diff options
Diffstat (limited to 'fs')
98 files changed, 1304 insertions, 916 deletions
diff --git a/fs/Kconfig b/fs/Kconfig index 462253ae483a..a55bda4233bb 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -203,7 +203,7 @@ config TMPFS_XATTR config TMPFS_INODE64 bool "Use 64-bit ino_t by default in tmpfs" - depends on TMPFS && 64BIT && !(S390 || ALPHA) + depends on TMPFS && 64BIT default n help tmpfs has historically used only inode numbers as wide as an unsigned diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 714fcca9af99..17548c1faf02 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -70,7 +70,6 @@ const struct inode_operations afs_dir_inode_operations = { .permission = afs_permission, .getattr = afs_getattr, .setattr = afs_setattr, - .listxattr = afs_listxattr, }; const struct address_space_operations afs_dir_aops = { diff --git a/fs/afs/file.c b/fs/afs/file.c index 85f5adf21aa0..960b64268623 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -43,7 +43,6 @@ const struct inode_operations afs_file_inode_operations = { .getattr = afs_getattr, .setattr = afs_setattr, .permission = afs_permission, - .listxattr = afs_listxattr, }; const struct address_space_operations afs_fs_aops = { diff --git a/fs/afs/fs_operation.c b/fs/afs/fs_operation.c index 97cab12b0a6c..71c58723763d 100644 --- a/fs/afs/fs_operation.c +++ b/fs/afs/fs_operation.c @@ -181,10 +181,13 @@ void afs_wait_for_operation(struct afs_operation *op) if (test_bit(AFS_SERVER_FL_IS_YFS, &op->server->flags) && op->ops->issue_yfs_rpc) op->ops->issue_yfs_rpc(op); - else + else if (op->ops->issue_afs_rpc) op->ops->issue_afs_rpc(op); + else + op->ac.error = -ENOTSUPP; - op->error = afs_wait_for_call_to_complete(op->call, &op->ac); + if (op->call) + op->error = afs_wait_for_call_to_complete(op->call, &op->ac); } switch (op->error) { diff --git a/fs/afs/inode.c b/fs/afs/inode.c index 1156b2df28d3..12be88716e4c 100644 --- a/fs/afs/inode.c +++ b/fs/afs/inode.c @@ -27,7 +27,6 @@ static const struct inode_operations afs_symlink_inode_operations = { .get_link = page_get_link, - .listxattr = afs_listxattr, }; static noinline void dump_vnode(struct afs_vnode *vnode, struct afs_vnode *parent_vnode) diff --git a/fs/afs/internal.h b/fs/afs/internal.h index b626e38e9ab5..1627b1872812 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -1509,7 +1509,6 @@ extern int afs_launder_page(struct page *); * xattr.c */ extern const struct xattr_handler *afs_xattr_handlers[]; -extern ssize_t afs_listxattr(struct dentry *, char *, size_t); /* * yfsclient.c diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index 052dab2f5c03..bbb2c210d139 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -32,7 +32,6 @@ const struct inode_operations afs_mntpt_inode_operations = { .lookup = afs_mntpt_lookup, .readlink = page_readlink, .getattr = afs_getattr, - .listxattr = afs_listxattr, }; const struct inode_operations afs_autocell_inode_operations = { diff --git a/fs/afs/write.c b/fs/afs/write.c index c9195fc67fd8..eb737ed63afb 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -851,8 +851,7 @@ vm_fault_t afs_page_mkwrite(struct vm_fault *vmf) fscache_wait_on_page_write(vnode->cache, vmf->page); #endif - if (PageWriteback(vmf->page) && - wait_on_page_bit_killable(vmf->page, PG_writeback) < 0) + if (wait_on_page_writeback_killable(vmf->page)) return VM_FAULT_RETRY; if (lock_page_killable(vmf->page) < 0) diff --git a/fs/afs/xattr.c b/fs/afs/xattr.c index c629caae5002..7751b0b3f81d 100644 --- a/fs/afs/xattr.c +++ b/fs/afs/xattr.c @@ -11,29 +11,6 @@ #include <linux/xattr.h> #include "internal.h" -static const char afs_xattr_list[] = - "afs.acl\0" - "afs.cell\0" - "afs.fid\0" - "afs.volume\0" - "afs.yfs.acl\0" - "afs.yfs.acl_inherited\0" - "afs.yfs.acl_num_cleaned\0" - "afs.yfs.vol_acl"; - -/* - * Retrieve a list of the supported xattrs. - */ -ssize_t afs_listxattr(struct dentry *dentry, char *buffer, size_t size) -{ - if (size == 0) - return sizeof(afs_xattr_list); - if (size < sizeof(afs_xattr_list)) - return -ERANGE; - memcpy(buffer, afs_xattr_list, sizeof(afs_xattr_list)); - return sizeof(afs_xattr_list); -} - /* * Deal with the result of a successful fetch ACL operation. */ @@ -231,6 +208,8 @@ static int afs_xattr_get_yfs(const struct xattr_handler *handler, else ret = -ERANGE; } + } else if (ret == -ENOTSUPP) { + ret = -ENODATA; } error_yacl: @@ -256,6 +235,7 @@ static int afs_xattr_set_yfs(const struct xattr_handler *handler, { struct afs_operation *op; struct afs_vnode *vnode = AFS_FS_I(inode); + int ret; if (flags == XATTR_CREATE || strcmp(name, "acl") != 0) @@ -270,7 +250,10 @@ static int afs_xattr_set_yfs(const struct xattr_handler *handler, return afs_put_operation(op); op->ops = &yfs_store_opaque_acl2_operation; - return afs_do_sync_operation(op); + ret = afs_do_sync_operation(op); + if (ret == -ENOTSUPP) + ret = -ENODATA; + return ret; } static const struct xattr_handler afs_xattr_yfs_handler = { diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index c457334de43f..e1eae7ea823a 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -649,12 +649,24 @@ static ssize_t bm_register_write(struct file *file, const char __user *buffer, struct super_block *sb = file_inode(file)->i_sb; struct dentry *root = sb->s_root, *dentry; int err = 0; + struct file *f = NULL; e = create_entry(buffer, count); if (IS_ERR(e)) return PTR_ERR(e); + if (e->flags & MISC_FMT_OPEN_FILE) { + f = open_exec(e->interpreter); + if (IS_ERR(f)) { + pr_notice("register: failed to install interpreter file %s\n", + e->interpreter); + kfree(e); + return PTR_ERR(f); + } + e->interp_file = f; + } + inode_lock(d_inode(root)); dentry = lookup_one_len(e->name, root, strlen(e->name)); err = PTR_ERR(dentry); @@ -678,21 +690,6 @@ static ssize_t bm_register_write(struct file *file, const char __user *buffer, goto out2; } - if (e->flags & MISC_FMT_OPEN_FILE) { - struct file *f; - - f = open_exec(e->interpreter); - if (IS_ERR(f)) { - err = PTR_ERR(f); - pr_notice("register: failed to install interpreter file %s\n", e->interpreter); - simple_release_fs(&bm_mnt, &entry_count); - iput(inode); - inode = NULL; - goto out2; - } - e->interp_file = f; - } - e->dentry = dget(dentry); inode->i_private = e; inode->i_fop = &bm_entry_operations; @@ -709,6 +706,8 @@ out: inode_unlock(d_inode(root)); if (err) { + if (f) + filp_close(f, NULL); kfree(e); return err; } diff --git a/fs/block_dev.c b/fs/block_dev.c index 4aa1f88d5bf8..92ed7d5df677 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -118,13 +118,22 @@ int truncate_bdev_range(struct block_device *bdev, fmode_t mode, if (!(mode & FMODE_EXCL)) { int err = bd_prepare_to_claim(bdev, truncate_bdev_range); if (err) - return err; + goto invalidate; } truncate_inode_pages_range(bdev->bd_inode->i_mapping, lstart, lend); if (!(mode & FMODE_EXCL)) bd_abort_claiming(bdev, truncate_bdev_range); return 0; + +invalidate: + /* + * Someone else has handle exclusively open. Try invalidating instead. + * The 'end' argument is inclusive so the rounding is safe. + */ + return invalidate_inode_pages2_range(bdev->bd_inode->i_mapping, + lstart >> PAGE_SHIFT, + lend >> PAGE_SHIFT); } static void set_init_blocksize(struct block_device *bdev) @@ -423,7 +432,7 @@ static ssize_t __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, dio->size += bio->bi_iter.bi_size; pos += bio->bi_iter.bi_size; - nr_pages = bio_iov_vecs_to_alloc(iter, BIO_MAX_PAGES); + nr_pages = bio_iov_vecs_to_alloc(iter, BIO_MAX_VECS); if (!nr_pages) { bool polled = false; @@ -491,8 +500,8 @@ blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter) if (!iov_iter_count(iter)) return 0; - nr_pages = bio_iov_vecs_to_alloc(iter, BIO_MAX_PAGES + 1); - if (is_sync_kiocb(iocb) && nr_pages <= BIO_MAX_PAGES) + nr_pages = bio_iov_vecs_to_alloc(iter, BIO_MAX_VECS + 1); + if (is_sync_kiocb(iocb) && nr_pages <= BIO_MAX_VECS) return __blkdev_direct_IO_simple(iocb, iter, nr_pages); return __blkdev_direct_IO(iocb, iter, bio_max_segs(nr_pages)); diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index d56730a67885..34b929bd5c1a 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -1365,7 +1365,9 @@ get_old_root(struct btrfs_root *root, u64 time_seq) "failed to read tree block %llu from get_old_root", logical); } else { + btrfs_tree_read_lock(old); eb = btrfs_clone_extent_buffer(old); + btrfs_tree_read_unlock(old); free_extent_buffer(old); } } else if (old_root) { diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 78ad31a59e59..36a3c973fda1 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3323,6 +3323,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, if (last_ref && btrfs_header_generation(buf) == trans->transid) { struct btrfs_block_group *cache; + bool must_pin = false; if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) { ret = check_ref_cleanup(trans, buf->start); @@ -3340,7 +3341,27 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, goto out; } - if (btrfs_is_zoned(fs_info)) { + /* + * If this is a leaf and there are tree mod log users, we may + * have recorded mod log operations that point to this leaf. + * So we must make sure no one reuses this leaf's extent before + * mod log operations are applied to a node, otherwise after + * rewinding a node using the mod log operations we get an + * inconsistent btree, as the leaf's extent may now be used as + * a node or leaf for another different btree. + * We are safe from races here because at this point no other + * node or root points to this extent buffer, so if after this + * check a new tree mod log user joins, it will not be able to + * find a node pointing to this leaf and record operations that + * point to this leaf. + */ + if (btrfs_header_level(buf) == 0) { + read_lock(&fs_info->tree_mod_log_lock); + must_pin = !list_empty(&fs_info->tree_mod_seq_list); + read_unlock(&fs_info->tree_mod_log_lock); + } + + if (must_pin || btrfs_is_zoned(fs_info)) { btrfs_redirty_list_add(trans->transaction, buf); pin_down_extent(trans, cache, buf->start, buf->len, 1); btrfs_put_block_group(cache); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 4671c99d468d..910769d5fcdb 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2886,6 +2886,35 @@ static void end_page_read(struct page *page, bool uptodate, u64 start, u32 len) } /* + * Find extent buffer for a givne bytenr. + * + * This is for end_bio_extent_readpage(), thus we can't do any unsafe locking + * in endio context. + */ +static struct extent_buffer *find_extent_buffer_readpage( + struct btrfs_fs_info *fs_info, struct page *page, u64 bytenr) +{ + struct extent_buffer *eb; + + /* + * For regular sectorsize, we can use page->private to grab extent + * buffer + */ + if (fs_info->sectorsize == PAGE_SIZE) { + ASSERT(PagePrivate(page) && page->private); + return (struct extent_buffer *)page->private; + } + + /* For subpage case, we need to lookup buffer radix tree */ + rcu_read_lock(); + eb = radix_tree_lookup(&fs_info->buffer_radix, + bytenr >> fs_info->sectorsize_bits); + rcu_read_unlock(); + ASSERT(eb); + return eb; +} + +/* * after a readpage IO is done, we need to: * clear the uptodate bits on error * set the uptodate bits if things worked @@ -2996,7 +3025,7 @@ static void end_bio_extent_readpage(struct bio *bio) } else { struct extent_buffer *eb; - eb = (struct extent_buffer *)page->private; + eb = find_extent_buffer_readpage(fs_info, page, start); set_bit(EXTENT_BUFFER_READ_ERR, &eb->bflags); eb->read_mirror = mirror; atomic_dec(&eb->io_pages); @@ -3020,7 +3049,7 @@ readpage_ok: */ if (page->index == end_index && i_size <= end) { u32 zero_start = max(offset_in_page(i_size), - offset_in_page(end)); + offset_in_page(start)); zero_user_segment(page, zero_start, offset_in_page(end) + 1); @@ -3059,7 +3088,7 @@ struct bio *btrfs_bio_alloc(u64 first_byte) { struct bio *bio; - bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_PAGES, &btrfs_bioset); + bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_VECS, &btrfs_bioset); bio->bi_iter.bi_sector = first_byte >> 9; btrfs_io_bio_init(btrfs_io_bio(bio)); return bio; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 35bfa0533f23..7cdf65be3707 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -9008,7 +9008,7 @@ int __init btrfs_init_cachep(void) btrfs_free_space_bitmap_cachep = kmem_cache_create("btrfs_free_space_bitmap", PAGE_SIZE, PAGE_SIZE, - SLAB_RED_ZONE, NULL); + SLAB_MEM_SPREAD, NULL); if (!btrfs_free_space_bitmap_cachep) goto fail; @@ -9877,6 +9877,7 @@ static struct btrfs_trans_handle *insert_prealloc_file_extent( struct btrfs_path *path; u64 start = ins->objectid; u64 len = ins->offset; + int qgroup_released; int ret; memset(&stack_fi, 0, sizeof(stack_fi)); @@ -9889,16 +9890,16 @@ static struct btrfs_trans_handle *insert_prealloc_file_extent( btrfs_set_stack_file_extent_compression(&stack_fi, BTRFS_COMPRESS_NONE); /* Encryption and other encoding is reserved and all 0 */ - ret = btrfs_qgroup_release_data(inode, file_offset, len); - if (ret < 0) - return ERR_PTR(ret); + qgroup_released = btrfs_qgroup_release_data(inode, file_offset, len); + if (qgroup_released < 0) + return ERR_PTR(qgroup_released); if (trans) { ret = insert_reserved_file_extent(trans, inode, file_offset, &stack_fi, - true, ret); + true, qgroup_released); if (ret) - return ERR_PTR(ret); + goto free_qgroup; return trans; } @@ -9909,21 +9910,35 @@ static struct btrfs_trans_handle *insert_prealloc_file_extent( extent_info.file_offset = file_offset; extent_info.extent_buf = (char *)&stack_fi; extent_info.is_new_extent = true; - extent_info.qgroup_reserved = ret; + extent_info.qgroup_reserved = qgroup_released; extent_info.insertions = 0; path = btrfs_alloc_path(); - if (!path) - return ERR_PTR(-ENOMEM); + if (!path) { + ret = -ENOMEM; + goto free_qgroup; + } ret = btrfs_replace_file_extents(&inode->vfs_inode, path, file_offset, file_offset + len - 1, &extent_info, &trans); btrfs_free_path(path); if (ret) - return ERR_PTR(ret); - + goto free_qgroup; return trans; + +free_qgroup: + /* + * We have released qgroup data range at the beginning of the function, + * and normally qgroup_released bytes will be freed when committing + * transaction. + * But if we error out early, we have to free what we have released + * or we leak qgroup data reservation. + */ + btrfs_qgroup_free_refroot(inode->root->fs_info, + inode->root->root_key.objectid, qgroup_released, + BTRFS_QGROUP_RSV_DATA); + return ERR_PTR(ret); } static int __btrfs_prealloc_file_range(struct inode *inode, int mode, diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c index 20fd4aa48a8c..06713a8fe26b 100644 --- a/fs/btrfs/reada.c +++ b/fs/btrfs/reada.c @@ -209,7 +209,7 @@ int btree_readahead_hook(struct extent_buffer *eb, int err) /* find extent */ spin_lock(&fs_info->reada_lock); re = radix_tree_lookup(&fs_info->reada_tree, - eb->start >> PAGE_SHIFT); + eb->start >> fs_info->sectorsize_bits); if (re) re->refcnt++; spin_unlock(&fs_info->reada_lock); @@ -240,7 +240,7 @@ static struct reada_zone *reada_find_zone(struct btrfs_device *dev, u64 logical, zone = NULL; spin_lock(&fs_info->reada_lock); ret = radix_tree_gang_lookup(&dev->reada_zones, (void **)&zone, - logical >> PAGE_SHIFT, 1); + logical >> fs_info->sectorsize_bits, 1); if (ret == 1 && logical >= zone->start && logical <= zone->end) { kref_get(&zone->refcnt); spin_unlock(&fs_info->reada_lock); @@ -283,13 +283,13 @@ static struct reada_zone *reada_find_zone(struct btrfs_device *dev, u64 logical, spin_lock(&fs_info->reada_lock); ret = radix_tree_insert(&dev->reada_zones, - (unsigned long)(zone->end >> PAGE_SHIFT), - zone); + (unsigned long)(zone->end >> fs_info->sectorsize_bits), + zone); if (ret == -EEXIST) { kfree(zone); ret = radix_tree_gang_lookup(&dev->reada_zones, (void **)&zone, - logical >> PAGE_SHIFT, 1); + logical >> fs_info->sectorsize_bits, 1); if (ret == 1 && logical >= zone->start && logical <= zone->end) kref_get(&zone->refcnt); else @@ -315,7 +315,7 @@ static struct reada_extent *reada_find_extent(struct btrfs_fs_info *fs_info, u64 length; int real_stripes; int nzones = 0; - unsigned long index = logical >> PAGE_SHIFT; + unsigned long index = logical >> fs_info->sectorsize_bits; int dev_replace_is_ongoing; int have_zone = 0; @@ -497,7 +497,7 @@ static void reada_extent_put(struct btrfs_fs_info *fs_info, struct reada_extent *re) { int i; - unsigned long index = re->logical >> PAGE_SHIFT; + unsigned long index = re->logical >> fs_info->sectorsize_bits; spin_lock(&fs_info->reada_lock); if (--re->refcnt) { @@ -538,11 +538,12 @@ static void reada_extent_put(struct btrfs_fs_info *fs_info, static void reada_zone_release(struct kref *kref) { struct reada_zone *zone = container_of(kref, struct reada_zone, refcnt); + struct btrfs_fs_info *fs_info = zone->device->fs_info; - lockdep_assert_held(&zone->device->fs_info->reada_lock); + lockdep_assert_held(&fs_info->reada_lock); radix_tree_delete(&zone->device->reada_zones, - zone->end >> PAGE_SHIFT); + zone->end >> fs_info->sectorsize_bits); kfree(zone); } @@ -593,7 +594,7 @@ static int reada_add_block(struct reada_control *rc, u64 logical, static void reada_peer_zones_set_lock(struct reada_zone *zone, int lock) { int i; - unsigned long index = zone->end >> PAGE_SHIFT; + unsigned long index = zone->end >> zone->device->fs_info->sectorsize_bits; for (i = 0; i < zone->ndevs; ++i) { struct reada_zone *peer; @@ -628,7 +629,7 @@ static int reada_pick_zone(struct btrfs_device *dev) (void **)&zone, index, 1); if (ret == 0) break; - index = (zone->end >> PAGE_SHIFT) + 1; + index = (zone->end >> dev->fs_info->sectorsize_bits) + 1; if (zone->locked) { if (zone->elems > top_locked_elems) { top_locked_elems = zone->elems; @@ -709,7 +710,7 @@ static int reada_start_machine_dev(struct btrfs_device *dev) * plugging to speed things up */ ret = radix_tree_gang_lookup(&dev->reada_extents, (void **)&re, - dev->reada_next >> PAGE_SHIFT, 1); + dev->reada_next >> fs_info->sectorsize_bits, 1); if (ret == 0 || re->logical > dev->reada_curr_zone->end) { ret = reada_pick_zone(dev); if (!ret) { @@ -718,7 +719,7 @@ static int reada_start_machine_dev(struct btrfs_device *dev) } re = NULL; ret = radix_tree_gang_lookup(&dev->reada_extents, (void **)&re, - dev->reada_next >> PAGE_SHIFT, 1); + dev->reada_next >> fs_info->sectorsize_bits, 1); } if (ret == 0) { spin_unlock(&fs_info->reada_lock); @@ -885,7 +886,7 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all) pr_cont(" curr off %llu", device->reada_next - zone->start); pr_cont("\n"); - index = (zone->end >> PAGE_SHIFT) + 1; + index = (zone->end >> fs_info->sectorsize_bits) + 1; } cnt = 0; index = 0; @@ -910,7 +911,7 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all) } } pr_cont("\n"); - index = (re->logical >> PAGE_SHIFT) + 1; + index = (re->logical >> fs_info->sectorsize_bits) + 1; if (++cnt > 15) break; } @@ -926,7 +927,7 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all) if (ret == 0) break; if (!re->scheduled) { - index = (re->logical >> PAGE_SHIFT) + 1; + index = (re->logical >> fs_info->sectorsize_bits) + 1; continue; } pr_debug("re: logical %llu size %u list empty %d scheduled %d", @@ -942,7 +943,7 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all) } } pr_cont("\n"); - index = (re->logical >> PAGE_SHIFT) + 1; + index = (re->logical >> fs_info->sectorsize_bits) + 1; } spin_unlock(&fs_info->reada_lock); } diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index c2900ebf767a..3d9088eab2fc 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -1428,7 +1428,7 @@ static void scrub_recheck_block_on_raid56(struct btrfs_fs_info *fs_info, if (!first_page->dev->bdev) goto out; - bio = btrfs_io_bio_alloc(BIO_MAX_PAGES); + bio = btrfs_io_bio_alloc(BIO_MAX_VECS); bio_set_dev(bio, first_page->dev->bdev); for (page_num = 0; page_num < sblock->page_count; page_num++) { diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 2f1acc9aea9e..92a368627791 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -3169,10 +3169,6 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, mutex_lock(&log_root_tree->log_mutex); - index2 = log_root_tree->log_transid % 2; - list_add_tail(&root_log_ctx.list, &log_root_tree->log_ctxs[index2]); - root_log_ctx.log_transid = log_root_tree->log_transid; - if (btrfs_is_zoned(fs_info)) { if (!log_root_tree->node) { ret = btrfs_alloc_log_tree_node(trans, log_root_tree); @@ -3183,6 +3179,10 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, } } + index2 = log_root_tree->log_transid % 2; + list_add_tail(&root_log_ctx.list, &log_root_tree->log_ctxs[index2]); + root_log_ctx.log_transid = log_root_tree->log_transid; + /* * Now we are safe to update the log_root_tree because we're under the * log_mutex, and we're a current writer so we're holding the commit diff --git a/fs/cachefiles/bind.c b/fs/cachefiles/bind.c index dfb14dbddf51..38bb7764b454 100644 --- a/fs/cachefiles/bind.c +++ b/fs/cachefiles/bind.c @@ -118,6 +118,12 @@ static int cachefiles_daemon_add_cache(struct cachefiles_cache *cache) cache->mnt = path.mnt; root = path.dentry; + ret = -EINVAL; + if (mnt_user_ns(path.mnt) != &init_user_ns) { + pr_warn("File cache on idmapped mounts not supported"); + goto error_unsupported; + } + /* check parameters */ ret = -EOPNOTSUPP; if (d_is_negative(root) || diff --git a/fs/cachefiles/rdwr.c b/fs/cachefiles/rdwr.c index e027c718ca01..8ffc40e84a59 100644 --- a/fs/cachefiles/rdwr.c +++ b/fs/cachefiles/rdwr.c @@ -24,17 +24,16 @@ static int cachefiles_read_waiter(wait_queue_entry_t *wait, unsigned mode, container_of(wait, struct cachefiles_one_read, monitor); struct cachefiles_object *object; struct fscache_retrieval *op = monitor->op; - struct wait_bit_key *key = _key; + struct wait_page_key *key = _key; struct page *page = wait->private; ASSERT(key); _enter("{%lu},%u,%d,{%p,%u}", monitor->netfs_page->index, mode, sync, - key->flags, key->bit_nr); + key->page, key->bit_nr); - if (key->flags != &page->flags || - key->bit_nr != PG_locked) + if (key->page != page || key->bit_nr != PG_locked) return 0; _debug("--- monitor %p %lx ---", page, page->flags); diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index 3aedc484e440..88a7958170ee 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -207,7 +207,7 @@ static int cifs_debug_files_proc_show(struct seq_file *m, void *v) from_kuid(&init_user_ns, cfile->uid), cfile->dentry); #ifdef CONFIG_CIFS_DEBUG2 - seq_printf(m, " 0x%llx\n", cfile->fid.mid); + seq_printf(m, " %llu\n", cfile->fid.mid); #else seq_printf(m, "\n"); #endif /* CIFS_DEBUG2 */ diff --git a/fs/cifs/cifs_swn.c b/fs/cifs/cifs_swn.c index f2d730fffccb..d829b8bf833e 100644 --- a/fs/cifs/cifs_swn.c +++ b/fs/cifs/cifs_swn.c @@ -248,7 +248,7 @@ nlmsg_fail: /* * Try to find a matching registration for the tcon's server name and share name. - * Calls to this funciton must be protected by cifs_swnreg_idr_mutex. + * Calls to this function must be protected by cifs_swnreg_idr_mutex. * TODO Try to avoid memory allocations */ static struct cifs_swn_reg *cifs_find_swn_reg(struct cifs_tcon *tcon) diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c index 9d29eb9660c2..2be22a5c690f 100644 --- a/fs/cifs/cifsacl.c +++ b/fs/cifs/cifsacl.c @@ -1118,7 +1118,6 @@ static int set_chmod_dacl(struct cifs_acl *pdacl, struct cifs_acl *pndacl, /* Retain old ACEs which we can retain */ for (i = 0; i < src_num_aces; ++i) { pntace = (struct cifs_ace *) (acl_base + size); - pnntace = (struct cifs_ace *) (nacl_base + nsize); if (!new_aces_set && (pntace->flags & INHERITED_ACE)) { /* Place the new ACEs in between existing explicit and inherited */ @@ -1131,14 +1130,18 @@ static int set_chmod_dacl(struct cifs_acl *pdacl, struct cifs_acl *pndacl, } /* If it's any one of the ACE we're replacing, skip! */ - if ((compare_sids(&pntace->sid, &sid_unix_NFS_mode) == 0) || + if (!mode_from_sid && + ((compare_sids(&pntace->sid, &sid_unix_NFS_mode) == 0) || (compare_sids(&pntace->sid, pownersid) == 0) || (compare_sids(&pntace->sid, pgrpsid) == 0) || (compare_sids(&pntace->sid, &sid_everyone) == 0) || - (compare_sids(&pntace->sid, &sid_authusers) == 0)) { + (compare_sids(&pntace->sid, &sid_authusers) == 0))) { goto next_ace; } + /* update the pointer to the next ACE to populate*/ + pnntace = (struct cifs_ace *) (nacl_base + nsize); + nsize += cifs_copy_ace(pnntace, pntace, NULL); num_aces++; diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index d43e935d2df4..099ad9f3660b 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -290,7 +290,7 @@ cifs_statfs(struct dentry *dentry, struct kstatfs *buf) rc = server->ops->queryfs(xid, tcon, cifs_sb, buf); free_xid(xid); - return 0; + return rc; } static long cifs_fallocate(struct file *file, int mode, loff_t off, loff_t len) diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 3de3c5908a72..31fc8695abd6 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -257,7 +257,7 @@ struct smb_version_operations { /* verify the message */ int (*check_message)(char *, unsigned int, struct TCP_Server_Info *); bool (*is_oplock_break)(char *, struct TCP_Server_Info *); - int (*handle_cancelled_mid)(char *, struct TCP_Server_Info *); + int (*handle_cancelled_mid)(struct mid_q_entry *, struct TCP_Server_Info *); void (*downgrade_oplock)(struct TCP_Server_Info *server, struct cifsInodeInfo *cinode, __u32 oplock, unsigned int epoch, bool *purge_cache); @@ -1705,16 +1705,17 @@ static inline bool is_retryable_error(int error) #define CIFS_NO_RSP_BUF 0x040 /* no response buffer required */ /* Type of request operation */ -#define CIFS_ECHO_OP 0x080 /* echo request */ -#define CIFS_OBREAK_OP 0x0100 /* oplock break request */ -#define CIFS_NEG_OP 0x0200 /* negotiate request */ +#define CIFS_ECHO_OP 0x080 /* echo request */ +#define CIFS_OBREAK_OP 0x0100 /* oplock break request */ +#define CIFS_NEG_OP 0x0200 /* negotiate request */ +#define CIFS_CP_CREATE_CLOSE_OP 0x0400 /* compound create+close request */ /* Lower bitmask values are reserved by others below. */ -#define CIFS_SESS_OP 0x2000 /* session setup request */ -#define CIFS_OP_MASK 0x2380 /* mask request type */ +#define CIFS_SESS_OP 0x2000 /* session setup request */ +#define CIFS_OP_MASK 0x2780 /* mask request type */ -#define CIFS_HAS_CREDITS 0x0400 /* already has credits */ -#define CIFS_TRANSFORM_REQ 0x0800 /* transform request before sending */ -#define CIFS_NO_SRV_RSP 0x1000 /* there is no server response */ +#define CIFS_HAS_CREDITS 0x0400 /* already has credits */ +#define CIFS_TRANSFORM_REQ 0x0800 /* transform request before sending */ +#define CIFS_NO_SRV_RSP 0x1000 /* there is no server response */ /* Security Flags: indicate type of session setup needed */ #define CIFSSEC_MAY_SIGN 0x00001 diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 112692300fb6..eec8a2052da2 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -741,7 +741,7 @@ static void clean_demultiplex_info(struct TCP_Server_Info *server) spin_lock(&GlobalMid_Lock); list_for_each_safe(tmp, tmp2, &server->pending_mid_q) { mid_entry = list_entry(tmp, struct mid_q_entry, qhead); - cifs_dbg(FYI, "Clearing mid 0x%llx\n", mid_entry->mid); + cifs_dbg(FYI, "Clearing mid %llu\n", mid_entry->mid); kref_get(&mid_entry->refcount); mid_entry->mid_state = MID_SHUTDOWN; list_move(&mid_entry->qhead, &dispose_list); @@ -752,7 +752,7 @@ static void clean_demultiplex_info(struct TCP_Server_Info *server) /* now walk dispose list and issue callbacks */ list_for_each_safe(tmp, tmp2, &dispose_list) { mid_entry = list_entry(tmp, struct mid_q_entry, qhead); - cifs_dbg(FYI, "Callback mid 0x%llx\n", mid_entry->mid); + cifs_dbg(FYI, "Callback mid %llu\n", mid_entry->mid); list_del_init(&mid_entry->qhead); mid_entry->callback(mid_entry); cifs_mid_q_entry_release(mid_entry); @@ -1429,6 +1429,11 @@ smbd_connected: tcp_ses->min_offload = ctx->min_offload; tcp_ses->tcpStatus = CifsNeedNegotiate; + if ((ctx->max_credits < 20) || (ctx->max_credits > 60000)) + tcp_ses->max_credits = SMB2_MAX_CREDITS_AVAILABLE; + else + tcp_ses->max_credits = ctx->max_credits; + tcp_ses->nr_targets = 1; tcp_ses->ignore_signature = ctx->ignore_signature; /* thread spawned, put it on the list */ @@ -2832,11 +2837,6 @@ static int mount_get_conns(struct smb3_fs_context *ctx, struct cifs_sb_info *cif *nserver = server; - if ((ctx->max_credits < 20) || (ctx->max_credits > 60000)) - server->max_credits = SMB2_MAX_CREDITS_AVAILABLE; - else - server->max_credits = ctx->max_credits; - /* get a reference to a SMB session */ ses = cifs_get_smb_ses(server, ctx); if (IS_ERR(ses)) { diff --git a/fs/cifs/fs_context.c b/fs/cifs/fs_context.c index 892f51a21278..78889024a7ed 100644 --- a/fs/cifs/fs_context.c +++ b/fs/cifs/fs_context.c @@ -1196,9 +1196,11 @@ static int smb3_fs_context_parse_param(struct fs_context *fc, pr_warn_once("Witness protocol support is experimental\n"); break; case Opt_rootfs: -#ifdef CONFIG_CIFS_ROOT - ctx->rootfs = true; +#ifndef CONFIG_CIFS_ROOT + cifs_dbg(VFS, "rootfs support requires CONFIG_CIFS_ROOT config option\n"); + goto cifs_parse_mount_err; #endif + ctx->rootfs = true; break; case Opt_posixpaths: if (result.negated) diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 7c61bc9573c0..f2df4422e54a 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -2395,7 +2395,7 @@ int cifs_getattr(struct user_namespace *mnt_userns, const struct path *path, * We need to be sure that all dirty pages are written and the server * has actual ctime, mtime and file length. */ - if ((request_mask & (STATX_CTIME | STATX_MTIME | STATX_SIZE)) && + if ((request_mask & (STATX_CTIME | STATX_MTIME | STATX_SIZE | STATX_BLOCKS)) && !CIFS_CACHE_READ(CIFS_I(inode)) && inode->i_mapping && inode->i_mapping->nrpages != 0) { rc = filemap_fdatawait(inode->i_mapping); @@ -2585,6 +2585,14 @@ set_size_out: if (rc == 0) { cifsInode->server_eof = attrs->ia_size; cifs_setsize(inode, attrs->ia_size); + /* + * i_blocks is not related to (i_size / i_blksize), but instead + * 512 byte (2**9) size is required for calculating num blocks. + * Until we can query the server for actual allocation size, + * this is best estimate we have for blocks allocated for a file + * Number of blocks must be rounded up so size 1 is not 0 blocks + */ + inode->i_blocks = (512 - 1 + attrs->ia_size) >> 9; /* * The man page of truncate says if the size changed, diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 183a3a868d7b..63d517b9f2ff 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -230,6 +230,7 @@ cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, ctx.noautotune = ses->server->noautotune; ctx.sockopt_tcp_nodelay = ses->server->tcp_nodelay; ctx.echo_interval = ses->server->echo_interval / HZ; + ctx.max_credits = ses->server->max_credits; /* * This will be used for encoding/decoding user/domain/pw diff --git a/fs/cifs/smb2inode.c b/fs/cifs/smb2inode.c index 1f900b81c34a..a718dc77e604 100644 --- a/fs/cifs/smb2inode.c +++ b/fs/cifs/smb2inode.c @@ -358,6 +358,7 @@ smb2_compound_op(const unsigned int xid, struct cifs_tcon *tcon, if (cfile) goto after_close; /* Close */ + flags |= CIFS_CP_CREATE_CLOSE_OP; rqst[num_rqst].rq_iov = &vars->close_iov[0]; rqst[num_rqst].rq_nvec = 1; rc = SMB2_close_init(tcon, server, diff --git a/fs/cifs/smb2misc.c b/fs/cifs/smb2misc.c index 60d4bd1eae2b..b50164e2c88d 100644 --- a/fs/cifs/smb2misc.c +++ b/fs/cifs/smb2misc.c @@ -767,7 +767,7 @@ smb2_cancelled_close_fid(struct work_struct *work) int rc; if (cancelled->mid) - cifs_tcon_dbg(VFS, "Close unmatched open for MID:%llx\n", + cifs_tcon_dbg(VFS, "Close unmatched open for MID:%llu\n", cancelled->mid); else cifs_tcon_dbg(VFS, "Close interrupted close\n"); @@ -844,14 +844,14 @@ smb2_handle_cancelled_close(struct cifs_tcon *tcon, __u64 persistent_fid, } int -smb2_handle_cancelled_mid(char *buffer, struct TCP_Server_Info *server) +smb2_handle_cancelled_mid(struct mid_q_entry *mid, struct TCP_Server_Info *server) { - struct smb2_sync_hdr *sync_hdr = (struct smb2_sync_hdr *)buffer; - struct smb2_create_rsp *rsp = (struct smb2_create_rsp *)buffer; + struct smb2_sync_hdr *sync_hdr = mid->resp_buf; + struct smb2_create_rsp *rsp = mid->resp_buf; struct cifs_tcon *tcon; int rc; - if (sync_hdr->Command != SMB2_CREATE || + if ((mid->optype & CIFS_CP_CREATE_CLOSE_OP) || sync_hdr->Command != SMB2_CREATE || sync_hdr->Status != STATUS_SUCCESS) return 0; diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index f5087295424c..9bae7e8deb09 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -1195,7 +1195,7 @@ smb2_set_ea(const unsigned int xid, struct cifs_tcon *tcon, struct TCP_Server_Info *server = cifs_pick_channel(ses); __le16 *utf16_path = NULL; int ea_name_len = strlen(ea_name); - int flags = 0; + int flags = CIFS_CP_CREATE_CLOSE_OP; int len; struct smb_rqst rqst[3]; int resp_buftype[3]; @@ -1573,7 +1573,7 @@ smb2_ioctl_query_info(const unsigned int xid, struct smb_query_info qi; struct smb_query_info __user *pqi; int rc = 0; - int flags = 0; + int flags = CIFS_CP_CREATE_CLOSE_OP; struct smb2_query_info_rsp *qi_rsp = NULL; struct smb2_ioctl_rsp *io_rsp = NULL; void *buffer = NULL; @@ -2577,7 +2577,7 @@ smb2_query_info_compound(const unsigned int xid, struct cifs_tcon *tcon, { struct cifs_ses *ses = tcon->ses; struct TCP_Server_Info *server = cifs_pick_channel(ses); - int flags = 0; + int flags = CIFS_CP_CREATE_CLOSE_OP; struct smb_rqst rqst[3]; int resp_buftype[3]; struct kvec rsp_iov[3]; @@ -2975,7 +2975,7 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon, unsigned int sub_offset; unsigned int print_len; unsigned int print_offset; - int flags = 0; + int flags = CIFS_CP_CREATE_CLOSE_OP; struct smb_rqst rqst[3]; int resp_buftype[3]; struct kvec rsp_iov[3]; @@ -3157,7 +3157,7 @@ smb2_query_reparse_tag(const unsigned int xid, struct cifs_tcon *tcon, struct cifs_open_parms oparms; struct cifs_fid fid; struct TCP_Server_Info *server = cifs_pick_channel(tcon->ses); - int flags = 0; + int flags = CIFS_CP_CREATE_CLOSE_OP; struct smb_rqst rqst[3]; int resp_buftype[3]; struct kvec rsp_iov[3]; diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index 4bbb6126b14d..2199a9bfae8f 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -4041,8 +4041,7 @@ smb2_async_readv(struct cifs_readdata *rdata) if (rdata->credits.value > 0) { shdr->CreditCharge = cpu_to_le16(DIV_ROUND_UP(rdata->bytes, SMB2_MAX_BUFFER_SIZE)); - shdr->CreditRequest = - cpu_to_le16(le16_to_cpu(shdr->CreditCharge) + 1); + shdr->CreditRequest = cpu_to_le16(le16_to_cpu(shdr->CreditCharge) + 8); rc = adjust_credits(server, &rdata->credits, rdata->bytes); if (rc) @@ -4348,8 +4347,7 @@ smb2_async_writev(struct cifs_writedata *wdata, if (wdata->credits.value > 0) { shdr->CreditCharge = cpu_to_le16(DIV_ROUND_UP(wdata->bytes, SMB2_MAX_BUFFER_SIZE)); - shdr->CreditRequest = - cpu_to_le16(le16_to_cpu(shdr->CreditCharge) + 1); + shdr->CreditRequest = cpu_to_le16(le16_to_cpu(shdr->CreditCharge) + 8); rc = adjust_credits(server, &wdata->credits, wdata->bytes); if (rc) diff --git a/fs/cifs/smb2proto.h b/fs/cifs/smb2proto.h index 9565e27681a5..a2eb34a8d9c9 100644 --- a/fs/cifs/smb2proto.h +++ b/fs/cifs/smb2proto.h @@ -246,8 +246,7 @@ extern int SMB2_oplock_break(const unsigned int xid, struct cifs_tcon *tcon, extern int smb2_handle_cancelled_close(struct cifs_tcon *tcon, __u64 persistent_fid, __u64 volatile_fid); -extern int smb2_handle_cancelled_mid(char *buffer, - struct TCP_Server_Info *server); +extern int smb2_handle_cancelled_mid(struct mid_q_entry *mid, struct TCP_Server_Info *server); void smb2_cancelled_close_fid(struct work_struct *work); extern int SMB2_QFS_info(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_file_id, u64 volatile_file_id, diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index e90a1d1380b0..c1725b55f364 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -101,7 +101,7 @@ static void _cifs_mid_q_entry_release(struct kref *refcount) if (midEntry->resp_buf && (midEntry->mid_flags & MID_WAIT_CANCELLED) && midEntry->mid_state == MID_RESPONSE_RECEIVED && server->ops->handle_cancelled_mid) - server->ops->handle_cancelled_mid(midEntry->resp_buf, server); + server->ops->handle_cancelled_mid(midEntry, server); midEntry->mid_state = MID_FREE; atomic_dec(&midCount); @@ -1196,9 +1196,12 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, /* * Compounding is never used during session establish. */ - if ((ses->status == CifsNew) || (optype & CIFS_NEG_OP) || (optype & CIFS_SESS_OP)) + if ((ses->status == CifsNew) || (optype & CIFS_NEG_OP) || (optype & CIFS_SESS_OP)) { + mutex_lock(&server->srv_mutex); smb311_update_preauth_hash(ses, rqst[0].rq_iov, rqst[0].rq_nvec); + mutex_unlock(&server->srv_mutex); + } for (i = 0; i < num_rqst; i++) { rc = wait_for_response(server, midQ[i]); @@ -1207,7 +1210,7 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, } if (rc != 0) { for (; i < num_rqst; i++) { - cifs_server_dbg(VFS, "Cancelling wait for mid %llu cmd: %d\n", + cifs_server_dbg(FYI, "Cancelling wait for mid %llu cmd: %d\n", midQ[i]->mid, le16_to_cpu(midQ[i]->command)); send_cancel(server, &rqst[i], midQ[i]); spin_lock(&GlobalMid_Lock); @@ -1266,7 +1269,9 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, .iov_base = resp_iov[0].iov_base, .iov_len = resp_iov[0].iov_len }; + mutex_lock(&server->srv_mutex); smb311_update_preauth_hash(ses, &iov, 1); + mutex_unlock(&server->srv_mutex); } out: diff --git a/fs/configfs/file.c b/fs/configfs/file.c index 1f0270229d7b..da8351d1e455 100644 --- a/fs/configfs/file.c +++ b/fs/configfs/file.c @@ -378,7 +378,7 @@ static int __configfs_open_file(struct inode *inode, struct file *file, int type attr = to_attr(dentry); if (!attr) - goto out_put_item; + goto out_free_buffer; if (type & CONFIGFS_ITEM_BIN_ATTR) { buffer->bin_attr = to_bin_attr(dentry); @@ -391,7 +391,7 @@ static int __configfs_open_file(struct inode *inode, struct file *file, int type /* Grab the module reference for this attribute if we have one */ error = -ENODEV; if (!try_module_get(buffer->owner)) - goto out_put_item; + goto out_free_buffer; error = -EACCES; if (!buffer->item->ci_type) @@ -435,8 +435,6 @@ static int __configfs_open_file(struct inode *inode, struct file *file, int type out_put_module: module_put(buffer->owner); -out_put_item: - config_item_put(buffer->item); out_free_buffer: up_read(&frag->frag_sem); kfree(buffer); diff --git a/fs/crypto/bio.c b/fs/crypto/bio.c index b048a0e38516..68a2de6b5a9b 100644 --- a/fs/crypto/bio.c +++ b/fs/crypto/bio.c @@ -52,7 +52,7 @@ static int fscrypt_zeroout_range_inline_crypt(const struct inode *inode, int num_pages = 0; /* This always succeeds since __GFP_DIRECT_RECLAIM is set. */ - bio = bio_alloc(GFP_NOFS, BIO_MAX_PAGES); + bio = bio_alloc(GFP_NOFS, BIO_MAX_VECS); while (len) { unsigned int blocks_this_page = min(len, blocks_per_page); @@ -74,7 +74,7 @@ static int fscrypt_zeroout_range_inline_crypt(const struct inode *inode, len -= blocks_this_page; lblk += blocks_this_page; pblk += blocks_this_page; - if (num_pages == BIO_MAX_PAGES || !len || + if (num_pages == BIO_MAX_VECS || !len || !fscrypt_mergeable_bio(bio, inode, lblk)) { err = submit_bio_wait(bio); if (err) @@ -126,7 +126,7 @@ int fscrypt_zeroout_range(const struct inode *inode, pgoff_t lblk, return fscrypt_zeroout_range_inline_crypt(inode, lblk, pblk, len); - BUILD_BUG_ON(ARRAY_SIZE(pages) > BIO_MAX_PAGES); + BUILD_BUG_ON(ARRAY_SIZE(pages) > BIO_MAX_VECS); nr_pages = min_t(unsigned int, ARRAY_SIZE(pages), (len + blocks_per_page - 1) >> blocks_per_page_bits); diff --git a/fs/erofs/data.c b/fs/erofs/data.c index f88851c5c250..1249e74b3bf0 100644 --- a/fs/erofs/data.c +++ b/fs/erofs/data.c @@ -129,6 +129,7 @@ static inline struct bio *erofs_read_raw_page(struct bio *bio, struct page *page, erofs_off_t *last_block, unsigned int nblocks, + unsigned int *eblks, bool ra) { struct inode *const inode = mapping->host; @@ -145,8 +146,7 @@ static inline struct bio *erofs_read_raw_page(struct bio *bio, /* note that for readpage case, bio also equals to NULL */ if (bio && - /* not continuous */ - *last_block + 1 != current_block) { + (*last_block + 1 != current_block || !