XFS_AGI_UNLINKED 0x00000400 -#define XFS_AGI_NUM_BITS 11 -#define XFS_AGI_ALL_BITS ((1 << XFS_AGI_NUM_BITS) - 1) +#define XFS_AGI_MAGICNUM (1 << 0) +#define XFS_AGI_VERSIONNUM (1 << 1) +#define XFS_AGI_SEQNO (1 << 2) +#define XFS_AGI_LENGTH (1 << 3) +#define XFS_AGI_COUNT (1 << 4) +#define XFS_AGI_ROOT (1 << 5) +#define XFS_AGI_LEVEL (1 << 6) +#define XFS_AGI_FREECOUNT (1 << 7) +#define XFS_AGI_NEWINO (1 << 8) +#define XFS_AGI_DIRINO (1 << 9) +#define XFS_AGI_UNLINKED (1 << 10) +#define XFS_AGI_NUM_BITS_R1 11 /* end of the 1st agi logging region */ +#define XFS_AGI_ALL_BITS_R1 ((1 << XFS_AGI_NUM_BITS_R1) - 1) +#define XFS_AGI_FREE_ROOT (1 << 11) +#define XFS_AGI_FREE_LEVEL (1 << 12) +#define XFS_AGI_NUM_BITS_R2 13 /* disk block (xfs_daddr_t) in the AG */ #define XFS_AGI_DADDR(mp) ((xfs_daddr_t)(2 << (mp)->m_sectbb_log)) diff --git a/fs/xfs/xfs_alloc_btree.c b/fs/xfs/xfs_alloc_btree.c index cc1eadcbb049..8358f1ded94d 100644 --- a/fs/xfs/xfs_alloc_btree.c +++ b/fs/xfs/xfs_alloc_btree.c @@ -70,7 +70,6 @@ xfs_allocbt_alloc_block( struct xfs_btree_cur *cur, union xfs_btree_ptr *start, union xfs_btree_ptr *new, - int length, int *stat) { int error; diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 75df77d09f75..0479c32c5eb1 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -1344,6 +1344,14 @@ __xfs_get_blocks( /* * If this is O_DIRECT or the mpage code calling tell them how large * the mapping is, so that we can avoid repeated get_blocks calls. + * + * If the mapping spans EOF, then we have to break the mapping up as the + * mapping for blocks beyond EOF must be marked new so that sub block + * regions can be correctly zeroed. We can't do this for mappings within + * EOF unless the mapping was just allocated or is unwritten, otherwise + * the callers would overwrite existing data with zeros. Hence we have + * to split the mapping into a range up to and including EOF, and a + * second mapping for beyond EOF. */ if (direct || size > (1 << inode->i_blkbits)) { xfs_off_t mapping_size; @@ -1354,6 +1362,12 @@ __xfs_get_blocks( ASSERT(mapping_size > 0); if (mapping_size > size) mapping_size = size; + if (offset < i_size_read(inode) && + offset + mapping_size >= i_size_read(inode)) { + /* limit mapping to block that spans EOF */ + mapping_size = roundup_64(i_size_read(inode) - offset, + 1 << inode->i_blkbits); + } if (mapping_size > LONG_MAX) mapping_size = LONG_MAX; @@ -1566,6 +1580,16 @@ xfs_vm_write_failed( xfs_vm_kill_delalloc_range(inode, block_offset, block_offset + bh->b_size); + + /* + * This buffer does not contain data anymore. make sure anyone + * who finds it knows that for certain. + */ + clear_buffer_delay(bh); + clear_buffer_uptodate(bh); + clear_buffer_mapped(bh); + clear_buffer_new(bh); + clear_buffer_dirty(bh); } } @@ -1599,12 +1623,21 @@ xfs_vm_write_begin( status = __block_write_begin(page, pos, len, xfs_get_blocks); if (unlikely(status)) { struct inode *inode = mapping->host; + size_t isize = i_size_read(inode); xfs_vm_write_failed(inode, page, pos, len); unlock_page(page); - if (pos + len > i_size_read(inode)) - truncate_pagecache(inode, i_size_read(inode)); + /* + * If the write is beyond EOF, we only want to kill blocks + * allocated in this write, not blocks that were previously + * written successfully. + */ + if (pos + len > isize) { + ssize_t start = max_t(ssize_t, pos, isize); + + truncate_pagecache_range(inode, start, pos + len); + } page_cache_release(page); page = NULL; @@ -1615,9 +1648,12 @@ xfs_vm_write_begin( } /* - * On failure, we only need to kill delalloc blocks beyond EOF because they - * will never be written. For blocks within EOF, generic_write_end() zeros them - * so they are safe to leave alone and be written with all the other valid data. + * On failure, we only need to kill delalloc blocks beyond EOF in the range of + * this specific write because they will never be written. Previous writes + * beyond EOF where block allocation succeeded do not need to be trashed, so + * only new blocks from this write should be trashed. For blocks within + * EOF, generic_write_end() zeros them so they are safe to leave alone and be + * written with all the other valid data. */ STATIC int xfs_vm_write_end( @@ -1640,8 +1676,11 @@ xfs_vm_write_end( loff_t to = pos + len; if (to > isize) { - truncate_pagecache(inode, isize); + /* only kill blocks in this write beyond EOF */ + if (pos > isize) + isize = pos; xfs_vm_kill_delalloc_range(inode, isize, to); + truncate_pagecache_range(inode, isize, to); } } return ret; diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c index 86f482e5798f..1fc1f06277da 100644 --- a/fs/xfs/xfs_attr.c +++ b/fs/xfs/xfs_attr.c @@ -187,7 +187,7 @@ xfs_attr_calc_size( * Out of line attribute, cannot double split, but * make room for the attribute value itself. */ - uint dblocks = XFS_B_TO_FSB(mp, args->valuelen); + uint dblocks = xfs_attr3_rmt_blocks(mp, args->valuelen); nblks += dblocks; nblks += XFS_NEXTENTADD_SPACE_RES(mp, dblocks, XFS_ATTR_FORK); } @@ -604,11 +604,22 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) trace_xfs_attr_leaf_replace(args); + /* save the attribute state for later removal*/ args->op_flags |= XFS_DA_OP_RENAME; /* an atomic rename */ args->blkno2 = args->blkno; /* set 2nd entry info*/ args->index2 = args->index; args->rmtblkno2 = args->rmtblkno; args->rmtblkcnt2 = args->rmtblkcnt; + args->rmtvaluelen2 = args->rmtvaluelen; + + /* + * clear the remote attr state now that it is saved so that the + * values reflect the state of the attribute we are about to + * add, not the attribute we just found and will remove later. + */ + args->rmtblkno = 0; + args->rmtblkcnt = 0; + args->rmtvaluelen = 0; } /* @@ -700,6 +711,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) args->blkno = args->blkno2; args->rmtblkno = args->rmtblkno2; args->rmtblkcnt = args->rmtblkcnt2; + args->rmtvaluelen = args->rmtvaluelen2; if (args->rmtblkno) { error = xfs_attr_rmtval_remove(args); if (error) @@ -905,13 +917,22 @@ restart: trace_xfs_attr_node_replace(args); + /* save the attribute state for later removal*/ args->op_flags |= XFS_DA_OP_RENAME; /* atomic rename op */ args->blkno2 = args->blkno; /* set 2nd entry info*/ args->index2 = args->index; args->rmtblkno2 = args->rmtblkno; args->rmtblkcnt2 = args->rmtblkcnt; + args->rmtvaluelen2 = args->rmtvaluelen; + + /* + * clear the remote attr state now that it is saved so that the + * values reflect the state of the attribute we are about to + * add, not the attribute we just found and will remove later. + */ args->rmtblkno = 0; args->rmtblkcnt = 0; + args->rmtvaluelen = 0; } retval = xfs_attr3_leaf_add(blk->bp, state->args); @@ -1039,6 +1060,7 @@ restart: args->blkno = args->blkno2; args->rmtblkno = args->rmtblkno2; args->rmtblkcnt = args->rmtblkcnt2; + args->rmtvaluelen = args->rmtvaluelen2; if (args->rmtblkno) { error = xfs_attr_rmtval_remove(args); if (error) diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index fe9587fab17a..511c283459b1 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -1229,6 +1229,7 @@ xfs_attr3_leaf_add_work( name_rmt->valueblk = 0; args->rmtblkno = 1; args->rmtblkcnt = xfs_attr3_rmt_blocks(mp, args->valuelen); + args->rmtvaluelen = args->valuelen; } xfs_trans_log_buf(args->trans, bp, XFS_DA_LOGRANGE(leaf, xfs_attr3_leaf_name(leaf, args->index), @@ -2167,11 +2168,11 @@ 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->rmtvaluelen = be32_to_cpu(name_rmt->valuelen); args->rmtblkno = be32_to_cpu(name_rmt->valueblk); args->rmtblkcnt = xfs_attr3_rmt_blocks( args->dp->i_mount, - args->valuelen); + args->rmtvaluelen); return XFS_ERROR(EEXIST); } } @@ -2220,19 +2221,19 @@ xfs_attr3_leaf_getvalue( name_rmt = xfs_attr3_leaf_name_remote(leaf, args->index); ASSERT(name_rmt->namelen == args->namelen); ASSERT(memcmp(args->name, name_rmt->name, args->namelen) == 0); - valuelen = be32_to_cpu(name_rmt->valuelen); + args->rmtvaluelen = be32_to_cpu(name_rmt->valuelen); args->rmtblkno = be32_to_cpu(name_rmt->valueblk); args->rmtblkcnt = xfs_attr3_rmt_blocks(args->dp->i_mount, - valuelen); + args->rmtvaluelen); if (args->flags & ATTR_KERNOVAL) { - args->valuelen = valuelen; + args->valuelen = args->rmtvaluelen; return 0; } - if (args->valuelen < valuelen) { - args->valuelen = valuelen; + if (args->valuelen < args->rmtvaluelen) { + args->valuelen = args->rmtvaluelen; return XFS_ERROR(ERANGE); } - args->valuelen = valuelen; + args->valuelen = args->rmtvaluelen; } return 0; } @@ -2519,7 +2520,7 @@ xfs_attr3_leaf_clearflag( ASSERT((entry->flags & XFS_ATTR_LOCAL) == 0); name_rmt = xfs_attr3_leaf_name_remote(leaf, args->index); name_rmt->valueblk = cpu_to_be32(args->rmtblkno); - name_rmt->valuelen = cpu_to_be32(args->valuelen); + name_rmt->valuelen = cpu_to_be32(args->rmtvaluelen); xfs_trans_log_buf(args->trans, bp, XFS_DA_LOGRANGE(leaf, name_rmt, sizeof(*name_rmt))); } @@ -2677,7 +2678,7 @@ xfs_attr3_leaf_flipflags( ASSERT((entry1->flags & XFS_ATTR_LOCAL) == 0); name_rmt = xfs_attr3_leaf_name_remote(leaf1, args->index); name_rmt->valueblk = cpu_to_be32(args->rmtblkno); - name_rmt->valuelen = cpu_to_be32(args->valuelen); + name_rmt->valuelen = cpu_to_be32(args->rmtvaluelen); xfs_trans_log_buf(args->trans, bp1, XFS_DA_LOGRANGE(leaf1, name_rmt, sizeof(*name_rmt))); } diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c index 01db96f60cf0..833fe5d98d80 100644 --- a/fs/xfs/xfs_attr_list.c +++ b/fs/xfs/xfs_attr_list.c @@ -447,6 +447,7 @@ xfs_attr3_leaf_list_int( args.dp = context->dp; args.whichfork = XFS_ATTR_FORK; args.valuelen = valuelen; + args.rmtvaluelen = valuelen; args.value = kmem_alloc(valuelen, KM_SLEEP | KM_NOFS); args.rmtblkno = be32_to_cpu(name_rmt->valueblk); args.rmtblkcnt = xfs_attr3_rmt_blocks( diff --git a/fs/xfs/xfs_attr_remote.c b/fs/xfs/xfs_attr_remote.c index 6e37823e2932..0f0679a134e2 100644 --- a/fs/xfs/xfs_attr_remote.c +++ b/fs/xfs/xfs_attr_remote.c @@ -68,7 +68,6 @@ xfs_attr3_rmt_blocks( */ static bool xfs_attr3_rmt_hdr_ok( - struct xfs_mount *mp, void *ptr, xfs_ino_t ino, uint32_t offset, @@ -251,7 +250,7 @@ xfs_attr_rmtval_copyout( byte_cnt = min(*valuelen, byte_cnt); if (xfs_sb_version_hascrc(&mp->m_sb)) { - if (!xfs_attr3_rmt_hdr_ok(mp, src, ino, *offset, + if (!xfs_attr3_rmt_hdr_ok(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)", @@ -337,7 +336,7 @@ xfs_attr_rmtval_get( struct xfs_buf *bp; xfs_dablk_t lblkno = args->rmtblkno; __uint8_t *dst = args->value; - int valuelen = args->valuelen; + int valuelen; int nmap; int error; int blkcnt = args->rmtblkcnt; @@ -347,7 +346,9 @@ xfs_attr_rmtval_get( trace_xfs_attr_rmtval_get(args); ASSERT(!(args->flags & ATTR_KERNOVAL)); + ASSERT(args->rmtvaluelen == args->valuelen); + valuelen = args->rmtvaluelen; while (valuelen > 0) { nmap = ATTR_RMTVALUE_MAPSIZE; error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno, @@ -415,7 +416,7 @@ xfs_attr_rmtval_set( * attributes have headers, we can't just do a straight byte to FSB * conversion and have to take the header space into account. */ - blkcnt = xfs_attr3_rmt_blocks(mp, args->valuelen); + blkcnt = xfs_attr3_rmt_blocks(mp, args->rmtvaluelen); error = xfs_bmap_first_unused(args->trans, args->dp, blkcnt, &lfileoff, XFS_ATTR_FORK); if (error) @@ -480,7 +481,7 @@ xfs_attr_rmtval_set( */ lblkno = args->rmtblkno; blkcnt = args->rmtblkcnt; - valuelen = args->valuelen; + valuelen = args->rmtvaluelen; while (valuelen > 0) { struct xfs_buf *bp; xfs_daddr_t dblkno; diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 5b6092ef51ef..1ff0da6e2bf9 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -94,7 +94,7 @@ xfs_bmap_compute_maxlevels( maxleafents = MAXAEXTNUM; sz = XFS_BMDR_SPACE_CALC(MINABTPTRS); } - maxrootrecs = xfs_bmdr_maxrecs(mp, sz, 0); + maxrootrecs = xfs_bmdr_maxrecs(sz, 0); minleafrecs = mp->m_bmap_dmnr[0]; minnoderecs = mp->m_bmap_dmnr[1]; maxblocks = (maxleafents + minleafrecs - 1) / minleafrecs; @@ -233,7 +233,6 @@ xfs_default_attroffset( */ STATIC void xfs_bmap_forkoff_reset( - xfs_mount_t *mp, xfs_inode_t *ip, int whichfork) { @@ -905,7 +904,7 @@ xfs_bmap_local_to_extents_empty( ASSERT(ifp->if_bytes == 0); ASSERT(XFS_IFORK_NEXTENTS(ip, whichfork) == 0); - xfs_bmap_forkoff_reset(ip->i_mount, ip, whichfork); + xfs_bmap_forkoff_reset(ip, whichfork); ifp->if_flags &= ~XFS_IFINLINE; ifp->if_flags |= XFS_IFEXTENTS; XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS); @@ -1675,7 +1674,6 @@ xfs_bmap_isaeof( */ int xfs_bmap_last_offset( - struct xfs_trans *tp, struct xfs_inode *ip, xfs_fileoff_t *last_block, int whichfork) @@ -3517,6 +3515,67 @@ xfs_bmap_adjacent( #undef ISVALID } +static int +xfs_bmap_longest_free_extent( + struct xfs_trans *tp, + xfs_agnumber_t ag, + xfs_extlen_t *blen, + int *notinit) +{ + struct xfs_mount *mp = tp->t_mountp; + struct xfs_perag *pag; + xfs_extlen_t longest; + int error = 0; + + pag = xfs_perag_get(mp, ag); + if (!pag->pagf_init) { + error = xfs_alloc_pagf_init(mp, tp, ag, XFS_ALLOC_FLAG_TRYLOCK); + if (error) + goto out; + + if (!pag->pagf_init) { + *notinit = 1; + goto out; + } + } + + longest = xfs_alloc_longest_free_extent(mp, pag); + if (*blen < longest) + *blen = longest; + +out: + xfs_perag_put(pag); + return error; +} + +static void +xfs_bmap_select_minlen( + struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args, + xfs_extlen_t *blen, + int notinit) +{ + if (notinit || *blen < ap->minlen) { + /* + * Since we did a BUF_TRYLOCK above, it is possible that + * there is space for this request. + */ + args->minlen = ap->minlen; + } else if (*blen < args->maxlen) { + /* + * If the best seen length is less than the request length, + * use the best as the minimum. + */ + args->minlen = *blen; + } else { + /* + * Otherwise we've seen an extent as big as maxlen, use that + * as the minimum. + */ + args->minlen = args->maxlen; + } +} + STATIC int xfs_bmap_btalloc_nullfb( struct xfs_bmalloca *ap, @@ -3524,111 +3583,74 @@ xfs_bmap_btalloc_nullfb( xfs_extlen_t *blen) { struct xfs_mount *mp = ap->ip->i_mount; - struct xfs_perag *pag; xfs_agnumber_t ag, startag; int notinit = 0; int error; - if (ap->userdata && xfs_inode_is_filestream(ap->ip)) - args->type = XFS_ALLOCTYPE_NEAR_BNO; - else - args->type = XFS_ALLOCTYPE_START_BNO; + args->type = XFS_ALLOCTYPE_START_BNO; args->total = ap->total; - /* - * Search for an allocation group with a single extent large enough - * for the request. If one isn't found, then adjust the minimum - * allocation size to the largest space found. - */ startag = ag = XFS_FSB_TO_AGNO(mp, args->fsbno); if (startag == NULLAGNUMBER) startag = ag = 0; - pag = xfs_perag_get(mp, ag); while (*blen < args->maxlen) { - if (!pag->pagf_init) { - error = xfs_alloc_pagf_init(mp, args->tp, ag, - XFS_ALLOC_FLAG_TRYLOCK); - if (error) { - xfs_perag_put(pag); - return error; - } - } - - /* - * See xfs_alloc_fix_freelist... - */ - if (pag->pagf_init) { - xfs_extlen_t longest; - longest = xfs_alloc_longest_free_extent(mp, pag); - if (*blen < longest) - *blen = longest; - } else - notinit = 1; - - if (xfs_inode_is_filestream(ap->ip)) { - if (*blen >= args->maxlen) - break; - - if (ap->userdata) { - /* - * If startag is an invalid AG, we've - * come here once before and - * xfs_filestream_new_ag picked the - * best currently available. - * - * Don't continue looping, since we - * could loop forever. - */ - if (startag == NULLAGNUMBER) - break; - - error = xfs_filestream_new_ag(ap, &ag); - xfs_perag_put(pag); - if (error) - return error; + error = xfs_bmap_longest_free_extent(args->tp, ag, blen, + ¬init); + if (error) + return error; - /* loop again to set 'blen'*/ - startag = NULLAGNUMBER; - pag = xfs_perag_get(mp, ag); - continue; - } - } if (++ag == mp->m_sb.sb_agcount) ag = 0; if (ag == startag) break; - xfs_perag_put(pag); - pag = xfs_perag_get(mp, ag); } - xfs_perag_put(pag); - /* - * Since the above loop did a BUF_TRYLOCK, it is - * possible that there is space for this request. - */ - if (notinit || *blen < ap->minlen) - args->minlen = ap->minlen; - /* - * If the best seen length is less than the request - * length, use the best as the minimum. - */ - else if (*blen < args->maxlen) - args->minlen = *blen; - /* - * Otherwise we've seen an extent as big as maxlen, - * use that as the minimum. - */ - else - args->minlen = args->maxlen; + xfs_bmap_select_minlen(ap, args, blen, notinit); + return 0; +} + +STATIC int +xfs_bmap_btalloc_filestreams( + struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args, + xfs_extlen_t *blen) +{ + struct xfs_mount *mp = ap->ip->i_mount; + xfs_agnumber_t ag; + int notinit = 0; + int error; + + args->type = XFS_ALLOCTYPE_NEAR_BNO; + args->total = ap->total; + + ag = XFS_FSB_TO_AGNO(mp, args->fsbno); + if (ag == NULLAGNUMBER) + ag = 0; + + error = xfs_bmap_longest_free_extent(args->tp, ag, blen, ¬init); + if (error) + return error; + + if (*blen < args->maxlen) { + error = xfs_filestream_new_ag(ap, &ag); + if (error) + return error; + + error = xfs_bmap_longest_free_extent(args->tp, ag, blen, + ¬init); + if (error) + return error; + + } + + xfs_bmap_select_minlen(ap, args, blen, notinit); /* - * set the failure fallback case to look in the selected - * AG as the stream may have moved. + * Set the failure fallback case to look in the selected AG as stream + * may have moved. */ - if (xfs_inode_is_filestream(ap->ip)) - ap->blkno = args->fsbno = XFS_AGB_TO_FSB(mp, ag, 0); - + ap->blkno = args->fsbno = XFS_AGB_TO_FSB(mp, ag, 0); return 0; } @@ -3708,7 +3730,15 @@ xfs_bmap_btalloc( args.firstblock = *ap->firstblock; blen = 0; if (nullfb) { - error = xfs_bmap_btalloc_nullfb(ap, &args, &blen); + /* + * Search for an allocation group with a single extent large + * enough for the request. If one isn't found, then adjust + * the minimum allocation size to the largest space found. + */ + if (ap->userdata && xfs_inode_is_filestream(ap->ip)) + error = xfs_bmap_btalloc_filestreams(ap, &args, &blen); + else + error = xfs_bmap_btalloc_nullfb(ap, &args, &blen); if (error) return error; } else if (ap->flist->xbf_low) { @@ -5413,6 +5443,7 @@ xfs_bmap_shift_extents( int whichfork = XFS_DATA_FORK; int logflags; xfs_filblks_t blockcount = 0; + int total_extents; if (unlikely(XFS_TEST_ERROR( (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS && @@ -5429,7 +5460,6 @@ xfs_bmap_shift_extents( ASSERT(current_ext != NULL); ifp = XFS_IFORK_PTR(ip, whichfork); - if (!(ifp->if_flags & XFS_IFEXTENTS)) { /* Read in all the extents */ error = xfs_iread_extents(tp, ip, whichfork); @@ -5456,7 +5486,6 @@ xfs_bmap_shift_extents( /* We are going to change core inode */ logflags = XFS_ILOG_CORE; - if (ifp->if_flags & XFS_IFBROOT) { cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork); cur->bc_private.b.firstblock = *firstblock; @@ -5467,8 +5496,14 @@ xfs_bmap_shift_extents( logflags |= XFS_ILOG_DEXT; } - while (nexts++ < num_exts && - *current_ext < XFS_IFORK_NEXTENTS(ip, whichfork)) { + /* + * There may be delalloc extents in the data fork before the range we + * are collapsing out, so we cannot + * use the count of real extents here. Instead we have to calculate it + * from the incore fork. + */ + total_extents = ifp->if_bytes / sizeof(xfs_bmbt_rec_t); + while (nexts++ < num_exts && *current_ext < total_extents) { gotp = xfs_iext_get_ext(ifp, *current_ext); xfs_bmbt_get_all(gotp, &got); @@ -5556,10 +5591,11 @@ xfs_bmap_shift_extents( } (*current_ext)++; + total_extents = ifp->if_bytes / sizeof(xfs_bmbt_rec_t); } /* Check if we are done */ - if (*current_ext == XFS_IFORK_NEXTENTS(ip, whichfork)) + if (*current_ext == total_extents) *done = 1; del_cursor: @@ -5568,6 +5604,5 @@ del_cursor: error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR); xfs_trans_log_inode(tp, ip, logflags); - return error; } diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h index f84bd7af43be..38ba36e9b2f0 100644 --- a/fs/xfs/xfs_bmap.h +++ b/fs/xfs/xfs_bmap.h @@ -156,8 +156,8 @@ int xfs_bmap_first_unused(struct xfs_trans *tp, struct xfs_inode *ip, xfs_extlen_t len, xfs_fileoff_t *unused, int whichfork); int xfs_bmap_last_before(struct xfs_trans *tp, struct xfs_inode *ip, xfs_fileoff_t *last_block, int whichfork); -int xfs_bmap_last_offset(struct xfs_trans *tp, struct xfs_inode *ip, - xfs_fileoff_t *unused, int whichfork); +int xfs_bmap_last_offset(struct xfs_inode *ip, xfs_fileoff_t *unused, + int whichfork); int xfs_bmap_one_block(struct xfs_inode *ip, int whichfork); int xfs_bmap_read_extents(struct xfs_trans *tp, struct xfs_inode *ip, int whichfork); diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c index 818d546664e7..948836c4fd90 100644 --- a/fs/xfs/xfs_bmap_btree.c +++ b/fs/xfs/xfs_bmap_btree.c @@ -84,7 +84,7 @@ xfs_bmdr_to_bmbt( rblock->bb_level = dblock->bb_level; ASSERT(be16_to_cpu(rblock->bb_level) > 0); rblock->bb_numrecs = dblock->bb_numrecs; - dmxr = xfs_bmdr_maxrecs(mp, dblocklen, 0); + dmxr = xfs_bmdr_maxrecs(dblocklen, 0); fkp = XFS_BMDR_KEY_ADDR(dblock, 1); tkp = XFS_BMBT_KEY_ADDR(mp, rblock, 1); fpp = XFS_BMDR_PTR_ADDR(dblock, 1, dmxr); @@ -443,7 +443,7 @@ xfs_bmbt_to_bmdr( ASSERT(rblock->bb_level != 0); dblock->bb_level = rblock->bb_level; dblock->bb_numrecs = rblock->bb_numrecs; - dmxr = xfs_bmdr_maxrecs(mp, dblocklen, 0); + dmxr = xfs_bmdr_maxrecs(dblocklen, 0); fkp = XFS_BMBT_KEY_ADDR(mp, rblock, 1); tkp = XFS_BMDR_KEY_ADDR(dblock, 1); fpp = XFS_BMAP_BROOT_PTR_ADDR(mp, rblock, 1, rblocklen); @@ -519,7 +519,6 @@ xfs_bmbt_alloc_block( struct xfs_btree_cur *cur, union xfs_btree_ptr *start, union xfs_btree_ptr *new, - int length, int *stat) { xfs_alloc_arg_t args; /* block allocation args */ @@ -672,8 +671,7 @@ xfs_bmbt_get_dmaxrecs( { if (level != cur->bc_nlevels - 1) return cur->bc_mp->m_bmap_dmxr[level != 0]; - return xfs_bmdr_maxrecs(cur->bc_mp, cur->bc_private.b.forksize, - level == 0); + return xfs_bmdr_maxrecs(cur->bc_private.b.forksize, level == 0); } STATIC void @@ -914,7 +912,6 @@ xfs_bmbt_maxrecs( */ int xfs_bmdr_maxrecs( - struct xfs_mount *mp, int blocklen, int leaf) { diff --git a/fs/xfs/xfs_bmap_btree.h b/fs/xfs/xfs_bmap_btree.h index 6e42e1e50b89..819a8a4dee95 100644 --- a/fs/xfs/xfs_bmap_btree.h +++ b/fs/xfs/xfs_bmap_btree.h @@ -130,7 +130,7 @@ extern void xfs_bmbt_to_bmdr(struct xfs_mount *, struct xfs_btree_block *, int, xfs_bmdr_block_t *, int); extern int xfs_bmbt_get_maxrecs(struct xfs_btree_cur *, int level); -extern int xfs_bmdr_maxrecs(struct xfs_mount *, int blocklen, int leaf); +extern int xfs_bmdr_maxrecs(int blocklen, int leaf); extern int xfs_bmbt_maxrecs(struct xfs_mount *, int blocklen, int leaf); extern int xfs_bmbt_change_owner(struct xfs_trans *tp, struct xfs_inode *ip, diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index 01f6a646caa1..296160b8e78c 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -1418,6 +1418,8 @@ xfs_zero_file_space( xfs_off_t end_boundary; int error; + trace_xfs_zero_file_space(ip); + granularity = max_t(uint, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE); /* @@ -1432,9 +1434,18 @@ xfs_zero_file_space( ASSERT(end_boundary <= offset + len); if (start_boundary < end_boundary - 1) { - /* punch out the page cache over the conversion range */ + /* + * punch out delayed allocation blocks and the page cache over + * the conversion range + */ + xfs_ilock(ip, XFS_ILOCK_EXCL); + error = xfs_bmap_punch_delalloc_range(ip, + XFS_B_TO_FSBT(mp, start_boundary), + XFS_B_TO_FSB(mp, end_boundary - start_boundary)); + xfs_iunlock(ip, XFS_ILOCK_EXCL); truncate_pagecache_range(VFS_I(ip), start_boundary, end_boundary - 1); + /* convert the blocks */ error = xfs_alloc_file_space(ip, start_boundary, end_boundary - start_boundary - 1, diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c index e80d59fdf89a..182bac2bb276 100644 --- a/fs/xfs/xfs_btree.c +++ b/fs/xfs/xfs_btree.c @@ -43,9 +43,10 @@ kmem_zone_t *xfs_btree_cur_zone; * Btree magic numbers. */ static const __uint32_t xfs_magics[2][XFS_BTNUM_MAX] = { - { XFS_ABTB_MAGIC, XFS_ABTC_MAGIC, XFS_BMAP_MAGIC, XFS_IBT_MAGIC }, + { XFS_ABTB_MAGIC, XFS_ABTC_MAGIC, XFS_BMAP_MAGIC, XFS_IBT_MAGIC, + XFS_FIBT_MAGIC }, { XFS_ABTB_CRC_MAGIC, XFS_ABTC_CRC_MAGIC, - XFS_BMAP_CRC_MAGIC, XFS_IBT_CRC_MAGIC } + XFS_BMAP_CRC_MAGIC, XFS_IBT_CRC_MAGIC, XFS_FIBT_CRC_MAGIC } }; #define xfs_btree_magic(cur) \ xfs_magics[!!