diff options
Diffstat (limited to 'fs')
67 files changed, 1196 insertions, 657 deletions
@@ -141,6 +141,7 @@ struct kioctx { struct { unsigned tail; + unsigned completed_events; spinlock_t completion_lock; } ____cacheline_aligned_in_smp; @@ -792,6 +793,8 @@ void exit_aio(struct mm_struct *mm) for (i = 0; i < table->nr; ++i) { struct kioctx *ctx = table->table[i]; + struct completion requests_done = + COMPLETION_INITIALIZER_ONSTACK(requests_done); if (!ctx) continue; @@ -803,7 +806,10 @@ void exit_aio(struct mm_struct *mm) * that it needs to unmap the area, just set it to 0. */ ctx->mmap_size = 0; - kill_ioctx(mm, ctx, NULL); + kill_ioctx(mm, ctx, &requests_done); + + /* Wait until all IO for the context are done. */ + wait_for_completion(&requests_done); } RCU_INIT_POINTER(mm->ioctx_table, NULL); @@ -857,6 +863,68 @@ out: return ret; } +/* refill_reqs_available + * Updates the reqs_available reference counts used for tracking the + * number of free slots in the completion ring. This can be called + * from aio_complete() (to optimistically update reqs_available) or + * from aio_get_req() (the we're out of events case). It must be + * called holding ctx->completion_lock. + */ +static void refill_reqs_available(struct kioctx *ctx, unsigned head, + unsigned tail) +{ + unsigned events_in_ring, completed; + + /* Clamp head since userland can write to it. */ + head %= ctx->nr_events; + if (head <= tail) + events_in_ring = tail - head; + else + events_in_ring = ctx->nr_events - (head - tail); + + completed = ctx->completed_events; + if (events_in_ring < completed) + completed -= events_in_ring; + else + completed = 0; + + if (!completed) + return; + + ctx->completed_events -= completed; + put_reqs_available(ctx, completed); +} + +/* user_refill_reqs_available + * Called to refill reqs_available when aio_get_req() encounters an + * out of space in the completion ring. + */ +static void user_refill_reqs_available(struct kioctx *ctx) +{ + spin_lock_irq(&ctx->completion_lock); + if (ctx->completed_events) { + struct aio_ring *ring; + unsigned head; + + /* Access of ring->head may race with aio_read_events_ring() + * here, but that's okay since whether we read the old version + * or the new version, and either will be valid. The important + * part is that head cannot pass tail since we prevent + * aio_complete() from updating tail by holding + * ctx->completion_lock. Even if head is invalid, the check + * against ctx->completed_events below will make sure we do the + * safe/right thing. + */ + ring = kmap_atomic(ctx->ring_pages[0]); + head = ring->head; + kunmap_atomic(ring); + + refill_reqs_available(ctx, head, ctx->tail); + } + + spin_unlock_irq(&ctx->completion_lock); +} + /* aio_get_req * Allocate a slot for an aio request. * Returns NULL if no requests are free. @@ -865,8 +933,11 @@ static inline struct kiocb *aio_get_req(struct kioctx *ctx) { struct kiocb *req; - if (!get_reqs_available(ctx)) - return NULL; + if (!get_reqs_available(ctx)) { + user_refill_reqs_available(ctx); + if (!get_reqs_available(ctx)) + return NULL; + } req = kmem_cache_alloc(kiocb_cachep, GFP_KERNEL|__GFP_ZERO); if (unlikely(!req)) @@ -925,8 +996,8 @@ void aio_complete(struct kiocb *iocb, long res, long res2) struct kioctx *ctx = iocb->ki_ctx; struct aio_ring *ring; struct io_event *ev_page, *event; + unsigned tail, pos, head; unsigned long flags; - unsigned tail, pos; /* * Special case handling for sync iocbs: @@ -987,10 +1058,14 @@ void aio_complete(struct kiocb *iocb, long res, long res2) ctx->tail = tail; ring = kmap_atomic(ctx->ring_pages[0]); + head = ring->head; ring->tail = tail; kunmap_atomic(ring); flush_dcache_page(ctx->ring_pages[0]); + ctx->completed_events++; + if (ctx->completed_events > 1) + refill_reqs_available(ctx, head, tail); spin_unlock_irqrestore(&ctx->completion_lock, flags); pr_debug("added to ring %p at [%u]\n", iocb, tail); @@ -1005,7 +1080,6 @@ void aio_complete(struct kiocb *iocb, long res, long res2) /* everything turned out well, dispose of the aiocb. */ kiocb_free(iocb); - put_reqs_available(ctx, 1); /* * We have to order our ring_info tail store above and test @@ -1042,6 +1116,12 @@ static long aio_read_events_ring(struct kioctx *ctx, tail = ring->tail; kunmap_atomic(ring); + /* + * Ensure that once we've read the current tail pointer, that + * we also see the events that were stored up to the tail. + */ + smp_rmb(); + pr_debug("h%u t%u m%u\n", head, tail, ctx->nr_events); if (head == tail) diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c index 5a201d8..fbd76de 100644 --- a/fs/btrfs/async-thread.c +++ b/fs/btrfs/async-thread.c @@ -22,7 +22,6 @@ #include <linux/list.h> #include <linux/spinlock.h> #include <linux/freezer.h> -#include <linux/workqueue.h> #include "async-thread.h" #include "ctree.h" @@ -55,8 +54,39 @@ struct btrfs_workqueue { struct __btrfs_workqueue *high; }; -static inline struct __btrfs_workqueue -*__btrfs_alloc_workqueue(const char *name, int flags, int max_active, +static void normal_work_helper(struct btrfs_work *work); + +#define BTRFS_WORK_HELPER(name) \ +void btrfs_##name(struct work_struct *arg) \ +{ \ + struct btrfs_work *work = container_of(arg, struct btrfs_work, \ + normal_work); \ + normal_work_helper(work); \ +} + +BTRFS_WORK_HELPER(worker_helper); +BTRFS_WORK_HELPER(delalloc_helper); +BTRFS_WORK_HELPER(flush_delalloc_helper); +BTRFS_WORK_HELPER(cache_helper); +BTRFS_WORK_HELPER(submit_helper); +BTRFS_WORK_HELPER(fixup_helper); +BTRFS_WORK_HELPER(endio_helper); +BTRFS_WORK_HELPER(endio_meta_helper); +BTRFS_WORK_HELPER(endio_meta_write_helper); +BTRFS_WORK_HELPER(endio_raid56_helper); +BTRFS_WORK_HELPER(rmw_helper); +BTRFS_WORK_HELPER(endio_write_helper); +BTRFS_WORK_HELPER(freespace_write_helper); +BTRFS_WORK_HELPER(delayed_meta_helper); +BTRFS_WORK_HELPER(readahead_helper); +BTRFS_WORK_HELPER(qgroup_rescan_helper); +BTRFS_WORK_HELPER(extent_refs_helper); +BTRFS_WORK_HELPER(scrub_helper); +BTRFS_WORK_HELPER(scrubwrc_helper); +BTRFS_WORK_HELPER(scrubnc_helper); + +static struct __btrfs_workqueue * +__btrfs_alloc_workqueue(const char *name, int flags, int max_active, int thresh) { struct __btrfs_workqueue *ret = kzalloc(sizeof(*ret), GFP_NOFS); @@ -232,13 +262,11 @@ static void run_ordered_work(struct __btrfs_workqueue *wq) spin_unlock_irqrestore(lock, flags); } -static void normal_work_helper(struct work_struct *arg) +static void normal_work_helper(struct btrfs_work *work) { - struct btrfs_work *work; struct __btrfs_workqueue *wq; int need_order = 0; - work = container_of(arg, struct btrfs_work, normal_work); /* * We should not touch things inside work in the following cases: * 1) after work->func() if it has no ordered_free @@ -262,7 +290,7 @@ static void normal_work_helper(struct work_struct *arg) trace_btrfs_all_work_done(work); } -void btrfs_init_work(struct btrfs_work *work, +void btrfs_init_work(struct btrfs_work *work, btrfs_work_func_t uniq_func, btrfs_func_t func, btrfs_func_t ordered_func, btrfs_func_t ordered_free) @@ -270,7 +298,7 @@ void btrfs_init_work(struct btrfs_work *work, work->func = func; work->ordered_func = ordered_func; work->ordered_free = ordered_free; - INIT_WORK(&work->normal_work, normal_work_helper); + INIT_WORK(&work->normal_work, uniq_func); INIT_LIST_HEAD(&work->ordered_list); work->flags = 0; } diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h index 9c6b66d1..e9e31c9 100644 --- a/fs/btrfs/async-thread.h +++ b/fs/btrfs/async-thread.h @@ -19,12 +19,14 @@ #ifndef __BTRFS_ASYNC_THREAD_ #define __BTRFS_ASYNC_THREAD_ +#include <linux/workqueue.h> struct btrfs_workqueue; /* Internal use only */ struct __btrfs_workqueue; struct btrfs_work; typedef void (*btrfs_func_t)(struct btrfs_work *arg); +typedef void (*btrfs_work_func_t)(struct work_struct *arg); struct btrfs_work { btrfs_func_t func; @@ -38,11 +40,35 @@ struct btrfs_work { unsigned long flags; }; +#define BTRFS_WORK_HELPER_PROTO(name) \ +void btrfs_##name(struct work_struct *arg) + +BTRFS_WORK_HELPER_PROTO(worker_helper); +BTRFS_WORK_HELPER_PROTO(delalloc_helper); +BTRFS_WORK_HELPER_PROTO(flush_delalloc_helper); +BTRFS_WORK_HELPER_PROTO(cache_helper); +BTRFS_WORK_HELPER_PROTO(submit_helper); +BTRFS_WORK_HELPER_PROTO(fixup_helper); +BTRFS_WORK_HELPER_PROTO(endio_helper); +BTRFS_WORK_HELPER_PROTO(endio_meta_helper); +BTRFS_WORK_HELPER_PROTO(endio_meta_write_helper); +BTRFS_WORK_HELPER_PROTO(endio_raid56_helper); +BTRFS_WORK_HELPER_PROTO(rmw_helper); +BTRFS_WORK_HELPER_PROTO(endio_write_helper); +BTRFS_WORK_HELPER_PROTO(freespace_write_helper); +BTRFS_WORK_HELPER_PROTO(delayed_meta_helper); +BTRFS_WORK_HELPER_PROTO(readahead_helper); +BTRFS_WORK_HELPER_PROTO(qgroup_rescan_helper); +BTRFS_WORK_HELPER_PROTO(extent_refs_helper); +BTRFS_WORK_HELPER_PROTO(scrub_helper); +BTRFS_WORK_HELPER_PROTO(scrubwrc_helper); +BTRFS_WORK_HELPER_PROTO(scrubnc_helper); + struct btrfs_workqueue *btrfs_alloc_workqueue(const char *name, int flags, int max_active, int thresh); -void btrfs_init_work(struct btrfs_work *work, +void btrfs_init_work(struct btrfs_work *work, btrfs_work_func_t helper, btrfs_func_t func, btrfs_func_t ordered_func, btrfs_func_t ordered_free); diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index da775bf..a2e90f8 