diff options
Diffstat (limited to 'fs')
123 files changed, 4519 insertions, 1639 deletions
diff --git a/fs/afs/file.c b/fs/afs/file.c index 7a1d942ef68d..0149dab365e7 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -102,6 +102,7 @@ int afs_release(struct inode *inode, struct file *file) return 0; } +#ifdef CONFIG_AFS_FSCACHE /* * deal with notification that a page was read from the cache */ @@ -117,6 +118,7 @@ static void afs_file_readpage_read_complete(struct page *page, SetPageUptodate(page); unlock_page(page); } +#endif /* * AFS read page from file, directory or symlink diff --git a/fs/afs/netdevices.c b/fs/afs/netdevices.c index 49f189423063..7ad36506c256 100644 --- a/fs/afs/netdevices.c +++ b/fs/afs/netdevices.c @@ -20,8 +20,7 @@ int afs_get_MAC_address(u8 *mac, size_t maclen) struct net_device *dev; int ret = -ENODEV; - if (maclen != ETH_ALEN) - BUG(); + BUG_ON(maclen != ETH_ALEN); rtnl_lock(); dev = __dev_getfirstbyhwtype(&init_net, ARPHRD_ETHER); diff --git a/fs/autofs/dirhash.c b/fs/autofs/dirhash.c index bf8c8af98004..4eb4d8dfb2f1 100644 --- a/fs/autofs/dirhash.c +++ b/fs/autofs/dirhash.c @@ -39,10 +39,12 @@ struct autofs_dir_ent *autofs_expire(struct super_block *sb, { struct autofs_dirhash *dh = &sbi->dirhash; struct autofs_dir_ent *ent; - struct dentry *dentry; unsigned long timeout = sbi->exp_timeout; while (1) { + struct path path; + int umount_ok; + if ( list_empty(&dh->expiry_head) || sbi->catatonic ) return NULL; /* No entries */ /* We keep the list sorted by last_usage and want old stuff */ @@ -57,17 +59,17 @@ struct autofs_dir_ent *autofs_expire(struct super_block *sb, return ent; /* Symlinks are always expirable */ /* Get the dentry for the autofs subdirectory */ - dentry = ent->dentry; + path.dentry = ent->dentry; - if ( !dentry ) { + if (!path.dentry) { /* Should only happen in catatonic mode */ printk("autofs: dentry == NULL but inode range is directory, entry %s\n", ent->name); autofs_delete_usage(ent); continue; } - if ( !dentry->d_inode ) { - dput(dentry); + if (!path.dentry->d_inode) { + dput(path.dentry); printk("autofs: negative dentry on expiry queue: %s\n", ent->name); autofs_delete_usage(ent); @@ -76,29 +78,29 @@ struct autofs_dir_ent *autofs_expire(struct super_block *sb, /* Make sure entry is mounted and unused; note that dentry will point to the mounted-on-top root. */ - if (!S_ISDIR(dentry->d_inode->i_mode)||!d_mountpoint(dentry)) { + if (!S_ISDIR(path.dentry->d_inode->i_mode) || + !d_mountpoint(path.dentry)) { DPRINTK(("autofs: not expirable (not a mounted directory): %s\n", ent->name)); continue; } - mntget(mnt); - dget(dentry); - if (!follow_down(&mnt, &dentry)) { - dput(dentry); - mntput(mnt); + path.mnt = mnt; + path_get(&path); + if (!follow_down(&path.mnt, &path.dentry)) { + path_put(&path); DPRINTK(("autofs: not expirable (not a mounted directory): %s\n", ent->name)); continue; } - while (d_mountpoint(dentry) && follow_down(&mnt, &dentry)) + while (d_mountpoint(path.dentry) && + follow_down(&path.mnt, &path.dentry)) ; - dput(dentry); + umount_ok = may_umount(path.mnt); + path_put(&path); - if ( may_umount(mnt) ) { - mntput(mnt); + if (umount_ok) { DPRINTK(("autofs: signaling expire on %s\n", ent->name)); return ent; /* Expirable! */ } DPRINTK(("autofs: didn't expire due to may_umount: %s\n", ent->name)); - mntput(mnt); } return NULL; /* No expirable entries */ } diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c index 9e5ae8a4f5c8..84168c0dcc2d 100644 --- a/fs/autofs4/dev-ioctl.c +++ b/fs/autofs4/dev-ioctl.c @@ -54,11 +54,10 @@ static int check_name(const char *name) * Check a string doesn't overrun the chunk of * memory we copied from user land. */ -static int invalid_str(char *str, void *end) +static int invalid_str(char *str, size_t size) { - while ((void *) str <= end) - if (!*str++) - return 0; + if (memchr(str, 0, size)) + return 0; return -EINVAL; } @@ -138,8 +137,7 @@ static int validate_dev_ioctl(int cmd, struct autofs_dev_ioctl *param) } if (param->size > sizeof(*param)) { - err = invalid_str(param->path, - (void *) ((size_t) param + param->size)); + err = invalid_str(param->path, param->size - sizeof(*param)); if (err) { AUTOFS_WARN( "path string terminator missing for cmd(0x%08x)", @@ -488,7 +486,7 @@ static int autofs_dev_ioctl_requester(struct file *fp, } path = param->path; - devid = sbi->sb->s_dev; + devid = new_encode_dev(sbi->sb->s_dev); param->requester.uid = param->requester.gid = -1; diff --git a/fs/befs/super.c b/fs/befs/super.c index 41f2b4d0093e..ca40f828f64d 100644 --- a/fs/befs/super.c +++ b/fs/befs/super.c @@ -8,6 +8,7 @@ */ #include <linux/fs.h> +#include <asm/page.h> /* for PAGE_SIZE */ #include "befs.h" #include "super.h" @@ -348,6 +348,24 @@ err: return NULL; } +/** + * bio_alloc - allocate a bio for I/O + * @gfp_mask: the GFP_ mask given to the slab allocator + * @nr_iovecs: number of iovecs to pre-allocate + * + * Description: + * bio_alloc will allocate a bio and associated bio_vec array that can hold + * at least @nr_iovecs entries. Allocations will be done from the + * fs_bio_set. Also see @bio_alloc_bioset. + * + * If %__GFP_WAIT is set, then bio_alloc will always be able to allocate + * a bio. This is due to the mempool guarantees. To make this work, callers + * must never allocate more than 1 bio at the time from this pool. Callers + * that need to allocate more than 1 bio must always submit the previously + * allocate bio for IO before attempting to allocate a new one. Failure to + * do so can cause livelocks under memory pressure. + * + **/ struct bio *bio_alloc(gfp_t gfp_mask, int nr_iovecs) { struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, fs_bio_set); diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c index 51bfdfc8fcda..502c3d61de62 100644 --- a/fs/btrfs/async-thread.c +++ b/fs/btrfs/async-thread.c @@ -25,6 +25,7 @@ #define WORK_QUEUED_BIT 0 #define WORK_DONE_BIT 1 #define WORK_ORDER_DONE_BIT 2 +#define WORK_HIGH_PRIO_BIT 3 /* * container for the kthread task pointer and the list of pending work @@ -36,6 +37,7 @@ struct btrfs_worker_thread { /* list of struct btrfs_work that are waiting for service */ struct list_head pending; + struct list_head prio_pending; /* list of worker threads from struct btrfs_workers */ struct list_head worker_list; @@ -103,10 +105,16 @@ static noinline int run_ordered_completions(struct btrfs_workers *workers, spin_lock_irqsave(&workers->lock, flags); - while (!list_empty(&workers->order_list)) { - work = list_entry(workers->order_list.next, - struct btrfs_work, order_list); - + while (1) { + if (!list_empty(&workers->prio_order_list)) { + work = list_entry(workers->prio_order_list.next, + struct btrfs_work, order_list); + } else if (!list_empty(&workers->order_list)) { + work = list_entry(workers->order_list.next, + struct btrfs_work, order_list); + } else { + break; + } if (!test_bit(WORK_DONE_BIT, &work->flags)) break; @@ -143,8 +151,14 @@ static int worker_loop(void *arg) do { spin_lock_irq(&worker->lock); again_locked: - while (!list_empty(&worker->pending)) { - cur = worker->pending.next; + while (1) { + if (!list_empty(&worker->prio_pending)) + cur = worker->prio_pending.next; + else if (!list_empty(&worker->pending)) + cur = worker->pending.next; + else + break; + work = list_entry(cur, struct btrfs_work, list); list_del(&work->list); clear_bit(WORK_QUEUED_BIT, &work->flags); @@ -163,7 +177,6 @@ again_locked: spin_lock_irq(&worker->lock); check_idle_worker(worker); - } if (freezing(current)) { worker->working = 0; @@ -178,7 +191,8 @@ again_locked: * jump_in? */ smp_mb(); - if (!list_empty(&worker->pending)) + if (!list_empty(&worker->pending) || + !list_empty(&worker->prio_pending)) continue; /* @@ -191,7 +205,8 @@ again_locked: */ schedule_timeout(1); smp_mb(); - if (!list_empty(&worker->pending)) + if (!list_empty(&worker->pending) || + !list_empty(&worker->prio_pending)) continue; if (kthread_should_stop()) @@ -200,7 +215,8 @@ again_locked: /* still no more work?, sleep for real */ spin_lock_irq(&worker->lock); set_current_state(TASK_INTERRUPTIBLE); - if (!list_empty(&worker->pending)) + if (!list_empty(&worker->pending) || + !list_empty(&worker->prio_pending)) goto again_locked; /* @@ -248,6 +264,7 @@ void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max) INIT_LIST_HEAD(&workers->worker_list); INIT_LIST_HEAD(&workers->idle_list); INIT_LIST_HEAD(&workers->order_list); + INIT_LIST_HEAD(&workers->prio_order_list); spin_lock_init(&workers->lock); workers->max_workers = max; workers->idle_thresh = 32; @@ -273,6 +290,7 @@ int btrfs_start_workers(struct btrfs_workers *workers, int num_workers) } INIT_LIST_HEAD(&worker->pending); + INIT_LIST_HEAD(&worker->prio_pending); INIT_LIST_HEAD(&worker->worker_list); spin_lock_init(&worker->lock); atomic_set(&worker->num_pending, 0); @@ -396,7 +414,10 @@ int btrfs_requeue_work(struct btrfs_work *work) goto out; spin_lock_irqsave(&worker->lock, flags); - list_add_tail(&work->list, &worker->pending); + if (test_bit(WORK_HIGH_PRIO_BIT, &work->flags)) + list_add_tail(&work->list, &worker->prio_pending); + else + list_add_tail(&work->list, &worker->pending); atomic_inc(&worker->num_pending); /* by definition we're busy, take ourselves off the idle @@ -422,6 +443,11 @@ out: return 0; } +void btrfs_set_work_high_prio(struct btrfs_work *work) +{ + set_bit(WORK_HIGH_PRIO_BIT, &work->flags); +} + /* * places a struct btrfs_work into the pending queue of one of the kthreads */ @@ -438,7 +464,12 @@ int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work) worker = find_worker(workers); if (workers->ordered) { spin_lock_irqsave(&workers->lock, flags); - list_add_tail(&work->order_list, &workers->order_list); + if (test_bit(WORK_HIGH_PRIO_BIT, &work->flags)) { + list_add_tail(&work->order_list, + &workers->prio_order_list); + } else { + list_add_tail(&work->order_list, &workers->order_list); + } spin_unlock_irqrestore(&workers->lock, flags); } else { INIT_LIST_HEAD(&work->order_list); @@ -446,7 +477,10 @@ int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work) spin_lock_irqsave(&worker->lock, flags); - list_add_tail(&work->list, &worker->pending); + if (test_bit(WORK_HIGH_PRIO_BIT, &work->flags)) + list_add_tail(&work->list, &worker->prio_pending); + else + list_add_tail(&work->list, &worker->pending); atomic_inc(&worker->num_pending); check_busy_worker(worker); diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h index 31be4ed8b63e..1b511c109db6 100644 --- a/fs/btrfs/async-thread.h +++ b/fs/btrfs/async-thread.h @@ -85,6 +85,7 @@ struct btrfs_workers { * of work items waiting for completion */ struct list_head order_list; + struct list_head prio_order_list; /* lock for finding the next worker thread to queue on */ spinlock_t lock; @@ -98,4 +99,5 @@ int btrfs_start_workers(struct btrfs_workers *workers, int num_workers); int btrfs_stop_workers(struct btrfs_workers *workers); void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max); int btrfs_requeue_work(struct btrfs_work *work); +void btrfs_set_work_high_prio(struct btrfs_work *work); #endif diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index e5b2533b691a..a99f1c2a710d 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -1325,12 +1325,12 @@ static noinline int reada_for_balance(struct btrfs_root *root, int ret = 0; int blocksize; - parent = path->nodes[level - 1]; + parent = path->nodes[level + 1]; if (!parent) return 0; nritems = btrfs_header_nritems(parent); - slot = path->slots[level]; + slot = path->slots[level + 1]; blocksize = btrfs_level_size(root, level); if (slot > 0) { @@ -1341,7 +1341,7 @@ static noinline int reada_for_balance(struct btrfs_root *root, block1 = 0; free_extent_buffer(eb); } - if (slot < nritems) { + if (slot + 1 < nritems) { block2 = btrfs_node_blockptr(parent, slot + 1); gen = btrfs_node_ptr_generation(parent, slot + 1); eb = btrfs_find_tree_block(root, block2, blocksize); @@ -1351,7 +1351,11 @@ static noinline int reada_for_balance(struct btrfs_root *root, } if (block1 || block2) { ret = -EAGAIN; + + /* release the whole path */ btrfs_release_path(root, path); + + /* read the blocks */ if (block1) readahead_tree_block(root, block1, blocksize, 0); if (block2) @@ -1361,7 +1365,7 @@ static noinline int reada_for_balance(struct btrfs_root *root, eb = read_tree_block(root, block1, blocksize, 0); free_extent_buffer(eb); } - if (block1) { + if (block2) { eb = read_tree_block(root, block2, blocksize, 0); free_extent_buffer(eb); } @@ -1481,12 +1485,15 @@ read_block_for_search(struct btrfs_trans_handle *trans, * of the btree by dropping locks before * we read. */ - btrfs_release_path(NULL, p); + btrfs_unlock_up_safe(p, level + 1); + btrfs_set_path_blocking(p); + if (tmp) free_extent_buffer(tmp); if (p->reada) reada_for_search(root, p, level, slot, key->objectid); + btrfs_release_path(NULL, p); tmp = read_tree_block(root, blocknr, blocksize, gen); if (tmp) free_extent_buffer(tmp); diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 92caa8035f36..a6b83744b05d 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -579,6 +579,10 @@ int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode, async->bio_flags = bio_flags; atomic_inc(&fs_info->nr_async_submits); + + if (rw & (1 << BIO_RW_SYNCIO)) + btrfs_set_work_high_prio(&async->work); + btrfs_queue_worker(&fs_info->workers, &async->work); #if 0 int limit = btrfs_async_submit_limit(fs_info); @@ -656,6 +660,7 @@ static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio, return btrfs_map_bio(BTRFS_I(inode)->root, rw, bio, mirror_num, 0); } + /* * kthread helpers are used to submit writes so that checksumming * can happen in parallel across all CPUs @@ -2095,10 +2100,10 @@ static int write_dev_supers(struct btrfs_device *device, device->barriers = 0; get_bh(bh); lock_buffer(bh); - ret = submit_bh(WRITE, bh); + ret = submit_bh(WRITE_SYNC, bh); } } else { - ret = submit_bh(WRITE, bh); + ret = submit_bh(WRITE_SYNC, bh); } if (!ret && wait) { diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index eb2bee8b7fbf..05a1c42e25bf 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -50,7 +50,10 @@ struct extent_page_data { /* tells writepage not to lock the state bits for this range * it still does the unlocking */ - int extent_locked; + unsigned int extent_locked:1; + + /* tells the submit_bio code to use a WRITE_SYNC */ + unsigned int sync_io:1; }; int __init extent_io_init(void) @@ -2101,6 +2104,16 @@ int extent_read_full_page(struct extent_io_tree *tree, struct page *page, return ret; } +static noinline void update_nr_written(struct page *page, + struct writeback_control *wbc, + unsigned long nr_written) +{ + wbc->nr_to_write -= nr_written; + if (wbc->range_cyclic || (wbc->nr_to_write > 0 && + wbc->range_start == 0 && wbc->range_end == LLONG_MAX)) + page->mapping->writeback_index = page->index + nr_written; +} + /* * the writepage semantics are similar to regular writepage. extent * records are inserted to lock ranges in the tree, and as dirty areas @@ -2136,8 +2149,14 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, u64 delalloc_end; int page_started; int compressed; + int write_flags; unsigned long nr_written = 0; + if (wbc->sync_mode == WB_SYNC_ALL) + write_flags = WRITE_SYNC_PLUG; + else + write_flags = WRITE; + WARN_ON(!PageLocked(page)); pg_offset = i_size & (PAGE_CACHE_SIZE - 1); if (page->index > end_index || @@ -2164,6 +2183,12 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, delalloc_end = 0; page_started = 0; if (!epd->extent_locked) { + /* + * make sure the wbc mapping index is at least updated + * to this page. + */ + update_nr_written(page, wbc, 0); + while (delalloc_end < page_end) { nr_delalloc = find_lock_delalloc_range(inode, tree, page, @@ -2185,7 +2210,13 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, */ if (page_started) { ret = 0; - goto update_nr_written; + /* + * we've unlocked the page, so we can't update + * the mapping's writeback index, just update + * nr_to_write. + */ + wbc->nr_to_write -= nr_written; + goto done_unlocked; } } lock_extent(tree, start, page_end, GFP_NOFS); @@ -2198,13 +2229,18 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, if (ret == -EAGAIN) { unlock_extent(tree, start, page_end, GFP_NOFS); redirty_page_for_writepage(wbc, page); + update_nr_written(page, wbc, nr_written); unlock_page(page); ret = 0; - goto update_nr_written; + goto done_unlocked; } } - nr_written++; + /* + * we don't want to touch the inode after unlocking the page, + * so we update the mapping writeback index now + */ + update_nr_written(page, wbc, nr_written + 1); end = page_end; if (test_range_bit(tree, start, page_end, EXTENT_DELALLOC, 0)) @@ -2314,9 +2350,9 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, (unsigned long long)end); } - ret = submit_extent_page(WRITE, tree, page, sector, - iosize, pg_offset, bdev, - &epd->bio, max_nr, + ret = submit_extent_page(write_flags, tree, page, + sector, iosize, pg_offset, + bdev, &epd->bio, max_nr, end_bio_extent_writepage, 0, 0, 0); if (ret) @@ -2336,11 +2372,8 @@ done: unlock_extent(tree, unlock_start, page_end, GFP_NOFS); unlock_page(page); -update_nr_written: - wbc->nr_to_write -= nr_written; - if (wbc->range_cyclic || (wbc->nr_to_write > 0 && - wbc->range_start == 0 && wbc->range_end == LLONG_MAX)) - page->mapping->writeback_index = page->index + nr_written; +done_unlocked: + return 0; } @@ -2460,15 +2493,23 @@ retry: return ret; } -static noinline void flush_write_bio(void *data) +static void flush_epd_write_bio(struct extent_page_data *epd) { - struct extent_page_data *epd = data; if (epd->bio) { - submit_one_bio(WRITE, epd->bio, 0, 0); + if (epd->sync_io) + submit_one_bio(WRITE_SYNC, epd->bio, 0, 0); + else + submit_one_bio(WRITE, epd->bio, 0, 0); epd->bio = NULL; } } +static noinline void flush_write_bio(void *data) +{ + struct extent_page_data *epd = data; + flush_epd_write_bio(epd); +} + int extent_write_full_page(struct extent_io_tree *tree, struct page *page, get_extent_t *get_extent, struct writeback_control *wbc) @@ -2480,23 +2521,22 @@ int extent_write_full_page(struct extent_io_tree *tree, struct page *page, .tree = tree, .get_extent = get_extent, .extent_locked = 0, + .sync_io = wbc->sync_mode == WB_SYNC_ALL, }; struct writeback_control wbc_writepages = { .bdi = wbc->bdi, - .sync_mode = WB_SYNC_NONE, + .sync_mode = wbc->sync_mode, .older_than_this = NULL, .nr_to_write = 64, .range_start = page_offset(page) + PAGE_CACHE_SIZE, .range_end = (loff_t)-1, }; - ret = __extent_writepage(page, wbc, &epd); extent_write_cache_pages(tree, mapping, &wbc_writepages, __extent_writepage, &epd, flush_write_bio); - if (epd.bio) - submit_one_bio(WRITE, epd.bio, 0, 0); + flush_epd_write_bio(&epd); return ret; } @@ -2515,6 +2555,7 @@ int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode, .tree = tree, .get_extent = get_extent, .extent_locked = 1, + .sync_io = mode == WB_SYNC_ALL, }; struct writeback_control wbc_writepages = { .bdi = inode->i_mapping->backing_dev_info, @@ -2540,8 +2581,7 @@ int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode, start += PAGE_CACHE_SIZE; } - if (epd.bio) - submit_one_bio(WRITE, epd.bio, 0, 0); + flush_epd_write_bio(&epd); return ret; } @@ -2556,13 +2596,13 @@ int extent_writepages(struct extent_io_tree *tree, .tree = tree, .get_extent = get_extent, .extent_locked = 0, + .sync_io = wbc->sync_mode == WB_SYNC_ALL, }; ret = extent_write_cache_pages(tree, mapping, wbc, __extent_writepage, &epd, flush_write_bio); - if (epd.bio) - submit_one_bio(WRITE, epd.bio, 0, 0); + flush_epd_write_bio(&epd); return ret; } diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 9c9fb46ccd08..482f8db2cfd0 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -830,7 +830,7 @@ again: ret = btrfs_del_items(trans, root, path, del_slot, del_nr); BUG_ON(ret); - goto done; + goto release; } else if (split == start) { if (locked_end < extent_end) { ret = try_lock_extent(&BTRFS_I(inode)->io_tree, @@ -926,6 +926,8 @@ again: } done: btrfs_mark_buffer_dirty(leaf); + +release: btrfs_release_path(root, path); if (split_end && split == start) { split = end; @@ -1131,7 +1133,7 @@ static ssize_t btrfs_file_write(struct file *file, const char __user *buf, if (will_write) { btrfs_fdatawrite_range(inode->i_mapping, pos, pos + write_bytes - 1, - WB_SYNC_NONE); + WB_SYNC_ALL); } else { balance_dirty_pages_ratelimited_nr(inode->i_mapping, num_pages); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index a0d1dd492a58..65219f6a16a1 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4970,10 +4970,10 @@ out_fail: return err; } -static int prealloc_file_range(struct inode *inode, u64 start, u64 end, +static int prealloc_file_range(struct btrfs_trans_handle *trans, + struct inode *inode, u64 start, u64 end, u64 alloc_hint, int mode) { - struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_key ins; u64 alloc_size; @@ -4981,10 +4981,6 @@ static int prealloc_file_range(struct inode *inode, u64 start, u64 end, u64 num_bytes = end - start; int ret = 0; - trans = btrfs_join_transaction(root, 1); - BUG_ON(!trans); - btrfs_set_trans_block_group(trans, inode); - while (num_bytes > 0) { alloc_size = min(num_bytes, root->fs_info->max_extent); ret = btrfs_reserve_extent(trans, root, alloc_size, @@ -5015,7 +5011,6 @@ out: BUG_ON(ret); } - btrfs_end_transaction(trans, root); return ret; } @@ -5029,11 +5024,18 @@ static long btrfs_fallocate(struct inode *inode, int mode, u64 alloc_hint = 0; u64 mask = BTRFS_I(inode)->root->sectorsize - 1; struct extent_map *em; + struct btrfs_trans_handle *trans; int ret; alloc_start = offset & ~mask; alloc_end = (offset + len + mask) & ~mask; + /* + * wait for ordered IO before we have any locks. We'll loop again + * below with the locks held. + */ + btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start); + mutex_lock(&inode->i_mutex); if (alloc_start > inode->i_size) { ret = btrfs_cont_expand(inode, alloc_start); @@ -5043,6 +5045,16 @@ static long btrfs_fallocate(struct inode *inode, int mode, while (1) { struct btrfs_ordered_extent *ordered; + + trans = btrfs_start_transaction(BTRFS_I(inode)->root, 1); + if (!trans) { + ret = -EIO; + goto out; + } + + /* the extent lock is ordered inside the running + * transaction + */ lock_extent(&BTRFS_I(inode)->io_tree, alloc_start, alloc_end - 1, GFP_NOFS); ordered = btrfs_lookup_first_ordered_extent(inode, @@ -5053,6 +5065,12 @@ static long btrfs_fallocate(struct inode *inode, int mode, btrfs_put_ordered_extent(ordered); unlock_extent(&BTRFS_I(inode)->io_tree, alloc_start, alloc_end - 1, GFP_NOFS); + btrfs_end_transaction(trans, BTRFS_I(inode)->root); + + /* + * we can't wait on the range with the transaction + * running or with the extent lock held + */ btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start); } else { @@ -5070,7 +5088,7 @@ static long btrfs_fallocate(struct inode *inode, int mode, last_byte = min(extent_map_end(em), alloc_end); last_byte = (last_byte + mask) & ~mask; if (em->block_start == EXTENT_MAP_HOLE) { - ret = prealloc_file_range(inode, cur_offset, + ret = prealloc_file_range(trans, inode, cur_offset, last_byte, alloc_hint, mode); if (ret < 0) { free_extent_map(em); @@ -5089,6 +5107,8 @@ static long btrfs_fallocate(struct inode *inode, int mode, } unlock_extent(&BTRFS_I(inode)->io_tree, alloc_start, alloc_end - 1, GFP_NOFS); + + btrfs_end_transaction(trans, BTRFS_I(inode)->root); out: mutex_unlock(&inode->i_mutex); return ret; diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 7594bec1be10..9f135e878507 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -461,15 +461,9 @@ static int btrfs_ioctl_resize(struct btrfs_root *root, void __user *arg) if (!capable(CAP_SYS_ADMIN)) return -EPERM; - vol_args = kmalloc(sizeof(*vol_args), GFP_NOFS); - - if (!vol_args) - return -ENOMEM; - - if (copy_from_user(vol_args, arg, sizeof(*vol_args))) { - ret = -EFAULT; - goto out; - } + vol_args = memdup_user(arg, sizeof(*vol_args)); + if (IS_ERR(vol_args)) + return PTR_ERR(vol_args); vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; namelen = strlen(vol_args->name); @@ -545,7 +539,6 @@ static int btrfs_ioctl_resize(struct btrfs_root *root, void __user *arg) out_unlock: mutex_unlock(&root->fs_info->volume_mutex); -out: kfree(vol_args); return ret; } @@ -565,15 +558,9 @@ static noinline int btrfs_ioctl_snap_create(struct file *file, if (root->fs_info->sb->s_flags & MS_RDONLY) return -EROFS; - vol_args = kmalloc(sizeof(*vol_args), GFP_NOFS); - - if (!vol_args) - return -ENOMEM; - - if (copy_from_user(vol_args, arg, sizeof(*vol_args))) { - ret = -EFAULT; - goto out; - } + vol_args = memdup_user(arg, sizeof(*vol_args)); + if (IS_ERR(vol_args)) + return PTR_ERR(vol_args); vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; namelen = strlen(vol_args->name); @@ -675,19 +662,13 @@ static long btrfs_ioctl_add_dev(struct btrfs_root *root, void __user *arg) if (!capable(CAP_SYS_ADMIN)) return -EPERM; - vol_args = kmalloc(sizeof(*vol_args), GFP_NOFS); + vol_args = memdup_user(arg, sizeof(*vol_args)); + if (IS_ERR(vol_args)) + return PTR_ERR(vol_args); - if (!vol_args) - return -ENOMEM; - - if (copy_from_user(vol_args, arg, sizeof(*vol_args))) { - ret = -EFAULT; - goto out; - } vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; ret = btrfs_init_new_device(root, vol_args->name); -out: kfree(vol_args); return ret; } @@ -703,19 +684,13 @@ static long btrfs_ioctl_rm_dev(struct btrfs_root *root, void __user *arg) if (root->fs_info->sb->s_flags & MS_RDONLY) return -EROFS; - vol_args = kmalloc(sizeof(*vol_args), GFP_NOFS); + vol_args = memdup_user(arg, sizeof(*vol_args)); + if (IS_ERR(vol_args)) + return PTR_ERR(vol_args); - if (!vol_args) - return -ENOMEM; - - if (copy_from_user(vol_args, arg, sizeof(*vol_args))) { - ret = -EFAULT; - goto out; - } vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; ret = btrfs_rm_device(root, vol_args->name); -out: kfree(vol_args); return ret; } diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index 53c87b197d70..d6f0806c682f 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -489,7 +489,7 @@ again: /* start IO across the range first to instantiate any delalloc * extents */ - btrfs_fdatawrite_range(inode->i_mapping, start, orig_end, WB_SYNC_NONE); + btrfs_fdatawrite_range(inode->i_mapping, start, orig_end, WB_SYNC_ALL); /* The compression code will leave pages locked but return from * writepage without setting the page writeback. Starting again diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 9744af9d71e9..a7acfe639a44 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -635,14 +635,9 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd, if (!capable(CAP_SYS_ADMIN)) return -EPERM; - vol = kmalloc(sizeof(*vol), GFP_KERNEL); - if (!vol) - return -ENOMEM; - - if (copy_from_user(vol, (void __user *)arg, sizeof(*vol))) { - ret = -EFAULT; - goto out; - } + vol = memdup_user((void __user *)arg, sizeof(*vol)); + if (IS_ERR(vol)) + return PTR_ERR(vol); switch (cmd) { case BTRFS_IOC_SCAN_DEV: @@ -650,7 +645,7 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd, &btrfs_fs_type, &fs_devices); break; } -out: + kfree(vol); return ret; } diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index e0913e469728..e53835b88594 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -125,6 +125,20 @@ static noinline struct btrfs_fs_devices *find_fsid(u8 *fsid) return NULL; } +static void requeue_list(struct btrfs_pending_bios *pending_bios, + struct bio *head, struct bio *tail) +{ + + struct bio *old_head; + + old_head = pending_bios->head; + pending_bios->head = head; + if (pending_bios->tail) + tail->bi_next = old_head; + else + pending_bios->tail = tail; +} + /* * we try to collect pending bios for a device so we don't get a large * number of procs sending bios down to the same device. This greatly @@ -141,10 +155,12 @@ static noinline int run_scheduled_bios(struct btrfs_device *device) struct bio *pending; struct backing_dev_info *bdi; struct btrfs_fs_info *fs_info; + struct btrfs_pending_bios *pending_bios; struct bio *tail; struct bio *cur; int again = 0; - unsigned long num_run = 0; + unsigned long num_run; + unsigned long num_sync_run; unsigned long limit; unsigned long last_waited = 0; @@ -153,20 +169,30 @@ static noinline int run_scheduled_bios(struct btrfs_device *device) limit = btrfs_async_submit_limit(fs_info); limit = limit * 2 / 3; + /* we want to make sure that every time we switch from the sync + * list to the normal list, we unplug + */ + num_sync_run = 0; + loop: spin_lock(&device->io_lock); + num_run = 0; loop_lock: + /* take all the bios off the list at once and process them * later on (without the lock held). But, remember the * tail and other pointers so the bios can be properly reinserted * into the list if we hit congestion */ - pending = device->pending_bios; - tail = device->pending_bio_tail; + if (device->pending_sync_bios.head) + pending_bios = &device->pending_sync_bios; + else + pending_bios = &device->pending_bios; + + pending = pending_bios->head; + tail = pending_bios->tail; WARN_ON(pending && !tail); - device->pending_bios = NULL; - device->pending_bio_tail = NULL; /* * if pending was null this time around, no bios need processing @@ -176,16 +202,41 @@ loop_lock: * device->running_pending is used to synchronize with the * schedule_bio code. */ - if (pending) { - again = 1; - device->running_pending = 1; - } else { + if (device->pending_sync_bios.head == NULL && + device->pending_bios.head == NULL) { again = 0; device->running_pending = 0; + } else { + again = 1; + device->running_pending = 1; } + + pending_bios->head = NULL; + pending_bios->tail = NULL; + spin_unlock(&device->io_lock); + /* + * if we're doing the regular priority list, make sure we unplug + * for any high prio bios we've sent down + */ + if (pending_bios == &device->pending_bios && num_sync_run > 0) { + num_sync_run = 0; + blk_run_backing_dev(bdi, NULL); + } + while (pending) { + + rmb(); + if (pending_bios != &device->pending_sync_bios && + device->pending_sync_bios.head && + num_run > 16) { + cond_resched(); + spin_lock(&device->io_lock); + requeue_list(pending_bios, pending, tail); + goto loop_lock; + } + cur = pending; pending = pending->bi_next; cur->bi_next = NULL; @@ -196,10 +247,18 @@ loop_lock: wake_up(&fs_info->async_submit_wait); BUG_ON(atomic_read(&cur->bi_cnt) == 0); - bio_get(cur); submit_bio(cur->bi_rw, cur); - bio_put(cur); num_run++; + if (bio_sync(cur)) + num_sync_run++; + + if (need_resched()) { + if (num_sync_run) { + blk_run_backing_dev(bdi, NULL); + num_sync_run = 0; + } + cond_resched(); + } /* * we made progress, there is more work to do and the bdi @@ -208,7 +267,6 @@ loop_lock: */ if (pending && bdi_write_congested(bdi) && num_run > 16 && fs_info->fs_devices->open_devices > 1) { - struct bio *old_head; struct io_context *ioc; ioc = current->io_context; @@ -233,17 +291,17 @@ loop_lock: * against it before looping */ last_waited = ioc->last_waited; + if (need_resched()) { + if (num_sync_run) { + blk_run_backing_dev(bdi, NULL); + num_sync_run = 0; + } + cond_resched(); + } continue; } spin_lock(&device->io_lock); - - old_head = device->pending_bios; - device->pending_bios = pending; - if (device->pending_bio_tail) - tail->bi_next = old_head; - else - device->pending_bio_tail = tail; - + requeue_list(pending_bios, pending, tail); device->running_pending = 1; spin_unlock(&device->io_lock); @@ -251,11 +309,18 @@ loop_lock: goto done; } } + + if (num_sync_run) { + num_sync_run = 0; + blk_run_backing_dev(bdi, NULL); + } + + cond_resched(); if (again) goto loop; spin_lock(&device->io_lock); - if (device->pending_bios) + if (device->pending_bios.head || device->pending_sync_bios.head) goto loop_lock; spin_unlock(&device->io_lock); @@ -2497,7 +2562,7 @@ again: max_errors = 1; } } - if (multi_ret && rw == WRITE && + if (multi_ret && (rw & (1 << BIO_RW)) && stripes_allocated < stripes_required) { stripes_allocated = map->num_stripes; free_extent_map(em); @@ -2762,6 +2827,7 @@ static noinline int schedule_bio(struct btrfs_root *root, int rw, struct bio *bio) { int should_queue = 1; + struct btrfs_pending_bios *pending_bios; /* don't bother with additional async steps for reads, right now */ if (!(rw & (1 << BIO_RW))) { @@ -2783,13 +2849,17 @@ static noinline int schedule_bio(struct btrfs_root *root, bio->bi_rw |= rw; spin_lock(&device->io_lock); + if (bio_sync(bio)) + pending_bios = &device->pending_sync_bios; + else + pending_bios = &device->pending_bios; - if (device->pending_bio_tail) - device->pending_bio_tail->bi_next = bio; + if (pending_bios->tail) + pending_bios->tail->bi_next = bio; - device->pending_bio_tail = bio; - if (!device->pending_bios) - device->pending_bios = bio; + pending_bios->tail = bio; + if (!pending_bios->head) + pending_bios->head = bio; if (device->running_pending) should_queue = 0; diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index 2185de72ff7d..5836327ba5dd 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -23,13 +23,22 @@ #include "async-thread.h" struct buffer_head; +struct btrfs_pending_bios { + struct bio *head; + struct bio *tail; +}; + struct btrfs_device { struct list_head dev_list; struct list_head dev_alloc_list; struct btrfs_fs_devices *fs_devices; struct btrfs_root *dev_root; - struct bio *pending_bios; - struct bio *pending_bio_tail; + + /* regular prio bios */ + struct btrfs_pending_bios pending_bios; + /* WRITE_SYNC bios */ + struct btrfs_pending_bios pending_sync_bios; + int running_pending; u64 generation; diff --git a/fs/buffer.c b/fs/buffer.c index 6e35762b6169..b3e5be7514f5 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -360,7 +360,7 @@ still_busy: * Completion handler for block_write_full_page() - pages which are unlocked * during I/O, and which have PageWriteback cleared upon I/O completion. */ -static void end_buffer_async_write(struct buffer_head *bh, int uptodate) +void end_buffer_async_write(struct buffer_head *bh, int uptodate) { char b[BDEVNAME_SIZE]; unsigned long flags; @@ -438,11 +438,17 @@ static void mark_buffer_async_read(struct buffer_head *bh) set_buffer_async_read(bh); } -void mark_buffer_async_write(struct buffer_head *bh) +void mark_buffer_async_write_endio(struct buffer_head *bh, + bh_end_io_t *handler) { - bh->b_end_io = end_buffer_async_write; + bh->b_end_io = handler; set_buffer_async_write(bh); } + +void mark_buffer_async_write(struct buffer_head *bh) +{ + mark_buffer_async_write_endio(bh, end_buffer_async_write); +} EXPORT_SYMBOL(mark_buffer_async_write); @@ -547,7 +553,7 @@ repeat: return err; } -void do_thaw_all(unsigned long unused) +void do_thaw_all(struct work_struct *work) { struct super_block *sb; char b[BDEVNAME_SIZE]; @@ -567,6 +573,7 @@ restart: goto restart; } spin_unlock(&sb_lock); + kfree(work); printk(KERN_WARNING "Emergency Thaw complete\n"); } @@ -577,7 +584,13 @@ restart: */ void emergency_thaw_all(void) { - pdflush_operation(do_thaw_all, 0); + struct work_struct *work; + + work = kmalloc(sizeof(*work), GFP_ATOMIC); + if (work) { + INIT_WORK(work, do_thaw_all); + schedule_work(work); + } } /** @@ -1596,9 +1609,20 @@ EXPORT_SYMBOL(unmap_underlying_metadata); * locked buffer. This only can happen if someone has written the buffer * directly, with submit_bh(). At the address_space level PageWriteback * prevents this contention from occurring. + * + * If block_write_full_page() is called with wbc->sync_mode == + * WB_SYNC_ALL, the writes are posted using WRITE_SYNC_PLUG; this + * causes the writes to be flagged as synchronous writes, but the + * block device queue will NOT be unplugged, since usually many pages + * will be pushed to the out before the higher-level caller actually + * waits for the writes to be completed. The various wait functions, + * such as wait_on_writeback_range() will ultimately call sync_page() + * which will ultimately call blk_run_backing_dev(), which will end up + * unplugging the device queue. */ static int __block_write_full_page(struct inode *inode, struct page *page, - get_block_t *get_block, struct writeback_control *wbc) + get_block_t *get_block, struct writeback_control *wbc, + bh_end_io_t *handler) { int err; sector_t block; @@ -1606,7 +1630,8 @@ static int __block_write_full_page(struct inode *inode, struct page *page, struct buffer_head *bh, *head; const unsigned blocksize = 1 << inode->i_blkbits; int nr_underway = 0; - int write_op = (wbc->sync_mode == WB_SYNC_ALL ? WRITE_SYNC : WRITE); + int write_op = (wbc->sync_mode == WB_SYNC_ALL ? + WRITE_SYNC_PLUG : WRITE); BUG_ON(!PageLocked(page)); @@ -1682,7 +1707,7 @@ static int __block_write_full_page(struct inode *inode, struct page *page, continue; } if (test_clear_buffer_dirty(bh)) { - mark_buffer_async_write(bh); + mark_buffer_async_write_endio(bh, handler); } else { unlock_buffer(bh); } @@ -1735,7 +1760,7 @@ recover: if (buffer_mapped(bh) && buffer_dirty(bh) && !buffer_delay(bh)) { lock_buffer(bh); - mark_buffer_async_write(bh); + mark_buffer_async_write_endio(bh, handler); } else { /* * The buffer may have been set dirty during @@ -2661,7 +2686,8 @@ int nobh_writepage(struct page *page, get_block_t *get_block, out: ret = mpage_writepage(page, get_block, wbc); if (ret == -EAGAIN) - ret = __block_write_full_page(inode, page, get_block, wbc); + ret = __block_write_full_page(inode, page, get_block, wbc, + end_buffer_async_write); return ret; } EXPORT_SYMBOL(nobh_writepage); @@ -2819,9 +2845,10 @@ out: /* * The generic ->writepage function for buffer-backed address_spaces + * this form passes in the end_io handler used to finish the IO. */ -int block_write_full_page(struct page *page, get_block_t *get_block, - struct writeback_control *wbc) +int block_write_full_page_endio(struct page *page, get_block_t *get_block, + struct writeback_control *wbc, bh_end_io_t *handler) { struct inode * const inode = page->mapping->host; loff_t i_size = i_size_read(inode); @@ -2830,7 +2857,8 @@ int block_write_full_page(struct page *page, get_block_t *get_block, /* Is the page fully inside i_size? */ if (page->index < end_index) - return __block_write_full_page(inode, page, get_block, wbc); + return __block_write_full_page(inode, page, get_block, wbc, + handler); /* Is the page fully outside i_size? (truncate in progress) */ offset = i_size & (PAGE_CACHE_SIZE-1); @@ -2853,9 +2881,20 @@ int block_write_full_page(struct page *page, get_block_t *get_block, * writes to that region are not written out to the file." */ zero_user_segment(page, offset, PAGE_CACHE_SIZE); - return __block_write_full_page(inode, page, get_block, wbc); + return __block_write_full_page(inode, page, get_block, wbc, handler); +} + +/* + * The generic ->writepage function for buffer-backed address_spaces + */ +int block_write_full_page(struct page *page, get_block_t *get_block, + struct writeback_control *wbc) +{ + return block_write_full_page_endio(page, get_block, wbc, + end_buffer_async_write); } + sector_t generic_block_bmap(struct address_space *mapping, sector_t block, get_block_t *get_block) { @@ -3324,9 +3363,11 @@ EXPORT_SYMBOL(block_read_full_page); EXPORT_SYMBOL(block_sync_page); EXPORT_SYMBOL(block_truncate_page); EXPORT_SYMBOL(block_write_full_page); +EXPORT_SYMBOL(block_write_full_page_endio); EXPORT_SYMBOL(cont_write_begin); EXPORT_SYMBOL(end_buffer_read_sync); EXPORT_SYMBOL(end_buffer_write_sync); +EXPORT_SYMBOL(end_buffer_async_write); EXPORT_SYMBOL(file_fsync); EXPORT_SYMBOL(generic_block_bmap); EXPORT_SYMBOL(generic_cont_expand_simple); diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 65984006192c..9d1fb6ec8a5a 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -15,7 +15,8 @@ Posix file open support added (turned off after one attempt if server fails to support it properly, as with Samba server versions prior to 3.3.2) Fix "redzone overwritten" bug in cifs_put_tcon (CIFSTcon may allocate too little memory for the "nativeFileSystem" field returned by the server -during mount). +during mount). Endian convert inode numbers if necessary (makes it easier +to compare inode numbers on network files from big endian systems). Version 1.56 ------------ diff --git a/fs/cifs/cifs_spnego.c b/fs/cifs/cifs_spnego.c index 3fd3a9df043a..67bf93a40d2e 100644 --- a/fs/cifs/cifs_spnego.c +++ b/fs/cifs/cifs_spnego.c @@ -41,7 +41,7 @@ cifs_spnego_key_instantiate(struct key *key, const void *data, size_t datalen) /* attach the data */ memcpy(payload, data, datalen); - rcu_assign_pointer(key->payload.data, payload); + key->payload.data = payload; ret = 0; error: diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 38491fd3871d..0d6d8b573652 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -66,9 +66,6 @@ unsigned int sign_CIFS_PDUs = 1; extern struct task_struct *oplockThread; /* remove sparse warning */ struct task_struct *oplockThread = NULL; /* extern struct task_struct * dnotifyThread; remove sparse warning */ -#ifdef CONFIG_CIFS_EXPERIMENTAL -static struct task_struct *dnotifyThread = NULL; -#endif static const struct super_operations cifs_super_ops; unsigned int CIFSMaxBufSize = CIFS_MAX_MSGSIZE; module_param(CIFSMaxBufSize, int, 0); @@ -316,6 +313,7 @@ cifs_alloc_inode(struct super_block *sb) cifs_inode->clientCanCacheAll = false; cifs_inode->delete_pending = false; cifs_inode->vfs_inode.i_blkbits = 14; /* 2**14 = CIFS_MAX_MSGSIZE */ + cifs_inode->server_eof = 0; /* Can not set i_flags here - they get immediately overwritten to zero by the VFS */ @@ -1040,34 +1038,6 @@ static int cifs_oplock_thread(void *dummyarg) return 0; } -#ifdef CONFIG_CIFS_EXPERIMENTAL -static int cifs_dnotify_thread(void *dummyarg) -{ - struct list_head *tmp; - struct TCP_Server_Info *server; - - do { - if (try_to_freeze()) - continue; - set_current_state(TASK_INTERRUPTIBLE); - schedule_timeout(15*HZ); - /* check if any stuck requests that need - to be woken up and wakeq so the - thread can wake up and error out */ - read_lock(&cifs_tcp_ses_lock); - list_for_each(tmp, &cifs_tcp_ses_list) { - server = list_entry(tmp, struct TCP_Server_Info, - tcp_ses_list); - if (atomic_read(&server->inFlight)) - wake_up_all(&server->response_q); - } - read_unlock(&cifs_tcp_ses_lock); - } while (!kthread_should_stop()); - - return 0; -} -#endif - static int __init init_cifs(void) { @@ -1144,21 +1114,8 @@ init_cifs(void) goto out_unregister_dfs_key_type; } -#ifdef CONFIG_CIFS_EXPERIMENTAL - dnotifyThread = kthread_run(cifs_dnotify_thread, NULL, "cifsdnotifyd"); - if (IS_ERR(dnotifyThread)) { - rc = PTR_ERR(dnotifyThread); - cERROR(1, ("error %d create dnotify thread", rc)); - goto out_stop_oplock_thread; - } -#endif - return 0; -#ifdef CONFIG_CIFS_EXPERIMENTAL - out_stop_oplock_thread: -#endif - kthread_stop(oplockThread); out_unregister_dfs_key_type: #ifdef CONFIG_CIFS_DFS_UPCALL unregister_key_type(&key_type_dns_resolver); @@ -1196,9 +1153,6 @@ exit_cifs(void) cifs_destroy_inodecache(); cifs_destroy_mids(); cifs_destroy_request_bufs(); -#ifdef CONFIG_CIFS_EXPERIMENTAL - kthread_stop(dnotifyThread); -#endif kthread_stop(oplockThread); } diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 9fbf4dff5da6..df40ab64cd95 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -350,7 +350,7 @@ struct cifsFileInfo { bool invalidHandle:1; /* file closed via session abend */ bool messageMode:1; /* for pipes: message vs byte mode */ atomic_t wrtPending; /* handle in use - defer close */ - struct semaphore fh_sem; /* prevents reopen race after dead ses*/ + struct mutex fh_mutex; /* prevents reopen race after dead ses*/ struct cifs_search_info srch_inf; }; @@ -370,6 +370,7 @@ struct cifsInodeInfo { bool clientCanCacheAll:1; /* read and writebehind oplock */ bool oplockPending:1; bool delete_pending:1; /* DELETE_ON_CLOSE is set */ + u64 server_eof; /* current file size on server */ struct inode vfs_inode; }; diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index b370489c8da5..a785f69dbc9f 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -2163,7 +2163,7 @@ typedef struct { __le32 Type; __le64 DevMajor; __le64 DevMinor; - __u64 UniqueId; + __le64 UniqueId; __le64 Permissions; __le64 Nlinks; } __attribute__((packed)) FILE_UNIX_BASIC_INFO; /* level 0x200 QPathInfo */ @@ -2308,7 +2308,7 @@ struct unlink_psx_rq { /* level 0x20a SetPathInfo */ } __attribute__((packed)); struct file_internal_info { - __u64 UniqueId; /* inode number */ + __le64 UniqueId; /* inode number */ } __attribute__((packed)); /* level 0x3ee */ struct file_mode_info { @@ -2338,7 +2338,7 @@ typedef struct { __le32 Type; __le64 DevMajor; __le64 DevMinor; - __u64 UniqueId; + __le64 UniqueId; __le64 Permissions; __le64 Nlinks; char FileName[1]; @@ -2386,7 +2386,7 @@ typedef struct { __le32 FileNameLength; __le32 EaSize; /* EA size */ __le32 Reserved; - __u64 UniqueId; /* inode num - le since Samba puts ino in low 32 bit*/ + __le64 UniqueId; /* inode num - le since Samba puts ino in low 32 bit*/ char FileName[1]; } __attribute__((packed)) SEARCH_ID_FULL_DIR_INFO; /* level 0x105 FF rsp data */ diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index bc09c998631f..a0845dc7b8a9 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -1626,6 +1626,8 @@ CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, int smb_hdr_len; int resp_buf_type = 0; + *nbytes = 0; + cFYI(1, ("write2 at %lld %d bytes", (long long)offset, count)); if (tcon->ses->capabilities & CAP_LARGE_FILES) { @@ -1682,11 +1684,9 @@ CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, cifs_stats_inc(&tcon->num_writes); if (rc) { cFYI(1, ("Send error Write2 = %d", rc)); - *nbytes = 0; } else if (resp_buf_type == 0) { /* presumably this can not happen, but best to be safe */ rc = -EIO; - *nbytes = 0; } else { WRITE_RSP *pSMBr = (WRITE_RSP *)iov[0].iov_base; *nbytes = le16_to_cpu(pSMBr->CountHigh); @@ -3918,7 +3918,7 @@ GetInodeNumberRetry: } pfinfo = (struct file_internal_info *) (data_offset + (char *) &pSMBr->hdr.Protocol); - *inode_number = pfinfo->UniqueId; + *inode_number = le64_to_cpu(pfinfo->UniqueId); } } GetInodeNumOut: diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 0de3b5615a22..bacdef1546b7 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -2214,9 +2214,58 @@ is_path_accessible(int xid, struct cifsTconInfo *tcon, return rc; } +static void +cleanup_volume_info(struct smb_vol **pvolume_info) +{ + struct smb_vol *volume_info; + + if (!pvolume_info && !