*eblks)) { submit_bio_retry: submit_bio(bio); bio = NULL; @@ -216,7 +216,8 @@ submit_bio_retry: if (nblocks > DIV_ROUND_UP(map.m_plen, PAGE_SIZE)) nblocks = DIV_ROUND_UP(map.m_plen, PAGE_SIZE); - bio = bio_alloc(GFP_NOIO, bio_max_segs(nblocks)); + *eblks = bio_max_segs(nblocks); + bio = bio_alloc(GFP_NOIO, *eblks); bio->bi_end_io = erofs_readendio; bio_set_dev(bio, sb->s_bdev); @@ -229,16 +230,8 @@ submit_bio_retry: /* out of the extent or bio is full */ if (err < PAGE_SIZE) goto submit_bio_retry; - + --*eblks; *last_block = current_block; - - /* shift in advance in case of it followed by too many gaps */ - if (bio->bi_iter.bi_size >= bio->bi_max_vecs * PAGE_SIZE) { - /* err should reassign to 0 after submitting */ - err = 0; - goto submit_bio_out; - } - return bio; err_out: @@ -252,7 +245,6 @@ has_updated: /* if updated manually, continuous pages has a gap */ if (bio) -submit_bio_out: submit_bio(bio); return err ? ERR_PTR(err) : NULL; } @@ -264,23 +256,26 @@ submit_bio_out: static int erofs_raw_access_readpage(struct file *file, struct page *page) { erofs_off_t last_block; + unsigned int eblks; struct bio *bio; trace_erofs_readpage(page, true); bio = erofs_read_raw_page(NULL, page->mapping, - page, &last_block, 1, false); + page, &last_block, 1, &eblks, false); if (IS_ERR(bio)) return PTR_ERR(bio); - DBG_BUGON(bio); /* since we have only one bio -- must be NULL */ + if (bio) + submit_bio(bio); return 0; } static void erofs_raw_access_readahead(struct readahead_control *rac) { erofs_off_t last_block; + unsigned int eblks; struct bio *bio = NULL; struct page *page; @@ -291,7 +286,7 @@ static void erofs_raw_access_readahead(struct readahead_control *rac) prefetchw(&page->flags); bio = erofs_read_raw_page(bio, rac->mapping, page, &last_block, - readahead_count(rac), true); + readahead_count(rac), &eblks, true); /* all the page errors are ignored when readahead */ if (IS_ERR(bio)) { @@ -305,7 +300,6 @@ static void erofs_raw_access_readahead(struct readahead_control *rac) put_page(page); } - /* the rare case (end in gaps) */ if (bio) submit_bio(bio); } diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 6cb356c4217b..3851e1a64f73 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -1235,7 +1235,7 @@ submit_bio_retry: } if (!bio) { - bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES); + bio = bio_alloc(GFP_NOIO, BIO_MAX_VECS); bio->bi_end_io = z_erofs_decompressqueue_endio; bio_set_dev(bio, sb->s_bdev); diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index f45f9feebe59..74a5172c2d83 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -626,27 +626,41 @@ int ext4_claim_free_clusters(struct ext4_sb_info *sbi, /** * ext4_should_retry_alloc() - check if a block allocation should be retried - * @sb: super block - * @retries: number of attemps has been made + * @sb: superblock + * @retries: number of retry attempts made so far * - * ext4_should_retry_alloc() is called when ENOSPC is returned, and if - * it is profitable to retry the operation, this function will wait - * for the current or committing transaction to complete, and then - * return TRUE. We will only retry once. + * ext4_should_retry_alloc() is called when ENOSPC is returned while + * attempting to allocate blocks. If there's an indication that a pending + * journal transaction might free some space and allow another attempt to + * succeed, this function will wait for the current or committing transaction + * to complete and then return TRUE. */ int ext4_should_retry_alloc(struct super_block *sb, int *retries) { - if (!ext4_has_free_clusters(EXT4_SB(sb), 1, 0) || - (*retries)++ > 1 || - !EXT4_SB(sb)->s_journal) + struct ext4_sb_info *sbi = EXT4_SB(sb); + + if (!sbi->s_journal) return 0; - smp_mb(); - if (EXT4_SB(sb)->s_mb_free_pending == 0) + if (++(*retries) > 3) { + percpu_counter_inc(&sbi->s_sra_exceeded_retry_limit); return 0; + } + /* + * if there's no indication that blocks are about to be freed it's + * possible we just missed a transaction commit that did so + */ + smp_mb(); + if (sbi->s_mb_free_pending == 0) + return ext4_has_free_clusters(sbi, 1, 0); + + /* + * it's possible we've just missed a transaction commit here, + * so ignore the returned status + */ jbd_debug(1, "%s: retrying operation after ENOSPC\n", sb->s_id); - jbd2_journal_force_commit_nested(EXT4_SB(sb)->s_journal); + (void) jbd2_journal_force_commit_nested(sbi->s_journal); return 1; } diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 644fd69185d3..826a56e3bbd2 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -1484,6 +1484,7 @@ struct ext4_sb_info { struct percpu_counter s_freeinodes_counter; struct percpu_counter s_dirs_counter; struct percpu_counter s_dirtyclusters_counter; + struct percpu_counter s_sra_exceeded_retry_limit; struct blockgroup_lock *s_blockgroup_lock; struct proc_dir_entry *s_proc; struct kobject s_kobj; @@ -2793,6 +2794,8 @@ void __ext4_fc_track_link(handle_t *handle, struct inode *inode, struct dentry *dentry); void ext4_fc_track_unlink(handle_t *handle, struct dentry *dentry); void ext4_fc_track_link(handle_t *handle, struct dentry *dentry); +void __ext4_fc_track_create(handle_t *handle, struct inode *inode, + struct dentry *dentry); void ext4_fc_track_create(handle_t *handle, struct dentry *dentry); void ext4_fc_track_inode(handle_t *handle, struct inode *inode); void ext4_fc_mark_ineligible(struct super_block *sb, int reason); diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 77c7c8a54da7..77c84d6f1af6 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -4382,7 +4382,7 @@ static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset, { struct inode *inode = file_inode(file); handle_t *handle; - int ret, ret2 = 0, ret3 = 0; + int ret = 0, ret2 = 0, ret3 = 0; int retries = 0; int depth = 0; struct ext4_map_blocks map; diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c index 6c4f19b0a556..7541d0b5d706 100644 --- a/fs/ext4/fast_commit.c +++ b/fs/ext4/fast_commit.c @@ -513,10 +513,10 @@ void ext4_fc_track_link(handle_t *handle, struct dentry *dentry) __ext4_fc_track_link(handle, d_inode(dentry), dentry); } -void ext4_fc_track_create(handle_t *handle, struct dentry *dentry) +void __ext4_fc_track_create(handle_t *handle, struct inode *inode, + struct dentry *dentry) { struct __track_dentry_update_args args; - struct inode *inode = d_inode(dentry); int ret; args.dentry = dentry; @@ -527,6 +527,11 @@ void ext4_fc_track_create(handle_t *handle, struct dentry *dentry) trace_ext4_fc_track_create(inode, dentry, ret); } +void ext4_fc_track_create(handle_t *handle, struct dentry *dentry) +{ + __ext4_fc_track_create(handle, d_inode(dentry), dentry); +} + /* __track_fn for inode tracking */ static int __track_inode(struct inode *inode, void *arg, bool update) { diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 650c5acd2f2d..0948a43f1b3d 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1938,13 +1938,13 @@ static int __ext4_journalled_writepage(struct page *page, if (!ret) ret = err; - if (!ext4_has_inline_data(inode)) - ext4_walk_page_buffers(NULL, page_bufs, 0, len, - NULL, bput_one); ext4_set_inode_state(inode, EXT4_STATE_JDATA); out: unlock_page(page); out_no_pagelock: + if (!inline_data && page_bufs) + ext4_walk_page_buffers(NULL, page_bufs, 0, len, + NULL, bput_one); brelse(inode_bh); return ret; } @@ -5026,7 +5026,7 @@ static int ext4_do_update_inode(handle_t *handle, struct ext4_inode_info *ei = EXT4_I(inode); struct buffer_head *bh = iloc->bh; struct super_block *sb = inode->i_sb; - int err = 0, rc, block; + int err = 0, block; int need_datasync = 0, set_large_file = 0; uid_t i_uid; gid_t i_gid; @@ -5138,9 +5138,9 @@ static int ext4_do_update_inode(handle_t *handle, bh->b_data); BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata"); - rc = ext4_handle_dirty_metadata(handle, NULL, bh); - if (!err) - err = rc; + err = ext4_handle_dirty_metadata(handle, NULL, bh); + if (err) + goto out_brelse; ext4_clear_inode_state(inode, EXT4_STATE_NEW); if (set_large_file) { BUFFER_TRACE(EXT4_SB(sb)->s_sbh, "get write access"); @@ -5387,8 +5387,10 @@ int ext4_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, inode->i_gid = attr->ia_gid; error = ext4_mark_inode_dirty(handle, inode); ext4_journal_stop(handle); - if (unlikely(error)) + if (unlikely(error)) { + ext4_fc_stop_update(inode); return error; + } } if (attr->ia_valid & ATTR_SIZE) { diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 99bf091fee10..a02fadf4fc84 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2709,8 +2709,15 @@ static int ext4_mb_init_backend(struct super_block *sb) } if (ext4_has_feature_flex_bg(sb)) { - /* a single flex group is supposed to be read by a single IO */ - sbi->s_mb_prefetch = min(1 << sbi->s_es->s_log_groups_per_flex, + /* a single flex group is supposed to be read by a single IO. + * 2 ^ s_log_groups_per_flex != UINT_MAX as s_mb_prefetch is + * unsigned integer, so the maximum shift is 32. + */ + if (sbi->s_es->s_log_groups_per_flex >= 32) { + ext4_msg(sb, KERN_ERR, "too many log groups per flexible block group"); + goto err_freesgi; + } + sbi->s_mb_prefetch = min_t(uint, 1 << sbi->s_es->s_log_groups_per_flex, BLK_MAX_SEGMENT_SIZE >> (sb->s_blocksize_bits - 9)); sbi->s_mb_prefetch *= 8; /* 8 prefetch IOs in flight at most */ } else { diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 686bf982c84e..883e2a7cd4ab 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -3613,6 +3613,31 @@ static int ext4_setent(handle_t *handle, struct ext4_renament *ent, return retval; } +static void ext4_resetent(handle_t *handle, struct ext4_renament *ent, + unsigned ino, unsigned file_type) +{ + struct ext4_renament old = *ent; + int retval = 0; + + /* + * old->de could have moved from under us during make indexed dir, + * so the old->de may no longer valid and need to find it again + * before reset old inode info. + */ + old.bh = ext4_find_entry(old.dir, &old.dentry->d_name, &old.de, NULL); + if (IS_ERR(old.bh)) + retval = PTR_ERR(old.bh); + if (!old.bh) + retval = -ENOENT; + if (retval) { + ext4_std_error(old.dir->i_sb, retval); + return; + } + + ext4_setent(handle, &old, ino, file_type); + brelse(old.bh); +} + static int ext4_find_delete_entry(handle_t *handle, struct inode *dir, const struct qstr *d_name) { @@ -3774,14 +3799,14 @@ static int ext4_rename(struct user_namespace *mnt_userns, struct inode *old_dir, */ retval = -ENOENT; if (!old.bh || le32_to_cpu(old.de->inode) != old.inode->i_ino) - goto end_rename; + goto release_bh; new.bh = ext4_find_entry(new.dir, &new.dentry->d_name, &new.de, &new.inlined); if (IS_ERR(new.bh)) { retval = PTR_ERR(new.bh); new.bh = NULL; - goto end_rename; + goto release_bh; } if (new.bh) { if (!new.inode) { @@ -3798,15 +3823,13 @@ static int ext4_rename(struct user_namespace *mnt_userns, struct inode *old_dir, handle = ext4_journal_start(old.dir, EXT4_HT_DIR, credits); if (IS_ERR(handle)) { retval = PTR_ERR(handle); - handle = NULL; - goto end_rename; + goto release_bh; } } else { whiteout = ext4_whiteout_for_rename(mnt_userns, &old, credits, &handle); if (IS_ERR(whiteout)) { retval = PTR_ERR(whiteout); - whiteout = NULL; - goto end_rename; + goto release_bh; } } @@ -3850,6 +3873,7 @@ static int ext4_rename(struct user_namespace *mnt_userns, struct inode *old_dir, retval = ext4_mark_inode_dirty(handle, whiteout); if (unlikely(retval)) goto end_rename; + } if (!new.bh) { retval = ext4_add_entry(handle, new.dentry, old.inode); @@ -3923,6 +3947,8 @@ static int ext4_rename(struct user_namespace *mnt_userns, struct inode *old_dir, ext4_fc_track_unlink(handle, new.dentry); __ext4_fc_track_link(handle, old.inode, new.dentry); __ext4_fc_track_unlink(handle, old.inode, old.dentry); + if (whiteout) + __ext4_fc_track_create(handle, whiteout, old.dentry); } if (new.inode) { @@ -3937,19 +3963,21 @@ static int ext4_rename(struct user_namespace *mnt_userns, struct inode *old_dir, end_rename: if (whiteout) { if (retval) { - ext4_setent(handle, &old, - old.inode->i_ino, old_file_type); + ext4_resetent(handle, &old, + old.inode->i_ino, old_file_type); drop_nlink(whiteout); + ext4_orphan_add(handle, whiteout); } unlock_new_inode(whiteout); + ext4_journal_stop(handle); iput(whiteout); - + } else { + ext4_journal_stop(handle); } +release_bh: brelse(old.dir_bh); brelse(old.bh); brelse(new.bh); - if (handle) - ext4_journal_stop(handle); return retval; } diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index 03a44a0de86a..f038d578d8d8 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -398,7 +398,7 @@ static void io_submit_init_bio(struct ext4_io_submit *io, * bio_alloc will _always_ be able to allocate a bio if * __GFP_DIRECT_RECLAIM is set, see comments for bio_alloc_bioset(). */ - bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES); + bio = bio_alloc(GFP_NOIO, BIO_MAX_VECS); fscrypt_set_bio_crypt_ctx_bh(bio, bh, GFP_NOIO); bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9); bio_set_dev(bio, bh->b_bdev); diff --git a/fs/ext4/super.c b/fs/ext4/super.c index ad34a37278cd..b9693680463a 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1210,6 +1210,7 @@ static void ext4_put_super(struct super_block *sb) percpu_counter_destroy(&sbi->s_freeinodes_counter); percpu_counter_destroy(&sbi->s_dirs_counter); percpu_counter_destroy(&sbi->s_dirtyclusters_counter); + percpu_counter_destroy(&sbi->s_sra_exceeded_retry_limit); percpu_free_rwsem(&sbi->s_writepages_rwsem); #ifdef CONFIG_QUOTA for (i = 0; i < EXT4_MAXQUOTAS; i++) @@ -5012,6 +5013,9 @@ no_journal: err = percpu_counter_init(&sbi->s_dirtyclusters_counter, 0, GFP_KERNEL); if (!err) + err = percpu_counter_init(&sbi->s_sra_exceeded_retry_limit, 0, + GFP_KERNEL); + if (!err) err = percpu_init_rwsem(&sbi->s_writepages_rwsem); if (err) { @@ -5124,6 +5128,7 @@ failed_mount6: percpu_counter_destroy(&sbi->s_freeinodes_counter); percpu_counter_destroy(&sbi->s_dirs_counter); percpu_counter_destroy(&sbi->s_dirtyclusters_counter); + percpu_counter_destroy(&sbi->s_sra_exceeded_retry_limit); percpu_free_rwsem(&sbi->s_writepages_rwsem); failed_mount5: ext4_ext_release(sb); @@ -5149,8 +5154,8 @@ failed_mount_wq: failed_mount3a: ext4_es_unregister_shrinker(sbi); failed_mount3: - del_timer_sync(&sbi->s_err_report); flush_work(&sbi->s_error_work); + del_timer_sync(&sbi->s_err_report); if (sbi->s_mmp_tsk) kthread_stop(sbi->s_mmp_tsk); failed_mount2: diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c index 075aa3a19ff5..a3d08276d441 100644 --- a/fs/ext4/sysfs.c +++ b/fs/ext4/sysfs.c @@ -24,6 +24,7 @@ typedef enum { attr_session_write_kbytes, attr_lifetime_write_kbytes, attr_reserved_clusters, + attr_sra_exceeded_retry_limit, attr_inode_readahead, attr_trigger_test_error, attr_first_error_time, @@ -202,6 +203,7 @@ EXT4_ATTR_FUNC(delayed_allocation_blocks, 0444); EXT4_ATTR_FUNC(session_write_kbytes, 0444); EXT4_ATTR_FUNC(lifetime_write_kbytes, 0444); EXT4_ATTR_FUNC(reserved_clusters, 0644); +EXT4_ATTR_FUNC(sra_exceeded_retry_limit, 0444); EXT4_ATTR_OFFSET(inode_readahead_blks, 0644, inode_readahead, ext4_sb_info, s_inode_readahead_blks); @@ -251,6 +253,7 @@ static struct attribute *ext4_attrs[] = { ATTR_LIST(session_write_kbytes), ATTR_LIST(lifetime_write_kbytes), ATTR_LIST(reserved_clusters), + ATTR_LIST(sra_exceeded_retry_limit), ATTR_LIST(inode_readahead_blks), ATTR_LIST(inode_goal), ATTR_LIST(mb_stats), @@ -374,6 +377,10 @@ static ssize_t ext4_attr_show(struct kobject *kobj, return snprintf(buf, PAGE_SIZE, "%llu\n", (unsigned long long) atomic64_read(&sbi->s_resv_clusters)); + case attr_sra_exceeded_retry_limit: + return snprintf(buf, PAGE_SIZE, "%llu\n", + (unsigned long long) + percpu_counter_sum(&sbi->s_sra_exceeded_retry_limit)); case attr_inode_readahead: case attr_pointer_ui: if (!ptr) diff --git a/fs/ext4/verity.c b/fs/ext4/verity.c index 5b7ba8f71153..00e3cbde472e 100644 --- a/fs/ext4/verity.c +++ b/fs/ext4/verity.c @@ -201,55 +201,76 @@ static int ext4_end_enable_verity(struct file *filp, const void *desc, struct inode *inode = file_inode(filp); const int credits = 2; /* superblock and inode for ext4_orphan_del() */ handle_t *handle; + struct ext4_iloc iloc; int err = 0; - int err2; - if (desc != NULL) { - /* Succeeded; write the verity descriptor. */ - err = ext4_write_verity_descriptor(inode, desc, desc_size, - merkle_tree_size); - - /* Write all pages before clearing VERITY_IN_PROGRESS. */ - if (!err) - err = filemap_write_and_wait(inode->i_mapping); - } + /* + * If an error already occurred (which fs/verity/ signals by passing + * desc == NULL), then only clean-up is needed. + */ + if (desc == NULL) + goto cleanup; - /* If we failed, truncate anything we wrote past i_size. */ - if (desc == NULL || err) - ext4_truncate(inode); + /* Append the verity descriptor. */ + err = ext4_write_verity_descriptor(inode, desc, desc_size, + merkle_tree_size); + if (err) + goto cleanup; /* - * We must always clean up by clearing EXT4_STATE_VERITY_IN_PROGRESS and - * deleting the inode from the orphan list, even if something failed. - * If everything succeeded, we'll also set the verity bit in the same - * transaction. + * Write all pages (both data and verity metadata). Note that this must + * happen before clearing EXT4_STATE_VERITY_IN_PROGRESS; otherwise pages + * beyond i_size won't be written properly. For crash consistency, this + * also must happen before the verity inode flag gets persisted. */ + err = filemap_write_and_wait(inode->i_mapping); + if (err) + goto cleanup; - ext4_clear_inode_state(inode, EXT4_STATE_VERITY_IN_PROGRESS); + /* + * Finally, set the verity inode flag and remove the inode from the + * orphan list (in a single transaction). + */ handle = ext4_journal_start(inode, EXT4_HT_INODE, credits); if (IS_ERR(handle)) { - ext4_orphan_del(NULL, inode); - return PTR_ERR(handle); + err = PTR_ERR(handle); + goto cleanup; } - err2 = ext4_orphan_del(handle, inode); - if (err2) - goto out_stop; + err = ext4_orphan_del(handle, inode); + if (err) + goto stop_and_cleanup; - if (desc != NULL && !err) { - struct ext4_iloc iloc; + err = ext4_reserve_inode_write(handle, inode, &iloc); + if (err) + goto stop_and_cleanup; - err = ext4_reserve_inode_write(handle, inode, &iloc); - if (err) - goto out_stop; - ext4_set_inode_flag(inode, EXT4_INODE_VERITY); - ext4_set_inode_flags(inode, false); - err = ext4_mark_iloc_dirty(handle, inode, &iloc); - } -out_stop: + ext4_set_inode_flag(inode, EXT4_INODE_VERITY); + ext4_set_inode_flags(inode, false); + err = ext4_mark_iloc_dirty(handle, inode, &iloc); + if (err) + goto stop_and_cleanup; + + ext4_journal_stop(handle); + + ext4_clear_inode_state(inode, EXT4_STATE_VERITY_IN_PROGRESS); + return 0; + +stop_and_cleanup: ext4_journal_stop(handle); - return err ?: err2; +cleanup: + /* + * Verity failed to be enabled, so clean up by truncating any verity + * metadata that was written beyond i_size (both from cache and from + * disk), removing the inode from the orphan list (if it wasn't done + * already), and clearing EXT4_STATE_VERITY_IN_PROGRESS. + */ + truncate_inode_pages(inode->i_mapping, inode->i_size); + ext4_truncate(inode); + ext4_orphan_del(NULL, inode); + ext4_clear_inode_state(inode, EXT4_STATE_VERITY_IN_PROGRESS); + return err; } static int ext4_get_verity_descriptor_location(struct inode *inode, diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 372208500f4e..6c1018223c54 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -1462,6 +1462,9 @@ ext4_xattr_inode_cache_find(struct inode *inode, const void *value, if (!ce) return NULL; + WARN_ON_ONCE(ext4_handle_valid(journal_current_handle()) && + !(current->flags & PF_MEMALLOC_NOFS)); + ea_data = kvmalloc(value_len, GFP_KERNEL); if (!ea_data) { mb_cache_entry_put(ea_inode_cache, ce); @@ -2327,6 +2330,7 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, error = -ENOSPC; goto cleanup; } + WARN_ON_ONCE(!