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -1395,8 +1395,8 @@ static int btrfs_wq_run_delayed_node(struct btrfs_delayed_root *delayed_root, return -ENOMEM; async_work->delayed_root = delayed_root; - btrfs_init_work(&async_work->work, btrfs_async_run_delayed_root, - NULL, NULL); + btrfs_init_work(&async_work->work, btrfs_delayed_meta_helper, + btrfs_async_run_delayed_root, NULL, NULL); async_work->nr = nr; btrfs_queue_work(root->fs_info->delayed_workers, &async_work->work); diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index d0ed9e6..a1d36e6 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -39,7 +39,6 @@ #include "btrfs_inode.h" #include "volumes.h" #include "print-tree.h" -#include "async-thread.h" #include "locking.h" #include "tree-log.h" #include "free-space-cache.h" @@ -693,35 +692,41 @@ static void end_workqueue_bio(struct bio *bio, int err) { struct end_io_wq *end_io_wq = bio->bi_private; struct btrfs_fs_info *fs_info; + struct btrfs_workqueue *wq; + btrfs_work_func_t func; fs_info = end_io_wq->info; end_io_wq->error = err; - btrfs_init_work(&end_io_wq->work, end_workqueue_fn, NULL, NULL); if (bio->bi_rw & REQ_WRITE) { - if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA) - btrfs_queue_work(fs_info->endio_meta_write_workers, - &end_io_wq->work); - else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_FREE_SPACE) - btrfs_queue_work(fs_info->endio_freespace_worker, - &end_io_wq->work); - else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56) - btrfs_queue_work(fs_info->endio_raid56_workers, - &end_io_wq->work); - else - btrfs_queue_work(fs_info->endio_write_workers, - &end_io_wq->work); + if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA) { + wq = fs_info->endio_meta_write_workers; + func = btrfs_endio_meta_write_helper; + } else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_FREE_SPACE) { + wq = fs_info->endio_freespace_worker; + func = btrfs_freespace_write_helper; + } else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56) { + wq = fs_info->endio_raid56_workers; + func = btrfs_endio_raid56_helper; + } else { + wq = fs_info->endio_write_workers; + func = btrfs_endio_write_helper; + } } else { - if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56) - btrfs_queue_work(fs_info->endio_raid56_workers, - &end_io_wq->work); - else if (end_io_wq->metadata) - btrfs_queue_work(fs_info->endio_meta_workers, - &end_io_wq->work); - else - btrfs_queue_work(fs_info->endio_workers, - &end_io_wq->work); + if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56) { + wq = fs_info->endio_raid56_workers; + func = btrfs_endio_raid56_helper; + } else if (end_io_wq->metadata) { + wq = fs_info->endio_meta_workers; + func = btrfs_endio_meta_helper; + } else { + wq = fs_info->endio_workers; + func = btrfs_endio_helper; + } } + + btrfs_init_work(&end_io_wq->work, func, end_workqueue_fn, NULL, NULL); + btrfs_queue_work(wq, &end_io_wq->work); } /* @@ -828,7 +833,7 @@ int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode, async->submit_bio_start = submit_bio_start; async->submit_bio_done = submit_bio_done; - btrfs_init_work(&async->work, run_one_async_start, + btrfs_init_work(&async->work, btrfs_worker_helper, run_one_async_start, run_one_async_done, run_one_async_free); async->bio_flags = bio_flags; @@ -3450,7 +3455,8 @@ static int write_all_supers(struct btrfs_root *root, int max_mirrors) btrfs_set_stack_device_generation(dev_item, 0); btrfs_set_stack_device_type(dev_item, dev->type); btrfs_set_stack_device_id(dev_item, dev->devid); - btrfs_set_stack_device_total_bytes(dev_item, dev->total_bytes); + btrfs_set_stack_device_total_bytes(dev_item, + dev->disk_total_bytes); btrfs_set_stack_device_bytes_used(dev_item, dev->bytes_used); btrfs_set_stack_device_io_align(dev_item, dev->io_align); btrfs_set_stack_device_io_width(dev_item, dev->io_width); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 102ed31..3efe1c3 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -552,7 +552,8 @@ static int cache_block_group(struct btrfs_block_group_cache *cache, caching_ctl->block_group = cache; caching_ctl->progress = cache->key.objectid; atomic_set(&caching_ctl->count, 1); - btrfs_init_work(&caching_ctl->work, caching_thread, NULL, NULL); + btrfs_init_work(&caching_ctl->work, btrfs_cache_helper, + caching_thread, NULL, NULL); spin_lock(&cache->lock); /* @@ -2749,8 +2750,8 @@ int btrfs_async_run_delayed_refs(struct btrfs_root *root, async->sync = 0; init_completion(&async->wait); - btrfs_init_work(&async->work, delayed_ref_async_start, - NULL, NULL); + btrfs_init_work(&async->work, btrfs_extent_refs_helper, + delayed_ref_async_start, NULL, NULL); btrfs_queue_work(root->fs_info->extent_workers, &async->work); @@ -3586,13 +3587,7 @@ static u64 get_restripe_target(struct btrfs_fs_info *fs_info, u64 flags) */ static u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags) { - /* - * we add in the count of missing devices because we want - * to make sure that any RAID levels on a degraded FS - * continue to be honored. - */ - u64 num_devices = root->fs_info->fs_devices->rw_devices + - root->fs_info->fs_devices->missing_devices; + u64 num_devices = root->fs_info->fs_devices->rw_devices; u64 target; u64 tmp; @@ -8440,13 +8435,7 @@ static u64 update_block_group_flags(struct btrfs_root *root, u64 flags) if (stripped) return extended_to_chunk(stripped); - /* - * we add in the count of missing devices because we want - * to make sure that any RAID levels on a degraded FS - * continue to be honored. - */ - num_devices = root->fs_info->fs_devices->rw_devices + - root->fs_info->fs_devices->missing_devices; + num_devices = root->fs_info->fs_devices->rw_devices; stripped = BTRFS_BLOCK_GROUP_RAID0 | BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6 | diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 3e11aab..af0359d 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2532,6 +2532,7 @@ static void end_bio_extent_readpage(struct bio *bio, int err) test_bit(BIO_UPTODATE, &bio->bi_flags); if (err) uptodate = 0; + offset += len; continue; } } @@ -4207,8 +4208,8 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, return -ENOMEM; path->leave_spinning = 1; - start = ALIGN(start, BTRFS_I(inode)->root->sectorsize); - len = ALIGN(len, BTRFS_I(inode)->root->sectorsize); + start = round_down(start, BTRFS_I(inode)->root->sectorsize); + len = round_up(max, BTRFS_I(inode)->root->sectorsize) - start; /* * lookup the last file extent. We're not using i_size here diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index d3afac2..36861b7 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1840,7 +1840,15 @@ int btrfs_release_file(struct inode *inode, struct file *filp) { if (filp->private_data) btrfs_ioctl_trans_end(filp); - filemap_flush(inode->i_mapping); + /* + * ordered_data_close is set by settattr when we are about to truncate + * a file from a non-zero size to a zero size. This tries to + * flush down new bytes that may have been written if the + * application were using truncate to replace a file in place. + */ + if (test_and_clear_bit(BTRFS_INODE_ORDERED_DATA_CLOSE, + &BTRFS_I(inode)->runtime_flags)) + filemap_flush(inode->i_mapping); return 0; } @@ -2088,10 +2096,9 @@ static int fill_holes(struct btrfs_trans_handle *trans, struct inode *inode, goto out; } - if (hole_mergeable(inode, leaf, path->slots[0]+1, offset, end)) { + if (hole_mergeable(inode, leaf, path->slots[0], offset, end)) { u64 num_bytes; - path->slots[0]++; key.offset = offset; btrfs_set_item_key_safe(root, path, &key); fi = btrfs_item_ptr(leaf, path->slots[0], @@ -2216,7 +2223,7 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) goto out_only_mutex; } - lockstart = round_up(offset , BTRFS_I(inode)->root->sectorsize); + lockstart = round_up(offset, BTRFS_I(inode)->root->sectorsize); lockend = round_down(offset + len, BTRFS_I(inode)->root->sectorsize) - 1; same_page = ((offset >> PAGE_CACHE_SHIFT) == @@ -2277,7 +2284,7 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) tail_start + tail_len, 0, 1); if (ret) goto out_only_mutex; - } + } } } diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 03708ef..9c194bd 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1096,8 +1096,10 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page, async_cow->end = cur_end; INIT_LIST_HEAD(&async_cow->extents); - btrfs_init_work(&async_cow->work, async_cow_start, - async_cow_submit, async_cow_free); + btrfs_init_work(&async_cow->work, + btrfs_delalloc_helper, + async_cow_start, async_cow_submit, + async_cow_free); nr_pages = (cur_end - start + PAGE_CACHE_SIZE) >> PAGE_CACHE_SHIFT; @@ -1881,7 +1883,8 @@ static int btrfs_writepage_start_hook(struct page *page, u64 start, u64 end) SetPageChecked(page); page_cache_get(page); - btrfs_init_work(&fixup->work, btrfs_writepage_fixup_worker, NULL, NULL); + btrfs_init_work(&fixup->work, btrfs_fixup_helper, + btrfs_writepage_fixup_worker, NULL, NULL); fixup->page = page; btrfs_queue_work(root->fs_info->fixup_workers, &fixup->work); return -EBUSY; @@ -2822,7 +2825,8 @@ static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end, struct inode *inode = page->mapping->host; struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_ordered_extent *ordered_extent = NULL; - struct btrfs_workqueue *workers; + struct btrfs_workqueue *wq; + btrfs_work_func_t func; trace_btrfs_writepage_end_io_hook(page, start, end, uptodate); @@ -2831,13 +2835,17 @@ static int btrfs_writepage_end_io_hook(struct page *page, u64 start, u64 end, end - start + 1, uptodate)) return 0; - btrfs_init_work(&ordered_extent->work, finish_ordered_fn, NULL, NULL); + if (btrfs_is_free_space_inode(inode)) { + wq = root->fs_info->endio_freespace_worker; + func = btrfs_freespace_write_helper; + } else { + wq = root->fs_info->endio_write_workers; + func = btrfs_endio_write_helper; + } - if (btrfs_is_free_space_inode(inode)) - workers = root->fs_info->endio_freespace_worker; - else - workers = root->fs_info->endio_write_workers; - btrfs_queue_work(workers, &ordered_extent->work); + btrfs_init_work(&ordered_extent->work, func, finish_ordered_fn, NULL, + NULL); + btrfs_queue_work(wq, &ordered_extent->work); return 0; } @@ -4674,6 +4682,11 @@ static void evict_inode_truncate_pages(struct inode *inode) clear_bit(EXTENT_FLAG_LOGGING, &em->flags); remove_extent_mapping(map_tree, em); free_extent_map(em); + if (need_resched()) { + write_unlock(&map_tree->lock); + cond_resched(); + write_lock(&map_tree->lock); + } } write_unlock(&map_tree->lock); @@ -4696,6 +4709,7 @@ static void evict_inode_truncate_pages(struct inode *inode) &cached_state, GFP_NOFS); free_extent_state(state); + cond_resched(); spin_lock(&io_tree->lock); } spin_unlock(&io_tree->lock); @@ -5181,6 +5195,42 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry) iput(inode); inode = ERR_PTR(ret); } + /* + * If orphan cleanup did remove any orphans, it means the tree + * was modified and therefore the commit root is not the same as + * the current root anymore. This is a problem, because send + * uses the commit root and therefore can see inode items that + * don't exist in the current root anymore, and for example make + * calls to btrfs_iget, which will do tree lookups based on the + * current root and not on the commit root. Those lookups will + * fail, returning a -ESTALE error, and making send fail with + * that error. So make sure a send does not see any orphans we + * have just removed, and that it will see the same inodes + * regardless of whether a transaction commit happened before + * it started (meaning that the commit root will be the same as + * the current root) or not. + */ + if (sub_root->node != sub_root->commit_root) { + u64 sub_flags = btrfs_root_flags(&sub_root->root_item); + + if (sub_flags & BTRFS_ROOT_SUBVOL_RDONLY) { + struct extent_buffer *eb; + + /* + * Assert we can't have races between dentry + * lookup called through the snapshot creation + * ioctl and the VFS. + */ + ASSERT(mutex_is_locked(&dir->i_mutex)); + + down_write(&root->fs_info->commit_root_sem); + eb = sub_root->commit_root; + sub_root->commit_root = + btrfs_root_node(sub_root); + up_write(&root->fs_info->commit_root_sem); + free_extent_buffer(eb); + } + } } return inode; @@ -5606,6 +5656,13 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, } /* + * O_TMPFILE, set link count to 0, so that after this point, + * we fill in an inode item with the correct link count. + */ + if (!name) + set_nlink(inode, 0); + + /* * we have to initialize this early, so we can reclaim the inode * number if we fail afterwards in this function. */ @@ -6097,14 +6154,14 @@ out_fail: static int merge_extent_mapping(struct extent_map_tree *em_tree, struct extent_map *existing, struct extent_map *em, - u64 map_start, u64 map_len) + u64 map_start) { u64 start_diff; BUG_ON(map_start < em->start || map_start >= extent_map_end(em)); start_diff = map_start - em->start; em->start = map_start; - em->len = map_len; + em->len = existing->start - em->start; if (em->block_start < EXTENT_MAP_LAST_BYTE && !test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) { em->block_start += start_diff; @@ -6275,6 +6332,8 @@ next: goto not_found; if (start + len <= found_key.offset) goto not_found; + if (start > found_key.offset) + goto next; em->start = start; em->orig_start = start; em->len = found_key.offset - start; @@ -6390,8 +6449,7 @@ insert: em->len); if (existing) { err = merge_extent_mapping(em_tree, existing, - em, start, - root->sectorsize); + em, start); free_extent_map(existing); if (err) { free_extent_map(em); @@ -7158,7 +7216,8 @@ again: if (!ret) goto out_test; - btrfs_init_work(&ordered->work, finish_ordered_fn, NULL, NULL); + btrfs_init_work(&ordered->work, btrfs_endio_write_helper, + finish_ordered_fn, NULL, NULL); btrfs_queue_work(root->fs_info->endio_write_workers, &ordered->work); out_test: @@ -7306,10 +7365,8 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip, map_length = orig_bio->bi_iter.bi_size; ret = btrfs_map_block(root->fs_info, rw, start_sector << 9, &map_length, NULL, 0); - if (ret) { - bio_put(orig_bio); + if (ret) return -EIO; - } if (map_length >= orig_bio->bi_iter.bi_size) { bio = orig_bio; @@ -7326,6 +7383,7 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip, bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, start_sector, GFP_NOFS); if (!bio) return -ENOMEM; + bio->bi_private = dip; bio->bi_end_io = btrfs_end_dio_bio; atomic_inc(&dip->pending_bios); @@ -7534,7 +7592,8 @@ static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb, count = iov_iter_count(iter); if (test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT, &BTRFS_I(inode)->runtime_flags)) - filemap_fdatawrite_range(inode->i_mapping, offset, count); + filemap_fdatawrite_range(inode->i_mapping, offset, + offset + count - 1); if (rw & WRITE) { /* @@ -8495,7 +8554,9 @@ struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode, work->inode = inode; work->wait = wait; work->delay_iput = delay_iput; - btrfs_init_work(&work->work, btrfs_run_delalloc_work, NULL, NULL); + WARN_ON_ONCE(!inode); + btrfs_init_work(&work->work, btrfs_flush_delalloc_helper, + btrfs_run_delalloc_work, NULL, NULL); return work; } @@ -8979,6 +9040,14 @@ static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) if (ret) goto out; + /* + * We set number of links to 0 in btrfs_new_inode(), and here we set + * it to 1 because d_tmpfile() will issue a warning if the count is 0, + * through: + * + * d_tmpfile() -> inode_dec_link_count() -> drop_nlink() + */ + set_nlink(inode, 1); d_tmpfile(dentry, inode); mark_inode_dirty(inode); diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 47aceb4..fce6fd0e 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -711,39 +711,6 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, if (ret) goto fail; - ret = btrfs_orphan_cleanup(pending_snapshot->snap); - if (ret) - goto fail; - - /* - * If orphan cleanup did remove any orphans, it means the tree was - * modified and therefore the commit root is not the same as the - * current root anymore. This is a problem, because send uses the - * commit root and therefore can see inode items that don't exist - * in the current root anymore, and for example make calls to - * btrfs_iget, which will do tree lookups based on the current root - * and not on the commit root. Those lookups will fail, returning a - * -ESTALE error, and making send fail with that error. So make sure - * a send does not see any orphans we have just removed, and that it - * will see the same inodes regardless of whether a transaction - * commit happened before it started (meaning that the commit root - * will be the same as the current root) or not. - */ - if (readonly && pending_snapshot->snap->node != - pending_snapshot->snap->commit_root) { - trans = btrfs_join_transaction(pending_snapshot->snap); - if (IS_ERR(trans) && PTR_ERR(trans) != -ENOENT) { - ret = PTR_ERR(trans); - goto fail; - } - if (!IS_ERR(trans)) { - ret = btrfs_commit_transaction(trans, - pending_snapshot->snap); - if (ret) - goto fail; - } - } - inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry); if (IS_ERR(inode)) { ret = PTR_ERR(inode); @@ -3527,7 +3494,8 @@ process_slot: btrfs_mark_buffer_dirty(leaf); btrfs_release_path(path); - last_dest_end = new_key.offset + datal; + last_dest_end = ALIGN(new_key.offset + datal, + root->sectorsize); ret = clone_finish_inode_update(trans, inode, last_dest_end, destoff, olen); diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index 963895c..ac734ec 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -615,6 +615,7 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr) spin_unlock(&root->ordered_extent_lock); btrfs_init_work(&ordered->flush_work, + btrfs_flush_delalloc_helper, btrfs_run_ordered_extent_work, NULL, NULL); list_add_tail(&ordered->work_list, &works); btrfs_queue_work(root->fs_info->flush_workers, diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index b497498..ded5c60 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -1973,7 +1973,7 @@ static int qgroup_subtree_accounting(struct btrfs_trans_handle *trans, elem.seq, &roots); btrfs_put_tree_mod_seq(fs_info, &elem); if (ret < 0) - return ret; + goto out; if (roots->nnodes != 1) goto out; @@ -2720,6 +2720,7 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid, memset(&fs_info->qgroup_rescan_work, 0, sizeof(fs_info->qgroup_rescan_work)); btrfs_init_work(&fs_info->qgroup_rescan_work, + btrfs_qgroup_rescan_helper, btrfs_qgroup_rescan_worker, NULL, NULL); if (ret) { diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c index 4a88f07..0a6b6e4 100644 --- a/fs/btrfs/raid56.c +++ b/fs/btrfs/raid56.c @@ -1416,7 +1416,8 @@ cleanup: static void async_rmw_stripe(struct btrfs_raid_bio *rbio) { - btrfs_init_work(&rbio->work, rmw_work, NULL, NULL); + btrfs_init_work(&rbio->work, btrfs_rmw_helper, + rmw_work, NULL, NULL); btrfs_queue_work(rbio->fs_info->rmw_workers, &rbio->work); @@ -1424,7 +1425,8 @@ static void async_rmw_stripe(struct btrfs_raid_bio *rbio) static void async_read_rebuild(struct btrfs_raid_bio *rbio) { - btrfs_init_work(&rbio->work, read_rebuild_work, NULL, NULL); + btrfs_init_work(&rbio->work, btrfs_rmw_helper, + read_rebuild_work, NULL, NULL); btrfs_queue_work(rbio->fs_info->rmw_workers, &rbio->work); @@ -1665,7 +1667,8 @@ static void btrfs_raid_unplug(struct blk_plug_cb *cb, bool from_schedule) plug = container_of(cb, struct btrfs_plug_cb, cb); if (from_schedule) { - btrfs_init_work(&plug->work, unplug_work, NULL, NULL); + btrfs_init_work(&plug->work, btrfs_rmw_helper, + unplug_work, NULL, NULL); btrfs_queue_work(plug->info->rmw_workers, &plug->work); return; diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c index 09230cf..20408c6 100644 --- a/fs/btrfs/reada.c +++ b/fs/btrfs/reada.c @@ -798,7 +798,8 @@ static void reada_start_machine(struct btrfs_fs_info *fs_info) /* FIXME we cannot handle this properly right now */ BUG(); } - btrfs_init_work(&rmw->work, reada_start_machine_worker, NULL, NULL); + btrfs_init_work(&rmw->work, btrfs_readahead_helper, + reada_start_machine_worker, NULL, NULL); rmw->fs_info = fs_info; btrfs_queue_work(fs_info->readahead_workers, &rmw->work); diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index b6d198f..f4a41f3 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -428,8 +428,8 @@ struct scrub_ctx *scrub_setup_ctx(struct btrfs_device *dev, int is_dev_replace) sbio->index = i; sbio->sctx = sctx; sbio->page_count = 0; - btrfs_init_work(&sbio->work, scrub_bio_end_io_worker, - NULL, NULL); + btrfs_init_work(&sbio->work, btrfs_scrub_helper, + scrub_bio_end_io_worker, NULL, NULL); if (i != SCRUB_BIOS_PER_SCTX - 1) sctx->bios[i]->next_free = i + 1; @@ -999,8 +999,8 @@ nodatasum_case: fixup_nodatasum->root = fs_info->extent_root; fixup_nodatasum->mirror_num = failed_mirror_index + 1; scrub_pending_trans_workers_inc(sctx); - btrfs_init_work(&fixup_nodatasum->work, scrub_fixup_nodatasum, - NULL, NULL); + btrfs_init_work(&fixup_nodatasum->work, btrfs_scrub_helper, + scrub_fixup_nodatasum, NULL, NULL); btrfs_queue_work(fs_info->scrub_workers, &fixup_nodatasum->work); goto out; @@ -1616,7 +1616,8 @@ static void scrub_wr_bio_end_io(struct bio *bio, int err) sbio->err = err; sbio->bio = bio; - btrfs_init_work(&sbio->work, scrub_wr_bio_end_io_worker, NULL, NULL); + btrfs_init_work(&sbio->work, btrfs_scrubwrc_helper, + scrub_wr_bio_end_io_worker, NULL, NULL); btrfs_queue_work(fs_info->scrub_wr_completion_workers, &sbio->work); } @@ -2904,6 +2905,7 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start, struct scrub_ctx *sctx; int ret; struct btrfs_device *dev; + struct rcu_string *name; if (btrfs_fs_closing(fs_info)) return -EINVAL; @@ -2965,6 +2967,16 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start, return -ENODEV; } + if (!is_dev_replace && !readonly && !dev->writeable) { + mutex_unlock(&fs_info->fs_devices->device_list_mutex); + rcu_read_lock(); + name = rcu_dereference(dev->name); + btrfs_err(fs_info, "scrub: device %s is not writable", + name->str); + rcu_read_unlock(); + return -EROFS; + } + mutex_lock(&fs_info->scrub_lock); if (!dev->in_fs_metadata || dev->is_tgtdev_for_dev_replace) { mutex_unlock(&fs_info->scrub_lock); @@ -3203,7 +3215,8 @@ static int copy_nocow_pages(struct scrub_ctx *sctx, u64 logical, u64 len, nocow_ctx->len = len; nocow_ctx->mirror_num = mirror_num; nocow_ctx->physical_for_dev_replace = physical_for_dev_replace; - btrfs_init_work(&nocow_ctx->work, copy_nocow_pages_worker, NULL, NULL); + btrfs_init_work(&nocow_ctx->work, btrfs_scrubnc_helper, + copy_nocow_pages_worker, NULL, NULL); INIT_LIST_HEAD(&nocow_ctx->inodes); btrfs_queue_work(fs_info->scrub_nocow_workers, &nocow_ctx->work); diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c index 7869936..12e5355 100644 --- a/fs/btrfs/sysfs.c +++ b/fs/btrfs/sysfs.c @@ -614,7 +614,7 @@ int btrfs_kobj_rm_device(struct btrfs_fs_info *fs_info, if (!fs_info->device_dir_kobj) return -EINVAL; - if (one_device) { + if (one_device && one_device->bdev) { disk = one_device->bdev->bd_part; disk_kobj = &part_to_dev(disk)->kobj; diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 9e1f2cd..7e0e6e3 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -3298,7 +3298,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans, struct list_head ordered_sums; int skip_csum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM; bool has_extents = false; - bool need_find_last_extent = (*last_extent == 0); + bool need_find_last_extent = true; bool done = false; INIT_LIST_HEAD(&ordered_sums); @@ -3352,8 +3352,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans, */ if (ins_keys[i].type == BTRFS_EXTENT_DATA_KEY) { has_extents = true; - if (need_find_last_extent && - first_key.objectid == (u64)-1) + if (first_key.objectid == (u64)-1) first_key = ins_keys[i]; } else { need_find_last_extent = false; @@ -3427,6 +3426,16 @@ static noinline int copy_items(struct btrfs_trans_handle *trans, if (!has_extents) return ret; + if (need_find_last_extent && *last_extent == first_key.offset) { + /* + * We don't have any leafs between our current one and the one + * we processed before that can have file extent items for our + * inode (and have a generation number smaller than our current + * transaction id). + */ + need_find_last_extent = false; + } + /* * Because we use btrfs_search_forward we could skip leaves that were * not modified and then assume *last_extent is valid when it really @@ -3537,7 +3546,7 @@ fill_holes: 0, 0); if (ret) break; - *last_extent = offset + len; + *last_extent = extent_end; } /* * Need to let the callers know we dropped the path so they should diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 6cb82f6..340a92d 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -508,6 +508,44 @@ static noinline int device_list_add(const char *path, ret = 1; device->fs_devices = fs_devices; } else if (!device->name || strcmp(device->name->str, path)) { + /* + * When FS is already mounted. + * 1. If you are here and if the device->name is NULL that + * means this device was missing at time of FS mount. + * 2. If you are here and if the device->name is different + * from 'path' that means either + * a. The same device disappeared and reappeared with + * different name. or + * b. The missing-disk-which-was-replaced, has + * reappeared now. + * + * We must allow 1 and 2a above. But 2b would be a spurious + * and unintentional. + * + * Further in case of 1 and 2a above, the disk at 'path' + * would have missed some transaction when it was away and + * in case of 2a the stale bdev has to be updated as well. + * 2b must not be allowed at all time. + */ + + /* + * As of now don't allow update to btrfs_fs_device through + * the btrfs dev scan cli, after FS has been mounted. + */ + if (fs_devices->opened) { + return -EBUSY; + } else { + /* + * That is if the FS is _not_ mounted and if you + * are here, that means there is more than one + * disk with same uuid and devid.We keep the one + * with larger generation number or the last-in if + * generation are equal. + */ + if (found_transid < device->generation) + return -EEXIST; + } + name = rcu_string_strdup(path, GFP_NOFS); if (!name) return -ENOMEM; @@ -519,6 +557,15 @@ static noinline int device_list_add(const char *path, } } + /* + * Unmount does not free the btrfs_device struct but would zero + * generation along with most of the other members. So just update + * it back. We need it to pick the disk with largest generation + * (as above). + */ + if (!fs_devices->opened) + device->generation = found_transid; + if (found_transid > fs_devices->latest_trans) { fs_devices->latest_devid = devid; fs_devices->latest_trans = found_transid; @@ -1436,7 +1483,7 @@ static int btrfs_add_device(struct btrfs_trans_handle *trans, btrfs_set_device_io_align(leaf, dev_item, device->io_align); btrfs_set_device_io_width(leaf, dev_item, device->io_width); btrfs_set_device_sector_size(leaf, dev_item, device->sector_size); - btrfs_set_device_total_bytes(leaf, dev_item, device->total_bytes); + btrfs_set_device_total_bytes(leaf, dev_item, device->disk_total_bytes); btrfs_set_device_bytes_used(leaf, dev_item, device->bytes_used); btrfs_set_device_group(leaf, dev_item, 0); btrfs_set_device_seek_speed(leaf, dev_item, 0); @@ -1671,7 +1718,7 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path) device->fs_devices->total_devices--; if (device->missing) - root->fs_info->fs_devices->missing_devices--; + device->fs_devices->missing_devices--; next_device = list_entry(root->fs_info->fs_devices->devices.next, struct btrfs_device, dev_list); @@ -1801,8 +1848,12 @@ void btrfs_rm_dev_replace_srcdev(struct