((cur)->bc_flags & XFS_BTREE_CRC_BLOCKS)][cur->bc_btnum] @@ -1115,6 +1116,7 @@ xfs_btree_set_refs( xfs_buf_set_ref(bp, XFS_ALLOC_BTREE_REF); break; case XFS_BTNUM_INO: + case XFS_BTNUM_FINO: xfs_buf_set_ref(bp, XFS_INO_BTREE_REF); break; case XFS_BTNUM_BMAP: @@ -1159,7 +1161,6 @@ STATIC int xfs_btree_read_buf_block( struct xfs_btree_cur *cur, union xfs_btree_ptr *ptr, - int level, int flags, struct xfs_btree_block **block, struct xfs_buf **bpp) @@ -1517,8 +1518,8 @@ xfs_btree_increment( union xfs_btree_ptr *ptrp; ptrp = xfs_btree_ptr_addr(cur, cur->bc_ptrs[lev], block); - error = xfs_btree_read_buf_block(cur, ptrp, --lev, - 0, &block, &bp); + --lev; + error = xfs_btree_read_buf_block(cur, ptrp, 0, &block, &bp); if (error) goto error0; @@ -1616,8 +1617,8 @@ xfs_btree_decrement( union xfs_btree_ptr *ptrp; ptrp = xfs_btree_ptr_addr(cur, cur->bc_ptrs[lev], block); - error = xfs_btree_read_buf_block(cur, ptrp, --lev, - 0, &block, &bp); + --lev; + error = xfs_btree_read_buf_block(cur, ptrp, 0, &block, &bp); if (error) goto error0; xfs_btree_setbuf(cur, lev, bp); @@ -1667,7 +1668,7 @@ xfs_btree_lookup_get_block( return 0; } - error = xfs_btree_read_buf_block(cur, pp, level, 0, blkp, &bp); + error = xfs_btree_read_buf_block(cur, pp, 0, blkp, &bp); if (error) return error; @@ -2018,7 +2019,7 @@ xfs_btree_lshift( goto out0; /* Set up the left neighbor as "left". */ - error = xfs_btree_read_buf_block(cur, &lptr, level, 0, &left, &lbp); + error = xfs_btree_read_buf_block(cur, &lptr, 0, &left, &lbp); if (error) goto error0; @@ -2202,7 +2203,7 @@ xfs_btree_rshift( goto out0; /* Set up the right neighbor as "right". */ - error = xfs_btree_read_buf_block(cur, &rptr, level, 0, &right, &rbp); + error = xfs_btree_read_buf_block(cur, &rptr, 0, &right, &rbp); if (error) goto error0; @@ -2372,7 +2373,7 @@ xfs_btree_split( xfs_btree_buf_to_ptr(cur, lbp, &lptr); /* Allocate the new block. If we can't do it, we're toast. Give up. */ - error = cur->bc_ops->alloc_block(cur, &lptr, &rptr, 1, stat); + error = cur->bc_ops->alloc_block(cur, &lptr, &rptr, stat); if (error) goto error0; if (*stat == 0) @@ -2470,7 +2471,7 @@ xfs_btree_split( * point back to right instead of to left. */ if (!xfs_btree_ptr_is_null(cur, &rrptr)) { - error = xfs_btree_read_buf_block(cur, &rrptr, level, + error = xfs_btree_read_buf_block(cur, &rrptr, 0, &rrblock, &rrbp); if (error) goto error0; @@ -2545,7 +2546,7 @@ xfs_btree_new_iroot( pp = xfs_btree_ptr_addr(cur, 1, block); /* Allocate the new block. If we can't do it, we're toast. Give up. */ - error = cur->bc_ops->alloc_block(cur, pp, &nptr, 1, stat); + error = cur->bc_ops->alloc_block(cur, pp, &nptr, stat); if (error) goto error0; if (*stat == 0) { @@ -2649,7 +2650,7 @@ xfs_btree_new_root( cur->bc_ops->init_ptr_from_cur(cur, &rptr); /* Allocate the new block. If we can't do it, we're toast. Give up. */ - error = cur->bc_ops->alloc_block(cur, &rptr, &lptr, 1, stat); + error = cur->bc_ops->alloc_block(cur, &rptr, &lptr, stat); if (error) goto error0; if (*stat == 0) @@ -2684,8 +2685,7 @@ xfs_btree_new_root( lbp = bp; xfs_btree_buf_to_ptr(cur, lbp, &lptr); left = block; - error = xfs_btree_read_buf_block(cur, &rptr, - cur->bc_nlevels - 1, 0, &right, &rbp); + error = xfs_btree_read_buf_block(cur, &rptr, 0, &right, &rbp); if (error) goto error0; bp = rbp; @@ -2696,8 +2696,7 @@ xfs_btree_new_root( xfs_btree_buf_to_ptr(cur, rbp, &rptr); right = block; xfs_btree_get_sibling(cur, right, &lptr, XFS_BB_LEFTSIB); - error = xfs_btree_read_buf_block(cur, &lptr, - cur->bc_nlevels - 1, 0, &left, &lbp); + error = xfs_btree_read_buf_block(cur, &lptr, 0, &left, &lbp); if (error) goto error0; bp = lbp; @@ -3649,8 +3648,7 @@ xfs_btree_delrec( rptr = cptr; right = block; rbp = bp; - error = xfs_btree_read_buf_block(cur, &lptr, level, - 0, &left, &lbp); + error = xfs_btree_read_buf_block(cur, &lptr, 0, &left, &lbp); if (error) goto error0; @@ -3667,8 +3665,7 @@ xfs_btree_delrec( lptr = cptr; left = block; lbp = bp; - error = xfs_btree_read_buf_block(cur, &rptr, level, - 0, &right, &rbp); + error = xfs_btree_read_buf_block(cur, &rptr, 0, &right, &rbp); if (error) goto error0; @@ -3740,8 +3737,7 @@ xfs_btree_delrec( /* If there is a right sibling, point it to the remaining block. */ xfs_btree_get_sibling(cur, left, &cptr, XFS_BB_RIGHTSIB); if (!xfs_btree_ptr_is_null(cur, &cptr)) { - error = xfs_btree_read_buf_block(cur, &cptr, level, - 0, &rrblock, &rrbp); + error = xfs_btree_read_buf_block(cur, &cptr, 0, &rrblock, &rrbp); if (error) goto error0; xfs_btree_set_sibling(cur, rrblock, &lptr, XFS_BB_LEFTSIB); diff --git a/fs/xfs/xfs_btree.h b/fs/xfs/xfs_btree.h index 91e34f21bace..a04b69422f67 100644 --- a/fs/xfs/xfs_btree.h +++ b/fs/xfs/xfs_btree.h @@ -62,6 +62,7 @@ union xfs_btree_rec { #define XFS_BTNUM_CNT ((xfs_btnum_t)XFS_BTNUM_CNTi) #define XFS_BTNUM_BMAP ((xfs_btnum_t)XFS_BTNUM_BMAPi) #define XFS_BTNUM_INO ((xfs_btnum_t)XFS_BTNUM_INOi) +#define XFS_BTNUM_FINO ((xfs_btnum_t)XFS_BTNUM_FINOi) /* * For logging record fields. @@ -92,6 +93,7 @@ do { \ case XFS_BTNUM_CNT: __XFS_BTREE_STATS_INC(abtc, stat); break; \ case XFS_BTNUM_BMAP: __XFS_BTREE_STATS_INC(bmbt, stat); break; \ case XFS_BTNUM_INO: __XFS_BTREE_STATS_INC(ibt, stat); break; \ + case XFS_BTNUM_FINO: __XFS_BTREE_STATS_INC(fibt, stat); break; \ case XFS_BTNUM_MAX: ASSERT(0); /* fucking gcc */ ; break; \ } \ } while (0) @@ -105,6 +107,7 @@ do { \ case XFS_BTNUM_CNT: __XFS_BTREE_STATS_ADD(abtc, stat, val); break; \ case XFS_BTNUM_BMAP: __XFS_BTREE_STATS_ADD(bmbt, stat, val); break; \ case XFS_BTNUM_INO: __XFS_BTREE_STATS_ADD(ibt, stat, val); break; \ + case XFS_BTNUM_FINO: __XFS_BTREE_STATS_ADD(fibt, stat, val); break; \ case XFS_BTNUM_MAX: ASSERT(0); /* fucking gcc */ ; break; \ } \ } while (0) @@ -129,7 +132,7 @@ struct xfs_btree_ops { int (*alloc_block)(struct xfs_btree_cur *cur, union xfs_btree_ptr *start_bno, union xfs_btree_ptr *new_bno, - int length, int *stat); + int *stat); int (*free_block)(struct xfs_btree_cur *cur, struct xfs_buf *bp); /* update last record information */ diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index 107f2fdfe41f..7a34a1ae6552 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -216,8 +216,7 @@ _xfs_buf_alloc( STATIC int _xfs_buf_get_pages( xfs_buf_t *bp, - int page_count, - xfs_buf_flags_t flags) + int page_count) { /* Make sure that we have a page list */ if (bp->b_pages == NULL) { @@ -330,7 +329,7 @@ use_alloc_page: end = (BBTOB(bp->b_maps[0].bm_bn + bp->b_length) + PAGE_SIZE - 1) >> PAGE_SHIFT; page_count = end - start; - error = _xfs_buf_get_pages(bp, page_count, flags); + error = _xfs_buf_get_pages(bp, page_count); if (unlikely(error)) return error; @@ -778,7 +777,7 @@ xfs_buf_associate_memory( bp->b_pages = NULL; bp->b_addr = mem; - rval = _xfs_buf_get_pages(bp, page_count, 0); + rval = _xfs_buf_get_pages(bp, page_count); if (rval) return rval; @@ -811,7 +810,7 @@ xfs_buf_get_uncached( goto fail; page_count = PAGE_ALIGN(numblks << BBSHIFT) >> PAGE_SHIFT; - error = _xfs_buf_get_pages(bp, page_count, 0); + error = _xfs_buf_get_pages(bp, page_count); if (error) goto fail_free_buf; @@ -1372,21 +1371,29 @@ xfs_buf_iorequest( xfs_buf_wait_unpin(bp); xfs_buf_hold(bp); - /* Set the count to 1 initially, this will stop an I/O + /* + * Set the count to 1 initially, this will stop an I/O * completion callout which happens before we have started * all the I/O from calling xfs_buf_ioend too early. */ atomic_set(&bp->b_io_remaining, 1); _xfs_buf_ioapply(bp); - _xfs_buf_ioend(bp, 1); + /* + * If _xfs_buf_ioapply failed, we'll get back here with + * only the reference we took above. _xfs_buf_ioend will + * drop it to zero, so we'd better not queue it for later, + * or we'll free it before it's done. + */ + _xfs_buf_ioend(bp, bp->b_error ? 0 : 1); xfs_buf_rele(bp); } /* * Waits for I/O to complete on the buffer supplied. It returns immediately if - * no I/O is pending or there is already a pending error on the buffer. It - * returns the I/O error code, if any, or 0 if there was no error. + * no I/O is pending or there is already a pending error on the buffer, in which + * case nothing will ever complete. It returns the I/O error code, if any, or + * 0 if there was no error. */ int xfs_buf_iowait( @@ -1607,7 +1614,6 @@ xfs_free_buftarg( int xfs_setsize_buftarg( xfs_buftarg_t *btp, - unsigned int blocksize, unsigned int sectorsize) { /* Set up metadata sector size info */ @@ -1642,16 +1648,13 @@ xfs_setsize_buftarg_early( xfs_buftarg_t *btp, struct block_device *bdev) { - return xfs_setsize_buftarg(btp, PAGE_SIZE, - bdev_logical_block_size(bdev)); + return xfs_setsize_buftarg(btp, bdev_logical_block_size(bdev)); } xfs_buftarg_t * xfs_alloc_buftarg( struct xfs_mount *mp, - struct block_device *bdev, - int external, - const char *fsname) + struct block_device *bdev) { xfs_buftarg_t *btp; diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h index b8a3abf6cf47..0e47fd1fedba 100644 --- a/fs/xfs/xfs_buf.h +++ b/fs/xfs/xfs_buf.h @@ -387,10 +387,10 @@ xfs_buf_update_cksum(struct xfs_buf *bp, unsigned long cksum_offset) * Handling of buftargs. */ extern xfs_buftarg_t *xfs_alloc_buftarg(struct xfs_mount *, - struct block_device *, int, const char *); + struct block_device *); extern void xfs_free_buftarg(struct xfs_mount *, struct xfs_buftarg *); extern void xfs_wait_buftarg(xfs_buftarg_t *); -extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int, unsigned int); +extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int); #define xfs_getsize_buftarg(buftarg) block_size((buftarg)->bt_bdev) #define xfs_readonly_buftarg(buftarg) bdev_read_only((buftarg)->bt_bdev) diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index 8752821443be..64b17f5bed9a 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -812,7 +812,6 @@ xfs_buf_item_init( */ static void xfs_buf_item_log_segment( - struct xfs_buf_log_item *bip, uint first, uint last, uint *map) @@ -920,7 +919,7 @@ xfs_buf_item_log( if (end > last) end = last; - xfs_buf_item_log_segment(bip, first, end, + xfs_buf_item_log_segment(first, end, &bip->bli_formats[i].blf_data_map[0]); start += bp->b_maps[i].bm_len; diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c index 6cc5f6785a77..9eec594cc25a 100644 --- a/fs/xfs/xfs_da_btree.c +++ b/fs/xfs/xfs_da_btree.c @@ -2462,7 +2462,6 @@ xfs_buf_map_from_irec( */ static int xfs_dabuf_map( - struct xfs_trans *trans, struct xfs_inode *dp, xfs_dablk_t bno, xfs_daddr_t mappedbno, @@ -2558,7 +2557,7 @@ xfs_da_get_buf( *bpp = NULL; mapp = ↦ nmap = 1; - error = xfs_dabuf_map(trans, dp, bno, mappedbno, whichfork, + error = xfs_dabuf_map(dp, bno, mappedbno, whichfork, &mapp, &nmap); if (error) { /* mapping a hole is not an error, but we don't continue */ @@ -2606,7 +2605,7 @@ xfs_da_read_buf( *bpp = NULL; mapp = ↦ nmap = 1; - error = xfs_dabuf_map(trans, dp, bno, mappedbno, whichfork, + error = xfs_dabuf_map(dp, bno, mappedbno, whichfork, &mapp, &nmap); if (error) { /* mapping a hole is not an error, but we don't continue */ @@ -2679,7 +2678,6 @@ out_free: */ xfs_daddr_t xfs_da_reada_buf( - struct xfs_trans *trans, struct xfs_inode *dp, xfs_dablk_t bno, xfs_daddr_t mappedbno, @@ -2693,7 +2691,7 @@ xfs_da_reada_buf( mapp = ↦ nmap = 1; - error = xfs_dabuf_map(trans, dp, bno, mappedbno, whichfork, + error = xfs_dabuf_map(dp, bno, mappedbno, whichfork, &mapp, &nmap); if (error) { /* mapping a hole is not an error, but we don't continue */ diff --git a/fs/xfs/xfs_da_btree.h b/fs/xfs/xfs_da_btree.h index 6e95ea79f5d7..c824a0aa039f 100644 --- a/fs/xfs/xfs_da_btree.h +++ b/fs/xfs/xfs_da_btree.h @@ -60,10 +60,12 @@ typedef struct xfs_da_args { int index; /* index of attr of interest in blk */ xfs_dablk_t rmtblkno; /* remote attr value starting blkno */ int rmtblkcnt; /* remote attr value block count */ + int rmtvaluelen; /* remote attr value length in bytes */ xfs_dablk_t blkno2; /* blkno of 2nd attr leaf of interest */ int index2; /* index of 2nd attr in blk */ xfs_dablk_t rmtblkno2; /* remote attr value starting blkno */ int rmtblkcnt2; /* remote attr value block count */ + int rmtvaluelen2; /* remote attr value length in bytes */ int op_flags; /* operation flags */ enum xfs_dacmp cmpresult; /* name compare result for lookups */ } xfs_da_args_t; @@ -183,9 +185,9 @@ int xfs_da_read_buf(struct xfs_trans *trans, struct xfs_inode *dp, xfs_dablk_t bno, xfs_daddr_t mappedbno, struct xfs_buf **bpp, int whichfork, const struct xfs_buf_ops *ops); -xfs_daddr_t xfs_da_reada_buf(struct xfs_trans *trans, struct xfs_inode *dp, - xfs_dablk_t bno, xfs_daddr_t mapped_bno, - int whichfork, const struct xfs_buf_ops *ops); +xfs_daddr_t xfs_da_reada_buf(struct xfs_inode *dp, xfs_dablk_t bno, + xfs_daddr_t mapped_bno, int whichfork, + const struct xfs_buf_ops *ops); int xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno, struct xfs_buf *dead_buf); diff --git a/fs/xfs/xfs_da_format.h b/fs/xfs/xfs_da_format.h index a19d3f8f639c..1432b576b4a7 100644 --- a/fs/xfs/xfs_da_format.h +++ b/fs/xfs/xfs_da_format.h @@ -541,7 +541,7 @@ xfs_dir2_leaf_bests_p(struct xfs_dir2_leaf_tail *ltp) * Convert dataptr to byte in file space */ static inline xfs_dir2_off_t -xfs_dir2_dataptr_to_byte(struct xfs_mount *mp, xfs_dir2_dataptr_t dp) +xfs_dir2_dataptr_to_byte(xfs_dir2_dataptr_t dp) { return (xfs_dir2_off_t)dp << XFS_DIR2_DATA_ALIGN_LOG; } @@ -550,7 +550,7 @@ xfs_dir2_dataptr_to_byte(struct xfs_mount *mp, xfs_dir2_dataptr_t dp) * Convert byte in file space to dataptr. It had better be aligned. */ static inline xfs_dir2_dataptr_t -xfs_dir2_byte_to_dataptr(struct xfs_mount *mp, xfs_dir2_off_t by) +xfs_dir2_byte_to_dataptr(xfs_dir2_off_t by) { return (xfs_dir2_dataptr_t)(by >> XFS_DIR2_DATA_ALIGN_LOG); } @@ -571,7 +571,7 @@ xfs_dir2_byte_to_db(struct xfs_mount *mp, xfs_dir2_off_t by) static inline xfs_dir2_db_t xfs_dir2_dataptr_to_db(struct xfs_mount *mp, xfs_dir2_dataptr_t dp) { - return xfs_dir2_byte_to_db(mp, xfs_dir2_dataptr_to_byte(mp, dp)); + return xfs_dir2_byte_to_db(mp, xfs_dir2_dataptr_to_byte(dp)); } /* @@ -590,7 +590,7 @@ xfs_dir2_byte_to_off(struct xfs_mount *mp, xfs_dir2_off_t by) static inline xfs_dir2_data_aoff_t xfs_dir2_dataptr_to_off(struct xfs_mount *mp, xfs_dir2_dataptr_t dp) { - return xfs_dir2_byte_to_off(mp, xfs_dir2_dataptr_to_byte(mp, dp)); + return xfs_dir2_byte_to_off(mp, xfs_dir2_dataptr_to_byte(dp)); } /* @@ -629,7 +629,7 @@ static inline xfs_dir2_dataptr_t xfs_dir2_db_off_to_dataptr(struct xfs_mount *mp, xfs_dir2_db_t db, xfs_dir2_data_aoff_t o) { - return xfs_dir2_byte_to_dataptr(mp, xfs_dir2_db_off_to_byte(mp, db, o)); + return xfs_dir2_byte_to_dataptr(xfs_dir2_db_off_to_byte(mp, db, o)); } /* diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c index fda46253966a..e365c98c0f1e 100644 --- a/fs/xfs/xfs_dir2.c +++ b/fs/xfs/xfs_dir2.c @@ -244,7 +244,7 @@ xfs_dir_createname( goto out_free; } - rval = xfs_dir2_isblock(tp, dp, &v); + rval = xfs_dir2_isblock(dp, &v); if (rval) goto out_free; if (v) { @@ -252,7 +252,7 @@ xfs_dir_createname( goto out_free; } - rval = xfs_dir2_isleaf(tp, dp, &v); + rval = xfs_dir2_isleaf(dp, &v); if (rval) goto out_free; if (v) @@ -336,7 +336,7 @@ xfs_dir_lookup( goto out_check_rval; } - rval = xfs_dir2_isblock(tp, dp, &v); + rval = xfs_dir2_isblock(dp, &v); if (rval) goto out_free; if (v) { @@ -344,7 +344,7 @@ xfs_dir_lookup( goto out_check_rval; } - rval = xfs_dir2_isleaf(tp, dp, &v); + rval = xfs_dir2_isleaf(dp, &v); if (rval) goto out_free; if (v) @@ -408,7 +408,7 @@ xfs_dir_removename( goto out_free; } - rval = xfs_dir2_isblock(tp, dp, &v); + rval = xfs_dir2_isblock(dp, &v); if (rval) goto out_free; if (v) { @@ -416,7 +416,7 @@ xfs_dir_removename( goto out_free; } - rval = xfs_dir2_isleaf(tp, dp, &v); + rval = xfs_dir2_isleaf(dp, &v); if (rval) goto out_free; if (v) @@ -472,7 +472,7 @@ xfs_dir_replace( goto out_free; } - rval = xfs_dir2_isblock(tp, dp, &v); + rval = xfs_dir2_isblock(dp, &v); if (rval) goto out_free; if (v) { @@ -480,7 +480,7 @@ xfs_dir_replace( goto out_free; } - rval = xfs_dir2_isleaf(tp, dp, &v); + rval = xfs_dir2_isleaf(dp, &v); if (rval) goto out_free; if (v) @@ -531,7 +531,7 @@ xfs_dir_canenter( goto out_free; } - rval = xfs_dir2_isblock(tp, dp, &v); + rval = xfs_dir2_isblock(dp, &v); if (rval) goto out_free; if (v) { @@ -539,7 +539,7 @@ xfs_dir_canenter( goto out_free; } - rval = xfs_dir2_isleaf(tp, dp, &v); + rval = xfs_dir2_isleaf(dp, &v); if (rval) goto out_free; if (v) @@ -607,7 +607,6 @@ xfs_dir2_grow_inode( */ int xfs_dir2_isblock( - xfs_trans_t *tp, xfs_inode_t *dp, int *vp) /* out: 1 is block, 0 is not block */ { @@ -616,7 +615,7 @@ xfs_dir2_isblock( int rval; mp = dp->i_mount; - if ((rval = xfs_bmap_last_offset(tp, dp, &last, XFS_DATA_FORK))) + if ((rval = xfs_bmap_last_offset(dp, &last, XFS_DATA_FORK))) return rval; rval = XFS_FSB_TO_B(mp, last) == mp->m_dirblksize; ASSERT(rval == 0 || dp->i_d.di_size == mp->m_dirblksize); @@ -629,7 +628,6 @@ xfs_dir2_isblock( */ int xfs_dir2_isleaf( - xfs_trans_t *tp, xfs_inode_t *dp, int *vp) /* out: 1 is leaf, 0 is not leaf */ { @@ -638,7 +636,7 @@ xfs_dir2_isleaf( int rval; mp = dp->i_mount; - if ((rval = xfs_bmap_last_offset(tp, dp, &last, XFS_DATA_FORK))) + if ((rval = xfs_bmap_last_offset(dp, &last, XFS_DATA_FORK))) return rval; *vp = last == mp->m_dirleafblk + (1 << mp->m_sb.sb_dirblklog); return 0; diff --git a/fs/xfs/xfs_dir2.h b/fs/xfs/xfs_dir2.h index cec70e0781ab..64a6b19c2fd0 100644 --- a/fs/xfs/xfs_dir2.h +++ b/fs/xfs/xfs_dir2.h @@ -142,8 +142,8 @@ extern int xfs_dir2_sf_to_block(struct xfs_da_args *args); /* * Interface routines used by userspace utilities */ -extern int xfs_dir2_isblock(struct xfs_trans *tp, struct xfs_inode *dp, int *r); -extern int xfs_dir2_isleaf(struct xfs_trans *tp, struct xfs_inode *dp, int *r); +extern int xfs_dir2_isblock(struct xfs_inode *dp, int *r); +extern int xfs_dir2_isleaf(struct xfs_inode *dp, int *r); extern int xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db, struct xfs_buf *bp); diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c index 4f6a38cb83a4..dd9d00515582 100644 --- a/fs/xfs/xfs_dir2_block.c +++ b/fs/xfs/xfs_dir2_block.c @@ -319,7 +319,6 @@ xfs_dir2_block_compact( (xfs_dir2_data_aoff_t)((char *)blp - (char *)hdr), (xfs_dir2_data_aoff_t)((be32_to_cpu(btp->stale) - 1) * sizeof(*blp)), needlog, &needscan); - blp += be32_to_cpu(btp->stale) - 1; btp->stale = cpu_to_be32(1); /* * If we now need to rebuild the bestfree map, do so. @@ -537,7 +536,7 @@ xfs_dir2_block_addname( * Fill in the leaf entry. */ blp[mid].hashval = cpu_to_be32(args->hashval); - blp[mid].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp, + blp[mid].address = cpu_to_be32(xfs_dir2_byte_to_dataptr( (char *)dep - (char *)hdr)); xfs_dir2_block_log_leaf(tp, bp, lfloglow, lfloghigh); /* @@ -1170,7 +1169,7 @@ xfs_dir2_sf_to_block( *tagp = cpu_to_be16((char *)dep - (char *)hdr); xfs_dir2_data_log_entry(tp, dp, bp, dep); blp[0].hashval = cpu_to_be32(xfs_dir_hash_dot); - blp[0].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp, + blp[0].address = cpu_to_be32(xfs_dir2_byte_to_dataptr( (char *)dep - (char *)hdr)); /* * Create entry for .. @@ -1184,7 +1183,7 @@ xfs_dir2_sf_to_block( *tagp = cpu_to_be16((char *)dep - (char *)hdr); xfs_dir2_data_log_entry(tp, dp, bp, dep); blp[1].hashval = cpu_to_be32(xfs_dir_hash_dotdot); - blp[1].