*pvolume_info) + return; + + volume_info = *pvolume_info; + kzfree(volume_info->password); + kfree(volume_info->UNC); + kfree(volume_info->prepath); + kfree(volume_info); + *pvolume_info = NULL; + return; +} + +#ifdef CONFIG_CIFS_DFS_UPCALL +/* build_path_to_root returns full path to root when + * we do not have an exiting connection (tcon) */ +static char * +build_unc_path_to_root(const struct smb_vol *volume_info, + const struct cifs_sb_info *cifs_sb) +{ + char *full_path; + + int unc_len = strnlen(volume_info->UNC, MAX_TREE_SIZE + 1); + full_path = kmalloc(unc_len + cifs_sb->prepathlen + 1, GFP_KERNEL); + if (full_path == NULL) + return ERR_PTR(-ENOMEM); + + strncpy(full_path, volume_info->UNC, unc_len); + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) { + int i; + for (i = 0; i < unc_len; i++) { + if (full_path[i] == '\\') + full_path[i] = '/'; + } + } + + if (cifs_sb->prepathlen) + strncpy(full_path + unc_len, cifs_sb->prepath, + cifs_sb->prepathlen); + + full_path[unc_len + cifs_sb->prepathlen] = 0; /* add trailing null */ + return full_path; +} +#endif + int cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, - char *mount_data, const char *devname) + char *mount_data_global, const char *devname) { int rc = 0; int xid; @@ -2225,6 +2274,13 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, struct cifsTconInfo *tcon = NULL; struct TCP_Server_Info *srvTcp = NULL; char *full_path; + char *mount_data = mount_data_global; +#ifdef CONFIG_CIFS_DFS_UPCALL + struct dfs_info3_param *referrals = NULL; + unsigned int num_referrals = 0; +try_mount_again: +#endif + full_path = NULL; xid = GetXid(); @@ -2371,11 +2427,9 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, } } - /* check for null share name ie connect to dfs root */ if ((strchr(volume_info->UNC + 3, '\\') == NULL) && (strchr(volume_info->UNC + 3, '/') == NULL)) { - /* rc = connect_to_dfs_path(...) */ - cFYI(1, ("DFS root not supported")); + cERROR(1, ("Missing share name")); rc = -ENODEV; goto mount_fail_check; } else { @@ -2392,7 +2446,7 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, } } if (rc) - goto mount_fail_check; + goto remote_path_check; tcon->seal = volume_info->seal; write_lock(&cifs_tcp_ses_lock); list_add(&tcon->tcon_list, &pSesInfo->tcon_list); @@ -2417,19 +2471,9 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, /* BB FIXME fix time_gran to be larger for LANMAN sessions */ sb->s_time_gran = 100; -mount_fail_check: - /* on error free sesinfo and tcon struct if needed */ - if (rc) { - /* If find_unc succeeded then rc == 0 so we can not end */ - /* up accidently freeing someone elses tcon struct */ - if (tcon) - cifs_put_tcon(tcon); - else if (pSesInfo) - cifs_put_smb_ses(pSesInfo); - else - cifs_put_tcp_session(srvTcp); - goto out; - } + if (rc) + goto remote_path_check; + cifs_sb->tcon = tcon; /* do not care if following two calls succeed - informational */ @@ -2461,7 +2505,9 @@ mount_fail_check: cifs_sb->rsize = min(cifs_sb->rsize, (tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE)); - if (!rc && cifs_sb->prepathlen) { +remote_path_check: + /* check if a whole path (including prepath) is not remote */ + if (!rc && cifs_sb->prepathlen && tcon) { /* build_path_to_root works only when we have a valid tcon */ full_path = cifs_build_path_to_root(cifs_sb); if (full_path == NULL) { @@ -2469,31 +2515,79 @@ mount_fail_check: goto mount_fail_check; } rc = is_path_accessible(xid, tcon, cifs_sb, full_path); - if (rc) { - cERROR(1, ("Path %s in not accessible: %d", - full_path, rc)); + if (rc != -EREMOTE) { kfree(full_path); goto mount_fail_check; } kfree(full_path); } + /* get referral if needed */ + if (rc == -EREMOTE) { +#ifdef CONFIG_CIFS_DFS_UPCALL + /* convert forward to back slashes in prepath here if needed */ + if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) == 0) + convert_delimiter(cifs_sb->prepath, + CIFS_DIR_SEP(cifs_sb)); + full_path = build_unc_path_to_root(volume_info, cifs_sb); + if (IS_ERR(full_path)) { + rc = PTR_ERR(full_path); + goto mount_fail_check; + } + + cFYI(1, ("Getting referral for: %s", full_path)); + rc = get_dfs_path(xid, pSesInfo , full_path + 1, + cifs_sb->local_nls, &num_referrals, &referrals, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + if (!rc && num_referrals > 0) { + char *fake_devname = NULL; + + if (mount_data != mount_data_global) + kfree(mount_data); + mount_data = cifs_compose_mount_options( + cifs_sb->mountdata, full_path + 1, + referrals, &fake_devname); + kfree(fake_devname); + free_dfs_info_array(referrals, num_referrals); + + if (tcon) + cifs_put_tcon(tcon); + else if (pSesInfo) + cifs_put_smb_ses(pSesInfo); + + cleanup_volume_info(&volume_info); + FreeXid(xid); + kfree(full_path); + goto try_mount_again; + } +#else /* No DFS support, return error on mount */ + rc = -EOPNOTSUPP; +#endif + } + +mount_fail_check: + /* on error free sesinfo and tcon struct if needed */ + if (rc) { + if (mount_data != mount_data_global) + kfree(mount_data); + /* If find_unc succeeded then rc == 0 so we can not end */ + /* up accidently freeing someone elses tcon struct */ + if (tcon) + cifs_put_tcon(tcon); + else if (pSesInfo) + cifs_put_smb_ses(pSesInfo); + else + cifs_put_tcp_session(srvTcp); + goto out; + } + /* volume_info->password is freed above when existing session found (in which case it is not needed anymore) but when new sesion is created the password ptr is put in the new session structure (in which case the password will be freed at unmount time) */ out: /* zero out password before freeing */ - if (volume_info) { - if (volume_info->password != NULL) { - memset(volume_info->password, 0, - strlen(volume_info->password)); - kfree(volume_info->password); - } - kfree(volume_info->UNC); - kfree(volume_info->prepath); - kfree(volume_info); - } + cleanup_volume_info(&volume_info); FreeXid(xid); return rc; } @@ -2673,8 +2767,7 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, /* We look for obvious messed up bcc or strings in response so we do not go off the end since (at least) WIN2K and Windows XP have a major bug in not null terminating last Unicode string in response */ - if (ses->serverOS) - kfree(ses->serverOS); + kfree(ses->serverOS); ses->serverOS = kzalloc(2 * (len + 1), GFP_KERNEL); if (ses->serverOS == NULL) @@ -2710,8 +2803,7 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, len = UniStrnlen((wchar_t *) bcc_ptr, remaining_words); /* last string is not always null terminated (for e.g. for Windows XP & 2000) */ - if (ses->serverDomain) - kfree(ses->serverDomain); + kfree(ses->serverDomain); ses->serverDomain = kzalloc(2*(len+1), GFP_KERNEL); @@ -2725,8 +2817,7 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, ses->serverDomain[1+(2*len)] = 0; } else { /* else no more room so create dummy domain string */ - if (ses->serverDomain) - kfree(ses->serverDomain); + kfree(ses->serverDomain); ses->serverDomain = kzalloc(2, GFP_KERNEL); } @@ -2772,8 +2863,7 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, bcc_ptr++; len = strnlen(bcc_ptr, 1024); - if (ses->serverDomain) - kfree(ses->serverDomain); + kfree(ses->serverDomain); ses->serverDomain = kzalloc(len + 1, GFP_KERNEL); if (ses->serverDomain == NULL) @@ -3013,8 +3103,7 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, /* We look for obvious messed up bcc or strings in response so we do not go off the end since (at least) WIN2K and Windows XP have a major bug in not null terminating last Unicode string in response */ - if (ses->serverOS) - kfree(ses->serverOS); + kfree(ses->serverOS); ses->serverOS = kzalloc(2 * (len + 1), GFP_KERNEL); cifs_strfromUCS_le(ses->serverOS, @@ -3086,8 +3175,7 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, if (((long) bcc_ptr + len) - (long) pByteArea(smb_buffer_response) <= BCC(smb_buffer_response)) { - if (ses->serverOS) - kfree(ses->serverOS); + kfree(ses->serverOS); ses->serverOS = kzalloc(len + 1, GFP_KERNEL); @@ -3414,8 +3502,7 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, /* We look for obvious messed up bcc or strings in response so we do not go off the end since (at least) WIN2K and Windows XP have a major bug in not null terminating last Unicode string in response */ - if (ses->serverOS) - kfree(ses->serverOS); + kfree(ses->serverOS); ses->serverOS = kzalloc(2 * (len + 1), GFP_KERNEL); cifs_strfromUCS_le(ses->serverOS, @@ -3448,8 +3535,7 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, if (remaining_words > 0) { len = UniStrnlen((wchar_t *) bcc_ptr, remaining_words); /* last string not always null terminated (e.g. for Windows XP & 2000) */ - if (ses->serverDomain) - kfree(ses->serverDomain); + kfree(ses->serverDomain); ses->serverDomain = kzalloc(2 * (len + @@ -3476,13 +3562,11 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, = 0; } /* else no more room so create dummy domain string */ else { - if (ses->serverDomain) - kfree(ses->serverDomain); + kfree(ses->serverDomain); ses->serverDomain = kzalloc(2,GFP_KERNEL); } } else { /* no room so create dummy domain and NOS string */ - if (ses->serverDomain) - kfree(ses->serverDomain); + kfree(ses->serverDomain); ses->serverDomain = kzalloc(2, GFP_KERNEL); kfree(ses->serverNOS); ses->serverNOS = kzalloc(2, GFP_KERNEL); @@ -3492,8 +3576,7 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, if (((long) bcc_ptr + len) - (long) pByteArea(smb_buffer_response) <= BCC(smb_buffer_response)) { - if (ses->serverOS) - kfree(ses->serverOS); + kfree(ses->serverOS); ses->serverOS = kzalloc(len + 1, GFP_KERNEL); strncpy(ses->serverOS,bcc_ptr, len); @@ -3512,8 +3595,7 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, bcc_ptr++; len = strnlen(bcc_ptr, 1024); - if (ses->serverDomain) - kfree(ses->serverDomain); + kfree(ses->serverDomain); ses->serverDomain = kzalloc(len+1, GFP_KERNEL); @@ -3674,16 +3756,15 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, BCC(smb_buffer_response)) { kfree(tcon->nativeFileSystem); tcon->nativeFileSystem = - kzalloc(2*(length + 1), GFP_KERNEL); - if (tcon->nativeFileSystem) + kzalloc((4 * length) + 2, GFP_KERNEL); + if (tcon->nativeFileSystem) { cifs_strfromUCS_le( tcon->nativeFileSystem, (__le16 *) bcc_ptr, length, nls_codepage); - bcc_ptr += 2 * length; - bcc_ptr[0] = 0; /* null terminate the string */ - bcc_ptr[1] = 0; - bcc_ptr += 2; + cFYI(1, ("nativeFileSystem=%s", + tcon->nativeFileSystem)); + } } /* else do not bother copying these information fields*/ } else { diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 54dce78fbb73..461750e01364 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -129,12 +129,62 @@ cifs_bp_rename_retry: return full_path; } +static void +cifs_fill_fileinfo(struct inode *newinode, __u16 fileHandle, + struct cifsTconInfo *tcon, bool write_only) +{ + int oplock = 0; + struct cifsFileInfo *pCifsFile; + struct cifsInodeInfo *pCifsInode; + + pCifsFile = kzalloc(sizeof(struct cifsFileInfo), GFP_KERNEL); + + if (pCifsFile == NULL) + return; + + if (oplockEnabled) + oplock = REQ_OPLOCK; + + pCifsFile->netfid = fileHandle; + pCifsFile->pid = current->tgid; + pCifsFile->pInode = newinode; + pCifsFile->invalidHandle = false; + pCifsFile->closePend = false; + mutex_init(&pCifsFile->fh_mutex); + mutex_init(&pCifsFile->lock_mutex); + INIT_LIST_HEAD(&pCifsFile->llist); + atomic_set(&pCifsFile->wrtPending, 0); + + /* set the following in open now + pCifsFile->pfile = file; */ + write_lock(&GlobalSMBSeslock); + list_add(&pCifsFile->tlist, &tcon->openFileList); + pCifsInode = CIFS_I(newinode); + if (pCifsInode) { + /* if readable file instance put first in list*/ + if (write_only) + list_add_tail(&pCifsFile->flist, + &pCifsInode->openFileList); + else + list_add(&pCifsFile->flist, &pCifsInode->openFileList); + + if ((oplock & 0xF) == OPLOCK_EXCLUSIVE) { + pCifsInode->clientCanCacheAll = true; + pCifsInode->clientCanCacheRead = true; + cFYI(1, ("Exclusive Oplock inode %p", newinode)); + } else if ((oplock & 0xF) == OPLOCK_READ) + pCifsInode->clientCanCacheRead = true; + } + write_unlock(&GlobalSMBSeslock); +} + int cifs_posix_open(char *full_path, struct inode **pinode, struct super_block *sb, int mode, int oflags, int *poplock, __u16 *pnetfid, int xid) { int rc; __u32 oplock; + bool write_only = false; FILE_UNIX_BASIC_INFO *presp_data; __u32 posix_flags = 0; struct cifs_sb_info *cifs_sb = CIFS_SB(sb); @@ -172,6 +222,8 @@ int cifs_posix_open(char *full_path, struct inode **pinode, if (oflags & O_DIRECT) posix_flags |= SMB_O_DIRECT; + if (!(oflags & FMODE_READ)) + write_only = true; rc = CIFSPOSIXCreate(xid, cifs_sb->tcon, posix_flags, mode, pnetfid, presp_data, &oplock, full_path, @@ -187,8 +239,10 @@ int cifs_posix_open(char *full_path, struct inode **pinode, if (!pinode) goto posix_open_ret; /* caller does not need info */ - if (*pinode == NULL) - *pinode = cifs_new_inode(sb, &presp_data->UniqueId); + if (*pinode == NULL) { + __u64 unique_id = le64_to_cpu(presp_data->UniqueId); + *pinode = cifs_new_inode(sb, &unique_id); + } /* else an inode was passed in. Update its info, don't create one */ /* We do not need to close the file if new_inode fails since @@ -198,6 +252,8 @@ int cifs_posix_open(char *full_path, struct inode **pinode, posix_fill_in_inode(*pinode, presp_data, 1); + cifs_fill_fileinfo(*pinode, *pnetfid, cifs_sb->tcon, write_only); + posix_open_ret: kfree(presp_data); return rc; @@ -239,7 +295,6 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, char *full_path = NULL; FILE_ALL_INFO *buf = NULL; struct inode *newinode = NULL; - struct cifsInodeInfo *pCifsInode; int disposition = FILE_OVERWRITE_IF; bool write_only = false; @@ -410,44 +465,8 @@ cifs_create_set_dentry: /* mknod case - do not leave file open */ CIFSSMBClose(xid, tcon, fileHandle); } else if (newinode) { - struct cifsFileInfo *pCifsFile = - kzalloc(sizeof(struct cifsFileInfo), GFP_KERNEL); - - if (pCifsFile == NULL) - goto cifs_create_out; - pCifsFile->netfid = fileHandle; - pCifsFile->pid = current->tgid; - pCifsFile->pInode = newinode; - pCifsFile->invalidHandle = false; - pCifsFile->closePend = false; - init_MUTEX(&pCifsFile->fh_sem); - mutex_init(&pCifsFile->lock_mutex); - INIT_LIST_HEAD(&pCifsFile->llist); - atomic_set(&pCifsFile->wrtPending, 0); - - /* set the following in open now - pCifsFile->pfile = file; */ - write_lock(&GlobalSMBSeslock); - list_add(&pCifsFile->tlist, &tcon->openFileList); - pCifsInode = CIFS_I(newinode); - if (pCifsInode) { - /* if readable file instance put first in list*/ - if (write_only) { - list_add_tail(&pCifsFile->flist, - &pCifsInode->openFileList); - } else { - list_add(&pCifsFile->flist, - &pCifsInode->openFileList); - } - if ((oplock & 0xF) == OPLOCK_EXCLUSIVE) { - pCifsInode->clientCanCacheAll = true; - pCifsInode->clientCanCacheRead = true; - cFYI(1, ("Exclusive Oplock inode %p", - newinode)); - } else if ((oplock & 0xF) == OPLOCK_READ) - pCifsInode->clientCanCacheRead = true; - } - write_unlock(&GlobalSMBSeslock); + cifs_fill_fileinfo(newinode, fileHandle, + cifs_sb->tcon, write_only); } cifs_create_out: kfree(buf); @@ -580,17 +599,21 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, return rc; } - struct dentry * cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, struct nameidata *nd) { int xid; int rc = 0; /* to get around spurious gcc warning, set to zero here */ + int oplock = 0; + int mode; + __u16 fileHandle = 0; + bool posix_open = false; struct cifs_sb_info *cifs_sb; struct cifsTconInfo *pTcon; struct inode *newInode = NULL; char *full_path = NULL; + struct file *filp; xid = GetXid(); @@ -632,12 +655,37 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, } cFYI(1, ("Full path: %s inode = 0x%p", full_path, direntry->d_inode)); - if (pTcon->unix_ext) - rc = cifs_get_inode_info_unix(&newInode, full_path, - parent_dir_inode->i_sb, xid); - else + if (pTcon->unix_ext) { + if (!(nd->flags & (LOOKUP_PARENT | LOOKUP_DIRECTORY)) && + (nd->flags & LOOKUP_OPEN)) { + if (!((nd->intent.open.flags & O_CREAT) && + (nd->intent.open.flags & O_EXCL))) { + mode = nd->intent.open.create_mode & + ~current_umask(); + rc = cifs_posix_open(full_path, &newInode, + parent_dir_inode->i_sb, mode, + nd->intent.open.flags, &oplock, + &fileHandle, xid); + /* + * This code works around a bug in + * samba posix open in samba versions 3.3.1 + * and earlier where create works + * but open fails with invalid parameter. + * If either of these error codes are + * returned, follow the normal lookup. + * Otherwise, the error during posix open + * is handled. + */ + if ((rc != -EINVAL) && (rc != -EOPNOTSUPP)) + posix_open = true; + } + } + if (!posix_open) + rc = cifs_get_inode_info_unix(&newInode, full_path, + parent_dir_inode->i_sb, xid); + } else rc = cifs_get_inode_info(&newInode, full_path, NULL, - parent_dir_inode->i_sb, xid, NULL); + parent_dir_inode->i_sb, xid, NULL); if ((rc == 0) && (newInode != NULL)) { if (pTcon->nocase) @@ -645,7 +693,8 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, else direntry->d_op = &cifs_dentry_ops; d_add(direntry, newInode); - + if (posix_open) + filp = lookup_instantiate_filp(nd, direntry, NULL); /* since paths are not looked up by component - the parent directories are presumed to be good here */ renew_parental_timestamps(direntry); diff --git a/fs/cifs/dns_resolve.c b/fs/cifs/dns_resolve.c index 1e0c1bd8f2e4..df4a306f697e 100644 --- a/fs/cifs/dns_resolve.c +++ b/fs/cifs/dns_resolve.c @@ -78,7 +78,7 @@ dns_resolver_instantiate(struct key *key, const void *data, } key->type_data.x[0] = datalen; - rcu_assign_pointer(key->payload.data, ip); + key->payload.data = ip; return rc; } diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 81747acca4c4..50ca088d8860 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -46,7 +46,7 @@ static inline struct cifsFileInfo *cifs_init_private( memset(private_data, 0, sizeof(struct cifsFileInfo)); private_data->netfid = netfid; private_data->pid = current->tgid; - init_MUTEX(&private_data->fh_sem); + mutex_init(&private_data->fh_mutex); mutex_init(&private_data->lock_mutex); INIT_LIST_HEAD(&private_data->llist); private_data->pfile = file; /* needed for writepage */ @@ -284,35 +284,32 @@ int cifs_open(struct inode *inode, struct file *file) cifs_sb = CIFS_SB(inode->i_sb); tcon = cifs_sb->tcon; - if (file->f_flags & O_CREAT) { - /* search inode for this file and fill in file->private_data */ - pCifsInode = CIFS_I(file->f_path.dentry->d_inode); - read_lock(&GlobalSMBSeslock); - list_for_each(tmp, &pCifsInode->openFileList) { - pCifsFile = list_entry(tmp, struct cifsFileInfo, - flist); - if ((pCifsFile->pfile == NULL) && - (pCifsFile->pid == current->tgid)) { - /* mode set in cifs_create */ - - /* needed for writepage */ - pCifsFile->pfile = file; - - file->private_data = pCifsFile; - break; - } - } - read_unlock(&GlobalSMBSeslock); - if (file->private_data != NULL) { - rc = 0; - FreeXid(xid); - return rc; - } else { - if (file->f_flags & O_EXCL) - cERROR(1, ("could not find file instance for " - "new file %p", file)); + /* search inode for this file and fill in file->private_data */ + pCifsInode = CIFS_I(file->f_path.dentry->d_inode); + read_lock(&GlobalSMBSeslock); + list_for_each(tmp, &pCifsInode->openFileList) { + pCifsFile = list_entry(tmp, struct cifsFileInfo, + flist); + if ((pCifsFile->pfile == NULL) && + (pCifsFile->pid == current->tgid)) { + /* mode set in cifs_create */ + + /* needed for writepage */ + pCifsFile->pfile = file; + + file->private_data = pCifsFile; + break; } } + read_unlock(&GlobalSMBSeslock); + + if (file->private_data != NULL) { + rc = 0; + FreeXid(xid); + return rc; + } else if ((file->f_flags & O_CREAT) && (file->f_flags & O_EXCL)) + cERROR(1, ("could not find file instance for " + "new file %p", file)); full_path = build_path_from_dentry(file->f_path.dentry); if (full_path == NULL) { @@ -500,9 +497,9 @@ static int cifs_reopen_file(struct file *file, bool can_flush) return -EBADF; xid = GetXid(); - down(&pCifsFile->fh_sem); + mutex_unlock(&pCifsFile->fh_mutex); if (!pCifsFile->invalidHandle) { - up(&pCifsFile->fh_sem); + mutex_lock(&pCifsFile->fh_mutex); FreeXid(xid); return 0; } @@ -533,7 +530,7 @@ static int cifs_reopen_file(struct file *file, bool can_flush) if (full_path == NULL) { rc = -ENOMEM; reopen_error_exit: - up(&pCifsFile->fh_sem); + mutex_lock(&pCifsFile->fh_mutex); FreeXid(xid); return rc; } @@ -575,14 +572,14 @@ reopen_error_exit: cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc) { - up(&pCifsFile->fh_sem); + mutex_lock(&pCifsFile->fh_mutex); cFYI(1, ("cifs_open returned 0x%x", rc)); cFYI(1, ("oplock: %d", oplock)); } else { reopen_success: pCifsFile->netfid = netfid; pCifsFile->invalidHandle = false; - up(&pCifsFile->fh_sem); + mutex_lock(&pCifsFile->fh_mutex); pCifsInode = CIFS_I(inode); if (pCifsInode) { if (can_flush) { @@ -971,6 +968,40 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) return rc; } +/* + * Set the timeout on write requests past EOF. For some servers (Windows) + * these calls can be very long. + * + * If we're writing >10M past the EOF we give a 180s timeout. Anything less + * than that gets a 45s timeout. Writes not past EOF get 15s timeouts. + * The 10M cutoff is totally arbitrary. A better scheme for this would be + * welcome if someone wants to suggest one. + * + * We may be able to do a better job with this if there were some way to + * declare that a file should be sparse. + */ +static int +cifs_write_timeout(struct cifsInodeInfo *cifsi, loff_t offset) +{ + if (offset <= cifsi->server_eof) + return CIFS_STD_OP; + else if (offset > (cifsi->server_eof + (10 * 1024 * 1024))) + return CIFS_VLONG_OP; + else + return CIFS_LONG_OP; +} + +/* update the file size (if needed) after a write */ +static void +cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset, + unsigned int bytes_written) +{ + loff_t end_of_write = offset + bytes_written; + + if (end_of_write > cifsi->server_eof) + cifsi->server_eof = end_of_write; +} + ssize_t cifs_user_write(struct file *file, const char __user *write_data, size_t write_size, loff_t *poffset) { @@ -981,6 +1012,7 @@ ssize_t cifs_user_write(struct file *file, const char __user *write_data, struct cifsTconInfo *pTcon; int xid, long_op; struct cifsFileInfo *open_file; + struct cifsInodeInfo *cifsi = CIFS_I(file->f_path.dentry->d_inode); cifs_sb = CIFS_SB(file->f_path.dentry->d_sb); @@ -1000,11 +1032,7 @@ ssize_t cifs_user_write(struct file *file, const char __user *write_data, xid = GetXid(); - if (*poffset > file->f_path.dentry->d_inode->i_size) - long_op = CIFS_VLONG_OP; /* writes past EOF take long time */ - else - long_op = CIFS_LONG_OP; - + long_op = cifs_write_timeout(cifsi, *poffset); for (total_written = 0; write_size > total_written; total_written += bytes_written) { rc = -EAGAIN; @@ -1048,8 +1076,10 @@ ssize_t cifs_user_write(struct file *file, const char __user *write_data, FreeXid(xid); return rc; } - } else + } else { + cifs_update_eof(cifsi, *poffset, bytes_written); *poffset += bytes_written; + } long_op = CIFS_STD_OP; /* subsequent writes fast - 15 seconds is plenty */ } @@ -1085,6 +1115,7 @@ static ssize_t cifs_write(struct file *file, const char *write_data, struct cifsTconInfo *pTcon; int xid, long_op; struct cifsFileInfo *open_file; + struct cifsInodeInfo *cifsi = CIFS_I(file->f_path.dentry->d_inode); cifs_sb = CIFS_SB(file->f_path.dentry->d_sb); @@ -1099,11 +1130,7 @@ static ssize_t cifs_write(struct file *file, const char *write_data, xid = GetXid(); - if (*poffset > file->f_path.dentry->d_inode->i_size) - long_op = CIFS_VLONG_OP; /* writes past EOF can be slow */ - else - long_op = CIFS_LONG_OP; - + long_op = cifs_write_timeout(cifsi, *poffset); for (total_written = 0; write_size > total_written; total_written += bytes_written) { rc = -EAGAIN; @@ -1166,8 +1193,10 @@ static ssize_t cifs_write(struct file *file, const char *write_data, FreeXid(xid); return rc; } - } else + } else { + cifs_update_eof(cifsi, *poffset, bytes_written); *poffset += bytes_written; + } long_op = CIFS_STD_OP; /* subsequent writes fast - 15 seconds is plenty */ } @@ -1380,11 +1409,12 @@ static int cifs_writepages(struct address_space *mapping, int nr_pages; __u64 offset = 0; struct cifsFileInfo *open_file; + struct cifsInodeInfo *cifsi = CIFS_I(mapping->host); struct page *page; struct pagevec pvec; int rc = 0; int scanned = 0; - int xid; + int xid, long_op; cifs_sb = CIFS_SB(mapping->host->i_sb); @@ -1528,12 +1558,15 @@ retry: cERROR(1, ("No writable handles for inode")); rc = -EBADF; } else { + long_op = cifs_write_timeout(cifsi, offset); rc = CIFSSMBWrite2(xid, cifs_sb->tcon, open_file->netfid, bytes_to_write, offset, &bytes_written, iov, n_iov, - CIFS_LONG_OP); + long_op); atomic_dec(&open_file->wrtPending); + cifs_update_eof(cifsi, offset, bytes_written); + if (rc || bytes_written < bytes_to_write) { cERROR(1, ("Write2 ret %d, wrote %d", rc, bytes_written)); diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index f121a80fdd6f..f36b4e40e443 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -143,6 +143,7 @@ static void cifs_unix_info_to_inode(struct inode *inode, inode->i_nlink = le64_to_cpu(info->Nlinks); + cifsInfo->server_eof = end_of_file; spin_lock(&inode->i_lock); if (is_size_safe_to_change(cifsInfo, end_of_file)) { /* @@ -276,7 +277,8 @@ int cifs_get_inode_info_unix(struct inode **pinode, /* get new inode */ if (*pinode == NULL) { - *pinode = cifs_new_inode(sb, &find_data.UniqueId); + __u64 unique_id = le64_to_cpu(find_data.UniqueId); + *pinode = cifs_new_inode(sb, &unique_id); if (*pinode == NULL) { rc = -ENOMEM; goto cgiiu_exit; @@ -605,12 +607,12 @@ int cifs_get_inode_info(struct inode **pinode, inode->i_mode |= S_IFREG; } + cifsInfo->server_eof = le64_to_cpu(pfindData->EndOfFile); spin_lock(&inode->i_lock); - if (is_size_safe_to_change(cifsInfo, - le64_to_cpu(pfindData->EndOfFile))) { + if (is_size_safe_to_change(cifsInfo, cifsInfo->server_eof)) { /* can not safely shrink the file size here if the client is writing to it due to potential races */ - i_size_write(inode, le64_to_cpu(pfindData->EndOfFile)); + i_size_write(inode, cifsInfo->server_eof); /* 512 bytes (2**9) is the fake blocksize that must be used for this calculation */ @@ -1138,6 +1140,7 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) cFYI(1, ("posix mkdir returned 0x%x", rc)); d_drop(direntry); } else { + __u64 unique_id; if (pInfo->Type == cpu_to_le32(-1)) { /* no return info, go query for it */ kfree(pInfo); @@ -1151,8 +1154,8 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) else direntry->d_op = &cifs_dentry_ops; - newinode = cifs_new_inode(inode->i_sb, - &pInfo->UniqueId); + unique_id = le64_to_cpu(pInfo->UniqueId); + newinode = cifs_new_inode(inode->i_sb, &unique_id); if (newinode == NULL) { kfree(pInfo); goto mkdir_get_info; @@ -1450,7 +1453,8 @@ int cifs_rename(struct inode *source_dir, struct dentry *source_dentry, checking the UniqueId via FILE_INTERNAL_INFO */ unlink_target: - if ((rc == -EACCES) || (rc == -EEXIST)) { + /* Try unlinking the target dentry if it's not negative */ + if (target_dentry->d_inode && (rc == -EACCES || rc == -EEXIST)) { tmprc = cifs_unlink(target_dir, target_dentry); if (tmprc) goto cifs_rename_exit; @@ -1753,6 +1757,7 @@ cifs_set_file_size(struct inode *inode, struct iattr *attrs, } if (rc == 0) { + cifsInode->server_eof = attrs->ia_size; rc = cifs_vmtruncate(inode, attrs->ia_size); cifs_truncate_page(inode->i_mapping, inode->i_size); } @@ -1792,20 +1797,21 @@ cifs_setattr_unix(struct dentry *direntry, struct iattr *attrs) goto out; } - if ((attrs->ia_valid & ATTR_MTIME) || (attrs->ia_valid & ATTR_SIZE)) { - /* - Flush data before changing file size or changing the last - write time of the file on the server. If the - flush returns error, store it to report later and continue. - BB: This should be smarter. Why bother flushing pages that - will be truncated anyway? Also, should we error out here if - the flush returns error? - */ - rc = filemap_write_and_wait(inode->i_mapping); - if (rc != 0) { - cifsInode->write_behind_rc = rc; - rc = 0; - } + /* + * Attempt to flush data before changing attributes. We need to do + * this for ATTR_SIZE and ATTR_MTIME for sure, and if we change the + * ownership or mode then we may also need to do this. Here, we take + * the safe way out and just do the flush on all setattr requests. If + * the flush returns error, store it to report later and continue. + * + * BB: This should be smarter. Why bother flushing pages that + * will be truncated anyway? Also, should we error out here if + * the flush returns error? + */ + rc = filemap_write_and_wait(inode->i_mapping); + if (rc != 0) { + cifsInode->write_behind_rc = rc; + rc = 0; } if (attrs->ia_valid & ATTR_SIZE) { @@ -1903,20 +1909,21 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) return -ENOMEM; } - if ((attrs->ia_valid & ATTR_MTIME) || (attrs->ia_valid & ATTR_SIZE)) { - /* - Flush data before changing file size or changing the last - write time of the file on the server. If the - flush returns error, store it to report later and continue. - BB: This should be smarter. Why bother flushing pages that - will be truncated anyway? Also, should we error out here if - the flush returns error? - */ - rc = filemap_write_and_wait(inode->i_mapping); - if (rc != 0) { - cifsInode->write_behind_rc = rc; - rc = 0; - } + /* + * Attempt to flush data before changing attributes. We need to do + * this for ATTR_SIZE and ATTR_MTIME for sure, and if we change the + * ownership or mode then we may also need to do this. Here, we take + * the safe way out and just do the flush on all setattr requests. If + * the flush returns error, store it to report later and continue. + * + * BB: This should be smarter. Why bother flushing pages that + * will be truncated anyway? Also, should we error out here if + * the flush returns error? + */ + rc = filemap_write_and_wait(inode->i_mapping); + if (rc != 0) { + cifsInode->write_behind_rc = rc; + rc = 0; } if (attrs->ia_valid & ATTR_SIZE) { diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index c2c01ff4c32c..1a8be6228333 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -239,6 +239,7 @@ static void fill_in_inode(struct inode *tmp_inode, int new_buf_type, if (atomic_read(&cifsInfo->inUse) == 0) atomic_set(&cifsInfo->inUse, 1); + cifsInfo->server_eof = end_of_file; spin_lock(&tmp_inode->i_lock); if (is_size_safe_to_change(cifsInfo, end_of_file)) { /* can not safely change the file size here if the @@ -375,6 +376,7 @@ static void unix_fill_in_inode(struct inode *tmp_inode, tmp_inode->i_gid = le64_to_cpu(pfindData->Gid); tmp_inode->i_nlink = le64_to_cpu(pfindData->Nlinks); + cifsInfo->server_eof = end_of_file; spin_lock(&tmp_inode->i_lock); if (is_size_safe_to_change(cifsInfo, end_of_file)) { /* can not safely change the file size here if the @@ -840,7 +842,7 @@ static int cifs_get_name_from_search_buf(struct qstr *pqst, len = strnlen(filename, PATH_MAX); } - *pinum = pFindData->UniqueId; + *pinum = le64_to_cpu(pFindData->UniqueId); } else if (level == SMB_FIND_FILE_DIRECTORY_INFO) { FILE_DIRECTORY_INFO *pFindData = (FILE_DIRECTORY_INFO *)current_entry; @@ -856,7 +858,7 @@ static int cifs_get_name_from_search_buf(struct qstr *pqst, (SEARCH_ID_FULL_DIR_INFO *)current_entry; filename = &pFindData->FileName[0]; len = le32_to_cpu(pFindData->FileNameLength); - *pinum = pFindData->UniqueId; + *pinum = le64_to_cpu(pFindData->UniqueId); } else if (level == SMB_FIND_FILE_BOTH_DIRECTORY_INFO) { FILE_BOTH_DIRECTORY_INFO *pFindData = (FILE_BOTH_DIRECTORY_INFO *)current_entry; diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 5c68b4282be9..c652c73760dd 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -285,35 +285,36 @@ static int decode_unicode_ssetup(char **pbcc_area, int bleft, int words_left, len; char *data = *pbcc_area; - - cFYI(1, ("bleft %d", bleft)); - - /* SMB header is unaligned, so cifs servers word align start of - Unicode strings */ - data++; - bleft--; /* Windows servers do not always double null terminate - their final Unicode string - in which case we - now will not attempt to decode the byte of junk - which follows it */ + /* + * Windows servers do not always double null terminate their final + * Unicode string. Check to see if there are an uneven number of bytes + * left. If so, then add an extra NULL pad byte to the end of the + * response. + * + * See section 2.7.2 in "Implementing CIFS" for details + */ + if (bleft % 2) { + data[bleft] = 0; + ++bleft; + } words_left = bleft / 2; /* save off server operating system */ len = UniStrnlen((wchar_t *) data, words_left); -/* We look for obvious messed up bcc or strings in response so we do not go off - the end since (at least) WIN2K and Windows XP have a major bug in not null - terminating last Unicode string in response */ if (len >= words_left) return rc; kfree(ses->serverOS); /* UTF-8 string will not grow more than four times as big as UCS-16 */ ses->serverOS = kzalloc((4 * len) + 2 /* trailing null */, GFP_KERNEL); - if (ses->serverOS != NULL) + if (ses->serverOS != NULL) { cifs_strfromUCS_le(ses->serverOS, (__le16 *)data, len, nls_cp); + cFYI(1, ("serverOS=%s", ses->serverOS)); + } data += 2 * (len + 1); words_left -= len + 1; @@ -328,6 +329,7 @@ static int decode_unicode_ssetup(char **pbcc_area, int bleft, if (ses->serverNOS != NULL) { cifs_strfromUCS_le(ses->serverNOS, (__le16 *)data, len, nls_cp); + cFYI(1, ("serverNOS=%s", ses->serverNOS)); if (strncmp(ses->serverNOS, "NT LAN Manager 4", 16) == 0) { cFYI(1, ("NT4 server")); ses->flags |= CIFS_SES_NT4; @@ -343,12 +345,11 @@ static int decode_unicode_ssetup(char **pbcc_area, int bleft, return rc; kfree(ses->serverDomain); - ses->serverDomain = kzalloc(2 * (len + 1), GFP_KERNEL); /* BB FIXME wrong length */ + ses->serverDomain = kzalloc((4 * len) + 2, GFP_KERNEL); if (ses->serverDomain != NULL) { cifs_strfromUCS_le(ses->serverDomain, (__le16 *)data, len, nls_cp); - ses->serverDomain[2*len] = 0; - ses->serverDomain[(2*len) + 1] = 0; + cFYI(1, ("serverDomain=%s", ses->serverDomain)); } data += 2 * (len + 1); words_left -= len + 1; @@ -702,12 +703,18 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, } /* BB check if Unicode and decode strings */ - if (smb_buf->Flags2 & SMBFLG2_UNICODE) + if (smb_buf->Flags2 & SMBFLG2_UNICODE) { + /* unicode string area must be word-aligned */ + if (((unsigned long) bcc_ptr - (unsigned long) smb_buf) % 2) { + ++bcc_ptr; + --bytes_remaining; + } rc = decode_unicode_ssetup(&bcc_ptr, bytes_remaining, - ses, nls_cp); - else + ses, nls_cp); + } else { rc = decode_ascii_ssetup(&bcc_ptr, bytes_remaining, ses, nls_cp); + } ssetup_exit: if (spnego_key) { diff --git a/fs/compat.c b/fs/compat.c index 3f84d5f15889..379a399bf5c3 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -181,22 +181,24 @@ asmlinkage long compat_sys_newstat(char __user * filename, struct compat_stat __user *statbuf) { struct kstat stat; - int error = vfs_stat_fd(AT_FDCWD, filename, &stat); + int error; - if (!error) - error = cp_compat_stat(&stat, statbuf); - return error; + error = vfs_stat(filename, &stat); + if (error) + return error; + return cp_compat_stat(&stat, statbuf); } asmlinkage long compat_sys_newlstat(char __user * filename, struct compat_stat __user *statbuf) { struct kstat stat; - int error = vfs_lstat_fd(AT_FDCWD, filename, &stat); + int error; - if (!error) - error = cp_compat_stat(&stat, statbuf); - return error; + error = vfs_lstat(filename, &stat); + if (error) + return error; + return cp_compat_stat(&stat, statbuf); } #ifndef __ARCH_WANT_STAT64 @@ -204,21 +206,12 @@ asmlinkage long compat_sys_newfstatat(unsigned int dfd, char __user *filename, struct compat_stat __user *statbuf, int flag) { struct kstat stat; - int error = -EINVAL; - - if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) - goto out; - - if (flag & AT_SYMLINK_NOFOLLOW) - error = vfs_lstat_fd(dfd, filename, &stat); - else - error = vfs_stat_fd(dfd, filename, &stat); - - if (!error) - error = cp_compat_stat(&stat, statbuf); + int error; -out: - return error; + error = vfs_fstatat(dfd, filename, &stat, flag); + if (error) + return error; + return cp_compat_stat(&stat, statbuf); } #endif diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 3e87ce443ea2..b83f6bcfa51a 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -58,7 +58,6 @@ #include <linux/i2c.h> #include <linux/i2c-dev.h> #include <linux/atalk.h> -#include <linux/loop.h> #include <net/bluetooth/bluetooth.h> #include <net/bluetooth/hci.h> @@ -68,6 +67,7 @@ #include <linux/gigaset_dev.h> #ifdef CONFIG_BLOCK +#include <linux/loop.h> #include <scsi/scsi.h> #include <scsi/scsi_ioctl.h> #include <scsi/sg.h> @@ -2660,6 +2660,8 @@ HANDLE_IOCTL(SONET_GETFRAMING, do_atm_ioctl) HANDLE_IOCTL(SONET_GETFRSENSE, do_atm_ioctl) /* block stuff */ #ifdef CONFIG_BLOCK +/* loop */ +IGNORE_IOCTL(LOOP_CLR_FD) /* Raw devices */ HANDLE_IOCTL(RAW_SETBIND, raw_ioctl) HANDLE_IOCTL(RAW_GETBIND, raw_ioctl) @@ -2728,9 +2730,6 @@ HANDLE_IOCTL(LPSETTIMEOUT, lp_timeout_trans) IGNORE_IOCTL(VFAT_IOCTL_READDIR_BOTH32) IGNORE_IOCTL(VFAT_IOCTL_READDIR_SHORT32) -/* loop */ -IGNORE_IOCTL(LOOP_CLR_FD) - #ifdef CONFIG_SPARC /* Sparc framebuffers, handled in sbusfb_compat_ioctl() */ IGNORE_IOCTL(FBIOGTYPE) diff --git a/fs/configfs/symlink.c b/fs/configfs/symlink.c index 932a92b31483..c8afa6b1d91d 100644 --- a/fs/configfs/symlink.c +++ b/fs/configfs/symlink.c @@ -135,7 +135,7 @@ int configfs_symlink(struct inode *dir, struct dentry *dentry, const char *symna struct path path; struct configfs_dirent *sd; struct config_item *parent_item; - struct config_item *target_item; + struct config_item *target_item = NULL; struct config_item_type *type; ret = -EPERM; /* What lack-of-symlink returns */ diff --git a/fs/dcache.c b/fs/dcache.c index 761d30be2683..1fcffebfb44f 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -2149,7 +2149,6 @@ int is_subdir(struct dentry *new_dentry, struct dentry *old_dentry) int result; unsigned long seq; - /* FIXME: This is old behavior, needed? Please check callers. */ if (new_dentry == old_dentry) return 1; diff --git a/fs/direct-io.c b/fs/direct-io.c index da258e7249cc..05763bbc2050 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -307,8 +307,6 @@ dio_bio_alloc(struct dio *dio, struct block_device *bdev, struct bio *bio; bio = bio_alloc(GFP_KERNEL, nr_vecs); - if (bio == NULL) - return -ENOMEM; bio->bi_bdev = bdev; bio->bi_sector = first_sector; diff --git a/fs/ecryptfs/miscdev.c b/fs/ecryptfs/miscdev.c index a67fea655f49..dda3c58eefc0 100644 --- a/fs/ecryptfs/miscdev.c +++ b/fs/ecryptfs/miscdev.c @@ -418,18 +418,13 @@ ecryptfs_miscdev_write(struct file *file, const char __user *buf, if (count == 0) goto out; - data = kmalloc(count, GFP_KERNEL); - if (!data) { - printk(KERN_ERR "%s: Out of memory whilst attempting to " - "kmalloc([%zd], GFP_KERNEL)\n", __func__, count); + + data = memdup_user(buf, count); + if (IS_ERR(data)) { + printk(KERN_ERR "%s: memdup_user returned error [%ld]\n", + __func__, PTR_ERR(data)); goto out; } - rc = copy_from_user(data, buf, count); - if (rc) { - printk(KERN_ERR "%s: copy_from_user returned error [%d]\n", - __func__, rc); - goto out_free; - } sz = count; i = 0; switch (data[i++]) { diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index b43b95563663..acf678831103 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -590,9 +590,8 @@ static int ext2_get_blocks(struct inode *inode, if (depth == 0) return (err); -reread: - partial = ext2_get_branch(inode, depth, offsets, chain, &err); + partial = ext2_get_branch(inode, depth, offsets, chain, &err); /* Simplest case - block found, no allocation needed */ if (!partial) { first_block = le32_to_cpu(chain[depth - 1].key); @@ -602,15 +601,16 @@ reread: while (count < maxblocks && count <= blocks_to_boundary) { ext2_fsblk_t blk; - if (!verify_chain(chain, partial)) { + if (!verify_chain(chain, chain + depth - 1)) { /* * Indirect block might be removed by * truncate while we were reading it. * Handling of that case: forget what we've * got now, go to reread. */ + err = -EAGAIN; count = 0; - goto changed; + break; } blk = le32_to_cpu(*(chain[depth-1].p + count)); if (blk == first_block + count) @@ -618,7 +618,8 @@ reread: else break; } - goto got_it; + if (err != -EAGAIN) + goto got_it; } /* Next simple case - plain lookup or failed read of indirect block */ @@ -626,6 +627,33 @@ reread: goto cleanup; mutex_lock(&ei->truncate_mutex); + /* + * If the indirect block is missing while we are reading + * the chain(ext3_get_branch() returns -EAGAIN err), or + * if the chain has been changed after we grab the semaphore, + * (either because another process truncated this branch, or + * another get_block allocated this branch) re-grab the chain to see if + * the request block has been allocated or not. + * + * Since we already block the truncate/other get_block + * at this point, we will have the current copy of the chain when we + * splice the branch into the tree. + */ + if (err == -EAGAIN || !verify_chain(chain, partial)) { + while (partial > chain) { + brelse(partial->bh); + partial--; + } + partial = ext2_get_branch(inode, depth, offsets, chain, &err); + if (!partial) { + count++; + mutex_unlock(&ei->truncate_mutex); + if (err) + goto cleanup; + clear_buffer_new(bh_result); + goto got_it; + } + } /* * Okay, we need to do block allocation. Lazily initialize the block @@ -683,12 +711,6 @@ cleanup: partial--; } return err; -changed: - while (partial > chain) { - brelse(partial->bh); - partial--; - } - goto reread; } int ext2_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 466a332e0bd1..fcfa24361856 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -1521,12 +1521,16 @@ static int ext3_ordered_writepage(struct page *page, if (!page_has_buffers(page)) { create_empty_buffers(page, inode->i_sb->s_blocksize, (1 << BH_Dirty)|(1 << BH_Uptodate)); - } else if (!walk_page_buffers(NULL, page_buffers(page), 0, PAGE_CACHE_SIZE, NULL, buffer_unmapped)) { - /* Provide NULL instead of get_block so that we catch bugs if buffers weren't really mapped */ - return block_write_full_page(page, NULL, wbc); + page_bufs = page_buffers(page); + } else { + page_bufs = page_buffers(page); + if (!walk_page_buffers(NULL, page_bufs, 0, PAGE_CACHE_SIZE, + NULL, buffer_unmapped)) { + /* Provide NULL get_block() to catch bugs if buffers + * weren't really mapped */ + return block_write_full_page(page, NULL, wbc); + } } - page_bufs = page_buffers(page); - handle = ext3_journal_start(inode, ext3_writepage_trans_blocks(inode)); if (IS_ERR(handle)) { @@ -1581,6 +1585,15 @@ static int ext3_writeback_writepage(struct page *page, if (ext3_journal_current_handle()) goto out_fail; + if (page_has_buffers(page)) { + if (!walk_page_buffers(NULL, page_buffers(page), 0, + PAGE_CACHE_SIZE, NULL, buffer_unmapped)) { + /* Provide NULL get_block() to catch bugs if buffers + * weren't really mapped */ + return block_write_full_page(page, NULL, wbc); + } + } + handle = ext3_journal_start(inode, ext3_writepage_trans_blocks(inode)); if (IS_ERR(handle)) { ret = PTR_ERR(handle); diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index ac77d8b8251d..e40332158340 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -326,11 +326,14 @@ ext4_ext_max_entries(struct inode *inode, int depth) static int ext4_valid_extent(struct inode *inode, struct ext4_extent *ext) { - ext4_fsblk_t block = ext_pblock(ext); + ext4_fsblk_t block = ext_pblock(ext), valid_block; int len = ext4_ext_get_actual_len(ext); struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es; - if (unlikely(block < le32_to_cpu(es->s_first_data_block) || - ((block + len) > ext4_blocks_count(es)))) + + valid_block = le32_to_cpu(es->s_first_data_block) + + EXT4_SB(inode->i_sb)->s_gdb_count; + if (unlikely(block <= valid_block || + ((block + len) > ext4_blocks_count(es)))) return 0; else return 1; @@ -339,10 +342,13 @@ static int ext4_valid_extent(struct inode *inode, struct ext4_extent *ext) static int ext4_valid_extent_idx(struct inode *inode, struct ext4_extent_idx *ext_idx) { - ext4_fsblk_t block = idx_pblock(ext_idx); + ext4_fsblk_t block = idx_pblock(ext_idx), valid_block; struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es; - if (unlikely(block < le32_to_cpu(es->s_first_data_block) || - (block > ext4_blocks_count(es)))) + + valid_block = le32_to_cpu(es->s_first_data_block) + + EXT4_SB(inode->i_sb)->s_gdb_count; + if (unlikely(block <= valid_block || + (block >= ext4_blocks_count(es)))) return 0; else return 1; @@ -2416,8 +2422,6 @@ static int ext4_ext_zeroout(struct inode *inode, struct ext4_extent *ex) len = ee_len; bio = bio_alloc(GFP_NOIO, len); - if (!bio) - return -ENOMEM; bio->bi_sector = ee_pblock; bio->bi_bdev = inode->i_sb->s_bdev; diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index 47b84e8df568..f18e0a08a6b5 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -585,6 +585,7 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent, fallback: ngroups = sbi->s_groups_count; avefreei = freei / ngroups; +fallback_retry: parent_group = EXT4_I(parent)->i_block_group; for (i = 0; i < ngroups; i++) { grp = (parent_group + i) % ngroups; @@ -602,7 +603,7 @@ fallback: * filesystems the above test can fail to find any blockgroups */ avefreei = 0; - goto fallback; + goto fallback_retry; } return -1; @@ -831,11 +832,12 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, int mode) ret2 = find_group_flex(sb, dir, &group); if (ret2 == -1) { ret2 = find_group_other(sb, dir, &group, mode); - if (ret2 == 0 && once) + if (ret2 == 0 && once) { once = 0; printk(KERN_NOTICE "ext4: find_group_flex " "failed, fallback succeeded dir %lu\n", dir->i_ino); + } } goto got_group; } diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index a2e7952bc5f9..c6bd6ced3bb7 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -372,16 +372,16 @@ static int ext4_block_to_path(struct inode *inode, } static int __ext4_check_blockref(const char *function, struct inode *inode, - unsigned int *p, unsigned int max) { + __le32 *p, unsigned int max) { unsigned int maxblocks = ext4_blocks_count(EXT4_SB(inode->i_sb)->s_es); - unsigned int *bref = p; + __le32 *bref = p; while (bref < p+max) { - if (unlikely(*bref >= maxblocks)) { + if (unlikely(le32_to_cpu(*bref) >= maxblocks)) { ext4_error(inode->i_sb, function, "block reference %u >= max (%u) " "in inode #%lu, offset=%d", - *bref, maxblocks, + le32_to_cpu(*bref), maxblocks, inode->i_ino, (int)(bref-p)); return -EIO; } diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 9987bba99db3..2958f4e6f222 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -2508,6 +2508,15 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) if (EXT4_BLOCKS_PER_GROUP(sb) == 0) goto cantfind_ext4; + /* check blocks count against device size */ + blocks_count = sb->s_bdev->bd_inode->i_size >> sb->s_blocksize_bits; + if (blocks_count && ext4_blocks_count(es) > blocks_count) { + printk(KERN_WARNING "EXT4-fs: bad geometry: block count %llu " + "exceeds size of device (%llu blocks)\n", + ext4_blocks_count(es), blocks_count); + goto failed_mount; + } + /* * It makes no sense for the first data block to be beyond the end * of the filesystem. diff --git a/fs/fat/Kconfig b/fs/fat/Kconfig index d0a69ff25375..182f9ffe2b51 100644 --- a/fs/fat/Kconfig +++ b/fs/fat/Kconfig @@ -95,3 +95,6 @@ config FAT_DEFAULT_IOCHARSET Note that "utf8" is not recommended for FAT filesystems. If unsure, you shouldn't set "utf8" here. See <file:Documentation/filesystems/vfat.txt> for more information. + + Enable any character sets you need in File Systems/Native Language + Support. diff --git a/fs/filesystems.c b/fs/filesystems.c index 1aa70260e6d1..a24c58e181db 100644 --- a/fs/filesystems.c +++ b/fs/filesystems.c @@ -199,7 +199,7 @@ SYSCALL_DEFINE3(sysfs, int, option, unsigned long, arg1, unsigned long, arg2) return retval; } -int get_filesystem_list(char * buf) +int __init get_filesystem_list(char *buf) { int len = 0; struct file_system_type * tmp; diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 2b25133524a3..06f30e965676 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -938,9 +938,9 @@ static void fuse_release_user_pages(struct fuse_req *req, int write) } static int fuse_get_user_pages(struct fuse_req *req, const char __user *buf, - unsigned *nbytesp, int write) + size_t *nbytesp, int write) { - unsigned nbytes = *nbytesp; + size_t nbytes = *nbytesp; unsigned long user_addr = (unsigned long) buf; unsigned offset = user_addr & ~PAGE_MASK; int npages; @@ -955,7 +955,7 @@ static int fuse_get_user_pages(struct fuse_req *req, const char __user *buf, return 0; } - nbytes = min(nbytes, (unsigned) FUSE_MAX_PAGES_PER_REQ << PAGE_SHIFT); + nbytes = min_t(size_t, nbytes, FUSE_MAX_PAGES_PER_REQ << PAGE_SHIFT); npages = (nbytes + offset + PAGE_SIZE - 1) >> PAGE_SHIFT; npages = clamp(npages, 1, FUSE_MAX_PAGES_PER_REQ); down_read(¤t->mm->mmap_sem); @@ -1298,6 +1298,8 @@ static int fuse_direct_mmap(struct file *file, struct vm_area_struct *vma) if (vma->vm_flags & VM_MAYSHARE) return -ENODEV; + invalidate_inode_pages2(file->f_mapping); + return generic_file_mmap(file, vma); } diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 3984e47d1d33..1afd9f26bcb1 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -597,7 +597,6 @@ __acquires(&gl->gl_spin) GLOCK_BUG_ON(gl, test_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags)); - down_read(&gfs2_umount_flush_sem); if (test_bit(GLF_DEMOTE, &gl->gl_flags) && gl->gl_demote_state != gl->gl_state) { if (find_first_holder(gl)) @@ -614,15 +613,14 @@ __acquires(&gl->gl_spin) if (ret == 0) goto out_unlock; if (ret == 2) - goto out_sem; + goto out; gh = find_first_waiter(gl); gl->gl_target = gh->gh_state; if (!(gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))) do_error(gl, 0); /* Fail queued try locks */ } do_xmote(gl, gh, gl->gl_target); -out_sem: - up_read(&gfs2_umount_flush_sem); +out: return; out_sched: @@ -631,7 +629,7 @@ out_sched: gfs2_glock_put(gl); out_unlock: clear_bit(GLF_LOCK, &gl->gl_flags); - goto out_sem; + goto out; } static void glock_work_func(struct work_struct *work) @@ -641,6 +639,7 @@ static void glock_work_func(struct work_struct *work) if (test_and_clear_bit(GLF_REPLY_PENDING, &gl->gl_flags)) finish_xmote(gl, gl->gl_reply); + down_read(&gfs2_umount_flush_sem); spin_lock(&gl->gl_spin); if (test_and_clear_bit(GLF_PENDING_DEMOTE, &gl->gl_flags) && gl->gl_state != LM_ST_UNLOCKED && @@ -653,6 +652,7 @@ static void glock_work_func(struct work_struct *work) } run_queue(gl, 0); spin_unlock(&gl->gl_spin); + up_read(&gfs2_umount_flush_sem); if (!delay || queue_delayed_work(glock_workqueue, &gl->gl_work, delay) == 0) gfs2_glock_put(gl); diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index bf23a62aa925..70f87f43afa2 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -156,6 +156,12 @@ static void inode_go_sync(struct gfs2_glock *gl) error = filemap_fdatawait(metamapping); mapping_set_error(metamapping, error); gfs2_ail_empty_gl(gl); + /* + * Writeback of the data mapping may cause the dirty flag to be set + * so we have to clear it again here. + */ + smp_mb__before_clear_bit(); + clear_bit(GLF_DIRTY, &gl->gl_flags); } /** diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 7b277d449155..5a31d426116f 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -137,15 +137,15 @@ void gfs2_set_iop(struct inode *inode) if (S_ISREG(mode)) { inode->i_op = &gfs2_file_iops; if (gfs2_localflocks(sdp)) - inode->i_fop = gfs2_file_fops_nolock; + inode->i_fop = &gfs2_file_fops_nolock; else - inode->i_fop = gfs2_file_fops; + inode->i_fop = &gfs2_file_fops; } else if (S_ISDIR(mode)) { inode->i_op = &gfs2_dir_iops; if (gfs2_localflocks(sdp)) - inode->i_fop = gfs2_dir_fops_nolock; + inode->i_fop = &gfs2_dir_fops_nolock; else - inode->i_fop = gfs2_dir_fops; + inode->i_fop = &gfs2_dir_fops; } else if (S_ISLNK(mode)) { inode->i_op = &gfs2_symlink_iops; } else { diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index dca4fee3078b..c30be2b66580 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -101,21 +101,23 @@ void gfs2_dinode_print(const struct gfs2_inode *ip); extern const struct inode_operations gfs2_file_iops; extern const struct inode_operations gfs2_dir_iops; extern const struct inode_operations gfs2_symlink_iops; -extern const struct file_operations *gfs2_file_fops_nolock; -extern const struct file_operations *gfs2_dir_fops_nolock; +extern const struct file_operations gfs2_file_fops_nolock; +extern const struct file_operations gfs2_dir_fops_nolock; extern void gfs2_set_inode_flags(struct inode *inode); #ifdef CONFIG_GFS2_FS_LOCKING_DLM -extern const struct file_operations *gfs2_file_fops; -extern const struct file_operations *gfs2_dir_fops; +extern const struct file_operations gfs2_file_fops; +extern const struct file_operations gfs2_dir_fops; + static inline int gfs2_localflocks(const struct gfs2_sbd *sdp) { return sdp->sd_args.ar_localflocks; } #else /* Single node only */ -#define gfs2_file_fops NULL -#define gfs2_dir_fops NULL +#define gfs2_file_fops gfs2_file_fops_nolock +#define gfs2_dir_fops gfs2_dir_fops_nolock + static inline int gfs2_localflocks(const struct gfs2_sbd *sdp) { return 1; diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 98918a756410..aa62cf5976e8 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -120,7 +120,7 @@ __acquires(&sdp->sd_log_lock) lock_buffer(bh); if (test_clear_buffer_dirty(bh)) { bh->b_end_io = end_buffer_write_sync; - submit_bh(WRITE, bh); + submit_bh(WRITE_SYNC_PLUG, bh); } else { unlock_buffer(bh); brelse(bh); @@ -604,7 +604,7 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags, int pull) if (test_bit(SDF_NOBARRIERS, &sdp->sd_flags)) goto skip_barrier; get_bh(bh); - submit_bh(WRITE_BARRIER | (1 << BIO_RW_META), bh); + submit_bh(WRITE_SYNC | (1 << BIO_RW_BARRIER) | (1 << BIO_RW_META), bh); wait_on_buffer(bh); if (buffer_eopnotsupp(bh)) { clear_buffer_eopnotsupp(bh); @@ -664,7 +664,7 @@ static void gfs2_ordered_write(struct gfs2_sbd *sdp) lock_buffer(bh); if (buffer_mapped(bh) && test_clear_buffer_dirty(bh)) { bh->b_end_io = end_buffer_write_sync; - submit_bh(WRITE, bh); + submit_bh(WRITE_SYNC_PLUG, bh); } else { unlock_buffer(bh); brelse(bh); diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index 80e4f5f898bb..00315f50fa46 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -13,6 +13,8 @@ #include <linux/completion.h> #include <linux/buffer_head.h> #include <linux/gfs2_ondisk.h> +#include <linux/bio.h> +#include <linux/fs.h> #include "gfs2.h" #include "incore.h" @@ -189,7 +191,7 @@ static void buf_lo_before_commit(struct gfs2_sbd *sdp) } gfs2_log_unlock(sdp); - submit_bh(WRITE, bh); + submit_bh(WRITE_SYNC_PLUG, bh); gfs2_log_lock(sdp); n = 0; @@ -199,7 +201,7 @@ static void buf_lo_before_commit(struct gfs2_sbd *sdp) gfs2_log_unlock(sdp); lock_buffer(bd2->bd_bh); bh = gfs2_log_fake_buf(sdp, bd2->bd_bh); - submit_bh(WRITE, bh); + submit_bh(WRITE_SYNC_PLUG, bh); gfs2_log_lock(sdp); if (++n >= num) break; @@ -341,7 +343,7 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp) sdp->sd_log_num_revoke--; if (offset + sizeof(u64) > sdp->sd_sb.sb_bsize) { - submit_bh(WRITE, bh); + submit_bh(WRITE_SYNC_PLUG, bh); bh = gfs2_log_get_buf(sdp); mh = (struct gfs2_meta_header *)bh->b_data; @@ -358,7 +360,7 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp) } gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke); - submit_bh(WRITE, bh); + submit_bh(WRITE_SYNC_PLUG, bh); } static void revoke_lo_before_scan(struct gfs2_jdesc *jd, @@ -560,7 +562,7 @@ static void gfs2_write_blocks(struct gfs2_sbd *sdp, struct buffer_head *bh, ptr = bh_log_ptr(bh); get_bh(bh); - submit_bh(WRITE, bh); + submit_bh(WRITE_SYNC_PLUG, bh); gfs2_log_lock(sdp); while(!list_empty(list)) { bd = list_entry(list->next, struct gfs2_bufdata, bd_le.le_list); @@ -586,7 +588,7 @@ static void gfs2_write_blocks(struct gfs2_sbd *sdp, struct buffer_head *bh, } else { bh1 = gfs2_log_fake_buf(sdp, bd->bd_bh); } - submit_bh(WRITE, bh1); + submit_bh(WRITE_SYNC_PLUG, bh1); gfs2_log_lock(sdp); ptr += 2; } diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c index 8d6f13256b26..78a5f4312667 100644 --- a/fs/gfs2/meta_io.c +++ b/fs/gfs2/meta_io.c @@ -33,17 +33,65 @@ #include "util.h" #include "ops_address.h" -static int aspace_get_block(struct inode *inode, sector_t lblock, - struct buffer_head *bh_result, int create) +static int gfs2_aspace_writepage(struct page *page, struct writeback_control *wbc) { - gfs2_assert_warn(inode->i_sb->s_fs_info, 0); - return -EOPNOTSUPP; -} + int err; + struct buffer_head *bh, *head; + int nr_underway = 0; + int write_op = (1 << BIO_RW_META) | ((wbc->sync_mode == WB_SYNC_ALL ? + WRITE_SYNC_PLUG : WRITE)); + + BUG_ON(!PageLocked(page)); + BUG_ON(!page_has_buffers(page)); + + head = page_buffers(page); + bh = head; + + do { + if (!buffer_mapped(bh)) + continue; + /* + * If it's a fully non-blocking write attempt and we cannot + * lock the buffer then redirty the page. Note that this can + * potentially cause a busy-wait loop from pdflush and kswapd + * activity, but those code paths have their own higher-level + * throttling. + */ + if (wbc->sync_mode != WB_SYNC_NONE || !wbc->nonblocking) { + lock_buffer(bh); + } else if (!trylock_buffer(bh)) { + redirty_page_for_writepage(wbc, page); + continue; + } + if (test_clear_buffer_dirty(bh)) { + mark_buffer_async_write(bh); + } else { + unlock_buffer(bh); + } + } while ((bh = bh->b_this_page) != head); + + /* + * The page and its buffers are protected by PageWriteback(), so we can + * drop the bh refcounts early. + */ + BUG_ON(PageWriteback(page)); + set_page_writeback(page); + + do { + struct buffer_head *next = bh->b_this_page; + if (buffer_async_write(bh)) { + submit_bh(write_op, bh); + nr_underway++; + } + bh = next; + } while (bh != head); + unlock_page(page); -static int gfs2_aspace_writepage(struct page *page, - struct writeback_control *wbc) -{ - return block_write_full_page(page, aspace_get_block, wbc); + err = 0; + if (nr_underway == 0) + end_page_writeback(page); + + return err; } static const struct address_space_operations aspace_aops = { @@ -201,16 +249,32 @@ struct buffer_head *gfs2_meta_new(struct gfs2_glock *gl, u64 blkno) int gfs2_meta_read(struct gfs2_glock *gl, u64 blkno, int flags, struct buffer_head **bhp) { - *bhp = gfs2_getbuf(gl, blkno, CREATE); - if (!buffer_uptodate(*bhp)) { - ll_rw_block(READ_META, 1, bhp); - if (flags & DIO_WAIT) { - int error = gfs2_meta_wait(gl->gl_sbd, *bhp); - if (error) { - brelse(*bhp); - return error; - } - } + struct gfs2_sbd *sdp = gl->gl_sbd; + struct buffer_head *bh; + + if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags))) + return -EIO; + + *bhp = bh = gfs2_getbuf(gl, blkno, CREATE); + + lock_buffer(bh); + if (buffer_uptodate(bh)) { + unlock_buffer(bh); + return 0; + } + bh->b_end_io = end_buffer_read_sync; + get_bh(bh); + submit_bh(READ_SYNC | (1 << BIO_RW_META), bh); + if (!(flags & DIO_WAIT)) + return 0; + + wait_on_buffer(bh); + if (unlikely(!buffer_uptodate(bh))) { + struct gfs2_trans *tr = current->journal_info; + if (tr && tr->tr_touched) + gfs2_io_error_bh(sdp, bh); + brelse(bh); + return -EIO; } return 0; @@ -404,7 +468,7 @@ struct buffer_head *gfs2_meta_ra(struct gfs2_glock *gl, u64 dblock, u32 extlen) if (buffer_uptodate(first_bh)) goto out; if (!buffer_locked(first_bh)) - ll_rw_block(READ_META, 1, &first_bh); + ll_rw_block(READ_SYNC | (1 << BIO_RW_META), 1, &first_bh); dblock++; extlen--; diff --git a/fs/gfs2/ops_file.c b/fs/gfs2/ops_file.c index 70b9b8548945..5d82e91887e3 100644 --- a/fs/gfs2/ops_file.c +++ b/fs/gfs2/ops_file.c @@ -413,7 +413,9 @@ out_unlock: gfs2_glock_dq(&gh); out: gfs2_holder_uninit(&gh); - if (ret) + if (ret == -ENOMEM) + ret = VM_FAULT_OOM; + else if (ret) ret = VM_FAULT_SIGBUS; return ret; } @@ -705,7 +707,7 @@ static int gfs2_flock(struct file *file, int cmd, struct file_lock *fl) } } -const struct file_operations *gfs2_file_fops = &(const struct file_operations){ +const struct file_operations gfs2_file_fops = { .llseek = gfs2_llseek, .read = do_sync_read, .aio_read = generic_file_aio_read, @@ -723,7 +725,7 @@ const struct file_operations *gfs2_file_fops = &(const struct file_operations){ .setlease = gfs2_setlease, }; -const struct file_operations *gfs2_dir_fops = &(const struct file_operations){ +const struct file_operations gfs2_dir_fops = { .readdir = gfs2_readdir, .unlocked_ioctl = gfs2_ioctl, .open = gfs2_open, @@ -735,7 +737,7 @@ const struct file_operations *gfs2_dir_fops = &(const struct file_operations){ #endif /* CONFIG_GFS2_FS_LOCKING_DLM */ -const struct file_operations *gfs2_file_fops_nolock = &(const struct file_operations){ +const struct file_operations gfs2_file_fops_nolock = { .llseek = gfs2_llseek, .read = do_sync_read, .aio_read = generic_file_aio_read, @@ -751,7 +753,7 @@ const struct file_operations *gfs2_file_fops_nolock = &(const struct file_operat .setlease = generic_setlease, }; -const struct file_operations *gfs2_dir_fops_nolock = &(const struct file_operations){ +const struct file_operations gfs2_dir_fops_nolock = { .readdir = gfs2_readdir, .unlocked_ioctl = gfs2_ioctl, .open = gfs2_open, diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 51883b3ad89c..650a730707b7 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -272,11 +272,6 @@ static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector) lock_page(page); bio = bio_alloc(GFP_NOFS, 1); - if (unlikely(!bio)) { - __free_page(page); - return -ENOBUFS; - } - bio->bi_sector = sector * (sb->s_blocksize >> 9); bio->bi_bdev = sb->s_bdev; bio_add_page(bio, page, PAGE_SIZE, 0); diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index abd5429ae285..1c70fa5168d6 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -371,6 +371,7 @@ static int gfs2_symlink(struct inode *dir, struct dentry *dentry, ip = ghs[1].gh_gl->gl_object; ip->i_disksize = size; + i_size_write(inode, size); error = gfs2_meta_inode_buffer(ip, &dibh); diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index 8d53f66b5bcc..152e6c4a0dca 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -81,7 +81,7 @@ struct gfs2_quota_change_host { static LIST_HEAD(qd_lru_list); static atomic_t qd_lru_count = ATOMIC_INIT(0); -static spinlock_t qd_lru_lock = SPIN_LOCK_UNLOCKED; +static DEFINE_SPINLOCK(qd_lru_lock); int gfs2_shrink_qd_memory(int nr, gfp_t gfp_mask) { @@ -1364,7 +1364,7 @@ int gfs2_quotad(void *data) refrigerator(); t = min(quotad_timeo, statfs_timeo); - prepare_to_wait(&sdp->sd_quota_wait, &wait, TASK_UNINTERRUPTIBLE); + prepare_to_wait(&sdp->sd_quota_wait, &wait, TASK_INTERRUPTIBLE); spin_lock(&sdp->sd_trunc_lock); empty = list_empty(&sdp->sd_trunc_list); spin_unlock(&sdp->sd_trunc_lock); diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index f03d024038ea..565038243fa2 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -212,8 +212,7 @@ static u32 gfs2_bitfit(const u8 *buf, const unsigned int len, if (tmp == 0) return BFITNOENT; ptr--; - bit = fls64(tmp); - bit--; /* fls64 always adds one to the bit count */ + bit = __ffs64(tmp); bit /= 2; /* two bits per entry in the bitmap */ return (((const unsigned char *)ptr - buf) * GFS2_NBBY) + bit; } @@ -1445,10 +1444,12 @@ static struct gfs2_rgrpd *rgblk_free(struct gfs2_sbd *sdp, u64 bstart, u64 gfs2_alloc_block(struct gfs2_inode *ip, unsigned int *n) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); + struct buffer_head *dibh; struct gfs2_alloc *al = ip->i_alloc; struct gfs2_rgrpd *rgd = al->al_rgd; u32 goal, blk; u64 block; + int error; if (rgrp_contains_block(rgd, ip->i_goal)) goal = ip->i_goal - rgd->rd_data0; @@ -1461,7 +1462,13 @@ u64 gfs2_alloc_block(struct gfs2_inode *ip, unsigned int *n) rgd->rd_last_alloc = blk; block = rgd->rd_data0 + blk; ip->i_goal = block; - + error = gfs2_meta_inode_buffer(ip, &dibh); + if (error == 0) { + struct gfs2_dinode *di = (struct gfs2_dinode *)dibh->b_data; + gfs2_trans_add_bh(ip->i_gl, dibh, 1); + di->di_goal_meta = di->di_goal_data = cpu_to_be64(ip->i_goal); + brelse(dibh); + } gfs2_assert_withdraw(sdp, rgd->rd_free >= *n); rgd->rd_free -= *n; diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index 9435dda8f1e0..a1cbff2b4d99 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -70,6 +70,10 @@ static int hfs_releasepage(struct page *page, gfp_t mask) BUG(); return 0; } + + if (!tree) + return 0; + if (tree->node_size >= PAGE_CACHE_SIZE) { nidx = page->index >> (tree->node_size_shift - PAGE_CACHE_SHIFT); spin_lock(&tree->hash_lock); diff --git a/fs/hfs/mdb.c b/fs/hfs/mdb.c index 36ca2e1a4fa3..7b6165f25fbe 100644 --- a/fs/hfs/mdb.c +++ b/fs/hfs/mdb.c @@ -349,6 +349,7 @@ void hfs_mdb_put(struct super_block *sb) if (HFS_SB(sb)->nls_disk) unload_nls(HFS_SB(sb)->nls_disk); + free_pages((unsigned long)HFS_SB(sb)->bitmap, PAGE_SIZE < 8192 ? 1 : 0); kfree(HFS_SB(sb)); sb->s_fs_info = NULL; } diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 23a3c76711e0..153d9681192b 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -26,7 +26,6 @@ #include <linux/pagevec.h> #include <linux/parser.h> #include <linux/mman.h> -#include <linux/quotaops.h> #include <linux/slab.h> #include <linux/dnotify.h> #include <linux/statfs.h> @@ -842,7 +841,7 @@ hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig) bad_val: printk(KERN_ERR "hugetlbfs: Bad value '%s' for mount option '%s'\n", args[0].from, p); - return 1; + return -EINVAL; } static int diff --git a/fs/inode.c b/fs/inode.c index d06d6d268de9..6ad14a1cd8c9 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1470,42 +1470,6 @@ static void __wait_on_freeing_inode(struct inode *inode) spin_lock(&inode_lock); } -/* - * We rarely want to lock two inodes that do not have a parent/child - * relationship (such as directory, child inode) simultaneously. The - * vast majority of file systems should be able to get along fine - * without this. Do not use these functions except as a last resort. - */ -void inode_double_lock(struct inode *inode1, struct inode *inode2) -{ - if (inode1 == NULL || inode2 == NULL || inode1 == inode2) { - if (inode1) - mutex_lock(&inode1->i_mutex); - else if (inode2) - mutex_lock(&inode2->i_mutex); - return; - } - - if (inode1 < inode2) { - mutex_lock_nested(&inode1->i_mutex, I_MUTEX_PARENT); - mutex_lock_nested(&inode2->i_mutex, I_MUTEX_CHILD); - } else { - mutex_lock_nested(&inode2->i_mutex, I_MUTEX_PARENT); - mutex_lock_nested(&inode1->i_mutex, I_MUTEX_CHILD); - } -} -EXPORT_SYMBOL(inode_double_lock); - -void inode_double_unlock(struct inode *inode1, struct inode *inode2) -{ - if (inode1) - mutex_unlock(&inode1->i_mutex); - - if (inode2 && inode2 != inode1) - mutex_unlock(&inode2->i_mutex); -} -EXPORT_SYMBOL(inode_double_unlock); - static __initdata unsigned long ihash_entries; static int __init set_ihash_entries(char *str) { diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c index a8e8513a78a9..06560c520f49 100644 --- a/fs/jbd/commit.c +++ b/fs/jbd/commit.c @@ -502,7 +502,7 @@ void journal_commit_transaction(journal_t *journal) err = 0; } - journal_write_revoke_records(journal, commit_transaction); + journal_write_revoke_records(journal, commit_transaction, write_op); /* * If we found any dirty or locked buffers, then we should have diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c index c7bd649bbbdc..da6cd9bdaabc 100644 --- a/fs/jbd/revoke.c +++ b/fs/jbd/revoke.c @@ -55,6 +55,25 @@ * need do nothing. * RevokeValid set, Revoked set: * buffer has been revoked. + * + * Locking rules: + * We keep two hash tables of revoke records. One hashtable belongs to the + * running transaction (is pointed to by journal->j_revoke), the other one + * belongs to the committing transaction. Accesses to the second hash table + * happen only from the kjournald and no other thread touches this table. Also + * journal_switch_revoke_table() which switches which hashtable belongs to the + * running and which to the committing transaction is called only from + * kjournald. Therefore we need no locks when accessing the hashtable belonging + * to the committing transaction. + * + * All users operating on the hash table belonging to the running transaction + * have a handle to the transaction. Therefore they are safe from kjournald + * switching hash tables under them. For operations on the lists of entries in + * the hash table j_revoke_lock is used. + * + * Finally, also replay code uses the hash tables but at this moment noone else + * can touch them (filesystem isn't mounted yet) and hence no locking is + * needed. */ #ifndef __KERNEL__ @@ -67,6 +86,7 @@ #include <linux/slab.h> #include <linux/list.h> #include <linux/init.h> +#include <linux/bio.h> #endif #include <linux/log2.h> @@ -99,8 +119,8 @@ struct jbd_revoke_table_s #ifdef __KERNEL__ static void write_one_revoke_record(journal_t *, transaction_t *, struct journal_head **, int *, - struct jbd_revoke_record_s *); -static void flush_descriptor(journal_t *, struct journal_head *, int); + struct jbd_revoke_record_s *, int); +static void flush_descriptor(journal_t *, struct journal_head *, int, int); #endif /* Utility functions to maintain the revoke table */ @@ -402,8 +422,6 @@ int journal_revoke(handle_t *handle, unsigned long blocknr, * the second time we would still have a pending revoke to cancel. So, * do not trust the Revoked bit on buffers unless RevokeValid is also * set. - * - * The caller must have the journal locked. */ int journal_cancel_revoke(handle_t *handle, struct journal_head *jh) { @@ -481,12 +499,9 @@ void journal_switch_revoke_table(journal_t *journal) /* * Write revoke records to the journal for all entries in the current * revoke hash, deleting the entries as we go. - * - * Called with the journal lock held. */ - void journal_write_revoke_records(journal_t *journal, - transaction_t *transaction) + transaction_t *transaction, int write_op) { struct journal_head *descriptor; struct jbd_revoke_record_s *record; @@ -510,14 +525,14 @@ void journal_write_revoke_records(journal_t *journal, hash_list->next; write_one_revoke_record(journal, transaction, &descriptor, &offset, - record); + record, write_op); count++; list_del(&record->hash); kmem_cache_free(revoke_record_cache, record); } } if (descriptor) - flush_descriptor(journal, descriptor, offset); + flush_descriptor(journal, descriptor, offset, write_op); jbd_debug(1, "Wrote %d revoke records\n", count); } @@ -530,7 +545,8 @@ static void write_one_revoke_record(journal_t *journal, transaction_t *transaction, struct journal_head **descriptorp, int *offsetp, - struct jbd_revoke_record_s *record) + struct jbd_revoke_record_s *record, + int write_op) { struct journal_head *descriptor; int offset; @@ -549,7 +565,7 @@ static void write_one_revoke_record(journal_t *journal, /* Make sure we have a descriptor with space left for the record */ if (descriptor) { if (offset == journal->j_blocksize) { - flush_descriptor(journal, descriptor, offset); + flush_descriptor(journal, descriptor, offset, write_op); descriptor = NULL; } } @@ -586,7 +602,7 @@ static void write_one_revoke_record(journal_t *journal, static void flush_descriptor(journal_t *journal, struct journal_head *descriptor, - int offset) + int offset, int write_op) { journal_revoke_header_t *header; struct buffer_head *bh = jh2bh(descriptor); @@ -601,7 +617,7 @@ static void flush_descriptor(journal_t *journal, set_buffer_jwrite(bh); BUFFER_TRACE(bh, "write"); set_buffer_dirty(bh); - ll_rw_block(SWRITE, 1, &bh); + ll_rw_block((write_op == WRITE) ? SWRITE : SWRITE_SYNC_PLUG, 1, &bh); } #endif diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 073c8c3df7cd..0b7d3b8226fd 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -506,7 +506,8 @@ void jbd2_journal_commit_transaction(journal_t *journal) if (err) jbd2_journal_abort(journal, err); - jbd2_journal_write_revoke_records(journal, commit_transaction); + jbd2_journal_write_revoke_records(journal, commit_transaction, + write_op); jbd_debug(3, "JBD: commit phase 2\n"); diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index bbe6d592d8b3..a360b06af2e3 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -86,6 +86,7 @@ #include <linux/slab.h> #include <linux/list.h> #include <linux/init.h> +#include <linux/bio.h> #endif #include <linux/log2.h> @@ -118,8 +119,8 @@ struct jbd2_revoke_table_s #ifdef __KERNEL__ static void write_one_revoke_record(journal_t *, transaction_t *, struct journal_head **, int *, - struct jbd2_revoke_record_s *); -static void flush_descriptor(journal_t *, struct journal_head *, int); + struct jbd2_revoke_record_s *, int); +static void flush_descriptor(journal_t *, struct journal_head *, int, int); #endif /* Utility functions to maintain the revoke table */ @@ -499,7 +500,8 @@ void jbd2_journal_switch_revoke_table(journal_t *journal) * revoke hash, deleting the entries as we go. */ void jbd2_journal_write_revoke_records(journal_t *journal, - transaction_t *transaction) + transaction_t *transaction, + int write_op) { struct journal_head *descriptor; struct jbd2_revoke_record_s *record; @@ -523,14 +525,14 @@ void jbd2_journal_write_revoke_records(journal_t *journal, hash_list->next; write_one_revoke_record(journal, transaction, &descriptor, &offset, - record); + record, write_op); count++; list_del(&record->hash); kmem_cache_free(jbd2_revoke_record_cache, record); } } if (descriptor) - flush_descriptor(journal, descriptor, offset); + flush_descriptor(journal, descriptor, offset, write_op); jbd_debug(1, "Wrote %d revoke records\n", count); } @@ -543,7 +545,8 @@ static void write_one_revoke_record(journal_t *journal, transaction_t *transaction, struct journal_head **descriptorp, int *offsetp, - struct jbd2_revoke_record_s *record) + struct jbd2_revoke_record_s *record, + int write_op) { struct journal_head *descriptor; int offset; @@ -562,7 +565,7 @@ static void write_one_revoke_record(journal_t *journal, /* Make sure we have a descriptor with space left for the record */ if (descriptor) { if (offset == journal->j_blocksize) { - flush_descriptor(journal, descriptor, offset); + flush_descriptor(journal, descriptor, offset, write_op); descriptor = NULL; } } @@ -607,7 +610,7 @@ static void write_one_revoke_record(journal_t *journal, static void flush_descriptor(journal_t *journal, struct journal_head *descriptor, - int offset) + int offset, int write_op) { jbd2_journal_revoke_header_t *header; struct buffer_head *bh = jh2bh(descriptor); @@ -622,7 +625,7 @@ static void flush_descriptor(journal_t *journal, set_buffer_jwrite(bh); BUFFER_TRACE(bh, "write"); set_buffer_dirty(bh); - ll_rw_block(SWRITE, 1, &bh); + ll_rw_block((write_op == WRITE) ? SWRITE : SWRITE_SYNC_PLUG, 1, &bh); } #endif diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 6f21adf9479a..d9b0e92b3602 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -720,8 +720,10 @@ static ssize_t jfs_quota_write(struct super_block *sb, int type, blk++; } out: - if (len == towrite) + if (len == towrite) { + mutex_unlock(&inode->i_mutex); return err; + } if (inode->i_size < off+len-towrite) i_size_write(inode, off+len-towrite); inode->i_version++; diff --git a/fs/namei.c b/fs/namei.c index b8433ebfae05..78f253cd2d4f 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1248,6 +1248,8 @@ struct dentry *lookup_one_len(const char *name, struct dentry *base, int len) int err; struct qstr this; + WARN_ON_ONCE(!mutex_is_locked(&base->d_inode->i_mutex)); + err = __lookup_one_len(name, &this, base, len); if (err) return ERR_PTR(err); diff --git a/fs/namespace.c b/fs/namespace.c index c6f54e4c4290..41196209a906 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1377,7 +1377,7 @@ static int attach_recursive_mnt(struct vfsmount *source_mnt, if (parent_path) { detach_mnt(source_mnt, parent_path); attach_mnt(source_mnt, path); - touch_mnt_namespace(current->nsproxy->mnt_ns); + touch_mnt_namespace(parent_path->mnt->mnt_ns); } else { mnt_set_mountpoint(dest_mnt, dest_dentry, source_mnt); commit_tree(source_mnt); @@ -1920,8 +1920,9 @@ long do_mount(char *dev_name, char *dir_name, char *type_page, if (data_page) ((char *)data_page)[PAGE_SIZE - 1] = 0; - /* Default to relatime */ - mnt_flags |= MNT_RELATIME; + /* Default to relatime unless overriden */ + if (!(flags & MS_NOATIME)) + mnt_flags |= MNT_RELATIME; /* Separate the per-mountpoint flags */ if (flags & MS_NOSUID) diff --git a/fs/ncpfs/ioctl.c b/fs/ncpfs/ioctl.c index f54360f50a9c..fa038df63ac8 100644 --- a/fs/ncpfs/ioctl.c +++ b/fs/ncpfs/ioctl.c @@ -660,13 +660,10 @@ outrel: if (user.object_name_len > NCP_OBJECT_NAME_MAX_LEN) return -ENOMEM; if (user.object_name_len) { - newname = kmalloc(user.object_name_len, GFP_USER); - if (!newname) - return -ENOMEM; - if (copy_from_user(newname, user.object_name, user.object_name_len)) { - kfree(newname); - return -EFAULT; - } + newname = memdup_user(user.object_name, + user.object_name_len); + if (IS_ERR(newname)) + return PTR_ERR(newname); } else { newname = NULL; } @@ -760,13 +757,9 @@ outrel: if (user.len > NCP_PRIVATE_DATA_MAX_LEN) return -ENOMEM; if (user.len) { - new = kmalloc(user.len, GFP_USER); - if (!new) - return -ENOMEM; - if (copy_from_user(new, user.data, user.len)) { - kfree(new); - return -EFAULT; - } + new = memdup_user(user.data, user.len); + if (IS_ERR(new)) + return PTR_ERR(new); } else { new = NULL; } diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig index e67f3ec07736..5d6d6f415935 100644 --- a/fs/nfs/Kconfig +++ b/fs/nfs/Kconfig @@ -74,6 +74,15 @@ config NFS_V4 If unsure, say N. +config NFS_V4_1 + bool "NFS client support for NFSv4.1 (DEVELOPER ONLY)" + depends on NFS_V4 && EXPERIMENTAL + help + This option enables support for minor version 1 of the NFSv4 protocol + (draft-ietf-nfsv4-minorversion1) in the kernel's NFS client. + + Unless you're an NFS developer, say N. + config ROOT_NFS bool "Root file system on NFS" depends on NFS_FS=y && IP_PNP diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 75c9cd2aa119..3e232bf56dfb 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -102,6 +102,7 @@ struct nfs_client_initdata { size_t addrlen; const struct nfs_rpc_ops *rpc_ops; int proto; + u32 minorversion; }; /* @@ -150,6 +151,7 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_ rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS client"); clp->cl_boot_time = CURRENT_TIME; clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED; + clp->cl_minorversion = cl_init->minorversion; #endif cred = rpc_lookup_machine_cred(); if (!IS_ERR(cred)) @@ -182,12 +184,29 @@ static void nfs4_shutdown_client(struct nfs_client *clp) } /* + * Clears/puts all minor version specific parts from an nfs_client struct + * reverting it to minorversion 0. + */ +static void nfs4_clear_client_minor_version(struct nfs_client *clp) +{ +#ifdef CONFIG_NFS_V4_1 + if (nfs4_has_session(clp)) { + nfs4_destroy_session(clp->cl_session); + clp->cl_session = NULL; + } + + clp->cl_call_sync = _nfs4_call_sync; +#endif /* CONFIG_NFS_V4_1 */ +} + +/* * Destroy a shared client record */ static void nfs_free_client(struct nfs_client *clp) { dprintk("--> nfs_free_client(%u)\n", clp->rpc_ops->version); + nfs4_clear_client_minor_version(clp); nfs4_shutdown_client(clp); nfs_fscache_release_client_cookie(clp); @@ -347,7 +366,8 @@ struct nfs_client *nfs_find_client(const struct sockaddr *addr, u32 nfsversion) struct sockaddr *clap = (struct sockaddr *)&clp->cl_addr; /* Don't match clients that failed to initialise properly */ - if (clp->cl_cons_state != NFS_CS_READY) + if (!