(current->flags & PF_MEMALLOC_NOFS)); } error = ext4_reserve_inode_write(handle, inode, &is.iloc); @@ -2400,7 +2404,7 @@ retry_inode: * external inode if possible. */ if (ext4_has_feature_ea_inode(inode->i_sb) && - !i.in_inode) { + i.value_len && !i.in_inode) { i.in_inode = 1; goto retry_inode; } diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c index 174a0819ad96..be5415a0dbbc 100644 --- a/fs/f2fs/checkpoint.c +++ b/fs/f2fs/checkpoint.c @@ -292,7 +292,7 @@ void f2fs_ra_meta_pages_cond(struct f2fs_sb_info *sbi, pgoff_t index) f2fs_put_page(page, 0); if (readahead) - f2fs_ra_meta_pages(sbi, index, BIO_MAX_PAGES, META_POR, true); + f2fs_ra_meta_pages(sbi, index, BIO_MAX_VECS, META_POR, true); } static int __f2fs_write_meta_page(struct page *page, diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 7c95818639a6..4e5257c763d0 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -857,7 +857,7 @@ int f2fs_merge_page_bio(struct f2fs_io_info *fio) f2fs_submit_merged_ipu_write(fio->sbi, &bio, NULL); alloc_new: if (!bio) { - bio = __bio_alloc(fio, BIO_MAX_PAGES); + bio = __bio_alloc(fio, BIO_MAX_VECS); __attach_io_flag(fio); f2fs_set_bio_crypt_ctx(bio, fio->page->mapping->host, fio->page->index, fio, GFP_NOIO); @@ -932,7 +932,7 @@ alloc_new: fio->retry = true; goto skip; } - io->bio = __bio_alloc(fio, BIO_MAX_PAGES); + io->bio = __bio_alloc(fio, BIO_MAX_VECS); f2fs_set_bio_crypt_ctx(io->bio, fio->page->mapping->host, bio_page->index, fio, GFP_NOIO); io->fio = *fio; diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 993004f06a77..c2866561263e 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -4381,7 +4381,7 @@ static int build_sit_entries(struct f2fs_sb_info *sbi) block_t total_node_blocks = 0; do { - readed = f2fs_ra_meta_pages(sbi, start_blk, BIO_MAX_PAGES, + readed = f2fs_ra_meta_pages(sbi, start_blk, BIO_MAX_VECS, META_SIT, true); start = start_blk * sit_i->sents_per_block; diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index 229814b4f4a6..e9a7a637d688 100644 --- a/fs/f2fs/segment.h +++ b/fs/f2fs/segment.h @@ -851,7 +851,7 @@ static inline int nr_pages_to_skip(struct f2fs_sb_info *sbi, int type) else if (type == NODE) return 8 * sbi->blocks_per_seg; else if (type == META) - return 8 * BIO_MAX_PAGES; + return 8 * BIO_MAX_VECS; else return 0; } @@ -868,7 +868,7 @@ static inline long nr_pages_to_write(struct f2fs_sb_info *sbi, int type, return 0; nr_to_write = wbc->nr_to_write; - desired = BIO_MAX_PAGES; + desired = BIO_MAX_VECS; if (type == NODE) desired <<= 1; diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 7069793752f1..82592b19b4e0 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -753,9 +753,9 @@ static int parse_options(struct super_block *sb, char *options, bool is_remount) case Opt_io_size_bits: if (args->from && match_int(args, &arg)) return -EINVAL; - if (arg <= 0 || arg > __ilog2_u32(BIO_MAX_PAGES)) { + if (arg <= 0 || arg > __ilog2_u32(BIO_MAX_VECS)) { f2fs_warn(sbi, "Not support %d, larger than %d", - 1 << arg, BIO_MAX_PAGES); + 1 << arg, BIO_MAX_VECS); return -EINVAL; } F2FS_OPTION(sbi).write_io_size_bits = arg; diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index c6636b4c4ccf..c0fee830a34e 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -2229,19 +2229,21 @@ static int fuse_device_clone(struct fuse_conn *fc, struct file *new) static long fuse_dev_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { - int err = -ENOTTY; + int res; + int oldfd; + struct fuse_dev *fud = NULL; - if (cmd == FUSE_DEV_IOC_CLONE) { - int oldfd; + if (_IOC_TYPE(cmd) != FUSE_DEV_IOC_MAGIC) + return -ENOTTY; - err = -EFAULT; - if (!get_user(oldfd, (__u32 __user *) arg)) { + switch (_IOC_NR(cmd)) { + case _IOC_NR(FUSE_DEV_IOC_CLONE): + res = -EFAULT; + if (!get_user(oldfd, (__u32 __user *)arg)) { struct file *old = fget(oldfd); - err = -EINVAL; + res = -EINVAL; if (old) { - struct fuse_dev *fud = NULL; - /* * Check against file->f_op because CUSE * uses the same ioctl handler. @@ -2252,14 +2254,18 @@ static long fuse_dev_ioctl(struct file *file, unsigned int cmd, if (fud) { mutex_lock(&fuse_mutex); - err = fuse_device_clone(fud->fc, file); + res = fuse_device_clone(fud->fc, file); mutex_unlock(&fuse_mutex); } fput(old); } } + break; + default: + res = -ENOTTY; + break; } - return err; + return res; } const struct file_operations fuse_dev_operations = { diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 68cca8d4db6e..63d97a15ffde 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -863,6 +863,7 @@ static inline u64 fuse_get_attr_version(struct fuse_conn *fc) static inline void fuse_make_bad(struct inode *inode) { + remove_inode_hash(inode); set_bit(FUSE_I_BAD, &get_fuse_inode(inode)->state); } diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c index 8868ac31a3c0..4ee6f734ba83 100644 --- a/fs/fuse/virtio_fs.c +++ b/fs/fuse/virtio_fs.c @@ -1324,8 +1324,15 @@ static int virtio_fs_fill_super(struct super_block *sb, struct fs_context *fsc) /* virtiofs allocates and installs its own fuse devices */ ctx->fudptr = NULL; - if (ctx->dax) + if (ctx->dax) { + if (!fs->dax_dev) { + err = -EINVAL; + pr_err("virtio-fs: dax can't be enabled as filesystem" + " device does not support it.\n"); + goto err_free_fuse_devs; + } ctx->dax_dev = fs->dax_dev; + } err = fuse_fill_super_common(sb, ctx); if (err < 0) goto err_free_fuse_devs; diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 16937ebb2a3e..6410281546f9 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -998,12 +998,16 @@ static void trans_drain(struct gfs2_trans *tr) while (!list_empty(head)) { bd = list_first_entry(head, struct gfs2_bufdata, bd_list); list_del_init(&bd->bd_list); + if (!list_empty(&bd->bd_ail_st_list)) + gfs2_remove_from_ail(bd); kmem_cache_free(gfs2_bufdata_cachep, bd); } head = &tr->tr_databuf; while (!list_empty(head)) { bd = list_first_entry(head, struct gfs2_bufdata, bd_list); list_del_init(&bd->bd_list); + if (!list_empty(&bd->bd_ail_st_list)) + gfs2_remove_from_ail(bd); kmem_cache_free(gfs2_bufdata_cachep, bd); } } @@ -1032,7 +1036,7 @@ repeat: * Do this check while holding the log_flush_lock to prevent new * buffers from being added to the ail via gfs2_pin() */ - if (gfs2_withdrawn(sdp)) + if (gfs2_withdrawn(sdp) || !test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) goto out; /* Log might have been flushed while we waited for the flush lock */ diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index dc1b93a877c6..a82f4747aa8d 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -267,7 +267,7 @@ static struct bio *gfs2_log_alloc_bio(struct gfs2_sbd *sdp, u64 blkno, bio_end_io_t *end_io) { struct super_block *sb = sdp->sd_vfs; - struct bio *bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES); + struct bio *bio = bio_alloc(GFP_NOIO, BIO_MAX_VECS); bio->bi_iter.bi_sector = blkno << sdp->sd_fsb2bb_shift; bio_set_dev(bio, sb->s_bdev); diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 74c7d01723b9..aa4136055a83 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -1539,9 +1539,7 @@ static int gfs2_reconfigure(struct fs_context *fc) return -EINVAL; if (fc->sb_flags & SB_RDONLY) { - error = gfs2_make_fs_ro(sdp); - if (error) - errorfc(fc, "unable to remount read-only"); + gfs2_make_fs_ro(sdp); } else { error = gfs2_make_fs_rw(sdp); if (error) diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 861ed5fe02a5..97076d3f562f 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -587,9 +587,8 @@ out: * Returns: errno */ -int gfs2_make_fs_ro(struct gfs2_sbd *sdp) +void gfs2_make_fs_ro(struct gfs2_sbd *sdp) { - int error = 0; int log_write_allowed = test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); gfs2_flush_delete_work(sdp); @@ -624,8 +623,6 @@ int gfs2_make_fs_ro(struct gfs2_sbd *sdp) if (!log_write_allowed) sdp->sd_vfs->s_flags |= SB_RDONLY; - - return error; } /** @@ -637,7 +634,6 @@ int gfs2_make_fs_ro(struct gfs2_sbd *sdp) static void gfs2_put_super(struct super_block *sb) { struct gfs2_sbd *sdp = sb->s_fs_info; - int error; struct gfs2_jdesc *jd; /* No more recovery requests */ @@ -658,9 +654,7 @@ restart: spin_unlock(&sdp->sd_jindex_spin); if (!sb_rdonly(sb)) { - error = gfs2_make_fs_ro(sdp); - if (error) - gfs2_io_error(sdp); + gfs2_make_fs_ro(sdp); } WARN_ON(gfs2_withdrawing(sdp)); diff --git a/fs/gfs2/super.h b/fs/gfs2/super.h index 08e502dec7ec..ec4affb33ed5 100644 --- a/fs/gfs2/super.h +++ b/fs/gfs2/super.h @@ -34,7 +34,7 @@ extern int gfs2_lookup_in_master_dir(struct gfs2_sbd *sdp, char *filename, struct gfs2_inode **ipp); extern int gfs2_make_fs_rw(struct gfs2_sbd *sdp); -extern int gfs2_make_fs_ro(struct gfs2_sbd *sdp); +extern void gfs2_make_fs_ro(struct gfs2_sbd *sdp); extern void gfs2_online_uevent(struct gfs2_sbd *sdp); extern int gfs2_statfs_init(struct gfs2_sbd *sdp); extern void gfs2_statfs_change(struct gfs2_sbd *sdp, s64 total, s64 free, diff --git a/fs/gfs2/trans.c b/fs/gfs2/trans.c index ab96cf0bf26b..63fec11ef2ce 100644 --- a/fs/gfs2/trans.c +++ b/fs/gfs2/trans.c @@ -169,6 +169,8 @@ static struct gfs2_bufdata *gfs2_alloc_bufdata(struct gfs2_glock *gl, bd->bd_bh = bh; bd->bd_gl = gl; INIT_LIST_HEAD(&bd->bd_list); + INIT_LIST_HEAD(&bd->bd_ail_st_list); + INIT_LIST_HEAD(&bd->bd_ail_gl_list); bh->b_private = bd; return bd; } diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c index 8d3c670c990f..4f034b87b427 100644 --- a/fs/gfs2/util.c +++ b/fs/gfs2/util.c @@ -119,17 +119,22 @@ void gfs2_freeze_unlock(struct gfs2_holder *freeze_gh) static void signal_our_withdraw(struct gfs2_sbd *sdp) { struct gfs2_glock *live_gl = sdp->sd_live_gh.gh_gl; - struct inode *inode = sdp->sd_jdesc->jd_inode; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_glock *i_gl = ip->i_gl; - u64 no_formal_ino = ip->i_no_formal_ino; + struct inode *inode; + struct gfs2_inode *ip; + struct gfs2_glock *i_gl; + u64 no_formal_ino; int log_write_allowed = test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); int ret = 0; int tries; - if (test_bit(SDF_NORECOVERY, &sdp->sd_flags)) + if (test_bit(SDF_NORECOVERY, &sdp->sd_flags) || !sdp->sd_jdesc) return; + inode = sdp->sd_jdesc->jd_inode; + ip = GFS2_I(inode); + i_gl = ip->i_gl; + no_formal_ino = ip->i_no_formal_ino; + /* Prevent any glock dq until withdraw recovery is complete */ set_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags); /* @@ -156,7 +161,7 @@ static void signal_our_withdraw(struct gfs2_sbd *sdp) ret = 0; } if (!ret) - ret = gfs2_make_fs_ro(sdp); + gfs2_make_fs_ro(sdp); gfs2_freeze_unlock(&freeze_gh); } diff --git a/fs/io-wq.c b/fs/io-wq.c index 28868eb4cd09..3dc10bfd8c3b 100644 --- a/fs/io-wq.c +++ b/fs/io-wq.c @@ -110,7 +110,6 @@ struct io_wq { io_wq_work_fn *do_work; struct task_struct *manager; - struct user_struct *user; struct io_wq_hash *hash; @@ -387,13 +386,16 @@ static struct io_wq_work *io_get_next_work(struct io_wqe *wqe) return NULL; } -static void io_flush_signals(void) +static bool io_flush_signals(void) { if (unlikely(test_tsk_thread_flag(current, TIF_NOTIFY_SIGNAL))) { + __set_current_state(TASK_RUNNING); if (current->task_works) task_work_run(); clear_tsk_thread_flag(current, TIF_NOTIFY_SIGNAL); + return true; } + return false; } static void io_assign_current_work(struct io_worker *worker, @@ -489,6 +491,8 @@ static int io_wqe_worker(void *data) set_task_comm(current, buf); while (!test_bit(IO_WQ_BIT_EXIT, &wq->state)) { + long ret; + set_current_state(TASK_INTERRUPTIBLE); loop: raw_spin_lock_irq(&wqe->lock); @@ -498,8 +502,10 @@ loop: } __io_worker_idle(wqe, worker); raw_spin_unlock_irq(&wqe->lock); - io_flush_signals(); - if (schedule_timeout(WORKER_IDLE_TIMEOUT)) + if (io_flush_signals()) + continue; + ret = schedule_timeout(WORKER_IDLE_TIMEOUT); + if (try_to_freeze() || ret) continue; if (fatal_signal_pending(current)) break; @@ -592,7 +598,7 @@ static bool create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index) tsk->pf_io_worker = worker; worker->task = tsk; set_cpus_allowed_ptr(tsk, cpumask_of_node(wqe->node)); - tsk->flags |= PF_NOFREEZE | PF_NO_SETAFFINITY; + tsk->flags |= PF_NO_SETAFFINITY; raw_spin_lock_irq(&wqe->lock); hlist_nulls_add_head_rcu(&worker->nulls_node, &wqe->free_list); @@ -722,9 +728,9 @@ static int io_wq_manager(void *data) io_wq_for_each_worker(wq->wqes[node], io_wq_worker_wake, NULL); rcu_read_unlock(); - /* we might not ever have created any workers */ - if (atomic_read(&wq->worker_refs)) - wait_for_completion(&wq->worker_done); + if (atomic_dec_and_test(&wq->worker_refs)) + complete(&wq->worker_done); + wait_for_completion(&wq->worker_done); spin_lock_irq(&wq->hash->wait.lock); for_each_node(node) @@ -774,7 +780,10 @@ static int io_wq_fork_manager(struct io_wq *wq) if (wq->manager) return 0; - reinit_completion(&wq->worker_done); + WARN_ON_ONCE(test_bit(IO_WQ_BIT_EXIT, &wq->state)); + + init_completion(&wq->worker_done); + atomic_set(&wq->worker_refs, 1); tsk = create_io_thread(io_wq_manager, wq, NUMA_NO_NODE); if (!IS_ERR(tsk)) { wq->manager = get_task_struct(tsk); @@ -782,6 +791,9 @@ static int io_wq_fork_manager(struct io_wq *wq) return 0; } + if (atomic_dec_and_test(&wq->worker_refs)) + complete(&wq->worker_done); + return PTR_ERR(tsk); } @@ -794,8 +806,7 @@ static void io_wqe_enqueue(struct io_wqe *wqe, struct io_wq_work *work) /* Can only happen if manager creation fails after exec */ if (io_wq_fork_manager(wqe->wq) || test_bit(IO_WQ_BIT_EXIT, &wqe->wq->state)) { - work->flags |= IO_WQ_WORK_CANCEL; - wqe->wq->do_work(work); + io_run_cancel(work, wqe); return; } @@ -1018,13 +1029,9 @@ struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) init_completion(&wq->exited); refcount_set(&wq->refs, 1); - init_completion(&wq->worker_done); - atomic_set(&wq->worker_refs, 0); - ret = io_wq_fork_manager(wq); if (!ret) return wq; - err: io_wq_put_hash(data->hash); cpuhp_state_remove_instance_nocalls(io_wq_online, &wq->cpuhp_node); diff --git a/fs/io-wq.h b/fs/io-wq.h index 5fbf7997149e..80d590564ff9 100644 --- a/fs/io-wq.h +++ b/fs/io-wq.h @@ -2,7 +2,6 @@ #define INTERNAL_IO_WQ_H #include <linux/refcount.h> -#include <linux/io_uring.h> struct io_wq; @@ -21,6 +20,15 @@ enum io_wq_cancel { IO_WQ_CANCEL_NOTFOUND, /* work not found */ }; +struct io_wq_work_node { + struct io_wq_work_node *next; +}; + +struct io_wq_work_list { + struct io_wq_work_node *first; + struct io_wq_work_node *last; +}; + static inline void wq_list_add_after(struct io_wq_work_node *node, struct io_wq_work_node *pos, struct io_wq_work_list *list) @@ -79,8 +87,8 @@ static inline void wq_list_del(struct io_wq_work_list *list, struct io_wq_work { struct io_wq_work_node list; + const struct cred *creds; unsigned flags; - unsigned short personality; }; static inline struct io_wq_work *wq_next_work(struct io_wq_work *work) diff --git a/fs/io_uring.c b/fs/io_uring.c index 92c25b5f1349..543551d70327 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -258,12 +258,11 @@ enum { struct io_sq_data { refcount_t refs; + atomic_t park_pending; struct mutex lock; /* ctx's that are using this sqd */ struct list_head ctx_list; - struct list_head ctx_new_list; - struct mutex ctx_lock; struct task_struct *thread; struct wait_queue_head wait; @@ -271,11 +270,11 @@ struct io_sq_data { unsigned sq_thread_idle; int sq_cpu; pid_t task_pid; + pid_t task_tgid; unsigned long state; - struct completion startup; - struct completion parked; struct completion exited; + struct callback_head *park_task_work; }; #define IO_IOPOLL_BATCH 8 @@ -336,7 +335,6 @@ struct io_ring_ctx { unsigned int drain_next: 1; unsigned int eventfd_async: 1; unsigned int restricted: 1; - unsigned int sqo_exec: 1; /* * Ring buffer of indices into array of io_uring_sqe, which is @@ -380,6 +378,7 @@ struct io_ring_ctx { /* Only used for accounting purposes */ struct mm_struct *mm_account; + const struct cred *sq_creds; /* cred used for __io_sq_thread() */ struct io_sq_data *sq_data; /* if using sq thread polling */ struct wait_queue_head sqo_sq_wait; @@ -400,15 +399,15 @@ struct io_ring_ctx { struct user_struct *user; struct completion ref_comp; - struct completion sq_thread_comp; #if defined(CONFIG_UNIX) struct socket *ring_sock; #endif - struct idr io_buffer_idr; + struct xarray io_buffers; - struct idr personality_idr; + struct xarray personalities; + u32 pers_next; struct { unsigned cached_cq_tail; @@ -454,6 +453,23 @@ struct io_ring_ctx { /* Keep this last, we don't need it for the fast path */ struct work_struct exit_work; + struct list_head tctx_list; +}; + +struct io_uring_task { + /* submission side */ + struct xarray xa; + struct wait_queue_head wait; + const struct io_ring_ctx *last; + struct io_wq *io_wq; + struct percpu_counter inflight; + atomic_t in_idle; + bool sqpoll; + + spinlock_t task_lock; + struct io_wq_work_list task_list; + unsigned long task_state; + struct callback_head task_work; }; /* @@ -805,6 +821,12 @@ struct io_kiocb { struct io_wq_work work; }; +struct io_tctx_node { + struct list_head ctx_node; + struct task_struct *task; + struct io_ring_ctx *ctx; +}; + struct io_defer_entry { struct list_head list; struct io_kiocb *req; @@ -979,6 +1001,8 @@ static const struct io_op_def io_op_defs[] = { [IORING_OP_UNLINKAT] = {}, }; +static bool io_disarm_next(struct io_kiocb *req); +static void io_uring_del_task_file(unsigned long index); static void io_uring_try_cancel_requests(struct io_ring_ctx *ctx, struct task_struct *task, struct files_struct *files); @@ -1129,9 +1153,8 @@ static struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p) init_waitqueue_head(&ctx->cq_wait); INIT_LIST_HEAD(&ctx->cq_overflow_list); init_completion(&ctx->ref_comp); - init_completion(&ctx->sq_thread_comp); - idr_init(&ctx->io_buffer_idr); - idr_init(&ctx->personality_idr); + xa_init_flags(&ctx->io_buffers, XA_FLAGS_ALLOC1); + xa_init_flags(&ctx->personalities, XA_FLAGS_ALLOC1); mutex_init(&ctx->uring_lock); init_waitqueue_head(&ctx->wait); spin_lock_init(&ctx->completion_lock); @@ -1144,6 +1167,7 @@ static struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p) INIT_LIST_HEAD(&ctx->rsrc_ref_list); INIT_DELAYED_WORK(&ctx->rsrc_put_work, io_rsrc_put_work); init_llist_head(&ctx->rsrc_put_llist); + INIT_LIST_HEAD(&ctx->tctx_list); INIT_LIST_HEAD(&ctx->submit_state.comp.free_list); INIT_LIST_HEAD(&ctx->submit_state.comp.locked_free_list); return ctx; @@ -1183,6 +1207,9 @@ static void io_prep_async_work(struct io_kiocb *req) const struct io_op_def *def = &io_op_defs[req->opcode]; struct io_ring_ctx *ctx = req->ctx; + if (!req->work.creds) + req->work.creds = get_current_cred(); + if (req->flags & REQ_F_FORCE_ASYNC) req->work.flags |= IO_WQ_WORK_CONCURRENT; @@ -1514,15 +1541,14 @@ static void io_cqring_fill_event(struct io_kiocb *req, long res) __io_cqring_fill_event(req, res, 0); } -static inline void io_req_complete_post(struct io_kiocb *req, long res, - unsigned int cflags) +static void io_req_complete_post(struct io_kiocb *req, long res, + unsigned int cflags) { struct io_ring_ctx *ctx = req->ctx; unsigned long flags; spin_lock_irqsave(&ctx->completion_lock, flags); __io_cqring_fill_event(req, res, cflags); - io_commit_cqring(ctx); /* * If we're the last reference to this request, add to our locked * free_list cache. @@ -1530,17 +1556,27 @@ static inline void io_req_complete_post(struct io_kiocb *req, long res, if (refcount_dec_and_test(&req->refs)) { struct io_comp_state *cs = &ctx->submit_state.comp; + if (req->flags & (REQ_F_LINK | REQ_F_HARDLINK)) { + if (req->flags & (REQ_F_LINK_TIMEOUT | REQ_F_FAIL_LINK)) + io_disarm_next(req); + if (req->link) { + io_req_task_queue(req->link); + req->link = NULL; + } + } io_dismantle_req(req); io_put_task(req->task, 1); list_add(&req->compl.list, &cs->locked_free_list); cs->locked_free_nr++; - } else - req = NULL; + } else { + if (!percpu_ref_tryget(&ctx->refs)) + req = NULL; + } + io_commit_cqring(ctx); spin_unlock_irqrestore(&ctx->completion_lock, flags); - io_cqring_ev_posted(ctx); if (req) { - io_queue_next(req); + io_cqring_ev_posted(ctx); percpu_ref_put(&ctx->refs); } } @@ -1648,6 +1684,10 @@ static void io_dismantle_req(struct io_kiocb *req) io_put_file(req, req->file, (req->flags & REQ_F_FIXED_FILE)); if (req->fixed_rsrc_refs) percpu_ref_put(req->fixed_rsrc_refs); + if (req->work.creds) { + put_cred(req->work.creds); + req->work.creds = NULL; + } if (req->flags & REQ_F_INFLIGHT) { struct io_ring_ctx *ctx = req->ctx; @@ -1690,15 +1730,11 @@ static inline void io_remove_next_linked(struct io_kiocb *req) nxt->link = NULL; } -static void io_kill_linked_timeout(struct io_kiocb *req) +static bool io_kill_linked_timeout(struct io_kiocb *req) + __must_hold(&req->ctx->completion_lock) { - struct io_ring_ctx *ctx = req->ctx; - struct io_kiocb *link; + struct io_kiocb *link = req->link; bool cancelled = false; - unsigned long flags; - - spin_lock_irqsave(&ctx->completion_lock, flags); - link = req->link; /* * Can happen if a linked timeout fired and link had been like @@ -1713,50 +1749,48 @@ static void io_kill_linked_timeout(struct io_kiocb *req) ret = hrtimer_try_to_cancel(&io->timer); if (ret != -1) { io_cqring_fill_event(link, -ECANCELED); - io_commit_cqring(ctx); + io_put_req_deferred(link, 1); cancelled = true; } } req->flags &= ~REQ_F_LINK_TIMEOUT; - spin_unlock_irqrestore(&ctx->completion_lock, flags); - - if (cancelled) { - io_cqring_ev_posted(ctx); - io_put_req(link); - } + return cancelled; } - static void io_fail_links(struct io_kiocb *req) + __must_hold(&req->ctx->completion_lock) { - struct io_kiocb *link, *nxt; - struct io_ring_ctx *ctx = req->ctx; - unsigned long flags; + struct io_kiocb *nxt, *link = req->link; - spin_lock_irqsave(&ctx->completion_lock, flags); - link = req->link; req->link = NULL; - while (link) { nxt = link->link; link->link = NULL; trace_io_uring_fail_link(req, link); io_cqring_fill_event(link, -ECANCELED); - io_put_req_deferred(link, 2); link = nxt; } - io_commit_cqring(ctx); - spin_unlock_irqrestore(&ctx->completion_lock, flags); +} - io_cqring_ev_posted(ctx); +static bool io_disarm_next(struct io_kiocb *req) + __must_hold(&req->ctx->completion_lock) +{ + bool posted = false; + + if (likely(req->flags & REQ_F_LINK_TIMEOUT)) + posted = io_kill_linked_timeout(req); + if (unlikely(req->flags & REQ_F_FAIL_LINK)) { + posted |= (req->link != NULL); + io_fail_links(req); + } + return posted; } static struct io_kiocb *__io_req_find_next(struct io_kiocb *req) { - if (req->flags & REQ_F_LINK_TIMEOUT) - io_kill_linked_timeout(req); + struct io_kiocb *nxt; /* * If LINK is set, we have dependent requests in this chain. If we @@ -1764,14 +1798,22 @@ static struct io_kiocb *__io_req_find_next(struct io_kiocb *req) * dependencies to the next request. In case of failure, fail the rest * of the chain. */ - if (likely(!