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp, + blp[1].address = cpu_to_be32(xfs_dir2_byte_to_dataptr( (char *)dep - (char *)hdr)); offset = dp->d_ops->data_first_offset; /* @@ -1238,7 +1237,7 @@ xfs_dir2_sf_to_block( name.len = sfep->namelen; blp[2 + i].hashval = cpu_to_be32(mp->m_dirnameops-> hashname(&name)); - blp[2 + i].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp, + blp[2 + i].address = cpu_to_be32(xfs_dir2_byte_to_dataptr( (char *)dep - (char *)hdr)); offset = (int)((char *)(tagp + 1) - (char *)hdr); if (++i == sfp->count) diff --git a/fs/xfs/xfs_dir2_data.c b/fs/xfs/xfs_dir2_data.c index afa4ad523f3f..bae8b5b8d1c2 100644 --- a/fs/xfs/xfs_dir2_data.c +++ b/fs/xfs/xfs_dir2_data.c @@ -329,12 +329,11 @@ xfs_dir3_data_read( int xfs_dir3_data_readahead( - struct xfs_trans *tp, struct xfs_inode *dp, xfs_dablk_t bno, xfs_daddr_t mapped_bno) { - return xfs_da_reada_buf(tp, dp, bno, mapped_bno, + return xfs_da_reada_buf(dp, bno, mapped_bno, XFS_DATA_FORK, &xfs_dir3_data_reada_buf_ops); } diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c index d36e97df1187..f571723e2378 100644 --- a/fs/xfs/xfs_dir2_leaf.c +++ b/fs/xfs/xfs_dir2_leaf.c @@ -1708,7 +1708,7 @@ xfs_dir2_node_to_leaf( /* * Get the last offset in the file. */ - if ((error = xfs_bmap_last_offset(tp, dp, &fo, XFS_DATA_FORK))) { + if ((error = xfs_bmap_last_offset(dp, &fo, XFS_DATA_FORK))) { return error; } fo -= mp->m_dirblkfsbs; diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c index cb434d732681..9cb91ee0914b 100644 --- a/fs/xfs/xfs_dir2_node.c +++ b/fs/xfs/xfs_dir2_node.c @@ -1727,7 +1727,7 @@ xfs_dir2_node_addname_int( if (dbno == -1) { xfs_fileoff_t fo; /* freespace block number */ - if ((error = xfs_bmap_last_offset(tp, dp, &fo, XFS_DATA_FORK))) + if ((error = xfs_bmap_last_offset(dp, &fo, XFS_DATA_FORK))) return error; lastfbno = xfs_dir2_da_to_db(mp, (xfs_dablk_t)fo); fbno = ifbno; diff --git a/fs/xfs/xfs_dir2_priv.h b/fs/xfs/xfs_dir2_priv.h index 8b9d2281f85b..2429960739e9 100644 --- a/fs/xfs/xfs_dir2_priv.h +++ b/fs/xfs/xfs_dir2_priv.h @@ -54,8 +54,8 @@ extern int xfs_dir2_leaf_to_block(struct xfs_da_args *args, extern int __xfs_dir3_data_check(struct xfs_inode *dp, struct xfs_buf *bp); extern int xfs_dir3_data_read(struct xfs_trans *tp, struct xfs_inode *dp, xfs_dablk_t bno, xfs_daddr_t mapped_bno, struct xfs_buf **bpp); -extern int xfs_dir3_data_readahead(struct xfs_trans *tp, struct xfs_inode *dp, - xfs_dablk_t bno, xfs_daddr_t mapped_bno); +extern int xfs_dir3_data_readahead(struct xfs_inode *dp, xfs_dablk_t bno, + xfs_daddr_t mapped_bno); extern struct xfs_dir2_data_free * xfs_dir2_data_freeinsert(struct xfs_dir2_data_hdr *hdr, diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c index aead369e1c30..bf7a5cee7adc 100644 --- a/fs/xfs/xfs_dir2_readdir.c +++ b/fs/xfs/xfs_dir2_readdir.c @@ -434,7 +434,7 @@ xfs_dir2_leaf_readbuf( */ if (i > mip->ra_current && map[mip->ra_index].br_blockcount >= mp->m_dirblkfsbs) { - xfs_dir3_data_readahead(NULL, dp, + xfs_dir3_data_readahead(dp, map[mip->ra_index].br_startoff + mip->ra_offset, XFS_FSB_TO_DADDR(mp, map[mip->ra_index].br_startblock + @@ -447,7 +447,7 @@ xfs_dir2_leaf_readbuf( * use our mapping, but this is a very rare case. */ else if (i > mip->ra_current) { - xfs_dir3_data_readahead(NULL, dp, + xfs_dir3_data_readahead(dp, map[mip->ra_index].br_startoff + mip->ra_offset, -1); mip->ra_current = i; @@ -456,7 +456,7 @@ xfs_dir2_leaf_readbuf( /* * Advance offset through the mapping table. */ - for (j = 0; j < mp->m_dirblkfsbs; j++) { + for (j = 0; j < mp->m_dirblkfsbs; j += length ) { /* * The rest of this extent but not more than a dir * block. @@ -464,7 +464,6 @@ xfs_dir2_leaf_readbuf( length = min_t(int, mp->m_dirblkfsbs, map[mip->ra_index].br_blockcount - mip->ra_offset); - j += length; mip->ra_offset += length; /* @@ -531,7 +530,7 @@ xfs_dir2_leaf_getdents( * Inside the loop we keep the main offset value as a byte offset * in the directory file. */ - curoff = xfs_dir2_dataptr_to_byte(mp, ctx->pos); + curoff = xfs_dir2_dataptr_to_byte(ctx->pos); /* * Force this conversion through db so we truncate the offset @@ -635,7 +634,7 @@ xfs_dir2_leaf_getdents( length = dp->d_ops->data_entsize(dep->namelen); filetype = dp->d_ops->data_get_ftype(dep); - ctx->pos = xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff; + ctx->pos = xfs_dir2_byte_to_dataptr(curoff) & 0x7fffffff; if (!dir_emit(ctx, (char *)dep->name, dep->namelen, be64_to_cpu(dep->inumber), xfs_dir3_get_dtype(mp, filetype))) @@ -653,10 +652,10 @@ 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)) + if (curoff > xfs_dir2_dataptr_to_byte(XFS_DIR2_MAX_DATAPTR)) ctx->pos = XFS_DIR2_MAX_DATAPTR & 0x7fffffff; else - ctx->pos = xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff; + ctx->pos = xfs_dir2_byte_to_dataptr(curoff) & 0x7fffffff; kmem_free(map_info); if (bp) xfs_trans_brelse(NULL, bp); @@ -687,7 +686,7 @@ xfs_readdir( lock_mode = xfs_ilock_data_map_shared(dp); if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) rval = xfs_dir2_sf_getdents(dp, ctx); - else if ((rval = xfs_dir2_isblock(NULL, dp, &v))) + else if ((rval = xfs_dir2_isblock(dp, &v))) ; else if (v) rval = xfs_dir2_block_getdents(dp, ctx); diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/xfs_dir2_sf.c index 3725fb1b902b..7aab8ec117ad 100644 --- a/fs/xfs/xfs_dir2_sf.c +++ b/fs/xfs/xfs_dir2_sf.c @@ -285,14 +285,12 @@ int /* error */ xfs_dir2_sf_addname( xfs_da_args_t *args) /* operation arguments */ { - int add_entsize; /* size of the new entry */ xfs_inode_t *dp; /* incore directory inode */ int error; /* error return value */ int incr_isize; /* total change in size */ int new_isize; /* di_size after adding name */ int objchange; /* changing to 8-byte inodes */ xfs_dir2_data_aoff_t offset = 0; /* offset for new entry */ - int old_isize; /* di_size before adding name */ int pick; /* which algorithm to use */ xfs_dir2_sf_hdr_t *sfp; /* shortform structure */ xfs_dir2_sf_entry_t *sfep = NULL; /* shortform entry */ @@ -316,8 +314,7 @@ xfs_dir2_sf_addname( /* * Compute entry (and change in) size. */ - add_entsize = dp->d_ops->sf_entsize(sfp, args->namelen); - incr_isize = add_entsize; + incr_isize = dp->d_ops->sf_entsize(sfp, args->namelen); objchange = 0; #if XFS_BIG_INUMS /* @@ -325,11 +322,8 @@ xfs_dir2_sf_addname( */ if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && sfp->i8count == 0) { /* - * Yes, adjust the entry size and the total size. + * Yes, adjust the inode size. old count + (parent + new) */ - add_entsize += - (uint)sizeof(xfs_dir2_ino8_t) - - (uint)sizeof(xfs_dir2_ino4_t); incr_isize += (sfp->count + 2) * ((uint)sizeof(xfs_dir2_ino8_t) - @@ -337,8 +331,7 @@ xfs_dir2_sf_addname( objchange = 1; } #endif - old_isize = (int)dp->i_d.di_size; - new_isize = old_isize + incr_isize; + new_isize = (int)dp->i_d.di_size + incr_isize; /* * Won't fit as shortform any more (due to size), * or the pick routine says it won't (due to offset values). @@ -1110,9 +1103,9 @@ xfs_dir2_sf_toino4( } /* - * Convert from 4-byte inode numbers to 8-byte inode numbers. - * The new 8-byte inode number is not there yet, we leave with the - * count 1 but no corresponding entry. + * Convert existing entries from 4-byte inode numbers to 8-byte inode numbers. + * The new entry w/ an 8-byte inode number is not there yet; we leave with + * i8count set to 1, but no corresponding 8-byte entry. */ static void xfs_dir2_sf_toino8( @@ -1145,7 +1138,7 @@ xfs_dir2_sf_toino8( ASSERT(oldsfp->i8count == 0); memcpy(buf, oldsfp, oldsize); /* - * Compute the new inode size. + * Compute the new inode size (nb: entry count + 1 for parent) */ newsize = oldsize + diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index 868b19f096bf..5fec738f1f2e 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -832,47 +832,6 @@ restart: return (0); } - -STATIC void -xfs_qm_dqput_final( - struct xfs_dquot *dqp) -{ - struct xfs_quotainfo *qi = dqp->q_mount->m_quotainfo; - struct xfs_dquot *gdqp; - struct xfs_dquot *pdqp; - - trace_xfs_dqput_free(dqp); - - if (list_lru_add(&qi->qi_lru, &dqp->q_lru)) - XFS_STATS_INC(xs_qm_dquot_unused); - - /* - * If we just added a udquot to the freelist, then we want to release - * the gdquot/pdquot reference that it (probably) has. Otherwise it'll - * keep the gdquot/pdquot from getting reclaimed. - */ - gdqp = dqp->q_gdquot; - if (gdqp) { - xfs_dqlock(gdqp); - dqp->q_gdquot = NULL; - } - - pdqp = dqp->q_pdquot; - if (pdqp) { - xfs_dqlock(pdqp); - dqp->q_pdquot = NULL; - } - xfs_dqunlock(dqp); - - /* - * If we had a group/project quota hint, release it now. - */ - if (gdqp) - xfs_qm_dqput(gdqp); - if (pdqp) - xfs_qm_dqput(pdqp); -} - /* * Release a reference to the dquot (decrement ref-count) and unlock it. * @@ -888,10 +847,14 @@ xfs_qm_dqput( trace_xfs_dqput(dqp); - if (--dqp->q_nrefs > 0) - xfs_dqunlock(dqp); - else - xfs_qm_dqput_final(dqp); + if (--dqp->q_nrefs == 0) { + struct xfs_quotainfo *qi = dqp->q_mount->m_quotainfo; + trace_xfs_dqput_free(dqp); + + if (list_lru_add(&qi->qi_lru, &dqp->q_lru)) + XFS_STATS_INC(xs_qm_dquot_unused); + } + xfs_dqunlock(dqp); } /* diff --git a/fs/xfs/xfs_dquot.h b/fs/xfs/xfs_dquot.h index d22ed0053c32..68a68f704837 100644 --- a/fs/xfs/xfs_dquot.h +++ b/fs/xfs/xfs_dquot.h @@ -52,8 +52,6 @@ typedef struct xfs_dquot { int q_bufoffset; /* off of dq in buffer (# dquots) */ xfs_fileoff_t q_fileoffset; /* offset in quotas file */ - struct xfs_dquot*q_gdquot; /* group dquot, hint only */ - struct xfs_dquot*q_pdquot; /* project dquot, hint only */ xfs_disk_dquot_t q_core; /* actual usage & quotas */ xfs_dq_logitem_t q_logitem; /* dquot log item */ xfs_qcnt_t q_res_bcount; /* total regular nblks used+reserved */ diff --git a/fs/xfs/xfs_dquot_buf.c b/fs/xfs/xfs_dquot_buf.c index 610da8177737..c2ac0c611ad8 100644 --- a/fs/xfs/xfs_dquot_buf.c +++ b/fs/xfs/xfs_dquot_buf.c @@ -35,7 +35,6 @@ int xfs_calc_dquots_per_chunk( - struct xfs_mount *mp, unsigned int nbblks) /* basic block units */ { unsigned int ndquots; @@ -194,7 +193,7 @@ xfs_dquot_buf_verify_crc( if (mp->m_quotainfo) ndquots = mp->m_quotainfo->qi_dqperchunk; else - ndquots = xfs_calc_dquots_per_chunk(mp, + ndquots = xfs_calc_dquots_per_chunk( XFS_BB_TO_FSB(mp, bp->b_length)); for (i = 0; i < ndquots; i++, d++) { @@ -225,7 +224,7 @@ xfs_dquot_buf_verify( if (mp->m_quotainfo) ndquots = mp->m_quotainfo->qi_dqperchunk; else - ndquots = xfs_calc_dquots_per_chunk(mp, bp->b_length); + ndquots = xfs_calc_dquots_per_chunk(bp->b_length); /* * On the first read of the buffer, verify that each dquot is valid. diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c index 1399e187d425..753e467aa1a5 100644 --- a/fs/xfs/xfs_export.c +++ b/fs/xfs/xfs_export.c @@ -237,7 +237,7 @@ xfs_fs_nfs_commit_metadata( if (!lsn) return 0; - return _xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL); + return -_xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL); } const struct export_operations xfs_export_operations = { diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 79e96ce98733..1b8160dc04d1 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -155,7 +155,7 @@ xfs_dir_fsync( if (!lsn) return 0; - return _xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL); + return -_xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL); } STATIC int @@ -295,7 +295,7 @@ xfs_file_aio_read( xfs_rw_ilock(ip, XFS_IOLOCK_EXCL); if (inode->i_mapping->nrpages) { - ret = -filemap_write_and_wait_range( + ret = filemap_write_and_wait_range( VFS_I(ip)->i_mapping, pos, -1); if (ret) { @@ -679,7 +679,7 @@ xfs_file_dio_aio_write( goto out; if (mapping->nrpages) { - ret = -filemap_write_and_wait_range(VFS_I(ip)->i_mapping, + ret = filemap_write_and_wait_range(VFS_I(ip)->i_mapping, pos, -1); if (ret) goto out; @@ -837,11 +837,19 @@ xfs_file_fallocate( unsigned blksize_mask = (1 << inode->i_blkbits) - 1; if (offset & blksize_mask || len & blksize_mask) { - error = -EINVAL; + error = EINVAL; + goto out_unlock; + } + + /* + * There is no need to overlap collapse range with EOF, + * in which case it is effectively a truncate operation + */ + if (offset + len >= i_size_read(inode)) { + error = EINVAL; goto out_unlock; } - ASSERT(offset + len < i_size_read(inode)); new_size = i_size_read(inode) - len; error = xfs_collapse_file_space(ip, offset, len); @@ -936,7 +944,7 @@ xfs_dir_open( */ mode = xfs_ilock_data_map_shared(ip); if (ip->i_d.di_nextents > 0) - xfs_dir3_data_readahead(NULL, ip, 0, -1); + xfs_dir3_data_readahead(ip, 0, -1); xfs_iunlock(ip, mode); return 0; } diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 12b6e7701985..8ec81bed7992 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -1,5 +1,6 @@ /* * Copyright (c) 2006-2007 Silicon Graphics, Inc. + * Copyright (c) 2014 Christoph Hellwig. * All Rights Reserved. * * This program is free software; you can redistribute it and/or @@ -32,100 +33,20 @@ #include "xfs_filestream.h" #include "xfs_trace.h" -#ifdef XFS_FILESTREAMS_TRACE - -ktrace_t *xfs_filestreams_trace_buf; - -STATIC void -xfs_filestreams_trace( - xfs_mount_t *mp, /* mount point */ - int type, /* type of trace */ - const char *func, /* source function */ - int line, /* source line number */ - __psunsigned_t arg0, - __psunsigned_t arg1, - __psunsigned_t arg2, - __psunsigned_t arg3, - __psunsigned_t arg4, - __psunsigned_t arg5) -{ - ktrace_enter(xfs_filestreams_trace_buf, - (void *)(__psint_t)(type | (line << 16)), - (void *)func, - (void *)(__psunsigned_t)current_pid(), - (void *)mp, - (void *)(__psunsigned_t)arg0, - (void *)(__psunsigned_t)arg1, - (void *)(__psunsigned_t)arg2, - (void *)(__psunsigned_t)arg3, - (void *)(__psunsigned_t)arg4, - (void *)(__psunsigned_t)arg5, - NULL, NULL, NULL, NULL, NULL, NULL); -} - -#define TRACE0(mp,t) TRACE6(mp,t,0,0,0,0,0,0) -#define TRACE1(mp,t,a0) TRACE6(mp,t,a0,0,0,0,0,0) -#define TRACE2(mp,t,a0,a1) TRACE6(mp,t,a0,a1,0,0,0,0) -#define TRACE3(mp,t,a0,a1,a2) TRACE6(mp,t,a0,a1,a2,0,0,0) -#define TRACE4(mp,t,a0,a1,a2,a3) TRACE6(mp,t,a0,a1,a2,a3,0,0) -#define TRACE5(mp,t,a0,a1,a2,a3,a4) TRACE6(mp,t,a0,a1,a2,a3,a4,0) -#define TRACE6(mp,t,a0,a1,a2,a3,a4,a5) \ - xfs_filestreams_trace(mp, t, __func__, __LINE__, \ - (__psunsigned_t)a0, (__psunsigned_t)a1, \ - (__psunsigned_t)a2, (__psunsigned_t)a3, \ - (__psunsigned_t)a4, (__psunsigned_t)a5) - -#define TRACE_AG_SCAN(mp, ag, ag2) \ - TRACE2(mp, XFS_FSTRM_KTRACE_AGSCAN, ag, ag2); -#define TRACE_AG_PICK1(mp, max_ag, maxfree) \ - TRACE2(mp, XFS_FSTRM_KTRACE_AGPICK1, max_ag, maxfree); -#define TRACE_AG_PICK2(mp, ag, ag2, cnt, free, scan, flag) \ - TRACE6(mp, XFS_FSTRM_KTRACE_AGPICK2, ag, ag2, \ - cnt, free, scan, flag) -#define TRACE_UPDATE(mp, ip, ag, cnt, ag2, cnt2) \ - TRACE5(mp, XFS_FSTRM_KTRACE_UPDATE, ip, ag, cnt, ag2, cnt2) -#define TRACE_FREE(mp, ip, pip, ag, cnt) \ - TRACE4(mp, XFS_FSTRM_KTRACE_FREE, ip, pip, ag, cnt) -#define TRACE_LOOKUP(mp, ip, pip, ag, cnt) \ - TRACE4(mp, XFS_FSTRM_KTRACE_ITEM_LOOKUP, ip, pip, ag, cnt) -#define TRACE_ASSOCIATE(mp, ip, pip, ag, cnt) \ - TRACE4(mp, XFS_FSTRM_KTRACE_ASSOCIATE, ip, pip, ag, cnt) -#define TRACE_MOVEAG(mp, ip, pip, oag, ocnt, nag, ncnt) \ - TRACE6(mp, XFS_FSTRM_KTRACE_MOVEAG, ip, pip, oag, ocnt, nag, ncnt) -#define TRACE_ORPHAN(mp, ip, ag) \ - TRACE2(mp, XFS_FSTRM_KTRACE_ORPHAN, ip, ag); - - -#else -#define TRACE_AG_SCAN(mp, ag, ag2) -#define TRACE_AG_PICK1(mp, max_ag, maxfree) -#define TRACE_AG_PICK2(mp, ag, ag2, cnt, free, scan, flag) -#define TRACE_UPDATE(mp, ip, ag, cnt, ag2, cnt2) -#define TRACE_FREE(mp, ip, pip, ag, cnt) -#define TRACE_LOOKUP(mp, ip, pip, ag, cnt) -#define TRACE_ASSOCIATE(mp, ip, pip, ag, cnt) -#define TRACE_MOVEAG(mp, ip, pip, oag, ocnt, nag, ncnt) -#define TRACE_ORPHAN(mp, ip, ag) -#endif - -static kmem_zone_t *item_zone; +struct xfs_fstrm_item { + struct xfs_mru_cache_elem mru; + struct xfs_inode *ip; + xfs_agnumber_t ag; /* AG in use for this directory */ +}; -/* - * Structure for associating a file or a directory with an allocation group. - * The parent directory pointer is only needed for files, but since there will - * generally be vastly more files than directories in the cache, using the same - * data structure simplifies the code with very little memory overhead. - */ -typedef struct fstrm_item -{ - xfs_agnumber_t ag; /* AG currently in use for the file/directory. */ - xfs_inode_t *ip; /* inode self-pointer. */ - xfs_inode_t *pip; /* Parent directory inode pointer. */ -} fstrm_item_t; +enum xfs_fstrm_alloc { + XFS_PICK_USERDATA = 1, + XFS_PICK_LOWSPACE = 2, +}; /* * Allocation group filestream associations are tracked with per-ag atomic - * counters. These counters allow _xfs_filestream_pick_ag() to tell whether a + * counters. These counters allow xfs_filestream_pick_ag() to tell whether a * particular AG already has active filestreams associated with it. The mount * point's m_peraglock is used to protect these counters from per-ag array * re-allocation during a growfs operation. When xfs_growfs_data_private() is @@ -160,7 +81,7 @@ typedef struct fstrm_item * the cache that reference per-ag array elements that have since been * reallocated. */ -static int +int xfs_filestream_peek_ag( xfs_mount_t *mp, xfs_agnumber_t agno) @@ -200,23 +121,40 @@ xfs_filestream_put_ag( xfs_perag_put(pag); } +static void +xfs_fstrm_free_func( + struct xfs_mru_cache_elem *mru) +{ + struct xfs_fstrm_item *item = + container_of(mru, struct xfs_fstrm_item, mru); + + xfs_filestream_put_ag(item->ip->i_mount, item->ag); + + trace_xfs_filestream_free(item->ip, item->ag); + + kmem_free(item); +} + /* * Scan the AGs starting at startag looking for an AG that isn't in use and has * at least minlen blocks free. */ static int -_xfs_filestream_pick_ag( - xfs_mount_t *mp, - xfs_agnumber_t startag, - xfs_agnumber_t *agp, - int flags, - xfs_extlen_t minlen) +xfs_filestream_pick_ag( + struct xfs_inode *ip, + xfs_agnumber_t startag, + xfs_agnumber_t *agp, + int flags, + xfs_extlen_t minlen) { - int streams, max_streams; - int err, trylock, nscan; - xfs_extlen_t longest, free, minfree, maxfree = 0; - xfs_agnumber_t ag, max_ag = NULLAGNUMBER; - struct xfs_perag *pag; + struct xfs_mount *mp = ip->i_mount; + struct xfs_fstrm_item *item; + struct xfs_perag *pag; + xfs_extlen_t longest, free = 0, minfree, maxfree = 0; + xfs_agnumber_t ag, max_ag = NULLAGNUMBER; + int err, trylock, nscan; + + ASSERT(S_ISDIR(ip->i_d.di_mode)); /* 2% of an AG's blocks must be free for it to be chosen. */ minfree = mp->m_sb.sb_agblocks / 50; @@ -228,8 +166,9 @@ _xfs_filestream_pick_ag( trylock = XFS_ALLOC_FLAG_TRYLOCK; for (nscan = 0; 1; nscan++) { + trace_xfs_filestream_scan(ip, ag); + pag = xfs_perag_get(mp, ag); - TRACE_AG_SCAN(mp, ag, atomic_read(&pag->pagf_fstrms)); if (!pag->pagf_init) { err = xfs_alloc_pagf_init(mp, NULL, ag, trylock); @@ -246,7 +185,6 @@ _xfs_filestream_pick_ag( /* Keep track of the AG with the most free blocks. */ if (pag->pagf_freeblks > maxfree) { maxfree = pag->pagf_freeblks; - max_streams = atomic_read(&pag->pagf_fstrms); max_ag = ag; } @@ -269,7 +207,6 @@ _xfs_filestream_pick_ag( /* Break out, retaining the reference on the AG. */ free = pag->pagf_freeblks; - streams = atomic_read(&pag->pagf_fstrms); xfs_perag_put(pag); *agp = ag; break; @@ -305,317 +242,98 @@ next_ag: */ if (max_ag != NULLAGNUMBER) { xfs_filestream_get_ag(mp, max_ag); - TRACE_AG_PICK1(mp, max_ag, maxfree); - streams = max_streams; free = maxfree; *agp = max_ag; break; } /* take AG 0 if none matched */ - TRACE_AG_PICK1(mp, max_ag, maxfree); + trace_xfs_filestream_pick(ip, *agp, free, nscan); *agp = 0; return 0; } - TRACE_AG_PICK2(mp, startag, *agp, streams, free, nscan, flags); - - return 0; -} + trace_xfs_filestream_pick(ip, *agp, free, nscan); -/* - * Set the allocation group number for a file or a directory, updating inode - * references and per-AG references as appropriate. - */ -static int -_xfs_filestream_update_ag( - xfs_inode_t *ip, - xfs_inode_t *pip, - xfs_agnumber_t ag) -{ - int err = 0; - xfs_mount_t *mp; - xfs_mru_cache_t *cache; - fstrm_item_t *item; - xfs_agnumber_t old_ag; - xfs_inode_t *old_pip; - - /* - * Either ip is a regular file and pip is a directory, or ip is a - * directory and pip is NULL. - */ - ASSERT(ip && ((S_ISREG(ip->i_d.di_mode) && pip && - S_ISDIR(pip->i_d.di_mode)) || - (S_ISDIR(ip->i_d.di_mode) && !pip))); - - mp = ip->i_mount; - cache = mp->m_filestream; - - item = xfs_mru_cache_lookup(cache, ip->i_ino); - if (item) { - ASSERT(item->ip == ip); - old_ag = item->ag; - item->ag = ag; - old_pip = item->pip; - item->pip = pip; - xfs_mru_cache_done(cache); - - /* - * If the AG has changed, drop the old ref and take a new one, - * effectively transferring the reference from old to new AG. - */ - if (ag != old_ag) { - xfs_filestream_put_ag(mp, old_ag); - xfs_filestream_get_ag(mp, ag); - } - - /* - * If ip is a file and its pip has changed, drop the old ref and - * take a new one. - */ - if (pip && pip != old_pip) { - IRELE(old_pip); - IHOLD(pip); - } - - TRACE_UPDATE(mp, ip, old_ag, xfs_filestream_peek_ag(mp, old_ag), - ag, xfs_filestream_peek_ag(mp, ag)); + if (*agp == NULLAGNUMBER) return 0; - } - item = kmem_zone_zalloc(item_zone, KM_MAYFAIL); + err = ENOMEM; + item = kmem_alloc(sizeof(*item), KM_MAYFAIL); if (!item) - return ENOMEM; + goto out_put_ag; - item->ag = ag; + item->ag = *agp; item->ip = ip; - item->pip = pip; - err = xfs_mru_cache_insert(cache, ip->i_ino, item); + err = xfs_mru_cache_insert(mp->m_filestream, ip->i_ino, &item->mru); if (err) { - kmem_zone_free(item_zone, item); - return