(clp->cl_cons_state == NFS_CS_READY || + clp->cl_cons_state == NFS_CS_SESSION_INITING)) continue; /* Different NFS versions cannot share the same nfs_client */ @@ -420,7 +440,9 @@ static struct nfs_client *nfs_match_client(const struct nfs_client_initdata *dat if (clp->cl_proto != data->proto) continue; - + /* Match nfsv4 minorversion */ + if (clp->cl_minorversion != data->minorversion) + continue; /* Match the full socket address */ if (!nfs_sockaddr_cmp(sap, clap)) continue; @@ -478,7 +500,7 @@ found_client: nfs_free_client(new); error = wait_event_killable(nfs_client_active_wq, - clp->cl_cons_state != NFS_CS_INITING); + clp->cl_cons_state < NFS_CS_INITING); if (error < 0) { nfs_put_client(clp); return ERR_PTR(-ERESTARTSYS); @@ -499,13 +521,29 @@ found_client: /* * Mark a server as ready or failed */ -static void nfs_mark_client_ready(struct nfs_client *clp, int state) +void nfs_mark_client_ready(struct nfs_client *clp, int state) { clp->cl_cons_state = state; wake_up_all(&nfs_client_active_wq); } /* + * With sessions, the client is not marked ready until after a + * successful EXCHANGE_ID and CREATE_SESSION. + * + * Map errors cl_cons_state errors to EPROTONOSUPPORT to indicate + * other versions of NFS can be tried. + */ +int nfs4_check_client_ready(struct nfs_client *clp) +{ + if (!nfs4_has_session(clp)) + return 0; + if (clp->cl_cons_state < NFS_CS_READY) + return -EPROTONOSUPPORT; + return 0; +} + +/* * Initialise the timeout values for a connection */ static void nfs_init_timeout_values(struct rpc_timeout *to, int proto, @@ -1050,6 +1088,33 @@ error: #ifdef CONFIG_NFS_V4 /* + * Initialize the minor version specific parts of an NFS4 client record + */ +static int nfs4_init_client_minor_version(struct nfs_client *clp) +{ + clp->cl_call_sync = _nfs4_call_sync; + +#if defined(CONFIG_NFS_V4_1) + if (clp->cl_minorversion) { + struct nfs4_session *session = NULL; + /* + * Create the session and mark it expired. + * When a SEQUENCE operation encounters the expired session + * it will do session recovery to initialize it. + */ + session = nfs4_alloc_session(clp); + if (!session) + return -ENOMEM; + + clp->cl_session = session; + clp->cl_call_sync = _nfs4_call_sync_session; + } +#endif /* CONFIG_NFS_V4_1 */ + + return 0; +} + +/* * Initialise an NFS4 client record */ static int nfs4_init_client(struct nfs_client *clp, @@ -1083,7 +1148,12 @@ static int nfs4_init_client(struct nfs_client *clp, } __set_bit(NFS_CS_IDMAP, &clp->cl_res_state); - nfs_mark_client_ready(clp, NFS_CS_READY); + error = nfs4_init_client_minor_version(clp); + if (error < 0) + goto error; + + if (!nfs4_has_session(clp)) + nfs_mark_client_ready(clp, NFS_CS_READY); return 0; error: @@ -1101,7 +1171,8 @@ static int nfs4_set_client(struct nfs_server *server, const size_t addrlen, const char *ip_addr, rpc_authflavor_t authflavour, - int proto, const struct rpc_timeout *timeparms) + int proto, const struct rpc_timeout *timeparms, + u32 minorversion) { struct nfs_client_initdata cl_init = { .hostname = hostname, @@ -1109,6 +1180,7 @@ static int nfs4_set_client(struct nfs_server *server, .addrlen = addrlen, .rpc_ops = &nfs_v4_clientops, .proto = proto, + .minorversion = minorversion, }; struct nfs_client *clp; int error; @@ -1138,6 +1210,36 @@ error: } /* + * Initialize a session. + * Note: save the mount rsize and wsize for create_server negotiation. + */ +static void nfs4_init_session(struct nfs_client *clp, + unsigned int wsize, unsigned int rsize) +{ +#if defined(CONFIG_NFS_V4_1) + if (nfs4_has_session(clp)) { + clp->cl_session->fc_attrs.max_rqst_sz = wsize; + clp->cl_session->fc_attrs.max_resp_sz = rsize; + } +#endif /* CONFIG_NFS_V4_1 */ +} + +/* + * Session has been established, and the client marked ready. + * Set the mount rsize and wsize with negotiated fore channel + * attributes which will be bound checked in nfs_server_set_fsinfo. + */ +static void nfs4_session_set_rwsize(struct nfs_server *server) +{ +#ifdef CONFIG_NFS_V4_1 + if (!nfs4_has_session(server->nfs_client)) + return; + server->rsize = server->nfs_client->cl_session->fc_attrs.max_resp_sz; + server->wsize = server->nfs_client->cl_session->fc_attrs.max_rqst_sz; +#endif /* CONFIG_NFS_V4_1 */ +} + +/* * Create a version 4 volume record */ static int nfs4_init_server(struct nfs_server *server, @@ -1164,7 +1266,8 @@ static int nfs4_init_server(struct nfs_server *server, data->client_address, data->auth_flavors[0], data->nfs_server.protocol, - &timeparms); + &timeparms, + data->minorversion); if (error < 0) goto error; @@ -1214,6 +1317,8 @@ struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data, BUG_ON(!server->nfs_client->rpc_ops); BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops); + nfs4_init_session(server->nfs_client, server->wsize, server->rsize); + /* Probe the root fh to retrieve its FSID */ error = nfs4_path_walk(server, mntfh, data->nfs_server.export_path); if (error < 0) @@ -1224,6 +1329,8 @@ struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data, (unsigned long long) server->fsid.minor); dprintk("Mount FH: %d\n", mntfh->size); + nfs4_session_set_rwsize(server); + error = nfs_probe_fsinfo(server, mntfh, &fattr); if (error < 0) goto error; @@ -1282,7 +1389,8 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data, parent_client->cl_ipaddr, data->authflavor, parent_server->client->cl_xprt->prot, - parent_server->client->cl_timeout); + parent_server->client->cl_timeout, + parent_client->cl_minorversion); if (error < 0) goto error; diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 08f6b040d289..489fc01a3204 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -259,6 +259,9 @@ static void nfs_direct_read_release(void *calldata) } static const struct rpc_call_ops nfs_read_direct_ops = { +#if defined(CONFIG_NFS_V4_1) + .rpc_call_prepare = nfs_read_prepare, +#endif /* CONFIG_NFS_V4_1 */ .rpc_call_done = nfs_direct_read_result, .rpc_release = nfs_direct_read_release, }; @@ -535,6 +538,9 @@ static void nfs_direct_commit_release(void *calldata) } static const struct rpc_call_ops nfs_commit_direct_ops = { +#if defined(CONFIG_NFS_V4_1) + .rpc_call_prepare = nfs_write_prepare, +#endif /* CONFIG_NFS_V4_1 */ .rpc_call_done = nfs_direct_commit_result, .rpc_release = nfs_direct_commit_release, }; @@ -673,6 +679,9 @@ out_unlock: } static const struct rpc_call_ops nfs_write_direct_ops = { +#if defined(CONFIG_NFS_V4_1) + .rpc_call_prepare = nfs_write_prepare, +#endif /* CONFIG_NFS_V4_1 */ .rpc_call_done = nfs_direct_write_result, .rpc_release = nfs_direct_write_release, }; diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index e4d6a8348adf..acee3274d275 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -2,6 +2,7 @@ * NFS internal definitions */ +#include "nfs4_fs.h" #include <linux/mount.h> #include <linux/security.h> @@ -17,6 +18,18 @@ struct nfs_string; */ #define NFS_MAX_READAHEAD (RPC_DEF_SLOT_TABLE - 1) +/* + * Determine if sessions are in use. + */ +static inline int nfs4_has_session(const struct nfs_client *clp) +{ +#ifdef CONFIG_NFS_V4_1 + if (clp->cl_session) + return 1; +#endif /* CONFIG_NFS_V4_1 */ + return 0; +} + struct nfs_clone_mount { const struct super_block *sb; const struct dentry *dentry; @@ -44,6 +57,7 @@ struct nfs_parsed_mount_data { unsigned int auth_flavor_len; rpc_authflavor_t auth_flavors[1]; char *client_address; + unsigned int minorversion; char *fscache_uniq; struct { @@ -99,6 +113,8 @@ extern void nfs_free_server(struct nfs_server *server); extern struct nfs_server *nfs_clone_server(struct nfs_server *, struct nfs_fh *, struct nfs_fattr *); +extern void nfs_mark_client_ready(struct nfs_client *clp, int state); +extern int nfs4_check_client_ready(struct nfs_client *clp); #ifdef CONFIG_PROC_FS extern int __init nfs_fs_proc_init(void); extern void nfs_fs_proc_exit(void); @@ -146,6 +162,20 @@ extern __be32 * nfs_decode_dirent(__be32 *, struct nfs_entry *, int); extern struct rpc_procinfo nfs3_procedures[]; extern __be32 *nfs3_decode_dirent(__be32 *, struct nfs_entry *, int); +/* nfs4proc.c */ +static inline void nfs4_restart_rpc(struct rpc_task *task, + const struct nfs_client *clp) +{ +#ifdef CONFIG_NFS_V4_1 + if (nfs4_has_session(clp) && + test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)) { + rpc_restart_call_prepare(task); + return; + } +#endif /* CONFIG_NFS_V4_1 */ + rpc_restart_call(task); +} + /* nfs4xdr.c */ #ifdef CONFIG_NFS_V4 extern __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus); @@ -205,6 +235,38 @@ extern int nfs4_path_walk(struct nfs_server *server, const char *path); #endif +/* read.c */ +extern void nfs_read_prepare(struct rpc_task *task, void *calldata); + +/* write.c */ +extern void nfs_write_prepare(struct rpc_task *task, void *calldata); + +/* nfs4proc.c */ +extern int _nfs4_call_sync(struct nfs_server *server, + struct rpc_message *msg, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + int cache_reply); +extern int _nfs4_call_sync_session(struct nfs_server *server, + struct rpc_message *msg, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + int cache_reply); + +#ifdef CONFIG_NFS_V4_1 +extern void nfs41_sequence_free_slot(const struct nfs_client *, + struct nfs4_sequence_res *res); +#endif /* CONFIG_NFS_V4_1 */ + +static inline void nfs4_sequence_free_slot(const struct nfs_client *clp, + struct nfs4_sequence_res *res) +{ +#ifdef CONFIG_NFS_V4_1 + if (nfs4_has_session(clp)) + nfs41_sequence_free_slot(clp, res); +#endif /* CONFIG_NFS_V4_1 */ +} + /* * Determine the device name as a string */ diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index e6a1932c7110..35869a4921f1 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -713,7 +713,8 @@ nfs3_xdr_setaclargs(struct rpc_rqst *req, __be32 *p, if (args->npages != 0) xdr_encode_pages(buf, args->pages, 0, args->len); else - req->rq_slen += args->len; + req->rq_slen = xdr_adjust_iovec(req->rq_svec, + p + XDR_QUADLEN(args->len)); err = nfsacl_encode(buf, base, args->inode, (args->mask & NFS_ACL) ? diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 84345deab26f..61bc3a32e1e2 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -44,6 +44,7 @@ enum nfs4_client_state { NFS4CLNT_RECLAIM_REBOOT, NFS4CLNT_RECLAIM_NOGRACE, NFS4CLNT_DELEGRETURN, + NFS4CLNT_SESSION_SETUP, }; /* @@ -177,6 +178,14 @@ struct nfs4_state_recovery_ops { int state_flag_bit; int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *); int (*recover_lock)(struct nfs4_state *, struct file_lock *); + int (*establish_clid)(struct nfs_client *, struct rpc_cred *); + struct rpc_cred * (*get_clid_cred)(struct nfs_client *); +}; + +struct nfs4_state_maintenance_ops { + int (*sched_state_renewal)(struct nfs_client *, struct rpc_cred *); + struct rpc_cred * (*get_state_renewal_cred_locked)(struct nfs_client *); + int (*renew_lease)(struct nfs_client *, struct rpc_cred *); }; extern const struct dentry_operations nfs4_dentry_operations; @@ -193,6 +202,7 @@ extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struc extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct rpc_cred *); extern int nfs4_proc_async_renew(struct nfs_client *, struct rpc_cred *); extern int nfs4_proc_renew(struct nfs_client *, struct rpc_cred *); +extern int nfs4_init_clientid(struct nfs_client *, struct rpc_cred *); extern int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait); extern struct dentry *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *); extern int nfs4_open_revalidate(struct inode *, struct dentry *, int, struct nameidata *); @@ -200,8 +210,26 @@ extern int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fh extern int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name, struct nfs4_fs_locations *fs_locations, struct page *page); -extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops; -extern struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops; +extern struct nfs4_state_recovery_ops *nfs4_reboot_recovery_ops[]; +extern struct nfs4_state_recovery_ops *nfs4_nograce_recovery_ops[]; +#if defined(CONFIG_NFS_V4_1) +extern int nfs4_setup_sequence(struct nfs_client *clp, + struct nfs4_sequence_args *args, struct nfs4_sequence_res *res, + int cache_reply, struct rpc_task *task); +extern void nfs4_destroy_session(struct nfs4_session *session); +extern struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp); +extern int nfs4_proc_create_session(struct nfs_client *, int reset); +extern int nfs4_proc_destroy_session(struct nfs4_session *); +#else /* CONFIG_NFS_v4_1 */ +static inline int nfs4_setup_sequence(struct nfs_client *clp, + struct nfs4_sequence_args *args, struct nfs4_sequence_res *res, + int cache_reply, struct rpc_task *task) +{ + return 0; +} +#endif /* CONFIG_NFS_V4_1 */ + +extern struct nfs4_state_maintenance_ops *nfs4_state_renewal_ops[]; extern const u32 nfs4_fattr_bitmap[2]; extern const u32 nfs4_statfs_bitmap[2]; @@ -216,7 +244,12 @@ extern void nfs4_kill_renewd(struct nfs_client *); extern void nfs4_renew_state(struct work_struct *); /* nfs4state.c */ +struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp); struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp); +#if defined(CONFIG_NFS_V4_1) +struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp); +struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp); +#endif /* CONFIG_NFS_V4_1 */ extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *); extern void nfs4_put_state_owner(struct nfs4_state_owner *); diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index a4d242680299..530ebb1e84fe 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -48,11 +48,13 @@ #include <linux/smp_lock.h> #include <linux/namei.h> #include <linux/mount.h> +#include <linux/module.h> #include "nfs4_fs.h" #include "delegation.h" #include "internal.h" #include "iostat.h" +#include "callback.h" #define NFSDBG_FACILITY NFSDBG_PROC @@ -247,7 +249,25 @@ static int nfs4_handle_exception(const struct nfs_server *server, int errorcode, ret = nfs4_wait_clnt_recover(clp); if (ret == 0) exception->retry = 1; +#if !defined(CONFIG_NFS_V4_1) break; +#else /* !defined(CONFIG_NFS_V4_1) */ + if (!nfs4_has_session(server->nfs_client)) + break; + /* FALLTHROUGH */ + case -NFS4ERR_BADSESSION: + case -NFS4ERR_BADSLOT: + case -NFS4ERR_BAD_HIGH_SLOT: + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + case -NFS4ERR_DEADSESSION: + case -NFS4ERR_SEQ_FALSE_RETRY: + case -NFS4ERR_SEQ_MISORDERED: + dprintk("%s ERROR: %d Reset session\n", __func__, + errorcode); + set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state); + exception->retry = 1; + /* FALLTHROUGH */ +#endif /* !defined(CONFIG_NFS_V4_1) */ case -NFS4ERR_FILE_OPEN: case -NFS4ERR_GRACE: case -NFS4ERR_DELAY: @@ -271,6 +291,353 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp spin_unlock(&clp->cl_lock); } +#if defined(CONFIG_NFS_V4_1) + +/* + * nfs4_free_slot - free a slot and efficiently update slot table. + * + * freeing a slot is trivially done by clearing its respective bit + * in the bitmap. + * If the freed slotid equals highest_used_slotid we want to update it + * so that the server would be able to size down the slot table if needed, + * otherwise we know that the highest_used_slotid is still in use. + * When updating highest_used_slotid there may be "holes" in the bitmap + * so we need to scan down from highest_used_slotid to 0 looking for the now + * highest slotid in use. + * If none found, highest_used_slotid is set to -1. + */ +static void +nfs4_free_slot(struct nfs4_slot_table *tbl, u8 free_slotid) +{ + int slotid = free_slotid; + + spin_lock(&tbl->slot_tbl_lock); + /* clear used bit in bitmap */ + __clear_bit(slotid, tbl->used_slots); + + /* update highest_used_slotid when it is freed */ + if (slotid == tbl->highest_used_slotid) { + slotid = find_last_bit(tbl->used_slots, tbl->max_slots); + if (slotid >= 0 && slotid < tbl->max_slots) + tbl->highest_used_slotid = slotid; + else + tbl->highest_used_slotid = -1; + } + rpc_wake_up_next(&tbl->slot_tbl_waitq); + spin_unlock(&tbl->slot_tbl_lock); + dprintk("%s: free_slotid %u highest_used_slotid %d\n", __func__, + free_slotid, tbl->highest_used_slotid); +} + +void nfs41_sequence_free_slot(const struct nfs_client *clp, + struct nfs4_sequence_res *res) +{ + struct nfs4_slot_table *tbl; + + if (!nfs4_has_session(clp)) { + dprintk("%s: No session\n", __func__); + return; + } + tbl = &clp->cl_session->fc_slot_table; + if (res->sr_slotid == NFS4_MAX_SLOT_TABLE) { + dprintk("%s: No slot\n", __func__); + /* just wake up the next guy waiting since + * we may have not consumed a slot after all */ + rpc_wake_up_next(&tbl->slot_tbl_waitq); + return; + } + nfs4_free_slot(tbl, res->sr_slotid); + res->sr_slotid = NFS4_MAX_SLOT_TABLE; +} + +static void nfs41_sequence_done(struct nfs_client *clp, + struct nfs4_sequence_res *res, + int rpc_status) +{ + unsigned long timestamp; + struct nfs4_slot_table *tbl; + struct nfs4_slot *slot; + + /* + * sr_status remains 1 if an RPC level error occurred. The server + * may or may not have processed the sequence operation.. + * Proceed as if the server received and processed the sequence + * operation. + */ + if (res->sr_status == 1) + res->sr_status = NFS_OK; + + /* -ERESTARTSYS can result in skipping nfs41_sequence_setup */ + if (res->sr_slotid == NFS4_MAX_SLOT_TABLE) + goto out; + + tbl = &clp->cl_session->fc_slot_table; + slot = tbl->slots + res->sr_slotid; + + if (res->sr_status == 0) { + /* Update the slot's sequence and clientid lease timer */ + ++slot->seq_nr; + timestamp = res->sr_renewal_time; + spin_lock(&clp->cl_lock); + if (time_before(clp->cl_last_renewal, timestamp)) + clp->cl_last_renewal = timestamp; + spin_unlock(&clp->cl_lock); + return; + } +out: + /* The session may be reset by one of the error handlers. */ + dprintk("%s: Error %d free the slot \n", __func__, res->sr_status); + nfs41_sequence_free_slot(clp, res); +} + +/* + * nfs4_find_slot - efficiently look for a free slot + * + * nfs4_find_slot looks for an unset bit in the used_slots bitmap. + * If found, we mark the slot as used, update the highest_used_slotid, + * and respectively set up the sequence operation args. + * The slot number is returned if found, or NFS4_MAX_SLOT_TABLE otherwise. + * + * Note: must be called with under the slot_tbl_lock. + */ +static u8 +nfs4_find_slot(struct nfs4_slot_table *tbl, struct rpc_task *task) +{ + int slotid; + u8 ret_id = NFS4_MAX_SLOT_TABLE; + BUILD_BUG_ON((u8)NFS4_MAX_SLOT_TABLE != (int)NFS4_MAX_SLOT_TABLE); + + dprintk("--> %s used_slots=%04lx highest_used=%d max_slots=%d\n", + __func__, tbl->used_slots[0], tbl->highest_used_slotid, + tbl->max_slots); + slotid = find_first_zero_bit(tbl->used_slots, tbl->max_slots); + if (slotid >= tbl->max_slots) + goto out; + __set_bit(slotid, tbl->used_slots); + if (slotid > tbl->highest_used_slotid) + tbl->highest_used_slotid = slotid; + ret_id = slotid; +out: + dprintk("<-- %s used_slots=%04lx highest_used=%d slotid=%d \n", + __func__, tbl->used_slots[0], tbl->highest_used_slotid, ret_id); + return ret_id; +} + +static int nfs4_recover_session(struct nfs4_session *session) +{ + struct nfs_client *clp = session->clp; + int ret; + + for (;;) { + ret = nfs4_wait_clnt_recover(clp); + if (ret != 0) + return ret; + if (!test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)) + break; + nfs4_schedule_state_manager(clp); + } + return 0; +} + +static int nfs41_setup_sequence(struct nfs4_session *session, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + int cache_reply, + struct rpc_task *task) +{ + struct nfs4_slot *slot; + struct nfs4_slot_table *tbl; + int status = 0; + u8 slotid; + + dprintk("--> %s\n", __func__); + /* slot already allocated? */ + if (res->sr_slotid != NFS4_MAX_SLOT_TABLE) + return 0; + + memset(res, 0, sizeof(*res)); + res->sr_slotid = NFS4_MAX_SLOT_TABLE; + tbl = &session->fc_slot_table; + + spin_lock(&tbl->slot_tbl_lock); + if (test_bit(NFS4CLNT_SESSION_SETUP, &session->clp->cl_state)) { + if (tbl->highest_used_slotid != -1) { + rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL); + spin_unlock(&tbl->slot_tbl_lock); + dprintk("<-- %s: Session reset: draining\n", __func__); + return -EAGAIN; + } + + /* The slot table is empty; start the reset thread */ + dprintk("%s Session Reset\n", __func__); + spin_unlock(&tbl->slot_tbl_lock); + status = nfs4_recover_session(session); + if (status) + return status; + spin_lock(&tbl->slot_tbl_lock); + } + + slotid = nfs4_find_slot(tbl, task); + if (slotid == NFS4_MAX_SLOT_TABLE) { + rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL); + spin_unlock(&tbl->slot_tbl_lock); + dprintk("<-- %s: no free slots\n", __func__); + return -EAGAIN; + } + spin_unlock(&tbl->slot_tbl_lock); + + slot = tbl->slots + slotid; + args->sa_session = session; + args->sa_slotid = slotid; + args->sa_cache_this = cache_reply; + + dprintk("<-- %s slotid=%d seqid=%d\n", __func__, slotid, slot->seq_nr); + + res->sr_session = session; + res->sr_slotid = slotid; + res->sr_renewal_time = jiffies; + /* + * sr_status is only set in decode_sequence, and so will remain + * set to 1 if an rpc level failure occurs. + */ + res->sr_status = 1; + return 0; +} + +int nfs4_setup_sequence(struct nfs_client *clp, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + int cache_reply, + struct rpc_task *task) +{ + int ret = 0; + + dprintk("--> %s clp %p session %p sr_slotid %d\n", + __func__, clp, clp->cl_session, res->sr_slotid); + + if (!nfs4_has_session(clp)) + goto out; + ret = nfs41_setup_sequence(clp->cl_session, args, res, cache_reply, + task); + if (ret != -EAGAIN) { + /* terminate rpc task */ + task->tk_status = ret; + task->tk_action = NULL; + } +out: + dprintk("<-- %s status=%d\n", __func__, ret); + return ret; +} + +struct nfs41_call_sync_data { + struct nfs_client *clp; + struct nfs4_sequence_args *seq_args; + struct nfs4_sequence_res *seq_res; + int cache_reply; +}; + +static void nfs41_call_sync_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs41_call_sync_data *data = calldata; + + dprintk("--> %s data->clp->cl_session %p\n", __func__, + data->clp->cl_session); + if (nfs4_setup_sequence(data->clp, data->seq_args, + data->seq_res, data->cache_reply, task)) + return; + rpc_call_start(task); +} + +static void nfs41_call_sync_done(struct rpc_task *task, void *calldata) +{ + struct nfs41_call_sync_data *data = calldata; + + nfs41_sequence_done(data->clp, data->seq_res, task->tk_status); + nfs41_sequence_free_slot(data->clp, data->seq_res); +} + +struct rpc_call_ops nfs41_call_sync_ops = { + .rpc_call_prepare = nfs41_call_sync_prepare, + .rpc_call_done = nfs41_call_sync_done, +}; + +static int nfs4_call_sync_sequence(struct nfs_client *clp, + struct rpc_clnt *clnt, + struct rpc_message *msg, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + int cache_reply) +{ + int ret; + struct rpc_task *task; + struct nfs41_call_sync_data data = { + .clp = clp, + .seq_args = args, + .seq_res = res, + .cache_reply = cache_reply, + }; + struct rpc_task_setup task_setup = { + .rpc_client = clnt, + .rpc_message = msg, + .callback_ops = &nfs41_call_sync_ops, + .callback_data = &data + }; + + res->sr_slotid = NFS4_MAX_SLOT_TABLE; + task = rpc_run_task(&task_setup); + if (IS_ERR(task)) + ret = PTR_ERR(task); + else { + ret = task->tk_status; + rpc_put_task(task); + } + return ret; +} + +int _nfs4_call_sync_session(struct nfs_server *server, + struct rpc_message *msg, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + int cache_reply) +{ + return nfs4_call_sync_sequence(server->nfs_client, server->client, + msg, args, res, cache_reply); +} + +#endif /* CONFIG_NFS_V4_1 */ + +int _nfs4_call_sync(struct nfs_server *server, + struct rpc_message *msg, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + int cache_reply) +{ + args->sa_session = res->sr_session = NULL; + return rpc_call_sync(server->client, msg, 0); +} + +#define nfs4_call_sync(server, msg, args, res, cache_reply) \ + (server)->nfs_client->cl_call_sync((server), (msg), &(args)->seq_args, \ + &(res)->seq_res, (cache_reply)) + +static void nfs4_sequence_done(const struct nfs_server *server, + struct nfs4_sequence_res *res, int rpc_status) +{ +#ifdef CONFIG_NFS_V4_1 + if (nfs4_has_session(server->nfs_client)) + nfs41_sequence_done(server->nfs_client, res, rpc_status); +#endif /* CONFIG_NFS_V4_1 */ +} + +/* no restart, therefore free slot here */ +static void nfs4_sequence_done_free_slot(const struct nfs_server *server, + struct nfs4_sequence_res *res, + int rpc_status) +{ + nfs4_sequence_done(server, res, rpc_status); + nfs4_sequence_free_slot(server->nfs_client, res); +} + static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo) { struct nfs_inode *nfsi = NFS_I(dir); @@ -343,6 +710,7 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct path *path, p->o_arg.server = server; p->o_arg.bitmask = server->attr_bitmask; p->o_arg.claim = NFS4_OPEN_CLAIM_NULL; + p->o_res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE; if (flags & O_EXCL) { u32 *s = (u32 *) p->o_arg.u.verifier.data; s[0] = jiffies; @@ -929,6 +1297,10 @@ static void nfs4_open_prepare(struct rpc_task *task, void *calldata) nfs_copy_fh(&data->o_res.fh, data->o_arg.fh); } data->timestamp = jiffies; + if (nfs4_setup_sequence(data->o_arg.server->nfs_client, + &data->o_arg.seq_args, + &data->o_res.seq_res, 1, task)) + return; rpc_call_start(task); return; out_no_action: @@ -941,6 +1313,10 @@ static void nfs4_open_done(struct rpc_task *task, void *calldata) struct nfs4_opendata *data = calldata; data->rpc_status = task->tk_status; + + nfs4_sequence_done_free_slot(data->o_arg.server, &data->o_res.seq_res, + task->tk_status); + if (RPC_ASSASSINATED(task)) return; if (task->tk_status == 0) { @@ -1269,7 +1645,7 @@ static int _nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred, } else memcpy(&arg.stateid, &zero_stateid, sizeof(arg.stateid)); - status = rpc_call_sync(server->client, &msg, 0); + status = nfs4_call_sync(server, &msg, &arg, &res, 1); if (status == 0 && state != NULL) renew_lease(server, timestamp); return status; @@ -1318,6 +1694,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data) struct nfs4_state *state = calldata->state; struct nfs_server *server = NFS_SERVER(calldata->inode); + nfs4_sequence_done(server, &calldata->res.seq_res, task->tk_status); if (RPC_ASSASSINATED(task)) return; /* hmm. we are done with the inode, and in the process of freeing @@ -1336,10 +1713,11 @@ static void nfs4_close_done(struct rpc_task *task, void *data) break; default: if (nfs4_async_handle_error(task, server, state) == -EAGAIN) { - rpc_restart_call(task); + nfs4_restart_rpc(task, server->nfs_client); return; } } + nfs4_sequence_free_slot(server->nfs_client, &calldata->res.seq_res); nfs_refresh_inode(calldata->inode, calldata->res.fattr); } @@ -1380,6 +1758,10 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data) calldata->arg.fmode = FMODE_WRITE; } calldata->timestamp = jiffies; + if (nfs4_setup_sequence((NFS_SERVER(calldata->inode))->nfs_client, + &calldata->arg.seq_args, &calldata->res.seq_res, + 1, task)) + return; rpc_call_start(task); } @@ -1419,13 +1801,15 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait) }; int status = -ENOMEM; - calldata = kmalloc(sizeof(*calldata), GFP_KERNEL); + calldata = kzalloc(sizeof(*calldata), GFP_KERNEL); if (calldata == NULL) goto out; calldata->inode = state->inode; calldata->state = state; calldata->arg.fh = NFS_FH(state->inode); calldata->arg.stateid = &state->open_stateid; + if (nfs4_has_session(server->nfs_client)) + memset(calldata->arg.stateid->data, 0, 4); /* clear seqid */ /* Serialization for the sequence id */ calldata->arg.seqid = nfs_alloc_seqid(&state->owner->so_seqid); if (calldata->arg.seqid == NULL) @@ -1435,6 +1819,7 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait) calldata->res.fattr = &calldata->fattr; calldata->res.seqid = calldata->arg.seqid; calldata->res.server = server; + calldata->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE; calldata->path.mnt = mntget(path->mnt); calldata->path.dentry = dget(path->dentry); @@ -1584,15 +1969,18 @@ void nfs4_close_context(struct nfs_open_context *ctx, int is_sync) static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle) { + struct nfs4_server_caps_arg args = { + .fhandle = fhandle, + }; struct nfs4_server_caps_res res = {}; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SERVER_CAPS], - .rpc_argp = fhandle, + .rpc_argp = &args, .rpc_resp = &res, }; int status; - status = rpc_call_sync(server->client, &msg, 0); + status = nfs4_call_sync(server, &msg, &args, &res, 0); if (status == 0) { memcpy(server->attr_bitmask, res.attr_bitmask, sizeof(server->attr_bitmask)); if (res.attr_bitmask[0] & FATTR4_WORD0_ACL) @@ -1606,6 +1994,7 @@ static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *f server->cache_consistency_bitmask[1] &= FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY; server->acl_bitmask = res.acl_bitmask; } + return status; } @@ -1637,8 +2026,15 @@ static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle, .rpc_argp = &args, .rpc_resp = &res, }; + int status; + nfs_fattr_init(info->fattr); - return rpc_call_sync(server->client, &msg, 0); + status = nfs4_recover_expired_lease(server); + if (!status) + status = nfs4_check_client_ready(server->nfs_client); + if (!status) + status = nfs4_call_sync(server, &msg, &args, &res, 0); + return status; } static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle, @@ -1728,7 +2124,7 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, }; nfs_fattr_init(fattr); - return rpc_call_sync(server->client, &msg, 0); + return nfs4_call_sync(server, &msg, &args, &res, 0); } static int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr) @@ -1812,7 +2208,7 @@ static int _nfs4_proc_lookupfh(struct nfs_server *server, const struct nfs_fh *d nfs_fattr_init(fattr); dprintk("NFS call lookupfh %s\n", name->name); - status = rpc_call_sync(server->client, &msg, 0); + status = nfs4_call_sync(server, &msg, &args, &res, 0); dprintk("NFS reply lookupfh: %d\n", status); return status; } @@ -1898,7 +2294,7 @@ static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry args.access |= NFS4_ACCESS_EXECUTE; } nfs_fattr_init(&fattr); - status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); + status = nfs4_call_sync(server, &msg, &args, &res, 0); if (!status) { entry->mask = 0; if (res.access & NFS4_ACCESS_READ) @@ -1957,13 +2353,14 @@ static int _nfs4_proc_readlink(struct inode *inode, struct page *page, .pglen = pglen, .pages = &page, }; + struct nfs4_readlink_res res; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READLINK], .rpc_argp = &args, - .rpc_resp = NULL, + .rpc_resp = &res, }; - return rpc_call_sync(NFS_CLIENT(inode), &msg, 0); + return nfs4_call_sync(NFS_SERVER(inode), &msg, &args, &res, 0); } static int nfs4_proc_readlink(struct inode *inode, struct page *page, @@ -2057,7 +2454,7 @@ static int _nfs4_proc_remove(struct inode *dir, struct qstr *name) int status; nfs_fattr_init(&res.dir_attr); - status = rpc_call_sync(server->client, &msg, 0); + status = nfs4_call_sync(server, &msg, &args, &res, 1); if (status == 0) { update_changeattr(dir, &res.cinfo); nfs_post_op_update_inode(dir, &res.dir_attr); @@ -2092,8 +2489,10 @@ static int nfs4_proc_unlink_done(struct rpc_task *task, struct inode *dir) { struct nfs_removeres *res = task->tk_msg.rpc_resp; + nfs4_sequence_done(res->server, &res->seq_res, task->tk_status); if (nfs4_async_handle_error(task, res->server, NULL) == -EAGAIN) return 0; + nfs4_sequence_free_slot(res->server->nfs_client, &res->seq_res); update_changeattr(dir, &res->cinfo); nfs_post_op_update_inode(dir, &res->dir_attr); return 1; @@ -2125,7 +2524,7 @@ static int _nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name, nfs_fattr_init(res.old_fattr); nfs_fattr_init(res.new_fattr); - status = rpc_call_sync(server->client, &msg, 0); + status = nfs4_call_sync(server, &msg, &arg, &res, 1); if (!status) { update_changeattr(old_dir, &res.old_cinfo); @@ -2174,7 +2573,7 @@ static int _nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr * nfs_fattr_init(res.fattr); nfs_fattr_init(res.dir_attr); - status = rpc_call_sync(server->client, &msg, 0); + status = nfs4_call_sync(server, &msg, &arg, &res, 1); if (!status) { update_changeattr(dir, &res.cinfo); nfs_post_op_update_inode(dir, res.dir_attr); @@ -2235,7 +2634,8 @@ static struct nfs4_createdata *nfs4_alloc_createdata(struct inode *dir, static int nfs4_do_create(struct inode *dir, struct dentry *dentry, struct nfs4_createdata *data) { - int status = rpc_call_sync(NFS_CLIENT(dir), &data->msg, 0); + int status = nfs4_call_sync(NFS_SERVER(dir), &data->msg, + &data->arg, &data->res, 1); if (status == 0) { update_changeattr(dir, &data->res.dir_cinfo); nfs_post_op_update_inode(dir, data->res.dir_fattr); @@ -2344,7 +2744,7 @@ static int _nfs4_proc_readdir(struct dentry *dentry, struct rpc_cred *cred, (unsigned long long)cookie); nfs4_setup_readdir(cookie, NFS_COOKIEVERF(dir), dentry, &args); res.pgbase = args.pgbase; - status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + status = nfs4_call_sync(NFS_SERVER(dir), &msg, &args, &res, 0); if (status == 0) memcpy(NFS_COOKIEVERF(dir), res.verifier.data, NFS4_VERIFIER_SIZE); @@ -2422,14 +2822,17 @@ static int _nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, .fh = fhandle, .bitmask = server->attr_bitmask, }; + struct nfs4_statfs_res res = { + .fsstat = fsstat, + }; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_STATFS], .rpc_argp = &args, - .rpc_resp = fsstat, + .rpc_resp = &res, }; nfs_fattr_init(fsstat->fattr); - return rpc_call_sync(server->client, &msg, 0); + return nfs4_call_sync(server, &msg, &args, &res, 0); } static int nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsstat *fsstat) @@ -2451,13 +2854,16 @@ static int _nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, .fh = fhandle, .bitmask = server->attr_bitmask, }; + struct nfs4_fsinfo_res res = { + .fsinfo = fsinfo, + }; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FSINFO], .rpc_argp = &args, - .rpc_resp = fsinfo, + .rpc_resp = &res, }; - return rpc_call_sync(server->client, &msg, 0); + return nfs4_call_sync(server, &msg, &args, &res, 0); } static int nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *fsinfo) @@ -2486,10 +2892,13 @@ static int _nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle .fh = fhandle, .bitmask = server->attr_bitmask, }; + struct nfs4_pathconf_res res = { + .pathconf = pathconf, + }; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_PATHCONF], .rpc_argp = &args, - .rpc_resp = pathconf, + .rpc_resp = &res, }; /* None of the pathconf attributes are mandatory to implement */ @@ -2499,7 +2908,7 @@ static int _nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle } nfs_fattr_init(pathconf->fattr); - return rpc_call_sync(server->client, &msg, 0); + return nfs4_call_sync(server, &msg, &args, &res, 0); } static int nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, @@ -2520,8 +2929,13 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data) { struct nfs_server *server = NFS_SERVER(data->inode); + dprintk("--> %s\n", __func__); + + /* nfs4_sequence_free_slot called in the read rpc_call_done */ + nfs4_sequence_done(server, &data->res.seq_res, task->tk_status); + if (nfs4_async_handle_error(task, server, data->args.context->state) == -EAGAIN) { - rpc_restart_call(task); + nfs4_restart_rpc(task, server->nfs_client); return -EAGAIN; } @@ -2541,8 +2955,12 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data) { struct inode *inode = data->inode; + /* slot is freed in nfs_writeback_done */ + nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res, + task->tk_status); + if (nfs4_async_handle_error(task, NFS_SERVER(inode), data->args.context->state) == -EAGAIN) { - rpc_restart_call(task); + nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client); return -EAGAIN; } if (task->tk_status >= 0) { @@ -2567,10 +2985,14 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data) { struct inode *inode = data->inode; + nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res, + task->tk_status); if (nfs4_async_handle_error(task, NFS_SERVER(inode), NULL) == -EAGAIN) { - rpc_restart_call(task); + nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client); return -EAGAIN; } + nfs4_sequence_free_slot(NFS_SERVER(inode)->nfs_client, + &data->res.seq_res); nfs_refresh_inode(inode, data->res.fattr); return 0; } @@ -2606,6 +3028,9 @@ static void nfs4_renew_done(struct rpc_task *task, void *data) if (time_before(clp->cl_last_renewal,timestamp)) clp->cl_last_renewal = timestamp; spin_unlock(&clp->cl_lock); + dprintk("%s calling put_rpccred on rpc_cred %p\n", __func__, + task->tk_msg.rpc_cred); + put_rpccred(task->tk_msg.rpc_cred); } static const struct rpc_call_ops nfs4_renew_ops = { @@ -2745,12 +3170,14 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu .acl_pages = pages, .acl_len = buflen, }; - size_t resp_len = buflen; + struct nfs_getaclres res = { + .acl_len = buflen, + }; void *resp_buf; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GETACL], .rpc_argp = &args, - .rpc_resp = &resp_len, + .rpc_resp = &res, }; struct page *localpage = NULL; int ret; @@ -2764,26 +3191,26 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu return -ENOMEM; args.acl_pages[0] = localpage; args.acl_pgbase = 0; - resp_len = args.acl_len = PAGE_SIZE; + args.acl_len = PAGE_SIZE; } else { resp_buf = buf; buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase); } - ret = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); + ret = nfs4_call_sync(NFS_SERVER(inode), &msg, &args, &res, 0); if (ret) goto out_free; - if (resp_len > args.acl_len) - nfs4_write_cached_acl(inode, NULL, resp_len); + if (res.acl_len > args.acl_len) + nfs4_write_cached_acl(inode, NULL, res.acl_len); else - nfs4_write_cached_acl(inode, resp_buf, resp_len); + nfs4_write_cached_acl(inode, resp_buf, res.acl_len); if (buf) { ret = -ERANGE; - if (resp_len > buflen) + if (res.acl_len > buflen) goto out_free; if (localpage) - memcpy(buf, resp_buf, resp_len); + memcpy(buf, resp_buf, res.acl_len); } - ret = resp_len; + ret = res.acl_len; out_free: if (localpage) __free_page(localpage); @@ -2830,10 +3257,11 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl .acl_pages = pages, .acl_len = buflen, }; + struct nfs_setaclres res; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SETACL], .rpc_argp = &arg, - .rpc_resp = NULL, + .rpc_resp = &res, }; int ret; @@ -2841,7 +3269,7 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl return -EOPNOTSUPP; nfs_inode_return_delegation(inode); buf_to_pages(buf, buflen, arg.acl_pages, &arg.acl_pgbase); - ret = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); + ret = nfs4_call_sync(server, &msg, &arg, &res, 1); nfs_access_zap_cache(inode); nfs_zap_acl_cache(inode); return ret; @@ -2860,10 +3288,8 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen } static int -nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, struct nfs4_state *state) +_nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, struct nfs_client *clp, struct nfs4_state *state) { - struct nfs_client *clp = server->nfs_client; - if (!clp || task->tk_status >= 0) return 0; switch(task->tk_status) { @@ -2882,8 +3308,23 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, rpc_wake_up_queued_task(&clp->cl_rpcwaitq, task); task->tk_status = 0; return -EAGAIN; +#if defined(CONFIG_NFS_V4_1) + case -NFS4ERR_BADSESSION: + case -NFS4ERR_BADSLOT: + case -NFS4ERR_BAD_HIGH_SLOT: + case -NFS4ERR_DEADSESSION: + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + case -NFS4ERR_SEQ_FALSE_RETRY: + case -NFS4ERR_SEQ_MISORDERED: + dprintk("%s ERROR %d, Reset session\n", __func__, + task->tk_status); + set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state); + task->tk_status = 0; + return -EAGAIN; +#endif /* CONFIG_NFS_V4_1 */ case -NFS4ERR_DELAY: - nfs_inc_server_stats(server, NFSIOS_DELAY); + if (server) + nfs_inc_server_stats(server, NFSIOS_DELAY); case -NFS4ERR_GRACE: rpc_delay(task, NFS4_POLL_RETRY_MAX); task->tk_status = 0; @@ -2896,6 +3337,12 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, return 0; } +static int +nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, struct nfs4_state *state) +{ + return _nfs4_async_handle_error(task, server, server->nfs_client, state); +} + int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short port, struct rpc_cred *cred) { nfs4_verifier sc_verifier; @@ -3003,6 +3450,10 @@ struct nfs4_delegreturndata { static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata) { struct nfs4_delegreturndata *data = calldata; + + nfs4_sequence_done_free_slot(data->res.server, &data->res.seq_res, + task->tk_status); + data->rpc_status = task->tk_status; if (data->rpc_status == 0) renew_lease(data->res.server, data->timestamp); @@ -3013,7 +3464,25 @@ static void nfs4_delegreturn_release(void *calldata) kfree(calldata); } +#if defined(CONFIG_NFS_V4_1) +static void nfs4_delegreturn_prepare(struct rpc_task *task, void *data) +{ + struct nfs4_delegreturndata *d_data; + + d_data = (struct nfs4_delegreturndata *)data; + + if (nfs4_setup_sequence(d_data->res.server->nfs_client, + &d_data->args.seq_args, + &d_data->res.seq_res, 1, task)) + return; + rpc_call_start(task); +} +#endif /* CONFIG_NFS_V4_1 */ + static const struct rpc_call_ops nfs4_delegreturn_ops = { +#if defined(CONFIG_NFS_V4_1) + .rpc_call_prepare = nfs4_delegreturn_prepare, +#endif /* CONFIG_NFS_V4_1 */ .rpc_call_done = nfs4_delegreturn_done, .rpc_release = nfs4_delegreturn_release, }; @@ -3035,7 +3504,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co }; int status = 0; - data = kmalloc(sizeof(*data), GFP_KERNEL); + data = kzalloc(sizeof(*data), GFP_KERNEL); if (data == NULL) return -ENOMEM; data->args.fhandle = &data->fh; @@ -3045,6 +3514,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co memcpy(&data->stateid, stateid, sizeof(data->stateid)); data->res.fattr = &data->fattr; data->res.server = server; + data->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE; nfs_fattr_init(data->res.fattr); data->timestamp = jiffies; data->rpc_status = 0; @@ -3130,7 +3600,7 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock goto out; lsp = request->fl_u.nfs4_fl.owner; arg.lock_owner.id = lsp->ls_id.id; - status = rpc_call_sync(server->client, &msg, 0); + status = nfs4_call_sync(server, &msg, &arg, &res, 1); switch (status) { case 0: request->fl_type = F_UNLCK; @@ -3190,13 +3660,14 @@ static struct nfs4_unlockdata *nfs4_alloc_unlockdata(struct file_lock *fl, struct nfs4_unlockdata *p; struct inode *inode = lsp->ls_state->inode; - p = kmalloc(sizeof(*p), GFP_KERNEL); + p = kzalloc(sizeof(*p), GFP_KERNEL); if (p == NULL) return NULL; p->arg.fh = NFS_FH(inode); p->arg.fl = &p->fl; p->arg.seqid = seqid; p->res.seqid = seqid; + p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE; p->arg.stateid = &lsp->ls_stateid; p->lsp = lsp; atomic_inc(&lsp->ls_count); @@ -3220,6 +3691,8 @@ static void nfs4_locku_done(struct rpc_task *task, void *data) { struct nfs4_unlockdata *calldata = data; + nfs4_sequence_done(calldata->server, &calldata->res.seq_res, + task->tk_status); if (RPC_ASSASSINATED(task)) return; switch (task->tk_status) { @@ -3236,8 +3709,11 @@ static void nfs4_locku_done(struct rpc_task *task, void *data) break; default: if (nfs4_async_handle_error(task, calldata->server, NULL) == -EAGAIN) - rpc_restart_call(task); + nfs4_restart_rpc(task, + calldata->server->nfs_client); } + nfs4_sequence_free_slot(calldata->server->nfs_client, + &calldata->res.seq_res); } static void nfs4_locku_prepare(struct rpc_task *task, void *data) @@ -3252,6 +3728,10 @@ static void nfs4_locku_prepare(struct rpc_task *task, void *data) return; } calldata->timestamp = jiffies; + if (nfs4_setup_sequence(calldata->server->nfs_client, + &calldata->arg.seq_args, + &calldata->res.seq_res, 1, task)) + return; rpc_call_start(task); } @@ -3344,6 +3824,7 @@ struct nfs4_lockdata { unsigned long timestamp; int rpc_status; int cancelled; + struct nfs_server *server; }; static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl, @@ -3369,7 +3850,9 @@ static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl, p->arg.lock_owner.clientid = server->nfs_client->cl_clientid; p->arg.lock_owner.id = lsp->ls_id.id; p->res.lock_seqid = p->arg.lock_seqid; + p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE; p->lsp = lsp; + p->server = server; atomic_inc(&lsp->ls_count); p->ctx = get_nfs_open_context(ctx); memcpy(&p->fl, fl, sizeof(p->fl)); @@ -3399,6 +3882,9 @@ static void nfs4_lock_prepare(struct rpc_task *task, void *calldata) } else data->arg.new_lock_owner = 0; data->timestamp = jiffies; + if (nfs4_setup_sequence(data->server->nfs_client, &data->arg.seq_args, + &data->res.seq_res, 1, task)) + return; rpc_call_start(task); dprintk("%s: done!, ret = %d\n", __func__, data->rpc_status); } @@ -3409,6 +3895,9 @@ static void nfs4_lock_done(struct rpc_task *task, void *calldata) dprintk("%s: begin!