(req->flags & REQ_F_FAIL_LINK))) { - struct io_kiocb *nxt = req->link; + if (req->flags & (REQ_F_LINK_TIMEOUT | REQ_F_FAIL_LINK)) { + struct io_ring_ctx *ctx = req->ctx; + unsigned long flags; + bool posted; - req->link = NULL; - return nxt; + spin_lock_irqsave(&ctx->completion_lock, flags); + posted = io_disarm_next(req); + if (posted) + io_commit_cqring(req->ctx); + spin_unlock_irqrestore(&ctx->completion_lock, flags); + if (posted) + io_cqring_ev_posted(ctx); } - io_fail_links(req); - return NULL; + nxt = req->link; + req->link = NULL; + return nxt; } static inline struct io_kiocb *io_req_find_next(struct io_kiocb *req) @@ -1904,17 +1946,44 @@ static int io_req_task_work_add(struct io_kiocb *req) return ret; } -static void io_req_task_work_add_fallback(struct io_kiocb *req, - task_work_func_t cb) +static bool io_run_task_work_head(struct callback_head **work_head) +{ + struct callback_head *work, *next; + bool executed = false; + + do { + work = xchg(work_head, NULL); + if (!work) + break; + + do { + next = work->next; + work->func(work); + work = next; + cond_resched(); + } while (work); + executed = true; + } while (1); + + return executed; +} + +static void io_task_work_add_head(struct callback_head **work_head, + struct callback_head *task_work) { - struct io_ring_ctx *ctx = req->ctx; struct callback_head *head; - init_task_work(&req->task_work, cb); do { - head = READ_ONCE(ctx->exit_task_work); - req->task_work.next = head; - } while (cmpxchg(&ctx->exit_task_work, head, &req->task_work) != head); + head = READ_ONCE(*work_head); + task_work->next = head; + } while (cmpxchg(work_head, head, task_work) != head); +} + +static void io_req_task_work_add_fallback(struct io_kiocb *req, + task_work_func_t cb) +{ + init_task_work(&req->task_work, cb); + io_task_work_add_head(&req->ctx->exit_task_work, &req->task_work); } static void __io_req_task_cancel(struct io_kiocb *req, int error) @@ -2822,7 +2891,7 @@ static struct io_buffer *io_buffer_select(struct io_kiocb *req, size_t *len, lockdep_assert_held(&req->ctx->uring_lock); - head = idr_find(&req->ctx->io_buffer_idr, bgid); + head = xa_load(&req->ctx->io_buffers, bgid); if (head) { if (!list_empty(&head->list)) { kbuf = list_last_entry(&head->list, struct io_buffer, @@ -2830,7 +2899,7 @@ static struct io_buffer *io_buffer_select(struct io_kiocb *req, size_t *len, list_del(&kbuf->list); } else { kbuf = head; - idr_remove(&req->ctx->io_buffer_idr, bgid); + xa_erase(&req->ctx->io_buffers, bgid); } if (*len > kbuf->len) *len = kbuf->len; @@ -3871,7 +3940,7 @@ static int __io_remove_buffers(struct io_ring_ctx *ctx, struct io_buffer *buf, } i++; kfree(buf); - idr_remove(&ctx->io_buffer_idr, bgid); + xa_erase(&ctx->io_buffers, bgid); return i; } @@ -3889,7 +3958,7 @@ static int io_remove_buffers(struct io_kiocb *req, unsigned int issue_flags) lockdep_assert_held(&ctx->uring_lock); ret = -ENOENT; - head = idr_find(&ctx->io_buffer_idr, p->bgid); + head = xa_load(&ctx->io_buffers, p->bgid); if (head) ret = __io_remove_buffers(ctx, head, p->bgid, p->nbufs); if (ret < 0) @@ -3972,21 +4041,14 @@ static int io_provide_buffers(struct io_kiocb *req, unsigned int issue_flags) lockdep_assert_held(&ctx->uring_lock); - list = head = idr_find(&ctx->io_buffer_idr, p->bgid); + list = head = xa_load(&ctx->io_buffers, p->bgid); ret = io_add_buffers(p, &head); - if (ret < 0) - goto out; - - if (!list) { - ret = idr_alloc(&ctx->io_buffer_idr, head, p->bgid, p->bgid + 1, - GFP_KERNEL); - if (ret < 0) { + if (ret >= 0 && !list) { + ret = xa_insert(&ctx->io_buffers, p->bgid, head, GFP_KERNEL); + if (ret < 0) __io_remove_buffers(ctx, head, p->bgid, -1U); - goto out; - } } -out: if (ret < 0) req_set_fail_links(req); @@ -4324,6 +4386,7 @@ static int io_sendmsg(struct io_kiocb *req, unsigned int issue_flags) struct io_async_msghdr iomsg, *kmsg; struct socket *sock; unsigned flags; + int min_ret = 0; int ret; sock = sock_from_file(req->file); @@ -4338,12 +4401,15 @@ static int io_sendmsg(struct io_kiocb *req, unsigned int issue_flags) kmsg = &iomsg; } - flags = req->sr_msg.msg_flags; + flags = req->sr_msg.msg_flags | MSG_NOSIGNAL; if (flags & MSG_DONTWAIT) req->flags |= REQ_F_NOWAIT; else if (issue_flags & IO_URING_F_NONBLOCK) flags |= MSG_DONTWAIT; + if (flags & MSG_WAITALL) + min_ret = iov_iter_count(&kmsg->msg.msg_iter); + ret = __sys_sendmsg_sock(sock, &kmsg->msg, flags); if ((issue_flags & IO_URING_F_NONBLOCK) && ret == -EAGAIN) return io_setup_async_msg(req, kmsg); @@ -4354,7 +4420,7 @@ static int io_sendmsg(struct io_kiocb *req, unsigned int issue_flags) if (kmsg->free_iov) kfree(kmsg->free_iov); req->flags &= ~REQ_F_NEED_CLEANUP; - if (ret < 0) + if (ret < min_ret) req_set_fail_links(req); __io_req_complete(req, issue_flags, ret, 0); return 0; @@ -4367,6 +4433,7 @@ static int io_send(struct io_kiocb *req, unsigned int issue_flags) struct iovec iov; struct socket *sock; unsigned flags; + int min_ret = 0; int ret; sock = sock_from_file(req->file); @@ -4382,12 +4449,15 @@ static int io_send(struct io_kiocb *req, unsigned int issue_flags) msg.msg_controllen = 0; msg.msg_namelen = 0; - flags = req->sr_msg.msg_flags; + flags = req->sr_msg.msg_flags | MSG_NOSIGNAL; if (flags & MSG_DONTWAIT) req->flags |= REQ_F_NOWAIT; else if (issue_flags & IO_URING_F_NONBLOCK) flags |= MSG_DONTWAIT; + if (flags & MSG_WAITALL) + min_ret = iov_iter_count(&msg.msg_iter); + msg.msg_flags = flags; ret = sock_sendmsg(sock, &msg); if ((issue_flags & IO_URING_F_NONBLOCK) && ret == -EAGAIN) @@ -4395,7 +4465,7 @@ static int io_send(struct io_kiocb *req, unsigned int issue_flags) if (ret == -ERESTARTSYS) ret = -EINTR; - if (ret < 0) + if (ret < min_ret) req_set_fail_links(req); __io_req_complete(req, issue_flags, ret, 0); return 0; @@ -4547,6 +4617,7 @@ static int io_recvmsg(struct io_kiocb *req, unsigned int issue_flags) struct socket *sock; struct io_buffer *kbuf; unsigned flags; + int min_ret = 0; int ret, cflags = 0; bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; @@ -4572,12 +4643,15 @@ static int io_recvmsg(struct io_kiocb *req, unsigned int issue_flags) 1, req->sr_msg.len); } - flags = req->sr_msg.msg_flags; + flags = req->sr_msg.msg_flags | MSG_NOSIGNAL; if (flags & MSG_DONTWAIT) req->flags |= REQ_F_NOWAIT; else if (force_nonblock) flags |= MSG_DONTWAIT; + if (flags & MSG_WAITALL) + min_ret = iov_iter_count(&kmsg->msg.msg_iter); + ret = __sys_recvmsg_sock(sock, &kmsg->msg, req->sr_msg.umsg, kmsg->uaddr, flags); if (force_nonblock && ret == -EAGAIN) @@ -4591,7 +4665,7 @@ static int io_recvmsg(struct io_kiocb *req, unsigned int issue_flags) if (kmsg->free_iov) kfree(kmsg->free_iov); req->flags &= ~REQ_F_NEED_CLEANUP; - if (ret < 0) + if (ret < min_ret || ((flags & MSG_WAITALL) && (kmsg->msg.msg_flags & (MSG_TRUNC | MSG_CTRUNC)))) req_set_fail_links(req); __io_req_complete(req, issue_flags, ret, cflags); return 0; @@ -4606,6 +4680,7 @@ static int io_recv(struct io_kiocb *req, unsigned int issue_flags) struct socket *sock; struct iovec iov; unsigned flags; + int min_ret = 0; int ret, cflags = 0; bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; @@ -4631,12 +4706,15 @@ static int io_recv(struct io_kiocb *req, unsigned int issue_flags) msg.msg_iocb = NULL; msg.msg_flags = 0; - flags = req->sr_msg.msg_flags; + flags = req->sr_msg.msg_flags | MSG_NOSIGNAL; if (flags & MSG_DONTWAIT) req->flags |= REQ_F_NOWAIT; else if (force_nonblock) flags |= MSG_DONTWAIT; + if (flags & MSG_WAITALL) + min_ret = iov_iter_count(&msg.msg_iter); + ret = sock_recvmsg(sock, &msg, flags); if (force_nonblock && ret == -EAGAIN) return -EAGAIN; @@ -4645,7 +4723,7 @@ static int io_recv(struct io_kiocb *req, unsigned int issue_flags) out_free: if (req->flags & REQ_F_BUFFER_SELECTED) cflags = io_put_recv_kbuf(req); - if (ret < 0) + if (ret < min_ret || ((flags & MSG_WAITALL) && (msg.msg_flags & (MSG_TRUNC | MSG_CTRUNC)))) req_set_fail_links(req); __io_req_complete(req, issue_flags, ret, cflags); return 0; @@ -5559,22 +5637,30 @@ add: return 0; } +struct io_cancel_data { + struct io_ring_ctx *ctx; + u64 user_data; +}; + static bool io_cancel_cb(struct io_wq_work *work, void *data) { struct io_kiocb *req = container_of(work, struct io_kiocb, work); + struct io_cancel_data *cd = data; - return req->user_data == (unsigned long) data; + return req->ctx == cd->ctx && req->user_data == cd->user_data; } -static int io_async_cancel_one(struct io_uring_task *tctx, void *sqe_addr) +static int io_async_cancel_one(struct io_uring_task *tctx, u64 user_data, + struct io_ring_ctx *ctx) { + struct io_cancel_data data = { .ctx = ctx, .user_data = user_data, }; enum io_wq_cancel cancel_ret; int ret = 0; - if (!tctx->io_wq) + if (!tctx || !tctx->io_wq) return -ENOENT; - cancel_ret = io_wq_cancel_cb(tctx->io_wq, io_cancel_cb, sqe_addr, false); + cancel_ret = io_wq_cancel_cb(tctx->io_wq, io_cancel_cb, &data, false); switch (cancel_ret) { case IO_WQ_CANCEL_OK: ret = 0; @@ -5597,8 +5683,7 @@ static void io_async_find_and_cancel(struct io_ring_ctx *ctx, unsigned long flags; int ret; - ret = io_async_cancel_one(req->task->io_uring, - (void *) (unsigned long) sqe_addr); + ret = io_async_cancel_one(req->task->io_uring, sqe_addr, ctx); if (ret != -ENOENT) { spin_lock_irqsave(&ctx->completion_lock, flags); goto done; @@ -5639,8 +5724,47 @@ static int io_async_cancel_prep(struct io_kiocb *req, static int io_async_cancel(struct io_kiocb *req, unsigned int issue_flags) { struct io_ring_ctx *ctx = req->ctx; + u64 sqe_addr = req->cancel.addr; + struct io_tctx_node *node; + int ret; - io_async_find_and_cancel(ctx, req, req->cancel.addr, 0); + /* tasks should wait for their io-wq threads, so safe w/o sync */ + ret = io_async_cancel_one(req->task->io_uring, sqe_addr, ctx); + spin_lock_irq(&ctx->completion_lock); + if (ret != -ENOENT) + goto done; + ret = io_timeout_cancel(ctx, sqe_addr); + if (ret != -ENOENT) + goto done; + ret = io_poll_cancel(ctx, sqe_addr); + if (ret != -ENOENT) + goto done; + spin_unlock_irq(&ctx->completion_lock); + + /* slow path, try all io-wq's */ + io_ring_submit_lock(ctx, !(issue_flags & IO_URING_F_NONBLOCK)); + ret = -ENOENT; + list_for_each_entry(node, &ctx->tctx_list, ctx_node) { + struct io_uring_task *tctx = node->task->io_uring; + + if (!tctx || !tctx->io_wq) + continue; + ret = io_async_cancel_one(tctx, req->cancel.addr, ctx); + if (ret != -ENOENT) + break; + } + io_ring_submit_unlock(ctx, !(issue_flags & IO_URING_F_NONBLOCK)); + + spin_lock_irq(&ctx->completion_lock); +done: + io_cqring_fill_event(req, ret); + io_commit_cqring(ctx); + spin_unlock_irq(&ctx->completion_lock); + io_cqring_ev_posted(ctx); + + if (ret < 0) + req_set_fail_links(req); + io_put_req(req); return 0; } @@ -5916,18 +6040,8 @@ static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags) const struct cred *creds = NULL; int ret; - if (req->work.personality) { - const struct cred *new_creds; - - if (!(issue_flags & IO_URING_F_NONBLOCK)) - mutex_lock(&ctx->uring_lock); - new_creds = idr_find(&ctx->personality_idr, req->work.personality); - if (!(issue_flags & IO_URING_F_NONBLOCK)) - mutex_unlock(&ctx->uring_lock); - if (!new_creds) - return -EINVAL; - creds = override_creds(new_creds); - } + if (req->work.creds && req->work.creds != current_cred()) + creds = override_creds(req->work.creds); switch (req->opcode) { case IORING_OP_NOP: @@ -6147,7 +6261,6 @@ static enum hrtimer_restart io_link_timeout_fn(struct hrtimer *timer) spin_unlock_irqrestore(&ctx->completion_lock, flags); if (prev) { - req_set_fail_links(prev); io_async_find_and_cancel(ctx, req, prev->user_data, -ETIME); io_put_req_deferred(prev, 1); } else { @@ -6291,7 +6404,7 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req, { struct io_submit_state *state; unsigned int sqe_flags; - int ret = 0; + int personality, ret = 0; req->opcode = READ_ONCE(sqe->opcode); /* same numerical values with corresponding REQ_F_*, safe to copy */ @@ -6306,6 +6419,9 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req, refcount_set(&req->refs, 2); req->task = current; req->result = 0; + req->work.list.next = NULL; + req->work.creds = NULL; + req->work.flags = 0; /* enforce forwards compatibility on users */ if (unlikely(sqe_flags & ~SQE_VALID_FLAGS)) { @@ -6323,9 +6439,13 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req, !io_op_defs[req->opcode].buffer_select) return -EOPNOTSUPP; - req->work.list.next = NULL; - req->work.flags = 0; - req->work.personality = READ_ONCE(sqe->personality); + personality = READ_ONCE(sqe->personality); + if (personality) { + req->work.creds = xa_load(&ctx->personalities, personality); + if (!req->work.creds) + return -EINVAL; + get_cred(req->work.creds); + } state = &ctx->submit_state; /* @@ -6587,7 +6707,8 @@ static int __io_sq_thread(struct io_ring_ctx *ctx, bool cap_entries) if (!list_empty(&ctx->iopoll_list)) io_do_iopoll(ctx, &nr_events, 0); - if (to_submit && likely(!percpu_ref_is_dying(&ctx->refs))) + if (to_submit && likely(!percpu_ref_is_dying(&ctx->refs)) && + !(ctx->flags & IORING_SETUP_R_DISABLED)) ret = io_submit_sqes(ctx, to_submit); mutex_unlock(&ctx->uring_lock); } @@ -6611,58 +6732,6 @@ static void io_sqd_update_thread_idle(struct io_sq_data *sqd) sqd->sq_thread_idle = sq_thread_idle; } -static void io_sqd_init_new(struct io_sq_data *sqd) -{ - struct io_ring_ctx *ctx; - - while (!list_empty(&sqd->ctx_new_list)) { - ctx = list_first_entry(&sqd->ctx_new_list, struct io_ring_ctx, sqd_list); - list_move_tail(&ctx->sqd_list, &sqd->ctx_list); - complete(&ctx->sq_thread_comp); - } - - io_sqd_update_thread_idle(sqd); -} - -static bool io_sq_thread_should_stop(struct io_sq_data *sqd) -{ - return test_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state); -} - -static bool io_sq_thread_should_park(struct io_sq_data *sqd) -{ - return test_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state); -} - -static void io_sq_thread_parkme(struct io_sq_data *sqd) -{ - for (;;) { - /* - * TASK_PARKED is a special state; we must serialize against - * possible pending wakeups to avoid store-store collisions on - * task->state. - * - * Such a collision might possibly result in the task state - * changin from TASK_PARKED and us failing the - * wait_task_inactive() in kthread_park(). - */ - set_special_state(TASK_PARKED); - if (!test_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state)) - break; - - /* - * Thread is going to call schedule(), do not preempt it, - * or the caller of kthread_park() may spend more time in - * wait_task_inactive(). - */ - preempt_disable(); - complete(&sqd->parked); - schedule_preempt_disabled(); - preempt_enable(); - } - __set_current_state(TASK_RUNNING); -} - static int io_sq_thread(void *data) { struct io_sq_data *sqd = data; @@ -6681,31 +6750,32 @@ static int io_sq_thread(void *data) set_cpus_allowed_ptr(current, cpu_online_mask); current->flags |= PF_NO_SETAFFINITY; - wait_for_completion(&sqd->startup); - - while (!io_sq_thread_should_stop(sqd)) { + mutex_lock(&sqd->lock); + while (!test_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state)) { int ret; bool cap_entries, sqt_spin, needs_sched; - /* - * Any changes to the sqd lists are synchronized through the - * thread parking. This synchronizes the thread vs users, - * the users are synchronized on the sqd->ctx_lock. - */ - if (io_sq_thread_should_park(sqd)) { - io_sq_thread_parkme(sqd); - continue; - } - if (unlikely(!list_empty(&sqd->ctx_new_list))) { - io_sqd_init_new(sqd); + if (test_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state)) { + mutex_unlock(&sqd->lock); + cond_resched(); + mutex_lock(&sqd->lock); + io_run_task_work(); + io_run_task_work_head(&sqd->park_task_work); timeout = jiffies + sqd->sq_thread_idle; + continue; } if (fatal_signal_pending(current)) break; sqt_spin = false; cap_entries = !list_is_singular(&sqd->ctx_list); list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) { + const struct cred *creds = NULL; + + if (ctx->sq_creds != current_cred()) + creds = override_creds(ctx->sq_creds); ret = __io_sq_thread(ctx, cap_entries); + if (creds) + revert_creds(creds); if (!sqt_spin && (ret > 0 || !list_empty(&ctx->iopoll_list))) sqt_spin = true; } @@ -6732,45 +6802,33 @@ static int io_sq_thread(void *data) } } - if (needs_sched && !io_sq_thread_should_park(sqd)) { + if (needs_sched && !test_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state)) { list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) io_ring_set_wakeup_flag(ctx); + mutex_unlock(&sqd->lock); schedule(); try_to_freeze(); + mutex_lock(&sqd->lock); list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) io_ring_clear_wakeup_flag(ctx); } finish_wait(&sqd->wait, &wait); + io_run_task_work_head(&sqd->park_task_work); timeout = jiffies + sqd->sq_thread_idle; } list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) io_uring_cancel_sqpoll(ctx); - - io_run_task_work(); - - /* - * Ensure that we park properly if racing with someone trying to park - * while we're exiting. If we fail to grab the lock, check park and - * park if necessary. The ordering with the park bit and the lock - * ensures that we catch this reliably. - */ - if (!mutex_trylock(&sqd->lock)) { - if (io_sq_thread_should_park(sqd)) - io_sq_thread_parkme(sqd); - mutex_lock(&sqd->lock); - } - sqd->thread = NULL; - list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) { - ctx->sqo_exec = 1; + list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) io_ring_set_wakeup_flag(ctx); - } + mutex_unlock(&sqd->lock); + io_run_task_work(); + io_run_task_work_head(&sqd->park_task_work); complete(&sqd->exited); - mutex_unlock(&sqd->lock); do_exit(0); } @@ -7071,47 +7129,47 @@ static int io_sqe_files_unregister(struct io_ring_ctx *ctx) static void io_sq_thread_unpark(struct io_sq_data *sqd) __releases(&sqd->lock) { - if (sqd->thread == current) - return; + WARN_ON_ONCE(sqd->thread == current); + + /* + * Do the dance but not conditional clear_bit() because it'd race with + * other threads incrementing park_pending and setting the bit. + */ clear_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state); - if (sqd->thread) - wake_up_state(sqd->thread, TASK_PARKED); + if (atomic_dec_return(&sqd->park_pending)) + set_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state); mutex_unlock(&sqd->lock); } static void io_sq_thread_park(struct io_sq_data *sqd) __acquires(&sqd->lock) { - if (sqd->thread == current) - return; + WARN_ON_ONCE(sqd->thread == current); + + atomic_inc(&sqd->park_pending); set_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state); mutex_lock(&sqd->lock); - if (sqd->thread) { + if (sqd->thread) wake_up_process(sqd->thread); - wait_for_completion(&sqd->parked); - } } static void io_sq_thread_stop(struct io_sq_data *sqd) { - if (test_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state)) - return; + WARN_ON_ONCE(sqd->thread == current); + mutex_lock(&sqd->lock); - if (sqd->thread) { - set_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state); - WARN_ON_ONCE(test_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state)); + set_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state); + if (sqd->thread) wake_up_process(sqd->thread); - mutex_unlock(&sqd->lock); - wait_for_completion(&sqd->exited); - WARN_ON_ONCE(sqd->thread); - } else { - mutex_unlock(&sqd->lock); - } + mutex_unlock(&sqd->lock); + wait_for_completion(&sqd->exited); } static void io_put_sq_data(struct io_sq_data *sqd) { if (refcount_dec_and_test(&sqd->refs)) { + WARN_ON_ONCE(atomic_read(&sqd->park_pending)); + io_sq_thread_stop(sqd); kfree(sqd); } @@ -7122,22 +7180,15 @@ static void io_sq_thread_finish(struct io_ring_ctx *ctx) struct io_sq_data *sqd = ctx->sq_data; if (sqd) { - complete(&sqd->startup); - if (sqd->thread) { - wait_for_completion(&ctx->sq_thread_comp); - io_sq_thread_park(sqd); - } - - mutex_lock(&sqd->ctx_lock); - list_del(&ctx->sqd_list); + io_sq_thread_park(sqd); + list_del_init(&ctx->sqd_list); io_sqd_update_thread_idle(sqd); - mutex_unlock(&sqd->ctx_lock); - - if (sqd->thread) - io_sq_thread_unpark(sqd); + io_sq_thread_unpark(sqd); io_put_sq_data(sqd); ctx->sq_data = NULL; + if (ctx->sq_creds) + put_cred(ctx->sq_creds); } } @@ -7161,31 +7212,42 @@ static struct io_sq_data *io_attach_sq_data(struct io_uring_params *p) fdput(f); return ERR_PTR(-EINVAL); } + if (sqd->task_tgid != current->tgid) { + fdput(f); + return ERR_PTR(-EPERM); + } refcount_inc(&sqd->refs); fdput(f); return sqd; } -static struct io_sq_data *io_get_sq_data(struct io_uring_params *p) +static struct io_sq_data *io_get_sq_data(struct io_uring_params *p, + bool *attached) { struct io_sq_data *sqd; - if (p->flags & IORING_SETUP_ATTACH_WQ) - return io_attach_sq_data(p); + *attached = false; + if (p->flags & IORING_SETUP_ATTACH_WQ) { + sqd = io_attach_sq_data(p); + if (!IS_ERR(sqd)) { + *attached = true; + return sqd; + } + /* fall through for EPERM case, setup new sqd/task */ + if (PTR_ERR(sqd) != -EPERM) + return sqd; + } sqd = kzalloc(sizeof(*sqd), GFP_KERNEL); if (!sqd) return ERR_PTR(-ENOMEM); + atomic_set(&sqd->park_pending, 0); refcount_set(&sqd->refs, 1); INIT_LIST_HEAD(&sqd->ctx_list); - INIT_LIST_HEAD(&sqd->ctx_new_list); - mutex_init(&sqd->ctx_lock); mutex_init(&sqd->lock); init_waitqueue_head(&sqd->wait); - init_completion(&sqd->startup); - init_completion(&sqd->parked); init_completion(&sqd->exited); return sqd; } @@ -7802,7 +7864,6 @@ static int io_uring_alloc_task_context(struct task_struct *task, init_waitqueue_head(&tctx->wait); tctx->last = NULL; atomic_set(&tctx->in_idle, 0); - tctx->sqpoll = false; task->io_uring = tctx; spin_lock_init(&tctx->task_lock); INIT_WQ_LIST(&tctx->task_list); @@ -7823,26 +7884,6 @@ void __io_uring_free(struct task_struct *tsk) tsk->io_uring = NULL; } -static int io_sq_thread_fork(struct io_sq_data *sqd, struct io_ring_ctx *ctx) -{ - struct task_struct *tsk; - int ret; - - clear_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state); - reinit_completion(&sqd->parked); - ctx->sqo_exec = 0; - sqd->task_pid = current->pid; - tsk = create_io_thread(io_sq_thread, sqd, NUMA_NO_NODE); - if (IS_ERR(tsk)) - return PTR_ERR(tsk); - ret = io_uring_alloc_task_context(tsk, ctx); - if (ret) - set_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state); - sqd->thread = tsk; - wake_up_new_task(tsk); - return ret; -} - static int io_sq_offload_create(struct io_ring_ctx *ctx, struct io_uring_params *p) { @@ -7865,29 +7906,36 @@ static int io_sq_offload_create(struct io_ring_ctx *ctx, if (ctx->flags & IORING_SETUP_SQPOLL) { struct task_struct *tsk; struct io_sq_data *sqd; + bool attached; ret = -EPERM; if (!capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_NICE)) goto err; - sqd = io_get_sq_data(p); + sqd = io_get_sq_data(p, &attached); if (IS_ERR(sqd)) { ret = PTR_ERR(sqd); goto err; } + ctx->sq_creds = get_current_cred(); ctx->sq_data = sqd; - io_sq_thread_park(sqd); - mutex_lock(&sqd->ctx_lock); - list_add(&ctx->sqd_list, &sqd->ctx_new_list); - mutex_unlock(&sqd->ctx_lock); - io_sq_thread_unpark(sqd); - ctx->sq_thread_idle = msecs_to_jiffies(p->sq_thread_idle); if (!ctx->sq_thread_idle) ctx->sq_thread_idle = HZ; - if (sqd->thread) + ret = 0; + io_sq_thread_park(sqd); + list_add(&ctx->sqd_list, &sqd->ctx_list); + io_sqd_update_thread_idle(sqd); + /* don't attach to a dying SQPOLL thread, would be racy */ + if (attached && !sqd->thread) + ret = -ENXIO; + io_sq_thread_unpark(sqd); + + if (ret < 0) + goto err; + if (attached) return 0; if (p->flags & IORING_SETUP_SQ_AFF) { @@ -7895,9 +7943,9 @@ static int io_sq_offload_create(struct io_ring_ctx *ctx, ret = -EINVAL; if (cpu >= nr_cpu_ids) - goto err; + goto err_sqpoll; if (!cpu_online(cpu)) - goto err; + goto err_sqpoll; sqd->sq_cpu = cpu; } else { @@ -7905,15 +7953,15 @@ static int io_sq_offload_create(struct io_ring_ctx *ctx, } sqd->task_pid = current->pid; + sqd->task_tgid = current->tgid; tsk = create_io_thread(io_sq_thread, sqd, NUMA_NO_NODE); if (IS_ERR(tsk)) { ret = PTR_ERR(tsk); - goto err; + goto err_sqpoll; } - ret = io_uring_alloc_task_context(tsk, ctx); - if (ret) - set_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state); + sqd->thread = tsk; + ret = io_uring_alloc_task_context(tsk, ctx); wake_up_new_task(tsk); if (ret) goto err; @@ -7927,15 +7975,9 @@ static int io_sq_offload_create(struct io_ring_ctx *ctx, err: io_sq_thread_finish(ctx); return ret; -} - -static void io_sq_offload_start(struct io_ring_ctx *ctx) -{ - struct io_sq_data *sqd = ctx->sq_data; - - ctx->flags &= ~IORING_SETUP_R_DISABLED; - if (ctx->flags & IORING_SETUP_SQPOLL) - complete(&sqd->startup); +err_sqpoll: + complete(&ctx->sq_data->exited); + goto err; } static inline void __io_unaccount_mem(struct user_struct *user, @@ -8345,19 +8387,13 @@ static int io_eventfd_unregister(struct io_ring_ctx *ctx) return -ENXIO; } -static int __io_destroy_buffers(int id, void *p, void *data) -{ - struct io_ring_ctx *ctx = data; - struct io_buffer *buf = p; - - __io_remove_buffers(ctx, buf, id, -1U); - return 0; -} - static void io_destroy_buffers(struct io_ring_ctx *ctx) { - idr_for_each(&ctx->io_buffer_idr, __io_destroy_buffers, ctx); - idr_destroy(&ctx->io_buffer_idr); + struct io_buffer *buf; + unsigned long index; + + xa_for_each(&ctx->io_buffers, index, buf) + __io_remove_buffers(ctx, buf, index, -1U); } static void io_req_cache_free(struct list_head *list, struct task_struct *tsk) @@ -8399,11 +8435,13 @@ static void io_ring_ctx_free(struct io_ring_ctx *ctx) { /* * Some may use context even when all refs and requests have been put, - * and they are free to do so while still holding uring_lock, see - * __io_req_task_submit(). Wait for them to finish. + * and they are free to do so while still holding uring_lock or + * completion_lock, see __io_req_task_submit(). Wait for them to finish. */ mutex_lock(&ctx->uring_lock); mutex_unlock(&ctx->uring_lock); + spin_lock_irq(&ctx->completion_lock); + spin_unlock_irq(&ctx->completion_lock); io_sq_thread_finish(ctx); io_sqe_buffers_unregister(ctx); @@ -8418,7 +8456,6 @@ static void io_ring_ctx_free(struct io_ring_ctx *ctx) mutex_unlock(&ctx->uring_lock); io_eventfd_unregister(ctx); io_destroy_buffers(ctx); - idr_destroy(&ctx->personality_idr); #if defined(CONFIG_UNIX) if (ctx->ring_sock) { @@ -8483,7 +8520,7 @@ static int io_unregister_personality(struct io_ring_ctx *ctx, unsigned id) { const struct cred *creds; - creds = idr_remove(&ctx->personality_idr, id); + creds = xa_erase(&ctx->personalities, id); if (creds) { put_cred(creds); return 0; @@ -8492,40 +8529,39 @@ static int io_unregister_personality(struct io_ring_ctx *ctx, unsigned id) return -EINVAL; } -static int io_remove_personalities(int id, void *p, void *data) +static inline bool io_run_ctx_fallback(struct io_ring_ctx *ctx) { - struct io_ring_ctx *ctx = data; - - io_unregister_personality(ctx, id); - return 0; + return io_run_task_work_head(&ctx->exit_task_work); } -static bool io_run_ctx_fallback(struct io_ring_ctx *ctx) -{ - struct callback_head *work, *next; - bool executed = false; - - do { - work = xchg(&ctx->exit_task_work, NULL); - if (!work) - break; +struct io_tctx_exit { + struct callback_head task_work; + struct completion completion; + struct io_ring_ctx *ctx; +}; - do { - next = work->next; - work->func(work); - work = next; - cond_resched(); - } while (work); - executed = true; - } while (1); +static void io_tctx_exit_cb(struct callback_head *cb) +{ + struct io_uring_task *tctx = current->io_uring; + struct io_tctx_exit *work; - return executed; + work = container_of(cb, struct io_tctx_exit, task_work); + /* + * When @in_idle, we're in cancellation and it's racy to remove the + * node. It'll be removed by the end of cancellation, just ignore it. + */ + if (!atomic_read(&tctx->in_idle)) + io_uring_del_task_file((unsigned long)work->ctx); + complete(&work->completion); } static void io_ring_exit_work(struct work_struct *work) { - struct io_ring_ctx *ctx = container_of(work, struct io_ring_ctx, - exit_work); + struct io_ring_ctx *ctx = container_of(work, struct io_ring_ctx, exit_work); + unsigned long timeout = jiffies + HZ * 60 * 5; + struct io_tctx_exit exit; + struct io_tctx_node *node; + int ret; /* * If we're doing polled IO and end up having requests being @@ -8535,21 +8571,57 @@ static void io_ring_exit_work(struct work_struct *work) */ do { io_uring_try_cancel_requests(ctx, NULL, NULL); + + WARN_ON_ONCE(time_after(jiffies, timeout)); } while (!wait_for_completion_timeout(&ctx->ref_comp, HZ/20)); + + mutex_lock(&ctx->uring_lock); + while (!list_empty(&ctx->tctx_list)) { + WARN_ON_ONCE(time_after(jiffies, timeout)); + + node = list_first_entry(&ctx->tctx_list, struct io_tctx_node, + ctx_node); + exit.ctx = ctx; + init_completion(&exit.completion); + init_task_work(&exit.task_work, io_tctx_exit_cb); + ret = task_work_add(node->task, &exit.task_work, TWA_SIGNAL); + if (WARN_ON_ONCE(ret)) + continue; + wake_up_process(node->task); + + mutex_unlock(&ctx->uring_lock); + wait_for_completion(&exit.completion); + cond_resched(); + mutex_lock(&ctx->uring_lock); + } + mutex_unlock(&ctx->uring_lock); + io_ring_ctx_free(ctx); } static void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx) { + unsigned long index; + struct creds *creds; + mutex_lock(&ctx->uring_lock); percpu_ref_kill(&ctx->refs); /* if force is set, the ring is going away. always drop after that */ ctx->cq_overflow_flushed = 1; if (ctx->rings) __io_cqring_overflow_flush(ctx, true, NULL, NULL); - idr_for_each(&ctx->personality_idr, io_remove_personalities, ctx); + xa_for_each(&ctx->personalities, index, creds) + io_unregister_personality(ctx, index); mutex_unlock(&ctx->uring_lock); + /* prevent SQPOLL from submitting new requests */ + if (ctx->sq_data) { + io_sq_thread_park(ctx->sq_data); + list_del_init(&ctx->sqd_list); + io_sqd_update_thread_idle(ctx->sq_data); + io_sq_thread_unpark(ctx->sq_data); + } + io_kill_timeouts(ctx, NULL, NULL); io_poll_remove_all(ctx, NULL, NULL); @@ -8600,11 +8672,11 @@ static bool io_cancel_task_cb(struct io_wq_work *work, void *data) return ret; } -static void io_cancel_defer_files(struct io_ring_ctx *ctx, +static bool io_cancel_defer_files(struct io_ring_ctx *ctx, struct task_struct *task, struct files_struct *files) { - struct io_defer_entry *de = NULL; + struct io_defer_entry *de; LIST_HEAD(list); spin_lock_irq(&ctx->completion_lock); @@ -8615,6 +8687,8 @@ static void io_cancel_defer_files(struct io_ring_ctx *ctx, } } spin_unlock_irq(&ctx->completion_lock); + if (list_empty(&list)) + return false; while (!list_empty(&list)) { de = list_first_entry(&list, struct io_defer_entry, list); @@ -8624,6 +8698,38 @@ static void io_cancel_defer_files(struct io_ring_ctx *ctx, io_req_complete(de->req, -ECANCELED); kfree(de); } + return true; +} + +static bool io_cancel_ctx_cb(struct io_wq_work *work, void *data) +{ + struct io_kiocb *req = container_of(work, struct io_kiocb, work); + + return req->ctx == data; +} + +static bool io_uring_try_cancel_iowq(struct io_ring_ctx *ctx) +{ + struct io_tctx_node *node; + enum io_wq_cancel cret; + bool ret = false; + + mutex_lock(&ctx->uring_lock); + list_for_each_entry(node, &ctx->tctx_list, ctx_node) { + struct io_uring_task *tctx = node->task->io_uring; + + /* + * io_wq will stay alive while we hold uring_lock, because it's + * killed after ctx nodes, which requires to take the lock. + */ + if (!tctx || !tctx->io_wq) + continue; + cret = io_wq_cancel_cb(tctx->io_wq, io_cancel_ctx_cb, ctx, true); + ret |= (cret != IO_WQ_CANCEL_NOTFOUND); + } + mutex_unlock(&ctx->uring_lock); + + return ret; } static void io_uring_try_cancel_requests(struct io_ring_ctx *ctx, @@ -8631,27 +8737,34 @@ static void io_uring_try_cancel_requests(struct io_ring_ctx *ctx, struct files_struct *files) { struct io_task_cancel cancel = { .task = task, .files = files, }; - struct task_struct *tctx_task = task ?: current; - struct io_uring_task *tctx = tctx_task->io_uring; + struct io_uring_task *tctx = task ? task->io_uring : NULL; while (1) { enum io_wq_cancel cret; bool ret = false; - if (tctx && tctx->io_wq) { + if (!task) { + ret |= io_uring_try_cancel_iowq(ctx); + } else if (tctx && tctx->io_wq) { + /* + * Cancels requests of all rings, not only @ctx, but + * it's fine as the task is in exit/exec. + */ cret = io_wq_cancel_cb(tctx->io_wq, io_cancel_task_cb, &cancel, true); ret |= (cret != IO_WQ_CANCEL_NOTFOUND); } /* SQPOLL thread does its own polling */ - if (!(ctx->flags & IORING_SETUP_SQPOLL) && !files) { + if ((!(ctx->flags & IORING_SETUP_SQPOLL) && !files) || + (ctx->sq_data && ctx->sq_data->thread == current)) { while (!list_empty_careful(&ctx->iopoll_list)) { io_iopoll_try_reap_events(ctx); ret = true; } } + ret |= io_cancel_defer_files(ctx, task, files); ret |= io_poll_remove_all(ctx, task, files); ret |= io_kill_timeouts(ctx, task, files); ret |= io_run_task_work(); @@ -8691,58 +8804,21 @@ static void io_uring_cancel_files(struct io_ring_ctx *ctx, io_uring_try_cancel_requests(ctx, task, files); - if (ctx->sq_data) - io_sq_thread_unpark(ctx->sq_data); prepare_to_wait(&task->io_uring->wait, &wait, TASK_UNINTERRUPTIBLE); if (inflight == io_uring_count_inflight(ctx, task, files)) schedule(); finish_wait(&task->io_uring->wait, &wait); - if (ctx->sq_data) - io_sq_thread_park(ctx->sq_data); - } -} - -/* - * We need to iteratively cancel requests, in case a request has dependent - * hard links. These persist even for failure of cancelations, hence keep - * looping until none are found. - */ -static void io_uring_cancel_task_requests(struct io_ring_ctx *ctx, - struct files_struct *files) -{ - struct task_struct *task = current; - - if ((ctx->flags & IORING_SETUP_SQPOLL) && ctx->sq_data) { - /* never started, nothing to cancel */ - if (ctx->flags & IORING_SETUP_R_DISABLED) { - io_sq_offload_start(ctx); - return; - } - io_sq_thread_park(ctx->sq_data); - task = ctx->sq_data->thread; - if (task) - atomic_inc(&task->io_uring->in_idle); } - - io_cancel_defer_files(ctx, task, files); - - io_uring_cancel_files(ctx, task, files); - if (!files) - io_uring_try_cancel_requests(ctx, task, NULL); - - if (task) - atomic_dec(&task->io_uring->in_idle); - if (ctx->sq_data) - io_sq_thread_unpark(ctx->sq_data); } /* * Note that this task has used io_uring. We use it for cancelation purposes. */ -static int io_uring_add_task_file(struct io_ring_ctx *ctx, struct file *file) +static int io_uring_add_task_file(struct io_ring_ctx *ctx) { struct io_uring_task *tctx = current->io_uring; + struct io_tctx_node *node; int ret; if (unlikely(!tctx)) { @@ -8751,102 +8827,151 @@ static int io_uring_add_task_file(struct io_ring_ctx *ctx, struct file *file) return ret; tctx = current->io_uring; } - if (tctx->last != file) { - void *old = xa_load(&tctx->xa, (unsigned long)file); + if (tctx->last != ctx) { + void *old = xa_load(&tctx->xa, (unsigned long)ctx); if (!old) { - get_file(file); - ret = xa_err(xa_store(&tctx->xa, (unsigned long)file, - file, GFP_KERNEL)); + node = kmalloc(sizeof(*node), GFP_KERNEL); + if (!node) + return -ENOMEM; + node->ctx = ctx; + node->task = current; + + ret = xa_err(xa_store(&tctx->xa, (unsigned long)ctx, + node, GFP_KERNEL)); if (ret) { - fput(file); + kfree(node); return ret; } + + mutex_lock(&ctx->uring_lock); + list_add(&node->ctx_node, &ctx->tctx_list); + mutex_unlock(&ctx->uring_lock); } - tctx->last = file; + tctx->last = ctx; } - - /* - * This is race safe in that the task itself is doing this, hence it - * cannot be going through the exit/cancel paths at the same time. - * This cannot be modified while exit/cancel is running. - */ - if (!tctx->sqpoll && (ctx->flags & IORING_SETUP_SQPOLL)) - tctx->sqpoll = true; - return 0; } /* * Remove this io_uring_file -> task mapping. */ -static void io_uring_del_task_file(struct file *file) +static void io_uring_del_task_file(unsigned long index) { struct io_uring_task *tctx = current->io_uring; + struct io_tctx_node *node; + + if (!tctx) + return; + node = xa_erase(&tctx->xa, index); + if (!node) + return; + + WARN_ON_ONCE(current != node->task); + WARN_ON_ONCE(list_empty(&node->ctx_node)); + + mutex_lock(&node->ctx->uring_lock); + list_del(&node->ctx_node); + mutex_unlock(&node->ctx->uring_lock); - if (tctx->last == file) + if (tctx->last == node->ctx) tctx->last = NULL; - file = xa_erase(&tctx->xa, (unsigned long)file); - if (file) - fput(file); + kfree(node); } static void io_uring_clean_tctx(struct io_uring_task *tctx) { - struct file *file; + struct io_tctx_node *node; unsigned long index; - xa_for_each(&tctx->xa, index, file) - io_uring_del_task_file(file); + xa_for_each(&tctx->xa, index, node) + io_uring_del_task_file(index); if (tctx->io_wq) { io_wq_put_and_exit(tctx->io_wq); tctx->io_wq = NULL; } } +static s64 tctx_inflight(struct io_uring_task *tctx) +{ + return percpu_counter_sum(&tctx->inflight); +} + +static void io_sqpoll_cancel_cb(struct callback_head *cb) +{ + struct io_tctx_exit *work = container_of(cb, struct io_tctx_exit, task_work); + struct io_ring_ctx *ctx = work->ctx; + struct io_sq_data *sqd = ctx->sq_data; + + if (sqd->thread) + io_uring_cancel_sqpoll(ctx); + complete(&work->completion); +} + +static void io_sqpoll_cancel_sync(struct io_ring_ctx *ctx) +{ + struct io_sq_data *sqd = ctx->sq_data; + struct io_tctx_exit work = { .ctx = ctx, }; + struct task_struct *task; + + io_sq_thread_park(sqd); + list_del_init(&ctx->sqd_list); + io_sqd_update_thread_idle(sqd); + task = sqd->thread; + if (task) { + init_completion(&work.completion); + init_task_work(&work.task_work, io_sqpoll_cancel_cb); + io_task_work_add_head(&sqd->park_task_work, &work.task_work); + wake_up_process(task); + } + io_sq_thread_unpark(sqd); + + if (task) + wait_for_completion(&work.completion); +} + void __io_uring_files_cancel(struct files_struct *files) { struct io_uring_task *tctx = current->io_uring; - struct file *file; + struct io_tctx_node *node; unsigned long index; /* make sure overflow events are dropped */ atomic_inc(&tctx->in_idle); - xa_for_each(&tctx->xa, index, file) - io_uring_cancel_task_requests(file->private_data, files); + xa_for_each(&tctx->xa, index, node) { + struct io_ring_ctx *ctx = node->ctx; + + if (ctx->sq_data) { + io_sqpoll_cancel_sync(ctx); + continue; + } + io_uring_cancel_files(ctx, current, files); + if (!files) + io_uring_try_cancel_requests(ctx, current, NULL); + } atomic_dec(&tctx->in_idle); if (files) io_uring_clean_tctx(tctx); } -static s64 tctx_inflight(struct io_uring_task *tctx) -{ - return percpu_counter_sum(&tctx->inflight); -} - +/* should only be called by SQPOLL task */ static void io_uring_cancel_sqpoll(struct io_ring_ctx *ctx) { struct io_sq_data *sqd = ctx->sq_data; - struct io_uring_task *tctx; + struct io_uring_task *tctx = current->io_uring; s64 inflight; DEFINE_WAIT(wait); - if (!sqd) - return; - io_sq_thread_park(sqd); - if (!sqd->thread || !sqd->thread->io_uring) { - io_sq_thread_unpark(sqd); - return; - } - tctx = ctx->sq_data->thread->io_uring; + WARN_ON_ONCE(!sqd || ctx->sq_data->thread != current); + atomic_inc(&tctx->in_idle); do { /* read completions before cancelations */ inflight = tctx_inflight(tctx); if (!inflight) break; - io_uring_cancel_task_requests(ctx, NULL); + io_uring_try_cancel_requests(ctx, current, NULL); prepare_to_wait(&tctx->wait, &wait, TASK_UNINTERRUPTIBLE); /* @@ -8859,7 +8984,6 @@ static void io_uring_cancel_sqpoll(struct io_ring_ctx *ctx) finish_wait(&tctx->wait, &wait); } while (1); atomic_dec(&tctx->in_idle); - io_sq_thread_unpark(sqd); } /* @@ -8874,15 +8998,6 @@ void __io_uring_task_cancel(void) /* make sure overflow events are dropped */ atomic_inc(&tctx->in_idle); - - if (tctx->sqpoll) { - struct file *file; - unsigned long index; - - xa_for_each(&tctx->xa, index, file) - io_uring_cancel_sqpoll(file->private_data); - } - do { /* read completions before cancelations */ inflight = tctx_inflight(tctx); @@ -8981,7 +9096,6 @@ static unsigned long io_uring_nommu_get_unmapped_area(struct file *file, static int io_sqpoll_wait_sq(struct io_ring_ctx *ctx) { - int ret = 0; DEFINE_WAIT(wait); do { @@ -8995,7 +9109,7 @@ static int io_sqpoll_wait_sq(struct io_ring_ctx *ctx) } while (!signal_pending(current)); finish_wait(&ctx->sqo_sq_wait, &wait); - return ret; + return 0; } static int io_get_ext_arg(unsigned flags, const void __user *argp, size_t *argsz, @@ -9069,13 +9183,10 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit, if (ctx->flags & IORING_SETUP_SQPOLL) { io_cqring_overflow_flush(ctx, false, NULL, NULL); - if (unlikely(ctx->sqo_exec)) { - ret = io_sq_thread_fork(ctx->sq_data, ctx); - if (ret) - goto out; - ctx->sqo_exec = 0; - } ret = -EOWNERDEAD; + if (unlikely(ctx->sq_data->thread == NULL)) { + goto out; + } if (flags & IORING_ENTER_SQ_WAKEUP) wake_up(&ctx->sq_data->wait); if (flags & IORING_ENTER_SQ_WAIT) { @@ -9085,7 +9196,7 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit, } submitted = to_submit; } else if (to_submit) { - ret = io_uring_add_task_file(ctx, f.file); + ret = io_uring_add_task_file(ctx); if (unlikely(ret)) goto out; mutex_lock(&ctx->uring_lock); @@ -9127,10 +9238,9 @@ out_fput: } #ifdef CONFIG_PROC_FS -static int io_uring_show_cred(int id, void *p, void *data) +static int io_uring_show_cred(struct seq_file *m, unsigned int id, + const struct cred *cred) { - const struct cred *cred = p; - struct seq_file *m = data; struct user_namespace *uns = seq_user_ns(m); struct group_info *gi; kernel_cap_t cap; @@ -9198,9 +9308,13 @@ static void __io_uring_show_fdinfo(struct io_ring_ctx *ctx, struct seq_file *m) seq_printf(m, "%5u: 0x%llx/%u\n", i, buf->ubuf, (unsigned int) buf->len); } - if (has_lock && !idr_is_empty(&ctx->personality_idr)) { + if (has_lock && !xa_empty(&ctx->personalities)) { + unsigned long index; + const struct cred *cred; + seq_printf(m, "Personalities:\n"); - idr_for_each(&ctx->personality_idr, io_uring_show_cred, m); + xa_for_each(&ctx->personalities, index, cred) + io_uring_show_cred(m, index, cred); } seq_printf(m, "PollList:\n"); spin_lock_irq(&ctx->completion_lock); @@ -9294,7 +9408,7 @@ static int io_uring_install_fd(struct io_ring_ctx *ctx, struct file *file) if (fd < 0) return fd; - ret = io_uring_add_task_file(ctx, file); + ret = io_uring_add_task_file(ctx); if (ret) { put_unused_fd(fd); return ret; @@ -9402,9 +9516,6 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p, if (ret) goto err; - if (!(p->flags & IORING_SETUP_R_DISABLED)) - io_sq_offload_start(ctx); - memset(&p->sq_off, 0, sizeof(p->sq_off)); p->sq_off.head = offsetof(struct io_rings, sq.head); p->sq_off.tail = offsetof(struct io_rings, sq.tail); @@ -9532,14 +9643,16 @@ out: static int io_register_personality(struct io_ring_ctx *ctx) { const struct cred *creds; + u32 id; int ret; creds = get_current_cred(); - ret = idr_alloc_cyclic(&ctx->personality_idr, (void *) creds, 1, - USHRT_MAX, GFP_KERNEL); - if (ret < 0) - put_cred(creds); + ret = xa_alloc_cyclic(&ctx->personalities, &id, (void *)creds, + XA_LIMIT(0, USHRT_MAX), &ctx->pers_next, GFP_KERNEL); + if (!ret) + return id; + put_cred(creds); return ret; } @@ -9621,7 +9734,9 @@ static int io_register_enable_rings(struct io_ring_ctx *ctx) if (ctx->restrictions.registered) ctx->restricted = 1; - io_sq_offload_start(ctx); + ctx->flags &= ~IORING_SETUP_R_DISABLED; + if (ctx->sq_data && wq_has_sleeper(&ctx->sq_data->wait)) + wake_up(&ctx->sq_data->wait); return 0; } diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index 7ffcd7ef33d4..414769a6ad11 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -1221,7 +1221,7 @@ iomap_alloc_ioend(struct inode *inode, struct iomap_writepage_ctx *wpc, struct iomap_ioend *ioend; struct bio *bio; - bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_PAGES, &iomap_ioend_bioset); + bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_VECS, &iomap_ioend_bioset); bio_set_dev(bio, wpc->iomap.bdev); bio->bi_iter.bi_sector = sector; bio->bi_opf = REQ_OP_WRITE | wbc_to_write_flags(wbc); @@ -1252,7 +1252,7 @@ iomap_chain_bio(struct bio *prev) { struct bio *new; - new = bio_alloc(GFP_NOFS, BIO_MAX_PAGES); + new = bio_alloc(GFP_NOFS, BIO_MAX_VECS); bio_copy_dev(new, prev);/* also copies over blkcg information */ new->bi_iter.bi_sector = bio_end_sector(prev); new->bi_opf = prev->bi_opf; diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c index e2c4991833b8..bdd0d89bbf0a 100644 --- a/fs/iomap/direct-io.c +++ b/fs/iomap/direct-io.c @@ -296,7 +296,7 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length, */ bio_opf = iomap_dio_bio_opflags(dio, iomap, use_fua); - nr_pages = bio_iov_vecs_to_alloc(dio->submit.iter, BIO_MAX_PAGES); + nr_pages = bio_iov_vecs_to_alloc(dio->submit.iter, BIO_MAX_VECS); do { size_t n; if (dio->error) { @@ -338,7 +338,7 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length, copied += n; nr_pages = bio_iov_vecs_to_alloc(dio->submit.iter, - BIO_MAX_PAGES); + BIO_MAX_VECS); iomap_dio_submit_bio(dio, iomap, bio, pos); pos += n; } while (nr_pages); diff --git a/fs/iomap/swapfile.c b/fs/iomap/swapfile.c index a648dbf6991e..a5e478de1417 100644 --- a/fs/iomap/swapfile.c +++ b/fs/iomap/swapfile.c @@ -170,6 +170,16 @@ int iomap_swapfile_activate(struct swap_info_struct *sis, return ret; } + /* + * If this swapfile doesn't contain even a single page-aligned + * contiguous range of blocks, reject this useless swapfile to + * prevent confusion later on. + */ + if (isi.nr_pages == 0) { + pr_warn("swapon: Cannot find a single usable page in file.\n"); + return -EINVAL; + } + *pagespan = 1 + isi.highest_ppage - isi.lowest_ppage; sis->max = isi.nr_pages; sis->pages = isi.nr_pages - 1; diff --git a/fs/locks.c b/fs/locks.c index 99ca97e81b7a..6125d2de39b8 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -1808,9 +1808,6 @@ check_conflicting_open(struct file *filp, const long arg, int flags) if (flags & FL_LAYOUT) return 0; - if (flags & FL_DELEG) - /* We leave these checks to the caller. */ - return 0; if (arg == F_RDLCK) return inode_is_open_for_write(inode) ? -EAGAIN : 0; diff --git a/fs/mpage.c b/fs/mpage.c index 961234d68779..334e7d09aa65 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -616,7 +616,7 @@ alloc_new: goto out; } bio = mpage_alloc(bdev, blocks[0] << (blkbits - 9), - BIO_MAX_PAGES, GFP_NOFS|__GFP_HIGH); + BIO_MAX_VECS, GFP_NOFS|__GFP_HIGH); if (bio == NULL) goto confused; diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig index e2a488d403a6..14a72224b657 100644 --- a/fs/nfs/Kconfig +++ b/fs/nfs/Kconfig @@ -127,7 +127,7 @@ config PNFS_BLOCK config PNFS_FLEXFILE_LAYOUT tristate depends on NFS_V4_1 && NFS_V3 - default m + default NFS_V4 config NFS_V4_1_IMPLEMENTATION_ID_DOMAIN string "NFSv4.1 Implementation ID Domain" diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 19a9f434442f..fc4f490f2d78 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -81,8 +81,9 @@ static struct nfs_open_dir_context *alloc_nfs_open_dir_context(struct inode *dir spin_lock(&dir->i_lock); if (list_empty(&nfsi->open_files) && (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER)) - nfsi->cache_validity |= NFS_INO_INVALID_DATA | - NFS_INO_REVAL_FORCED; + nfs_set_cache_invalid(dir, + NFS_INO_INVALID_DATA | + NFS_INO_REVAL_FORCED); list_add(&ctx->list, &nfsi->open_files); spin_unlock(&dir->i_lock); return ctx; @@ -1401,6 +1402,13 @@ out_force: goto out; } +static void nfs_mark_dir_for_revalidate(struct inode *inode) +{ + spin_lock(&inode->i_lock); + nfs_set_cache_invalid(inode, NFS_INO_REVAL_PAGECACHE); + spin_unlock(&inode->i_lock); +} + /* * We judge how long we want to trust negative * dentries by looking at the parent inode mtime. @@ -1435,19 +1443,14 @@ nfs_lookup_revalidate_done(struct inode *dir, struct dentry *dentry, __func__, dentry); return 1; case 0: - nfs_mark_for_revalidate(dir); - if (inode && S_ISDIR(inode->i_mode)) { - /* Purge readdir caches. */ - nfs_zap_caches(inode); - /* - * We can't d_drop the root of a disconnected tree: - * its d_hash is on the s_anon list and d_drop() would hide - * it from shrink_dcache_for_unmount(), leading to busy - * inodes on unmount and further oopses. - */ - if (IS_ROOT(dentry)) - return 1; - } + /* + * We can't d_drop the root of a disconnected tree: + * its d_hash is on the s_anon list and d_drop() would hide + * it from shrink_dcache_for_unmount(), leading to busy + * inodes on unmount and further oopses. + */ + if (inode && IS_ROOT(dentry)) + return 1; dfprintk(LOOKUPCACHE, "NFS: %s(%pd2) is invalid\n", __func__, dentry); return 0; @@ -1525,6 +1528,13 @@ out: nfs_free_fattr(fattr); nfs_free_fhandle(fhandle); nfs4_label_free(label); + + /* + * If the lookup failed despite the dentry change attribute being + * a match, then we should revalidate the directory cache. + */ + if (!ret && nfs_verify_change_attribute(dir, dentry->d_time)) + nfs_mark_dir_for_revalidate(dir); return nfs_lookup_revalidate_done(dir, dentry, inode, ret); } @@ -1567,7 +1577,7 @@ nfs_do_lookup_revalidate(struct inode *dir, struct dentry *dentry, error = nfs_lookup_verify_inode(inode, flags); if (error) { if (error == -ESTALE) - nfs_zap_caches(dir); + nfs_mark_dir_for_revalidate(dir); goto out_bad; } nfs_advise_use_readdirplus(dir); @@ -1691,10 +1701,9 @@ static void nfs_drop_nlink(struct inode *inode) if (inode->i_nlink > 0) drop_nlink(inode); NFS_I(inode)->attr_gencount = nfs_inc_attr_generation_counter(); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_CHANGE - | NFS_INO_INVALID_CTIME - | NFS_INO_INVALID_OTHER - | NFS_INO_REVAL_FORCED; + nfs_set_cache_invalid( + inode, NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_CTIME | + NFS_INO_INVALID_OTHER | NFS_INO_REVAL_FORCED); spin_unlock(&inode->i_lock); } @@ -1706,7 +1715,7 @@ static void nfs_dentry_iput(struct dentry *dentry, struct inode *inode) { if (S_ISDIR(inode->i_mode)) /* drop any readdir cache as it could easily be old */ - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_DATA; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA); if (dentry->d_flags & DCACHE_NFSFS_RENAMED) { nfs_complete_unlink(dentry, inode); @@ -2064,7 +2073,6 @@ out: dput(parent); return d; out_error: - nfs_mark_for_revalidate(dir); d = ERR_PTR(error); goto out; } @@ -2473,9 +2481,9 @@ int nfs_rename(struct user_namespace *mnt_userns, struct inode *old_dir, if (error == 0) { spin_lock(&old_inode->i_lock); NFS_I(old_inode)->attr_gencount = nfs_inc_attr_generation_counter(); - NFS_I(old_inode)->cache_validity |= NFS_INO_INVALID_CHANGE - | NFS_INO_INVALID_CTIME - | NFS_INO_REVAL_FORCED; + nfs_set_cache_invalid(old_inode, NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME | + NFS_INO_REVAL_FORCED); spin_unlock(&old_inode->i_lock); } out: diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 749bbea14d99..a7fb076a5f44 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -207,7 +207,7 @@ static bool nfs_has_xattr_cache(const struct nfs_inode *nfsi) } #endif -static void nfs_set_cache_invalid(struct inode *inode, unsigned long flags) +void nfs_set_cache_invalid(struct inode *inode, unsigned long flags) { struct nfs_inode *nfsi = NFS_I(inode); bool have_delegation = NFS_PROTO(inode)->have_delegation(inode, FMODE_READ); @@ -229,6 +229,7 @@ static void nfs_set_cache_invalid(struct inode *inode, unsigned long flags) if (flags & NFS_INO_INVALID_DATA) nfs_fscache_invalidate(inode); } +EXPORT_SYMBOL_GPL(nfs_set_cache_invalid); /* * Invalidate the local caches @@ -1067,8 +1068,8 @@ void nfs_inode_attach_open_context(struct nfs_open_context *ctx) spin_lock(&inode->i_lock); if (list_empty(&nfsi->open_files) && (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER)) - nfsi->cache_validity |= NFS_INO_INVALID_DATA | - NFS_INO_REVAL_FORCED; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA | + NFS_INO_REVAL_FORCED); list_add_tail_rcu(&ctx->list, &nfsi->open_files); spin_unlock(&inode->i_lock); } diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 25fb43b69e5a..7b644d6c09e4 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -411,7 +411,8 @@ extern int nfs_write_inode(struct inode *, struct writeback_control *); extern int nfs_drop_inode(struct inode *); extern void nfs_clear_inode(struct inode *); extern void nfs_evict_inode(struct inode *); -void nfs_zap_acl_cache(struct inode *inode); +extern void nfs_zap_acl_cache(struct inode *inode); +extern void nfs_set_cache_invalid(struct inode *inode, unsigned long flags); extern bool nfs_check_cache_invalid(struct inode *, unsigned long); extern int nfs_wait_bit_killable(struct wait_bit_key *key, int mode); extern int nfs_wait_atomic_killable(atomic_t *p, unsigned int mode); diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index ca10072644ff..ed1c83738c30 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -36,6 +36,7 @@ #define NFS3_pagepad_sz (1) /* Page padding */ #define NFS3_fhandle_sz (1+16) #define NFS3_fh_sz (NFS3_fhandle_sz) /* shorthand */ +#define NFS3_post_op_fh_sz (1+NFS3_fh_sz) #define NFS3_sattr_sz (15) #define NFS3_filename_sz (1+(NFS3_MAXNAMLEN>>2)) #define NFS3_path_sz (1+(NFS3_MAXPATHLEN>>2)) @@ -73,7 +74,7 @@ #define NFS3_readlinkres_sz (1+NFS3_post_op_attr_sz+1+NFS3_pagepad_sz) #define NFS3_readres_sz (1+NFS3_post_op_attr_sz+3+NFS3_pagepad_sz) #define NFS3_writeres_sz (1+NFS3_wcc_data_sz+4) -#define NFS3_createres_sz (1+NFS3_fh_sz+NFS3_post_op_attr_sz+NFS3_wcc_data_sz) +#define NFS3_createres_sz (1+NFS3_post_op_fh_sz+NFS3_post_op_attr_sz+NFS3_wcc_data_sz) #define NFS3_renameres_sz (1+(2 * NFS3_wcc_data_sz)) #define NFS3_linkres_sz (1+NFS3_post_op_attr_sz+NFS3_wcc_data_sz) #define NFS3_readdirres_sz (1+NFS3_post_op_attr_sz+2+NFS3_pagepad_sz) diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c index f3fd935620fc..094024b0aca1 100644 --- a/fs/nfs/nfs42proc.c +++ b/fs/nfs/nfs42proc.c @@ -357,13 +357,15 @@ static ssize_t _nfs42_proc_copy(struct file *src, truncate_pagecache_range(dst_inode, pos_dst, pos_dst + res->write_res.count); spin_lock(&dst_inode->i_lock); - NFS_I(dst_inode)->cache_validity |= (NFS_INO_REVAL_PAGECACHE | - NFS_INO_REVAL_FORCED | NFS_INO_INVALID_SIZE | - NFS_INO_INVALID_ATTR | NFS_INO_INVALID_DATA); + nfs_set_cache_invalid( + dst_inode, NFS_INO_REVAL_PAGECACHE | NFS_INO_REVAL_FORCED | + NFS_INO_INVALID_SIZE | NFS_INO_INVALID_ATTR | + NFS_INO_INVALID_DATA); spin_unlock(&dst_inode->i_lock); spin_lock(&src_inode->i_lock); - NFS_I(src_inode)->cache_validity |= (NFS_INO_REVAL_PAGECACHE | - NFS_INO_REVAL_FORCED | NFS_INO_INVALID_ATIME); + nfs_set_cache_invalid(src_inode, NFS_INO_REVAL_PAGECACHE | + NFS_INO_REVAL_FORCED | + NFS_INO_INVALID_ATIME); spin_unlock(&src_inode->i_lock); status = res->write_res.count; out: diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 74bc5120013d..c65c4b41e2c1 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1169,14 +1169,14 @@ int nfs4_call_sync(struct rpc_clnt *clnt, static void nfs4_inc_nlink_locked(struct inode *inode) { - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_OTHER; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER); inc_nlink(inode); } static void nfs4_dec_nlink_locked(struct inode *inode) { - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_OTHER; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER); drop_nlink(inode); } @@ -1187,35 +1187,31 @@ nfs4_update_changeattr_locked(struct inode *inode, { struct nfs_inode *nfsi = NFS_I(inode); - nfsi->cache_validity |= NFS_INO_INVALID_CTIME - | NFS_INO_INVALID_MTIME - | cache_validity; + cache_validity |= NFS_INO_INVALID_CTIME | NFS_INO_INVALID_MTIME; if (cinfo->atomic && cinfo->before == inode_peek_iversion_raw(inode)) { nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE; nfsi->attrtimeo_timestamp = jiffies; } else { if (S_ISDIR(inode->i_mode)) { - nfsi->cache_validity |= NFS_INO_INVALID_DATA; + cache_validity |= NFS_INO_INVALID_DATA; nfs_force_lookup_revalidate(inode); } else { if (!NFS_PROTO(inode)->have_delegation(inode, FMODE_READ)) - nfsi->cache_validity |= NFS_INO_REVAL_PAGECACHE; + cache_validity |= NFS_INO_REVAL_PAGECACHE; } if (cinfo->before != inode_peek_iversion_raw(inode)) - nfsi->cache_validity |= NFS_INO_INVALID_ACCESS | - NFS_INO_INVALID_ACL | - NFS_INO_INVALID_XATTR; + cache_validity |= NFS_INO_INVALID_ACCESS | + NFS_INO_INVALID_ACL | + NFS_INO_INVALID_XATTR; } inode_set_iversion_raw(inode, cinfo->after); nfsi->read_cache_jiffies = timestamp; nfsi->attr_gencount = nfs_inc_attr_generation_counter(); + nfs_set_cache_invalid(inode, cache_validity); nfsi->cache_validity &= ~NFS_INO_INVALID_CHANGE; - - if (nfsi->cache_validity & NFS_INO_INVALID_DATA) - nfs_fscache_invalidate(inode); } void @@ -5893,6 +5889,9 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl unsigned int npages = DIV_ROUND_UP(buflen, PAGE_SIZE); int ret, i; + /* You can't remove system.nfs4_acl: */ + if (buflen == 0) + return -EINVAL; if (!nfs4_server_supports_acls(server)) return -EOPNOTSUPP; if (npages > ARRAY_SIZE(pages)) @@ -5915,9 +5914,9 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl * so mark the attribute cache invalid. */ spin_lock(&inode->i_lock); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_CHANGE - | NFS_INO_INVALID_CTIME - | NFS_INO_REVAL_FORCED; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME | + NFS_INO_REVAL_FORCED); spin_unlock(&inode->i_lock); nfs_access_zap_cache(inode); nfs_zap_acl_cache(inode); @@ -5969,7 +5968,7 @@ static int _nfs4_get_security_label(struct inode *inode, void *buf, return ret; if (!