err; + if (err == EEXIST) + err = 0; + goto out_free_item; } - /* Take a reference on the AG. */ - xfs_filestream_get_ag(mp, ag); - - /* - * Take a reference on the inode itself regardless of whether it's a - * regular file or a directory. - */ - IHOLD(ip); - - /* - * In the case of a regular file, take a reference on the parent inode - * as well to ensure it remains in-core. - */ - if (pip) - IHOLD(pip); - - TRACE_UPDATE(mp, ip, ag, xfs_filestream_peek_ag(mp, ag), - ag, xfs_filestream_peek_ag(mp, ag)); - return 0; -} - -/* xfs_fstrm_free_func(): callback for freeing cached stream items. */ -STATIC void -xfs_fstrm_free_func( - unsigned long ino, - void *data) -{ - fstrm_item_t *item = (fstrm_item_t *)data; - xfs_inode_t *ip = item->ip; - - ASSERT(ip->i_ino == ino); - - xfs_iflags_clear(ip, XFS_IFILESTREAM); - - /* Drop the reference taken on the AG when the item was added. */ - xfs_filestream_put_ag(ip->i_mount, item->ag); - - TRACE_FREE(ip->i_mount, ip, item->pip, item->ag, - xfs_filestream_peek_ag(ip->i_mount, item->ag)); - - /* - * _xfs_filestream_update_ag() always takes a reference on the inode - * itself, whether it's a file or a directory. Release it here. - * This can result in the inode being freed and so we must - * not hold any inode locks when freeing filesstreams objects - * otherwise we can deadlock here. - */ - IRELE(ip); - - /* - * In the case of a regular file, _xfs_filestream_update_ag() also - * takes a ref on the parent inode to keep it in-core. Release that - * too. - */ - if (item->pip) - IRELE(item->pip); - - /* Finally, free the memory allocated for the item. */ - kmem_zone_free(item_zone, item); -} - -/* - * xfs_filestream_init() is called at xfs initialisation time to set up the - * memory zone that will be used for filestream data structure allocation. - */ -int -xfs_filestream_init(void) -{ - item_zone = kmem_zone_init(sizeof(fstrm_item_t), "fstrm_item"); - if (!item_zone) - return -ENOMEM; - - return 0; -} - -/* - * xfs_filestream_uninit() is called at xfs termination time to destroy the - * memory zone that was used for filestream data structure allocation. - */ -void -xfs_filestream_uninit(void) -{ - kmem_zone_destroy(item_zone); -} - -/* - * xfs_filestream_mount() is called when a file system is mounted with the - * filestream option. It is responsible for allocating the data structures - * needed to track the new file system's file streams. - */ -int -xfs_filestream_mount( - xfs_mount_t *mp) -{ - int err; - unsigned int lifetime, grp_count; - - /* - * The filestream timer tunable is currently fixed within the range of - * one second to four minutes, with five seconds being the default. The - * group count is somewhat arbitrary, but it'd be nice to adhere to the - * timer tunable to within about 10 percent. This requires at least 10 - * groups. - */ - lifetime = xfs_fstrm_centisecs * 10; - grp_count = 10; - - err = xfs_mru_cache_create(&mp->m_filestream, lifetime, grp_count, - xfs_fstrm_free_func); +out_free_item: + kmem_free(item); +out_put_ag: + xfs_filestream_put_ag(mp, *agp); return err; } -/* - * xfs_filestream_unmount() is called when a file system that was mounted with - * the filestream option is unmounted. It drains the data structures created - * to track the file system's file streams and frees all the memory that was - * allocated. - */ -void -xfs_filestream_unmount( - xfs_mount_t *mp) +static struct xfs_inode * +xfs_filestream_get_parent( + struct xfs_inode *ip) { - xfs_mru_cache_destroy(mp->m_filestream); -} + struct inode *inode = VFS_I(ip), *dir = NULL; + struct dentry *dentry, *parent; -/* - * Return the AG of the filestream the file or directory belongs to, or - * NULLAGNUMBER otherwise. - */ -xfs_agnumber_t -xfs_filestream_lookup_ag( - xfs_inode_t *ip) -{ - xfs_mru_cache_t *cache; - fstrm_item_t *item; - xfs_agnumber_t ag; - int ref; - - if (!S_ISREG(ip->i_d.di_mode) && !S_ISDIR(ip->i_d.di_mode)) { - ASSERT(0); - return NULLAGNUMBER; - } + dentry = d_find_alias(inode); + if (!dentry) + goto out; - cache = ip->i_mount->m_filestream; - item = xfs_mru_cache_lookup(cache, ip->i_ino); - if (!item) { - TRACE_LOOKUP(ip->i_mount, ip, NULL, NULLAGNUMBER, 0); - return NULLAGNUMBER; - } + parent = dget_parent(dentry); + if (!parent) + goto out_dput; - ASSERT(ip == item->ip); - ag = item->ag; - ref = xfs_filestream_peek_ag(ip->i_mount, ag); - xfs_mru_cache_done(cache); + dir = igrab(parent->d_inode); + dput(parent); - TRACE_LOOKUP(ip->i_mount, ip, item->pip, ag, ref); - return ag; +out_dput: + dput(dentry); +out: + return dir ? XFS_I(dir) : NULL; } /* - * xfs_filestream_associate() should only be called to associate a regular file - * with its parent directory. Calling it with a child directory isn't - * appropriate because filestreams don't apply to entire directory hierarchies. - * Creating a file in a child directory of an existing filestream directory - * starts a new filestream with its own allocation group association. + * Find the right allocation group for a file, either by finding an + * existing file stream or creating a new one. * - * Returns < 0 on error, 0 if successful association occurred, > 0 if - * we failed to get an association because of locking issues. + * Returns NULLAGNUMBER in case of an error. */ -int -xfs_filestream_associate( - xfs_inode_t *pip, - xfs_inode_t *ip) +xfs_agnumber_t +xfs_filestream_lookup_ag( + struct xfs_inode *ip) { - xfs_mount_t *mp; - xfs_mru_cache_t *cache; - fstrm_item_t *item; - xfs_agnumber_t ag, rotorstep, startag; - int err = 0; + struct xfs_mount *mp = ip->i_mount; + struct xfs_inode *pip = NULL; + xfs_agnumber_t startag, ag = NULLAGNUMBER; + struct xfs_mru_cache_elem *mru; - ASSERT(S_ISDIR(pip->i_d.di_mode)); ASSERT(S_ISREG(ip->i_d.di_mode)); - if (!S_ISDIR(pip->i_d.di_mode) || !S_ISREG(ip->i_d.di_mode)) - return -EINVAL; - mp = pip->i_mount; - cache = mp->m_filestream; + pip = xfs_filestream_get_parent(ip); + if (!pip) + goto out; - /* - * We have a problem, Houston. - * - * Taking the iolock here violates inode locking order - we already - * hold the ilock. Hence if we block getting this lock we may never - * wake. Unfortunately, that means if we can't get the lock, we're - * screwed in terms of getting a stream association - we can't spin - * waiting for the lock because someone else is waiting on the lock we - * hold and we cannot drop that as we are in a transaction here. - * - * Lucky for us, this inversion is not a problem because it's a - * directory inode that we are trying to lock here. - * - * So, if we can't get the iolock without sleeping then just give up - */ - if (!xfs_ilock_nowait(pip, XFS_IOLOCK_EXCL)) - return 1; - - /* If the parent directory is already in the cache, use its AG. */ - item = xfs_mru_cache_lookup(cache, pip->i_ino); - if (item) { - ASSERT(item->ip == pip); - ag = item->ag; - xfs_mru_cache_done(cache); - - TRACE_LOOKUP(mp, pip, pip, ag, xfs_filestream_peek_ag(mp, ag)); - err = _xfs_filestream_update_ag(ip, pip, ag); + mru = xfs_mru_cache_lookup(mp->m_filestream, pip->i_ino); + if (mru) { + ag = container_of(mru, struct xfs_fstrm_item, mru)->ag; + xfs_mru_cache_done(mp->m_filestream); - goto exit; + trace_xfs_filestream_lookup(ip, ag); + goto out; } /* @@ -623,202 +341,94 @@ xfs_filestream_associate( * use the directory inode's AG. */ if (mp->m_flags & XFS_MOUNT_32BITINODES) { - rotorstep = xfs_rotorstep; + xfs_agnumber_t rotorstep = xfs_rotorstep; startag = (mp->m_agfrotor / rotorstep) % mp->m_sb.sb_agcount; mp->m_agfrotor = (mp->m_agfrotor + 1) % (mp->m_sb.sb_agcount * rotorstep); } else startag = XFS_INO_TO_AGNO(mp, pip->i_ino); - /* Pick a new AG for the parent inode starting at startag. */ - err = _xfs_filestream_pick_ag(mp, startag, &ag, 0, 0); - if (err || ag == NULLAGNUMBER) - goto exit_did_pick; - - /* Associate the parent inode with the AG. */ - err = _xfs_filestream_update_ag(pip, NULL, ag); - if (err) - goto exit_did_pick; - - /* Associate the file inode with the AG. */ - err = _xfs_filestream_update_ag(ip, pip, ag); - if (err) - goto exit_did_pick; - - TRACE_ASSOCIATE(mp, ip, pip, ag, xfs_filestream_peek_ag(mp, ag)); - -exit_did_pick: - /* - * If _xfs_filestream_pick_ag() returned a valid AG, remove the - * reference it took on it, since the file and directory will have taken - * their own now if they were successfully cached. - */ - if (ag != NULLAGNUMBER) - xfs_filestream_put_ag(mp, ag); - -exit: - xfs_iunlock(pip, XFS_IOLOCK_EXCL); - return -err; + if (xfs_filestream_pick_ag(pip, startag, &ag, 0, 0)) + ag = NULLAGNUMBER; +out: + IRELE(pip); + return ag; } /* - * Pick a new allocation group for the current file and its file stream. This - * function is called by xfs_bmap_filestreams() with the mount point's per-ag - * lock held. + * Pick a new allocation group for the current file and its file stream. + * + * This is called when the allocator can't find a suitable extent in the + * current AG, and we have to move the stream into a new AG with more space. */ int xfs_filestream_new_ag( struct xfs_bmalloca *ap, xfs_agnumber_t *agp) { - int flags, err; - xfs_inode_t *ip, *pip = NULL; - xfs_mount_t *mp; - xfs_mru_cache_t *cache; - xfs_extlen_t minlen; - fstrm_item_t *dir, *file; - xfs_agnumber_t ag = NULLAGNUMBER; - - ip = ap->ip; - mp = ip->i_mount; - cache = mp->m_filestream; - minlen = ap->length; - *agp = NULLAGNUMBER; + struct xfs_inode *ip = ap->ip, *pip; + struct xfs_mount *mp = ip->i_mount; + xfs_extlen_t minlen = ap->length; + xfs_agnumber_t startag = 0; + int flags, err = 0; + struct xfs_mru_cache_elem *mru; - /* - * Look for the file in the cache, removing it if it's found. Doing - * this allows it to be held across the dir lookup that follows. - */ - file = xfs_mru_cache_remove(cache, ip->i_ino); - if (file) { - ASSERT(ip == file->ip); - - /* Save the file's parent inode and old AG number for later. */ - pip = file->pip; - ag = file->ag; - - /* Look for the file's directory in the cache. */ - dir = xfs_mru_cache_lookup(cache, pip->i_ino); - if (dir) { - ASSERT(pip == dir->ip); - - /* - * If the directory has already moved on to a new AG, - * use that AG as the new AG for the file. Don't - * forget to twiddle the AG refcounts to match the - * movement. - */ - if (dir->ag != file->ag) { - xfs_filestream_put_ag(mp, file->ag); - xfs_filestream_get_ag(mp, dir->ag); - *agp = file->ag = dir->ag; - } - - xfs_mru_cache_done(cache); - } + *agp = NULLAGNUMBER; - /* - * Put the file back in the cache. If this fails, the free - * function needs to be called to tidy up in the same way as if - * the item had simply expired from the cache. - */ - err = xfs_mru_cache_insert(cache, ip->i_ino, file); - if (err) { - xfs_fstrm_free_func(ip->i_ino, file); - return err; - } + pip = xfs_filestream_get_parent(ip); + if (!pip) + goto exit; - /* - * If the file's AG was moved to the directory's new AG, there's - * nothing more to be done. - */ - if (*agp != NULLAGNUMBER) { - TRACE_MOVEAG(mp, ip, pip, - ag, xfs_filestream_peek_ag(mp, ag), - *agp, xfs_filestream_peek_ag(mp, *agp)); - return 0; - } + mru = xfs_mru_cache_remove(mp->m_filestream, pip->i_ino); + if (mru) { + struct xfs_fstrm_item *item = + container_of(mru, struct xfs_fstrm_item, mru); + startag = (item->ag + 1) % mp->m_sb.sb_agcount; } - /* - * If the file's parent directory is known, take its iolock in exclusive - * mode to prevent two sibling files from racing each other to migrate - * themselves and their parent to different AGs. - * - * Note that we lock the parent directory iolock inside the child - * iolock here. That's fine as we never hold both parent and child - * iolock in any other place. This is different from the ilock, - * which requires locking of the child after the parent for namespace - * operations. - */ - if (pip) - xfs_ilock(pip, XFS_IOLOCK_EXCL | XFS_IOLOCK_PARENT); - - /* - * A new AG needs to be found for the file. If the file's parent - * directory is also known, it will be moved to the new AG as well to - * ensure that files created inside it in future use the new AG. - */ - ag = (ag == NULLAGNUMBER) ? 0 : (ag + 1) % mp->m_sb.sb_agcount; flags = (ap->userdata ? XFS_PICK_USERDATA : 0) | (ap->flist->xbf_low ? XFS_PICK_LOWSPACE : 0); - err = _xfs_filestream_pick_ag(mp, ag, agp, flags, minlen); - if (err || *agp == NULLAGNUMBER) - goto exit; + err = xfs_filestream_pick_ag(pip, startag, agp, flags, minlen); /* - * If the file wasn't found in the file cache, then its parent directory - * inode isn't known. For this to have happened, the file must either - * be pre-existing, or it was created long enough ago that its cache - * entry has expired. This isn't the sort of usage that the filestreams - * allocator is trying to optimise, so there's no point trying to track - * its new AG somehow in the filestream data structures. + * Only free the item here so we skip over the old AG earlier. */ - if (!pip) { - TRACE_ORPHAN(mp, ip, *agp); - goto exit; - } - - /* Associate the parent inode with the AG. */ - err = _xfs_filestream_update_ag(pip, NULL, *agp); - if (err) - goto exit; - - /* Associate the file inode with the AG. */ - err = _xfs_filestream_update_ag(ip, pip, *agp); - if (err) - goto exit; - - TRACE_MOVEAG(mp, ip, pip, NULLAGNUMBER, 0, - *agp, xfs_filestream_peek_ag(mp, *agp)); + if (mru) + xfs_fstrm_free_func(mru); + IRELE(pip); exit: - /* - * If _xfs_filestream_pick_ag() returned a valid AG, remove the - * reference it took on it, since the file and directory will have taken - * their own now if they were successfully cached. - */ - if (*agp != NULLAGNUMBER) - xfs_filestream_put_ag(mp, *agp); - else + if (*agp == NULLAGNUMBER) *agp = 0; - - if (pip) - xfs_iunlock(pip, XFS_IOLOCK_EXCL); - return err; } -/* - * Remove an association between an inode and a filestream object. - * Typically this is done on last close of an unlinked file. - */ void xfs_filestream_deassociate( - xfs_inode_t *ip) + struct xfs_inode *ip) { - xfs_mru_cache_t *cache = ip->i_mount->m_filestream; + xfs_mru_cache_delete(ip->i_mount->m_filestream, ip->i_ino); +} + +int +xfs_filestream_mount( + xfs_mount_t *mp) +{ + /* + * The filestream timer tunable is currently fixed within the range of + * one second to four minutes, with five seconds being the default. The + * group count is somewhat arbitrary, but it'd be nice to adhere to the + * timer tunable to within about 10 percent. This requires at least 10 + * groups. + */ + return xfs_mru_cache_create(&mp->m_filestream, xfs_fstrm_centisecs * 10, + 10, xfs_fstrm_free_func); +} - xfs_mru_cache_delete(cache, ip->i_ino); +void +xfs_filestream_unmount( + xfs_mount_t *mp) +{ + xfs_mru_cache_destroy(mp->m_filestream); } diff --git a/fs/xfs/xfs_filestream.h b/fs/xfs/xfs_filestream.h index 6d61dbee8564..2ef43406e53b 100644 --- a/fs/xfs/xfs_filestream.h +++ b/fs/xfs/xfs_filestream.h @@ -20,50 +20,20 @@ struct xfs_mount; struct xfs_inode; -struct xfs_perag; struct xfs_bmalloca; -#ifdef XFS_FILESTREAMS_TRACE -#define XFS_FSTRM_KTRACE_INFO 1 -#define XFS_FSTRM_KTRACE_AGSCAN 2 -#define XFS_FSTRM_KTRACE_AGPICK1 3 -#define XFS_FSTRM_KTRACE_AGPICK2 4 -#define XFS_FSTRM_KTRACE_UPDATE 5 -#define XFS_FSTRM_KTRACE_FREE 6 -#define XFS_FSTRM_KTRACE_ITEM_LOOKUP 7 -#define XFS_FSTRM_KTRACE_ASSOCIATE 8 -#define XFS_FSTRM_KTRACE_MOVEAG 9 -#define XFS_FSTRM_KTRACE_ORPHAN 10 - -#define XFS_FSTRM_KTRACE_SIZE 16384 -extern ktrace_t *xfs_filestreams_trace_buf; - -#endif - -/* allocation selection flags */ -typedef enum xfs_fstrm_alloc { - XFS_PICK_USERDATA = 1, - XFS_PICK_LOWSPACE = 2, -} xfs_fstrm_alloc_t; - -/* prototypes for filestream.c */ -int xfs_filestream_init(void); -void xfs_filestream_uninit(void); int xfs_filestream_mount(struct xfs_mount *mp); void xfs_filestream_unmount(struct xfs_mount *mp); -xfs_agnumber_t xfs_filestream_lookup_ag(struct xfs_inode *ip); -int xfs_filestream_associate(struct xfs_inode *dip, struct xfs_inode *ip); void xfs_filestream_deassociate(struct xfs_inode *ip); +xfs_agnumber_t xfs_filestream_lookup_ag(struct xfs_inode *ip); int xfs_filestream_new_ag(struct xfs_bmalloca *ap, xfs_agnumber_t *agp); +int xfs_filestream_peek_ag(struct xfs_mount *mp, xfs_agnumber_t agno); - -/* filestreams for the inode? */ static inline int xfs_inode_is_filestream( struct xfs_inode *ip) { return (ip->i_mount->m_flags & XFS_MOUNT_FILESTREAMS) || - xfs_iflags_test(ip, XFS_IFILESTREAM) || (ip->i_d.di_flags & XFS_DIFLAG_FILESTREAM); } diff --git a/fs/xfs/xfs_format.h b/fs/xfs/xfs_format.h index 9898f31d05d8..34d85aca3058 100644 --- a/fs/xfs/xfs_format.h +++ b/fs/xfs/xfs_format.h @@ -202,6 +202,8 @@ typedef __be32 xfs_alloc_ptr_t; */ #define XFS_IBT_MAGIC 0x49414254 /* 'IABT' */ #define XFS_IBT_CRC_MAGIC 0x49414233 /* 'IAB3' */ +#define XFS_FIBT_MAGIC 0x46494254 /* 'FIBT' */ +#define XFS_FIBT_CRC_MAGIC 0x46494233 /* 'FIB3' */ typedef __uint64_t xfs_inofree_t; #define XFS_INODES_PER_CHUNK (NBBY * sizeof(xfs_inofree_t)) @@ -244,7 +246,17 @@ typedef __be32 xfs_inobt_ptr_t; * block numbers in the AG. */ #define XFS_IBT_BLOCK(mp) ((xfs_agblock_t)(XFS_CNT_BLOCK(mp) + 1)) -#define XFS_PREALLOC_BLOCKS(mp) ((xfs_agblock_t)(XFS_IBT_BLOCK(mp) + 1)) +#define XFS_FIBT_BLOCK(mp) ((xfs_agblock_t)(XFS_IBT_BLOCK(mp) + 1)) + +/* + * The first data block of an AG depends on whether the filesystem was formatted + * with the finobt feature. If so, account for the finobt reserved root btree + * block. + */ +#define XFS_PREALLOC_BLOCKS(mp) \ + (xfs_sb_version_hasfinobt(&((mp)->m_sb)) ? \ + XFS_FIBT_BLOCK(mp) + 1 : \ + XFS_IBT_BLOCK(mp) + 1) diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h index c5fc116dfaa3..d34703dbcb42 100644 --- a/fs/xfs/xfs_fs.h +++ b/fs/xfs/xfs_fs.h @@ -238,6 +238,7 @@ typedef struct xfs_fsop_resblks { #define XFS_FSOP_GEOM_FLAGS_LAZYSB 0x4000 /* lazy superblock counters */ #define XFS_FSOP_GEOM_FLAGS_V5SB 0x8000 /* version 5 superblock */ #define XFS_FSOP_GEOM_FLAGS_FTYPE 0x10000 /* inode directory types */ +#define XFS_FSOP_GEOM_FLAGS_FINOBT 0x20000 /* free inode btree */ /* * Minimum and maximum sizes need for growth checks. diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index 02fb943cbf22..3445ead7c1fc 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -104,7 +104,9 @@ xfs_fs_geometry( (xfs_sb_version_hascrc(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_V5SB : 0) | (xfs_sb_version_hasftype(&mp->m_sb) ? - XFS_FSOP_GEOM_FLAGS_FTYPE : 0); + XFS_FSOP_GEOM_FLAGS_FTYPE : 0) | + (xfs_sb_version_hasfinobt(&mp->m_sb) ? + XFS_FSOP_GEOM_FLAGS_FINOBT : 0); geo->logsectsize = xfs_sb_version_hassector(&mp->m_sb) ? mp->m_sb.sb_logsectsize : BBSIZE; geo->rtsectsize = mp->m_sb.sb_blocksize; @@ -316,6 +318,10 @@ xfs_growfs_data_private( agi->agi_dirino = cpu_to_be32(NULLAGINO); if (xfs_sb_version_hascrc(&mp->m_sb)) uuid_copy(&agi->agi_uuid, &mp->m_sb.sb_uuid); + if (xfs_sb_version_hasfinobt(&mp->m_sb)) { + agi->agi_free_root = cpu_to_be32(XFS_FIBT_BLOCK(mp)); + agi->agi_free_level = cpu_to_be32(1); + } for (bucket = 0; bucket < XFS_AGI_UNLINKED_BUCKETS; bucket++) agi->agi_unlinked[bucket] = cpu_to_be32(NULLAGINO); @@ -407,6 +413,34 @@ xfs_growfs_data_private( xfs_buf_relse(bp); if (error) goto error0; + + /* + * FINO btree root block + */ + if (xfs_sb_version_hasfinobt(&mp->m_sb)) { + bp = xfs_growfs_get_hdr_buf(mp, + XFS_AGB_TO_DADDR(mp, agno, XFS_FIBT_BLOCK(mp)), + BTOBB(mp->m_sb.sb_blocksize), 0, + &xfs_inobt_buf_ops); + if (!bp) { + error = ENOMEM; + goto error0; + } + + if (xfs_sb_version_hascrc(&mp->m_sb)) + xfs_btree_init_block(mp, bp, XFS_FIBT_CRC_MAGIC, + 0, 0, agno, + XFS_BTREE_CRC_BLOCKS); + else + xfs_btree_init_block(mp, bp, XFS_FIBT_MAGIC, 0, + 0, agno, 0); + + error = xfs_bwrite(bp); + xfs_buf_relse(bp); + if (error) + goto error0; + } + } xfs_trans_agblocks_delta(tp, nfree); /* diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c index 8f711db61a0c..6ac0c2986c32 100644 --- a/fs/xfs/xfs_ialloc.c +++ b/fs/xfs/xfs_ialloc.c @@ -112,6 +112,66 @@ xfs_inobt_get_rec( } /* + * Insert a single inobt record. Cursor must already point to desired location. + */ +STATIC int +xfs_inobt_insert_rec( + struct xfs_btree_cur *cur, + __int32_t freecount, + xfs_inofree_t free, + int *stat) +{ + cur->bc_rec.i.ir_freecount = freecount; + cur->bc_rec.i.ir_free = free; + return xfs_btree_insert(cur, stat); +} + +/* + * Insert records describing a newly allocated inode chunk into the inobt. + */ +STATIC int +xfs_inobt_insert( + struct xfs_mount *mp, + struct xfs_trans *tp, + struct xfs_buf *agbp, + xfs_agino_t newino, + xfs_agino_t newlen, + xfs_btnum_t btnum) +{ + struct xfs_btree_cur *cur; + struct xfs_agi *agi = XFS_BUF_TO_AGI(agbp); + xfs_agnumber_t agno = be32_to_cpu(agi->agi_seqno); + xfs_agino_t thisino; + int i; + int error; + + cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, btnum); + + for (thisino = newino; + thisino < newino + newlen; + thisino += XFS_INODES_PER_CHUNK) { + error = xfs_inobt_lookup(cur, thisino, XFS_LOOKUP_EQ, &i); + if (error) { + xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); + return error; + } + ASSERT(i == 0); + + error = xfs_inobt_insert_rec(cur, XFS_INODES_PER_CHUNK, + XFS_INOBT_ALL_FREE, &i); + if (error) { + xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); + return error; + } + ASSERT(i == 1); + } + + xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR); + + return 0; +} + +/* * Verify that the number of free inodes in the AGI is correct. */ #ifdef DEBUG @@ -303,13 +363,10 @@ xfs_ialloc_ag_alloc( { xfs_agi_t *agi; /* allocation group header */ xfs_alloc_arg_t args; /* allocation argument structure */ - xfs_btree_cur_t *cur; /* inode btree cursor */ xfs_agnumber_t agno; int error; - int i; xfs_agino_t newino; /* new first inode's number */ xfs_agino_t newlen; /* new number of inodes */ - xfs_agino_t thisino; /* current inode number, for loop */ int isaligned = 0; /* inode allocation at stripe unit */ /* boundary */ struct xfs_perag *pag; @@ -459,29 +516,19 @@ xfs_ialloc_ag_alloc( agi->agi_newino = cpu_to_be32(newino); /* - * Insert records describing the new inode chunk into the btree. + * Insert records describing the new inode chunk into the btrees. */ - cur = xfs_inobt_init_cursor(args.mp, tp, agbp, agno); - for (thisino = newino; - thisino < newino + newlen; - thisino += XFS_INODES_PER_CHUNK) { - cur->bc_rec.i.ir_startino = thisino; - cur->bc_rec.i.ir_freecount = XFS_INODES_PER_CHUNK; - cur->bc_rec.i.ir_free = XFS_INOBT_ALL_FREE; - error = xfs_btree_lookup(cur, XFS_LOOKUP_EQ, &i); - if (error) { - xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); - return error; - } - ASSERT(i == 0); - error = xfs_btree_insert(cur, &i); - if (error) { - xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); + error = xfs_inobt_insert(args.mp, tp, agbp, newino, newlen, + XFS_BTNUM_INO); + if (error) + return error; + + if (xfs_sb_version_hasfinobt(&args.mp->m_sb)) { + error = xfs_inobt_insert(args.mp, tp, agbp, newino, newlen, + XFS_BTNUM_FINO); + if (error) return error; - } - ASSERT(i == 1); } - xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR); /* * Log allocation group header fields */ @@ -675,13 +722,10 @@ xfs_ialloc_get_rec( } /* - * Allocate an inode. - * - * The caller selected an AG for us, and made sure that free inodes are - * available. + * Allocate an inode using the inobt-only algorithm. */ STATIC int -xfs_dialloc_ag( +xfs_dialloc_ag_inobt( struct xfs_trans *tp, struct xfs_buf *agbp, xfs_ino_t parent, @@ -707,7 +751,7 @@ xfs_dialloc_ag( ASSERT(pag->pagi_freecount > 0); restart_pagno: - cur = xfs_inobt_init_cursor(mp, tp, agbp, agno); + cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO); /* * If pagino is 0 (this is the root inode allocation) use newino. * This must work because we've just allocated some. @@ -940,6 +984,294 @@ error0: } /* + * Use the free inode btree to allocate an inode based on distance from the + * parent. Note that the provided cursor may be deleted and replaced. + */ +STATIC int +xfs_dialloc_ag_finobt_near( + xfs_agino_t pagino, + struct xfs_btree_cur **ocur, + struct xfs_inobt_rec_incore *rec) +{ + struct xfs_btree_cur *lcur = *ocur; /* left search cursor */ + struct xfs_btree_cur *rcur; /* right search cursor */ + struct xfs_inobt_rec_incore rrec; + int error; + int i, j; + + error = xfs_inobt_lookup(lcur, pagino, XFS_LOOKUP_LE, &i); + if (error) + return error; + + if (i == 1) { + error = xfs_inobt_get_rec(lcur, rec, &i); + if (error) + return error; + XFS_WANT_CORRUPTED_RETURN(i == 1); + + /* + * See if we've landed in the parent inode record. The finobt + * only tracks chunks with at least one free inode, so record + * existence is enough. + */ + if (pagino >= rec->ir_startino && + pagino < (rec->ir_startino + XFS_INODES_PER_CHUNK)) + return 0; + } + + error = xfs_btree_dup_cursor(lcur, &rcur); + if (error) + return error; + + error = xfs_inobt_lookup(rcur, pagino, XFS_LOOKUP_GE, &j); + if (error) + goto error_rcur; + if (j == 1) { + error = xfs_inobt_get_rec(rcur, &rrec, &j); + if (error) + goto error_rcur; + XFS_WANT_CORRUPTED_GOTO(j == 1, error_rcur); + } + + XFS_WANT_CORRUPTED_GOTO(i == 1 || j == 1, error_rcur); + if (i == 1 && j == 1) { + /* + * Both the left and right records are valid. Choose the closer + * inode chunk to the target. + */ + if ((pagino - rec->ir_startino + XFS_INODES_PER_CHUNK - 1) > + (rrec.ir_startino - pagino)) { + *rec = rrec; + xfs_btree_del_cursor(lcur, XFS_BTREE_NOERROR); + *ocur = rcur; + } else { + xfs_btree_del_cursor(rcur, XFS_BTREE_NOERROR); + } + } else if (j == 1) { + /* only the right record is valid */ + *rec = rrec; + xfs_btree_del_cursor(lcur, XFS_BTREE_NOERROR); + *ocur = rcur; + } else if (i == 1) { + /* only the left record is valid */ + xfs_btree_del_cursor(rcur, XFS_BTREE_NOERROR); + } + + return 0; + +error_rcur: + xfs_btree_del_cursor(rcur, XFS_BTREE_ERROR); + return error; +} + +/* + * Use the free inode btree to find a free inode based on a newino hint. If + * the hint is NULL, find the first free inode in the AG. + */ +STATIC int +xfs_dialloc_ag_finobt_newino( + struct xfs_agi *agi, + struct xfs_btree_cur *cur, + struct xfs_inobt_rec_incore *rec) +{ + int error; + int i; + + if (agi->agi_newino != cpu_to_be32(NULLAGINO)) { + error = xfs_inobt_lookup(cur, agi->agi_newino, XFS_LOOKUP_EQ, + &i); + if (error) + return error; + if (i == 1) { + error = xfs_inobt_get_rec(cur, rec, &i); + if (error) + return error; + XFS_WANT_CORRUPTED_RETURN(i == 1); + + return 0; + } + } + + /* + * Find the first inode available in the AG. + */ + error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &i); + if (error) + return error; + XFS_WANT_CORRUPTED_RETURN(i == 1); + + error = xfs_inobt_get_rec(cur, rec, &i); + if (error) + return error; + XFS_WANT_CORRUPTED_RETURN(i == 1); + + return 0; +} + +/* + * Update the inobt based on a modification made to the finobt. Also ensure that + * the records from both trees are equivalent post-modification. + */ +STATIC int +xfs_dialloc_ag_update_inobt( + struct xfs_btree_cur *cur, /* inobt cursor */ + struct xfs_inobt_rec_incore *frec, /* finobt record */ + int offset) /* inode offset */ +{ + struct xfs_inobt_rec_incore rec; + int error; + int i; + + error = xfs_inobt_lookup(cur, frec->ir_startino, XFS_LOOKUP_EQ, &i); + if (error) + return error; + XFS_WANT_CORRUPTED_RETURN(i == 1); + + error = xfs_inobt_get_rec(cur, &rec, &i); + if (error) + return error; + XFS_WANT_CORRUPTED_RETURN(i == 1); + ASSERT((XFS_AGINO_TO_OFFSET(cur->bc_mp, rec.ir_startino) % + XFS_INODES_PER_CHUNK) == 0); + + rec.ir_free &= ~XFS_INOBT_MASK(offset); + rec.ir_freecount--; + + XFS_WANT_CORRUPTED_RETURN((rec.ir_free == frec->ir_free) && + (rec.ir_freecount == frec->ir_freecount)); + + error = xfs_inobt_update(cur, &rec); + if (error) + return error; + + return 0; +} + +/* + * Allocate an inode using the free inode btree, if available. Otherwise, fall + * back to the inobt search algorithm. + * + * The caller selected an AG for us, and made sure that free inodes are + * available. + */ +STATIC int +xfs_dialloc_ag( + struct xfs_trans *tp, + struct xfs_buf *agbp, + xfs_ino_t parent, + xfs_ino_t *inop) +{ + struct xfs_mount *mp = tp->t_mountp; + struct xfs_agi *agi = XFS_BUF_TO_AGI(agbp); + xfs_agnumber_t agno = be32_to_cpu(agi->agi_seqno); + xfs_agnumber_t pagno = XFS_INO_TO_AGNO(mp, parent); + xfs_agino_t pagino = XFS_INO_TO_AGINO(mp, parent); + struct xfs_perag *pag; + struct xfs_btree_cur *cur; /* finobt cursor */ + struct xfs_btree_cur *icur; /* inobt cursor */ + struct xfs_inobt_rec_incore rec; + xfs_ino_t ino; + int error; + int offset; + int i; + + if (!xfs_sb_version_hasfinobt(&mp->m_sb)) + return xfs_dialloc_ag_inobt(tp, agbp, parent, inop); + + pag = xfs_perag_get(mp, agno); + + /* + * If pagino is 0 (this is the root inode allocation) use newino. + * This must work because we've just allocated some. + */ + if (!pagino) + pagino = be32_to_cpu(agi->agi_newino); + + cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_FINO); + + error = xfs_check_agi_freecount(cur, agi); + if (error) + goto error_cur; + + /* + * The search algorithm depends on whether we're in the same AG as the + * parent. If so, find the closest available inode to the parent. If + * not, consider the agi hint or find the first free inode in the AG. + */ + if (agno == pagno) + error = xfs_dialloc_ag_finobt_near(pagino, &cur, &rec); + else + error = xfs_dialloc_ag_finobt_newino(agi, cur, &rec); + if (error) + goto error_cur; + + offset = xfs_lowbit64(rec.ir_free); + ASSERT(offset >= 0); + ASSERT(offset < XFS_INODES_PER_CHUNK); + ASSERT((XFS_AGINO_TO_OFFSET(mp, rec.ir_startino) % + XFS_INODES_PER_CHUNK) == 0); + ino = XFS_AGINO_TO_INO(mp, agno, rec.ir_startino + offset); + + /* + * Modify or remove the finobt record. + */ + rec.ir_free &= ~XFS_INOBT_MASK(offset); + rec.ir_freecount--; + if (rec.ir_freecount) + error = xfs_inobt_update(cur, &rec); + else + error = xfs_btree_delete(cur, &i); + if (error) + goto error_cur; + + /* + * The finobt has now been updated appropriately. We haven't updated the + * agi and superblock yet, so we can create an inobt cursor and validate + * the original freecount. If all is well, make the equivalent update to + * the inobt using the finobt record and offset information. + */ + icur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO); + + error = xfs_check_agi_freecount(icur, agi); + if (error) + goto error_icur; + + error = xfs_dialloc_ag_update_inobt(icur, &rec, offset); + if (error) + goto error_icur; + + /* + * Both trees have now been updated. We must update the perag and + * superblock before we can check the freecount for each btree. + */ + be32_add_cpu(&agi->agi_freecount, -1); + xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT); + pag->pagi_freecount--; + + xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, -1); + + error = xfs_check_agi_freecount(icur, agi); + if (error) + goto error_icur; + error = xfs_check_agi_freecount(cur, agi); + if (error) + goto error_icur; + + xfs_btree_del_cursor(icur, XFS_BTREE_NOERROR); + xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR); + xfs_perag_put(pag); + *inop = ino; + return 0; + +error_icur: + xfs_btree_del_cursor(icur, XFS_BTREE_ERROR); +error_cur: + xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); + xfs_perag_put(pag); + return error; +} + +/* * Allocate an inode on disk. * * Mode is used to tell whether the new inode will need space, and whether it @@ -1098,78 +1430,34 @@ out_error: return XFS_ERROR(error); } -/* - * Free disk inode. Carefully avoids touching the incore inode, all - * manipulations incore are the caller's responsibility. - * The on-disk inode is not changed by this operation, only the - * btree (free inode mask) is changed. - */ -int -xfs_difree( - xfs_trans_t *tp, /* transaction pointer */ - xfs_ino_t inode, /* inode to be freed */ - xfs_bmap_free_t *flist, /* extents to free */ - int *delete, /* set if inode cluster was deleted */ - xfs_ino_t *first_ino) /* first inode in deleted cluster */ +STATIC int +xfs_difree_inobt( + struct xfs_mount *mp, + struct xfs_trans *tp, + struct xfs_buf *agbp, + xfs_agino_t agino, + struct xfs_bmap_free *flist, + int *delete, + xfs_ino_t *first_ino, + struct xfs_inobt_rec_incore *orec) { - /* REFERENCED */ - xfs_agblock_t agbno; /* block number containing inode */ - xfs_buf_t *agbp; /* buffer containing allocation group header */ - xfs_agino_t agino; /* inode number relative to allocation group */ - xfs_agnumber_t agno; /* allocation group number */ - xfs_agi_t *agi; /* allocation group header */ - xfs_btree_cur_t *cur; /* inode btree cursor */ - int error; /* error return value */ - int i; /* result code */ - int ilen; /* inodes in an inode cluster */ - xfs_mount_t *mp; /* mount structure for filesystem */ - int off; /* offset of inode in inode chunk */ - xfs_inobt_rec_incore_t rec; /* btree record */ - struct xfs_perag *pag; - - mp = tp->t_mountp; + struct xfs_agi *agi = XFS_BUF_TO_AGI(agbp); + xfs_agnumber_t agno = be32_to_cpu(agi->agi_seqno); + struct xfs_perag *pag; + struct xfs_btree_cur *cur; + struct xfs_inobt_rec_incore rec; + int ilen; + int error; + int i; + int off; - /* - * Break up inode number into its components. - */ - agno = XFS_INO_TO_AGNO(mp, inode); - if (agno >= mp->m_sb.sb_agcount) { - xfs_warn(mp, "%s: agno >= mp->m_sb.sb_agcount (%d >= %d).", - __func__, agno, mp->m_sb.sb_agcount); - ASSERT(0); - return XFS_ERROR(EINVAL); - } - agino = XFS_INO_TO_AGINO(mp, inode); - if (inode != XFS_AGINO_TO_INO(mp, agno, agino)) { - xfs_warn(mp, "%s: inode != XFS_AGINO_TO_INO() (%llu != %llu).", - __func__, (unsigned long long)inode, - (unsigned long long)XFS_AGINO_TO_INO(mp, agno, agino)); - ASSERT(0); - return XFS_ERROR(EINVAL); - } - agbno = XFS_AGINO_TO_AGBNO(mp, agino); - if (agbno >= mp->m_sb.sb_agblocks) { - xfs_warn(mp, "%s: agbno >= mp->m_sb.sb_agblocks (%d >= %d).", - __func__, agbno, mp->m_sb.sb_agblocks); - ASSERT(0); - return XFS_ERROR(EINVAL); - } - /* - * Get the allocation group header. - */ - error = xfs_ialloc_read_agi(mp, tp, agno, &agbp); - if (error) { - xfs_warn(mp, "%s: xfs_ialloc_read_agi() returned error %d.", - __func__, error); - return error; - } - agi = XFS_BUF_TO_AGI(agbp); ASSERT(agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC)); - ASSERT(agbno < be32_to_cpu(agi->agi_length)); + ASSERT(XFS_AGINO_TO_AGBNO(mp, agino) < be32_to_cpu(agi->agi_length)); + /* * Initialize the cursor. */ - cur = xfs_inobt_init_cursor(mp, tp, agbp, agno); + cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO); error = xfs_check_agi_freecount(cur, agi); if (error) @@ -1261,6 +1549,7 @@ xfs_difree( if (error) goto error0; + *orec = rec; xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR); return 0; @@ -1269,6 +1558,182 @@ error0: return error; } +/* + * Free an inode in the free inode btree. + */ +STATIC int +xfs_difree_finobt( + struct xfs_mount *mp, + struct xfs_trans *tp, + struct xfs_buf *agbp, + xfs_agino_t agino, + struct xfs_inobt_rec_incore *ibtrec) /* inobt record */ +{ + struct xfs_agi *agi = XFS_BUF_TO_AGI(agbp); + xfs_agnumber_t agno = be32_to_cpu(agi->agi_seqno); + struct xfs_btree_cur *cur; + struct xfs_inobt_rec_incore rec; + int offset = agino - ibtrec->ir_startino; + int error; + int i; + + cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_FINO); + + error = xfs_inobt_lookup(cur, ibtrec->ir_startino, XFS_LOOKUP_EQ, &i); + if (error) + goto error; + if (i == 0) { + /* + * If the record does not exist in the finobt, we must have just + * freed an inode in a previously fully allocated chunk. If not, + * something is out of sync. + */ + XFS_WANT_CORRUPTED_GOTO(ibtrec->ir_freecount == 1, error); + + error = xfs_inobt_insert_rec(cur, ibtrec->ir_freecount, + ibtrec->ir_free, &i); + if (error) + goto error; + ASSERT(i == 1); + + goto out; + } + + /* + * Read and update the existing record. We could just copy the ibtrec + * across here, but that would defeat the purpose of having redundant + * metadata. By making the modifications independently, we can catch + * corruptions that we wouldn't see if we just copied from one record + * to another. + */ + error = xfs_inobt_get_rec(cur, &rec, &i); + if (error) + goto error; + XFS_WANT_CORRUPTED_GOTO(i == 1, error); + + rec.ir_free |= XFS_INOBT_MASK(offset); + rec.ir_freecount++; + + XFS_WANT_CORRUPTED_GOTO((rec.ir_free == ibtrec->ir_free) && + (rec.ir_freecount == ibtrec->ir_freecount), + error); + + /* + * The content of inobt records should always match between the inobt + * and finobt. The lifecycle of records in the finobt is different from + * the inobt in that the finobt only tracks records with at least one + * free inode. Hence, if all of the inodes are free and we aren't + * keeping inode chunks permanently on disk, remove the record. + * Otherwise, update the record with the new information. + */ + if (rec.ir_freecount == mp->m_ialloc_inos && + !(mp->m_flags & XFS_MOUNT_IKEEP)) { + error = xfs_btree_delete(cur, &i); + if (error) + goto error; + ASSERT(i == 1); + } else { + error = xfs_inobt_update(cur, &rec); + if (error) + goto error; + } + +out: + error = xfs_check_agi_freecount(cur, agi); + if (error) + goto error; + + xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR); + return 0; + +error: + xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); + return error; +} + +/* + * Free disk inode. Carefully avoids touching the incore inode, all + * manipulations incore are the caller's responsibility. + * The on-disk inode is not changed by this operation, only the + * btree (free inode mask) is changed. + */ +int +xfs_difree( + struct xfs_trans *tp, /* transaction pointer */ + xfs_ino_t inode, /* inode to be freed */ + struct xfs_bmap_free *flist, /* extents to free */ + int *delete,/* set if inode cluster was deleted */ + xfs_ino_t *first_ino)/* first inode in deleted cluster */ +{ + /* REFERENCED */ + xfs_agblock_t agbno; /* block number containing inode */ + struct xfs_buf *agbp; /* buffer for allocation group header */ + xfs_agino_t agino; /* allocation group inode number */ + xfs_agnumber_t agno; /* allocation group number */ + int error; /* error return value */ + struct xfs_mount *mp; /* mount structure for filesystem */ + struct xfs_inobt_rec_incore rec;/* btree record */ + + mp = tp->t_mountp; + + /* + * Break up inode number into its components. + */ + agno = XFS_INO_TO_AGNO(mp, inode); + if (agno >= mp->m_sb.sb_agcount) { + xfs_warn(mp, "%s: agno >= mp->m_sb.sb_agcount (%d >= %d).", + __func__, agno, mp->m_sb.sb_agcount); + ASSERT(0); + return XFS_ERROR(EINVAL); + } + agino = XFS_INO_TO_AGINO(mp, inode); + if (inode != XFS_AGINO_TO_INO(mp, agno, agino)) { + xfs_warn(mp, "%s: inode != XFS_AGINO_TO_INO() (%llu != %llu).", + __func__, (unsigned long long)inode, + (unsigned long long)XFS_AGINO_TO_INO(mp, agno, agino)); + ASSERT(0); + return XFS_ERROR(EINVAL); + } + agbno = XFS_AGINO_TO_AGBNO(mp, agino); + if (agbno >= mp->m_sb.sb_agblocks) { + xfs_warn(mp, "%s: agbno >= mp->m_sb.sb_agblocks (%d >= %d).", + __func__, agbno, mp->m_sb.sb_agblocks); + ASSERT(0); + return XFS_ERROR(EINVAL); + } + /* + * Get the allocation group header. + */ + error = xfs_ialloc_read_agi(mp, tp, agno, &agbp); + if (error) { + xfs_warn(mp, "%s: xfs_ialloc_read_agi() returned error %d.", + __func__, error); + return error; + } + + /* + * Fix up the inode allocation btree. + */ + error = xfs_difree_inobt(mp, tp, agbp, agino, flist, delete, first_ino, + &rec); + if (error) + goto error0; + + /* + * Fix up the free inode btree. + */ + if (xfs_sb_version_hasfinobt(&mp->m_sb)) { + error = xfs_difree_finobt(mp, tp, agbp, agino, &rec); + if (error) + goto error0; + } + + return 0; + +error0: + return error; +} + STATIC int xfs_imap_lookup( struct xfs_mount *mp, @@ -1300,7 +1765,7 @@ xfs_imap_lookup( * we have a record, we need to ensure it contains the inode number * we are looking up. */ - cur = xfs_inobt_init_cursor(mp, tp, agbp, agno); + cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO); error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &i); if (!error) { if (i) @@ -1488,7 +1953,16 @@ xfs_ialloc_compute_maxlevels( } /* - * Log specified fields for the ag hdr (inode section) + * Log specified fields for the ag hdr (inode section). The growth of the agi + * structure over time requires that we interpret the buffer as two logical + * regions delineated by the end of the unlinked list. This is due to the size + * of the hash table and its location in the middle of the agi. + * + * For example, a request to log a field before agi_unlinked and a field after + * agi_unlinked could cause us to log the entire hash table and use an excessive + * amount of log space. To avoid this behavior, log the region up through + * agi_unlinked in one call and the region after agi_unlinked through the end of + * the structure in another. */ void xfs_ialloc_log_agi( @@ -1511,6 +1985,8 @@ xfs_ialloc_log_agi( offsetof(xfs_agi_t, agi_newino), offsetof(xfs_agi_t, agi_dirino), offsetof(xfs_agi_t, agi_unlinked), + offsetof(xfs_agi_t, agi_free_root), + offsetof(xfs_agi_t, agi_free_level), sizeof(xfs_agi_t) }; #ifdef DEBUG @@ -1519,15 +1995,30 @@ xfs_ialloc_log_agi( agi = XFS_BUF_TO_AGI(bp); ASSERT(agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC)); #endif + + xfs_trans_buf_set_type(tp, bp, XFS_BLFT_AGI_BUF); + /* - * Compute byte offsets for the first and last fields. + * Compute byte offsets for the first and last fields in the first + * region and log the agi buffer. This only logs up through + * agi_unlinked. */ - xfs_btree_offsets(fields, offsets, XFS_AGI_NUM_BITS, &first, &last); + if (fields & XFS_AGI_ALL_BITS_R1) { + xfs_btree_offsets(fields, offsets, XFS_AGI_NUM_BITS_R1, + &first, &last); + xfs_trans_log_buf(tp, bp, first, last); + } + /* - * Log the allocation group inode header buffer. + * Mask off the bits in the first region and calculate the first and + * last field offsets for any bits in the second region. */ - xfs_trans_buf_set_type(tp, bp, XFS_BLFT_AGI_BUF); - xfs_trans_log_buf(tp, bp, first, last); + fields &= ~XFS_AGI_ALL_BITS_R1; + if (fields) { + xfs_btree_offsets(fields, offsets, XFS_AGI_NUM_BITS_R2, + &first, &last); + xfs_trans_log_buf(tp, bp, first, last); + } } #ifdef DEBUG diff --git a/fs/xfs/xfs_ialloc_btree.c b/fs/xfs/xfs_ialloc_btree.c index 7e309b11e87d..726f83a681a5 100644 --- a/fs/xfs/xfs_ialloc_btree.c +++ b/fs/xfs/xfs_ialloc_btree.c @@ -49,7 +49,8 @@ xfs_inobt_dup_cursor( struct xfs_btree_cur *cur) { return xfs_inobt_init_cursor(cur->bc_mp, cur->bc_tp, - cur->bc_private.a.agbp, cur->bc_private.a.agno); + cur->bc_private.a.agbp, cur->bc_private.a.agno, + cur->bc_btnum); } STATIC void @@ -66,12 +67,26 @@ xfs_inobt_set_root( xfs_ialloc_log_agi(cur->bc_tp, agbp, XFS_AGI_ROOT | XFS_AGI_LEVEL); } +STATIC void +xfs_finobt_set_root( + struct xfs_btree_cur *cur, + union xfs_btree_ptr *nptr, + int inc) /* level change */ +{ + struct xfs_buf *agbp = cur->bc_private.a.agbp; + struct xfs_agi *agi = XFS_BUF_TO_AGI(agbp); + + agi->agi_free_root = nptr->s; + be32_add_cpu(&agi->agi_free_level, inc); + xfs_ialloc_log_agi(cur->bc_tp, agbp, + XFS_AGI_FREE_ROOT | XFS_AGI_FREE_LEVEL); +} + STATIC int xfs_inobt_alloc_block( struct xfs_btree_cur *cur, union xfs_btree_ptr *start, union xfs_btree_ptr *new, - int length, int *stat) { xfs_alloc_arg_t args; /* block allocation args */ @@ -173,6 +188,17 @@ xfs_inobt_init_ptr_from_cur( ptr->s = agi->agi_root; } +STATIC void +xfs_finobt_init_ptr_from_cur( + struct xfs_btree_cur *cur, + union xfs_btree_ptr *ptr) +{ + struct xfs_agi *agi = XFS_BUF_TO_AGI(cur->bc_private.a.agbp); + + ASSERT(cur->bc_private.a.agno == be32_to_cpu(agi->agi_seqno)); + ptr->s = agi->agi_free_root; +} + STATIC __int64_t xfs_inobt_key_diff( struct xfs_btree_cur *cur, @@ -203,6 +229,7 @@ xfs_inobt_verify( */ switch (block->bb_magic) { case cpu_to_be32(XFS_IBT_CRC_MAGIC): + case cpu_to_be32(XFS_FIBT_CRC_MAGIC): if (!xfs_sb_version_hascrc(&mp->m_sb)) return false; if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_uuid)) @@ -214,6 +241,7 @@ xfs_inobt_verify( return false; /* fall through */ case cpu_to_be32(XFS_IBT_MAGIC): + case cpu_to_be32(XFS_FIBT_MAGIC): break; default: return 0; @@ -317,6 +345,28 @@ static const struct xfs_btree_ops xfs_inobt_ops = { #endif }; +static const struct xfs_btree_ops xfs_finobt_ops = { + .rec_len = sizeof(xfs_inobt_rec_t), + .key_len = sizeof(xfs_inobt_key_t), + + .dup_cursor = xfs_inobt_dup_cursor, + .set_root = xfs_finobt_set_root, + .alloc_block = xfs_inobt_alloc_block, + .free_block = xfs_inobt_free_block, + .get_minrecs = xfs_inobt_get_minrecs, + .get_maxrecs = xfs_inobt_get_maxrecs, + .init_key_from_rec = xfs_inobt_init_key_from_rec, + .init_rec_from_key = xfs_inobt_init_rec_from_key, + .init_rec_from_cur = xfs_inobt_init_rec_from_cur, + .init_ptr_from_cur = xfs_finobt_init_ptr_from_cur, + .key_diff = xfs_inobt_key_diff, + .buf_ops = &xfs_inobt_buf_ops, +#if defined(DEBUG) || defined(XFS_WARN) + .keys_inorder = xfs_inobt_keys_inorder, + .recs_inorder = xfs_inobt_recs_inorder, +#endif +}; + /* * Allocate a new inode btree cursor. */ @@ -325,7 +375,8 @@ xfs_inobt_init_cursor( struct xfs_mount *mp, /* file system mount point */ struct xfs_trans *tp, /* transaction pointer */ struct xfs_buf *agbp, /* buffer for agi structure */ - xfs_agnumber_t agno) /* allocation group number */ + xfs_agnumber_t agno, /* allocation group number */ + xfs_btnum_t btnum) /* ialloc or free ino btree */ { struct xfs_agi *agi = XFS_BUF_TO_AGI(agbp); struct xfs_btree_cur *cur; @@ -334,11 +385,17 @@ xfs_inobt_init_cursor( cur->bc_tp = tp; cur->bc_mp = mp; - cur->bc_nlevels = be32_to_cpu(agi->agi_level); - cur->bc_btnum = XFS_BTNUM_INO; + cur->bc_btnum = btnum; + if (btnum == XFS_BTNUM_INO) { + cur->bc_nlevels = be32_to_cpu(agi->agi_level); + cur->bc_ops = &xfs_inobt_ops; + } else { + cur->bc_nlevels = be32_to_cpu(agi->agi_free_level); + cur->bc_ops = &xfs_finobt_ops; + } + cur->bc_blocklog = mp->m_sb.sb_blocklog; - cur->bc_ops = &xfs_inobt_ops; if (xfs_sb_version_hascrc(&mp->m_sb)) cur->bc_flags |= XFS_BTREE_CRC_BLOCKS; diff --git a/fs/xfs/xfs_ialloc_btree.h b/fs/xfs/xfs_ialloc_btree.h index f38b22011c4e..d7ebea72c2d0 100644 --- a/fs/xfs/xfs_ialloc_btree.h +++ b/fs/xfs/xfs_ialloc_btree.h @@ -58,7 +58,8 @@ struct xfs_mount; ((index) - 1) * sizeof(xfs_inobt_ptr_t))) extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *, - struct xfs_trans *, struct xfs_buf *, xfs_agnumber_t); + struct xfs_trans *, struct xfs_buf *, xfs_agnumber_t, + xfs_btnum_t); extern int xfs_inobt_maxrecs(struct xfs_mount *, int, int); #endif /* __XFS_IALLOC_BTREE_H__ */ diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index 98d35244eecc..c48df5f25b9f 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -507,8 +507,7 @@ STATIC int xfs_inode_ag_walk( struct xfs_mount *mp, struct xfs_perag *pag, - int (*execute)(struct xfs_inode *ip, - struct xfs_perag *pag, int flags, + int (*execute)(struct xfs_inode *ip, int flags, void *args), int flags, void *args, @@ -582,7 +581,7 @@ restart: for (i = 0; i < nr_found; i++) { if (!batch[i]) continue; - error = execute(batch[i], pag, flags, args); + error = execute(batch[i], flags, args); IRELE(batch[i]); if (error == EAGAIN) { skipped++; @@ -636,8 +635,7 @@ xfs_eofblocks_worker( int xfs_inode_ag_iterator( struct xfs_mount *mp, - int (*execute)(struct xfs_inode *ip, - struct xfs_perag *pag, int flags, + int (*execute)(struct xfs_inode *ip, int flags, void *args), int flags, void *args) @@ -664,8 +662,7 @@ xfs_inode_ag_iterator( int xfs_inode_ag_iterator_tag( struct xfs_mount *mp, - int (*execute)(struct xfs_inode *ip, - struct xfs_perag *pag, int flags, + int (*execute)(struct xfs_inode *ip, int flags, void *args), int flags, void *args, @@ -1209,7 +1206,6 @@ xfs_inode_match_id( STATIC int xfs_inode_free_eofblocks( struct xfs_inode *ip, - struct xfs_perag *pag, int flags, void *args) { diff --git a/fs/xfs/xfs_icache.h b/fs/xfs/xfs_icache.h index 9ed68bb750f5..9cf017b899be 100644 --- a/fs/xfs/xfs_icache.h +++ b/fs/xfs/xfs_icache.h @@ -60,12 +60,10 @@ int xfs_icache_free_eofblocks(struct xfs_mount *, struct xfs_eofblocks *); void xfs_eofblocks_worker(struct work_struct *); int xfs_inode_ag_iterator(struct xfs_mount *mp, - int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag, - int flags, void *args), + int (*execute)(struct xfs_inode *ip, int flags, void *args), int flags, void *args); int xfs_inode_ag_iterator_tag(struct xfs_mount *mp, - int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag, - int flags, void *args), + int (*execute)(struct xfs_inode *ip, int flags, void *args), int flags, void *args, int tag); static inline int diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 5e7a38fa6ee6..6d6b44a508f9 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -655,7 +655,6 @@ xfs_ialloc( uint flags; int error; timespec_t tv; - int filestreams = 0; /* * Call the space management code to pick @@ -772,13 +771,6 @@ xfs_ialloc( flags |= XFS_ILOG_DEV; break; case S_IFREG: - /* - * we can't set up filestreams until after the VFS inode - * is set up properly. - */ - if (pip && xfs_inode_is_filestream(pip)) - filestreams = 1; - /* fall through */ case S_IFDIR: if (pip && (pip->i_d.di_flags & XFS_DIFLAG_ANY)) { uint di_flags = 0; @@ -844,15 +836,6 @@ xfs_ialloc( /* now that we have an i_mode we can setup inode ops and unlock */ xfs_setup_inode(ip); - /* now we have set up the vfs inode we can associate the filestream */ - if (filestreams) { - error = xfs_filestream_associate(pip, ip); - if (error < 0) - return -error; - if (!error) - xfs_iflags_set(ip, XFS_IFILESTREAM); - } - *ipp = ip; return 0; } @@ -1334,7 +1317,8 @@ int xfs_create_tmpfile( struct xfs_inode *dp, struct dentry *dentry, - umode_t mode) + umode_t mode, + struct xfs_inode **ipp) { struct xfs_mount *mp = dp->i_mount; struct xfs_inode *ip = NULL; @@ -1402,7 +1386,6 @@ xfs_create_tmpfile( xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp, pdqp); ip->i_d.di_nlink--; - d_tmpfile(dentry, VFS_I(ip)); error = xfs_iunlink(tp, ip); if (error) goto out_trans_abort; @@ -1415,6 +1398,7 @@ xfs_create_tmpfile( xfs_qm_dqrele(gdqp); xfs_qm_dqrele(pdqp); + *ipp = ip; return 0; out_trans_abort: @@ -1698,16 +1682,6 @@ xfs_release( int truncated; /* - * If we are using filestreams, and we have an unlinked - * file that we are processing the last close on, then nothing - * will be able to reopen and write to this file. Purge this - * inode from the filestreams cache so that it doesn't delay - * teardown of the inode. - */ - if ((ip->i_d.di_nlink == 0) && xfs_inode_is_filestream(ip)) - xfs_filestream_deassociate(ip); - - /* * If we previously truncated this file and removed old data * in the process, we want to initiate "early" writeout on * the last close. This is an attempt to combat the notorious @@ -1837,9 +1811,33 @@ xfs_inactive_ifree( int error; tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE); - error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ifree, 0, 0); + + /* + * The ifree transaction might need to allocate blocks for record + * insertion to the finobt. We don't want to fail here at ENOSPC, so + * allow ifree to dip into the reserved block pool if necessary. + * + * Freeing large sets of inodes generally means freeing inode chunks, + * directory and file data blocks, so this should be relatively safe. + * Only under severe circumstances should it be possible to free enough + * inodes to exhaust the reserve block pool via finobt expansion while + * at the same time not creating free space in the filesystem. + * + * Send a warning if the reservation does happen to fail, as the inode + * now remains allocated and sits on the unlinked list until the fs is + * repaired. + */ + tp->t_flags |= XFS_TRANS_RESERVE; + error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ifree, + XFS_IFREE_SPACE_RES(mp), 0); if (error) { - ASSERT(XFS_FORCED_SHUTDOWN(mp)); + if (error == ENOSPC) { + xfs_warn_ratelimited(mp, + "Failed to remove inode(s) from unlinked list. " + "Please free space, unmount and run xfs_repair."); + } else { + ASSERT(XFS_FORCED_SHUTDOWN(mp)); + } xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES); return error; } @@ -2663,13 +2661,7 @@ xfs_remove( if (error) goto std_return; - /* - * If we are using filestreams, kill the stream association. - * If the file is still open it may get a new one but that - * will get killed on last close in xfs_close() so we don't - * have to worry about that. - */ - if (!is_dir && link_zero && xfs_inode_is_filestream(ip)) + if (is_dir && xfs_inode_is_filestream(ip)) xfs_filestream_deassociate(ip); return 0; @@ -3371,9 +3363,9 @@ xfs_iflush_int( } } - xfs_iflush_fork(ip, dip, iip, XFS_DATA_FORK, bp); + xfs_iflush_fork(ip, dip, iip, XFS_DATA_FORK); if (XFS_IFORK_Q(ip)) - xfs_iflush_fork(ip, dip, iip, XFS_ATTR_FORK, bp); + xfs_iflush_fork(ip, dip, iip, XFS_ATTR_FORK); xfs_inobp_check(mp, bp); /* diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 396cc1fafd0d..13aea548206c 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -209,7 +209,6 @@ xfs_get_initial_prid(struct xfs_inode *dp) #define XFS_ISTALE (1 << 1) /* inode has been staled */ #define XFS_IRECLAIMABLE (1 << 2) /* inode can be reclaimed */ #define XFS_INEW (1 << 3) /* inode has just been allocated */ -#define XFS_IFILESTREAM (1 << 4) /* inode is in a filestream dir. */ #define XFS_ITRUNCATED (1 << 5) /* truncated down so flush-on-close */ #define XFS_IDIRTY_RELEASE (1 << 6) /* dirty release already seen */ #define __XFS_IFLOCK_BIT 7 /* inode is being flushed right now */ @@ -225,8 +224,7 @@ xfs_get_initial_prid(struct xfs_inode *dp) */ #define XFS_IRECLAIM_RESET_FLAGS \ (XFS_IRECLAIMABLE | XFS_IRECLAIM | \ - XFS_IDIRTY_RELEASE | XFS_ITRUNCATED | \ - XFS_IFILESTREAM); + XFS_IDIRTY_RELEASE | XFS_ITRUNCATED) /* * Synchronize processes attempting to flush the in-core inode back to disk. @@ -334,7 +332,7 @@ int xfs_lookup(struct xfs_inode *dp, struct xfs_name *name, int xfs_create(struct xfs_inode *dp, struct xfs_name *name, umode_t mode, xfs_dev_t rdev, struct xfs_inode **ipp); int xfs_create_tmpfile(struct xfs_inode *dp, struct dentry *dentry, - umode_t mode); + umode_t mode, struct xfs_inode **ipp); 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, diff --git a/fs/xfs/xfs_inode_fork.c b/fs/xfs/xfs_inode_fork.c index 73514c0486b7..b031e8d0d928 100644 --- a/fs/xfs/xfs_inode_fork.c +++ b/fs/xfs/xfs_inode_fork.c @@ -798,8 +798,7 @@ xfs_iflush_fork( xfs_inode_t *ip, xfs_dinode_t *dip, xfs_inode_log_item_t *iip, - int whichfork, - xfs_buf_t *bp) + int whichfork) { char *cp; xfs_ifork_t *ifp; diff --git a/fs/xfs/xfs_inode_fork.h b/fs/xfs/xfs_inode_fork.h index eb329a1ea888..7d3b1ed6dcbe 100644 --- a/fs/xfs/xfs_inode_fork.h +++ b/fs/xfs/xfs_inode_fork.h @@ -127,8 +127,7 @@ typedef struct xfs_ifork { int xfs_iformat_fork(struct xfs_inode *, struct xfs_dinode *); void xfs_iflush_fork(struct xfs_inode *, struct xfs_dinode *, - struct xfs_inode_log_item *, int, - struct xfs_buf *); + struct xfs_inode_log_item *, int); void xfs_idestroy_fork(struct xfs_inode *, int); void xfs_idata_realloc(struct xfs_inode *, int, int); void xfs_iroot_realloc(struct xfs_inode *, int, int); diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 0b18776b075e..2d8f4fdf07f9 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -543,10 +543,11 @@ xfs_attrmulti_by_handle( ops = memdup_user(am_hreq.ops, size); if (IS_ERR(ops)) { - error = PTR_ERR(ops); + error = -PTR_ERR(ops); goto out_dput; } + error = ENOMEM; attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL); if (!attr_name) goto out_kfree_ops; @@ -556,7 +557,7 @@ xfs_attrmulti_by_handle( ops[i].am_error = strncpy_from_user((char *)attr_name, ops[i].am_attrname, MAXNAMELEN); if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN) - error = -ERANGE; + error = ERANGE; if (ops[i].am_error < 0) break; diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c index a7992f8de9d3..944d5baa710a 100644 --- a/fs/xfs/xfs_ioctl32.c +++ b/fs/xfs/xfs_ioctl32.c @@ -424,10 +424,11 @@ xfs_compat_attrmulti_by_handle( ops = memdup_user(compat_ptr(am_hreq.ops), size); if (IS_ERR(ops)) { - error = PTR_ERR(ops); + error = -PTR_ERR(ops); goto out_dput; } + error = ENOMEM; attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL); if (!attr_name) goto out_kfree_ops; @@ -438,7 +439,7 @@ xfs_compat_attrmulti_by_handle( compat_ptr(ops[i].am_attrname), MAXNAMELEN); if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN) - error = -ERANGE; + error = ERANGE; if (ops[i].am_error < 0) break; diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 3b80ebae05f5..6c5eb4c551e3 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -730,7 +730,7 @@ xfs_iomap_write_allocate( */ nimaps = 1; end_fsb = XFS_B_TO_FSB(mp, XFS_ISIZE(ip)); - error = xfs_bmap_last_offset(NULL, ip, &last_block, + error = xfs_bmap_last_offset(ip, &last_block, XFS_DATA_FORK); if (error) goto trans_cancel; diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index 89b07e43ca28..205613a06068 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -72,8 +72,8 @@ xfs_initxattrs( int error = 0; for (xattr = xattr_array; xattr->name != NULL; xattr++) { - error = xfs_attr_set(ip, xattr->name, xattr->value, - xattr->value_len, ATTR_SECURE); + error = -xfs_attr_set(ip, xattr->name, xattr->value, + xattr->value_len, ATTR_SECURE); if (error < 0) break; } @@ -93,8 +93,8 @@ xfs_init_security( struct inode *dir, const struct qstr *qstr) { - return security_inode_init_security(inode, dir, qstr, - &xfs_initxattrs, NULL); + return -security_inode_init_security(inode, dir, qstr, + &xfs_initxattrs, NULL); } static void @@ -124,15 +124,15 @@ xfs_cleanup_inode( xfs_dentry_to_name(&teardown, dentry, 0); xfs_remove(XFS_I(dir), &teardown, XFS_I(inode)); - iput(inode); } STATIC int -xfs_vn_mknod( +xfs_generic_create( struct inode *dir, struct dentry *dentry, umode_t mode, - dev_t rdev) + dev_t rdev, + bool tmpfile) /* unnamed file */ { struct inode *inode; struct xfs_inode *ip = NULL; @@ -156,8 +156,12 @@ xfs_vn_mknod( if (error) return error; - xfs_dentry_to_name(&name, dentry, mode); - error = xfs_create(XFS_I(dir), &name, mode, rdev, &ip); + if (!tmpfile) { + xfs_dentry_to_name(&name, dentry, mode); + error = xfs_create(XFS_I(dir), &name, mode, rdev, &ip); + } else { + error = xfs_create_tmpfile(XFS_I(dir), dentry, mode, &ip); + } if (unlikely(error)) goto out_free_acl; @@ -169,18 +173,22 @@ xfs_vn_mknod( #ifdef CONFIG_XFS_POSIX_ACL if (default_acl) { - error = xfs_set_acl(inode, default_acl, ACL_TYPE_DEFAULT); + error = -xfs_set_acl(inode, default_acl, ACL_TYPE_DEFAULT); if (error) goto out_cleanup_inode; } if (acl) { - error = xfs_set_acl(inode, acl, ACL_TYPE_ACCESS); + error = -xfs_set_acl(inode, acl, ACL_TYPE_ACCESS); if (error) goto out_cleanup_inode; } #endif - d_instantiate(dentry, inode); + if (tmpfile) + d_tmpfile(dentry, inode); + else + d_instantiate(dentry, inode); + out_free_acl: if (default_acl) posix_acl_release(default_acl); @@ -189,11 +197,23 @@ xfs_vn_mknod( return -error; out_cleanup_inode: - xfs_cleanup_inode(dir, inode, dentry); + if (!tmpfile) + xfs_cleanup_inode(dir, inode, dentry); + iput(inode); goto out_free_acl; } STATIC int +xfs_vn_mknod( + struct inode *dir, + struct dentry *dentry, + umode_t mode, + dev_t rdev) +{ + return xfs_generic_create(dir, dentry, mode, rdev, false); +} + +STATIC int xfs_vn_create( struct inode *dir, struct dentry *dentry, @@ -353,6 +373,7 @@ xfs_vn_symlink( out_cleanup_inode: xfs_cleanup_inode(dir, inode, dentry); + iput(inode); out: return -error; } @@ -808,22 +829,34 @@ xfs_setattr_size( */ inode_dio_wait(inode); + /* + * Do all the page cache truncate work outside the transaction context + * as the "lock" order is page lock->log space reservation. i.e. + * locking pages inside the transaction can ABBA deadlock with + * writeback. We have to do the VFS inode size update before we truncate + * the pagecache, however, to avoid racing with page faults beyond the + * new EOF they are not serialised against truncate operations except by + * page locks and size updates. + * + * Hence we are in a situation where a truncate can fail with ENOMEM + * from xfs_trans_reserve(), but having already truncated the in-memory + * version of the file (i.e. made user visible changes). There's not + * much we can do about this, except to hope that the caller sees ENOMEM + * and retries the truncate operation. + */ error = -block_truncate_page(inode->i_mapping, newsize, xfs_get_blocks); if (error) return error; + truncate_setsize(inode, newsize); tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_SIZE); error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0); if (error) goto out_trans_cancel; - truncate_setsize(inode, newsize); - commit_flags = XFS_TRANS_RELEASE_LOG_RES; lock_flags |= XFS_ILOCK_EXCL; - xfs_ilock(ip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip, 0); /* @@ -1053,11 +1086,7 @@ xfs_vn_tmpfile( struct dentry *dentry, umode_t mode) { - int error; - - error = xfs_create_tmpfile(XFS_I(dir), dentry, mode); - - return -error; + return xfs_generic_create(dir, dentry, mode, 0, true); } static const struct inode_operations xfs_inode_operations = { diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index f46338285152..cb64f222d607 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -270,7 +270,8 @@ xfs_bulkstat( /* * Allocate and initialize a btree cursor for ialloc btree. */ - cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno); + cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno, + XFS_BTNUM_INO); irbp = irbuf; irbufend = irbuf + nirbuf; end_of_ag = 0; @@ -621,7 +622,8 @@ xfs_inumbers( agino = 0; continue; } - cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno); + cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno, + XFS_BTNUM_INO); error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_GE, &tmp); if (error) { diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 8497a00e399d..3554098692d8 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -616,11 +616,13 @@ xfs_log_mount( int error = 0; int min_logfsbs; - if (!(mp->m_flags & XFS_MOUNT_NORECOVERY)) - xfs_notice(mp, "Mounting Filesystem"); - else { + if (!