\n", __func__); + nfs4_sequence_done_free_slot(data->server, &data->res.seq_res, + task->tk_status); + data->rpc_status = task->tk_status; if (RPC_ASSASSINATED(task)) goto out; @@ -3709,10 +4198,13 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name, .page = page, .bitmask = bitmask, }; + struct nfs4_fs_locations_res res = { + .fs_locations = fs_locations, + }; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FS_LOCATIONS], .rpc_argp = &args, - .rpc_resp = fs_locations, + .rpc_resp = &res, }; int status; @@ -3720,24 +4212,663 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name, nfs_fattr_init(&fs_locations->fattr); fs_locations->server = server; fs_locations->nlocations = 0; - status = rpc_call_sync(server->client, &msg, 0); + status = nfs4_call_sync(server, &msg, &args, &res, 0); nfs_fixup_referral_attributes(&fs_locations->fattr); dprintk("%s: returned status = %d\n", __func__, status); return status; } -struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops = { +#ifdef CONFIG_NFS_V4_1 +/* + * nfs4_proc_exchange_id() + * + * Since the clientid has expired, all compounds using sessions + * associated with the stale clientid will be returning + * NFS4ERR_BADSESSION in the sequence operation, and will therefore + * be in some phase of session reset. + */ +static int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred) +{ + nfs4_verifier verifier; + struct nfs41_exchange_id_args args = { + .client = clp, + .flags = clp->cl_exchange_flags, + }; + struct nfs41_exchange_id_res res = { + .client = clp, + }; + int status; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_EXCHANGE_ID], + .rpc_argp = &args, + .rpc_resp = &res, + .rpc_cred = cred, + }; + __be32 *p; + + dprintk("--> %s\n", __func__); + BUG_ON(clp == NULL); + + p = (u32 *)verifier.data; + *p++ = htonl((u32)clp->cl_boot_time.tv_sec); + *p = htonl((u32)clp->cl_boot_time.tv_nsec); + args.verifier = &verifier; + + while (1) { + args.id_len = scnprintf(args.id, sizeof(args.id), + "%s/%s %u", + clp->cl_ipaddr, + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_ADDR), + clp->cl_id_uniquifier); + + status = rpc_call_sync(clp->cl_rpcclient, &msg, 0); + + if (status != NFS4ERR_CLID_INUSE) + break; + + if (signalled()) + break; + + if (++clp->cl_id_uniquifier == 0) + break; + } + + dprintk("<-- %s status= %d\n", __func__, status); + return status; +} + +struct nfs4_get_lease_time_data { + struct nfs4_get_lease_time_args *args; + struct nfs4_get_lease_time_res *res; + struct nfs_client *clp; +}; + +static void nfs4_get_lease_time_prepare(struct rpc_task *task, + void *calldata) +{ + int ret; + struct nfs4_get_lease_time_data *data = + (struct nfs4_get_lease_time_data *)calldata; + + dprintk("--> %s\n", __func__); + /* just setup sequence, do not trigger session recovery + since we're invoked within one */ + ret = nfs41_setup_sequence(data->clp->cl_session, + &data->args->la_seq_args, + &data->res->lr_seq_res, 0, task); + + BUG_ON(ret == -EAGAIN); + rpc_call_start(task); + dprintk("<-- %s\n", __func__); +} + +/* + * Called from nfs4_state_manager thread for session setup, so don't recover + * from sequence operation or clientid errors. + */ +static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_get_lease_time_data *data = + (struct nfs4_get_lease_time_data *)calldata; + + dprintk("--> %s\n", __func__); + nfs41_sequence_done(data->clp, &data->res->lr_seq_res, task->tk_status); + switch (task->tk_status) { + case -NFS4ERR_DELAY: + case -NFS4ERR_GRACE: + dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status); + rpc_delay(task, NFS4_POLL_RETRY_MIN); + task->tk_status = 0; + nfs4_restart_rpc(task, data->clp); + return; + } + nfs41_sequence_free_slot(data->clp, &data->res->lr_seq_res); + dprintk("<-- %s\n", __func__); +} + +struct rpc_call_ops nfs4_get_lease_time_ops = { + .rpc_call_prepare = nfs4_get_lease_time_prepare, + .rpc_call_done = nfs4_get_lease_time_done, +}; + +int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo) +{ + struct rpc_task *task; + struct nfs4_get_lease_time_args args; + struct nfs4_get_lease_time_res res = { + .lr_fsinfo = fsinfo, + }; + struct nfs4_get_lease_time_data data = { + .args = &args, + .res = &res, + .clp = clp, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GET_LEASE_TIME], + .rpc_argp = &args, + .rpc_resp = &res, + }; + struct rpc_task_setup task_setup = { + .rpc_client = clp->cl_rpcclient, + .rpc_message = &msg, + .callback_ops = &nfs4_get_lease_time_ops, + .callback_data = &data + }; + int status; + + res.lr_seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE; + dprintk("--> %s\n", __func__); + task = rpc_run_task(&task_setup); + + if (IS_ERR(task)) + status = PTR_ERR(task); + else { + status = task->tk_status; + rpc_put_task(task); + } + dprintk("<-- %s return %d\n", __func__, status); + + return status; +} + +/* Reset a slot table */ +static int nfs4_reset_slot_table(struct nfs4_session *session) +{ + struct nfs4_slot_table *tbl = &session->fc_slot_table; + int i, max_slots = session->fc_attrs.max_reqs; + int old_max_slots = session->fc_slot_table.max_slots; + int ret = 0; + + dprintk("--> %s: max_reqs=%u, tbl %p\n", __func__, + session->fc_attrs.max_reqs, tbl); + + /* Until we have dynamic slot table adjustment, insist + * upon the same slot table size */ + if (max_slots != old_max_slots) { + dprintk("%s reset slot table does't match old\n", + __func__); + ret = -EINVAL; /*XXX NFS4ERR_REQ_TOO_BIG ? */ + goto out; + } + spin_lock(&tbl->slot_tbl_lock); + for (i = 0; i < max_slots; ++i) + tbl->slots[i].seq_nr = 1; + tbl->highest_used_slotid = -1; + spin_unlock(&tbl->slot_tbl_lock); + dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__, + tbl, tbl->slots, tbl->max_slots); +out: + dprintk("<-- %s: return %d\n", __func__, ret); + return ret; +} + +/* + * Initialize slot table + */ +static int nfs4_init_slot_table(struct nfs4_session *session) +{ + struct nfs4_slot_table *tbl = &session->fc_slot_table; + int i, max_slots = session->fc_attrs.max_reqs; + struct nfs4_slot *slot; + int ret = -ENOMEM; + + BUG_ON(max_slots > NFS4_MAX_SLOT_TABLE); + + dprintk("--> %s: max_reqs=%u\n", __func__, + session->fc_attrs.max_reqs); + + slot = kcalloc(max_slots, sizeof(struct nfs4_slot), GFP_KERNEL); + if (!slot) + goto out; + for (i = 0; i < max_slots; ++i) + slot[i].seq_nr = 1; + ret = 0; + + spin_lock(&tbl->slot_tbl_lock); + if (tbl->slots != NULL) { + spin_unlock(&tbl->slot_tbl_lock); + dprintk("%s: slot table already initialized. tbl=%p slots=%p\n", + __func__, tbl, tbl->slots); + WARN_ON(1); + goto out_free; + } + tbl->max_slots = max_slots; + tbl->slots = slot; + tbl->highest_used_slotid = -1; /* no slot is currently used */ + spin_unlock(&tbl->slot_tbl_lock); + dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__, + tbl, tbl->slots, tbl->max_slots); +out: + dprintk("<-- %s: return %d\n", __func__, ret); + return ret; +out_free: + kfree(slot); + goto out; +} + +/* Destroy the slot table */ +static void nfs4_destroy_slot_table(struct nfs4_session *session) +{ + if (session->fc_slot_table.slots == NULL) + return; + kfree(session->fc_slot_table.slots); + session->fc_slot_table.slots = NULL; + return; +} + +struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp) +{ + struct nfs4_session *session; + struct nfs4_slot_table *tbl; + + session = kzalloc(sizeof(struct nfs4_session), GFP_KERNEL); + if (!session) + return NULL; + + set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state); + /* + * The create session reply races with the server back + * channel probe. Mark the client NFS_CS_SESSION_INITING + * so that the client back channel can find the + * nfs_client struct + */ + clp->cl_cons_state = NFS_CS_SESSION_INITING; + + tbl = &session->fc_slot_table; + spin_lock_init(&tbl->slot_tbl_lock); + rpc_init_wait_queue(&tbl->slot_tbl_waitq, "Slot table"); + session->clp = clp; + return session; +} + +void nfs4_destroy_session(struct nfs4_session *session) +{ + nfs4_destroy_slot_table(session); + kfree(session); +} + +/* + * Initialize the values to be used by the client in CREATE_SESSION + * If nfs4_init_session set the fore channel request and response sizes, + * use them. + * + * Set the back channel max_resp_sz_cached to zero to force the client to + * always set csa_cachethis to FALSE because the current implementation + * of the back channel DRC only supports caching the CB_SEQUENCE operation. + */ +static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args) +{ + struct nfs4_session *session = args->client->cl_session; + unsigned int mxrqst_sz = session->fc_attrs.max_rqst_sz, + mxresp_sz = session->fc_attrs.max_resp_sz; + + if (mxrqst_sz == 0) + mxrqst_sz = NFS_MAX_FILE_IO_SIZE; + if (mxresp_sz == 0) + mxresp_sz = NFS_MAX_FILE_IO_SIZE; + /* Fore channel attributes */ + args->fc_attrs.headerpadsz = 0; + args->fc_attrs.max_rqst_sz = mxrqst_sz; + args->fc_attrs.max_resp_sz = mxresp_sz; + args->fc_attrs.max_resp_sz_cached = mxresp_sz; + args->fc_attrs.max_ops = NFS4_MAX_OPS; + args->fc_attrs.max_reqs = session->clp->cl_rpcclient->cl_xprt->max_reqs; + + dprintk("%s: Fore Channel : max_rqst_sz=%u max_resp_sz=%u " + "max_resp_sz_cached=%u max_ops=%u max_reqs=%u\n", + __func__, + args->fc_attrs.max_rqst_sz, args->fc_attrs.max_resp_sz, + args->fc_attrs.max_resp_sz_cached, args->fc_attrs.max_ops, + args->fc_attrs.max_reqs); + + /* Back channel attributes */ + args->bc_attrs.headerpadsz = 0; + args->bc_attrs.max_rqst_sz = PAGE_SIZE; + args->bc_attrs.max_resp_sz = PAGE_SIZE; + args->bc_attrs.max_resp_sz_cached = 0; + args->bc_attrs.max_ops = NFS4_MAX_BACK_CHANNEL_OPS; + args->bc_attrs.max_reqs = 1; + + dprintk("%s: Back Channel : max_rqst_sz=%u max_resp_sz=%u " + "max_resp_sz_cached=%u max_ops=%u max_reqs=%u\n", + __func__, + args->bc_attrs.max_rqst_sz, args->bc_attrs.max_resp_sz, + args->bc_attrs.max_resp_sz_cached, args->bc_attrs.max_ops, + args->bc_attrs.max_reqs); +} + +static int _verify_channel_attr(char *chan, char *attr_name, u32 sent, u32 rcvd) +{ + if (rcvd <= sent) + return 0; + printk(KERN_WARNING "%s: Session INVALID: %s channel %s increased. " + "sent=%u rcvd=%u\n", __func__, chan, attr_name, sent, rcvd); + return -EINVAL; +} + +#define _verify_fore_channel_attr(_name_) \ + _verify_channel_attr("fore", #_name_, \ + args->fc_attrs._name_, \ + session->fc_attrs._name_) + +#define _verify_back_channel_attr(_name_) \ + _verify_channel_attr("back", #_name_, \ + args->bc_attrs._name_, \ + session->bc_attrs._name_) + +/* + * The server is not allowed to increase the fore channel header pad size, + * maximum response size, or maximum number of operations. + * + * The back channel attributes are only negotiatied down: We send what the + * (back channel) server insists upon. + */ +static int nfs4_verify_channel_attrs(struct nfs41_create_session_args *args, + struct nfs4_session *session) +{ + int ret = 0; + + ret |= _verify_fore_channel_attr(headerpadsz); + ret |= _verify_fore_channel_attr(max_resp_sz); + ret |= _verify_fore_channel_attr(max_ops); + + ret |= _verify_back_channel_attr(headerpadsz); + ret |= _verify_back_channel_attr(max_rqst_sz); + ret |= _verify_back_channel_attr(max_resp_sz); + ret |= _verify_back_channel_attr(max_resp_sz_cached); + ret |= _verify_back_channel_attr(max_ops); + ret |= _verify_back_channel_attr(max_reqs); + + return ret; +} + +static int _nfs4_proc_create_session(struct nfs_client *clp) +{ + struct nfs4_session *session = clp->cl_session; + struct nfs41_create_session_args args = { + .client = clp, + .cb_program = NFS4_CALLBACK, + }; + struct nfs41_create_session_res res = { + .client = clp, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CREATE_SESSION], + .rpc_argp = &args, + .rpc_resp = &res, + }; + int status; + + nfs4_init_channel_attrs(&args); + args.flags = (SESSION4_PERSIST); + + status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0); + + if (!status) + /* Verify the session's negotiated channel_attrs values */ + status = nfs4_verify_channel_attrs(&args, session); + if (!status) { + /* Increment the clientid slot sequence id */ + clp->cl_seqid++; + } + + return status; +} + +/* + * Issues a CREATE_SESSION operation to the server. + * It is the responsibility of the caller to verify the session is + * expired before calling this routine. + */ +int nfs4_proc_create_session(struct nfs_client *clp, int reset) +{ + int status; + unsigned *ptr; + struct nfs_fsinfo fsinfo; + struct nfs4_session *session = clp->cl_session; + + dprintk("--> %s clp=%p session=%p\n", __func__, clp, session); + + status = _nfs4_proc_create_session(clp); + if (status) + goto out; + + /* Init or reset the fore channel */ + if (reset) + status = nfs4_reset_slot_table(session); + else + status = nfs4_init_slot_table(session); + dprintk("fore channel slot table initialization returned %d\n", status); + if (status) + goto out; + + ptr = (unsigned *)&session->sess_id.data[0]; + dprintk("%s client>seqid %d sessionid %u:%u:%u:%u\n", __func__, + clp->cl_seqid, ptr[0], ptr[1], ptr[2], ptr[3]); + + if (reset) + /* Lease time is aleady set */ + goto out; + + /* Get the lease time */ + status = nfs4_proc_get_lease_time(clp, &fsinfo); + if (status == 0) { + /* Update lease time and schedule renewal */ + spin_lock(&clp->cl_lock); + clp->cl_lease_time = fsinfo.lease_time * HZ; + clp->cl_last_renewal = jiffies; + clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); + spin_unlock(&clp->cl_lock); + + nfs4_schedule_state_renewal(clp); + } +out: + dprintk("<-- %s\n", __func__); + return status; +} + +/* + * Issue the over-the-wire RPC DESTROY_SESSION. + * The caller must serialize access to this routine. + */ +int nfs4_proc_destroy_session(struct nfs4_session *session) +{ + int status = 0; + struct rpc_message msg; + + dprintk("--> nfs4_proc_destroy_session\n"); + + /* session is still being setup */ + if (session->clp->cl_cons_state != NFS_CS_READY) + return status; + + msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_DESTROY_SESSION]; + msg.rpc_argp = session; + msg.rpc_resp = NULL; + msg.rpc_cred = NULL; + status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0); + + if (status) + printk(KERN_WARNING + "Got error %d from the server on DESTROY_SESSION. " + "Session has been destroyed regardless...\n", status); + + dprintk("<-- nfs4_proc_destroy_session\n"); + return status; +} + +/* + * Renew the cl_session lease. + */ +static int nfs4_proc_sequence(struct nfs_client *clp, struct rpc_cred *cred) +{ + struct nfs4_sequence_args args; + struct nfs4_sequence_res res; + + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SEQUENCE], + .rpc_argp = &args, + .rpc_resp = &res, + .rpc_cred = cred, + }; + + args.sa_cache_this = 0; + + return nfs4_call_sync_sequence(clp, clp->cl_rpcclient, &msg, &args, + &res, 0); +} + +void nfs41_sequence_call_done(struct rpc_task *task, void *data) +{ + struct nfs_client *clp = (struct nfs_client *)data; + + nfs41_sequence_done(clp, task->tk_msg.rpc_resp, task->tk_status); + + if (task->tk_status < 0) { + dprintk("%s ERROR %d\n", __func__, task->tk_status); + + if (_nfs4_async_handle_error(task, NULL, clp, NULL) + == -EAGAIN) { + nfs4_restart_rpc(task, clp); + return; + } + } + nfs41_sequence_free_slot(clp, task->tk_msg.rpc_resp); + dprintk("%s rpc_cred %p\n", __func__, task->tk_msg.rpc_cred); + + put_rpccred(task->tk_msg.rpc_cred); + kfree(task->tk_msg.rpc_argp); + kfree(task->tk_msg.rpc_resp); + + dprintk("<-- %s\n", __func__); +} + +static void nfs41_sequence_prepare(struct rpc_task *task, void *data) +{ + struct nfs_client *clp; + struct nfs4_sequence_args *args; + struct nfs4_sequence_res *res; + + clp = (struct nfs_client *)data; + args = task->tk_msg.rpc_argp; + res = task->tk_msg.rpc_resp; + + if (nfs4_setup_sequence(clp, args, res, 0, task)) + return; + rpc_call_start(task); +} + +static const struct rpc_call_ops nfs41_sequence_ops = { + .rpc_call_done = nfs41_sequence_call_done, + .rpc_call_prepare = nfs41_sequence_prepare, +}; + +static int nfs41_proc_async_sequence(struct nfs_client *clp, + struct rpc_cred *cred) +{ + struct nfs4_sequence_args *args; + struct nfs4_sequence_res *res; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SEQUENCE], + .rpc_cred = cred, + }; + + args = kzalloc(sizeof(*args), GFP_KERNEL); + if (!args) + return -ENOMEM; + res = kzalloc(sizeof(*res), GFP_KERNEL); + if (!res) { + kfree(args); + return -ENOMEM; + } + res->sr_slotid = NFS4_MAX_SLOT_TABLE; + msg.rpc_argp = args; + msg.rpc_resp = res; + + return rpc_call_async(clp->cl_rpcclient, &msg, RPC_TASK_SOFT, + &nfs41_sequence_ops, (void *)clp); +} + +#endif /* CONFIG_NFS_V4_1 */ + +struct nfs4_state_recovery_ops nfs40_reboot_recovery_ops = { + .owner_flag_bit = NFS_OWNER_RECLAIM_REBOOT, + .state_flag_bit = NFS_STATE_RECLAIM_REBOOT, + .recover_open = nfs4_open_reclaim, + .recover_lock = nfs4_lock_reclaim, + .establish_clid = nfs4_init_clientid, + .get_clid_cred = nfs4_get_setclientid_cred, +}; + +#if defined(CONFIG_NFS_V4_1) +struct nfs4_state_recovery_ops nfs41_reboot_recovery_ops = { .owner_flag_bit = NFS_OWNER_RECLAIM_REBOOT, .state_flag_bit = NFS_STATE_RECLAIM_REBOOT, .recover_open = nfs4_open_reclaim, .recover_lock = nfs4_lock_reclaim, + .establish_clid = nfs4_proc_exchange_id, + .get_clid_cred = nfs4_get_exchange_id_cred, +}; +#endif /* CONFIG_NFS_V4_1 */ + +struct nfs4_state_recovery_ops nfs40_nograce_recovery_ops = { + .owner_flag_bit = NFS_OWNER_RECLAIM_NOGRACE, + .state_flag_bit = NFS_STATE_RECLAIM_NOGRACE, + .recover_open = nfs4_open_expired, + .recover_lock = nfs4_lock_expired, + .establish_clid = nfs4_init_clientid, + .get_clid_cred = nfs4_get_setclientid_cred, }; -struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops = { +#if defined(CONFIG_NFS_V4_1) +struct nfs4_state_recovery_ops nfs41_nograce_recovery_ops = { .owner_flag_bit = NFS_OWNER_RECLAIM_NOGRACE, .state_flag_bit = NFS_STATE_RECLAIM_NOGRACE, .recover_open = nfs4_open_expired, .recover_lock = nfs4_lock_expired, + .establish_clid = nfs4_proc_exchange_id, + .get_clid_cred = nfs4_get_exchange_id_cred, +}; +#endif /* CONFIG_NFS_V4_1 */ + +struct nfs4_state_maintenance_ops nfs40_state_renewal_ops = { + .sched_state_renewal = nfs4_proc_async_renew, + .get_state_renewal_cred_locked = nfs4_get_renew_cred_locked, + .renew_lease = nfs4_proc_renew, +}; + +#if defined(CONFIG_NFS_V4_1) +struct nfs4_state_maintenance_ops nfs41_state_renewal_ops = { + .sched_state_renewal = nfs41_proc_async_sequence, + .get_state_renewal_cred_locked = nfs4_get_machine_cred_locked, + .renew_lease = nfs4_proc_sequence, +}; +#endif + +/* + * Per minor version reboot and network partition recovery ops + */ + +struct nfs4_state_recovery_ops *nfs4_reboot_recovery_ops[] = { + &nfs40_reboot_recovery_ops, +#if defined(CONFIG_NFS_V4_1) + &nfs41_reboot_recovery_ops, +#endif +}; + +struct nfs4_state_recovery_ops *nfs4_nograce_recovery_ops[] = { + &nfs40_nograce_recovery_ops, +#if defined(CONFIG_NFS_V4_1) + &nfs41_nograce_recovery_ops, +#endif +}; + +struct nfs4_state_maintenance_ops *nfs4_state_renewal_ops[] = { + &nfs40_state_renewal_ops, +#if defined(CONFIG_NFS_V4_1) + &nfs41_state_renewal_ops, +#endif }; static const struct inode_operations nfs4_file_inode_operations = { diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c index f524e932ff7b..e27c6cef18f2 100644 --- a/fs/nfs/nfs4renewd.c +++ b/fs/nfs/nfs4renewd.c @@ -59,12 +59,14 @@ void nfs4_renew_state(struct work_struct *work) { + struct nfs4_state_maintenance_ops *ops; struct nfs_client *clp = container_of(work, struct nfs_client, cl_renewd.work); struct rpc_cred *cred; long lease, timeout; unsigned long last, now; + ops = nfs4_state_renewal_ops[clp->cl_minorversion]; dprintk("%s: start\n", __func__); /* Are there any active superblocks? */ if (list_empty(&clp->cl_superblocks)) @@ -76,7 +78,7 @@ nfs4_renew_state(struct work_struct *work) timeout = (2 * lease) / 3 + (long)last - (long)now; /* Are we close to a lease timeout? */ if (time_after(now, last + lease/3)) { - cred = nfs4_get_renew_cred_locked(clp); + cred = ops->get_state_renewal_cred_locked(clp); spin_unlock(&clp->cl_lock); if (cred == NULL) { if (list_empty(&clp->cl_delegations)) { @@ -86,7 +88,7 @@ nfs4_renew_state(struct work_struct *work) nfs_expire_all_delegations(clp); } else { /* Queue an asynchronous RENEW. */ - nfs4_proc_async_renew(clp, cred); + ops->sched_state_renewal(clp, cred); put_rpccred(cred); } timeout = (2 * lease) / 3; diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 0298e909559f..c16f24dd0e3e 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -60,7 +60,7 @@ const nfs4_stateid zero_stateid; static LIST_HEAD(nfs4_clientid_list); -static int nfs4_init_client(struct nfs_client *clp, struct rpc_cred *cred) +int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred) { unsigned short port; int status; @@ -77,7 +77,7 @@ static int nfs4_init_client(struct nfs_client *clp, struct rpc_cred *cred) return status; } -static struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp) +struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp) { struct rpc_cred *cred = NULL; @@ -114,17 +114,21 @@ struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp) return cred; } -static struct rpc_cred *nfs4_get_renew_cred(struct nfs_client *clp) +#if defined(CONFIG_NFS_V4_1) + +struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp) { struct rpc_cred *cred; spin_lock(&clp->cl_lock); - cred = nfs4_get_renew_cred_locked(clp); + cred = nfs4_get_machine_cred_locked(clp); spin_unlock(&clp->cl_lock); return cred; } -static struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp) +#endif /* CONFIG_NFS_V4_1 */ + +struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp) { struct nfs4_state_owner *sp; struct rb_node *pos; @@ -738,12 +742,14 @@ static void nfs_increment_seqid(int status, struct nfs_seqid *seqid) void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid) { - if (status == -NFS4ERR_BAD_SEQID) { - struct nfs4_state_owner *sp = container_of(seqid->sequence, - struct nfs4_state_owner, so_seqid); + struct nfs4_state_owner *sp = container_of(seqid->sequence, + struct nfs4_state_owner, so_seqid); + struct nfs_server *server = sp->so_server; + + if (status == -NFS4ERR_BAD_SEQID) nfs4_drop_state_owner(sp); - } - nfs_increment_seqid(status, seqid); + if (!nfs4_has_session(server->nfs_client)) + nfs_increment_seqid(status, seqid); } /* @@ -1042,6 +1048,14 @@ static void nfs4_recovery_handle_error(struct nfs_client *clp, int error) case -NFS4ERR_EXPIRED: set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); nfs4_state_start_reclaim_nograce(clp); + case -NFS4ERR_BADSESSION: + case -NFS4ERR_BADSLOT: + case -NFS4ERR_BAD_HIGH_SLOT: + case -NFS4ERR_DEADSESSION: + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + case -NFS4ERR_SEQ_FALSE_RETRY: + case -NFS4ERR_SEQ_MISORDERED: + set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state); } } @@ -1075,18 +1089,22 @@ restart: static int nfs4_check_lease(struct nfs_client *clp) { struct rpc_cred *cred; + struct nfs4_state_maintenance_ops *ops = + nfs4_state_renewal_ops[clp->cl_minorversion]; int status = -NFS4ERR_EXPIRED; /* Is the client already known to have an expired lease? */ if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) return 0; - cred = nfs4_get_renew_cred(clp); + spin_lock(&clp->cl_lock); + cred = ops->get_state_renewal_cred_locked(clp); + spin_unlock(&clp->cl_lock); if (cred == NULL) { cred = nfs4_get_setclientid_cred(clp); if (cred == NULL) goto out; } - status = nfs4_proc_renew(clp, cred); + status = ops->renew_lease(clp, cred); put_rpccred(cred); out: nfs4_recovery_handle_error(clp, status); @@ -1096,21 +1114,77 @@ out: static int nfs4_reclaim_lease(struct nfs_client *clp) { struct rpc_cred *cred; + struct nfs4_state_recovery_ops *ops = + nfs4_reboot_recovery_ops[clp->cl_minorversion]; int status = -ENOENT; - cred = nfs4_get_setclientid_cred(clp); + cred = ops->get_clid_cred(clp); if (cred != NULL) { - status = nfs4_init_client(clp, cred); + status = ops->establish_clid(clp, cred); put_rpccred(cred); /* Handle case where the user hasn't set up machine creds */ if (status == -EACCES && cred == clp->cl_machine_cred) { nfs4_clear_machine_cred(clp); status = -EAGAIN; } + if (status == -NFS4ERR_MINOR_VERS_MISMATCH) + status = -EPROTONOSUPPORT; } return status; } +#ifdef CONFIG_NFS_V4_1 +static void nfs4_session_recovery_handle_error(struct nfs_client *clp, int err) +{ + switch (err) { + case -NFS4ERR_STALE_CLIENTID: + set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); + set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state); + } +} + +static int nfs4_reset_session(struct nfs_client *clp) +{ + int status; + + status = nfs4_proc_destroy_session(clp->cl_session); + if (status && status != -NFS4ERR_BADSESSION && + status != -NFS4ERR_DEADSESSION) { + nfs4_session_recovery_handle_error(clp, status); + goto out; + } + + memset(clp->cl_session->sess_id.data, 0, NFS4_MAX_SESSIONID_LEN); + status = nfs4_proc_create_session(clp, 1); + if (status) + nfs4_session_recovery_handle_error(clp, status); + /* fall through*/ +out: + /* Wake up the next rpc task even on error */ + rpc_wake_up_next(&clp->cl_session->fc_slot_table.slot_tbl_waitq); + return status; +} + +static int nfs4_initialize_session(struct nfs_client *clp) +{ + int status; + + status = nfs4_proc_create_session(clp, 0); + if (!status) { + nfs_mark_client_ready(clp, NFS_CS_READY); + } else if (status == -NFS4ERR_STALE_CLIENTID) { + set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); + set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state); + } else { + nfs_mark_client_ready(clp, status); + } + return status; +} +#else /* CONFIG_NFS_V4_1 */ +static int nfs4_reset_session(struct nfs_client *clp) { return 0; } +static int nfs4_initialize_session(struct nfs_client *clp) { return 0; } +#endif /* CONFIG_NFS_V4_1 */ + static void nfs4_state_manager(struct nfs_client *clp) { int status = 0; @@ -1124,6 +1198,9 @@ static void nfs4_state_manager(struct nfs_client *clp) set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); if (status == -EAGAIN) continue; + if (clp->cl_cons_state == + NFS_CS_SESSION_INITING) + nfs_mark_client_ready(clp, status); goto out_error; } clear_bit(NFS4CLNT_CHECK_LEASE, &clp->cl_state); @@ -1134,25 +1211,44 @@ static void nfs4_state_manager(struct nfs_client *clp) if (status != 0) continue; } - + /* Initialize or reset the session */ + if (nfs4_has_session(clp) && + test_and_clear_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)) { + if (clp->cl_cons_state == NFS_CS_SESSION_INITING) + status = nfs4_initialize_session(clp); + else + status = nfs4_reset_session(clp); + if (status) { + if (status == -NFS4ERR_STALE_CLIENTID) + continue; + goto out_error; + } + } /* First recover reboot state... */ if (test_and_clear_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state)) { - status = nfs4_do_reclaim(clp, &nfs4_reboot_recovery_ops); + status = nfs4_do_reclaim(clp, + nfs4_reboot_recovery_ops[clp->cl_minorversion]); if (status == -NFS4ERR_STALE_CLIENTID) continue; + if (test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)) + continue; nfs4_state_end_reclaim_reboot(clp); continue; } /* Now recover expired state... */ if (test_and_clear_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state)) { - status = nfs4_do_reclaim(clp, &nfs4_nograce_recovery_ops); + status = nfs4_do_reclaim(clp, + nfs4_nograce_recovery_ops[clp->cl_minorversion]); if (status < 0) { set_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state); if (status == -NFS4ERR_STALE_CLIENTID) continue; if (status == -NFS4ERR_EXPIRED) continue; + if (test_bit(NFS4CLNT_SESSION_SETUP, + &clp->cl_state)) + continue; goto out_error; } else nfs4_state_end_reclaim_nograce(clp); diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 1690f0e44b91..80af0aeaaf7d 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -192,12 +192,16 @@ static int nfs4_stat_to_errno(int); decode_verifier_maxsz) #define encode_remove_maxsz (op_encode_hdr_maxsz + \ nfs4_name_maxsz) +#define decode_remove_maxsz (op_decode_hdr_maxsz + \ + decode_change_info_maxsz) #define encode_rename_maxsz (op_encode_hdr_maxsz + \ 2 * nfs4_name_maxsz) -#define decode_rename_maxsz (op_decode_hdr_maxsz + 5 + 5) +#define decode_rename_maxsz (op_decode_hdr_maxsz + \ + decode_change_info_maxsz + \ + decode_change_info_maxsz) #define encode_link_maxsz (op_encode_hdr_maxsz + \ nfs4_name_maxsz) -#define decode_link_maxsz (op_decode_hdr_maxsz + 5) +#define decode_link_maxsz (op_decode_hdr_maxsz + decode_change_info_maxsz) #define encode_lock_maxsz (op_encode_hdr_maxsz + \ 7 + \ 1 + encode_stateid_maxsz + 8) @@ -240,43 +244,115 @@ static int nfs4_stat_to_errno(int); (encode_getattr_maxsz) #define decode_fs_locations_maxsz \ (0) + +#if defined(CONFIG_NFS_V4_1) +#define NFS4_MAX_MACHINE_NAME_LEN (64) + +#define encode_exchange_id_maxsz (op_encode_hdr_maxsz + \ + encode_verifier_maxsz + \ + 1 /* co_ownerid.len */ + \ + XDR_QUADLEN(NFS4_EXCHANGE_ID_LEN) + \ + 1 /* flags */ + \ + 1 /* spa_how */ + \ + 0 /* SP4_NONE (for now) */ + \ + 1 /* zero implemetation id array */) +#define decode_exchange_id_maxsz (op_decode_hdr_maxsz + \ + 2 /* eir_clientid */ + \ + 1 /* eir_sequenceid */ + \ + 1 /* eir_flags */ + \ + 1 /* spr_how */ + \ + 0 /* SP4_NONE (for now) */ + \ + 2 /* eir_server_owner.so_minor_id */ + \ + /* eir_server_owner.so_major_id<> */ \ + XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 + \ + /* eir_server_scope<> */ \ + XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 + \ + 1 /* eir_server_impl_id array length */ + \ + 0 /* ignored eir_server_impl_id contents */) +#define encode_channel_attrs_maxsz (6 + 1 /* ca_rdma_ird.len (0) */) +#define decode_channel_attrs_maxsz (6 + \ + 1 /* ca_rdma_ird.len */ + \ + 1 /* ca_rdma_ird */) +#define encode_create_session_maxsz (op_encode_hdr_maxsz + \ + 2 /* csa_clientid */ + \ + 1 /* csa_sequence */ + \ + 1 /* csa_flags */ + \ + encode_channel_attrs_maxsz + \ + encode_channel_attrs_maxsz + \ + 1 /* csa_cb_program */ + \ + 1 /* csa_sec_parms.len (1) */ + \ + 1 /* cb_secflavor (AUTH_SYS) */ + \ + 1 /* stamp */ + \ + 1 /* machinename.len */ + \ + XDR_QUADLEN(NFS4_MAX_MACHINE_NAME_LEN) + \ + 1 /* uid */ + \ + 1 /* gid */ + \ + 1 /* gids.len (0) */) +#define decode_create_session_maxsz (op_decode_hdr_maxsz + \ + XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + \ + 1 /* csr_sequence */ + \ + 1 /* csr_flags */ + \ + decode_channel_attrs_maxsz + \ + decode_channel_attrs_maxsz) +#define encode_destroy_session_maxsz (op_encode_hdr_maxsz + 4) +#define decode_destroy_session_maxsz (op_decode_hdr_maxsz) +#define encode_sequence_maxsz (op_encode_hdr_maxsz + \ + XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 4) +#define decode_sequence_maxsz (op_decode_hdr_maxsz + \ + XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 5) +#else /* CONFIG_NFS_V4_1 */ +#define encode_sequence_maxsz 0 +#define decode_sequence_maxsz 0 +#endif /* CONFIG_NFS_V4_1 */ + #define NFS4_enc_compound_sz (1024) /* XXX: large enough? */ #define NFS4_dec_compound_sz (1024) /* XXX: large enough? */ #define NFS4_enc_read_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_read_maxsz) #define NFS4_dec_read_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_read_maxsz) #define NFS4_enc_readlink_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_readlink_maxsz) #define NFS4_dec_readlink_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_readlink_maxsz) #define NFS4_enc_readdir_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_readdir_maxsz) #define NFS4_dec_readdir_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_readdir_maxsz) #define NFS4_enc_write_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_write_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_write_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_write_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_commit_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_commit_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_commit_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_commit_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_open_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_savefh_maxsz + \ encode_open_maxsz + \ @@ -285,6 +361,7 @@ static int nfs4_stat_to_errno(int); encode_restorefh_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_open_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_savefh_maxsz + \ decode_open_maxsz + \ @@ -301,43 +378,53 @@ static int nfs4_stat_to_errno(int); decode_putfh_maxsz + \ decode_open_confirm_maxsz) #define NFS4_enc_open_noattr_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_open_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_open_noattr_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_open_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_open_downgrade_sz \ (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_open_downgrade_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_open_downgrade_sz \ (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_open_downgrade_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_close_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_close_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_close_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_close_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_setattr_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_setattr_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_setattr_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_setattr_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_fsinfo_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_fsinfo_maxsz) #define NFS4_dec_fsinfo_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_fsinfo_maxsz) #define NFS4_enc_renew_sz (compound_encode_hdr_maxsz + \ @@ -359,64 +446,81 @@ static int nfs4_stat_to_errno(int); decode_putrootfh_maxsz + \ decode_fsinfo_maxsz) #define NFS4_enc_lock_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_lock_maxsz) #define NFS4_dec_lock_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_lock_maxsz) #define NFS4_enc_lockt_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_lockt_maxsz) #define NFS4_dec_lockt_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_lockt_maxsz) #define NFS4_enc_locku_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_locku_maxsz) #define NFS4_dec_locku_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_locku_maxsz) #define NFS4_enc_access_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_access_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_access_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_access_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_getattr_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_getattr_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_lookup_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_lookup_maxsz + \ encode_getattr_maxsz + \ encode_getfh_maxsz) #define NFS4_dec_lookup_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_lookup_maxsz + \ decode_getattr_maxsz + \ decode_getfh_maxsz) #define NFS4_enc_lookup_root_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putrootfh_maxsz + \ encode_getattr_maxsz + \ encode_getfh_maxsz) #define NFS4_dec_lookup_root_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putrootfh_maxsz + \ decode_getattr_maxsz + \ decode_getfh_maxsz) #define NFS4_enc_remove_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_remove_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_remove_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 5 + \ + decode_remove_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_rename_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_savefh_maxsz + \ encode_putfh_maxsz + \ @@ -425,6 +529,7 @@ static int nfs4_stat_to_errno(int); encode_restorefh_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_rename_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_savefh_maxsz + \ decode_putfh_maxsz + \ @@ -433,6 +538,7 @@ static int nfs4_stat_to_errno(int); decode_restorefh_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_link_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_savefh_maxsz + \ encode_putfh_maxsz + \ @@ -441,6 +547,7 @@ static int nfs4_stat_to_errno(int); encode_restorefh_maxsz + \ decode_getattr_maxsz) #define NFS4_dec_link_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_savefh_maxsz + \ decode_putfh_maxsz + \ @@ -449,16 +556,19 @@ static int nfs4_stat_to_errno(int); decode_restorefh_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_symlink_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_symlink_maxsz + \ encode_getattr_maxsz + \ encode_getfh_maxsz) #define NFS4_dec_symlink_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_symlink_maxsz + \ decode_getattr_maxsz + \ decode_getfh_maxsz) #define NFS4_enc_create_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_savefh_maxsz + \ encode_create_maxsz + \ @@ -467,6 +577,7 @@ static int nfs4_stat_to_errno(int); encode_restorefh_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_create_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_savefh_maxsz + \ decode_create_maxsz + \ @@ -475,52 +586,98 @@ static int nfs4_stat_to_errno(int); decode_restorefh_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_pathconf_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_pathconf_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_statfs_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_statfs_maxsz) #define NFS4_dec_statfs_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_statfs_maxsz) #define NFS4_enc_server_caps_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_server_caps_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_delegreturn_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_delegreturn_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_delegreturn_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_delegreturn_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_getacl_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_getacl_maxsz) #define NFS4_dec_getacl_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_getacl_maxsz) #define NFS4_enc_setacl_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_setacl_maxsz) #define NFS4_dec_setacl_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_setacl_maxsz) #define NFS4_enc_fs_locations_sz \ (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_lookup_maxsz + \ encode_fs_locations_maxsz) #define NFS4_dec_fs_locations_sz \ (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_lookup_maxsz + \ decode_fs_locations_maxsz) +#if defined(CONFIG_NFS_V4_1) +#define NFS4_enc_exchange_id_sz \ + (compound_encode_hdr_maxsz + \ + encode_exchange_id_maxsz) +#define NFS4_dec_exchange_id_sz \ + (compound_decode_hdr_maxsz + \ + decode_exchange_id_maxsz) +#define NFS4_enc_create_session_sz \ + (compound_encode_hdr_maxsz + \ + encode_create_session_maxsz) +#define NFS4_dec_create_session_sz \ + (compound_decode_hdr_maxsz + \ + decode_create_session_maxsz) +#define NFS4_enc_destroy_session_sz (compound_encode_hdr_maxsz + \ + encode_destroy_session_maxsz) +#define NFS4_dec_destroy_session_sz (compound_decode_hdr_maxsz + \ + decode_destroy_session_maxsz) +#define NFS4_enc_sequence_sz \ + (compound_decode_hdr_maxsz + \ + encode_sequence_maxsz) +#define NFS4_dec_sequence_sz \ + (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz) +#define NFS4_enc_get_lease_time_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putrootfh_maxsz + \ + encode_fsinfo_maxsz) +#define NFS4_dec_get_lease_time_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putrootfh_maxsz + \ + decode_fsinfo_maxsz) +#endif /* CONFIG_NFS_V4_1 */ static const umode_t nfs_type2fmt[] = { [NF4BAD] = 0, @@ -541,6 +698,8 @@ struct compound_hdr { __be32 * nops_p; uint32_t taglen; char * tag; + uint32_t replen; /* expected reply words */ + u32 minorversion; }; /* @@ -576,22 +735,31 @@ static void encode_string(struct xdr_stream *xdr, unsigned int len, const char * xdr_encode_opaque(p, str, len); } -static void encode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr) +static void encode_compound_hdr(struct xdr_stream *xdr, + struct rpc_rqst *req, + struct compound_hdr *hdr) { __be32 *p; + struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth; + + /* initialize running count of expected bytes in reply. + * NOTE: the replied tag SHOULD be the same is the one sent, + * but this is not required as a MUST for the server to do so. */ + hdr->replen = RPC_REPHDRSIZE + auth->au_rslack + 3 + hdr->taglen; dprintk("encode_compound: tag=%.