btrfs_fs_info *fs_info, if (srcdev->bdev) { fs_info->fs_devices->open_devices--; - /* zero out the old super */ - btrfs_scratch_superblock(srcdev); + /* + * zero out the old super if it is not writable + * (e.g. seed device) + */ + if (srcdev->writeable) + btrfs_scratch_superblock(srcdev); } call_rcu(&srcdev->rcu, free_device); @@ -1941,6 +1992,9 @@ static int btrfs_prepare_sprout(struct btrfs_root *root) fs_devices->seeding = 0; fs_devices->num_devices = 0; fs_devices->open_devices = 0; + fs_devices->missing_devices = 0; + fs_devices->num_can_discard = 0; + fs_devices->rotating = 0; fs_devices->seed = seed_devices; generate_random_uuid(fs_devices->fsid); @@ -5800,7 +5854,8 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info, else generate_random_uuid(dev->uuid); - btrfs_init_work(&dev->work, pending_bios_fn, NULL, NULL); + btrfs_init_work(&dev->work, btrfs_submit_helper, + pending_bios_fn, NULL, NULL); return dev; } diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 5b19760..b0c225c 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -1825,7 +1825,7 @@ ext4_group_first_block_no(struct super_block *sb, ext4_group_t group_no) /* * Special error return code only used by dx_probe() and its callers. */ -#define ERR_BAD_DX_DIR -75000 +#define ERR_BAD_DX_DIR (-(MAX_ERRNO - 1)) /* * Timeout and state flag for lazy initialization inode thread. @@ -2454,6 +2454,22 @@ static inline void ext4_update_i_disksize(struct inode *inode, loff_t newsize) up_write(&EXT4_I(inode)->i_data_sem); } +/* Update i_size, i_disksize. Requires i_mutex to avoid races with truncate */ +static inline int ext4_update_inode_size(struct inode *inode, loff_t newsize) +{ + int changed = 0; + + if (newsize > inode->i_size) { + i_size_write(inode, newsize); + changed = 1; + } + if (newsize > EXT4_I(inode)->i_disksize) { + ext4_update_i_disksize(inode, newsize); + changed |= 2; + } + return changed; +} + struct ext4_group_info { unsigned long bb_state; struct rb_root bb_free_root; diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 76c2df3..74292a7 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -4665,7 +4665,8 @@ retry: } static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset, - ext4_lblk_t len, int flags, int mode) + ext4_lblk_t len, loff_t new_size, + int flags, int mode) { struct inode *inode = file_inode(file); handle_t *handle; @@ -4674,8 +4675,10 @@ static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset, int retries = 0; struct ext4_map_blocks map; unsigned int credits; + loff_t epos; map.m_lblk = offset; + map.m_len = len; /* * Don't normalize the request if it can fit in one extent so * that it doesn't get unnecessarily split into multiple @@ -4690,9 +4693,7 @@ static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset, credits = ext4_chunk_trans_blocks(inode, len); retry: - while (ret >= 0 && ret < len) { - map.m_lblk = map.m_lblk + ret; - map.m_len = len = len - ret; + while (ret >= 0 && len) { handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, credits); if (IS_ERR(handle)) { @@ -4709,6 +4710,21 @@ retry: ret2 = ext4_journal_stop(handle); break; } + map.m_lblk += ret; + map.m_len = len = len - ret; + epos = (loff_t)map.m_lblk << inode->i_blkbits; + inode->i_ctime = ext4_current_time(inode); + if (new_size) { + if (epos > new_size) + epos = new_size; + if (ext4_update_inode_size(inode, epos) & 0x1) + inode->i_mtime = inode->i_ctime; + } else { + if (epos > inode->i_size) + ext4_set_inode_flag(inode, + EXT4_INODE_EOFBLOCKS); + } + ext4_mark_inode_dirty(handle, inode); ret2 = ext4_journal_stop(handle); if (ret2) break; @@ -4731,7 +4747,8 @@ static long ext4_zero_range(struct file *file, loff_t offset, loff_t new_size = 0; int ret = 0; int flags; - int partial; + int credits; + int partial_begin, partial_end; loff_t start, end; ext4_lblk_t lblk; struct address_space *mapping = inode->i_mapping; @@ -4771,7 +4788,8 @@ static long ext4_zero_range(struct file *file, loff_t offset, if (start < offset || end > offset + len) return -EINVAL; - partial = (offset + len) & ((1 << blkbits) - 1); + partial_begin = offset & ((1 << blkbits) - 1); + partial_end = (offset + len) & ((1 << blkbits) - 1); lblk = start >> blkbits; max_blocks = (end >> blkbits); @@ -4805,7 +4823,7 @@ static long ext4_zero_range(struct file *file, loff_t offset, * If we have a partial block after EOF we have to allocate * the entire block. */ - if (partial) + if (partial_end) max_blocks += 1; } @@ -4813,6 +4831,7 @@ static long ext4_zero_range(struct file *file, loff_t offset, /* Now release the pages and zero block aligned part of pages*/ truncate_pagecache_range(inode, start, end - 1); + inode->i_mtime = inode->i_ctime = ext4_current_time(inode); /* Wait all existing dio workers, newcomers will block on i_mutex */ ext4_inode_block_unlocked_dio(inode); @@ -4825,13 +4844,22 @@ static long ext4_zero_range(struct file *file, loff_t offset, if (ret) goto out_dio; - ret = ext4_alloc_file_blocks(file, lblk, max_blocks, flags, - mode); + ret = ext4_alloc_file_blocks(file, lblk, max_blocks, new_size, + flags, mode); if (ret) goto out_dio; } + if (!partial_begin && !partial_end) + goto out_dio; - handle = ext4_journal_start(inode, EXT4_HT_MISC, 4); + /* + * In worst case we have to writeout two nonadjacent unwritten + * blocks and update the inode + */ + credits = (2 * ext4_ext_index_trans_blocks(inode, 2)) + 1; + if (ext4_should_journal_data(inode)) + credits += 2; + handle = ext4_journal_start(inode, EXT4_HT_MISC, credits); if (IS_ERR(handle)) { ret = PTR_ERR(handle); ext4_std_error(inode->i_sb, ret); @@ -4839,12 +4867,8 @@ static long ext4_zero_range(struct file *file, loff_t offset, } inode->i_mtime = inode->i_ctime = ext4_current_time(inode); - if (new_size) { - if (new_size > i_size_read(inode)) - i_size_write(inode, new_size); - if (new_size > EXT4_I(inode)->i_disksize) - ext4_update_i_disksize(inode, new_size); + ext4_update_inode_size(inode, new_size); } else { /* * Mark that we allocate beyond EOF so the subsequent truncate @@ -4853,7 +4877,6 @@ static long ext4_zero_range(struct file *file, loff_t offset, if ((offset + len) > i_size_read(inode)) ext4_set_inode_flag(inode, EXT4_INODE_EOFBLOCKS); } - ext4_mark_inode_dirty(handle, inode); /* Zero out partial block at the edges of the range */ @@ -4880,13 +4903,11 @@ out_mutex: long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len) { struct inode *inode = file_inode(file); - handle_t *handle; loff_t new_size = 0; unsigned int max_blocks; int ret = 0; int flags; ext4_lblk_t lblk; - struct timespec tv; unsigned int blkbits = inode->i_blkbits; /* Return error if mode is not supported */ @@ -4937,36 +4958,15 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len) goto out; } - ret = ext4_alloc_file_blocks(file, lblk, max_blocks, flags, mode); + ret = ext4_alloc_file_blocks(file, lblk, max_blocks, new_size, + flags, mode); if (ret) goto out; - handle = ext4_journal_start(inode, EXT4_HT_INODE, 2); - if (IS_ERR(handle)) - goto out; - - tv = inode->i_ctime = ext4_current_time(inode); - - if (new_size) { - if (new_size > i_size_read(inode)) { - i_size_write(inode, new_size); - inode->i_mtime = tv; - } - if (new_size > EXT4_I(inode)->i_disksize) - ext4_update_i_disksize(inode, new_size); - } else { - /* - * Mark that we allocate beyond EOF so the subsequent truncate - * can proceed even if the new size is the same as i_size. - */ - if ((offset + len) > i_size_read(inode)) - ext4_set_inode_flag(inode, EXT4_INODE_EOFBLOCKS); + if (file->f_flags & O_SYNC && EXT4_SB(inode->i_sb)->s_journal) { + ret = jbd2_complete_transaction(EXT4_SB(inode->i_sb)->s_journal, + EXT4_I(inode)->i_sync_tid); } - ext4_mark_inode_dirty(handle, inode); - if (file->f_flags & O_SYNC) - ext4_handle_sync(handle); - - ext4_journal_stop(handle); out: mutex_unlock(&inode->i_mutex); trace_ext4_fallocate_exit(inode, offset, max_blocks, ret); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 367a60c..3aa26e9 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1055,27 +1055,11 @@ static int ext4_write_end(struct file *file, } else copied = block_write_end(file, mapping, pos, len, copied, page, fsdata); - /* - * No need to use i_size_read() here, the i_size - * cannot change under us because we hole i_mutex. - * - * But it's important to update i_size while still holding page lock: + * it's important to update i_size while still holding page lock: * page writeout could otherwise come in and zero beyond i_size. */ - if (pos + copied > inode->i_size) { - i_size_write(inode, pos + copied); - i_size_changed = 1; - } - - if (pos + copied > EXT4_I(inode)->i_disksize) { - /* We need to mark inode dirty even if - * new_i_size is less that inode->i_size - * but greater than i_disksize. (hint delalloc) - */ - ext4_update_i_disksize(inode, (pos + copied)); - i_size_changed = 1; - } + i_size_changed = ext4_update_inode_size(inode, pos + copied); unlock_page(page); page_cache_release(page); @@ -1123,7 +1107,7 @@ static int ext4_journalled_write_end(struct file *file, int ret = 0, ret2; int partial = 0; unsigned from, to; - loff_t new_i_size; + int size_changed = 0; trace_ext4_journalled_write_end(inode, pos, len, copied); from = pos & (PAGE_CACHE_SIZE - 1); @@ -1146,20 +1130,18 @@ static int ext4_journalled_write_end(struct file *file, if (!partial) SetPageUptodate(page); } - new_i_size = pos + copied; - if (new_i_size > inode->i_size) - i_size_write(inode, pos+copied); + size_changed = ext4_update_inode_size(inode, pos + copied); ext4_set_inode_state(inode, EXT4_STATE_JDATA); EXT4_I(inode)->i_datasync_tid = handle->h_transaction->t_tid; - if (new_i_size > EXT4_I(inode)->i_disksize) { - ext4_update_i_disksize(inode, new_i_size); + unlock_page(page); + page_cache_release(page); + + if (size_changed) { ret2 = ext4_mark_inode_dirty(handle, inode); if (!ret) ret = ret2; } - unlock_page(page); - page_cache_release(page); if (pos + len > inode->i_size && ext4_can_truncate(inode)) /* if we have allocated more blocks and copied * less. We will have blocks allocated outside @@ -2095,6 +2077,7 @@ static int mpage_map_and_submit_extent(handle_t *handle, struct ext4_map_blocks *map = &mpd->map; int err; loff_t disksize; + int progress = 0; mpd->io_submit.io_end->offset = ((loff_t)map->m_lblk) << inode->i_blkbits; @@ -2111,8 +2094,11 @@ static int mpage_map_and_submit_extent(handle_t *handle, * is non-zero, a commit should free up blocks. */ if ((err == -ENOMEM) || - (err == -ENOSPC && ext4_count_free_clusters(sb))) + (err == -ENOSPC && ext4_count_free_clusters(sb))) { + if (progress) + goto update_disksize; return err; + } ext4_msg(sb, KERN_CRIT, "Delayed block allocation failed for " "inode %lu at logical offset %llu with" @@ -2129,15 +2115,17 @@ static int mpage_map_and_submit_extent(handle_t *handle, *give_up_on_write = true; return err; } + progress = 1; /* * Update buffer state, submit mapped pages, and get us new * extent to map */ err = mpage_map_and_submit_buffers(mpd); if (err < 0) - return err; + goto update_disksize; } while (map->m_len); +update_disksize: /* * Update on-disk size after IO is submitted. Races with * truncate are avoided by checking i_size under i_data_sem. diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 9560277..8b0f9ef 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -1412,6 +1412,8 @@ static void mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b, int last = first + count - 1; struct super_block *sb = e4b->bd_sb; + if (WARN_ON(count == 0)) + return; BUG_ON(last >= (sb->s_blocksize << 3)); assert_spin_locked(ext4_group_lock_ptr(sb, e4b->bd_group)); /* Don't bother if the block group is corrupt. */ @@ -3221,6 +3223,8 @@ static void ext4_discard_allocated_blocks(struct ext4_allocation_context *ac) int err; if (pa == NULL) { + if (ac->ac_f_ex.fe_len == 0) + return; err = ext4_mb_load_buddy(ac->ac_sb, ac->ac_f_ex.fe_group, &e4b); if (err) { /* @@ -3235,6 +3239,7 @@ static void ext4_discard_allocated_blocks(struct ext4_allocation_context *ac) mb_free_blocks(ac->ac_inode, &e4b, ac->ac_f_ex.fe_start, ac->ac_f_ex.fe_len); ext4_unlock_group(ac->ac_sb, ac->ac_f_ex.fe_group); + ext4_mb_unload_buddy(&e4b); return; } if (pa->pa_type == MB_INODE_PA) diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index b147a67..603e4eb 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -1227,7 +1227,7 @@ static struct buffer_head * ext4_find_entry (struct inode *dir, buffer */ int num = 0; ext4_lblk_t nblocks; - int i, err; + int i, err = 0; int namelen; *res_dir = NULL; @@ -1264,7 +1264,11 @@ static struct buffer_head * ext4_find_entry (struct inode *dir, * return. Otherwise, fall back to doing a search the * old fashioned way. */ - if (bh || (err != ERR_BAD_DX_DIR)) + if (err == -ENOENT) + return NULL; + if (err && err != ERR_BAD_DX_DIR) + return ERR_PTR(err); + if (bh) return bh; dxtrace(printk(KERN_DEBUG "ext4_find_entry: dx failed, " "falling back\n")); @@ -1295,6 +1299,11 @@ restart: } num++; bh = ext4_getblk(NULL, dir, b++, 0, &err); + if (unlikely(err)) { + if (ra_max == 0) + return ERR_PTR(err); + break; + } bh_use[ra_max] = bh; if (bh) ll_rw_block(READ | REQ_META | REQ_PRIO, @@ -1417,6 +1426,8 @@ static struct dentry *ext4_lookup(struct inode *dir, struct dentry *dentry, unsi return ERR_PTR(-ENAMETOOLONG); bh = ext4_find_entry(dir, &dentry->d_name, &de, NULL); + if (IS_ERR(bh)) + return (struct dentry *) bh; inode = NULL; if (bh) { __u32 ino = le32_to_cpu(de->inode); @@ -1450,6 +1461,8 @@ struct dentry *ext4_get_parent(struct dentry *child) struct buffer_head *bh; bh = ext4_find_entry(child->d_inode, &dotdot, &de, NULL); + if (IS_ERR(bh)) + return (struct dentry *) bh; if (!bh) return ERR_PTR(-ENOENT); ino = le32_to_cpu(de->inode); @@ -2727,6 +2740,8 @@ static int ext4_rmdir(struct inode *dir, struct dentry *dentry) retval = -ENOENT; bh = ext4_find_entry(dir, &dentry->d_name, &de, NULL); + if (IS_ERR(bh)) + return PTR_ERR(bh); if (!bh) goto end_rmdir; @@ -2794,6 +2809,8 @@ static int ext4_unlink(struct inode *dir, struct dentry *dentry) retval = -ENOENT; bh = ext4_find_entry(dir, &dentry->d_name, &de, NULL); + if (IS_ERR(bh)) + return PTR_ERR(bh); if (!bh) goto end_unlink; @@ -3121,6 +3138,8 @@ static int ext4_find_delete_entry(handle_t *handle, struct inode *dir, struct ext4_dir_entry_2 *de; bh = ext4_find_entry(dir, d_name, &de, NULL); + if (IS_ERR(bh)) + return PTR_ERR(bh); if (bh) { retval = ext4_delete_entry(handle, dir, de, bh); brelse(bh); @@ -3128,7 +3147,8 @@ static int ext4_find_delete_entry(handle_t *handle, struct inode *dir, return retval; } -static void ext4_rename_delete(handle_t *handle, struct ext4_renament *ent) +static void ext4_rename_delete(handle_t *handle, struct ext4_renament *ent, + int force_reread) { int retval; /* @@ -3140,7 +3160,8 @@ static void ext4_rename_delete(handle_t *handle, struct ext4_renament *ent) if (le32_to_cpu(ent->de->inode) != ent->inode->i_ino || ent->de->name_len != ent->dentry->d_name.len || strncmp(ent->de->name, ent->dentry->d_name.name, - ent->de->name_len)) { + ent->de->name_len) || + force_reread) { retval = ext4_find_delete_entry(handle, ent->dir, &ent->dentry->d_name); } else { @@ -3191,6 +3212,7 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, .dentry = new_dentry, .inode = new_dentry->d_inode, }; + int force_reread; int retval; dquot_initialize(old.dir); @@ -3202,6 +3224,8 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, dquot_initialize(new.inode); old.bh = ext4_find_entry(old.dir, &old.dentry->d_name, &old.de, NULL); + if (IS_ERR(old.bh)) + return PTR_ERR(old.bh); /* * Check for inode number is _not_ due to possible IO errors. * We might rmdir the source, keep it as pwd of some process @@ -3214,6 +3238,11 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, new.bh = ext4_find_entry(new.dir, &new.dentry->d_name, &new.de, &new.inlined); + if (IS_ERR(new.bh)) { + retval = PTR_ERR(new.bh); + new.bh = NULL; + goto end_rename; + } if (new.bh) { if (!new.inode) { brelse(new.bh); @@ -3246,6 +3275,15 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, if (retval) goto end_rename; } + /* + * If we're renaming a file within an inline_data dir and adding or + * setting the new dirent causes a conversion from inline_data to + * extents/blockmap, we need to force the dirent delete code to + * re-read the directory, or else we end up trying to delete a dirent + * from what is now the extent tree root (or a block map). + */ + force_reread = (new.dir->i_ino == old.dir->i_ino && + ext4_test_inode_flag(new.dir, EXT4_INODE_INLINE_DATA)); if (!new.bh) { retval = ext4_add_entry(handle, new.dentry, old.inode); if (retval) @@ -3256,6 +3294,9 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, if (retval) goto end_rename; } + if (force_reread) + force_reread = !ext4_test_inode_flag(new.dir, + EXT4_INODE_INLINE_DATA); /* * Like most other Unix systems, set the ctime for inodes on a @@ -3267,7 +3308,7 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, /* * ok, that's it */ - ext4_rename_delete(handle, &old); + ext4_rename_delete(handle, &old, force_reread); if (new.inode) { ext4_dec_count(handle, new.inode); @@ -3330,6 +3371,8 @@ static int ext4_cross_rename(struct inode *old_dir, struct dentry *old_dentry, old.bh = ext4_find_entry(old.dir, &old.dentry->d_name, &old.de, &old.inlined); + if (IS_ERR(old.bh)) + return PTR_ERR(old.bh); /* * Check for inode number is _not_ due to possible IO errors. * We might rmdir the source, keep it as pwd of some process @@ -3342,6 +3385,11 @@ static int ext4_cross_rename(struct inode *old_dir, struct dentry *old_dentry, new.bh = ext4_find_entry(new.dir, &new.dentry->d_name, &new.de, &new.inlined); + if (IS_ERR(new.bh)) { + retval = PTR_ERR(new.bh); + new.bh = NULL; + goto end_rename; + } /* RENAME_EXCHANGE case: old *and* new must both exist */ if (!new.bh || le32_to_cpu(new.de->inode) != new.inode->i_ino) diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index bb0e80f..1e43b90 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -575,6 +575,7 @@ handle_bb: bh = bclean(handle, sb, block); if (IS_ERR(bh)) { err = PTR_ERR(bh); + bh = NULL; goto out; } overhead = ext4_group_overhead_blocks(sb, group); @@ -603,6 +604,7 @@ handle_ib: bh = bclean(handle, sb, block); if (IS_ERR(bh)) { err = PTR_ERR(bh); + bh = NULL; goto out; } diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 32b43ad..0b28b36 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -3181,9 +3181,9 @@ static int set_journal_csum_feature_set(struct