(mp->m_flags & XFS_MOUNT_NORECOVERY)) { + xfs_notice(mp, "Mounting V%d Filesystem", + XFS_SB_VERSION_NUM(&mp->m_sb)); + } else { xfs_notice(mp, -"Mounting filesystem in no-recovery mode. Filesystem will be inconsistent."); +"Mounting V%d filesystem in no-recovery mode. Filesystem will be inconsistent.", + XFS_SB_VERSION_NUM(&mp->m_sb)); ASSERT(mp->m_flags & XFS_MOUNT_RDONLY); } @@ -1181,11 +1183,14 @@ xlog_iodone(xfs_buf_t *bp) /* log I/O is always issued ASYNC */ ASSERT(XFS_BUF_ISASYNC(bp)); xlog_state_done_syncing(iclog, aborted); + /* - * do not reference the buffer (bp) here as we could race - * with it being freed after writing the unmount record to the - * log. + * drop the buffer lock now that we are done. Nothing references + * the buffer after this, so an unmount waiting on this lock can now + * tear it down safely. As such, it is unsafe to reference the buffer + * (bp) after the unlock as we could race with it being freed. */ + xfs_buf_unlock(bp); } /* @@ -1368,8 +1373,16 @@ xlog_alloc_log( bp = xfs_buf_alloc(mp->m_logdev_targp, 0, BTOBB(log->l_iclog_size), 0); if (!bp) goto out_free_log; - bp->b_iodone = xlog_iodone; + + /* + * The iclogbuf buffer locks are held over IO but we are not going to do + * IO yet. Hence unlock the buffer so that the log IO path can grab it + * when appropriately. + */ ASSERT(xfs_buf_islocked(bp)); + xfs_buf_unlock(bp); + + bp->b_iodone = xlog_iodone; log->l_xbuf = bp; spin_lock_init(&log->l_icloglock); @@ -1398,6 +1411,9 @@ xlog_alloc_log( if (!bp) goto out_free_iclog; + ASSERT(xfs_buf_islocked(bp)); + xfs_buf_unlock(bp); + bp->b_iodone = xlog_iodone; iclog->ic_bp = bp; iclog->ic_data = bp->b_addr; @@ -1422,7 +1438,6 @@ xlog_alloc_log( iclog->ic_callback_tail = &(iclog->ic_callback); iclog->ic_datap = (char *)iclog->ic_data + log->l_iclog_hsize; - ASSERT(xfs_buf_islocked(iclog->ic_bp)); init_waitqueue_head(&iclog->ic_force_wait); init_waitqueue_head(&iclog->ic_write_wait); @@ -1631,6 +1646,12 @@ xlog_cksum( * we transition the iclogs to IOERROR state *after* flushing all existing * iclogs to disk. This is because we don't want anymore new transactions to be * started or completed afterwards. + * + * We lock the iclogbufs here so that we can serialise against IO completion + * during unmount. We might be processing a shutdown triggered during unmount, + * and that can occur asynchronously to the unmount thread, and hence we need to + * ensure that completes before tearing down the iclogbufs. Hence we need to + * hold the buffer lock across the log IO to acheive that. */ STATIC int xlog_bdstrat( @@ -1638,6 +1659,7 @@ xlog_bdstrat( { struct xlog_in_core *iclog = bp->b_fspriv; + xfs_buf_lock(bp); if (iclog->ic_state & XLOG_STATE_IOERROR) { xfs_buf_ioerror(bp, EIO); xfs_buf_stale(bp); @@ -1645,7 +1667,8 @@ xlog_bdstrat( /* * It would seem logical to return EIO here, but we rely on * the log state machine to propagate I/O errors instead of - * doing it here. + * doing it here. Similarly, IO completion will unlock the + * buffer, so we don't do it here. */ return 0; } @@ -1847,14 +1870,28 @@ xlog_dealloc_log( xlog_cil_destroy(log); /* - * always need to ensure that the extra buffer does not point to memory - * owned by another log buffer before we free it. + * Cycle all the iclogbuf locks to make sure all log IO completion + * is done before we tear down these buffers. + */ + iclog = log->l_iclog; + for (i = 0; i < log->l_iclog_bufs; i++) { + xfs_buf_lock(iclog->ic_bp); + xfs_buf_unlock(iclog->ic_bp); + iclog = iclog->ic_next; + } + + /* + * Always need to ensure that the extra buffer does not point to memory + * owned by another log buffer before we free it. Also, cycle the lock + * first to ensure we've completed IO on it. */ + xfs_buf_lock(log->l_xbuf); + xfs_buf_unlock(log->l_xbuf); xfs_buf_set_empty(log->l_xbuf, BTOBB(log->l_iclog_size)); xfs_buf_free(log->l_xbuf); iclog = log->l_iclog; - for (i=0; i<log->l_iclog_bufs; i++) { + for (i = 0; i < log->l_iclog_bufs; i++) { xfs_buf_free(iclog->ic_bp); next_iclog = iclog->ic_next; kmem_free(iclog); @@ -3915,11 +3952,14 @@ xfs_log_force_umount( retval = xlog_state_ioerror(log); spin_unlock(&log->l_icloglock); } + /* - * Wake up everybody waiting on xfs_log_force. - * Callback all log item committed functions as if the - * log writes were completed. + * Wake up everybody waiting on xfs_log_force. Wake the CIL push first + * as if the log writes were completed. The abort handling in the log + * item committed callback functions will do this again under lock to + * avoid races. */ + wake_up_all(&log->l_cilp->xc_commit_wait); xlog_state_do_callback(log, XFS_LI_ABORTED, NULL); #ifdef XFSERRORDEBUG diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c index 7e5455391176..039c873e6fb2 100644 --- a/fs/xfs/xfs_log_cil.c +++ b/fs/xfs/xfs_log_cil.c @@ -385,7 +385,15 @@ xlog_cil_committed( xfs_extent_busy_clear(mp, &ctx->busy_extents, (mp->m_flags & XFS_MOUNT_DISCARD) && !abort); + /* + * If we are aborting the commit, wake up anyone waiting on the + * committing list. If we don't, then a shutdown we can leave processes + * waiting in xlog_cil_force_lsn() waiting on a sequence commit that + * will never happen because we aborted it. + */ spin_lock(&ctx->cil->xc_push_lock); + if (abort) + wake_up_all(&ctx->cil->xc_commit_wait); list_del(&ctx->committing); spin_unlock(&ctx->cil->xc_push_lock); @@ -564,8 +572,18 @@ restart: spin_lock(&cil->xc_push_lock); list_for_each_entry(new_ctx, &cil->xc_committing, committing) { /* + * Avoid getting stuck in this loop because we were woken by the + * shutdown, but then went back to sleep once already in the + * shutdown state. + */ + if (XLOG_FORCED_SHUTDOWN(log)) { + spin_unlock(&cil->xc_push_lock); + goto out_abort_free_ticket; + } + + /* * Higher sequences will wait for this one so skip them. - * Don't wait for own own sequence, either. + * Don't wait for our own sequence, either. */ if (new_ctx->sequence >= ctx->sequence) continue; @@ -810,6 +828,13 @@ restart: */ spin_lock(&cil->xc_push_lock); list_for_each_entry(ctx, &cil->xc_committing, committing) { + /* + * Avoid getting stuck in this loop because we were woken by the + * shutdown, but then went back to sleep once already in the + * shutdown state. + */ + if (XLOG_FORCED_SHUTDOWN(log)) + goto out_shutdown; if (ctx->sequence > sequence) continue; if (!ctx->commit_lsn) { @@ -833,14 +858,12 @@ restart: * push sequence after the above wait loop and the CIL still contains * dirty objects. * - * When the push occurs, it will empty the CIL and - * atomically increment the currect sequence past the push sequence and - * move it into the committing list. Of course, if the CIL is clean at - * the time of the push, it won't have pushed the CIL at all, so in that - * case we should try the push for this sequence again from the start - * just in case. + * When the push occurs, it will empty the CIL and atomically increment + * the currect sequence past the push sequence and move it into the + * committing list. Of course, if the CIL is clean at the time of the + * push, it won't have pushed the CIL at all, so in that case we should + * try the push for this sequence again from the start just in case. */ - if (sequence == cil->xc_current_sequence && !list_empty(&cil->xc_cil)) { spin_unlock(&cil->xc_push_lock); @@ -849,6 +872,17 @@ restart: spin_unlock(&cil->xc_push_lock); return commit_lsn; + + /* + * We detected a shutdown in progress. We need to trigger the log force + * to pass through it's iclog state machine error handling, even though + * we are already in a shutdown state. Hence we can't return + * NULLCOMMITLSN here as that has special meaning to log forces (i.e. + * LSN is already stable), so we return a zero LSN instead. + */ +out_shutdown: + spin_unlock(&cil->xc_push_lock); + return 0; } /* diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index bce53ac81096..981af0f6504b 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -2138,7 +2138,9 @@ xlog_recover_validate_buf_type( bp->b_ops = &xfs_allocbt_buf_ops; break; case XFS_IBT_CRC_MAGIC: + case XFS_FIBT_CRC_MAGIC: case XFS_IBT_MAGIC: + case XFS_FIBT_MAGIC: bp->b_ops = &xfs_inobt_buf_ops; break; case XFS_BMAP_CRC_MAGIC: @@ -3145,7 +3147,7 @@ xlog_recover_efd_pass2( } lip = xfs_trans_ail_cursor_next(ailp, &cur); } - xfs_trans_ail_cursor_done(ailp, &cur); + xfs_trans_ail_cursor_done(&cur); spin_unlock(&ailp->xa_lock); return 0; @@ -3520,8 +3522,7 @@ out: STATIC int xlog_recover_unmount_trans( - struct xlog *log, - struct xlog_recover *trans) + struct xlog *log) { /* Do nothing now */ xfs_warn(log->l_mp, "%s: Unmount LR", __func__); @@ -3595,7 +3596,7 @@ xlog_recover_process_data( trans, pass); break; case XLOG_UNMOUNT_TRANS: - error = xlog_recover_unmount_trans(log, trans); + error = xlog_recover_unmount_trans(log); break; case XLOG_WAS_CONT_TRANS: error = xlog_recover_add_to_cont_trans(log, @@ -3757,7 +3758,7 @@ xlog_recover_process_efis( lip = xfs_trans_ail_cursor_next(ailp, &cur); } out: - xfs_trans_ail_cursor_done(ailp, &cur); + xfs_trans_ail_cursor_done(&cur); spin_unlock(&ailp->xa_lock); return error; } diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 993cb19e7d39..944f3d9456a8 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -743,8 +743,6 @@ xfs_mountfs( new_size *= mp->m_sb.sb_inodesize / XFS_DINODE_MIN_SIZE; if (mp->m_sb.sb_inoalignmt >= XFS_B_TO_FSBT(mp, new_size)) mp->m_inode_cluster_size = new_size; - xfs_info(mp, "Using inode cluster size of %d bytes", - mp->m_inode_cluster_size); } /* diff --git a/fs/xfs/xfs_mru_cache.c b/fs/xfs/xfs_mru_cache.c index 4aff56395732..f99b4933dc22 100644 --- a/fs/xfs/xfs_mru_cache.c +++ b/fs/xfs/xfs_mru_cache.c @@ -100,14 +100,20 @@ * likely result in a loop in one of the lists. That's a sure-fire recipe for * an infinite loop in the code. */ -typedef struct xfs_mru_cache_elem -{ - struct list_head list_node; - unsigned long key; - void *value; -} xfs_mru_cache_elem_t; +struct xfs_mru_cache { + struct radix_tree_root store; /* Core storage data structure. */ + struct list_head *lists; /* Array of lists, one per grp. */ + struct list_head reap_list; /* Elements overdue for reaping. */ + spinlock_t lock; /* Lock to protect this struct. */ + unsigned int grp_count; /* Number of discrete groups. */ + unsigned int grp_time; /* Time period spanned by grps. */ + unsigned int lru_grp; /* Group containing time zero. */ + unsigned long time_zero; /* Time first element was added. */ + xfs_mru_cache_free_func_t free_func; /* Function pointer for freeing. */ + struct delayed_work work; /* Workqueue data for reaping. */ + unsigned int queued; /* work has been queued */ +}; -static kmem_zone_t *xfs_mru_elem_zone; static struct workqueue_struct *xfs_mru_reap_wq; /* @@ -129,12 +135,12 @@ static struct workqueue_struct *xfs_mru_reap_wq; */ STATIC unsigned long _xfs_mru_cache_migrate( - xfs_mru_cache_t *mru, - unsigned long now) + struct xfs_mru_cache *mru, + unsigned long now) { - unsigned int grp; - unsigned int migrated = 0; - struct list_head *lru_list; + unsigned int grp; + unsigned int migrated = 0; + struct list_head *lru_list; /* Nothing to do if the data store is empty. */ if (!mru->time_zero) @@ -193,11 +199,11 @@ _xfs_mru_cache_migrate( */ STATIC void _xfs_mru_cache_list_insert( - xfs_mru_cache_t *mru, - xfs_mru_cache_elem_t *elem) + struct xfs_mru_cache *mru, + struct xfs_mru_cache_elem *elem) { - unsigned int grp = 0; - unsigned long now = jiffies; + unsigned int grp = 0; + unsigned long now = jiffies; /* * If the data store is empty, initialise time zero, leave grp set to @@ -231,10 +237,10 @@ _xfs_mru_cache_list_insert( */ STATIC void _xfs_mru_cache_clear_reap_list( - xfs_mru_cache_t *mru) __releases(mru->lock) __acquires(mru->lock) - + struct xfs_mru_cache *mru) + __releases(mru->lock) __acquires(mru->lock) { - xfs_mru_cache_elem_t *elem, *next; + struct xfs_mru_cache_elem *elem, *next; struct list_head tmp; INIT_LIST_HEAD(&tmp); @@ -252,15 +258,8 @@ _xfs_mru_cache_clear_reap_list( spin_unlock(&mru->lock); list_for_each_entry_safe(elem, next, &tmp, list_node) { - - /* Remove the element from the reap list. */ list_del_init(&elem->list_node); - - /* Call the client's free function with the key and value pointer. */ - mru->free_func(elem->key, elem->value); - - /* Free the element structure. */ - kmem_zone_free(xfs_mru_elem_zone, elem); + mru->free_func(elem); } spin_lock(&mru->lock); @@ -277,7 +276,8 @@ STATIC void _xfs_mru_cache_reap( struct work_struct *work) { - xfs_mru_cache_t *mru = container_of(work, xfs_mru_cache_t, work.work); + struct xfs_mru_cache *mru = + container_of(work, struct xfs_mru_cache, work.work); unsigned long now, next; ASSERT(mru && mru->lists); @@ -304,28 +304,16 @@ _xfs_mru_cache_reap( int xfs_mru_cache_init(void) { - xfs_mru_elem_zone = kmem_zone_init(sizeof(xfs_mru_cache_elem_t), - "xfs_mru_cache_elem"); - if (!xfs_mru_elem_zone) - goto out; - xfs_mru_reap_wq = alloc_workqueue("xfs_mru_cache", WQ_MEM_RECLAIM, 1); if (!xfs_mru_reap_wq) - goto out_destroy_mru_elem_zone; - + return -ENOMEM; return 0; - - out_destroy_mru_elem_zone: - kmem_zone_destroy(xfs_mru_elem_zone); - out: - return -ENOMEM; } void xfs_mru_cache_uninit(void) { destroy_workqueue(xfs_mru_reap_wq); - kmem_zone_destroy(xfs_mru_elem_zone); } /* @@ -336,14 +324,14 @@ xfs_mru_cache_uninit(void) */ int xfs_mru_cache_create( - xfs_mru_cache_t **mrup, + struct xfs_mru_cache **mrup, unsigned int lifetime_ms, unsigned int grp_count, xfs_mru_cache_free_func_t free_func) { - xfs_mru_cache_t *mru = NULL; - int err = 0, grp; - unsigned int grp_time; + struct xfs_mru_cache *mru = NULL; + int err = 0, grp; + unsigned int grp_time; if (mrup) *mrup = NULL; @@ -400,7 +388,7 @@ exit: */ static void xfs_mru_cache_flush( - xfs_mru_cache_t *mru) + struct xfs_mru_cache *mru) { if (!mru || !mru->lists) return; @@ -420,7 +408,7 @@ xfs_mru_cache_flush( void xfs_mru_cache_destroy( - xfs_mru_cache_t *mru) + struct xfs_mru_cache *mru) { if (!mru || !mru->lists) return; @@ -438,38 +426,30 @@ xfs_mru_cache_destroy( */ int xfs_mru_cache_insert( - xfs_mru_cache_t *mru, - unsigned long key, - void *value) + struct xfs_mru_cache *mru, + unsigned long key, + struct xfs_mru_cache_elem *elem) { - xfs_mru_cache_elem_t *elem; + int error; ASSERT(mru && mru->lists); if (!mru || !mru->lists) return EINVAL; - elem = kmem_zone_zalloc(xfs_mru_elem_zone, KM_SLEEP); - if (!elem) + if (radix_tree_preload(GFP_KERNEL)) return ENOMEM; - if (radix_tree_preload(GFP_KERNEL)) { - kmem_zone_free(xfs_mru_elem_zone, elem); - return ENOMEM; - } - INIT_LIST_HEAD(&elem->list_node); elem->key = key; - elem->value = value; spin_lock(&mru->lock); - - radix_tree_insert(&mru->store, key, elem); + error = -radix_tree_insert(&mru->store, key, elem); radix_tree_preload_end(); - _xfs_mru_cache_list_insert(mru, elem); - + if (!error) + _xfs_mru_cache_list_insert(mru, elem); spin_unlock(&mru->lock); - return 0; + return error; } /* @@ -478,13 +458,12 @@ xfs_mru_cache_insert( * the client data pointer for the removed element is returned, otherwise this * function will return a NULL pointer. */ -void * +struct xfs_mru_cache_elem * xfs_mru_cache_remove( - xfs_mru_cache_t *mru, - unsigned long key) + struct xfs_mru_cache *mru, + unsigned long key) { - xfs_mru_cache_elem_t *elem; - void *value = NULL; + struct xfs_mru_cache_elem *elem; ASSERT(mru && mru->lists); if (!mru || !mru->lists) @@ -492,17 +471,11 @@ xfs_mru_cache_remove( spin_lock(&mru->lock); elem = radix_tree_delete(&mru->store, key); - if (elem) { - value = elem->value; + if (elem) list_del(&elem->list_node); - } - spin_unlock(&mru->lock); - if (elem) - kmem_zone_free(xfs_mru_elem_zone, elem); - - return value; + return elem; } /* @@ -511,13 +484,14 @@ xfs_mru_cache_remove( */ void xfs_mru_cache_delete( - xfs_mru_cache_t *mru, - unsigned long key) + struct xfs_mru_cache *mru, + unsigned long key) { - void *value = xfs_mru_cache_remove(mru, key); + struct xfs_mru_cache_elem *elem; - if (value) - mru->free_func(key, value); + elem = xfs_mru_cache_remove(mru, key); + if (elem) + mru->free_func(elem); } /* @@ -540,12 +514,12 @@ xfs_mru_cache_delete( * status, we need to help it get it right by annotating the path that does * not release the lock. */ -void * +struct xfs_mru_cache_elem * xfs_mru_cache_lookup( - xfs_mru_cache_t *mru, - unsigned long key) + struct xfs_mru_cache *mru, + unsigned long key) { - xfs_mru_cache_elem_t *elem; + struct xfs_mru_cache_elem *elem; ASSERT(mru && mru->lists); if (!mru || !mru->lists) @@ -560,7 +534,7 @@ xfs_mru_cache_lookup( } else spin_unlock(&mru->lock); - return elem ? elem->value : NULL; + return elem; } /* @@ -570,7 +544,8 @@ xfs_mru_cache_lookup( */ void xfs_mru_cache_done( - xfs_mru_cache_t *mru) __releases(mru->lock) + struct xfs_mru_cache *mru) + __releases(mru->lock) { spin_unlock(&mru->lock); } diff --git a/fs/xfs/xfs_mru_cache.h b/fs/xfs/xfs_mru_cache.h index 36dd3ec8b4eb..fb5245ba5ff7 100644 --- a/fs/xfs/xfs_mru_cache.h +++ b/fs/xfs/xfs_mru_cache.h @@ -18,24 +18,15 @@ #ifndef __XFS_MRU_CACHE_H__ #define __XFS_MRU_CACHE_H__ +struct xfs_mru_cache; -/* Function pointer type for callback to free a client's data pointer. */ -typedef void (*xfs_mru_cache_free_func_t)(unsigned long, void*); +struct xfs_mru_cache_elem { + struct list_head list_node; + unsigned long key; +}; -typedef struct xfs_mru_cache -{ - struct radix_tree_root store; /* Core storage data structure. */ - struct list_head *lists; /* Array of lists, one per grp. */ - struct list_head reap_list; /* Elements overdue for reaping. */ - spinlock_t lock; /* Lock to protect this struct. */ - unsigned int grp_count; /* Number of discrete groups. */ - unsigned int grp_time; /* Time period spanned by grps. */ - unsigned int lru_grp; /* Group containing time zero. */ - unsigned long time_zero; /* Time first element was added. */ - xfs_mru_cache_free_func_t free_func; /* Function pointer for freeing. */ - struct delayed_work work; /* Workqueue data for reaping. */ - unsigned int queued; /* work has been queued */ -} xfs_mru_cache_t; +/* Function pointer type for callback to free a client's data pointer. */ +typedef void (*xfs_mru_cache_free_func_t)(struct xfs_mru_cache_elem *elem); int xfs_mru_cache_init(void); void xfs_mru_cache_uninit(void); @@ -44,10 +35,12 @@ int xfs_mru_cache_create(struct xfs_mru_cache **mrup, unsigned int lifetime_ms, xfs_mru_cache_free_func_t free_func); void xfs_mru_cache_destroy(struct xfs_mru_cache *mru); int xfs_mru_cache_insert(struct xfs_mru_cache *mru, unsigned long key, - void *value); -void * xfs_mru_cache_remove(struct xfs_mru_cache *mru, unsigned long key); + struct xfs_mru_cache_elem *elem); +struct xfs_mru_cache_elem * +xfs_mru_cache_remove(struct xfs_mru_cache *mru, unsigned long key); void xfs_mru_cache_delete(struct xfs_mru_cache *mru, unsigned long key); -void *xfs_mru_cache_lookup(struct xfs_mru_cache *mru, unsigned long key); +struct xfs_mru_cache_elem * +xfs_mru_cache_lookup(struct xfs_mru_cache *mru, unsigned long key); void xfs_mru_cache_done(struct xfs_mru_cache *mru); #endif /* __XFS_MRU_CACHE_H__ */ diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c index 348e4d2ed6e6..6d26759c779a 100644 --- a/fs/xfs/xfs_qm.c +++ b/fs/xfs/xfs_qm.c @@ -193,47 +193,6 @@ xfs_qm_dqpurge( } /* - * Release the group or project dquot pointers the user dquots maybe carrying - * around as a hint, and proceed to purge the user dquot cache if requested. -*/ -STATIC int -xfs_qm_dqpurge_hints( - struct xfs_dquot *dqp, - void *data) -{ - struct xfs_dquot *gdqp = NULL; - struct xfs_dquot *pdqp = NULL; - uint flags = *((uint *)data); - - xfs_dqlock(dqp); - if (dqp->dq_flags & XFS_DQ_FREEING) { - xfs_dqunlock(dqp); - return EAGAIN; - } - - /* If this quota has a hint attached, prepare for releasing it now */ - gdqp = dqp->q_gdquot; - if (gdqp) - dqp->q_gdquot = NULL; - - pdqp = dqp->q_pdquot; - if (pdqp) - dqp->q_pdquot = NULL; - - xfs_dqunlock(dqp); - - if (gdqp) - xfs_qm_dqrele(gdqp); - if (pdqp) - xfs_qm_dqrele(pdqp); - - if (flags & XFS_QMOPT_UQUOTA) - return xfs_qm_dqpurge(dqp, NULL); - - return 0; -} - -/* * Purge the dquot cache. */ void @@ -241,18 +200,8 @@ xfs_qm_dqpurge_all( struct xfs_mount *mp, uint flags) { - /* - * We have to release group/project dquot hint(s) from the user dquot - * at first if they are there, otherwise we would run into an infinite - * loop while walking through radix tree to purge other type of dquots - * since their refcount is not zero if the user dquot refers to them - * as hint. - * - * Call the special xfs_qm_dqpurge_hints() will end up go through the - * general xfs_qm_dqpurge() against user dquot cache if requested. - */ - xfs_qm_dquot_walk(mp, XFS_DQ_USER, xfs_qm_dqpurge_hints, &flags); - + if (flags & XFS_QMOPT_UQUOTA) + xfs_qm_dquot_walk(mp, XFS_DQ_USER, xfs_qm_dqpurge, NULL); if (flags & XFS_QMOPT_GQUOTA) xfs_qm_dquot_walk(mp, XFS_DQ_GROUP, xfs_qm_dqpurge, NULL); if (flags & XFS_QMOPT_PQUOTA) @@ -409,7 +358,6 @@ xfs_qm_dqattach_one( xfs_dqid_t id, uint type, uint doalloc, - xfs_dquot_t *udqhint, /* hint */ xfs_dquot_t **IO_idqpp) { xfs_dquot_t *dqp; @@ -419,9 +367,9 @@ xfs_qm_dqattach_one( error = 0; /* - * See if we already have it in the inode itself. IO_idqpp is - * &i_udquot or &i_gdquot. This made the code look weird, but - * made the logic a lot simpler. + * See if we already have it in the inode itself. IO_idqpp is &i_udquot + * or &i_gdquot. This made the code look weird, but made the logic a lot + * simpler. */ dqp = *IO_idqpp; if (dqp) { @@ -430,49 +378,10 @@ xfs_qm_dqattach_one( } /* - * udqhint is the i_udquot field in inode, and is non-NULL only - * when the type arg is group/project. Its purpose is to save a - * lookup by dqid (xfs_qm_dqget) by caching a group dquot inside - * the user dquot. - */ - if (udqhint) { - ASSERT(type == XFS_DQ_GROUP || type == XFS_DQ_PROJ); - xfs_dqlock(udqhint); - - /* - * No need to take dqlock to look at the id. - * - * The ID can't change until it gets reclaimed, and it won't - * be reclaimed as long as we have a ref from inode and we - * hold the ilock. - */ - if (type == XFS_DQ_GROUP) - dqp = udqhint->q_gdquot; - else - dqp = udqhint->q_pdquot; - if (dqp && be32_to_cpu(dqp->q_core.d_id) == id) { - ASSERT(*IO_idqpp == NULL); - - *IO_idqpp = xfs_qm_dqhold(dqp); - xfs_dqunlock(udqhint); - return 0; - } - - /* - * We can't hold a dquot lock when we call the dqget code. - * We'll deadlock in no time, because of (not conforming to) - * lock ordering - the inodelock comes before any dquot lock, - * and we may drop and reacquire the ilock in xfs_qm_dqget(). - */ - xfs_dqunlock(udqhint); - } - - /* - * Find the dquot from somewhere. This bumps the - * reference count of dquot and returns it locked. - * This can return ENOENT if dquot didn't exist on - * disk and we didn't ask it to allocate; - * ESRCH if quotas got turned off suddenly. + * Find the dquot from somewhere. This bumps the reference count of + * dquot and returns it locked. This can return ENOENT if dquot didn't + * exist on disk and we didn't ask it to allocate; ESRCH if quotas got + * turned off suddenly. */ error = xfs_qm_dqget(ip->i_mount, ip, id, type, doalloc | XFS_QMOPT_DOWARN, &dqp); @@ -490,48 +399,6 @@ xfs_qm_dqattach_one( return 0; } - -/* - * Given a udquot and group/project type, attach the group/project - * dquot pointer to the udquot as a hint for future lookups. - */ -STATIC void -xfs_qm_dqattach_hint( - struct xfs_inode *ip, - int type) -{ - struct xfs_dquot **dqhintp; - struct xfs_dquot *dqp; - struct xfs_dquot *udq = ip->i_udquot; - - ASSERT(type == XFS_DQ_GROUP || type == XFS_DQ_PROJ); - - xfs_dqlock(udq); - - if (type == XFS_DQ_GROUP) { - dqp = ip->i_gdquot; - dqhintp = &udq->q_gdquot; - } else { - dqp = ip->i_pdquot; - dqhintp = &udq->q_pdquot; - } - - if (*dqhintp) { - struct xfs_dquot *tmp; - - if (*dqhintp == dqp) - goto done; - - tmp = *dqhintp; - *dqhintp = NULL; - xfs_qm_dqrele(tmp); - } - - *dqhintp = xfs_qm_dqhold(dqp); -done: - xfs_dqunlock(udq); -} - static bool xfs_qm_need_dqattach( struct xfs_inode *ip) @@ -562,7 +429,6 @@ xfs_qm_dqattach_locked( uint flags) { xfs_mount_t *mp = ip->i_mount; - uint nquotas = 0; int error = 0; if (!xfs_qm_need_dqattach(ip)) @@ -570,77 +436,39 @@ xfs_qm_dqattach_locked( ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - if (XFS_IS_UQUOTA_ON(mp)) { + if (XFS_IS_UQUOTA_ON(mp) && !ip->i_udquot) { error = xfs_qm_dqattach_one(ip, ip->i_d.di_uid, XFS_DQ_USER, flags & XFS_QMOPT_DQALLOC, - NULL, &ip->i_udquot); + &ip->i_udquot); if (error) goto