*s\n", (int)hdr->taglen, hdr->tag); BUG_ON(hdr->taglen > NFS4_MAXTAGLEN); RESERVE_SPACE(12+(XDR_QUADLEN(hdr->taglen)<<2)); WRITE32(hdr->taglen); WRITEMEM(hdr->tag, hdr->taglen); - WRITE32(NFS4_MINOR_VERSION); + WRITE32(hdr->minorversion); hdr->nops_p = p; WRITE32(hdr->nops); } static void encode_nops(struct compound_hdr *hdr) { + BUG_ON(hdr->nops > NFS4_MAX_OPS); *hdr->nops_p = htonl(hdr->nops); } @@ -736,6 +904,7 @@ static void encode_access(struct xdr_stream *xdr, u32 access, struct compound_hd WRITE32(OP_ACCESS); WRITE32(access); hdr->nops++; + hdr->replen += decode_access_maxsz; } static void encode_close(struct xdr_stream *xdr, const struct nfs_closeargs *arg, struct compound_hdr *hdr) @@ -747,6 +916,7 @@ static void encode_close(struct xdr_stream *xdr, const struct nfs_closeargs *arg WRITE32(arg->seqid->sequence->counter); WRITEMEM(arg->stateid->data, NFS4_STATEID_SIZE); hdr->nops++; + hdr->replen += decode_close_maxsz; } static void encode_commit(struct xdr_stream *xdr, const struct nfs_writeargs *args, struct compound_hdr *hdr) @@ -758,6 +928,7 @@ static void encode_commit(struct xdr_stream *xdr, const struct nfs_writeargs *ar WRITE64(args->offset); WRITE32(args->count); hdr->nops++; + hdr->replen += decode_commit_maxsz; } static void encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg *create, struct compound_hdr *hdr) @@ -789,6 +960,7 @@ static void encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg * WRITE32(create->name->len); WRITEMEM(create->name->name, create->name->len); hdr->nops++; + hdr->replen += decode_create_maxsz; encode_attrs(xdr, create->attrs, create->server); } @@ -802,6 +974,7 @@ static void encode_getattr_one(struct xdr_stream *xdr, uint32_t bitmap, struct c WRITE32(1); WRITE32(bitmap); hdr->nops++; + hdr->replen += decode_getattr_maxsz; } static void encode_getattr_two(struct xdr_stream *xdr, uint32_t bm0, uint32_t bm1, struct compound_hdr *hdr) @@ -814,6 +987,7 @@ static void encode_getattr_two(struct xdr_stream *xdr, uint32_t bm0, uint32_t bm WRITE32(bm0); WRITE32(bm1); hdr->nops++; + hdr->replen += decode_getattr_maxsz; } static void encode_getfattr(struct xdr_stream *xdr, const u32* bitmask, struct compound_hdr *hdr) @@ -841,6 +1015,7 @@ static void encode_getfh(struct xdr_stream *xdr, struct compound_hdr *hdr) RESERVE_SPACE(4); WRITE32(OP_GETFH); hdr->nops++; + hdr->replen += decode_getfh_maxsz; } static void encode_link(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr) @@ -852,6 +1027,7 @@ static void encode_link(struct xdr_stream *xdr, const struct qstr *name, struct WRITE32(name->len); WRITEMEM(name->name, name->len); hdr->nops++; + hdr->replen += decode_link_maxsz; } static inline int nfs4_lock_type(struct file_lock *fl, int block) @@ -899,6 +1075,7 @@ static void encode_lock(struct xdr_stream *xdr, const struct nfs_lock_args *args WRITE32(args->lock_seqid->sequence->counter); } hdr->nops++; + hdr->replen += decode_lock_maxsz; } static void encode_lockt(struct xdr_stream *xdr, const struct nfs_lockt_args *args, struct compound_hdr *hdr) @@ -915,6 +1092,7 @@ static void encode_lockt(struct xdr_stream *xdr, const struct nfs_lockt_args *ar WRITEMEM("lock id:", 8); WRITE64(args->lock_owner.id); hdr->nops++; + hdr->replen += decode_lockt_maxsz; } static void encode_locku(struct xdr_stream *xdr, const struct nfs_locku_args *args, struct compound_hdr *hdr) @@ -929,6 +1107,7 @@ static void encode_locku(struct xdr_stream *xdr, const struct nfs_locku_args *ar WRITE64(args->fl->fl_start); WRITE64(nfs4_lock_length(args->fl)); hdr->nops++; + hdr->replen += decode_locku_maxsz; } static void encode_lookup(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr) @@ -941,6 +1120,7 @@ static void encode_lookup(struct xdr_stream *xdr, const struct qstr *name, struc WRITE32(len); WRITEMEM(name->name, len); hdr->nops++; + hdr->replen += decode_lookup_maxsz; } static void encode_share_access(struct xdr_stream *xdr, fmode_t fmode) @@ -1080,6 +1260,7 @@ static void encode_open(struct xdr_stream *xdr, const struct nfs_openargs *arg, BUG(); } hdr->nops++; + hdr->replen += decode_open_maxsz; } static void encode_open_confirm(struct xdr_stream *xdr, const struct nfs_open_confirmargs *arg, struct compound_hdr *hdr) @@ -1091,6 +1272,7 @@ static void encode_open_confirm(struct xdr_stream *xdr, const struct nfs_open_co WRITEMEM(arg->stateid->data, NFS4_STATEID_SIZE); WRITE32(arg->seqid->sequence->counter); hdr->nops++; + hdr->replen += decode_open_confirm_maxsz; } static void encode_open_downgrade(struct xdr_stream *xdr, const struct nfs_closeargs *arg, struct compound_hdr *hdr) @@ -1103,6 +1285,7 @@ static void encode_open_downgrade(struct xdr_stream *xdr, const struct nfs_close WRITE32(arg->seqid->sequence->counter); encode_share_access(xdr, arg->fmode); hdr->nops++; + hdr->replen += decode_open_downgrade_maxsz; } static void @@ -1116,6 +1299,7 @@ encode_putfh(struct xdr_stream *xdr, const struct nfs_fh *fh, struct compound_hd WRITE32(len); WRITEMEM(fh->data, len); hdr->nops++; + hdr->replen += decode_putfh_maxsz; } static void encode_putrootfh(struct xdr_stream *xdr, struct compound_hdr *hdr) @@ -1125,6 +1309,7 @@ static void encode_putrootfh(struct xdr_stream *xdr, struct compound_hdr *hdr) RESERVE_SPACE(4); WRITE32(OP_PUTROOTFH); hdr->nops++; + hdr->replen += decode_putrootfh_maxsz; } static void encode_stateid(struct xdr_stream *xdr, const struct nfs_open_context *ctx) @@ -1153,6 +1338,7 @@ static void encode_read(struct xdr_stream *xdr, const struct nfs_readargs *args, WRITE64(args->offset); WRITE32(args->count); hdr->nops++; + hdr->replen += decode_read_maxsz; } static void encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg *readdir, struct rpc_rqst *req, struct compound_hdr *hdr) @@ -1178,6 +1364,7 @@ static void encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg WRITE32(attrs[0] & readdir->bitmask[0]); WRITE32(attrs[1] & readdir->bitmask[1]); hdr->nops++; + hdr->replen += decode_readdir_maxsz; dprintk("%s: cookie = %Lu, verifier = %08x:%08x, bitmap = %08x:%08x\n", __func__, (unsigned long long)readdir->cookie, @@ -1194,6 +1381,7 @@ static void encode_readlink(struct xdr_stream *xdr, const struct nfs4_readlink * RESERVE_SPACE(4); WRITE32(OP_READLINK); hdr->nops++; + hdr->replen += decode_readlink_maxsz; } static void encode_remove(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr) @@ -1205,6 +1393,7 @@ static void encode_remove(struct xdr_stream *xdr, const struct qstr *name, struc WRITE32(name->len); WRITEMEM(name->name, name->len); hdr->nops++; + hdr->replen += decode_remove_maxsz; } static void encode_rename(struct xdr_stream *xdr, const struct qstr *oldname, const struct qstr *newname, struct compound_hdr *hdr) @@ -1220,6 +1409,7 @@ static void encode_rename(struct xdr_stream *xdr, const struct qstr *oldname, co WRITE32(newname->len); WRITEMEM(newname->name, newname->len); hdr->nops++; + hdr->replen += decode_rename_maxsz; } static void encode_renew(struct xdr_stream *xdr, const struct nfs_client *client_stateid, struct compound_hdr *hdr) @@ -1230,6 +1420,7 @@ static void encode_renew(struct xdr_stream *xdr, const struct nfs_client *client WRITE32(OP_RENEW); WRITE64(client_stateid->cl_clientid); hdr->nops++; + hdr->replen += decode_renew_maxsz; } static void @@ -1240,6 +1431,7 @@ encode_restorefh(struct xdr_stream *xdr, struct compound_hdr *hdr) RESERVE_SPACE(4); WRITE32(OP_RESTOREFH); hdr->nops++; + hdr->replen += decode_restorefh_maxsz; } static int @@ -1259,6 +1451,7 @@ encode_setacl(struct xdr_stream *xdr, struct nfs_setaclargs *arg, struct compoun WRITE32(arg->acl_len); xdr_write_pages(xdr, arg->acl_pages, arg->acl_pgbase, arg->acl_len); hdr->nops++; + hdr->replen += decode_setacl_maxsz; return 0; } @@ -1270,6 +1463,7 @@ encode_savefh(struct xdr_stream *xdr, struct compound_hdr *hdr) RESERVE_SPACE(4); WRITE32(OP_SAVEFH); hdr->nops++; + hdr->replen += decode_savefh_maxsz; } static void encode_setattr(struct xdr_stream *xdr, const struct nfs_setattrargs *arg, const struct nfs_server *server, struct compound_hdr *hdr) @@ -1280,6 +1474,7 @@ static void encode_setattr(struct xdr_stream *xdr, const struct nfs_setattrargs WRITE32(OP_SETATTR); WRITEMEM(arg->stateid.data, NFS4_STATEID_SIZE); hdr->nops++; + hdr->replen += decode_setattr_maxsz; encode_attrs(xdr, arg->iap, server); } @@ -1299,6 +1494,7 @@ static void encode_setclientid(struct xdr_stream *xdr, const struct nfs4_setclie RESERVE_SPACE(4); WRITE32(setclientid->sc_cb_ident); hdr->nops++; + hdr->replen += decode_setclientid_maxsz; } static void encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_client *client_state, struct compound_hdr *hdr) @@ -1310,6 +1506,7 @@ static void encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_ WRITE64(client_state->cl_clientid); WRITEMEM(client_state->cl_confirm.data, NFS4_VERIFIER_SIZE); hdr->nops++; + hdr->replen += decode_setclientid_confirm_maxsz; } static void encode_write(struct xdr_stream *xdr, const struct nfs_writeargs *args, struct compound_hdr *hdr) @@ -1328,6 +1525,7 @@ static void encode_write(struct xdr_stream *xdr, const struct nfs_writeargs *arg xdr_write_pages(xdr, args->pages, args->pgbase, args->count); hdr->nops++; + hdr->replen += decode_write_maxsz; } static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *stateid, struct compound_hdr *hdr) @@ -1339,11 +1537,163 @@ static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *state WRITE32(OP_DELEGRETURN); WRITEMEM(stateid->data, NFS4_STATEID_SIZE); hdr->nops++; + hdr->replen += decode_delegreturn_maxsz; +} + +#if defined(CONFIG_NFS_V4_1) +/* NFSv4.1 operations */ +static void encode_exchange_id(struct xdr_stream *xdr, + struct nfs41_exchange_id_args *args, + struct compound_hdr *hdr) +{ + uint32_t *p; + + RESERVE_SPACE(4 + sizeof(args->verifier->data)); + WRITE32(OP_EXCHANGE_ID); + WRITEMEM(args->verifier->data, sizeof(args->verifier->data)); + + encode_string(xdr, args->id_len, args->id); + + RESERVE_SPACE(12); + WRITE32(args->flags); + WRITE32(0); /* zero length state_protect4_a */ + WRITE32(0); /* zero length implementation id array */ + hdr->nops++; + hdr->replen += decode_exchange_id_maxsz; +} + +static void encode_create_session(struct xdr_stream *xdr, + struct nfs41_create_session_args *args, + struct compound_hdr *hdr) +{ + uint32_t *p; + char machine_name[NFS4_MAX_MACHINE_NAME_LEN]; + uint32_t len; + struct nfs_client *clp = args->client; + + RESERVE_SPACE(4); + WRITE32(OP_CREATE_SESSION); + + RESERVE_SPACE(8); + WRITE64(clp->cl_ex_clid); + + RESERVE_SPACE(8); + WRITE32(clp->cl_seqid); /*Sequence id */ + WRITE32(args->flags); /*flags */ + + RESERVE_SPACE(2*28); /* 2 channel_attrs */ + /* Fore Channel */ + WRITE32(args->fc_attrs.headerpadsz); /* header padding size */ + WRITE32(args->fc_attrs.max_rqst_sz); /* max req size */ + WRITE32(args->fc_attrs.max_resp_sz); /* max resp size */ + WRITE32(args->fc_attrs.max_resp_sz_cached); /* Max resp sz cached */ + WRITE32(args->fc_attrs.max_ops); /* max operations */ + WRITE32(args->fc_attrs.max_reqs); /* max requests */ + WRITE32(0); /* rdmachannel_attrs */ + + /* Back Channel */ + WRITE32(args->fc_attrs.headerpadsz); /* header padding size */ + WRITE32(args->bc_attrs.max_rqst_sz); /* max req size */ + WRITE32(args->bc_attrs.max_resp_sz); /* max resp size */ + WRITE32(args->bc_attrs.max_resp_sz_cached); /* Max resp sz cached */ + WRITE32(args->bc_attrs.max_ops); /* max operations */ + WRITE32(args->bc_attrs.max_reqs); /* max requests */ + WRITE32(0); /* rdmachannel_attrs */ + + RESERVE_SPACE(4); + WRITE32(args->cb_program); /* cb_program */ + + RESERVE_SPACE(4); /* # of security flavors */ + WRITE32(1); + + RESERVE_SPACE(4); + WRITE32(RPC_AUTH_UNIX); /* auth_sys */ + + /* authsys_parms rfc1831 */ + RESERVE_SPACE(4); + WRITE32((u32)clp->cl_boot_time.tv_nsec); /* stamp */ + len = scnprintf(machine_name, sizeof(machine_name), "%s", + clp->cl_ipaddr); + RESERVE_SPACE(16 + len); + WRITE32(len); + WRITEMEM(machine_name, len); + WRITE32(0); /* UID */ + WRITE32(0); /* GID */ + WRITE32(0); /* No more gids */ + hdr->nops++; + hdr->replen += decode_create_session_maxsz; } + +static void encode_destroy_session(struct xdr_stream *xdr, + struct nfs4_session *session, + struct compound_hdr *hdr) +{ + uint32_t *p; + RESERVE_SPACE(4 + NFS4_MAX_SESSIONID_LEN); + WRITE32(OP_DESTROY_SESSION); + WRITEMEM(session->sess_id.data, NFS4_MAX_SESSIONID_LEN); + hdr->nops++; + hdr->replen += decode_destroy_session_maxsz; +} +#endif /* CONFIG_NFS_V4_1 */ + +static void encode_sequence(struct xdr_stream *xdr, + const struct nfs4_sequence_args *args, + struct compound_hdr *hdr) +{ +#if defined(CONFIG_NFS_V4_1) + struct nfs4_session *session = args->sa_session; + struct nfs4_slot_table *tp; + struct nfs4_slot *slot; + __be32 *p; + + if (!session) + return; + + tp = &session->fc_slot_table; + + WARN_ON(args->sa_slotid == NFS4_MAX_SLOT_TABLE); + slot = tp->slots + args->sa_slotid; + + RESERVE_SPACE(4); + WRITE32(OP_SEQUENCE); + + /* + * Sessionid + seqid + slotid + max slotid + cache_this + */ + dprintk("%s: sessionid=%u:%u:%u:%u seqid=%d slotid=%d " + "max_slotid=%d cache_this=%d\n", + __func__, + ((u32 *)session->sess_id.data)[0], + ((u32 *)session->sess_id.data)[1], + ((u32 *)session->sess_id.data)[2], + ((u32 *)session->sess_id.data)[3], + slot->seq_nr, args->sa_slotid, + tp->highest_used_slotid, args->sa_cache_this); + RESERVE_SPACE(NFS4_MAX_SESSIONID_LEN + 16); + WRITEMEM(session->sess_id.data, NFS4_MAX_SESSIONID_LEN); + WRITE32(slot->seq_nr); + WRITE32(args->sa_slotid); + WRITE32(tp->highest_used_slotid); + WRITE32(args->sa_cache_this); + hdr->nops++; + hdr->replen += decode_sequence_maxsz; +#endif /* CONFIG_NFS_V4_1 */ +} + /* * END OF "GENERIC" ENCODE ROUTINES. */ +static u32 nfs4_xdr_minorversion(const struct nfs4_sequence_args *args) +{ +#if defined(CONFIG_NFS_V4_1) + if (args->sa_session) + return args->sa_session->clp->cl_minorversion; +#endif /* CONFIG_NFS_V4_1 */ + return 0; +} + /* * Encode an ACCESS request */ @@ -1351,11 +1701,12 @@ static int nfs4_xdr_enc_access(struct rpc_rqst *req, __be32 *p, const struct nfs { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_access(&xdr, args->access, &hdr); encode_getfattr(&xdr, args->bitmask, &hdr); @@ -1370,11 +1721,12 @@ static int nfs4_xdr_enc_lookup(struct rpc_rqst *req, __be32 *p, const struct nfs { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->dir_fh, &hdr); encode_lookup(&xdr, args->name, &hdr); encode_getfh(&xdr, &hdr); @@ -1390,11 +1742,12 @@ static int nfs4_xdr_enc_lookup_root(struct rpc_rqst *req, __be32 *p, const struc { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putrootfh(&xdr, &hdr); encode_getfh(&xdr, &hdr); encode_getfattr(&xdr, args->bitmask, &hdr); @@ -1409,11 +1762,12 @@ static int nfs4_xdr_enc_remove(struct rpc_rqst *req, __be32 *p, const struct nfs { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_remove(&xdr, &args->name, &hdr); encode_getfattr(&xdr, args->bitmask, &hdr); @@ -1428,11 +1782,12 @@ static int nfs4_xdr_enc_rename(struct rpc_rqst *req, __be32 *p, const struct nfs { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->old_dir, &hdr); encode_savefh(&xdr, &hdr); encode_putfh(&xdr, args->new_dir, &hdr); @@ -1451,11 +1806,12 @@ static int nfs4_xdr_enc_link(struct rpc_rqst *req, __be32 *p, const struct nfs4_ { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_savefh(&xdr, &hdr); encode_putfh(&xdr, args->dir_fh, &hdr); @@ -1474,11 +1830,12 @@ static int nfs4_xdr_enc_create(struct rpc_rqst *req, __be32 *p, const struct nfs { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->dir_fh, &hdr); encode_savefh(&xdr, &hdr); encode_create(&xdr, args, &hdr); @@ -1505,11 +1862,12 @@ static int nfs4_xdr_enc_getattr(struct rpc_rqst *req, __be32 *p, const struct nf { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_getfattr(&xdr, args->bitmask, &hdr); encode_nops(&hdr); @@ -1523,11 +1881,12 @@ static int nfs4_xdr_enc_close(struct rpc_rqst *req, __be32 *p, struct nfs_closea { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_close(&xdr, args, &hdr); encode_getfattr(&xdr, args->bitmask, &hdr); @@ -1542,11 +1901,12 @@ static int nfs4_xdr_enc_open(struct rpc_rqst *req, __be32 *p, struct nfs_openarg { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_savefh(&xdr, &hdr); encode_open(&xdr, args, &hdr); @@ -1569,7 +1929,7 @@ static int nfs4_xdr_enc_open_confirm(struct rpc_rqst *req, __be32 *p, struct nfs }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_open_confirm(&xdr, args, &hdr); encode_nops(&hdr); @@ -1583,11 +1943,12 @@ static int nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, __be32 *p, struct nfs_ { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_open(&xdr, args, &hdr); encode_getfattr(&xdr, args->bitmask, &hdr); @@ -1602,11 +1963,12 @@ static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, __be32 *p, struct n { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_open_downgrade(&xdr, args, &hdr); encode_getfattr(&xdr, args->bitmask, &hdr); @@ -1621,11 +1983,12 @@ static int nfs4_xdr_enc_lock(struct rpc_rqst *req, __be32 *p, struct nfs_lock_ar { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_lock(&xdr, args, &hdr); encode_nops(&hdr); @@ -1639,11 +2002,12 @@ static int nfs4_xdr_enc_lockt(struct rpc_rqst *req, __be32 *p, struct nfs_lockt_ { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_lockt(&xdr, args, &hdr); encode_nops(&hdr); @@ -1657,11 +2021,12 @@ static int nfs4_xdr_enc_locku(struct rpc_rqst *req, __be32 *p, struct nfs_locku_ { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_locku(&xdr, args, &hdr); encode_nops(&hdr); @@ -1675,22 +2040,16 @@ static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, __be32 *p, const struct n { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; - struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth; - unsigned int replen; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_readlink(&xdr, args, req, &hdr); - /* set up reply kvec - * toplevel_status + taglen + rescount + OP_PUTFH + status - * + OP_READLINK + status + string length = 8 - */ - replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_readlink_sz) << 2; - xdr_inline_pages(&req->rq_rcv_buf, replen, args->pages, + xdr_inline_pages(&req->rq_rcv_buf, hdr.replen << 2, args->pages, args->pgbase, args->pglen); encode_nops(&hdr); return 0; @@ -1703,25 +2062,19 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; - struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth; - int replen; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_readdir(&xdr, args, req, &hdr); - /* set up reply kvec - * toplevel_status + taglen + rescount + OP_PUTFH + status - * + OP_READDIR + status + verifer(2) = 9 - */ - replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_readdir_sz) << 2; - xdr_inline_pages(&req->rq_rcv_buf, replen, args->pages, + xdr_inline_pages(&req->rq_rcv_buf, hdr.replen << 2, args->pages, args->pgbase, args->count); dprintk("%s: inlined page args = (%u, %p, %u, %u)\n", - __func__, replen, args->pages, + __func__, hdr.replen << 2, args->pages, args->pgbase, args->count); encode_nops(&hdr); return 0; @@ -1732,24 +2085,18 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf */ static int nfs4_xdr_enc_read(struct rpc_rqst *req, __be32 *p, struct nfs_readargs *args) { - struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth; struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; - int replen; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_read(&xdr, args, &hdr); - /* set up reply kvec - * toplevel status + taglen=0 + rescount + OP_PUTFH + status - * + OP_READ + status + eof + datalen = 9 - */ - replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_read_sz) << 2; - xdr_inline_pages(&req->rq_rcv_buf, replen, + xdr_inline_pages(&req->rq_rcv_buf, hdr.replen << 2, args->pages, args->pgbase, args->count); req->rq_rcv_buf.flags |= XDRBUF_READ; encode_nops(&hdr); @@ -1763,11 +2110,12 @@ static int nfs4_xdr_enc_setattr(struct rpc_rqst *req, __be32 *p, struct nfs_seta { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_setattr(&xdr, args, args->server, &hdr); encode_getfattr(&xdr, args->bitmask, &hdr); @@ -1783,20 +2131,19 @@ nfs4_xdr_enc_getacl(struct rpc_rqst *req, __be32 *p, struct nfs_getaclargs *args) { struct xdr_stream xdr; - struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; - int replen; + uint32_t replen; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); + replen = hdr.replen + nfs4_fattr_bitmap_maxsz + 1; encode_getattr_two(&xdr, FATTR4_WORD0_ACL, 0, &hdr); - /* set up reply buffer: */ - replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_getacl_sz) << 2; - xdr_inline_pages(&req->rq_rcv_buf, replen, + xdr_inline_pages(&req->rq_rcv_buf, replen << 2, args->acl_pages, args->acl_pgbase, args->acl_len); encode_nops(&hdr); return 0; @@ -1809,11 +2156,12 @@ static int nfs4_xdr_enc_write(struct rpc_rqst *req, __be32 *p, struct nfs_writea { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_write(&xdr, args, &hdr); req->rq_snd_buf.flags |= XDRBUF_WRITE; @@ -1829,11 +2177,12 @@ static int nfs4_xdr_enc_commit(struct rpc_rqst *req, __be32 *p, struct nfs_write { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_commit(&xdr, args, &hdr); encode_getfattr(&xdr, args->bitmask, &hdr); @@ -1848,11 +2197,12 @@ static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs4_fsin { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_fsinfo(&xdr, args->bitmask, &hdr); encode_nops(&hdr); @@ -1866,11 +2216,12 @@ static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, __be32 *p, const struct n { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_getattr_one(&xdr, args->bitmask[0] & nfs4_pathconf_bitmap[0], &hdr); @@ -1885,11 +2236,12 @@ static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, __be32 *p, const struct nfs { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); encode_getattr_two(&xdr, args->bitmask[0] & nfs4_statfs_bitmap[0], args->bitmask[1] & nfs4_statfs_bitmap[1], &hdr); @@ -1900,16 +2252,18 @@ static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, __be32 *p, const struct nfs /* * GETATTR_BITMAP request */ -static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p, const struct nfs_fh *fhandle) +static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p, + struct nfs4_server_caps_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); - encode_putfh(&xdr, fhandle, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); + encode_putfh(&xdr, args->fhandle, &hdr); encode_getattr_one(&xdr, FATTR4_WORD0_SUPPORTED_ATTRS| FATTR4_WORD0_LINK_SUPPORT| FATTR4_WORD0_SYMLINK_SUPPORT| @@ -1929,7 +2283,7 @@ static int nfs4_xdr_enc_renew(struct rpc_rqst *req, __be32 *p, struct nfs_client }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); encode_renew(&xdr, clp, &hdr); encode_nops(&hdr); return 0; @@ -1946,7 +2300,7 @@ static int nfs4_xdr_enc_setclientid(struct rpc_rqst *req, __be32 *p, struct nfs4 }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); encode_setclientid(&xdr, sc, &hdr); encode_nops(&hdr); return 0; @@ -1964,7 +2318,7 @@ static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, __be32 *p, str const u32 lease_bitmap[2] = { FATTR4_WORD0_LEASE_TIME, 0 }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); encode_setclientid_confirm(&xdr, clp, &hdr); encode_putrootfh(&xdr, &hdr); encode_fsinfo(&xdr, lease_bitmap, &hdr); @@ -1979,11 +2333,12 @@ static int nfs4_xdr_enc_delegreturn(struct rpc_rqst *req, __be32 *p, const struc { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fhandle, &hdr); encode_delegreturn(&xdr, args->stateid, &hdr); encode_getfattr(&xdr, args->bitmask, &hdr); @@ -1998,28 +2353,119 @@ static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; - struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth; - int replen; + uint32_t replen; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->dir_fh, &hdr); encode_lookup(&xdr, args->name, &hdr); + replen = hdr.replen; /* get the attribute into args->page */ encode_fs_locations(&xdr, args->bitmask, &hdr); - /* set up reply - * toplevel_status + OP_PUTFH + status - * + OP_LOOKUP + status + OP_GETATTR + status = 7 - */ - replen = (RPC_REPHDRSIZE + auth->au_rslack + 7) << 2; - xdr_inline_pages(&req->rq_rcv_buf, replen, &args->page, + xdr_inline_pages(&req->rq_rcv_buf, replen << 2, &args->page, 0, PAGE_SIZE); encode_nops(&hdr); return 0; } +#if defined(CONFIG_NFS_V4_1) +/* + * EXCHANGE_ID request + */ +static int nfs4_xdr_enc_exchange_id(struct rpc_rqst *req, uint32_t *p, + struct nfs41_exchange_id_args *args) +{ + struct xdr_stream xdr; + struct compound_hdr hdr = { + .minorversion = args->client->cl_minorversion, + }; + + xdr_init_encode(&xdr, &req->rq_snd_buf, p); + encode_compound_hdr(&xdr, req, &hdr); + encode_exchange_id(&xdr, args, &hdr); + encode_nops(&hdr); + return 0; +} + +/* + * a CREATE_SESSION request + */ +static int nfs4_xdr_enc_create_session(struct rpc_rqst *req, uint32_t *p, + struct nfs41_create_session_args *args) +{ + struct xdr_stream xdr; + struct compound_hdr hdr = { + .minorversion = args->client->cl_minorversion, + }; + + xdr_init_encode(&xdr, &req->rq_snd_buf, p); + encode_compound_hdr(&xdr, req, &hdr); + encode_create_session(&xdr, args, &hdr); + encode_nops(&hdr); + return 0; +} + +/* + * a DESTROY_SESSION request + */ +static int nfs4_xdr_enc_destroy_session(struct rpc_rqst *req, uint32_t *p, + struct nfs4_session *session) +{ + struct xdr_stream xdr; + struct compound_hdr hdr = { + .minorversion = session->clp->cl_minorversion, + }; + + xdr_init_encode(&xdr, &req->rq_snd_buf, p); + encode_compound_hdr(&xdr, req, &hdr); + encode_destroy_session(&xdr, session, &hdr); + encode_nops(&hdr); + return 0; +} + +/* + * a SEQUENCE request + */ +static int nfs4_xdr_enc_sequence(struct rpc_rqst *req, uint32_t *p, + struct nfs4_sequence_args *args) +{ + struct xdr_stream xdr; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(args), + }; + + xdr_init_encode(&xdr, &req->rq_snd_buf, p); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, args, &hdr); + encode_nops(&hdr); + return 0; +} + +/* + * a GET_LEASE_TIME request + */ +static int nfs4_xdr_enc_get_lease_time(struct rpc_rqst *req, uint32_t *p, + struct nfs4_get_lease_time_args *args) +{ + struct xdr_stream xdr; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->la_seq_args), + }; + const u32 lease_bitmap[2] = { FATTR4_WORD0_LEASE_TIME, 0 }; + + xdr_init_encode(&xdr, &req->rq_snd_buf, p); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->la_seq_args, &hdr); + encode_putrootfh(&xdr, &hdr); + encode_fsinfo(&xdr, lease_bitmap, &hdr); + encode_nops(&hdr); + return 0; +} +#endif /* CONFIG_NFS_V4_1 */ + /* * START OF "GENERIC" DECODE ROUTINES. * These may look a little ugly since they are imported from a "generic" @@ -3657,7 +4103,7 @@ decode_savefh(struct xdr_stream *xdr) return decode_op_hdr(xdr, OP_SAVEFH); } -static int decode_setattr(struct xdr_stream *xdr, struct nfs_setattrres *res) +static int decode_setattr(struct xdr_stream *xdr) { __be32 *p; uint32_t bmlen; @@ -3735,6 +4181,175 @@ static int decode_delegreturn(struct xdr_stream *xdr) return decode_op_hdr(xdr, OP_DELEGRETURN); } +#if defined(CONFIG_NFS_V4_1) +static int decode_exchange_id(struct xdr_stream *xdr, + struct nfs41_exchange_id_res *res) +{ + uint32_t *p; + int status, dummy; + struct nfs_client *clp = res->client; + + status = decode_op_hdr(xdr, OP_EXCHANGE_ID); + if (status) + return status; + + READ_BUF(8); + READ64(clp->cl_ex_clid); + READ_BUF(12); + READ32(clp->cl_seqid); + READ32(clp->cl_exchange_flags); + + /* We ask for SP4_NONE */ + READ32(dummy); + if (dummy != SP4_NONE) + return -EIO; + + /* minor_id */ + READ_BUF(8); + READ64(res->server_owner.minor_id); + + /* Major id */ + READ_BUF(4); + READ32(res->server_owner.major_id_sz); + READ_BUF(res->server_owner.major_id_sz); + COPYMEM(res->server_owner.major_id, res->server_owner.major_id_sz); + + /* server_scope */ + READ_BUF(4); + READ32(res->server_scope.server_scope_sz); + READ_BUF(res->server_scope.server_scope_sz); + COPYMEM(res->server_scope.server_scope, + res->server_scope.server_scope_sz); + /* Throw away Implementation id array */ + READ_BUF(4); + READ32(dummy); + p += XDR_QUADLEN(dummy); + + return 0; +} + +static int decode_create_session(struct xdr_stream *xdr, + struct nfs41_create_session_res *res) +{ + uint32_t *p; + int status; + u32 nr_attrs; + struct nfs_client *clp = res->client; + struct nfs4_session *session = clp->cl_session; + + status = decode_op_hdr(xdr, OP_CREATE_SESSION); + + if (status) + return status; + + /* sessionid */ + READ_BUF(NFS4_MAX_SESSIONID_LEN); + COPYMEM(&session->sess_id, NFS4_MAX_SESSIONID_LEN); + + /* seqid, flags */ + READ_BUF(8); + READ32(clp->cl_seqid); + READ32(session->flags); + + /* Channel attributes */ + /* fore channel */ + READ_BUF(24); + READ32(session->fc_attrs.headerpadsz); + READ32(session->fc_attrs.max_rqst_sz); + READ32(session->fc_attrs.max_resp_sz); + READ32(session->fc_attrs.max_resp_sz_cached); + READ32(session->fc_attrs.max_ops); + READ32(session->fc_attrs.max_reqs); + READ_BUF(4); + READ32(nr_attrs); + if (nr_attrs == 1) { + READ_BUF(4); + p++; /* skip rdma_attrs */ + } + + /* back channel */ + READ_BUF(24); + READ32(session->bc_attrs.headerpadsz); + READ32(session->bc_attrs.max_rqst_sz); + READ32(session->bc_attrs.max_resp_sz); + READ32(session->bc_attrs.max_resp_sz_cached); + READ32(session->bc_attrs.max_ops); + READ32(session->bc_attrs.max_reqs); + READ_BUF(4); + READ32(nr_attrs); + if (nr_attrs == 1) { + READ_BUF(4); + p++; /* skip rdma_attrs */ + } + + return 0; +} + +static int decode_destroy_session(struct xdr_stream *xdr, void *dummy) +{ + return decode_op_hdr(xdr, OP_DESTROY_SESSION); +} +#endif /* CONFIG_NFS_V4_1 */ + +static int decode_sequence(struct xdr_stream *xdr, + struct nfs4_sequence_res *res, + struct rpc_rqst *rqstp) +{ +#if defined(CONFIG_NFS_V4_1) + struct nfs4_slot *slot; + struct nfs4_sessionid id; + u32 dummy; + int status; + __be32 *p; + + if (!res->sr_session) + return 0; + + status = decode_op_hdr(xdr, OP_SEQUENCE); + if (status) + goto out_err; + + /* + * If the server returns different values for sessionID, slotID or + * sequence number, the server is looney tunes. + */ + status = -ESERVERFAULT; + + slot = &res->sr_session->fc_slot_table.slots[res->sr_slotid]; + READ_BUF(NFS4_MAX_SESSIONID_LEN + 20); + COPYMEM(id.data, NFS4_MAX_SESSIONID_LEN); + if (memcmp(id.data, res->sr_session->sess_id.data, + NFS4_MAX_SESSIONID_LEN)) { + dprintk("%s Invalid session id\n", __func__); + goto out_err; + } + /* seqid */ + READ32(dummy); + if (dummy != slot->seq_nr) { + dprintk("%s Invalid sequence number\n", __func__); + goto out_err; + } + /* slot id */ + READ32(dummy); + if (dummy != res->sr_slotid) { + dprintk("%s Invalid slot id\n", __func__); + goto out_err; + } + /* highest slot id - currently not processed */ + READ32(dummy); + /* target highest slot id - currently not processed */ + READ32(dummy); + /* result flags - currently not processed */ + READ32(dummy); + status = 0; +out_err: + res->sr_status = status; + return status; +#else /* CONFIG_NFS_V4_1 */ + return 0; +#endif /* CONFIG_NFS_V4_1 */ +} + /* * END OF "GENERIC" DECODE ROUTINES. */ @@ -3752,6 +4367,9 @@ static int nfs4_xdr_dec_open_downgrade(struct rpc_rqst *rqstp, __be32 *p, struct status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; @@ -3773,7 +4391,11 @@ static int nfs4_xdr_dec_access(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_ac int status; xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); - if ((status = decode_compound_hdr(&xdr, &hdr)) != 0) + status = decode_compound_hdr(&xdr, &hdr); + if (status) + goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) goto out; status = decode_putfh(&xdr); if (status != 0) @@ -3796,7 +4418,11 @@ static int nfs4_xdr_dec_lookup(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_lo int status; xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); - if ((status = decode_compound_hdr(&xdr, &hdr)) != 0) + status = decode_compound_hdr(&xdr, &hdr); + if (status) + goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) goto out; if ((status = decode_putfh(&xdr)) != 0) goto out; @@ -3819,7 +4445,11 @@ static int nfs4_xdr_dec_lookup_root(struct rpc_rqst *rqstp, __be32 *p, struct nf int status; xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); - if ((status = decode_compound_hdr(&xdr, &hdr)) != 0) + status = decode_compound_hdr(&xdr, &hdr); + if (status) + goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) goto out; if ((status = decode_putrootfh(&xdr)) != 0) goto out; @@ -3839,7 +4469,11 @@ static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, __be32 *p, struct nfs_rem int status; xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); - if ((status = decode_compound_hdr(&xdr, &hdr)) != 0) + status = decode_compound_hdr(&xdr, &hdr); + if (status) + goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) goto out; if ((status = decode_putfh(&xdr)) != 0) goto out; @@ -3860,7 +4494,11 @@ static int nfs4_xdr_dec_rename(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_re int status; xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); - if ((status = decode_compound_hdr(&xdr, &hdr)) != 0) + status = decode_compound_hdr(&xdr, &hdr); + if (status) + goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) goto out; if ((status = decode_putfh(&xdr)) != 0) goto out; @@ -3890,7 +4528,11 @@ static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_link int status; xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); - if ((status = decode_compound_hdr(&xdr, &hdr)) != 0) + status = decode_compound_hdr(&xdr, &hdr); + if (status) + goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) goto out; if ((status = decode_putfh(&xdr)) != 0) goto out; @@ -3923,7 +4565,11 @@ static int nfs4_xdr_dec_create(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_cr int status; xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); - if ((status = decode_compound_hdr(&xdr, &hdr)) != 0) + status = decode_compound_hdr(&xdr, &hdr); + if (status) + goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) goto out; if ((status = decode_putfh(&xdr)) != 0) goto out; @@ -3963,6 +4609,9 @@ static int nfs4_xdr_dec_getattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_g status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; @@ -3979,12 +4628,13 @@ nfs4_xdr_enc_setacl(struct rpc_rqst *req, __be32 *p, struct nfs_setaclargs *args { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 0, + .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; int status; xdr_init_encode(&xdr, &req->rq_snd_buf, p); - encode_compound_hdr(&xdr, &hdr); + encode_compound_hdr(&xdr, req, &hdr); + encode_sequence(&xdr, &args->seq_args, &hdr); encode_putfh(&xdr, args->fh, &hdr); status = encode_setacl(&xdr, args, &hdr); encode_nops(&hdr); @@ -3995,7 +4645,8 @@ nfs4_xdr_enc_setacl(struct rpc_rqst *req, __be32 *p, struct nfs_setaclargs *args * Decode SETACL response */ static int -nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, __be32 *p, void *res) +nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, __be32 *p, + struct nfs_setaclres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4005,10 +4656,13 @@ nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, __be32 *p, void *res) status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; - status = decode_setattr(&xdr, res); + status = decode_setattr(&xdr); out: return status; } @@ -4017,7 +4671,8 @@ out: * Decode GETACL response */ static int -nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, __be32 *p, size_t *acl_len) +nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, __be32 *p, + struct nfs_getaclres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4027,10 +4682,13 @@ nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, __be32 *p, size_t *acl_len) status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; - status = decode_getacl(&xdr, rqstp, acl_len); + status = decode_getacl(&xdr, rqstp, &res->acl_len); out: return status; @@ -4049,6 +4707,9 @@ static int nfs4_xdr_dec_close(struct rpc_rqst *rqstp, __be32 *p, struct nfs_clos status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; @@ -4079,6 +4740,9 @@ static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, __be32 *p, struct nfs_openr status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; @@ -4133,6 +4797,9 @@ static int nfs4_xdr_dec_open_noattr(struct rpc_rqst *rqstp, __be32 *p, struct nf status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; @@ -4157,10 +4824,13 @@ static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs_se status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; - status = decode_setattr(&xdr, res); + status = decode_setattr(&xdr); if (status) goto out; decode_getfattr(&xdr, res->fattr, res->server); @@ -4181,6 +4851,9 @@ static int nfs4_xdr_dec_lock(struct rpc_rqst *rqstp, __be32 *p, struct nfs_lock_ status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; @@ -4202,6 +4875,9 @@ static int nfs4_xdr_dec_lockt(struct rpc_rqst *rqstp, __be32 *p, struct nfs_lock status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; @@ -4223,6 +4899,9 @@ static int nfs4_xdr_dec_locku(struct rpc_rqst *rqstp, __be32 *p, struct nfs_lock status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; @@ -4234,7 +4913,8 @@ out: /* * Decode READLINK response */ -static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, __be32 *p, void *res) +static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, __be32 *p, + struct nfs4_readlink_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4244,6 +4924,9 @@ static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, __be32 *p, void *res) status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; @@ -4265,6 +4948,9 @@ static int nfs4_xdr_dec_readdir(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_r status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; @@ -4286,6 +4972,9 @@ static int nfs4_xdr_dec_read(struct rpc_rqst *rqstp, __be32 *p, struct nfs_readr status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; @@ -4309,6 +4998,9 @@ static int nfs4_xdr_dec_write(struct rpc_rqst *rqstp, __be32 *p, struct nfs_writ status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; @@ -4335,6 +5027,9 @@ static int nfs4_xdr_dec_commit(struct rpc_rqst *rqstp, __be32 *p, struct nfs_wri status = decode_compound_hdr(&xdr, &hdr); if (status) goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) + goto out; status = decode_putfh(&xdr); if (status) goto out; @@ -4349,7 +5044,8 @@ out: /* * FSINFO request */ -static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs_fsinfo *fsinfo) +static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p, + struct nfs4_fsinfo_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4358,16 +5054,19 @@ static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs_fsinf xdr_init_decode(&xdr, &req->rq_rcv_buf, p); status = decode_compound_hdr(&xdr, &hdr); if (!status) + status = decode_sequence(&xdr, &res->seq_res, req); + if (!status) status = decode_putfh(&xdr); if (!status) - status = decode_fsinfo(&xdr, fsinfo); + status = decode_fsinfo(&xdr, res->fsinfo); return status; } /* * PATHCONF request */ -static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p, struct nfs_pathconf *pathconf) +static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p, + struct nfs4_pathconf_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4376,16 +5075,19 @@ static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p, struct nfs_pat xdr_init_decode(&xdr, &req->rq_rcv_buf, p); status = decode_compound_hdr(&xdr, &hdr); if (!status) + status = decode_sequence(&xdr, &res->seq_res, req); + if (!status) status = decode_putfh(&xdr); if (!status) - status = decode_pathconf(&xdr, pathconf); + status = decode_pathconf(&xdr, res->pathconf); return status; } /* * STATFS request */ -static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p, struct nfs_fsstat *fsstat) +static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p, + struct nfs4_statfs_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4394,9 +5096,11 @@ static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p, struct nfs_fssta xdr_init_decode(&xdr, &req->rq_rcv_buf, p); status = decode_compound_hdr(&xdr, &hdr); if (!status) + status = decode_sequence(&xdr, &res->seq_res, req); + if (!status) status = decode_putfh(&xdr); if (!status) - status = decode_statfs(&xdr, fsstat); + status = decode_statfs(&xdr, res->fsstat); return status; } @@ -4410,7 +5114,11 @@ static int nfs4_xdr_dec_server_caps(struct rpc_rqst *req, __be32 *p, struct nfs4 int status; xdr_init_decode(&xdr, &req->rq_rcv_buf, p); - if ((status = decode_compound_hdr(&xdr, &hdr)) != 0) + status = decode_compound_hdr(&xdr, &hdr); + if (status) + goto out; + status = decode_sequence(&xdr, &res->seq_res, req); + if (status) goto out; if ((status = decode_putfh(&xdr)) != 0) goto out; @@ -4483,7 +5191,10 @@ static int nfs4_xdr_dec_delegreturn(struct rpc_rqst *rqstp, __be32 *p, struct nf xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); status = decode_compound_hdr(&xdr, &hdr); - if (status != 0) + if (status) + goto out; + status = decode_sequence(&xdr, &res->seq_res, rqstp); + if (status) goto out; status = decode_putfh(&xdr); if (status != 0) @@ -4497,7 +5208,8 @@ out: /* * FS_LOCATIONS request */ -static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs4_fs_locations *res) +static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p, + struct nfs4_fs_locations_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4505,18 +5217,113 @@ static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs xdr_init_decode(&xdr, &req->rq_rcv_buf, p); status = decode_compound_hdr(&xdr, &hdr); - if (status != 0) + if (status) + goto out; + status = decode_sequence(&xdr, &res->seq_res, req); + if (status) goto out; if ((status = decode_putfh(&xdr)) != 0) goto out; if ((status = decode_lookup(&xdr)) != 0) goto out; xdr_enter_page(&xdr, PAGE_SIZE); - status = decode_getfattr(&xdr, &res->fattr, res->server); + status = decode_getfattr(&xdr, &res->fs_locations->fattr, + res->fs_locations->server); out: return status; } +#if defined(CONFIG_NFS_V4_1) +/* + * EXCHANGE_ID request + */ +static int nfs4_xdr_dec_exchange_id(struct rpc_rqst *rqstp, uint32_t *p, + void *res) +{ + struct xdr_stream xdr; + struct compound_hdr hdr; + int status; + + xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); + status = decode_compound_hdr(&xdr, &hdr); + if (!status) + status = decode_exchange_id(&xdr, res); + return status; +} + +/* + * a CREATE_SESSION request + */ +static int nfs4_xdr_dec_create_session(struct rpc_rqst *rqstp, uint32_t *p, + struct nfs41_create_session_res *res) +{ + struct xdr_stream xdr; + struct compound_hdr hdr; + int status; + + xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); + status = decode_compound_hdr(&xdr, &hdr); + if (!status) + status = decode_create_session(&xdr, res); + return status; +} + +/* + * a DESTROY_SESSION request + */ +static int nfs4_xdr_dec_destroy_session(struct rpc_rqst *rqstp, uint32_t *p, + void *dummy) +{ + struct xdr_stream xdr; + struct compound_hdr hdr; + int status; + + xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); + status = decode_compound_hdr(&xdr, &hdr); + if (!status) + status = decode_destroy_session(&xdr, dummy); + return status; +} + +/* + * a SEQUENCE request + */ +static int nfs4_xdr_dec_sequence(struct rpc_rqst *rqstp, uint32_t *p, + struct nfs4_sequence_res *res) +{ + struct xdr_stream xdr; + struct compound_hdr hdr; + int status; + + xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); + status = decode_compound_hdr(&xdr, &hdr); + if (!status) + status = decode_sequence(&xdr, res, rqstp); + return status; +} + +/* + * a GET_LEASE_TIME request + */ +static int nfs4_xdr_dec_get_lease_time(struct rpc_rqst *rqstp, uint32_t *p, + struct nfs4_get_lease_time_res *res) +{ + struct xdr_stream xdr; + struct compound_hdr hdr; + int status; + + xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); + status = decode_compound_hdr(&xdr, &hdr); + if (!status) + status = decode_sequence(&xdr, &res->lr_seq_res, rqstp); + if (!status) + status = decode_putrootfh(&xdr); + if (!status) + status = decode_fsinfo(&xdr, res->lr_fsinfo); + return status; +} +#endif /* CONFIG_NFS_V4_1 */ + __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus) { uint32_t bitmap[2] = {0}; @@ -4686,6 +5493,13 @@ struct rpc_procinfo nfs4_procedures[] = { PROC(GETACL, enc_getacl, dec_getacl), PROC(SETACL, enc_setacl, dec_setacl), PROC(FS_LOCATIONS, enc_fs_locations, dec_fs_locations), +#if defined(CONFIG_NFS_V4_1) + PROC(EXCHANGE_ID, enc_exchange_id, dec_exchange_id), + PROC(CREATE_SESSION, enc_create_session, dec_create_session), + PROC(DESTROY_SESSION, enc_destroy_session, dec_destroy_session), + PROC(SEQUENCE, enc_sequence, dec_sequence), + PROC(GET_LEASE_TIME, enc_get_lease_time, dec_get_lease_time), +#endif /* CONFIG_NFS_V4_1 */ }; struct rpc_version nfs_version4 = { diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 4ace3c50a8eb..96c4ebfa46f4 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -22,6 +22,7 @@ #include <asm/system.h> +#include "nfs4_fs.h" #include "internal.h" #include "iostat.h" #include "fscache.h" @@ -46,6 +47,7 @@ struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount) memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->pages); p->npages = pagecount; + p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE; if (pagecount <= ARRAY_SIZE(p->page_array)) p->pagevec = p->page_array; else { @@ -357,19 +359,25 @@ static void nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data struct nfs_readres *resp = &data->res; if (resp->eof || resp->count == argp->count) - return; + goto out; /* This is a short read! */ nfs_inc_stats(data->inode, NFSIOS_SHORTREAD); /* Has the server at least made some progress? */ if (resp->count == 0) - return; + goto out; /* Yes, so retry the read at the end of the data */ argp->offset += resp->count; argp->pgbase += resp->count; argp->count -= resp->count; - rpc_restart_call(task); + nfs4_restart_rpc(task, NFS_SERVER(data->inode)->nfs_client); + return; +out: + nfs4_sequence_free_slot(NFS_SERVER(data->inode)->nfs_client, + &data->res.seq_res); + return; + } /* @@ -406,7 +414,23 @@ static void nfs_readpage_release_partial(void *calldata) nfs_readdata_release(calldata); } +#if defined(CONFIG_NFS_V4_1) +void nfs_read_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs_read_data *data = calldata; + + if (nfs4_setup_sequence(NFS_SERVER(data->inode)->nfs_client, + &data->args.seq_args, &data->res.seq_res, + 0, task)) + return; + rpc_call_start(task); +} +#endif /* CONFIG_NFS_V4_1 */ + static const struct rpc_call_ops nfs_read_partial_ops = { +#if defined(CONFIG_NFS_V4_1) + .rpc_call_prepare = nfs_read_prepare, +#endif /* CONFIG_NFS_V4_1 */ .rpc_call_done = nfs_readpage_result_partial, .rpc_release = nfs_readpage_release_partial, }; @@ -470,6 +494,9 @@ static void nfs_readpage_release_full(void *calldata) } static const struct rpc_call_ops nfs_read_full_ops = { +#if defined(CONFIG_NFS_V4_1) + .rpc_call_prepare = nfs_read_prepare, +#endif /* CONFIG_NFS_V4_1 */ .rpc_call_done = nfs_readpage_result_full, .rpc_release = nfs_readpage_release_full, }; diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 6717200923fe..a3d3cf20776c 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -90,6 +90,7 @@ enum { Opt_mountport, Opt_mountvers, Opt_nfsvers, + Opt_minorversion, /* Mount options that take string arguments */ Opt_sec, Opt_proto, Opt_mountproto, Opt_mounthost, @@ -155,6 +156,7 @@ static const match_table_t nfs_mount_option_tokens = { { Opt_mountvers, "mountvers=%u" }, { Opt_nfsvers, "nfsvers=%u" }, { Opt_nfsvers, "vers=%u" }, + { Opt_minorversion, "minorversion=%u" }, { Opt_sec, "sec=%s" }, { Opt_proto, "proto=%s" }, @@ -1206,6 +1208,13 @@ static int nfs_parse_mount_options(char *raw, nfs_parse_invalid_value("nfsvers"); } break; + case Opt_minorversion: + if (match_int(args, &option)) + return 0; + if (option < 0 || option > NFS4_MAX_MINOR_VERSION) + return 0; + mnt->minorversion = option; + break; /* * options that take text values @@ -2258,6 +2267,7 @@ static int nfs4_validate_mount_data(void *options, args->nfs_server.port = NFS_PORT; /* 2049 unless user set port= */ args->auth_flavors[0] = RPC_AUTH_UNIX; args->auth_flavor_len = 0; + args->minorversion = 0; switch (data->version) { case 1: @@ -2473,12 +2483,14 @@ static void nfs4_kill_super(struct super_block *sb) { struct nfs_server *server = NFS_SB(sb); + dprintk("--> %s\n", __func__); nfs_super_return_all_delegations(sb); kill_anon_super(sb); nfs4_renewd_prepare_shutdown(server); nfs_fscache_release_super_cookie(sb); nfs_free_server(server); + dprintk("<-- %s\n", __func__); } /* diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index ecc295347775..1064c91ae810 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -15,6 +15,7 @@ #include <linux/wait.h> #include "internal.h" +#include "nfs4_fs.h" struct nfs_unlinkdata { struct hlist_node list; @@ -82,7 +83,7 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata) struct inode *dir = data->dir; if (!NFS_PROTO(dir)->unlink_done(task, dir)) - rpc_restart_call(task); + nfs4_restart_rpc(task, NFS_SERVER(dir)->nfs_client); } /** @@ -102,9 +103,25 @@ static void nfs_async_unlink_release(void *calldata) nfs_sb_deactive(sb); } +#if defined(CONFIG_NFS_V4_1) +void nfs_unlink_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs_unlinkdata *data = calldata; + struct nfs_server *server = NFS_SERVER(data->dir); + + if (nfs4_setup_sequence(server->nfs_client, &data->args.seq_args, + &data->res.seq_res, 1, task)) + return; + rpc_call_start(task); +} +#endif /* CONFIG_NFS_V4_1 */ + static const struct rpc_call_ops nfs_unlink_ops = { .rpc_call_done = nfs_async_unlink_done, .rpc_release = nfs_async_unlink_release, +#if defined(CONFIG_NFS_V4_1) + .rpc_call_prepare = nfs_unlink_prepare, +#endif /* CONFIG_NFS_V4_1 */ }; static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct nfs_unlinkdata *data) @@ -241,6 +258,7 @@ nfs_async_unlink(struct inode *dir, struct dentry *dentry) status = PTR_ERR(data->cred); goto out_free; } + data->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE; status = -EBUSY; spin_lock(&dentry->d_lock); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index e560a78995a3..ce728829f79a 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -25,6 +25,7 @@ #include "delegation.h" #include "internal.h" #include "iostat.h" +#include "nfs4_fs.h" #define NFSDBG_FACILITY NFSDBG_PAGECACHE @@ -52,6 +53,7 @@ struct nfs_write_data *nfs_commitdata_alloc(void) if (p) { memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->pages); + p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE; } return p; } @@ -71,6 +73,7 @@ struct nfs_write_data *nfs_writedata_alloc(unsigned int pagecount) memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->pages); p->npages = pagecount; + p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE; if (pagecount <= ARRAY_SIZE(p->page_array)) p->pagevec = p->page_array; else { @@ -1048,7 +1051,23 @@ out: nfs_writedata_release(calldata); } +#if defined(CONFIG_NFS_V4_1) +void nfs_write_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs_write_data *data = calldata; + struct nfs_client *clp = (NFS_SERVER(data->inode))->nfs_client; + + if (nfs4_setup_sequence(clp, &data->args.seq_args, + &data->res.seq_res, 1, task)) + return; + rpc_call_start(task); +} +#endif /* CONFIG_NFS_V4_1 */ + static const struct rpc_call_ops nfs_write_partial_ops = { +#if defined(CONFIG_NFS_V4_1) + .rpc_call_prepare = nfs_write_prepare, +#endif /* CONFIG_NFS_V4_1 */ .rpc_call_done = nfs_writeback_done_partial, .rpc_release = nfs_writeback_release_partial, }; @@ -1111,6 +1130,9 @@ remove_request: } static const struct rpc_call_ops nfs_write_full_ops = { +#if defined(CONFIG_NFS_V4_1) + .rpc_call_prepare = nfs_write_prepare, +#endif /* CONFIG_NFS_V4_1 */ .rpc_call_done = nfs_writeback_done_full, .rpc_release = nfs_writeback_release_full, }; @@ -1123,6 +1145,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) { struct nfs_writeargs *argp = &data->args; struct nfs_writeres *resp = &data->res; + struct nfs_server *server = NFS_SERVER(data->inode); int status; dprintk("NFS: %5u nfs_writeback_done (status %d)\n", @@ -1155,7 +1178,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) if (time_before(complain, jiffies)) { dprintk("NFS: faulty NFS server %s:" " (committed = %d) != (stable = %d)\n", - NFS_SERVER(data->inode)->nfs_client->cl_hostname, + server->nfs_client->cl_hostname, resp->verf->committed, argp->stable); complain = jiffies + 300 * HZ; } @@ -1181,7 +1204,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) */ argp->stable = NFS_FILE_SYNC; } - rpc_restart_call(task); + nfs4_restart_rpc(task, server->nfs_client); return -EAGAIN; } if (time_before(complain, jiffies)) { @@ -1193,6 +1216,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) /* Can't do anything about it except throw an error. */ task->tk_status = -EIO; } + nfs4_sequence_free_slot(server->nfs_client, &data->res.seq_res); return 0; } @@ -1349,6 +1373,9 @@ static void nfs_commit_release(void *calldata) } static const struct rpc_call_ops nfs_commit_ops = { +#if defined(CONFIG_NFS_V4_1) + .rpc_call_prepare = nfs_write_prepare, +#endif /* CONFIG_NFS_V4_1 */ .rpc_call_done = nfs_commit_done, .rpc_release = nfs_commit_release, }; diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 3444c0052a87..5275097a7565 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -229,21 +229,23 @@ nfsd4_list_rec_dir(struct dentry *dir, recdir_func *f) goto out; status = vfs_readdir(filp, nfsd4_build_namelist, &names); fput(filp); + mutex_lock(&dir->d_inode->i_mutex); while (!list_empty(&names)) { entry = list_entry(names.next, struct name_list, list); dentry = lookup_one_len(entry->name, dir, HEXDIR_LEN-1); if (IS_ERR(dentry)) { status = PTR_ERR(dentry); - goto out; + break; } status = f(dir, dentry); dput(dentry); if (status) - goto out; + break; list_del(&entry->list); kfree(entry); } + mutex_unlock(&dir->d_inode->i_mutex); out: while (!list_empty(&names)) { entry = list_entry(names.next, struct name_list, list); @@ -255,36 +257,6 @@ out: } static int -nfsd4_remove_clid_file(struct dentry *dir, struct dentry *dentry) -{ - int status; - - if (!S_ISREG(dir->d_inode->i_mode)) { - printk("nfsd4: non-file found in client recovery directory\n"); - return -EINVAL; - } - mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT); - status = vfs_unlink(dir->d_inode, dentry); - mutex_unlock(&dir->d_inode->i_mutex); - return status; -} - -static int -nfsd4_clear_clid_dir(struct dentry *dir, struct dentry *dentry) -{ - int status; - - /* For now this directory should already be empty, but we empty it of - * any regular files anyway, just in case the directory was created by - * a kernel from the future.... */ - nfsd4_list_rec_dir(dentry, nfsd4_remove_clid_file); - mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT); - status = vfs_rmdir(dir->d_inode, dentry); - mutex_unlock(&dir->d_inode->i_mutex); - return status; -} - -static int nfsd4_unlink_clid_dir(char *name, int namlen) { struct dentry *dentry; @@ -294,18 +266,18 @@ nfsd4_unlink_clid_dir(char *name, int namlen) mutex_lock(&rec_dir.dentry->d_inode->i_mutex); dentry = lookup_one_len(name, rec_dir.dentry, namlen); - mutex_unlock(&rec_dir.dentry->d_inode->i_mutex); if (IS_ERR(dentry)) { status = PTR_ERR(dentry); - return status; + goto out_unlock; } status = -ENOENT; if (!dentry->d_inode) goto out; - - status = nfsd4_clear_clid_dir(rec_dir.dentry, dentry); + status = vfs_rmdir(rec_dir.dentry->d_inode, dentry); out: dput(dentry); +out_unlock: + mutex_unlock(&rec_dir.dentry->d_inode->i_mutex); return status; } @@ -348,7 +320,7 @@ purge_old(struct dentry *parent, struct dentry *child) if (nfs4_has_reclaimed_state(child->d_name.name, false)) return 0; - status = nfsd4_clear_clid_dir(parent, child); + status = vfs_rmdir(parent->d_inode, child); if (status) printk("failed to remove client recovery directory %s\n", child->d_name.name); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index ab93fcfef254..6c68ffd6b4bb 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -116,10 +116,15 @@ nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp, } if ((exp->ex_flags & NFSEXP_CROSSMOUNT) || EX_NOHIDE(exp2)) { /* successfully crossed mount point */ - exp_put(exp); - *expp = exp2; + /* + * This is subtle: dentry is *not* under mnt at this point. + * The only reason we are safe is that original mnt is pinned + * down by exp, so we should dput before putting exp. + */ dput(dentry); *dpp = mounts; + exp_put(exp); + *expp = exp2; } else { exp_put(exp2); dput(mounts); @@ -1885,8 +1890,8 @@ static int nfsd_buffered_filldir(void *__buf, const char *name, int namlen, return 0; } -static int nfsd_buffered_readdir(struct file *file, filldir_t func, - struct readdir_cd *cdp, loff_t *offsetp) +static __be32 nfsd_buffered_readdir(struct file *file, filldir_t func, + struct readdir_cd *cdp, loff_t *offsetp) { struct readdir_data buf; struct buffered_dirent *de; @@ -1896,11 +1901,12 @@ static int nfsd_buffered_readdir(struct file *file, filldir_t func, buf.dirent = (void *)__get_free_page(GFP_KERNEL); if (!buf.dirent) - return -ENOMEM; + return nfserrno(-ENOMEM); offset = *offsetp; while (1) { + struct inode *dir_inode = file->f_path.dentry->d_inode; unsigned int reclen; cdp->err = nfserr_eof; /* will be cleared on successful read */ @@ -1919,26 +1925,38 @@ static int nfsd_buffered_readdir(struct file *file, filldir_t func, if (!size) break; + /* + * Various filldir functions may end up calling back into + * lookup_one_len() and the file system's ->lookup() method. + * These expect i_mutex to be held, as it would within readdir. + */ + host_err = mutex_lock_killable(&dir_inode->i_mutex); + if (host_err) + break; + de = (struct buffered_dirent *)buf.dirent; while (size > 0) { offset = de->offset; if (func(cdp, de->name, de->namlen, de->offset, de->ino, de->d_type)) - goto done; + break; if (cdp->err != nfs_ok) - goto done; + break; reclen = ALIGN(sizeof(*de) + de->namlen, sizeof(u64)); size -= reclen; de = (struct buffered_dirent *)((char *)de + reclen); } + mutex_unlock(&dir_inode->i_mutex); + if (size > 0) /* We bailed out early */ + break; + offset = vfs_llseek(file, 0, SEEK_CUR); } - done: free_page((unsigned long)(buf.dirent)); if (host_err) diff --git a/fs/nilfs2/bmap.c b/fs/nilfs2/bmap.c index 24638e059bf3..064279e33bbb 100644 --- a/fs/nilfs2/bmap.c +++ b/fs/nilfs2/bmap.c @@ -688,6 +688,8 @@ static const struct nilfs_bmap_ptr_operations nilfs_bmap_ptr_ops_gc = { .bpop_translate = NULL, }; +static struct lock_class_key nilfs_bmap_dat_lock_key; + /** * nilfs_bmap_read - read a bmap from an inode * @bmap: bmap @@ -715,6 +717,7 @@ int nilfs_bmap_read(struct nilfs_bmap *bmap, struct nilfs_inode *raw_inode) bmap->b_pops = &nilfs_bmap_ptr_ops_p; bmap->b_last_allocated_key = 0; /* XXX: use macro */ bmap->b_last_allocated_ptr = NILFS_BMAP_NEW_PTR_INIT; + lockdep_set_class(&bmap->b_sem, &nilfs_bmap_dat_lock_key); break; case NILFS_CPFILE_INO: case NILFS_SUFILE_INO: @@ -772,6 +775,7 @@ void nilfs_bmap_init_gcdat(struct nilfs_bmap *gcbmap, struct nilfs_bmap *bmap) { memcpy(gcbmap, bmap, sizeof(union nilfs_bmap_union)); init_rwsem(&gcbmap->b_sem); + lockdep_set_class(&bmap->b_sem, &nilfs_bmap_dat_lock_key); gcbmap->b_inode = &NILFS_BMAP_I(gcbmap)->vfs_inode; } @@ -779,5 +783,6 @@ void nilfs_bmap_commit_gcdat(struct nilfs_bmap *gcbmap, struct nilfs_bmap *bmap) { memcpy(bmap, gcbmap, sizeof(union nilfs_bmap_union)); init_rwsem(&bmap->b_sem); + lockdep_set_class(&bmap->b_sem, &nilfs_bmap_dat_lock_key); bmap->b_inode = &NILFS_BMAP_I(bmap)->vfs_inode; } diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h index 7558c977db02..3d0c18a16db1 100644 --- a/fs/nilfs2/nilfs.h +++ b/fs/nilfs2/nilfs.h @@ -35,11 +35,6 @@ #include "bmap_union.h" /* - * NILFS filesystem version - */ -#define NILFS_VERSION "2.0.5" - -/* * nilfs inode data in memory */ struct nilfs_inode_info { diff --git a/fs/nilfs2/recovery.c b/fs/nilfs2/recovery.c index 6ade0963fc1d..4fc081e47d70 100644 --- a/fs/nilfs2/recovery.c +++ b/fs/nilfs2/recovery.c @@ -413,7 +413,6 @@ static int nilfs_prepare_segment_for_recovery(struct the_nilfs *nilfs, struct nilfs_segment_entry *ent, *n; struct inode *sufile = nilfs->ns_sufile; __u64 segnum[4]; - time_t mtime; int err; int i; @@ -442,24 +441,13 @@ static int nilfs_prepare_segment_for_recovery(struct the_nilfs *nilfs, * Collecting segments written after the latest super root. * These are marked dirty to avoid being reallocated in the next write. */ - mtime = get_seconds(); list_for_each_entry_safe(ent, n, head, list) { - if (ent->segnum == segnum[0]) { - list_del(&ent->list); - nilfs_free_segment_entry(ent); - continue; - } - err = nilfs_open_segment_entry(ent, sufile); - if (unlikely(err)) - goto failed; - if (!nilfs_segment_usage_dirty(ent->raw_su)) { - /* make the segment garbage */ - ent->raw_su->su_nblocks = cpu_to_le32(0); - ent->raw_su->su_lastmod = cpu_to_le32(mtime); - nilfs_segment_usage_set_dirty(ent->raw_su); + if (ent->segnum != segnum[0]) { + err = nilfs_sufile_scrap(sufile, ent->segnum); + if (unlikely(err)) + goto failed; } list_del(&ent->list); - nilfs_close_segment_entry(ent, sufile); nilfs_free_segment_entry(ent); } diff --git a/fs/nilfs2/sufile.c b/fs/nilfs2/sufile.c index c774cf397e2f..98e68677f045 100644 --- a/fs/nilfs2/sufile.c +++ b/fs/nilfs2/sufile.c @@ -93,6 +93,52 @@ nilfs_sufile_get_segment_usage_block(struct inode *sufile, __u64 segnum, create, NULL, bhp); } +static void nilfs_sufile_mod_counter(struct buffer_head *header_bh, + u64 ncleanadd, u64 ndirtyadd) +{ + struct nilfs_sufile_header *header; + void *kaddr; + + kaddr = kmap_atomic(header_bh->b_page, KM_USER0); + header = kaddr + bh_offset(header_bh); + le64_add_cpu(&header->sh_ncleansegs, ncleanadd); + le64_add_cpu(&header->sh_ndirtysegs, ndirtyadd); + kunmap_atomic(kaddr, KM_USER0); + + nilfs_mdt_mark_buffer_dirty(header_bh); +} + +int nilfs_sufile_update(struct inode *sufile, __u64 segnum, int create, + void (*dofunc)(struct inode *, __u64, + struct buffer_head *, + struct buffer_head *)) +{ + struct buffer_head *header_bh, *bh; + int ret; + + if (unlikely(segnum >= nilfs_sufile_get_nsegments(sufile))) { + printk(KERN_WARNING "%s: invalid segment number: %llu\n", + __func__, (unsigned long long)segnum); + return -EINVAL; + } + down_write(&NILFS_MDT(sufile)->mi_sem); + + ret = nilfs_sufile_get_header_block(sufile, &header_bh); + if (ret < 0) + goto out_sem; + + ret = nilfs_sufile_get_segment_usage_block(sufile, segnum, create, &bh); + if (!ret) { + dofunc(sufile, segnum, header_bh, bh); + brelse(bh); + } + brelse(header_bh); + + out_sem: + up_write(&NILFS_MDT(sufile)->mi_sem); + return ret; +} + /** * nilfs_sufile_alloc - allocate a segment * @sufile: inode of segment usage file @@ -113,7 +159,6 @@ nilfs_sufile_get_segment_usage_block(struct inode *sufile, __u64 segnum, int nilfs_sufile_alloc(struct inode *sufile, __u64 *segnump) { struct buffer_head *header_bh, *su_bh; - struct the_nilfs *nilfs; struct nilfs_sufile_header *header; struct nilfs_segment_usage *su; size_t susz = NILFS_MDT(sufile)->mi_entry_size; @@ -124,8 +169,6 @@ int nilfs_sufile_alloc(struct inode *sufile, __u64 *segnump) down_write(&NILFS_MDT(sufile)->mi_sem); - nilfs = NILFS_MDT(sufile)->mi_nilfs; - ret = nilfs_sufile_get_header_block(sufile, &header_bh); if (ret < 0) goto out_sem; @@ -192,165 +235,84 @@ int nilfs_sufile_alloc(struct inode *sufile, __u64 *segnump) return ret; } -/** - * nilfs_sufile_cancel_free - - * @sufile: inode of segment usage file - * @segnum: segment number - * - * Description: - * - * Return Value: On success, 0 is returned. On error, one of the following - * negative error codes is returned. - * - * %-EIO - I/O error. - * - * %-ENOMEM - Insufficient amount of memory available. - */ -int nilfs_sufile_cancel_free(struct inode *sufile, __u64 segnum) +void nilfs_sufile_do_cancel_free(struct inode *sufile, __u64 segnum, + struct buffer_head *header_bh, + struct buffer_head *su_bh) { - struct buffer_head *header_bh, *su_bh; - struct the_nilfs *nilfs; - struct nilfs_sufile_header *header; struct nilfs_segment_usage *su; void *kaddr; - int ret; - - down_write(&NILFS_MDT(sufile)->mi_sem); - - nilfs = NILFS_MDT(sufile)->mi_nilfs; - - ret = nilfs_sufile_get_header_block(sufile, &header_bh); - if (ret < 0) - goto out_sem; - - ret = nilfs_sufile_get_segment_usage_block(sufile, segnum, 0, &su_bh); - if (ret < 0) - goto out_header; kaddr = kmap_atomic(su_bh->b_page, KM_USER0); - su = nilfs_sufile_block_get_segment_usage( - sufile, segnum, su_bh, kaddr); + su = nilfs_sufile_block_get_segment_usage(sufile, segnum, su_bh, kaddr); if (unlikely(!nilfs_segment_usage_clean(su))) { printk(KERN_WARNING "%s: segment %llu must be clean\n", __func__, (unsigned long long)segnum); kunmap_atomic(kaddr, KM_USER0); - goto out_su_bh; + return; } nilfs_segment_usage_set_dirty(su); kunmap_atomic(kaddr, KM_USER0); - kaddr = kmap_atomic(header_bh->b_page, KM_USER0); - header = nilfs_sufile_block_get_header(sufile, header_bh, kaddr); - le64_add_cpu(&header->sh_ncleansegs, -1); - le64_add_cpu(&header->sh_ndirtysegs, 1); - kunmap_atomic(kaddr, KM_USER0); - - nilfs_mdt_mark_buffer_dirty(header_bh); + nilfs_sufile_mod_counter(header_bh, -1, 1); nilfs_mdt_mark_buffer_dirty(su_bh); nilfs_mdt_mark_dirty(sufile); - - out_su_bh: - brelse(su_bh); - out_header: - brelse(header_bh); - out_sem: - up_write(&NILFS_MDT(sufile)->mi_sem); - return ret; } -/** - * nilfs_sufile_freev - free segments - * @sufile: inode of segment usage file - * @segnum: array of segment numbers - * @nsegs: number of segments - * - * Description: nilfs_sufile_freev() frees segments specified by @segnum and - * @nsegs, which must have been returned by a previous call to - * nilfs_sufile_alloc(). - * - * Return Value: On success, 0 is returned. On error, one of the following - * negative error codes is returned. - * - * %-EIO - I/O error. - * - * %-ENOMEM - Insufficient amount of memory available. - */ -#define NILFS_SUFILE_FREEV_PREALLOC 16 -int nilfs_sufile_freev(struct inode *sufile, __u64 *segnum, size_t nsegs) +void nilfs_sufile_do_scrap(struct inode *sufile, __u64 segnum, + struct buffer_head *header_bh, + struct buffer_head *su_bh) { - struct buffer_head *header_bh, **su_bh, - *su_bh_prealloc[NILFS_SUFILE_FREEV_PREALLOC]; - struct the_nilfs *nilfs; - struct nilfs_sufile_header *header; struct nilfs_segment_usage *su; void *kaddr; - int ret, i; + int clean, dirty; - down_write(&NILFS_MDT(sufile)->mi_sem); - - nilfs = NILFS_MDT(sufile)->mi_nilfs; - - /* prepare resources */ - if (nsegs <= NILFS_SUFILE_FREEV_PREALLOC) - su_bh = su_bh_prealloc; - else { - su_bh = kmalloc(sizeof(*su_bh) * nsegs, GFP_NOFS); - if (su_bh == NULL) { - ret = -ENOMEM; - goto out_sem; - } - } - - ret = nilfs_sufile_get_header_block(sufile, &header_bh); - if (ret < 0) - goto out_su_bh; - for (i = 0; i < nsegs; i++) { - ret = nilfs_sufile_get_segment_usage_block(sufile, segnum[i], - 0, &su_bh[i]); - if (ret < 0) - goto out_bh; - } - - /* free segments */ - for (i = 0; i < nsegs; i++) { - kaddr = kmap_atomic(su_bh[i]->b_page, KM_USER0); - su = nilfs_sufile_block_get_segment_usage( - sufile, segnum[i], su_bh[i], kaddr); - WARN_ON(nilfs_segment_usage_error(su)); - nilfs_segment_usage_set_clean(su); + kaddr = kmap_atomic(su_bh->b_page, KM_USER0); + su = nilfs_sufile_block_get_segment_usage(sufile, segnum, su_bh, kaddr); + if (su->su_flags == cpu_to_le32(1UL << NILFS_SEGMENT_USAGE_DIRTY) && + su->su_nblocks == cpu_to_le32(0)) { kunmap_atomic(kaddr, KM_USER0); - nilfs_mdt_mark_buffer_dirty(su_bh[i]); + return; } - kaddr = kmap_atomic(header_bh->b_page, KM_USER0); - header = nilfs_sufile_block_get_header(sufile, header_bh, kaddr); - le64_add_cpu(&header->sh_ncleansegs, nsegs); - le64_add_cpu(&header->sh_ndirtysegs, -(u64)nsegs); + clean = nilfs_segment_usage_clean(su); + dirty = nilfs_segment_usage_dirty(su); + + /* make the segment garbage */ + su->su_lastmod = cpu_to_le64(0); + su->su_nblocks = cpu_to_le32(0); + su->su_flags = cpu_to_le32(1UL << NILFS_SEGMENT_USAGE_DIRTY); kunmap_atomic(kaddr, KM_USER0); - nilfs_mdt_mark_buffer_dirty(header_bh); + + nilfs_sufile_mod_counter(header_bh, clean ? (u64)-1 : 0, dirty ? 0 : 1); + nilfs_mdt_mark_buffer_dirty(su_bh); nilfs_mdt_mark_dirty(sufile); +} - out_bh: - for (i--; i >= 0; i--) - brelse(su_bh[i]); - brelse(header_bh); +void nilfs_sufile_do_free(struct inode *sufile, __u64 segnum, + struct buffer_head *header_bh, + struct buffer_head *su_bh) +{ + struct nilfs_segment_usage *su; + void *kaddr; + int sudirty; - out_su_bh: - if (su_bh != su_bh_prealloc) - kfree(su_bh); + kaddr = kmap_atomic(su_bh->b_page, KM_USER0); + su = nilfs_sufile_block_get_segment_usage(sufile, segnum, su_bh, kaddr); + if (nilfs_segment_usage_clean(su)) { + printk(KERN_WARNING "%s: segment %llu is already clean\n", + __func__, (unsigned long long)segnum); + kunmap_atomic(kaddr, KM_USER0); + return; + } + WARN_ON(nilfs_segment_usage_error(su)); + WARN_ON(!nilfs_segment_usage_dirty(su)); - out_sem: - up_write(&NILFS_MDT(sufile)->mi_sem); - return ret; -} + sudirty = nilfs_segment_usage_dirty(su); + nilfs_segment_usage_set_clean(su); + kunmap_atomic(kaddr, KM_USER0); + nilfs_mdt_mark_buffer_dirty(su_bh); -/** - * nilfs_sufile_free - - * @sufile: - * @segnum: - */ -int nilfs_sufile_free(struct inode *sufile, __u64 segnum) -{ - return nilfs_sufile_freev(sufile, &segnum, 1); + nilfs_sufile_mod_counter(header_bh, 1, sudirty ? (u64)-1 : 0); + nilfs_mdt_mark_dirty(sufile); } /** @@ -500,72 +462,28 @@ int nilfs_sufile_get_ncleansegs(struct inode *sufile, unsigned long *nsegsp) return ret; } -/** - * nilfs_sufile_set_error - mark a segment as erroneous - * @sufile: inode of segment usage file - * @segnum: segment number - * - * Description: nilfs_sufile_set_error() marks the segment specified by - * @segnum as erroneous. The error segment will never be used again. - * - * Return Value: On success, 0 is returned. On error, one of the following - * negative error codes is returned. - * - * %-EIO - I/O error. - * - * %-ENOMEM - Insufficient amount of memory available. - * - * %-EINVAL - Invalid segment usage number. - */ -int nilfs_sufile_set_error(struct inode *sufile, __u64 segnum) +void nilfs_sufile_do_set_error(struct inode *sufile, __u64 segnum, + struct buffer_head *header_bh, + struct buffer_head *su_bh) { - struct buffer_head *header_bh, *su_bh; struct nilfs_segment_usage *su; - struct nilfs_sufile_header *header; void *kaddr; - int ret; - - if (unlikely(segnum >= nilfs_sufile_get_nsegments(sufile))) { - printk(KERN_WARNING "%s: invalid segment number: %llu\n", - __func__, (unsigned long long)segnum); - return -EINVAL; - } - down_write(&NILFS_MDT(sufile)->mi_sem); - - ret = nilfs_sufile_get_header_block(sufile, &header_bh); - if (ret < 0) - goto out_sem; - ret = nilfs_sufile_get_segment_usage_block(sufile, segnum, 0, &su_bh); - if (ret < 0) - goto out_header; + int suclean; kaddr = kmap_atomic(su_bh->b_page, KM_USER0); su = nilfs_sufile_block_get_segment_usage(sufile, segnum, su_bh, kaddr); if (nilfs_segment_usage_error(su)) { kunmap_atomic(kaddr, KM_USER0); - brelse(su_bh); - goto out_header; + return; } - + suclean = nilfs_segment_usage_clean(su); nilfs_segment_usage_set_error(su); kunmap_atomic(kaddr, KM_USER0); - brelse(su_bh); - kaddr = kmap_atomic(header_bh->b_page, KM_USER0); - header = nilfs_sufile_block_get_header(sufile, header_bh, kaddr); - le64_add_cpu(&header->sh_ndirtysegs, -1); - kunmap_atomic(kaddr, KM_USER0); - nilfs_mdt_mark_buffer_dirty(header_bh); + if (suclean) + nilfs_sufile_mod_counter(header_bh, -1, 0); nilfs_mdt_mark_buffer_dirty(su_bh); nilfs_mdt_mark_dirty(sufile); - brelse(su_bh); - - out_header: - brelse(header_bh); - - out_sem: - up_write(&NILFS_MDT(sufile)->mi_sem); - return ret; } /** @@ -625,7 +543,7 @@ ssize_t nilfs_sufile_get_suinfo(struct inode *sufile, __u64 segnum, si[i + j].sui_nblocks = le32_to_cpu(su->su_nblocks); si[i + j].sui_flags = le32_to_cpu(su->su_flags) & ~(1UL << NILFS_SEGMENT_USAGE_ACTIVE); - if (nilfs_segment_is_active(nilfs, segnum + i + j)) + if (nilfs_segment_is_active(nilfs, segnum + j)) si[i + j].sui_flags |= (1UL << NILFS_SEGMENT_USAGE_ACTIVE); } diff --git a/fs/nilfs2/sufile.h b/fs/nilfs2/sufile.h index d595f33a768d..a2e2efd4ade1 100644 --- a/fs/nilfs2/sufile.h +++ b/fs/nilfs2/sufile.h @@ -36,9 +36,6 @@ static inline unsigned long nilfs_sufile_get_nsegments(struct inode *sufile) } int nilfs_sufile_alloc(struct inode *, __u64 *); -int nilfs_sufile_cancel_free(struct inode *, __u64); -int nilfs_sufile_freev(struct inode *, __u64 *, size_t); -int nilfs_sufile_free(struct inode *, __u64); int nilfs_sufile_get_segment_usage(struct inode *, __u64, struct nilfs_segment_usage **, struct buffer_head **); @@ -46,9 +43,83 @@ void nilfs_sufile_put_segment_usage(struct inode *, __u64, struct buffer_head *); int nilfs_sufile_get_stat(struct inode *, struct nilfs_sustat *); int nilfs_sufile_get_ncleansegs(struct inode *, unsigned long *); -int nilfs_sufile_set_error(struct inode *, __u64); ssize_t nilfs_sufile_get_suinfo(struct inode *, __u64, struct nilfs_suinfo *, size_t); +int nilfs_sufile_update(struct inode *, __u64, int, + void (*dofunc)(struct inode *, __u64, + struct buffer_head *, + struct buffer_head *)); +void nilfs_sufile_do_cancel_free(struct inode *, __u64, struct buffer_head *, + struct buffer_head *); +void nilfs_sufile_do_scrap(struct inode *, __u64, struct buffer_head *, + struct buffer_head *); +void nilfs_sufile_do_free(struct inode *, __u64, struct buffer_head *, + struct buffer_head *); +void nilfs_sufile_do_set_error(struct inode *, __u64, struct buffer_head *, + struct buffer_head *); + +/** + * nilfs_sufile_cancel_free - + * @sufile: inode of segment usage file + * @segnum: segment number + * + * Description: + * + * Return Value: On success, 0 is returned. On error, one of the following + * negative error codes is returned. + * + * %-EIO - I/O error. + * + * %-ENOMEM - Insufficient amount of memory available. + */ +static inline int nilfs_sufile_cancel_free(struct inode *sufile, __u64 segnum) +{ + return nilfs_sufile_update(sufile, segnum, 0, + nilfs_sufile_do_cancel_free); +} + +/** + * nilfs_sufile_scrap - make a segment garbage + * @sufile: inode of segment usage file + * @segnum: segment number to be freed + */ +static inline int nilfs_sufile_scrap(struct inode *sufile, __u64 segnum) +{ + return nilfs_sufile_update(sufile, segnum, 1, nilfs_sufile_do_scrap); +} + +/** + * nilfs_sufile_free - free segment + * @sufile: inode of segment usage file + * @segnum: segment number to be freed + */ +static inline int nilfs_sufile_free(struct inode *sufile, __u64 segnum) +{ + return nilfs_sufile_update(sufile, segnum, 0, nilfs_sufile_do_free); +} + +/** + * nilfs_sufile_set_error - mark a segment as erroneous + * @sufile: inode of segment usage file + * @segnum: segment number + * + * Description: nilfs_sufile_set_error() marks the segment specified by + * @segnum as erroneous. The error segment will never be used again. + * + * Return Value: On success, 0 is returned. On error, one of the following + * negative error codes is returned. + * + * %-EIO - I/O error. + * + * %-ENOMEM - Insufficient amount of memory available. + * + * %-EINVAL - Invalid segment usage number. + */ +static inline int nilfs_sufile_set_error(struct inode *sufile, __u64 segnum) +{ + return nilfs_sufile_update(sufile, segnum, 0, + nilfs_sufile_do_set_error); +} #endif /* _NILFS_SUFILE_H */ diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c index e117e1ea9bff..6989b03e97ab 100644 --- a/fs/nilfs2/super.c +++ b/fs/nilfs2/super.c @@ -63,7 +63,6 @@ MODULE_AUTHOR("NTT Corp."); MODULE_DESCRIPTION("A New Implementation of the Log-structured Filesystem " "(NILFS)"); -MODULE_VERSION(NILFS_VERSION); MODULE_LICENSE("GPL"); static int nilfs_remount(struct super_block *sb, int *flags, char *data); @@ -476,11 +475,12 @@ static int nilfs_statfs(struct dentry *dentry, struct kstatfs *buf) { struct super_block *sb = dentry->d_sb; struct nilfs_sb_info *sbi = NILFS_SB(sb); + struct the_nilfs *nilfs = sbi->s_nilfs; + u64 id = huge_encode_dev(sb->s_bdev->bd_dev); unsigned long long blocks; unsigned long overhead; unsigned long nrsvblocks; sector_t nfreeblocks; - struct the_nilfs *nilfs = sbi->s_nilfs; int err; /* @@ -514,6 +514,9 @@ static int nilfs_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_files = atomic_read(&sbi->s_inodes_count); buf->f_ffree = 0; /* nilfs_count_free_inodes(sb); */ buf->f_namelen = NILFS_NAME_LEN; + buf->f_fsid.val[0] = (u32)id; + buf->f_fsid.val[1] = (u32)(id >> 32); + return 0; } diff --git a/fs/nilfs2/the_nilfs.c b/fs/nilfs2/the_nilfs.c index 33400cf0bbe2..7f65b3be4aa9 100644 --- a/fs/nilfs2/the_nilfs.c +++ b/fs/nilfs2/the_nilfs.c @@ -115,6 +115,7 @@ void put_nilfs(struct the_nilfs *nilfs) static int nilfs_load_super_root(struct the_nilfs *nilfs, struct nilfs_sb_info *sbi, sector_t sr_block) { + static struct lock_class_key dat_lock_key; struct buffer_head *bh_sr; struct nilfs_super_root *raw_sr; struct nilfs_super_block **sbp = nilfs->ns_sbp; @@ -163,6 +164,9 @@ static int nilfs_load_super_root(struct the_nilfs *nilfs, if (unlikely(err)) goto failed_sufile; + lockdep_set_class(&NILFS_MDT(nilfs->ns_dat)->mi_sem, &dat_lock_key); + lockdep_set_class(&NILFS_MDT(nilfs->ns_gc_dat)->mi_sem, &dat_lock_key); + nilfs_mdt_set_shadow(nilfs->ns_dat, nilfs->ns_gc_dat); nilfs_mdt_set_entry_size(nilfs->ns_cpfile, checkpoint_size, sizeof(struct nilfs_cpfile_header)); diff --git a/fs/ocfs2/dcache.c b/fs/ocfs2/dcache.c index 7d604480557a..b574431a031d 100644 --- a/fs/ocfs2/dcache.c +++ b/fs/ocfs2/dcache.c @@ -290,6 +290,21 @@ out_attach: else mlog_errno(ret); + /* + * In case of error, manually free the allocation and do the iput(). + * We need to do this because error here means no d_instantiate(), + * which means iput() will not be called during dput(dentry). + */ + if (ret < 0 && !alias) { + ocfs2_lock_res_free(&dl->dl_lockres); + BUG_ON(dl->dl_count != 1); + spin_lock(&dentry_attach_lock); + dentry->d_fsdata = NULL; + spin_unlock(&dentry_attach_lock); + kfree(dl); + iput(inode); + } + dput(alias); return ret; diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index e71160cda110..c5752305627c 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -2697,7 +2697,7 @@ static int ocfs2_dx_dir_index_block(struct inode *dir, u32 *num_dx_entries, struct buffer_head *dirent_bh) { - int ret, namelen, i; + int ret = 0, namelen, i; char *de_buf, *limit; struct ocfs2_dir_entry *de; struct buffer_head *dx_leaf_bh; @@ -2934,7 +2934,7 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh, */ BUG_ON(alloc > 2); - ret = ocfs2_reserve_clusters(osb, alloc, &data_ac); + ret = ocfs2_reserve_clusters(osb, alloc + dx_alloc, &data_ac); if (ret) { mlog_errno(ret); goto out; diff --git a/fs/ocfs2/export.c b/fs/ocfs2/export.c index de3da8eb558c..15713cbb865c 100644 --- a/fs/ocfs2/export.c +++ b/fs/ocfs2/export.c @@ -100,7 +100,8 @@ static struct dentry *ocfs2_get_dentry(struct super_block *sb, /* If the inode allocator bit is clear, this inode must be stale */ if (!set) { - mlog(0, "inode %llu suballoc bit is clear\n", blkno); + mlog(0, "inode %llu suballoc bit is clear\n", + (unsigned long long)blkno); status = -ESTALE; goto unlock_nfs_sync; } @@ -114,7 +115,7 @@ check_err: if (status < 0) { if (status == -ESTALE) { mlog(0, "stale inode ino: %llu generation: %u\n", - blkno, handle->ih_generation); + (unsigned long long)blkno, handle->ih_generation); } result = ERR_PTR(status); goto bail; @@ -129,8 +130,8 @@ check_err: check_gen: if (handle->ih_generation != inode->i_generation) { iput(inode); - mlog(0, "stale inode ino: %llu generation: %u\n", blkno, - handle->ih_generation); + mlog(0, "stale inode ino: %llu generation: %u\n", + (unsigned long long)blkno, handle->ih_generation); result = ERR_PTR(-ESTALE); goto bail; } diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 8672b9536039..c2a87c885b73 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1912,6 +1912,22 @@ out_sems: return written ? written : ret; } +static int ocfs2_splice_to_file(struct pipe_inode_info *pipe, + struct file *out, + struct splice_desc *sd) +{ + int ret; + + ret = ocfs2_prepare_inode_for_write(out->f_path.dentry, &sd->pos, + sd->total_len, 0, NULL); + if (ret < 0) { + mlog_errno(ret); + return ret; + } + + return splice_from_pipe_feed(pipe, sd, pipe_to_file); +} + static ssize_t ocfs2_file_splice_write(struct pipe_inode_info *pipe, struct file *out, loff_t *ppos, @@ -1919,38 +1935,76 @@ static ssize_t ocfs2_file_splice_write(struct pipe_inode_info *pipe, unsigned int flags) { int ret; - struct inode *inode = out->f_path.dentry->d_inode; + struct address_space *mapping = out->f_mapping; + struct inode *inode = mapping->host; + struct splice_desc sd = { + .total_len = len, + .flags = flags, + .pos = *ppos, + .u.file = out, + }; mlog_entry("(0x%p, 0x%p, %u, '%.*s')\n", out, pipe, (unsigned int)len, out->f_path.dentry->d_name.len, out->f_path.dentry->d_name.name); - mutex_lock_nested(&inode->i_mutex, I_MUTEX_PARENT); + if (pipe->inode) + mutex_lock_nested(&pipe->inode->i_mutex, I_MUTEX_PARENT); - ret = ocfs2_rw_lock(inode, 1); - if (ret < 0) { - mlog_errno(ret); - goto out; - } + splice_from_pipe_begin(&sd); + do { + ret = splice_from_pipe_next(pipe, &sd); + if (ret <= 0) + break; - ret = ocfs2_prepare_inode_for_write(out->f_path.dentry, ppos, len, 0, - NULL); - if (ret < 0) { - mlog_errno(ret); - goto out_unlock; - } + mutex_lock_nested(&inode->i_mutex, I_MUTEX_CHILD); + ret = ocfs2_rw_lock(inode, 1); + if (ret < 0) + mlog_errno(ret); + else { + ret = ocfs2_splice_to_file(pipe, out, &sd); + ocfs2_rw_unlock(inode, 1); + } + mutex_unlock(&inode->i_mutex); + } while (ret > 0); + splice_from_pipe_end(pipe, &sd); if (pipe->inode) - mutex_lock_nested(&pipe->inode->i_mutex, I_MUTEX_CHILD); - ret = generic_file_splice_write_nolock(pipe, out, ppos, len, flags); - if (pipe->inode) mutex_unlock(&pipe->inode->i_mutex); -out_unlock: - ocfs2_rw_unlock(inode, 1); -out: - mutex_unlock(&inode->i_mutex); + if (sd.num_spliced) + ret = sd.num_spliced; + + if (ret > 0) { + unsigned long nr_pages; + + *ppos += ret; + nr_pages = (ret + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + + /* + * If file or inode is SYNC and we actually wrote some data, + * sync it. + */ + if (unlikely((out->f_flags & O_SYNC) || IS_SYNC(inode))) { + int err; + + mutex_lock(&inode->i_mutex); + err = ocfs2_rw_lock(inode, 1); + if (err < 0) { + mlog_errno(err); + } else { + err = generic_osync_inode(inode, mapping, + OSYNC_METADATA|OSYNC_DATA); + ocfs2_rw_unlock(inode, 1); + } + mutex_unlock(&inode->i_mutex); + + if (err) + ret = err; + } + balance_dirty_pages_ratelimited_nr(mapping, nr_pages); + } mlog_exit(ret); return ret; diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index b4ca5911caaf..8439f6b324b9 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -2197,26 +2197,29 @@ static int ocfs2_get_suballoc_slot_bit(struct ocfs2_super *osb, u64 blkno, struct buffer_head *inode_bh = NULL; struct ocfs2_dinode *inode_fe; - mlog_entry("blkno: %llu\n", blkno); + mlog_entry("blkno: %llu\n", (unsigned long long)blkno); /* dirty read disk */ status = ocfs2_read_blocks_sync(osb, blkno, 1, &inode_bh); if (status < 0) { - mlog(ML_ERROR, "read block %llu failed %d\n", blkno, status); + mlog(ML_ERROR, "read block %llu failed %d\n", + (unsigned long long)blkno, status); goto bail; } inode_fe = (struct ocfs2_dinode *) inode_bh->b_data; if (!OCFS2_IS_VALID_DINODE(inode_fe)) { - mlog(ML_ERROR, "invalid inode %llu requested\n", blkno); + mlog(ML_ERROR, "invalid inode %llu requested\n", + (unsigned long long)blkno); status = -EINVAL; goto bail; } - if (le16_to_cpu(inode_fe->i_suballoc_slot) != OCFS2_INVALID_SLOT && + if (le16_to_cpu(inode_fe->i_suballoc_slot) != (u16)OCFS2_INVALID_SLOT && (u32)le16_to_cpu(inode_fe->i_suballoc_slot) > osb->max_slots - 1) { mlog(ML_ERROR, "inode %llu has invalid suballoc slot %u\n", - blkno, (u32)le16_to_cpu(inode_fe->i_suballoc_slot)); + (unsigned long long)blkno, + (u32)le16_to_cpu(inode_fe->i_suballoc_slot)); status = -EINVAL; goto bail; } @@ -2251,7 +2254,8 @@ static int ocfs2_test_suballoc_bit(struct ocfs2_super *osb, u64 bg_blkno; int status; - mlog_entry("blkno: %llu bit: %u\n", blkno, (unsigned int)bit); + mlog_entry("blkno: %llu bit: %u\n", (unsigned long long)blkno, + (unsigned int)bit); alloc_fe = (struct ocfs2_dinode *)alloc_bh->b_data; if ((bit + 1) > ocfs2_bits_per_group(&alloc_fe->id2.i_chain)) { @@ -2266,7 +2270,8 @@ static int ocfs2_test_suballoc_bit(struct ocfs2_super *osb, status = ocfs2_read_group_descriptor(suballoc, alloc_fe, bg_blkno, &group_bh); if (status < 0) { - mlog(ML_ERROR, "read group %llu failed %d\n", bg_blkno, status); + mlog(ML_ERROR, "read group %llu failed %d\n", + (unsigned long long)bg_blkno, status); goto bail; } @@ -2300,7 +2305,7 @@ int ocfs2_test_inode_bit(struct ocfs2_super *osb, u64 blkno, int *res) struct inode *inode_alloc_inode; struct buffer_head *alloc_bh = NULL; - mlog_entry("blkno: %llu", blkno); + mlog_entry("blkno: %llu", (unsigned long long)blkno); status = ocfs2_get_suballoc_slot_bit(osb, blkno, &suballoc_slot, &suballoc_bit); diff --git a/fs/pipe.c b/fs/pipe.c index 4af7aa521813..13414ec45b8d 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -37,6 +37,42 @@ * -- Manfred Spraul <manfred@colorfullife.com> 2002-05-09 */ +static void pipe_lock_nested(struct pipe_inode_info *pipe, int subclass) +{ + if (pipe->inode) + mutex_lock_nested(&pipe->inode->i_mutex, subclass); +} + +void pipe_lock(struct pipe_inode_info *pipe) +{ + /* + * pipe_lock() nests non-pipe inode locks (for writing to a file) + */ + pipe_lock_nested(pipe, I_MUTEX_PARENT); +} +EXPORT_SYMBOL(pipe_lock); + +void pipe_unlock(struct pipe_inode_info *pipe) +{ + if (pipe->inode) + mutex_unlock(&pipe->inode->i_mutex); +} +EXPORT_SYMBOL(pipe_unlock); + +void pipe_double_lock(struct pipe_inode_info *pipe1, + struct pipe_inode_info *pipe2) +{ + BUG_ON(pipe1 == pipe2); + + if (pipe1 < pipe2) { + pipe_lock_nested(pipe1, I_MUTEX_PARENT); + pipe_lock_nested(pipe2, I_MUTEX_CHILD); + } else { + pipe_lock_nested(pipe2, I_MUTEX_CHILD); + pipe_lock_nested(pipe1, I_MUTEX_PARENT); + } +} + /* Drop the inode semaphore and wait for a pipe event, atomically */ void pipe_wait(struct pipe_inode_info *pipe) { @@ -47,12 +83,10 @@ void pipe_wait(struct pipe_inode_info *pipe) * is considered a noninteractive wait: */ prepare_to_wait(&pipe->wait, &wait, TASK_INTERRUPTIBLE); - if (pipe->inode) - mutex_unlock(&pipe->inode->i_mutex); + pipe_unlock(pipe); schedule(); finish_wait(&pipe->wait, &wait); - if (pipe->inode) - mutex_lock(&pipe->inode->i_mutex); + pipe_lock(pipe); } static int diff --git a/fs/proc/base.c b/fs/proc/base.c index f71559784bfb..aa763ab00777 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -648,14 +648,14 @@ static unsigned mounts_poll(struct file *file, poll_table *wait) { struct proc_mounts *p = file->private_data; struct mnt_namespace *ns = p->ns; - unsigned res = 0; + unsigned res = POLLIN | POLLRDNORM; poll_wait(file, &ns->poll, wait); spin_lock(&vfsmount_lock); if (p->event != ns->event) { p->event = ns->event; - res = POLLERR; + res |= POLLERR | POLLPRI; } spin_unlock(&vfsmount_lock); diff --git a/fs/proc/stat.c b/fs/proc/stat.c index f75efa22df5e..81e4eb60972e 100644 --- a/fs/proc/stat.c +++ b/fs/proc/stat.c @@ -18,6 +18,9 @@ #ifndef arch_irq_stat #define arch_irq_stat() 0 #endif +#ifndef arch_idle_time +#define arch_idle_time(cpu) 0 +#endif static int show_stat(struct seq_file *p, void *v) { @@ -40,6 +43,7 @@ static int show_stat(struct seq_file *p, void *v) nice = cputime64_add(nice, kstat_cpu(i).cpustat.nice); system = cputime64_add(system, kstat_cpu(i).cpustat.system); idle = cputime64_add(idle, kstat_cpu(i).cpustat.idle); + idle = cputime64_add(idle, arch_idle_time(i)); iowait = cputime64_add(iowait, kstat_cpu(i).cpustat.iowait); irq = cputime64_add(irq, kstat_cpu(i).cpustat.irq); softirq = cputime64_add(softirq, kstat_cpu(i).cpustat.softirq); @@ -69,6 +73,7 @@ static int show_stat(struct seq_file *p, void *v) nice = kstat_cpu(i).cpustat.nice; system = kstat_cpu(i).cpustat.system; idle = kstat_cpu(i).cpustat.idle; + idle = cputime64_add(idle, arch_idle_time(i)); iowait = kstat_cpu(i).cpustat.iowait; irq = kstat_cpu(i).cpustat.irq; softirq = kstat_cpu(i).cpustat.softirq; diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c index 12c20377772d..64a72e2e7650 100644 --- a/fs/proc/task_nommu.c +++ b/fs/proc/task_nommu.c @@ -135,7 +135,7 @@ static int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma) struct inode *inode = vma->vm_file->f_path.dentry->d_inode; dev = inode->i_sb->s_dev; ino = inode->i_ino; - pgoff = (loff_t)vma->pg_off << PAGE_SHIFT; + pgoff = (loff_t)vma->vm_pgoff << PAGE_SHIFT; } seq_printf(m, diff --git a/fs/splice.c b/fs/splice.c index c18aa7e03e2b..666953d59a35 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -182,8 +182,7 @@ ssize_t splice_to_pipe(struct pipe_inode_info *pipe, do_wakeup = 0; page_nr = 0; - if (pipe->inode) - mutex_lock(&pipe->inode->i_mutex); + pipe_lock(pipe); for (;;) { if (!pipe->readers) { @@ -245,15 +244,13 @@ ssize_t splice_to_pipe(struct pipe_inode_info *pipe, pipe->waiting_writers--; } - if (pipe->inode) { - mutex_unlock(&pipe->inode->i_mutex); + pipe_unlock(pipe); - if (do_wakeup) { - smp_mb(); - if (waitqueue_active(&pipe->wait)) - wake_up_interruptible(&pipe->wait); - kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); - } + if (do_wakeup) { + smp_mb(); + if (waitqueue_active(&pipe->wait)) + wake_up_interruptible(&pipe->wait); + kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); } while (page_nr < spd_pages) @@ -555,8 +552,8 @@ static int pipe_to_sendpage(struct pipe_inode_info *pipe, * SPLICE_F_MOVE isn't set, or we cannot move the page, we simply create * a new page in the output file page cache and fill/dirty that. */ -static int pipe_to_file(struct pipe_inode_info *pipe, struct pipe_buffer *buf, - struct splice_desc *sd) +int pipe_to_file(struct pipe_inode_info *pipe, struct pipe_buffer *buf, + struct splice_desc *sd) { struct file *file = sd->u.file; struct address_space *mapping = file->f_mapping; @@ -600,108 +597,177 @@ static int pipe_to_file(struct pipe_inode_info *pipe, struct pipe_buffer *buf, out: return ret; } +EXPORT_SYMBOL(pipe_to_file); + +static void wakeup_pipe_writers(struct pipe_inode_info *pipe) +{ + smp_mb(); + if (waitqueue_active(&pipe->wait)) + wake_up_interruptible(&pipe->wait); + kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT); +} /** - * __splice_from_pipe - splice data from a pipe to given actor + * splice_from_pipe_feed - feed available data from a pipe to a file * @pipe: pipe to splice from * @sd: information to @actor * @actor: handler that splices the data * * Description: - * This function does little more than loop over the pipe and call - * @actor to do the actual moving of a single struct pipe_buffer to - * the desired destination. See pipe_to_file, pipe_to_sendpage, or - * pipe_to_user. + * This function loops over the pipe and calls @actor to do the + * actual moving of a single struct pipe_buffer to the desired + * destination. It returns when there's no more buffers left in + * the pipe or if the requested number of bytes (@sd->total_len) + * have been copied. It returns a positive number (one) if the + * pipe needs to be filled with more data, zero if the required + * number of bytes have been copied and -errno on error. * + * This, together with splice_from_pipe_{begin,end,next}, may be + * used to implement the functionality of __splice_from_pipe() when + * locking is required around copying the pipe buffers to the + * destination. */ -ssize_t __splice_from_pipe(struct pipe_inode_info *pipe, struct splice_desc *sd, - splice_actor *actor) +int splice_from_pipe_feed(struct pipe_inode_info *pipe, struct splice_desc *sd, + splice_actor *actor) { - int ret, do_wakeup, err; - - ret = 0; - do_wakeup = 0; - - for (;;) { - if (pipe->nrbufs) { - struct pipe_buffer *buf = pipe->bufs + pipe->curbuf; - const struct pipe_buf_operations *ops = buf->ops; + int ret; - sd->len = buf->len; - if (sd->len > sd->total_len) - sd->len = sd->total_len; + while (pipe->nrbufs) { + struct pipe_buffer *buf = pipe->bufs + pipe->curbuf; + const struct pipe_buf_operations *ops = buf->ops; - err = actor(pipe, buf, sd); - if (err <= 0) { - if (!ret && err != -ENODATA) - ret = err; + sd->len = buf->len; + if (sd->len > sd->total_len) + sd->len = sd->total_len; - break; - } + ret = actor(pipe, buf, sd); + if (ret <= 0) { + if (ret == -ENODATA) + ret = 0; + return ret; + } + buf->offset += ret; + buf->len -= ret; - ret += err; - buf->offset += err; - buf->len -= err; + sd->num_spliced += ret; + sd->len -= ret; + sd->pos += ret; + sd->total_len -= ret; - sd->len -= err; - sd->pos += err; - sd->total_len -= err; - if (sd->len) - continue; + if (!buf->len) { + buf->ops = NULL; + ops->release(pipe, buf); + pipe->curbuf = (pipe->curbuf + 1) & (PIPE_BUFFERS - 1); + pipe->nrbufs--; + if (pipe->inode) + sd->need_wakeup = true; + } - if (!buf->len) { - buf->ops = NULL; - ops->release(pipe, buf); - pipe->curbuf = (pipe->curbuf + 1) & (PIPE_BUFFERS - 1); - pipe->nrbufs--; - if (pipe->inode) - do_wakeup = 1; - } + if (!sd->total_len) + return 0; + } - if (!sd->total_len) - break; - } + return 1; +} +EXPORT_SYMBOL(splice_from_pipe_feed); - if (pipe->nrbufs) - continue; +/** + * splice_from_pipe_next - wait for some data to splice from + * @pipe: pipe to splice from + * @sd: information about the splice operation + * + * Description: + * This function will wait for some data and return a positive + * value (one) if pipe buffers are available. It will return zero + * or -errno if no more data needs to be spliced. + */ +int splice_from_pipe_next(struct pipe_inode_info *pipe, struct splice_desc *sd) +{ + while (!pipe->nrbufs) { if (!pipe->writers) - break; - if (!pipe->waiting_writers) { - if (ret) - break; - } + return 0; - if (sd->flags & SPLICE_F_NONBLOCK) { - if (!ret) - ret = -EAGAIN; - break; - } + if (!pipe->waiting_writers && sd->num_spliced) + return 0; - if (signal_pending(current)) { - if (!ret) - ret = -ERESTARTSYS; - break; - } + if (sd->flags & SPLICE_F_NONBLOCK) + return -EAGAIN; - if (do_wakeup) { - smp_mb(); - if (waitqueue_active(&pipe->wait)) - wake_up_interruptible_sync(&pipe->wait); - kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT); - do_wakeup = 0; + if (signal_pending(current)) + return -ERESTARTSYS; + + if (sd->need_wakeup) { + wakeup_pipe_writers(pipe); + sd->need_wakeup = false; } pipe_wait(pipe); } - if (do_wakeup) { - smp_mb(); - if (waitqueue_active(&pipe->wait)) - wake_up_interruptible(&pipe->wait); - kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT); - } + return 1; +} +EXPORT_SYMBOL(splice_from_pipe_next); - return ret; +/** + * splice_from_pipe_begin - start splicing from pipe + * @sd: information about the splice operation + * + * Description: + * This function should be called before a loop containing + * splice_from_pipe_next() and splice_from_pipe_feed() to + * initialize the necessary fields of @sd. + */ +void splice_from_pipe_begin(struct splice_desc *sd) +{ + sd->num_spliced = 0; + sd->need_wakeup = false; +} +EXPORT_SYMBOL(splice_from_pipe_begin); + +/** + * splice_from_pipe_end - finish splicing from pipe + * @pipe: pipe to splice from + * @sd: information about the splice operation + * + * Description: + * This function will wake up pipe writers if necessary. It should + * be called after a loop containing splice_from_pipe_next() and + * splice_from_pipe_feed(). + */ +void splice_from_pipe_end(struct pipe_inode_info *pipe, struct splice_desc *sd) +{ + if (sd->need_wakeup) + wakeup_pipe_writers(pipe); +} +EXPORT_SYMBOL(splice_from_pipe_end); + +/** + * __splice_from_pipe - splice data from a pipe to given actor + * @pipe: pipe to splice from + * @sd: information to @actor + * @actor: handler that splices the data + * + * Description: + * This function does little more than loop over the pipe and call + * @actor to do the actual moving of a single struct pipe_buffer to + * the desired destination. See pipe_to_file, pipe_to_sendpage, or + * pipe_to_user. + * + */ +ssize_t __splice_from_pipe(struct pipe_inode_info *pipe, struct splice_desc *sd, + splice_actor *actor) +{ + int ret; + + splice_from_pipe_begin(sd); + do { + ret = splice_from_pipe_next(pipe, sd); + if (ret > 0) + ret = splice_from_pipe_feed(pipe, sd, actor); + } while (ret > 0); + splice_from_pipe_end(pipe, sd); + + return sd->num_spliced ? sd->num_spliced : ret; } EXPORT_SYMBOL(__splice_from_pipe); @@ -715,7 +781,7 @@ EXPORT_SYMBOL(__splice_from_pipe); * @actor: handler that splices the data * * Description: - * See __splice_from_pipe. This function locks the input and output inodes, + * See __splice_from_pipe. This function locks the pipe inode, * otherwise it's identical to __splice_from_pipe(). * */ @@ -724,7 +790,6 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, splice_actor *actor) { ssize_t ret; - struct inode *inode = out->f_mapping->host; struct splice_desc sd = { .total_len = len, .flags = flags, @@ -732,30 +797,15 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, .u.file = out, }; - /* - * The actor worker might be calling ->write_begin and - * ->write_end. Most of the time, these expect i_mutex to - * be held. Since this may result in an ABBA deadlock with - * pipe->inode, we have to order lock acquiry here. - * - * Outer lock must be inode->i_mutex, as pipe_wait() will - * release and reacquire pipe->inode->i_mutex, AND inode must - * never be a pipe. - */ - WARN_ON(S_ISFIFO(inode->i_mode)); - mutex_lock_nested(&inode->i_mutex, I_MUTEX_PARENT); - if (pipe->inode) - mutex_lock_nested(&pipe->inode->i_mutex, I_MUTEX_CHILD); + pipe_lock(pipe); ret = __splice_from_pipe(pipe, &sd, actor); - if (pipe->inode) - mutex_unlock(&pipe->inode->i_mutex); - mutex_unlock(&inode->i_mutex); + pipe_unlock(pipe); return ret; } /** - * generic_file_splice_write_nolock - generic_file_splice_write without mutexes + * generic_file_splice_write - splice data from a pipe to a file * @pipe: pipe info * @out: file to write to * @ppos: position in @out @@ -764,13 +814,12 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, * * Description: * Will either move or copy pages (determined by @flags options) from - * the given pipe inode to the given file. The caller is responsible - * for acquiring i_mutex on both inodes. + * the given pipe inode to the given file. * */ ssize_t -generic_file_splice_write_nolock(struct pipe_inode_info *pipe, struct file *out, - loff_t *ppos, size_t len, unsigned int flags) +generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags) { struct address_space *mapping = out->f_mapping; struct inode *inode = mapping->host; @@ -781,76 +830,28 @@ generic_file_splice_write_nolock(struct pipe_inode_info *pipe, struct file *out, .u.file = out, }; ssize_t ret; - int err; - - err = file_remove_suid(out); - if (unlikely(err)) - return err; - - ret = __splice_from_pipe(pipe, &sd, pipe_to_file); - if (ret > 0) { - unsigned long nr_pages; - *ppos += ret; - nr_pages = (ret + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - - /* - * If file or inode is SYNC and we actually wrote some data, - * sync it. - */ - if (unlikely((out->f_flags & O_SYNC) || IS_SYNC(inode))) { - err = generic_osync_inode(inode, mapping, - OSYNC_METADATA|OSYNC_DATA); + pipe_lock(pipe); - if (err) - ret = err; - } - balance_dirty_pages_ratelimited_nr(mapping, nr_pages); - } + splice_from_pipe_begin(&sd); + do { + ret = splice_from_pipe_next(pipe, &sd); + if (ret <= 0) + break; - return ret; -} + mutex_lock_nested(&inode->i_mutex, I_MUTEX_CHILD); + ret = file_remove_suid(out); + if (!ret) + ret = splice_from_pipe_feed(pipe, &sd, pipe_to_file); + mutex_unlock(&inode->i_mutex); + } while (ret > 0); + splice_from_pipe_end(pipe, &sd); -EXPORT_SYMBOL(generic_file_splice_write_nolock); + pipe_unlock(pipe); -/** - * generic_file_splice_write - splice data from a pipe to a file - * @pipe: pipe info - * @out: file to write to - * @ppos: position in @out - * @len: number of bytes to splice - * @flags: splice modifier flags - * - * Description: - * Will either move or copy pages (determined by @flags options) from - * the given pipe inode to the given file. - * - */ -ssize_t -generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, - loff_t *ppos, size_t len, unsigned int flags) -{ - struct address_space *mapping = out->f_mapping; - struct inode *inode = mapping->host; - struct splice_desc sd = { - .total_len = len, - .flags = flags, - .pos = *ppos, - .u.file = out, - }; - ssize_t ret; + if (sd.num_spliced) + ret = sd.num_spliced; - WARN_ON(S_ISFIFO(inode->i_mode)); - mutex_lock_nested(&inode->i_mutex, I_MUTEX_PARENT); - ret = file_remove_suid(out); - if (likely(!ret)) { - if (pipe->inode) - mutex_lock_nested(&pipe->inode->i_mutex, I_MUTEX_CHILD); - ret = __splice_from_pipe(pipe, &sd, pipe_to_file); - if (pipe->inode) - mutex_unlock(&pipe->inode->i_mutex); - } - mutex_unlock(&inode->i_mutex); if (ret > 0) { unsigned long nr_pages; @@ -1339,8 +1340,7 @@ static long vmsplice_to_user(struct file *file, const struct iovec __user *iov, if (!pipe) return -EBADF; - if (pipe->inode) - mutex_lock(&pipe->inode->i_mutex); + pipe_lock(pipe); error = ret = 0; while (nr_segs) { @@ -1395,8 +1395,7 @@ static long vmsplice_to_user(struct file *file, const struct iovec __user *iov, iov++; } - if (pipe->inode) - mutex_unlock(&pipe->inode->i_mutex); + pipe_unlock(pipe); if (!ret) ret = error; @@ -1524,7 +1523,7 @@ static int link_ipipe_prep(struct pipe_inode_info *pipe, unsigned int flags) return 0; ret = 0; - mutex_lock(&pipe->inode->i_mutex); + pipe_lock(pipe); while (!pipe->nrbufs) { if (signal_pending(current)) { @@ -1542,7 +1541,7 @@ static int link_ipipe_prep(struct pipe_inode_info *pipe, unsigned int flags) pipe_wait(pipe); } - mutex_unlock(&pipe->inode->i_mutex); + pipe_unlock(pipe); return ret; } @@ -1562,7 +1561,7 @@ static int link_opipe_prep(struct pipe_inode_info *pipe, unsigned int flags) return 0; ret = 0; - mutex_lock(&pipe->inode->i_mutex); + pipe_lock(pipe); while (pipe->nrbufs >= PIPE_BUFFERS) { if (!pipe->readers) { @@ -1583,7 +1582,7 @@ static int link_opipe_prep(struct pipe_inode_info *pipe, unsigned int flags) pipe->waiting_writers--; } - mutex_unlock(&pipe->inode->i_mutex); + pipe_unlock(pipe); return ret; } @@ -1599,10 +1598,10 @@ static int link_pipe(struct pipe_inode_info *ipipe, /* * Potential ABBA deadlock, work around it by ordering lock - * grabbing by inode address. Otherwise two different processes + * grabbing by pipe info address. Otherwise two different processes * could deadlock (one doing tee from A -> B, the other from B -> A). */ - inode_double_lock(ipipe->inode, opipe->inode); + pipe_double_lock(ipipe, opipe); do { if (!opipe->readers) { @@ -1653,7 +1652,8 @@ static int link_pipe(struct pipe_inode_info *ipipe, if (!ret && ipipe->waiting_writers && (flags & SPLICE_F_NONBLOCK)) ret = -EAGAIN; - inode_double_unlock(ipipe->inode, opipe->inode); + pipe_unlock(ipipe); + pipe_unlock(opipe); /* * If we put data in the output pipe, wakeup any potential readers. diff --git a/fs/stat.c b/fs/stat.c index 2db740a0cfb5..075694e31d8b 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -55,59 +55,54 @@ int vfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) EXPORT_SYMBOL(vfs_getattr); -int vfs_stat_fd(int dfd, char __user *name, struct kstat *stat) +int vfs_fstat(unsigned int fd, struct kstat *stat) { - struct path path; - int error; + struct file *f = fget(fd); + int error = -EBADF; - error = user_path_at(dfd, name, LOOKUP_FOLLOW, &path); - if (!error) { - error = vfs_getattr(path.mnt, path.dentry, stat); - path_put(&path); + if (f) { + error = vfs_getattr(f->f_path.mnt, f->f_path.dentry, stat); + fput(f); } return error; } +EXPORT_SYMBOL(vfs_fstat); -int vfs_stat(char __user *name, struct kstat *stat) +int vfs_fstatat(int dfd, char __user *filename, struct kstat *stat, int flag) { - return vfs_stat_fd(AT_FDCWD, name, stat); -} + struct path path; + int error = -EINVAL; + int lookup_flags = 0; -EXPORT_SYMBOL(vfs_stat); + if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) + goto out; -int vfs_lstat_fd(int dfd, char __user *name, struct kstat *stat) -{ - struct path path; - int error; + if (!(flag & AT_SYMLINK_NOFOLLOW)) + lookup_flags |= LOOKUP_FOLLOW; - error = user_path_at(dfd, name, 0, &path); - if (!error) { - error = vfs_getattr(path.mnt, path.dentry, stat); - path_put(&path); - } + error = user_path_at(dfd, filename, lookup_flags, &path); + if (error) + goto out; + + error = vfs_getattr(path.mnt, path.dentry, stat); + path_put(&path); +out: return error; } +EXPORT_SYMBOL(vfs_fstatat); -int vfs_lstat(char __user *name, struct kstat *stat) +int vfs_stat(char __user *name, struct kstat *stat) { - return vfs_lstat_fd(AT_FDCWD, name, stat); + return vfs_fstatat(AT_FDCWD, name, stat, 0); } +EXPORT_SYMBOL(vfs_stat); -EXPORT_SYMBOL(vfs_lstat); - -int vfs_fstat(unsigned int fd, struct kstat *stat) +int vfs_lstat(char __user *name, struct kstat *stat) { - struct file *f = fget(fd); - int error = -EBADF; - - if (f) { - error = vfs_getattr(f->f_path.mnt, f->f_path.dentry, stat); - fput(f); - } - return error; + return vfs_fstatat(AT_FDCWD, name, stat, AT_SYMLINK_NOFOLLOW); } +EXPORT_SYMBOL(vfs_lstat); -EXPORT_SYMBOL(vfs_fstat); #ifdef __ARCH_WANT_OLD_STAT @@ -155,23 +150,25 @@ static int cp_old_stat(struct kstat *stat, struct __old_kernel_stat __user * sta SYSCALL_DEFINE2(stat, char __user *, filename, struct __old_kernel_stat __user *, statbuf) { struct kstat stat; - int error = vfs_stat_fd(AT_FDCWD, filename, &stat); + int error; - if (!error) - error = cp_old_stat(&stat, statbuf); + error = vfs_stat(filename, &stat); + if (error) + return error; - return error; + return cp_old_stat(&stat, statbuf); } SYSCALL_DEFINE2(lstat, char __user *, filename, struct __old_kernel_stat __user *, statbuf) { struct kstat stat; - int error = vfs_lstat_fd(AT_FDCWD, filename, &stat); + int error; - if (!error) - error = cp_old_stat(&stat, statbuf); + error = vfs_lstat(filename, &stat); + if (error) + return error; - return error; + return cp_old_stat(&stat, statbuf); } SYSCALL_DEFINE2(fstat, unsigned int, fd, struct __old_kernel_stat __user *, statbuf) @@ -240,23 +237,23 @@ static int cp_new_stat(struct kstat *stat, struct stat __user *statbuf) SYSCALL_DEFINE2(newstat, char __user *, filename, struct stat __user *, statbuf) { struct kstat stat; - int error = vfs_stat_fd(AT_FDCWD, filename, &stat); - - if (!error) - error = cp_new_stat(&stat, statbuf); + int error = vfs_stat(filename, &stat); - return error; + if (error) + return error; + return cp_new_stat(&stat, statbuf); } SYSCALL_DEFINE2(newlstat, char __user *, filename, struct stat __user *, statbuf) { struct kstat stat; - int error = vfs_lstat_fd(AT_FDCWD, filename, &stat); + int error; - if (!error) - error = cp_new_stat(&stat, statbuf); + error = vfs_lstat(filename, &stat); + if (error) + return error; - return error; + return cp_new_stat(&stat, statbuf); } #if !defined(__ARCH_WANT_STAT64) || defined(__ARCH_WANT_SYS_NEWFSTATAT) @@ -264,21 +261,12 @@ SYSCALL_DEFINE4(newfstatat, int, dfd, char __user *, filename, struct stat __user *, statbuf, int, flag) { struct kstat stat; - int error = -EINVAL; - - if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) - goto out; - - if (flag & AT_SYMLINK_NOFOLLOW) - error = vfs_lstat_fd(dfd, filename, &stat); - else - error = vfs_stat_fd(dfd, filename, &stat); - - if (!error) - error = cp_new_stat(&stat, statbuf); + int error; -out: - return error; + error = vfs_fstatat(dfd, filename, &stat, flag); + if (error) + return error; + return cp_new_stat(&stat, statbuf); } #endif @@ -404,21 +392,12 @@ SYSCALL_DEFINE4(fstatat64, int, dfd, char __user *, filename, struct stat64 __user *, statbuf, int, flag) { struct kstat stat; - int error = -EINVAL; - - if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) - goto out; - - if (flag & AT_SYMLINK_NOFOLLOW) - error = vfs_lstat_fd(dfd, filename, &stat); - else - error = vfs_stat_fd(dfd, filename, &stat); - - if (!error) - error = cp_new_stat64(&stat, statbuf); + int error; -out: - return error; + error = vfs_fstatat(dfd, filename, &stat, flag); + if (error) + return error; + return cp_new_stat64(&stat, statbuf); } #endif /* __ARCH_WANT_STAT64 */ diff --git a/fs/sysfs/bin.c b/fs/sysfs/bin.c index 93e0c0281d45..9345806c8853 100644 --- a/fs/sysfs/bin.c +++ b/fs/sysfs/bin.c @@ -157,14 +157,9 @@ static ssize_t write(struct file *file, const char __user *userbuf, count = size - offs; } - temp = kmalloc(count, GFP_KERNEL); - if (!temp) - return -ENOMEM; - - if (copy_from_user(temp, userbuf, count)) { - count = -EFAULT; - goto out_free; - } + temp = memdup_user(userbuf, count); + if (IS_ERR(temp)) + return PTR_ERR(temp); mutex_lock(&bb->mutex); @@ -176,8 +171,6 @@ static ssize_t write(struct file *file, const char __user *userbuf, if (count > 0) *off = offs + count; -out_free: - kfree(temp); return count; } diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 289c43a47263..b1606e07b7a3 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -446,11 +446,11 @@ static unsigned int sysfs_poll(struct file *filp, poll_table *wait) if (buffer->event != atomic_read(&od->event)) goto trigger; - return 0; + return DEFAULT_POLLMASK; trigger: buffer->needs_read_fill = 1; - return POLLERR|POLLPRI; + return DEFAULT_POLLMASK|POLLERR|POLLPRI; } void sysfs_notify_dirent(struct sysfs_dirent *sd) @@ -667,6 +667,7 @@ struct sysfs_schedule_callback_struct { struct work_struct work; }; +static struct workqueue_struct *sysfs_workqueue; static DEFINE_MUTEX(sysfs_workq_mutex); static LIST_HEAD(sysfs_workq); static void sysfs_schedule_callback_work(struct work_struct *work) @@ -715,11 +716,20 @@ int sysfs_schedule_callback(struct kobject *kobj, void (*func)(void *), mutex_lock(&sysfs_workq_mutex); list_for_each_entry_safe(ss, tmp, &sysfs_workq, workq_list) if (ss->kobj == kobj) { + module_put(owner); mutex_unlock(&sysfs_workq_mutex); return -EAGAIN; } mutex_unlock(&sysfs_workq_mutex); + if (sysfs_workqueue == NULL) { + sysfs_workqueue = create_workqueue("sysfsd"); + if (sysfs_workqueue == NULL) { + module_put(owner); + return -ENOMEM; + } + } + ss = kmalloc(sizeof(*ss), GFP_KERNEL); if (!ss) { module_put(owner); @@ -735,7 +745,7 @@ int sysfs_schedule_callback(struct kobject *kobj, void (*func)(void *), mutex_lock(&sysfs_workq_mutex); list_add_tail(&ss->workq_list, &sysfs_workq); mutex_unlock(&sysfs_workq_mutex); - schedule_work(&ss->work); + queue_work(sysfs_workqueue, &ss->work); return 0; } EXPORT_SYMBOL_GPL(sysfs_schedule_callback); diff --git a/fs/xattr.c b/fs/xattr.c index 197c4fcac032..d51b8f9db921 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -237,13 +237,9 @@ setxattr(struct dentry *d, const char __user *name, const void __user *value, if (size) { if (size > XATTR_SIZE_MAX) return -E2BIG; - kvalue = kmalloc(size, GFP_KERNEL); - if (!kvalue) - return -ENOMEM; - if (copy_from_user(kvalue, value, size)) { - kfree(kvalue); - return -EFAULT; - } + kvalue = memdup_user(value, size); + if (IS_ERR(kvalue)) + return PTR_ERR(kvalue); } error = vfs_setxattr(d, kname, kvalue, size, flags); diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index c13f67300fe7..7ec89fc05b2b 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -153,23 +153,6 @@ xfs_find_bdev_for_inode( } /* - * Schedule IO completion handling on a xfsdatad if this was - * the final hold on this ioend. If we are asked to wait, - * flush the workqueue. - */ -STATIC void -xfs_finish_ioend( - xfs_ioend_t *ioend, - int wait) -{ - if (atomic_dec_and_test(&ioend->io_remaining)) { - queue_work(xfsdatad_workqueue, &ioend->io_work); - if (wait) - flush_workqueue(xfsdatad_workqueue); - } -} - -/* * We're now finished for good with this ioend structure. * Update the page state via the associated buffer_heads, * release holds on the inode and bio, and finally free @@ -310,6 +293,27 @@ xfs_end_bio_read( } /* + * Schedule IO completion handling on a xfsdatad if this was + * the final hold on this ioend. If we are asked to wait, + * flush the workqueue. + */ +STATIC void +xfs_finish_ioend( + xfs_ioend_t *ioend, + int wait) +{ + if (atomic_dec_and_test(&ioend->io_remaining)) { + struct workqueue_struct *wq = xfsdatad_workqueue; + if (ioend->io_work.func == xfs_end_bio_unwritten) + wq = xfsconvertd_workqueue; + + queue_work(wq, &ioend->io_work); + if (wait) + flush_workqueue(wq); + } +} + +/* * Allocate and initialise an IO completion structure. * We need to track unwritten extent write completion here initially. * We'll need to extend this for updating the ondisk inode size later diff --git a/fs/xfs/linux-2.6/xfs_aops.h b/fs/xfs/linux-2.6/xfs_aops.h index 1dd528849755..221b3e66ceef 100644 --- a/fs/xfs/linux-2.6/xfs_aops.h +++ b/fs/xfs/linux-2.6/xfs_aops.h @@ -19,6 +19,7 @@ #define __XFS_AOPS_H__ extern struct workqueue_struct *xfsdatad_workqueue; +extern struct workqueue_struct *xfsconvertd_workqueue; extern mempool_t *xfs_ioend_pool; /* diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index aa1016bb9134..e28800a9f2b5 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -51,6 +51,7 @@ static struct shrinker xfs_buf_shake = { static struct workqueue_struct *xfslogd_workqueue; struct workqueue_struct *xfsdatad_workqueue; +struct workqueue_struct *xfsconvertd_workqueue; #ifdef XFS_BUF_TRACE void @@ -1775,6 +1776,7 @@ xfs_flush_buftarg( xfs_buf_t *bp, *n; int pincount = 0; + xfs_buf_runall_queues(xfsconvertd_workqueue); xfs_buf_runall_queues(xfsdatad_workqueue); xfs_buf_runall_queues(xfslogd_workqueue); @@ -1831,9 +1833,15 @@ xfs_buf_init(void) if (!xfsdatad_workqueue) goto out_destroy_xfslogd_workqueue; + xfsconvertd_workqueue = create_workqueue("xfsconvertd"); + if (!xfsconvertd_workqueue) + goto out_destroy_xfsdatad_workqueue; + register_shrinker(&xfs_buf_shake); return 0; + out_destroy_xfsdatad_workqueue: + destroy_workqueue(xfsdatad_workqueue); out_destroy_xfslogd_workqueue: destroy_workqueue(xfslogd_workqueue); out_free_buf_zone: @@ -1849,6 +1857,7 @@ void xfs_buf_terminate(void) { unregister_shrinker(&xfs_buf_shake); + destroy_workqueue(xfsconvertd_workqueue); destroy_workqueue(xfsdatad_workqueue); destroy_workqueue(xfslogd_workqueue); kmem_zone_destroy(xfs_buf_zone); diff --git a/fs/xfs/linux-2.6/xfs_fs_subr.c b/fs/xfs/linux-2.6/xfs_fs_subr.c index 5aeb77776961..08be36d7326c 100644 --- a/fs/xfs/linux-2.6/xfs_fs_subr.c +++ b/fs/xfs/linux-2.6/xfs_fs_subr.c @@ -74,14 +74,14 @@ xfs_flush_pages( if (mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) { xfs_iflags_clear(ip, XFS_ITRUNCATED); - ret = filemap_fdatawrite(mapping); - if (flags & XFS_B_ASYNC) - return -ret; - ret2 = filemap_fdatawait(mapping); - if (!ret) - ret = ret2; + ret = -filemap_fdatawrite(mapping); } - return -ret; + if (flags & XFS_B_ASYNC) + return ret; + ret2 = xfs_wait_on_pages(ip, first, last); + if (!ret) + ret = ret2; + return ret; } int diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index d0b499418a7d..34eaab608e6e 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -489,17 +489,12 @@ xfs_attrmulti_attr_set( if (len > XATTR_SIZE_MAX) return EINVAL; - kbuf = kmalloc(len, GFP_KERNEL); - if (!kbuf) - return ENOMEM; - - if (copy_from_user(kbuf, ubuf, len)) - goto out_kfree; + kbuf = memdup_user(ubuf, len); + if (IS_ERR(kbuf)) + return PTR_ERR(kbuf); error = xfs_attr_set(XFS_I(inode), name, kbuf, len, flags); - out_kfree: - kfree(kbuf); return error; } @@ -540,20 +535,16 @@ xfs_attrmulti_by_handle( if (!size || size > 16 * PAGE_SIZE) goto out_dput; - error = ENOMEM; - ops = kmalloc(size, GFP_KERNEL); - if (!ops) + ops = memdup_user(am_hreq.ops, size); + if (IS_ERR(ops)) { + error = PTR_ERR(ops); goto out_dput; - - error = EFAULT; - if (copy_from_user(ops, am_hreq.ops, size)) - goto out_kfree_ops; + } attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL); if (!attr_name) goto out_kfree_ops; - error = 0; for (i = 0; i < am_hreq.opcount; i++) { ops[i].am_error = strncpy_from_user(attr_name, diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.c b/fs/xfs/linux-2.6/xfs_ioctl32.c index c70c4e3db790..0882d166239a 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl32.c +++ b/fs/xfs/linux-2.6/xfs_ioctl32.c @@ -427,20 +427,16 @@ xfs_compat_attrmulti_by_handle( if (!size || size > 16 * PAGE_SIZE) goto out_dput; - error = ENOMEM; - ops = kmalloc(size, GFP_KERNEL); - if (!ops) + ops = memdup_user(compat_ptr(am_hreq.ops), size); + if (IS_ERR(ops)) { + error = PTR_ERR(ops); goto out_dput; - - error = EFAULT; - if (copy_from_user(ops, compat_ptr(am_hreq.ops), size)) - goto out_kfree_ops; + } attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL); if (!attr_name) goto out_kfree_ops; - error = 0; for (i = 0; i < am_hreq.opcount; i++) { ops[i].am_error = strncpy_from_user(attr_name, diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c index 7e90daa0d1d1..9142192ccbe6 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.c +++ b/fs/xfs/linux-2.6/xfs_lrw.c @@ -751,10 +751,26 @@ start: goto relock; } } else { + int enospc = 0; + ssize_t ret2 = 0; + +write_retry: xfs_rw_enter_trace(XFS_WRITE_ENTER, xip, (void *)iovp, segs, *offset, ioflags); - ret = generic_file_buffered_write(iocb, iovp, segs, + ret2 = generic_file_buffered_write(iocb, iovp, segs, pos, offset, count, ret); + /* + * if we just got an ENOSPC, flush the inode now we + * aren't holding any page locks and retry *once* + */ + if (ret2 == -ENOSPC && !enospc) { + error = xfs_flush_pages(xip, 0, -1, 0, FI_NONE); + if (error) + goto out_unlock_internal; + enospc = 1; + goto write_retry; + } + ret = ret2; } current->backing_dev_info = NULL; diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index a608e72fa405..f7ba76633c29 100644 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ b/fs/xfs/linux-2.6/xfs_sync.c @@ -62,12 +62,6 @@ xfs_sync_inodes_ag( uint32_t first_index = 0; int error = 0; int last_error = 0; - int fflag = XFS_B_ASYNC; - - if (flags & SYNC_DELWRI) - fflag = XFS_B_DELWRI; - if (flags & SYNC_WAIT) - fflag = 0; /* synchronous overrides all */ do { struct inode *inode; @@ -128,11 +122,23 @@ xfs_sync_inodes_ag( * If we have to flush data or wait for I/O completion * we need to hold the iolock. */ - if ((flags & SYNC_DELWRI) && VN_DIRTY(inode)) { - xfs_ilock(ip, XFS_IOLOCK_SHARED); - lock_flags |= XFS_IOLOCK_SHARED; - error = xfs_flush_pages(ip, 0, -1, fflag, FI_NONE); - if (flags & SYNC_IOWAIT) + if (flags & SYNC_DELWRI) { + if (VN_DIRTY(inode)) { + if (flags & SYNC_TRYLOCK) { + if (xfs_ilock_nowait(ip, XFS_IOLOCK_SHARED)) + lock_flags |= XFS_IOLOCK_SHARED; + } else { + xfs_ilock(ip, XFS_IOLOCK_SHARED); + lock_flags |= XFS_IOLOCK_SHARED; + } + if (lock_flags & XFS_IOLOCK_SHARED) { + error = xfs_flush_pages(ip, 0, -1, + (flags & SYNC_WAIT) ? 0 + : XFS_B_ASYNC, + FI_NONE); + } + } + if (VN_CACHED(inode) && (flags & SYNC_IOWAIT)) xfs_ioend_wait(ip); } xfs_ilock(ip, XFS_ILOCK_SHARED); @@ -398,15 +404,17 @@ STATIC void xfs_syncd_queue_work( struct xfs_mount *mp, void *data, - void (*syncer)(struct xfs_mount *, void *)) + void (*syncer)(struct xfs_mount *, void *), + struct completion *completion) { - struct bhv_vfs_sync_work *work; + struct xfs_sync_work *work; - work = kmem_alloc(sizeof(struct bhv_vfs_sync_work), KM_SLEEP); + work = kmem_alloc(sizeof(struct xfs_sync_work), KM_SLEEP); INIT_LIST_HEAD(&work->w_list); work->w_syncer = syncer; work->w_data = data; work->w_mount = mp; + work->w_completion = completion; spin_lock(&mp->m_sync_lock); list_add_tail(&work->w_list, &mp->m_sync_list); spin_unlock(&mp->m_sync_lock); @@ -420,49 +428,26 @@ xfs_syncd_queue_work( * heads, looking about for more room... */ STATIC void -xfs_flush_inode_work( - struct xfs_mount *mp, - void *arg) -{ - struct inode *inode = arg; - filemap_flush(inode->i_mapping); - iput(inode); -} - -void -xfs_flush_inode( - xfs_inode_t *ip) -{ - struct inode *inode = VFS_I(ip); - - igrab(inode); - xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_inode_work); - delay(msecs_to_jiffies(500)); -} - -/* - * This is the "bigger hammer" version of xfs_flush_inode_work... - * (IOW, "If at first you don't succeed, use a Bigger Hammer"). - */ -STATIC void -xfs_flush_device_work( +xfs_flush_inodes_work( struct xfs_mount *mp, void *arg) { struct inode *inode = arg; - sync_blockdev(mp->m_super->s_bdev); + xfs_sync_inodes(mp, SYNC_DELWRI | SYNC_TRYLOCK); + xfs_sync_inodes(mp, SYNC_DELWRI | SYNC_TRYLOCK | SYNC_IOWAIT); iput(inode); } void -xfs_flush_device( +xfs_flush_inodes( xfs_inode_t *ip) { struct inode *inode = VFS_I(ip); + DECLARE_COMPLETION_ONSTACK(completion); igrab(inode); - xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_device_work); - delay(msecs_to_jiffies(500)); + xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_inodes_work, &completion); + wait_for_completion(&completion); xfs_log_force(ip->i_mount, (xfs_lsn_t)0, XFS_LOG_FORCE|XFS_LOG_SYNC); } @@ -497,7 +482,7 @@ xfssyncd( { struct xfs_mount *mp = arg; long timeleft; - bhv_vfs_sync_work_t *work, *n; + xfs_sync_work_t *work, *n; LIST_HEAD (tmp); set_freezable(); @@ -532,6 +517,8 @@ xfssyncd( list_del(&work->w_list); if (work == &mp->m_sync_work) continue; + if (work->w_completion) + complete(work->w_completion); kmem_free(work); } } @@ -545,6 +532,7 @@ xfs_syncd_init( { mp->m_sync_work.w_syncer = xfs_sync_worker; mp->m_sync_work.w_mount = mp; + mp->m_sync_work.w_completion = NULL; mp->m_sync_task = kthread_run(xfssyncd, mp, "xfssyncd"); if (IS_ERR(mp->m_sync_task)) return -PTR_ERR(mp->m_sync_task); diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h index 04f058c848ae..308d5bf6dfbd 100644 --- a/fs/xfs/linux-2.6/xfs_sync.h +++ b/fs/xfs/linux-2.6/xfs_sync.h @@ -21,18 +21,20 @@ struct xfs_mount; struct xfs_perag; -typedef struct bhv_vfs_sync_work { +typedef struct xfs_sync_work { struct list_head w_list; struct xfs_mount *w_mount; void *w_data; /* syncer routine argument */ void (*w_syncer)(struct xfs_mount *, void *); -} bhv_vfs_sync_work_t; + struct completion *w_completion; +} xfs_sync_work_t; #define SYNC_ATTR 0x0001 /* sync attributes */ #define SYNC_DELWRI 0x0002 /* look at delayed writes */ #define SYNC_WAIT 0x0004 /* wait for i/o to complete */ #define SYNC_BDFLUSH 0x0008 /* BDFLUSH is calling -- don't block */ #define SYNC_IOWAIT 0x0010 /* wait for all I/O to complete */ +#define SYNC_TRYLOCK 0x0020 /* only try to lock inodes */ int xfs_syncd_init(struct xfs_mount *mp); void xfs_syncd_stop(struct xfs_mount *mp); @@ -43,8 +45,7 @@ int xfs_sync_fsdata(struct xfs_mount *mp, int flags); int xfs_quiesce_data(struct xfs_mount *mp); void xfs_quiesce_attr(struct xfs_mount *mp); -void xfs_flush_inode(struct xfs_inode *ip); -void xfs_flush_device(struct xfs_inode *ip); +void xfs_flush_inodes(struct xfs_inode *ip); int xfs_reclaim_inode(struct xfs_inode *ip, int locked, int sync_mode); int xfs_reclaim_inodes(struct xfs_mount *mp, int noblock, int mode); diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index 478e587087fe..89b81eedce6a 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -69,15 +69,6 @@ xfs_inode_alloc( ASSERT(!spin_is_locked(&ip->i_flags_lock)); ASSERT(completion_done(&ip->i_flush)); - /* - * initialise the VFS inode here to get failures - * out of the way early. - */ - if (!inode_init_always(mp->m_super, VFS_I(ip))) { - kmem_zone_free(xfs_inode_zone, ip); - return NULL; - } - /* initialise the xfs inode */ ip->i_ino = ino; ip->i_mount = mp; @@ -113,6 +104,20 @@ xfs_inode_alloc( #ifdef XFS_DIR2_TRACE ip->i_dir_trace = ktrace_alloc(XFS_DIR2_KTRACE_SIZE, KM_NOFS); #endif + /* + * Now initialise the VFS inode. We do this after the xfs_inode + * initialisation as internal failures will result in ->destroy_inode + * being called and that will pass down through the reclaim path and + * free the XFS inode. This path requires the XFS inode to already be + * initialised. Hence if this call fails, the xfs_inode has already + * been freed and we should not reference it at all in the error + * handling. + */ + if (!inode_init_always(mp->m_super, VFS_I(ip))) + return NULL; + + /* prevent anyone from using this yet */ + VFS_I(ip)->i_state = I_NEW|I_LOCK; return ip; } diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 08ce72316bfe..5aaa2d7ec155 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -338,38 +338,6 @@ xfs_iomap_eof_align_last_fsb( } STATIC int -xfs_flush_space( - xfs_inode_t *ip, - int *fsynced, - int *ioflags) -{ - switch (*fsynced) { - case 0: - if (ip->i_delayed_blks) { - xfs_iunlock(ip, XFS_ILOCK_EXCL); - xfs_flush_inode(ip); - xfs_ilock(ip, XFS_ILOCK_EXCL); - *fsynced = 1; - } else { - *ioflags |= BMAPI_SYNC; - *fsynced = 2; - } - return 0; - case 1: - *fsynced = 2; - *ioflags |= BMAPI_SYNC; - return 0; - case 2: - xfs_iunlock(ip, XFS_ILOCK_EXCL); - xfs_flush_device(ip); - xfs_ilock(ip, XFS_ILOCK_EXCL); - *fsynced = 3; - return 0; - } - return 1; -} - -STATIC int xfs_cmn_err_fsblock_zero( xfs_inode_t *ip, xfs_bmbt_irec_t *imap) @@ -538,15 +506,9 @@ error_out: } /* - * If the caller is doing a write at the end of the file, - * then extend the allocation out to the file system's write - * iosize. We clean up any extra space left over when the - * file is closed in xfs_inactive(). - * - * For sync writes, we are flushing delayed allocate space to - * try to make additional space available for allocation near - * the filesystem full boundary - preallocation hurts in that - * situation, of course. + * If the caller is doing a write at the end of the file, then extend the + * allocation out to the file system's write iosize. We clean up any extra + * space left over when the file is closed in xfs_inactive(). */ STATIC int xfs_iomap_eof_want_preallocate( @@ -565,7 +527,7 @@ xfs_iomap_eof_want_preallocate( int n, error, imaps; *prealloc = 0; - if ((ioflag & BMAPI_SYNC) || (offset + count) <= ip->i_size) + if ((offset + count) <= ip->i_size) return 0; /* @@ -611,7 +573,7 @@ xfs_iomap_write_delay( xfs_extlen_t extsz; int nimaps; xfs_bmbt_irec_t imap[XFS_WRITE_IMAPS]; - int prealloc, fsynced = 0; + int prealloc, flushed = 0; int error; ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); @@ -627,12 +589,12 @@ xfs_iomap_write_delay( extsz = xfs_get_extsz_hint(ip); offset_fsb = XFS_B_TO_FSBT(mp, offset); -retry: error = xfs_iomap_eof_want_preallocate(mp, ip, offset, count, ioflag, imap, XFS_WRITE_IMAPS, &prealloc); if (error) return error; +retry: if (prealloc) { aligned_offset = XFS_WRITEIO_ALIGN(mp, (offset + count - 1)); ioalign = XFS_B_TO_FSBT(mp, aligned_offset); @@ -659,15 +621,22 @@ retry: /* * If bmapi returned us nothing, and if we didn't get back EDQUOT, - * then we must have run out of space - flush delalloc, and retry.. + * then we must have run out of space - flush all other inodes with + * delalloc blocks and retry without EOF preallocation. */ if (nimaps == 0) { xfs_iomap_enter_trace(XFS_IOMAP_WRITE_NOSPACE, ip, offset, count); - if (xfs_flush_space(ip, &fsynced, &ioflag)) + if (flushed) return XFS_ERROR(ENOSPC); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + xfs_flush_inodes(ip); + xfs_ilock(ip, XFS_ILOCK_EXCL); + + flushed = 1; error = 0; + prealloc = 0; goto retry; } diff --git a/fs/xfs/xfs_iomap.h b/fs/xfs/xfs_iomap.h index a1cc1322fc0f..fdcf7b82747f 100644 --- a/fs/xfs/xfs_iomap.h +++ b/fs/xfs/xfs_iomap.h @@ -40,8 +40,7 @@ typedef enum { BMAPI_IGNSTATE = (1 << 4), /* ignore unwritten state on read */ BMAPI_DIRECT = (1 << 5), /* direct instead of buffered write */ BMAPI_MMAP = (1 << 6), /* allocate for mmap write */ - BMAPI_SYNC = (1 << 7), /* sync write to flush delalloc space */ - BMAPI_TRYLOCK = (1 << 8), /* non-blocking request */ + BMAPI_TRYLOCK = (1 << 7), /* non-blocking request */ } bmapi_flags_t; diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index f76c6d7cea21..3750f04ede0b 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -562,9 +562,8 @@ xfs_log_mount( } mp->m_log = xlog_alloc_log(mp, log_target, blk_offset, num_bblks); - if (!mp->m_log) { - cmn_err(CE_WARN, "XFS: Log allocation failed: No memory!"); - error = ENOMEM; + if (IS_ERR(mp->m_log)) { + error = -PTR_ERR(mp->m_log); goto out; } @@ -1180,10 +1179,13 @@ xlog_alloc_log(xfs_mount_t *mp, xfs_buf_t *bp; int i; int iclogsize; + int error = ENOMEM; log = kmem_zalloc(sizeof(xlog_t), KM_MAYFAIL); - if (!log) - return NULL; + if (!log) { + xlog_warn("XFS: Log allocation failed: No memory!"); + goto out; + } log->l_mp = mp; log->l_targ = log_target; @@ -1201,19 +1203,35 @@ xlog_alloc_log(xfs_mount_t *mp, log->l_grant_reserve_cycle = 1; log->l_grant_write_cycle = 1; + error = EFSCORRUPTED; if (xfs_sb_version_hassector(&mp->m_sb)) { log->l_sectbb_log = mp->m_sb.sb_logsectlog - BBSHIFT; - ASSERT(log->l_sectbb_log <= mp->m_sectbb_log); + if (log->l_sectbb_log < 0 || + log->l_sectbb_log > mp->m_sectbb_log) { + xlog_warn("XFS: Log sector size (0x%x) out of range.", + log->l_sectbb_log); + goto out_free_log; + } + /* for larger sector sizes, must have v2 or external log */ - ASSERT(log->l_sectbb_log == 0 || - log->l_logBBstart == 0 || - xfs_sb_version_haslogv2(&mp->m_sb)); - ASSERT(mp->m_sb.sb_logsectlog >= BBSHIFT); + if (log->l_sectbb_log != 0 && + (log->l_logBBstart != 0 && + !xfs_sb_version_haslogv2(&mp->m_sb))) { + xlog_warn("XFS: log sector size (0x%x) invalid " + "for configuration.", log->l_sectbb_log); + goto out_free_log; + } + if (mp->m_sb.sb_logsectlog < BBSHIFT) { + xlog_warn("XFS: Log sector log (0x%x) too small.", + mp->m_sb.sb_logsectlog); + goto out_free_log; + } } log->l_sectbb_mask = (1 << log->l_sectbb_log) - 1; xlog_get_iclog_buffer_size(mp, log); + error = ENOMEM; bp = xfs_buf_get_empty(log->l_iclog_size, mp->m_logdev_targp); if (!bp) goto out_free_log; @@ -1313,7 +1331,8 @@ out_free_iclog: xfs_buf_free(log->l_xbuf); out_free_log: kmem_free(log); - return NULL; +out: + return ERR_PTR(-error); } /* xlog_alloc_log */ @@ -2541,18 +2560,19 @@ redo: xlog_ins_ticketq(&log->l_reserve_headq, tic); xlog_trace_loggrant(log, tic, "xlog_grant_log_space: sleep 2"); + spin_unlock(&log->l_grant_lock); + xlog_grant_push_ail(log->l_mp, need_bytes); + spin_lock(&log->l_grant_lock); + XFS_STATS_INC(xs_sleep_logspace); sv_wait(&tic->t_wait, PINOD|PLTWAIT, &log->l_grant_lock, s); - if (XLOG_FORCED_SHUTDOWN(log)) { - spin_lock(&log->l_grant_lock); + spin_lock(&log->l_grant_lock); + if (XLOG_FORCED_SHUTDOWN(log)) goto error_return; - } xlog_trace_loggrant(log, tic, "xlog_grant_log_space: wake 2"); - xlog_grant_push_ail(log->l_mp, need_bytes); - spin_lock(&log->l_grant_lock); goto redo; } else if (tic->t_flags & XLOG_TIC_IN_Q) xlog_del_ticketq(&log->l_reserve_headq, tic); @@ -2631,7 +2651,7 @@ xlog_regrant_write_log_space(xlog_t *log, * for more free space, otherwise try to get some space for * this transaction. */ - + need_bytes = tic->t_unit_res; if ((ntic = log->l_write_headq)) { free_bytes = xlog_space_left(log, log->l_grant_write_cycle, log->l_grant_write_bytes); @@ -2651,26 +2671,25 @@ xlog_regrant_write_log_space(xlog_t *log, xlog_trace_loggrant(log, tic, "xlog_regrant_write_log_space: sleep 1"); + spin_unlock(&log->l_grant_lock); + xlog_grant_push_ail(log->l_mp, need_bytes); + spin_lock(&log->l_grant_lock); + XFS_STATS_INC(xs_sleep_logspace); sv_wait(&tic->t_wait, PINOD|PLTWAIT, &log->l_grant_lock, s); /* If we're shutting down, this tic is already * off the queue */ - if (XLOG_FORCED_SHUTDOWN(log)) { - spin_lock(&log->l_grant_lock); + spin_lock(&log->l_grant_lock); + if (XLOG_FORCED_SHUTDOWN(log)) goto error_return; - } xlog_trace_loggrant(log, tic, "xlog_regrant_write_log_space: wake 1"); - xlog_grant_push_ail(log->l_mp, tic->t_unit_res); - spin_lock(&log->l_grant_lock); } } - need_bytes = tic->t_unit_res; - redo: if (XLOG_FORCED_SHUTDOWN(log)) goto error_return; @@ -2680,19 +2699,20 @@ redo: if (free_bytes < need_bytes) { if ((tic->t_flags & XLOG_TIC_IN_Q) == 0) xlog_ins_ticketq(&log->l_write_headq, tic); + spin_unlock(&log->l_grant_lock); + xlog_grant_push_ail(log->l_mp, need_bytes); + spin_lock(&log->l_grant_lock); + XFS_STATS_INC(xs_sleep_logspace); sv_wait(&tic->t_wait, PINOD|PLTWAIT, &log->l_grant_lock, s); /* If we're shutting down, this tic is already off the queue */ - if (XLOG_FORCED_SHUTDOWN(log)) { - spin_lock(&log->l_grant_lock); + spin_lock(&log->l_grant_lock); + if (XLOG_FORCED_SHUTDOWN(log)) goto error_return; - } xlog_trace_loggrant(log, tic, "xlog_regrant_write_log_space: wake 2"); - xlog_grant_push_ail(log->l_mp, need_bytes); - spin_lock(&log->l_grant_lock); goto redo; } else if (tic->t_flags & XLOG_TIC_IN_Q) xlog_del_ticketq(&log->l_write_headq, tic); diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 7af44adffc8f..d6a64392f983 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -313,7 +313,7 @@ typedef struct xfs_mount { #endif struct xfs_mru_cache *m_filestream; /* per-mount filestream data */ struct task_struct *m_sync_task; /* generalised sync thread */ - bhv_vfs_sync_work_t m_sync_work; /* work item for VFS_SYNC */ + xfs_sync_work_t m_sync_work; /* work item for VFS_SYNC */ struct list_head m_sync_list; /* sync thread work item list */ spinlock_t m_sync_lock; /* work item list lock */ int m_sync_seq; /* sync thread generation no. */ diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 7394c7af5de5..19cf90a9c762 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -1457,6 +1457,13 @@ xfs_create( error = xfs_trans_reserve(tp, resblks, log_res, 0, XFS_TRANS_PERM_LOG_RES, log_count); if (error == ENOSPC) { + /* flush outstanding delalloc blocks and retry */ + xfs_flush_inodes(dp); + error = xfs_trans_reserve(tp, resblks, XFS_CREATE_LOG_RES(mp), 0, + XFS_TRANS_PERM_LOG_RES, XFS_CREATE_LOG_COUNT); + } + if (error == ENOSPC) { + /* No space at all so try a "no-allocation" reservation */ resblks = 0; error = xfs_trans_reserve(tp, 0, log_res, 0, XFS_TRANS_PERM_LOG_RES, log_count); |