(fattr.valid & NFS_ATTR_FATTR_V4_SECURITY_LABEL)) return -ENOENT; - return 0; + return label.len; } static int nfs4_get_security_label(struct inode *inode, void *buf, diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index b27ebdccef70..5fa11e1aca4c 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -500,9 +500,9 @@ nfs_sillyrename(struct inode *dir, struct dentry *dentry) nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); spin_lock(&inode->i_lock); NFS_I(inode)->attr_gencount = nfs_inc_attr_generation_counter(); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_CHANGE - | NFS_INO_INVALID_CTIME - | NFS_INO_REVAL_FORCED; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME | + NFS_INO_REVAL_FORCED); spin_unlock(&inode->i_lock); d_move(dentry, sdentry); break; diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 82bdcb982186..f05a90338a76 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -303,9 +303,9 @@ static void nfs_set_pageerror(struct address_space *mapping) nfs_zap_mapping(mapping->host, mapping); /* Force file size revalidation */ spin_lock(&inode->i_lock); - NFS_I(inode)->cache_validity |= NFS_INO_REVAL_FORCED | - NFS_INO_REVAL_PAGECACHE | - NFS_INO_INVALID_SIZE; + nfs_set_cache_invalid(inode, NFS_INO_REVAL_FORCED | + NFS_INO_REVAL_PAGECACHE | + NFS_INO_INVALID_SIZE); spin_unlock(&inode->i_lock); } @@ -1604,7 +1604,7 @@ static int nfs_writeback_done(struct rpc_task *task, /* Deal with the suid/sgid bit corner case */ if (nfs_should_remove_suid(inode)) { spin_lock(&inode->i_lock); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_OTHER; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER); spin_unlock(&inode->i_lock); } return 0; diff --git a/fs/nfsd/Kconfig b/fs/nfsd/Kconfig index 821e5913faee..d6cff5fbe705 100644 --- a/fs/nfsd/Kconfig +++ b/fs/nfsd/Kconfig @@ -73,6 +73,7 @@ config NFSD_V4 select NFSD_V3 select FS_POSIX_ACL select SUNRPC_GSS + select CRYPTO select CRYPTO_MD5 select CRYPTO_SHA256 select GRACE_PERIOD diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c index 53fcbf79bdca..7629248fdd53 100644 --- a/fs/nfsd/filecache.c +++ b/fs/nfsd/filecache.c @@ -898,6 +898,8 @@ nfsd_file_find_locked(struct inode *inode, unsigned int may_flags, continue; if (!nfsd_match_cred(nf->nf_cred, current_cred())) continue; + if (!test_bit(NFSD_FILE_HASHED, &nf->nf_flags)) + continue; if (nfsd_file_get(nf) != NULL) return nf; } diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 052be5bf9ef5..7325592b456e 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -1189,6 +1189,7 @@ static void nfsd4_cb_done(struct rpc_task *task, void *calldata) switch (task->tk_status) { case -EIO: case -ETIMEDOUT: + case -EACCES: nfsd4_mark_cb_down(clp, task->tk_status); } break; diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index acdb3cd806a1..dd9f38d072dd 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -1302,7 +1302,7 @@ nfsd4_cleanup_inter_ssc(struct vfsmount *ss_mnt, struct nfsd_file *src, struct nfsd_file *dst) { nfs42_ssc_close(src->nf_file); - /* 'src' is freed by nfsd4_do_async_copy */ + fput(src->nf_file); nfsd_file_put(dst); mntput(ss_mnt); } diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 423fd6683f3a..97447a64bad0 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -4940,31 +4940,6 @@ static struct file_lock *nfs4_alloc_init_lease(struct nfs4_delegation *dp, return fl; } -static int nfsd4_check_conflicting_opens(struct nfs4_client *clp, - struct nfs4_file *fp) -{ - struct nfs4_clnt_odstate *co; - struct file *f = fp->fi_deleg_file->nf_file; - struct inode *ino = locks_inode(f); - int writes = atomic_read(&ino->i_writecount); - - if (fp->fi_fds[O_WRONLY]) - writes--; - if (fp->fi_fds[O_RDWR]) - writes--; - if (writes > 0) - return -EAGAIN; - spin_lock(&fp->fi_lock); - list_for_each_entry(co, &fp->fi_clnt_odstate, co_perfile) { - if (co->co_client != clp) { - spin_unlock(&fp->fi_lock); - return -EAGAIN; - } - } - spin_unlock(&fp->fi_lock); - return 0; -} - static struct nfs4_delegation * nfs4_set_delegation(struct nfs4_client *clp, struct svc_fh *fh, struct nfs4_file *fp, struct nfs4_clnt_odstate *odstate) @@ -4984,12 +4959,9 @@ nfs4_set_delegation(struct nfs4_client *clp, struct svc_fh *fh, nf = find_readable_file(fp); if (!nf) { - /* - * We probably could attempt another open and get a read - * delegation, but for now, don't bother until the - * client actually sends us one. - */ - return ERR_PTR(-EAGAIN); + /* We should always have a readable file here */ + WARN_ON_ONCE(1); + return ERR_PTR(-EBADF); } spin_lock(&state_lock); spin_lock(&fp->fi_lock); @@ -5019,19 +4991,11 @@ nfs4_set_delegation(struct nfs4_client *clp, struct svc_fh *fh, if (!fl) goto out_clnt_odstate; - status = nfsd4_check_conflicting_opens(clp, fp); - if (status) { - locks_free_lock(fl); - goto out_clnt_odstate; - } status = vfs_setlease(fp->fi_deleg_file->nf_file, fl->fl_type, &fl, NULL); if (fl) locks_free_lock(fl); if (status) goto out_clnt_odstate; - status = nfsd4_check_conflicting_opens(clp, fp); - if (status) - goto out_clnt_odstate; spin_lock(&state_lock); spin_lock(&fp->fi_lock); @@ -5113,6 +5077,17 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, goto out_no_deleg; if (!cb_up || !(oo->oo_flags & NFS4_OO_CONFIRMED)) goto out_no_deleg; + /* + * Also, if the file was opened for write or + * create, there's a good chance the client's + * about to write to it, resulting in an + * immediate recall (since we don't support + * write delegations): + */ + if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE) + goto out_no_deleg; + if (open->op_create == NFS4_OPEN_CREATE) + goto out_no_deleg; break; default: goto out_no_deleg; @@ -5389,7 +5364,7 @@ nfs4_laundromat(struct nfsd_net *nn) idr_for_each_entry(&nn->s2s_cp_stateids, cps_t, i) { cps = container_of(cps_t, struct nfs4_cpntf_state, cp_stateid); if (cps->cp_stateid.sc_type == NFS4_COPYNOTIFY_STID && - cps->cpntf_time > cutoff) + cps->cpntf_time < cutoff) _free_cpntf_state_locked(nn, cps); } spin_unlock(&nn->s2s_cp_lock); diff --git a/fs/nilfs2/segbuf.c b/fs/nilfs2/segbuf.c index 1e75417bfe6e..56872e93823d 100644 --- a/fs/nilfs2/segbuf.c +++ b/fs/nilfs2/segbuf.c @@ -399,7 +399,7 @@ static void nilfs_segbuf_prepare_write(struct nilfs_segment_buffer *segbuf, { wi->bio = NULL; wi->rest_blocks = segbuf->sb_sum.nblocks; - wi->max_pages = BIO_MAX_PAGES; + wi->max_pages = BIO_MAX_VECS; wi->nr_vecs = min(wi->max_pages, wi->rest_blocks); wi->start = wi->end = 0; wi->blocknr = segbuf->sb_pseg_start; diff --git a/fs/pnode.h b/fs/pnode.h index 26f74e092bd9..988f1aa9b02a 100644 --- a/fs/pnode.h +++ b/fs/pnode.h @@ -12,7 +12,7 @@ #define IS_MNT_SHARED(m) ((m)->mnt.mnt_flags & MNT_SHARED) #define IS_MNT_SLAVE(m) ((m)->mnt_master) -#define IS_MNT_NEW(m) (!(m)->mnt_ns) +#define IS_MNT_NEW(m) (!(m)->mnt_ns || is_anon_ns((m)->mnt_ns)) #define CLEAR_MNT_SHARED(m) ((m)->mnt.mnt_flags &= ~MNT_SHARED) #define IS_MNT_UNBINDABLE(m) ((m)->mnt.mnt_flags & MNT_UNBINDABLE) #define IS_MNT_MARKED(m) ((m)->mnt.mnt_flags & MNT_MARKED) diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 3cec6fbef725..e862cab69583 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -1036,8 +1036,6 @@ struct clear_refs_private { #ifdef CONFIG_MEM_SOFT_DIRTY -#define is_cow_mapping(flags) (((flags) & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE) - static inline bool pte_is_pinned(struct vm_area_struct *vma, unsigned long addr, pte_t pte) { struct page *page; diff --git a/fs/select.c b/fs/select.c index 37aaa8317f3a..945896d0ac9e 100644 --- a/fs/select.c +++ b/fs/select.c @@ -1055,10 +1055,9 @@ static long do_restart_poll(struct restart_block *restart_block) ret = do_sys_poll(ufds, nfds, to); - if (ret == -ERESTARTNOHAND) { - restart_block->fn = do_restart_poll; - ret = -ERESTART_RESTARTBLOCK; - } + if (ret == -ERESTARTNOHAND) + ret = set_restart_fn(restart_block, do_restart_poll); + return ret; } @@ -1080,7 +1079,6 @@ SYSCALL_DEFINE3(poll, struct pollfd __user *, ufds, unsigned int, nfds, struct restart_block *restart_block; restart_block = ¤t->restart_block; - restart_block->fn = do_restart_poll; restart_block->poll.ufds = ufds; restart_block->poll.nfds = nfds; @@ -1091,7 +1089,7 @@ SYSCALL_DEFINE3(poll, struct pollfd __user *, ufds, unsigned int, nfds, } else restart_block->poll.has_timeout = 0; - ret = -ERESTART_RESTARTBLOCK; + ret = set_restart_fn(restart_block, do_restart_poll); } return ret; } diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c index 45f44425d856..b9e87ebb1060 100644 --- a/fs/squashfs/block.c +++ b/fs/squashfs/block.c @@ -87,7 +87,7 @@ static int squashfs_bio_read(struct super_block *sb, u64 index, int length, int error, i; struct bio *bio; - if (page_count <= BIO_MAX_PAGES) + if (page_count <= BIO_MAX_VECS) bio = bio_alloc(GFP_NOIO, page_count); else bio = bio_kmalloc(GFP_NOIO, page_count); diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 46a861d55e48..f93370bd7b1e 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1007,9 +1007,10 @@ xfs_create( /* * Make sure that we have allocated dquot(s) on disk. */ - error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid, - XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, - &udqp, &gdqp, &pdqp); + error = xfs_qm_vop_dqalloc(dp, fsuid_into_mnt(mnt_userns), + fsgid_into_mnt(mnt_userns), prid, + XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, + &udqp, &gdqp, &pdqp); if (error) return error; @@ -1157,9 +1158,10 @@ xfs_create_tmpfile( /* * Make sure that we have allocated dquot(s) on disk. */ - error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid, - XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, - &udqp, &gdqp, &pdqp); + error = xfs_qm_vop_dqalloc(dp, fsuid_into_mnt(mnt_userns), + fsgid_into_mnt(mnt_userns), prid, + XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, + &udqp, &gdqp, &pdqp); if (error) return error; diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index ca310a125d1e..3498b97fb06d 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -168,6 +168,12 @@ xfs_bulkstat_one( }; int error; + if (breq->mnt_userns != &init_user_ns) { + xfs_warn_ratelimited(breq->mp, + "bulkstat not supported inside of idmapped mounts."); + return -EINVAL; + } + ASSERT(breq->icount == 1); bc.buf = kmem_zalloc(sizeof(struct xfs_bulkstat), diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 52370d0a3f43..1c97b155a8ee 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -635,6 +635,47 @@ xfs_check_summary_counts( } /* + * Flush and reclaim dirty inodes in preparation for unmount. Inodes and + * internal inode structures can be sitting in the CIL and AIL at this point, + * so we need to unpin them, write them back and/or reclaim them before unmount + * can proceed. + * + * An inode cluster that has been freed can have its buffer still pinned in + * memory because the transaction is still sitting in a iclog. The stale inodes + * on that buffer will be pinned to the buffer until the transaction hits the + * disk and the callbacks run. Pushing the AIL will skip the stale inodes and + * may never see the pinned buffer, so nothing will push out the iclog and + * unpin the buffer. + * + * Hence we need to force the log to unpin everything first. However, log + * forces don't wait for the discards they issue to complete, so we have to + * explicitly wait for them to complete here as well. + * + * Then we can tell the world we are unmounting so that error handling knows + * that the filesystem is going away and we should error out anything that we + * have been retrying in the background. This will prevent never-ending + * retries in AIL pushing from hanging the unmount. + * + * Finally, we can push the AIL to clean all the remaining dirty objects, then + * reclaim the remaining inodes that are still in memory at this point in time. + */ +static void +xfs_unmount_flush_inodes( + struct xfs_mount *mp) +{ + xfs_log_force(mp, XFS_LOG_SYNC); + xfs_extent_busy_wait_all(mp); + flush_workqueue(xfs_discard_wq); + + mp->m_flags |= XFS_MOUNT_UNMOUNTING; + + xfs_ail_push_all_sync(mp->m_ail); + cancel_delayed_work_sync(&mp->m_reclaim_work); + xfs_reclaim_inodes(mp); + xfs_health_unmount(mp); +} + +/* * This function does the following on an initial mount of a file system: * - reads the superblock from disk and init the mount struct * - if we're a 32-bit kernel, do a size check on the superblock @@ -1008,7 +1049,7 @@ xfs_mountfs( /* Clean out dquots that might be in memory after quotacheck. */ xfs_qm_unmount(mp); /* - * Cancel all delayed reclaim work and reclaim the inodes directly. + * Flush all inode reclamation work and flush the log. * We have to do this /after/ rtunmount and qm_unmount because those * two will have scheduled delayed reclaim for the rt/quota inodes. * @@ -1018,11 +1059,8 @@ xfs_mountfs( * qm_unmount_quotas and therefore rely on qm_unmount to release the * quota inodes. */ - cancel_delayed_work_sync(&mp->m_reclaim_work); - xfs_reclaim_inodes(mp); - xfs_health_unmount(mp); + xfs_unmount_flush_inodes(mp); out_log_dealloc: - mp->m_flags |= XFS_MOUNT_UNMOUNTING; xfs_log_mount_cancel(mp); out_fail_wait: if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) @@ -1063,47 +1101,7 @@ xfs_unmountfs( xfs_rtunmount_inodes(mp); xfs_irele(mp->m_rootip); - /* - * We can potentially deadlock here if we have an inode cluster - * that has been freed has its buffer still pinned in memory because - * the transaction is still sitting in a iclog. The stale inodes - * on that buffer will be pinned to the buffer until the - * transaction hits the disk and the callbacks run. Pushing the AIL will - * skip the stale inodes and may never see the pinned buffer, so - * nothing will push out the iclog and unpin the buffer. Hence we - * need to force the log here to ensure all items are flushed into the - * AIL before we go any further. - */ - xfs_log_force(mp, XFS_LOG_SYNC); - - /* - * Wait for all busy extents to be freed, including completion of - * any discard operation. - */ - xfs_extent_busy_wait_all(mp); - flush_workqueue(xfs_discard_wq); - - /* - * We now need to tell the world we are unmounting. This will allow - * us to detect that the filesystem is going away and we should error - * out anything that we have been retrying in the background. This will - * prevent neverending retries in AIL pushing from hanging the unmount. - */ - mp->m_flags |= XFS_MOUNT_UNMOUNTING; - - /* - * Flush all pending changes from the AIL. - */ - xfs_ail_push_all_sync(mp->m_ail); - - /* - * Reclaim all inodes. At this point there should be no dirty inodes and - * none should be pinned or locked. Stop background inode reclaim here - * if it is still running. - */ - cancel_delayed_work_sync(&mp->m_reclaim_work); - xfs_reclaim_inodes(mp); - xfs_health_unmount(mp); + xfs_unmount_flush_inodes(mp); xfs_qm_unmount(mp); diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c index 1379013d74b8..7f368b10ded1 100644 --- a/fs/xfs/xfs_symlink.c +++ b/fs/xfs/xfs_symlink.c @@ -182,7 +182,8 @@ xfs_symlink( /* * Make sure that we have allocated dquot(s) on disk. */ - error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid, + error = xfs_qm_vop_dqalloc(dp, fsuid_into_mnt(mnt_userns), + fsgid_into_mnt(mnt_userns), prid, XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp, &pdqp); if (error) diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c index b6ff4a21abac..049e36c69ed7 100644 --- a/fs/zonefs/super.c +++ b/fs/zonefs/super.c @@ -165,6 +165,21 @@ static int zonefs_writepages(struct address_space *mapping, return iomap_writepages(mapping, wbc, &wpc, &zonefs_writeback_ops); } +static int zonefs_swap_activate(struct swap_info_struct *sis, + struct file *swap_file, sector_t *span) +{ + struct inode *inode = file_inode(swap_file); + struct zonefs_inode_info *zi = ZONEFS_I(inode); + + if (zi->i_ztype != ZONEFS_ZTYPE_CNV) { + zonefs_err(inode->i_sb, + "swap file: not a conventional zone file\n"); + return -EINVAL; + } + + return iomap_swapfile_activate(sis, swap_file, span, &zonefs_iomap_ops); +} + static const struct address_space_operations zonefs_file_aops = { .readpage = zonefs_readpage, .readahead = zonefs_readahead, @@ -177,6 +192,7 @@ static const struct address_space_operations zonefs_file_aops = { .is_partially_uptodate = iomap_is_partially_uptodate, .error_remove_page = generic_error_remove_page, .direct_IO = noop_direct_IO, + .swap_activate = zonefs_swap_activate, }; static void zonefs_update_stats(struct inode *inode, loff_t new_isize) @@ -684,7 +700,7 @@ static ssize_t zonefs_file_dio_append(struct kiocb *iocb, struct iov_iter *from) max = ALIGN_DOWN(max << SECTOR_SHIFT, inode->i_sb->s_blocksize); iov_iter_truncate(from, max); - nr_pages = iov_iter_npages(from, BIO_MAX_PAGES); + nr_pages = iov_iter_npages(from, BIO_MAX_VECS); if (!nr_pages) return 0; @@ -728,6 +744,68 @@ out_release: } /* + * Do not exceed the LFS limits nor the file zone size. If pos is under the + * limit it becomes a short access. If it exceeds the limit, return -EFBIG. + */ +static loff_t zonefs_write_check_limits(struct file *file, loff_t pos, + loff_t count) +{ + struct inode *inode = file_inode(file); + struct zonefs_inode_info *zi = ZONEFS_I(inode); + loff_t limit = rlimit(RLIMIT_FSIZE); + loff_t max_size = zi->i_max_size; + + if (limit != RLIM_INFINITY) { + if (pos >= limit) { + send_sig(SIGXFSZ, current, 0); + return -EFBIG; + } + count = min(count, limit - pos); + } + + if (!(file->f_flags & O_LARGEFILE)) + max_size = min_t(loff_t, MAX_NON_LFS, max_size); + + if (unlikely(pos >= max_size)) + return -EFBIG; + + return min(count, max_size - pos); +} + +static ssize_t zonefs_write_checks(struct kiocb *iocb, struct iov_iter *from) +{ + struct file *file = iocb->ki_filp; + struct inode *inode = file_inode(file); + struct zonefs_inode_info *zi = ZONEFS_I(inode); + loff_t count; + + if (IS_SWAPFILE(inode)) + return -ETXTBSY; + + if (!iov_iter_count(from)) + return 0; + + if ((iocb->ki_flags & IOCB_NOWAIT) && !(iocb->ki_flags & IOCB_DIRECT)) + return -EINVAL; + + if (iocb->ki_flags & IOCB_APPEND) { + if (zi->i_ztype != ZONEFS_ZTYPE_SEQ) + return -EINVAL; + mutex_lock(&zi->i_truncate_mutex); + iocb->ki_pos = zi->i_wpoffset; + mutex_unlock(&zi->i_truncate_mutex); + } + + count = zonefs_write_check_limits(file, iocb->ki_pos, + iov_iter_count(from)); + if (count < 0) + return count; + + iov_iter_truncate(from, count); + return iov_iter_count(from); +} + +/* * Handle direct writes. For sequential zone files, this is the only possible * write path. For these files, check that the user is issuing writes * sequentially from the end of the file. This code assumes that the block layer @@ -744,8 +822,7 @@ static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from) struct super_block *sb = inode->i_sb; bool sync = is_sync_kiocb(iocb); bool append = false; - size_t count; - ssize_t ret; + ssize_t ret, count; /* * For async direct IOs to sequential zone files, refuse IOCB_NOWAIT @@ -763,12 +840,11 @@ static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from) inode_lock(inode); } - ret = generic_write_checks(iocb, from); - if (ret <= 0) + count = zonefs_write_checks(iocb, from); + if (count <= 0) { + ret = count; goto inode_unlock; - - iov_iter_truncate(from, zi->i_max_size - iocb->ki_pos); - count = iov_iter_count(from); + } if ((iocb->ki_pos | count) & (sb->s_blocksize - 1)) { ret = -EINVAL; @@ -828,12 +904,10 @@ static ssize_t zonefs_file_buffered_write(struct kiocb *iocb, inode_lock(inode); } - ret = generic_write_checks(iocb, from); + ret = zonefs_write_checks(iocb, from); if (ret <= 0) goto inode_unlock; - iov_iter_truncate(from, zi->i_max_size - iocb->ki_pos); - ret = iomap_file_buffered_write(iocb, from, &zonefs_iomap_ops); if (ret > 0) iocb->ki_pos += ret; @@ -966,9 +1040,7 @@ static int zonefs_open_zone(struct inode *inode) mutex_lock(&zi->i_truncate_mutex); - zi->i_wr_refcnt++; - if (zi->i_wr_refcnt == 1) { - + if (!zi->i_wr_refcnt) { if (atomic_inc_return(&sbi->s_open_zones) > sbi->s_max_open_zones) { atomic_dec(&sbi->s_open_zones); ret = -EBUSY; @@ -978,7 +1050,6 @@ static int zonefs_open_zone(struct inode *inode) if (i_size_read(inode) < zi->i_max_size) { ret = zonefs_zone_mgmt(inode, REQ_OP_ZONE_OPEN); if (ret) { - zi->i_wr_refcnt--; atomic_dec(&sbi->s_open_zones); goto unlock; } @@ -986,6 +1057,8 @@ static int zonefs_open_zone(struct inode *inode) } } + zi->i_wr_refcnt++; + unlock: mutex_unlock(&zi->i_truncate_mutex); |