done; - nquotas++; + ASSERT(ip->i_udquot); } - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - if (XFS_IS_GQUOTA_ON(mp)) { + if (XFS_IS_GQUOTA_ON(mp) && !ip->i_gdquot) { error = xfs_qm_dqattach_one(ip, ip->i_d.di_gid, XFS_DQ_GROUP, flags & XFS_QMOPT_DQALLOC, - ip->i_udquot, &ip->i_gdquot); - /* - * Don't worry about the udquot that we may have - * attached above. It'll get detached, if not already. - */ + &ip->i_gdquot); if (error) goto done; - nquotas++; + ASSERT(ip->i_gdquot); } - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - if (XFS_IS_PQUOTA_ON(mp)) { + if (XFS_IS_PQUOTA_ON(mp) && !ip->i_pdquot) { error = xfs_qm_dqattach_one(ip, xfs_get_projid(ip), XFS_DQ_PROJ, flags & XFS_QMOPT_DQALLOC, - ip->i_udquot, &ip->i_pdquot); - /* - * Don't worry about the udquot that we may have - * attached above. It'll get detached, if not already. - */ + &ip->i_pdquot); if (error) goto done; - nquotas++; + ASSERT(ip->i_pdquot); } +done: /* - * Attach this group/project quota to the user quota as a hint. - * This WON'T, in general, result in a thrash. + * Don't worry about the dquots that we may have attached before any + * error - they'll get detached later if it has not already been done. */ - if (nquotas > 1 && ip->i_udquot) { - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - ASSERT(ip->i_gdquot || !XFS_IS_GQUOTA_ON(mp)); - ASSERT(ip->i_pdquot || !XFS_IS_PQUOTA_ON(mp)); - - /* - * We do not have i_udquot locked at this point, but this check - * is OK since we don't depend on the i_gdquot to be accurate - * 100% all the time. It is just a hint, and this will - * succeed in general. - */ - if (ip->i_udquot->q_gdquot != ip->i_gdquot) - xfs_qm_dqattach_hint(ip, XFS_DQ_GROUP); - - if (ip->i_udquot->q_pdquot != ip->i_pdquot) - xfs_qm_dqattach_hint(ip, XFS_DQ_PROJ); - } - - done: -#ifdef DEBUG - if (!error) { - if (XFS_IS_UQUOTA_ON(mp)) - ASSERT(ip->i_udquot); - if (XFS_IS_GQUOTA_ON(mp)) - ASSERT(ip->i_gdquot); - if (XFS_IS_PQUOTA_ON(mp)) - ASSERT(ip->i_pdquot); - } ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); -#endif return error; } @@ -843,22 +671,17 @@ xfs_qm_init_quotainfo( qinf = mp->m_quotainfo = kmem_zalloc(sizeof(xfs_quotainfo_t), KM_SLEEP); - if ((error = list_lru_init(&qinf->qi_lru))) { - kmem_free(qinf); - mp->m_quotainfo = NULL; - return error; - } + error = -list_lru_init(&qinf->qi_lru); + if (error) + goto out_free_qinf; /* * See if quotainodes are setup, and if not, allocate them, * and change the superblock accordingly. */ - if ((error = xfs_qm_init_quotainos(mp))) { - list_lru_destroy(&qinf->qi_lru); - kmem_free(qinf); - mp->m_quotainfo = NULL; - return error; - } + error = xfs_qm_init_quotainos(mp); + if (error) + goto out_free_lru; INIT_RADIX_TREE(&qinf->qi_uquota_tree, GFP_NOFS); INIT_RADIX_TREE(&qinf->qi_gquota_tree, GFP_NOFS); @@ -870,8 +693,7 @@ xfs_qm_init_quotainfo( /* Precalc some constants */ qinf->qi_dqchunklen = XFS_FSB_TO_BB(mp, XFS_DQUOT_CLUSTER_SIZE_FSB); - qinf->qi_dqperchunk = xfs_calc_dquots_per_chunk(mp, - qinf->qi_dqchunklen); + qinf->qi_dqperchunk = xfs_calc_dquots_per_chunk(qinf->qi_dqchunklen); mp->m_qflags |= (mp->m_sb.sb_qflags & XFS_ALL_QUOTA_CHKD); @@ -918,7 +740,7 @@ xfs_qm_init_quotainfo( qinf->qi_isoftlimit = be64_to_cpu(ddqp->d_ino_softlimit); qinf->qi_rtbhardlimit = be64_to_cpu(ddqp->d_rtb_hardlimit); qinf->qi_rtbsoftlimit = be64_to_cpu(ddqp->d_rtb_softlimit); - + xfs_qm_dqdestroy(dqp); } else { qinf->qi_btimelimit = XFS_QM_BTIMELIMIT; @@ -935,6 +757,13 @@ xfs_qm_init_quotainfo( qinf->qi_shrinker.flags = SHRINKER_NUMA_AWARE; register_shrinker(&qinf->qi_shrinker); return 0; + +out_free_lru: + list_lru_destroy(&qinf->qi_lru); +out_free_qinf: + kmem_free(qinf); + mp->m_quotainfo = NULL; + return error; } diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c index 3daf5ea1eb8d..bbc813caba4c 100644 --- a/fs/xfs/xfs_qm_syscalls.c +++ b/fs/xfs/xfs_qm_syscalls.c @@ -278,9 +278,10 @@ xfs_qm_scall_trunc_qfiles( xfs_mount_t *mp, uint flags) { - int error; + int error = EINVAL; - if (!xfs_sb_version_hasquota(&mp->m_sb) || flags == 0) { + if (!xfs_sb_version_hasquota(&mp->m_sb) || flags == 0 || + (flags & ~XFS_DQ_ALLTYPES)) { xfs_debug(mp, "%s: flags=%x m_qflags=%x", __func__, flags, mp->m_qflags); return XFS_ERROR(EINVAL); @@ -959,7 +960,6 @@ xfs_qm_export_flags( STATIC int xfs_dqrele_inode( struct xfs_inode *ip, - struct xfs_perag *pag, int flags, void *args) { diff --git a/fs/xfs/xfs_quota_defs.h b/fs/xfs/xfs_quota_defs.h index b3b2b1065c0f..137e20937077 100644 --- a/fs/xfs/xfs_quota_defs.h +++ b/fs/xfs/xfs_quota_defs.h @@ -156,6 +156,6 @@ typedef __uint16_t xfs_qwarncnt_t; extern int xfs_dqcheck(struct xfs_mount *mp, xfs_disk_dquot_t *ddq, xfs_dqid_t id, uint type, uint flags, char *str); -extern int xfs_calc_dquots_per_chunk(struct xfs_mount *mp, unsigned int nbblks); +extern int xfs_calc_dquots_per_chunk(unsigned int nbblks); #endif /* __XFS_QUOTA_H__ */ diff --git a/fs/xfs/xfs_quotaops.c b/fs/xfs/xfs_quotaops.c index af33cafe69b6..2ad1b9822e92 100644 --- a/fs/xfs/xfs_quotaops.c +++ b/fs/xfs/xfs_quotaops.c @@ -100,16 +100,36 @@ xfs_fs_set_xstate( if (!XFS_IS_QUOTA_ON(mp)) return -EINVAL; return -xfs_qm_scall_quotaoff(mp, flags); - case Q_XQUOTARM: - if (XFS_IS_QUOTA_ON(mp)) - return -EINVAL; - return -xfs_qm_scall_trunc_qfiles(mp, flags); } return -EINVAL; } STATIC int +xfs_fs_rm_xquota( + struct super_block *sb, + unsigned int uflags) +{ + struct xfs_mount *mp = XFS_M(sb); + unsigned int flags = 0; + + if (sb->s_flags & MS_RDONLY) + return -EROFS; + + if (XFS_IS_QUOTA_ON(mp)) + return -EINVAL; + + if (uflags & FS_USER_QUOTA) + flags |= XFS_DQ_USER; + if (uflags & FS_GROUP_QUOTA) + flags |= XFS_DQ_GROUP; + if (uflags & FS_USER_QUOTA) + flags |= XFS_DQ_PROJ; + + return -xfs_qm_scall_trunc_qfiles(mp, flags); +} + +STATIC int xfs_fs_get_dqblk( struct super_block *sb, struct kqid qid, @@ -149,6 +169,7 @@ const struct quotactl_ops xfs_quotactl_operations = { .get_xstatev = xfs_fs_get_xstatev, .get_xstate = xfs_fs_get_xstate, .set_xstate = xfs_fs_set_xstate, + .rm_xquota = xfs_fs_rm_xquota, .get_dqblk = xfs_fs_get_dqblk, .set_dqblk = xfs_fs_set_dqblk, }; diff --git a/fs/xfs/xfs_sb.c b/fs/xfs/xfs_sb.c index 0c0e41bbe4e3..8baf61afae1d 100644 --- a/fs/xfs/xfs_sb.c +++ b/fs/xfs/xfs_sb.c @@ -201,10 +201,6 @@ xfs_mount_validate_sb( * write validation, we don't need to check feature masks. */ 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!"); - if (xfs_sb_has_compat_feature(sbp, XFS_SB_FEAT_COMPAT_UNKNOWN)) { xfs_warn(mp, diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/xfs_sb.h index f7b2fe77c5a5..950d1ea058b2 100644 --- a/fs/xfs/xfs_sb.h +++ b/fs/xfs/xfs_sb.h @@ -587,7 +587,9 @@ xfs_sb_has_compat_feature( return (sbp->sb_features_compat & feature) != 0; } -#define XFS_SB_FEAT_RO_COMPAT_ALL 0 +#define XFS_SB_FEAT_RO_COMPAT_FINOBT (1 << 0) /* free inode btree */ +#define XFS_SB_FEAT_RO_COMPAT_ALL \ + (XFS_SB_FEAT_RO_COMPAT_FINOBT) #define XFS_SB_FEAT_RO_COMPAT_UNKNOWN ~XFS_SB_FEAT_RO_COMPAT_ALL static inline bool xfs_sb_has_ro_compat_feature( @@ -641,6 +643,12 @@ static inline int xfs_sb_version_hasftype(struct xfs_sb *sbp) (sbp->sb_features2 & XFS_SB_VERSION2_FTYPE)); } +static inline int xfs_sb_version_hasfinobt(xfs_sb_t *sbp) +{ + return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) && + (sbp->sb_features_ro_compat & XFS_SB_FEAT_RO_COMPAT_FINOBT); +} + /* * end of superblock version macros */ diff --git a/fs/xfs/xfs_shared.h b/fs/xfs/xfs_shared.h index 4484e5151395..82404da2ca67 100644 --- a/fs/xfs/xfs_shared.h +++ b/fs/xfs/xfs_shared.h @@ -238,7 +238,7 @@ int xfs_log_calc_minimum_size(struct xfs_mount *); int xfs_symlink_blocks(struct xfs_mount *mp, int pathlen); int xfs_symlink_hdr_set(struct xfs_mount *mp, xfs_ino_t ino, uint32_t offset, uint32_t size, struct xfs_buf *bp); -bool xfs_symlink_hdr_ok(struct xfs_mount *mp, xfs_ino_t ino, uint32_t offset, +bool xfs_symlink_hdr_ok(xfs_ino_t ino, uint32_t offset, uint32_t size, struct xfs_buf *bp); void xfs_symlink_local_to_remote(struct xfs_trans *tp, struct xfs_buf *bp, struct xfs_inode *ip, struct xfs_ifork *ifp); diff --git a/fs/xfs/xfs_stats.c b/fs/xfs/xfs_stats.c index ce372b7d5644..f2240383d4bb 100644 --- a/fs/xfs/xfs_stats.c +++ b/fs/xfs/xfs_stats.c @@ -59,6 +59,7 @@ static int xfs_stat_proc_show(struct seq_file *m, void *v) { "abtc2", XFSSTAT_END_ABTC_V2 }, { "bmbt2", XFSSTAT_END_BMBT_V2 }, { "ibt2", XFSSTAT_END_IBT_V2 }, + { "fibt2", XFSSTAT_END_FIBT_V2 }, /* we print both series of quota information together */ { "qm", XFSSTAT_END_QM }, }; diff --git a/fs/xfs/xfs_stats.h b/fs/xfs/xfs_stats.h index c03ad38ceaeb..c8f238b8299a 100644 --- a/fs/xfs/xfs_stats.h +++ b/fs/xfs/xfs_stats.h @@ -183,7 +183,23 @@ struct xfsstats { __uint32_t xs_ibt_2_alloc; __uint32_t xs_ibt_2_free; __uint32_t xs_ibt_2_moves; -#define XFSSTAT_END_XQMSTAT (XFSSTAT_END_IBT_V2+6) +#define XFSSTAT_END_FIBT_V2 (XFSSTAT_END_IBT_V2+15) + __uint32_t xs_fibt_2_lookup; + __uint32_t xs_fibt_2_compare; + __uint32_t xs_fibt_2_insrec; + __uint32_t xs_fibt_2_delrec; + __uint32_t xs_fibt_2_newroot; + __uint32_t xs_fibt_2_killroot; + __uint32_t xs_fibt_2_increment; + __uint32_t xs_fibt_2_decrement; + __uint32_t xs_fibt_2_lshift; + __uint32_t xs_fibt_2_rshift; + __uint32_t xs_fibt_2_split; + __uint32_t xs_fibt_2_join; + __uint32_t xs_fibt_2_alloc; + __uint32_t xs_fibt_2_free; + __uint32_t xs_fibt_2_moves; +#define XFSSTAT_END_XQMSTAT (XFSSTAT_END_FIBT_V2+6) __uint32_t xs_qm_dqreclaims; __uint32_t xs_qm_dqreclaim_misses; __uint32_t xs_qm_dquot_dups; diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 205376776377..8f0333b3f7a0 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -765,20 +765,18 @@ xfs_open_devices( * Setup xfs_mount buffer target pointers */ error = ENOMEM; - mp->m_ddev_targp = xfs_alloc_buftarg(mp, ddev, 0, mp->m_fsname); + mp->m_ddev_targp = xfs_alloc_buftarg(mp, ddev); if (!mp->m_ddev_targp) goto out_close_rtdev; if (rtdev) { - mp->m_rtdev_targp = xfs_alloc_buftarg(mp, rtdev, 1, - mp->m_fsname); + mp->m_rtdev_targp = xfs_alloc_buftarg(mp, rtdev); if (!mp->m_rtdev_targp) goto out_free_ddev_targ; } if (logdev && logdev != ddev) { - mp->m_logdev_targp = xfs_alloc_buftarg(mp, logdev, 1, - mp->m_fsname); + mp->m_logdev_targp = xfs_alloc_buftarg(mp, logdev); if (!mp->m_logdev_targp) goto out_free_rtdev_targ; } else { @@ -811,8 +809,7 @@ xfs_setup_devices( { int error; - error = xfs_setsize_buftarg(mp->m_ddev_targp, mp->m_sb.sb_blocksize, - mp->m_sb.sb_sectsize); + error = xfs_setsize_buftarg(mp->m_ddev_targp, mp->m_sb.sb_sectsize); if (error) return error; @@ -822,14 +819,12 @@ xfs_setup_devices( if (xfs_sb_version_hassector(&mp->m_sb)) log_sector_size = mp->m_sb.sb_logsectsize; error = xfs_setsize_buftarg(mp->m_logdev_targp, - mp->m_sb.sb_blocksize, log_sector_size); if (error) return error; } if (mp->m_rtdev_targp) { error = xfs_setsize_buftarg(mp->m_rtdev_targp, - mp->m_sb.sb_blocksize, mp->m_sb.sb_sectsize); if (error) return error; @@ -1433,11 +1428,11 @@ xfs_fs_fill_super( if (error) goto out_free_fsname; - error = xfs_init_mount_workqueues(mp); + error = -xfs_init_mount_workqueues(mp); if (error) goto out_close_devices; - error = xfs_icsb_init_counters(mp); + error = -xfs_icsb_init_counters(mp); if (error) goto out_destroy_workqueues; @@ -1754,13 +1749,9 @@ init_xfs_fs(void) if (error) goto out_destroy_wq; - error = xfs_filestream_init(); - if (error) - goto out_mru_cache_uninit; - error = xfs_buf_init(); if (error) - goto out_filestream_uninit; + goto out_mru_cache_uninit; error = xfs_init_procfs(); if (error) @@ -1787,8 +1778,6 @@ init_xfs_fs(void) xfs_cleanup_procfs(); out_buf_terminate: xfs_buf_terminate(); - out_filestream_uninit: - xfs_filestream_uninit(); out_mru_cache_uninit: xfs_mru_cache_uninit(); out_destroy_wq: @@ -1807,7 +1796,6 @@ exit_xfs_fs(void) xfs_sysctl_unregister(); xfs_cleanup_procfs(); xfs_buf_terminate(); - xfs_filestream_uninit(); xfs_mru_cache_uninit(); xfs_destroy_workqueues(); xfs_destroy_zones(); diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c index 52979aa90986..0816b4018dfc 100644 --- a/fs/xfs/xfs_symlink.c +++ b/fs/xfs/xfs_symlink.c @@ -92,7 +92,7 @@ xfs_readlink_bmap( cur_chunk = bp->b_addr; if (xfs_sb_version_hascrc(&mp->m_sb)) { - if (!xfs_symlink_hdr_ok(mp, ip->i_ino, offset, + if (!xfs_symlink_hdr_ok(ip->i_ino, offset, byte_cnt, bp)) { error = EFSCORRUPTED; xfs_alert(mp, diff --git a/fs/xfs/xfs_symlink_remote.c b/fs/xfs/xfs_symlink_remote.c index 9b32052ff65e..23c2f2577c8d 100644 --- a/fs/xfs/xfs_symlink_remote.c +++ b/fs/xfs/xfs_symlink_remote.c @@ -80,7 +80,6 @@ xfs_symlink_hdr_set( */ bool xfs_symlink_hdr_ok( - struct xfs_mount *mp, xfs_ino_t ino, uint32_t offset, uint32_t size, diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c index dee3279c095e..1e85bcd0e418 100644 --- a/fs/xfs/xfs_trace.c +++ b/fs/xfs/xfs_trace.c @@ -46,6 +46,7 @@ #include "xfs_log_recover.h" #include "xfs_inode_item.h" #include "xfs_bmap_btree.h" +#include "xfs_filestream.h" /* * We include this last to have the helpers above available for the trace diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index a4ae41c179a8..6910458915cf 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -538,6 +538,64 @@ DEFINE_BUF_ITEM_EVENT(xfs_trans_bhold_release); DEFINE_BUF_ITEM_EVENT(xfs_trans_binval); DEFINE_BUF_ITEM_EVENT(xfs_trans_buf_ordered); +DECLARE_EVENT_CLASS(xfs_filestream_class, + TP_PROTO(struct xfs_inode *ip, xfs_agnumber_t agno), + TP_ARGS(ip, agno), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_agnumber_t, agno) + __field(int, streams) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->ino = ip->i_ino; + __entry->agno = agno; + __entry->streams = xfs_filestream_peek_ag(ip->i_mount, agno); + ), + TP_printk("dev %d:%d ino 0x%llx agno %u streams %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->agno, + __entry->streams) +) +#define DEFINE_FILESTREAM_EVENT(name) \ +DEFINE_EVENT(xfs_filestream_class, name, \ + TP_PROTO(struct xfs_inode *ip, xfs_agnumber_t agno), \ + TP_ARGS(ip, agno)) +DEFINE_FILESTREAM_EVENT(xfs_filestream_free); +DEFINE_FILESTREAM_EVENT(xfs_filestream_lookup); +DEFINE_FILESTREAM_EVENT(xfs_filestream_scan); + +TRACE_EVENT(xfs_filestream_pick, + TP_PROTO(struct xfs_inode *ip, xfs_agnumber_t agno, + xfs_extlen_t free, int nscan), + TP_ARGS(ip, agno, free, nscan), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_agnumber_t, agno) + __field(int, streams) + __field(xfs_extlen_t, free) + __field(int, nscan) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->ino = ip->i_ino; + __entry->agno = agno; + __entry->streams = xfs_filestream_peek_ag(ip->i_mount, agno); + __entry->free = free; + __entry->nscan = nscan; + ), + TP_printk("dev %d:%d ino 0x%llx agno %u streams %d free %d nscan %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->agno, + __entry->streams, + __entry->free, + __entry->nscan) +); + DECLARE_EVENT_CLASS(xfs_lock_class, TP_PROTO(struct xfs_inode *ip, unsigned lock_flags, unsigned long caller_ip), @@ -603,6 +661,7 @@ DEFINE_INODE_EVENT(xfs_readlink); DEFINE_INODE_EVENT(xfs_inactive_symlink); DEFINE_INODE_EVENT(xfs_alloc_file_space); DEFINE_INODE_EVENT(xfs_free_file_space); +DEFINE_INODE_EVENT(xfs_zero_file_space); DEFINE_INODE_EVENT(xfs_collapse_file_space); DEFINE_INODE_EVENT(xfs_readdir); #ifdef CONFIG_XFS_POSIX_ACL diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 54a57326d85b..d03932564ccb 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -827,7 +827,7 @@ xfs_trans_committed_bulk( xfs_log_item_batch_insert(ailp, &cur, log_items, i, commit_lsn); spin_lock(&ailp->xa_lock); - xfs_trans_ail_cursor_done(ailp, &cur); + xfs_trans_ail_cursor_done(&cur); spin_unlock(&ailp->xa_lock); } diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c index a7287354e535..cb0f3a84cc68 100644 --- a/fs/xfs/xfs_trans_ail.c +++ b/fs/xfs/xfs_trans_ail.c @@ -173,7 +173,6 @@ xfs_trans_ail_cursor_next( */ void xfs_trans_ail_cursor_done( - struct xfs_ail *ailp, struct xfs_ail_cursor *cur) { cur->item = NULL; @@ -368,7 +367,7 @@ xfsaild_push( * If the AIL is empty or our push has reached the end we are * done now. */ - xfs_trans_ail_cursor_done(ailp, &cur); + xfs_trans_ail_cursor_done(&cur); spin_unlock(&ailp->xa_lock); goto out_done; } @@ -453,7 +452,7 @@ xfsaild_push( break; lsn = lip->li_lsn; } - xfs_trans_ail_cursor_done(ailp, &cur); + xfs_trans_ail_cursor_done(&cur); spin_unlock(&ailp->xa_lock); if (xfs_buf_delwri_submit_nowait(&ailp->xa_buf_list)) diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h index 12e86af9d9b9..bd1281862ad7 100644 --- a/fs/xfs/xfs_trans_priv.h +++ b/fs/xfs/xfs_trans_priv.h @@ -133,8 +133,7 @@ struct xfs_log_item * xfs_trans_ail_cursor_last(struct xfs_ail *ailp, xfs_lsn_t lsn); struct xfs_log_item * xfs_trans_ail_cursor_next(struct xfs_ail *ailp, struct xfs_ail_cursor *cur); -void xfs_trans_ail_cursor_done(struct xfs_ail *ailp, - struct xfs_ail_cursor *cur); +void xfs_trans_ail_cursor_done(struct xfs_ail_cursor *cur); #if BITS_PER_LONG != 64 static inline void diff --git a/fs/xfs/xfs_trans_resv.c b/fs/xfs/xfs_trans_resv.c index ae368165244d..52b6c3e3203e 100644 --- a/fs/xfs/xfs_trans_resv.c +++ b/fs/xfs/xfs_trans_resv.c @@ -106,6 +106,47 @@ xfs_calc_inode_res( } /* + * The free inode btree is a conditional feature and the log reservation + * requirements differ slightly from that of the traditional inode allocation + * btree. The finobt tracks records for inode chunks with at least one free + * inode. A record can be removed from the tree for an inode allocation + * or free and thus the finobt reservation is unconditional across: + * + * - inode allocation + * - inode free + * - inode chunk allocation + * + * The 'modify' param indicates to include the record modification scenario. The + * 'alloc' param indicates to include the reservation for free space btree + * modifications on behalf of finobt modifications. This is required only for + * transactions that do not already account for free space btree modifications. + * + * the free inode btree: max depth * block size + * the allocation btrees: 2 trees * (max depth - 1) * block size + * the free inode btree entry: block size + */ +STATIC uint +xfs_calc_finobt_res( + struct xfs_mount *mp, + int alloc, + int modify) +{ + uint res; + + if (!xfs_sb_version_hasfinobt(&mp->m_sb)) + return 0; + + res = xfs_calc_buf_res(mp->m_in_maxlevels, XFS_FSB_TO_B(mp, 1)); + if (alloc) + res += xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), + XFS_FSB_TO_B(mp, 1)); + if (modify) + res += (uint)XFS_FSB_TO_B(mp, 1); + + return res; +} + +/* * Various log reservation values. * * These are based on the size of the file system block because that is what @@ -302,6 +343,7 @@ xfs_calc_remove_reservation( * the superblock for the nlink flag: sector size * the directory btree: (max depth + v2) * dir block size * the directory inode's bmap btree: (max depth + v2) * block size + * the finobt (record modification and allocation btrees) */ STATIC uint xfs_calc_create_resv_modify( @@ -310,7 +352,8 @@ xfs_calc_create_resv_modify( return xfs_calc_inode_res(mp, 2) + xfs_calc_buf_res(1, mp->m_sb.sb_sectsize) + (uint)XFS_FSB_TO_B(mp, 1) + - xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), XFS_FSB_TO_B(mp, 1)); + xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), XFS_FSB_TO_B(mp, 1)) + + xfs_calc_finobt_res(mp, 1, 1); } /* @@ -348,6 +391,7 @@ __xfs_calc_create_reservation( * the superblock for the nlink flag: sector size * the inode btree: max depth * blocksize * the allocation btrees: 2 trees * (max depth - 1) * block size + * the finobt (record insertion) */ STATIC uint xfs_calc_icreate_resv_alloc( @@ -357,7 +401,8 @@ xfs_calc_icreate_resv_alloc( mp->m_sb.sb_sectsize + xfs_calc_buf_res(mp->m_in_maxlevels, XFS_FSB_TO_B(mp, 1)) + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), - XFS_FSB_TO_B(mp, 1)); + XFS_FSB_TO_B(mp, 1)) + + xfs_calc_finobt_res(mp, 0, 0); } STATIC uint @@ -425,6 +470,7 @@ xfs_calc_symlink_reservation( * the on disk inode before ours in the agi hash list: inode cluster size * the inode btree: max depth * blocksize * the allocation btrees: 2 trees * (max depth - 1) * block size + * the finobt (record insertion, removal or modification) */ STATIC uint xfs_calc_ifree_reservation( @@ -439,7 +485,8 @@ xfs_calc_ifree_reservation( xfs_calc_buf_res(2 + mp->m_ialloc_blks + mp->m_in_maxlevels, 0) + xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), - XFS_FSB_TO_B(mp, 1)); + XFS_FSB_TO_B(mp, 1)) + + xfs_calc_finobt_res(mp, 0, 1); } /* diff --git a/fs/xfs/xfs_trans_space.h b/fs/xfs/xfs_trans_space.h index af5dbe06cb65..df4c1f81884c 100644 --- a/fs/xfs/xfs_trans_space.h +++ b/fs/xfs/xfs_trans_space.h @@ -47,7 +47,9 @@ #define XFS_DIRREMOVE_SPACE_RES(mp) \ XFS_DAREMOVE_SPACE_RES(mp, XFS_DATA_FORK) #define XFS_IALLOC_SPACE_RES(mp) \ - ((mp)->m_ialloc_blks + (mp)->m_in_maxlevels - 1) + ((mp)->m_ialloc_blks + \ + (xfs_sb_version_hasfinobt(&mp->m_sb) ? 2 : 1 * \ + ((mp)->m_in_maxlevels - 1))) /* * Space reservation values for various transactions. @@ -82,5 +84,8 @@ (XFS_DIRREMOVE_SPACE_RES(mp) + XFS_DIRENTER_SPACE_RES(mp,nl)) #define XFS_SYMLINK_SPACE_RES(mp,nl,b) \ (XFS_IALLOC_SPACE_RES(mp) + XFS_DIRENTER_SPACE_RES(mp,nl) + (b)) +#define XFS_IFREE_SPACE_RES(mp) \ + (xfs_sb_version_hasfinobt(&mp->m_sb) ? (mp)->m_in_maxlevels : 0) + #endif /* __XFS_TRANS_SPACE_H__ */ diff --git a/fs/xfs/xfs_types.h b/fs/xfs/xfs_types.h index 82bbc34d54a3..65c6e6650b1a 100644 --- a/fs/xfs/xfs_types.h +++ b/fs/xfs/xfs_types.h @@ -134,7 +134,7 @@ typedef enum { typedef enum { XFS_BTNUM_BNOi, XFS_BTNUM_CNTi, XFS_BTNUM_BMAPi, XFS_BTNUM_INOi, - XFS_BTNUM_MAX + XFS_BTNUM_FINOi, XFS_BTNUM_MAX } xfs_btnum_t; struct xfs_name { |