super_block *sb) if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) { - /* journal checksum v2 */ + /* journal checksum v3 */ compat = 0; - incompat = JBD2_FEATURE_INCOMPAT_CSUM_V2; + incompat = JBD2_FEATURE_INCOMPAT_CSUM_V3; } else { /* journal checksum v1 */ compat = JBD2_FEATURE_COMPAT_CHECKSUM; @@ -3205,6 +3205,7 @@ static int set_journal_csum_feature_set(struct super_block *sb) jbd2_journal_clear_features(sbi->s_journal, JBD2_FEATURE_COMPAT_CHECKSUM, 0, JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT | + JBD2_FEATURE_INCOMPAT_CSUM_V3 | JBD2_FEATURE_INCOMPAT_CSUM_V2); } diff --git a/fs/f2fs/Kconfig b/fs/f2fs/Kconfig index 214fe10..736a348 100644 --- a/fs/f2fs/Kconfig +++ b/fs/f2fs/Kconfig @@ -23,7 +23,7 @@ config F2FS_STAT_FS mounted as f2fs. Each file shows the whole f2fs information. /sys/kernel/debug/f2fs/status includes: - - major file system information managed by f2fs currently + - major filesystem information managed by f2fs currently - average SIT information about whole segments - current memory footprint consumed by f2fs. @@ -68,6 +68,6 @@ config F2FS_CHECK_FS bool "F2FS consistency checking feature" depends on F2FS_FS help - Enables BUG_ONs which check the file system consistency in runtime. + Enables BUG_ONs which check the filesystem consistency in runtime. If you want to improve the performance, say N. diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c index 6aeed5b..ec3b7a5 100644 --- a/fs/f2fs/checkpoint.c +++ b/fs/f2fs/checkpoint.c @@ -160,14 +160,11 @@ static int f2fs_write_meta_page(struct page *page, goto redirty_out; if (wbc->for_reclaim) goto redirty_out; - - /* Should not write any meta pages, if any IO error was occurred */ - if (unlikely(is_set_ckpt_flags(F2FS_CKPT(sbi), CP_ERROR_FLAG))) - goto no_write; + if (unlikely(f2fs_cp_error(sbi))) + goto redirty_out; f2fs_wait_on_page_writeback(page, META); write_meta_page(sbi, page); -no_write: dec_page_count(sbi, F2FS_DIRTY_META); unlock_page(page); return 0; @@ -348,7 +345,7 @@ bool exist_written_data(struct f2fs_sb_info *sbi, nid_t ino, int mode) return e ? true : false; } -static void release_dirty_inode(struct f2fs_sb_info *sbi) +void release_dirty_inode(struct f2fs_sb_info *sbi) { struct ino_entry *e, *tmp; int i; @@ -446,8 +443,8 @@ static void write_orphan_inodes(struct f2fs_sb_info *sbi, block_t start_blk) struct f2fs_orphan_block *orphan_blk = NULL; unsigned int nentries = 0; unsigned short index; - unsigned short orphan_blocks = (unsigned short)((sbi->n_orphans + - (F2FS_ORPHANS_PER_BLOCK - 1)) / F2FS_ORPHANS_PER_BLOCK); + unsigned short orphan_blocks = + (unsigned short)GET_ORPHAN_BLOCKS(sbi->n_orphans); struct page *page = NULL; struct ino_entry *orphan = NULL; @@ -737,7 +734,7 @@ retry: /* * Freeze all the FS-operations for checkpoint. */ -static void block_operations(struct f2fs_sb_info *sbi) +static int block_operations(struct f2fs_sb_info *sbi) { struct writeback_control wbc = { .sync_mode = WB_SYNC_ALL, @@ -745,6 +742,7 @@ static void block_operations(struct f2fs_sb_info *sbi) .for_reclaim = 0, }; struct blk_plug plug; + int err = 0; blk_start_plug(&plug); @@ -754,11 +752,15 @@ retry_flush_dents: if (get_pages(sbi, F2FS_DIRTY_DENTS)) { f2fs_unlock_all(sbi); sync_dirty_dir_inodes(sbi); + if (unlikely(f2fs_cp_error(sbi))) { + err = -EIO; + goto out; + } goto retry_flush_dents; } /* - * POR: we should ensure that there is no dirty node pages + * POR: we should ensure that there are no dirty node pages * until finishing nat/sit flush. */ retry_flush_nodes: @@ -767,9 +769,16 @@ retry_flush_nodes: if (get_pages(sbi, F2FS_DIRTY_NODES)) { up_write(&sbi->node_write); sync_node_pages(sbi, 0, &wbc); + if (unlikely(f2fs_cp_error(sbi))) { + f2fs_unlock_all(sbi); + err = -EIO; + goto out; + } goto retry_flush_nodes; } +out: blk_finish_plug(&plug); + return err; } static void unblock_operations(struct f2fs_sb_info *sbi) @@ -813,8 +822,11 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount) discard_next_dnode(sbi, NEXT_FREE_BLKADDR(sbi, curseg)); /* Flush all the NAT/SIT pages */ - while (get_pages(sbi, F2FS_DIRTY_META)) + while (get_pages(sbi, F2FS_DIRTY_META)) { sync_meta_pages(sbi, META, LONG_MAX); + if (unlikely(f2fs_cp_error(sbi))) + return; + } next_free_nid(sbi, &last_nid); @@ -825,7 +837,7 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount) ckpt->elapsed_time = cpu_to_le64(get_mtime(sbi)); ckpt->valid_block_count = cpu_to_le64(valid_user_blocks(sbi)); ckpt->free_segment_count = cpu_to_le32(free_segments(sbi)); - for (i = 0; i < 3; i++) { + for (i = 0; i < NR_CURSEG_NODE_TYPE; i++) { ckpt->cur_node_segno[i] = cpu_to_le32(curseg_segno(sbi, i + CURSEG_HOT_NODE)); ckpt->cur_node_blkoff[i] = @@ -833,7 +845,7 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount) ckpt->alloc_type[i + CURSEG_HOT_NODE] = curseg_alloc_type(sbi, i + CURSEG_HOT_NODE); } - for (i = 0; i < 3; i++) { + for (i = 0; i < NR_CURSEG_DATA_TYPE; i++) { ckpt->cur_data_segno[i] = cpu_to_le32(curseg_segno(sbi, i + CURSEG_HOT_DATA)); ckpt->cur_data_blkoff[i] = @@ -848,24 +860,23 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount) /* 2 cp + n data seg summary + orphan inode blocks */ data_sum_blocks = npages_for_summary_flush(sbi); - if (data_sum_blocks < 3) + if (data_sum_blocks < NR_CURSEG_DATA_TYPE) set_ckpt_flags(ckpt, CP_COMPACT_SUM_FLAG); else clear_ckpt_flags(ckpt, CP_COMPACT_SUM_FLAG); - orphan_blocks = (sbi->n_orphans + F2FS_ORPHANS_PER_BLOCK - 1) - / F2FS_ORPHANS_PER_BLOCK; + orphan_blocks = GET_ORPHAN_BLOCKS(sbi->n_orphans); ckpt->cp_pack_start_sum = cpu_to_le32(1 + cp_payload_blks + orphan_blocks); if (is_umount) { set_ckpt_flags(ckpt, CP_UMOUNT_FLAG); - ckpt->cp_pack_total_block_count = cpu_to_le32(2 + + ckpt->cp_pack_total_block_count = cpu_to_le32(F2FS_CP_PACKS+ cp_payload_blks + data_sum_blocks + orphan_blocks + NR_CURSEG_NODE_TYPE); } else { clear_ckpt_flags(ckpt, CP_UMOUNT_FLAG); - ckpt->cp_pack_total_block_count = cpu_to_le32(2 + + ckpt->cp_pack_total_block_count = cpu_to_le32(F2FS_CP_PACKS + cp_payload_blks + data_sum_blocks + orphan_blocks); } @@ -924,6 +935,9 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount) /* wait for previous submitted node/meta pages writeback */ wait_on_all_pages_writeback(sbi); + if (unlikely(f2fs_cp_error(sbi))) + return; + filemap_fdatawait_range(NODE_MAPPING(sbi), 0, LONG_MAX); filemap_fdatawait_range(META_MAPPING(sbi), 0, LONG_MAX); @@ -934,15 +948,17 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount) /* Here, we only have one bio having CP pack */ sync_meta_pages(sbi, META_FLUSH, LONG_MAX); - if (!is_set_ckpt_flags(ckpt, CP_ERROR_FLAG)) { - clear_prefree_segments(sbi); - release_dirty_inode(sbi); - F2FS_RESET_SB_DIRT(sbi); - } + release_dirty_inode(sbi); + + if (unlikely(f2fs_cp_error(sbi))) + return; + + clear_prefree_segments(sbi); + F2FS_RESET_SB_DIRT(sbi); } /* - * We guarantee that this checkpoint procedure should not fail. + * We guarantee that this checkpoint procedure will not fail. */ void write_checkpoint(struct f2fs_sb_info *sbi, bool is_umount) { @@ -952,7 +968,13 @@ void write_checkpoint(struct f2fs_sb_info *sbi, bool is_umount) trace_f2fs_write_checkpoint(sbi->sb, is_umount, "start block_ops"); mutex_lock(&sbi->cp_mutex); - block_operations(sbi); + + if (!sbi->s_dirty) + goto out; + if (unlikely(f2fs_cp_error(sbi))) + goto out; + if (block_operations(sbi)) + goto out; trace_f2fs_write_checkpoint(sbi->sb, is_umount, "finish block_ops"); @@ -976,9 +998,9 @@ void write_checkpoint(struct f2fs_sb_info *sbi, bool is_umount) do_checkpoint(sbi, is_umount); unblock_operations(sbi); - mutex_unlock(&sbi->cp_mutex); - stat_inc_cp_count(sbi->stat_info); +out: + mutex_unlock(&sbi->cp_mutex); trace_f2fs_write_checkpoint(sbi->sb, is_umount, "finish checkpoint"); } @@ -999,8 +1021,8 @@ void init_ino_entry_info(struct f2fs_sb_info *sbi) * for cp pack we can have max 1020*504 orphan entries */ sbi->n_orphans = 0; - sbi->max_orphans = (sbi->blocks_per_seg - 2 - NR_CURSEG_TYPE) - * F2FS_ORPHANS_PER_BLOCK; + sbi->max_orphans = (sbi->blocks_per_seg - F2FS_CP_PACKS - + NR_CURSEG_TYPE) * F2FS_ORPHANS_PER_BLOCK; } int __init create_checkpoint_caches(void) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 03313099..76de83e 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -53,7 +53,7 @@ static void f2fs_write_end_io(struct bio *bio, int err) struct page *page = bvec->bv_page; if (unlikely(err)) { - SetPageError(page); + set_page_dirty(page); set_bit(AS_EIO, &page->mapping->flags); f2fs_stop_checkpoint(sbi); } @@ -691,7 +691,7 @@ get_next: allocated = true; blkaddr = dn.data_blkaddr; } - /* Give more consecutive addresses for the read ahead */ + /* Give more consecutive addresses for the readahead */ if (blkaddr == (bh_result->b_blocknr + ofs)) { ofs++; dn.ofs_in_node++; @@ -739,7 +739,7 @@ static int f2fs_read_data_page(struct file *file, struct page *page) trace_f2fs_readpage(page, DATA); - /* If the file has inline data, try to read it directlly */ + /* If the file has inline data, try to read it directly */ if (f2fs_has_inline_data(inode)) ret = f2fs_read_inline_data(inode, page); else @@ -836,10 +836,19 @@ write: /* Dentry blocks are controlled by checkpoint */ if (S_ISDIR(inode->i_mode)) { + if (unlikely(f2fs_cp_error(sbi))) + goto redirty_out; err = do_write_data_page(page, &fio); goto done; } + /* we should bypass data pages to proceed the kworkder jobs */ + if (unlikely(f2fs_cp_error(sbi))) { + SetPageError(page); + unlock_page(page); + return 0; + } + if (!wbc->for_reclaim) need_balance_fs = true; else if (has_not_enough_free_secs(sbi, 0)) @@ -927,7 +936,7 @@ static void f2fs_write_failed(struct address_space *mapping, loff_t to) if (to > inode->i_size) { truncate_pagecache(inode, inode->i_size); - truncate_blocks(inode, inode->i_size); + truncate_blocks(inode, inode->i_size, true); } } @@ -946,7 +955,7 @@ static int f2fs_write_begin(struct file *file, struct address_space *mapping, f2fs_balance_fs(sbi); repeat: - err = f2fs_convert_inline_data(inode, pos + len); + err = f2fs_convert_inline_data(inode, pos + len, NULL); if (err) goto fail; diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c index a441ba3..fecebdb 100644 --- a/fs/f2fs/debug.c +++ b/fs/f2fs/debug.c @@ -32,7 +32,7 @@ static void update_general_status(struct f2fs_sb_info *sbi) struct f2fs_stat_info *si = F2FS_STAT(sbi); int i; - /* valid check of the segment numbers */ + /* validation check of the segment numbers */ si->hit_ext = sbi->read_hit_ext; si->total_ext = sbi->total_hit_ext; si->ndirty_node = get_pages(sbi, F2FS_DIRTY_NODES); @@ -152,7 +152,7 @@ static void update_mem_info(struct f2fs_sb_info *sbi) si->base_mem += NR_DIRTY_TYPE * f2fs_bitmap_size(TOTAL_SEGS(sbi)); si->base_mem += f2fs_bitmap_size(TOTAL_SECS(sbi)); - /* buld nm */ + /* build nm */ si->base_mem += sizeof(struct f2fs_nm_info); si->base_mem += __bitmap_size(sbi, NAT_BITMAP); diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c index bcf893c..155fb05 100644 --- a/fs/f2fs/dir.c +++ b/fs/f2fs/dir.c @@ -124,7 +124,7 @@ static struct f2fs_dir_entry *find_in_block(struct page *dentry_page, /* * For the most part, it should be a bug when name_len is zero. - * We stop here for figuring out where the bugs are occurred. + * We stop here for figuring out where the bugs has occurred. */ f2fs_bug_on(!de->name_len); @@ -391,7 +391,7 @@ put_error: error: /* once the failed inode becomes a bad inode, i_mode is S_IFREG */ truncate_inode_pages(&inode->i_data, 0); - truncate_blocks(inode, 0); + truncate_blocks(inode, 0, false); remove_dirty_dir_inode(inode); remove_inode_page(inode); return ERR_PTR(err); @@ -563,7 +563,7 @@ fail: } /* - * It only removes the dentry from the dentry page,corresponding name + * It only removes the dentry from the dentry page, corresponding name * entry in name page does not need to be touched during deletion. */ void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page, diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 4dab533..e921242 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -24,7 +24,7 @@ #define f2fs_bug_on(condition) BUG_ON(condition) #define f2fs_down_write(x, y) down_write_nest_lock(x, y) #else -#define f2fs_bug_on(condition) +#define f2fs_bug_on(condition) WARN_ON(condition) #define f2fs_down_write(x, y) down_write(x) #endif @@ -395,7 +395,7 @@ enum count_type { }; /* - * The below are the page types of bios used in submti_bio(). + * The below are the page types of bios used in submit_bio(). * The available types are: * DATA User data pages. It operates as async mode. * NODE Node pages. It operates as async mode. @@ -470,7 +470,7 @@ struct f2fs_sb_info { struct list_head dir_inode_list; /* dir inode list */ spinlock_t dir_inode_lock; /* for dir inode list lock */ - /* basic file system units */ + /* basic filesystem units */ unsigned int log_sectors_per_block; /* log2 sectors per block */ unsigned int log_blocksize; /* log2 block size */ unsigned int blocksize; /* block size */ @@ -799,7 +799,7 @@ static inline block_t __start_cp_addr(struct f2fs_sb_info *sbi) /* * odd numbered checkpoint should at cp segment 0 - * and even segent must be at cp segment 1 + * and even segment must be at cp segment 1 */ if (!(ckpt_version & 1)) start_addr += sbi->blocks_per_seg; @@ -1096,6 +1096,11 @@ static inline int f2fs_readonly(struct super_block *sb) return sb->s_flags & MS_RDONLY; } +static inline bool f2fs_cp_error(struct f2fs_sb_info *sbi) +{ + return is_set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG); +} + static inline void f2fs_stop_checkpoint(struct f2fs_sb_info *sbi) { set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG); @@ -1117,7 +1122,7 @@ static inline void f2fs_stop_checkpoint(struct f2fs_sb_info *sbi) */ int f2fs_sync_file(struct file *, loff_t, loff_t, int); void truncate_data_blocks(struct dnode_of_data *); -int truncate_blocks(struct inode *, u64); +int truncate_blocks(struct inode *, u64, bool); void f2fs_truncate(struct inode *); int f2fs_getattr(struct vfsmount *, struct dentry *, struct kstat *); int f2fs_setattr(struct dentry *, struct iattr *); @@ -1202,10 +1207,8 @@ int sync_node_pages(struct f2fs_sb_info *, nid_t, struct writeback_control *); bool alloc_nid(struct f2fs_sb_info *, nid_t *); void alloc_nid_done(struct f2fs_sb_info *, nid_t); void alloc_nid_failed(struct f2fs_sb_info *, nid_t); -void recover_node_page(struct f2fs_sb_info *, struct page *, - struct f2fs_summary *, struct node_info *, block_t); void recover_inline_xattr(struct inode *, struct page *); -bool recover_xattr_data(struct inode *, struct page *, block_t); +void recover_xattr_data(struct inode *, struct page *, block_t); int recover_inode_page(struct f2fs_sb_info *, struct page *); int restore_node_summary(struct f2fs_sb_info *, unsigned int, struct f2fs_summary_block *); @@ -1238,8 +1241,6 @@ void write_data_page(struct page *, struct dnode_of_data *, block_t *, void rewrite_data_page(struct page *, block_t, struct f2fs_io_info *); void recover_data_page(struct f2fs_sb_info *, struct page *, struct f2fs_summary *, block_t, block_t); -void rewrite_node_page(struct f2fs_sb_info *, struct page *, - struct f2fs_summary *, block_t, block_t); void allocate_data_block(struct f2fs_sb_info *, struct page *, block_t, block_t *, struct f2fs_summary *, int); void f2fs_wait_on_page_writeback(struct page *, enum page_type); @@ -1262,6 +1263,7 @@ int ra_meta_pages(struct f2fs_sb_info *, int, int, int); long sync_meta_pages(struct f2fs_sb_info *, enum page_type, long); void add_dirty_inode(struct f2fs_sb_info *, nid_t, int type); void remove_dirty_inode(struct f2fs_sb_info *, nid_t, int type); +void release_dirty_inode(struct f2fs_sb_info *); bool exist_written_data(struct f2fs_sb_info *, nid_t, int); int acquire_orphan_inode(struct f2fs_sb_info *); void release_orphan_inode(struct f2fs_sb_info *); @@ -1439,8 +1441,8 @@ extern const struct inode_operations f2fs_special_inode_operations; */ bool f2fs_may_inline(struct inode *); int f2fs_read_inline_data(struct inode *, struct page *); -int f2fs_convert_inline_data(struct inode *, pgoff_t); +int f2fs_convert_inline_data(struct inode *, pgoff_t, struct page *); int f2fs_write_inline_data(struct inode *, struct page *, unsigned int); void truncate_inline_data(struct inode *, u64); -int recover_inline_data(struct inode *, struct page *); +bool recover_inline_data(struct inode *, struct page *); #endif diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 208f1a9..060aee6 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -41,6 +41,11 @@ static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma, sb_start_pagefault(inode->i_sb); + /* force to convert with normal data indices */ + err = f2fs_convert_inline_data(inode, MAX_INLINE_DATA + 1, page); + if (err) + goto out; + /* block allocation */ f2fs_lock_op(sbi); set_new_dnode(&dn, inode, NULL, NULL, 0); @@ -110,6 +115,25 @@ static int get_parent_ino(struct inode *inode, nid_t *pino) return 1; } +static inline bool need_do_checkpoint(struct inode *inode) +{ + struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); + bool need_cp = false; + + if (!S_ISREG(inode->i_mode) || inode->i_nlink != 1) + need_cp = true; + else if (file_wrong_pino(inode)) + need_cp = true; + else if (!space_for_roll_forward(sbi)) + need_cp = true; + else if (!is_checkpointed_node(sbi, F2FS_I(inode)->i_pino)) + need_cp = true; + else if (F2FS_I(inode)->xattr_ver == cur_cp_version(F2FS_CKPT(sbi))) + need_cp = true; + + return need_cp; +} + int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) { struct inode *inode = file->f_mapping->host; @@ -154,23 +178,12 @@ int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) /* guarantee free sections for fsync */ f2fs_balance_fs(sbi); - down_read(&fi->i_sem); - /* * Both of fdatasync() and fsync() are able to be recovered from * sudden-power-off. */ - if (!S_ISREG(inode->i_mode) || inode->i_nlink != 1) - need_cp = true; - else if (file_wrong_pino(inode)) - need_cp = true; - else if (!space_for_roll_forward(sbi)) - need_cp = true; - else if (!is_checkpointed_node(sbi, F2FS_I(inode)->i_pino)) - need_cp = true; - else if (F2FS_I(inode)->xattr_ver == cur_cp_version(F2FS_CKPT(sbi))) - need_cp = true; - + down_read(&fi->i_sem); + need_cp = need_do_checkpoint(inode); up_read(&fi->i_sem); if (need_cp) { @@ -288,7 +301,7 @@ static loff_t f2fs_seek_block(struct file *file, loff_t offset, int whence) if (err && err != -ENOENT) { goto fail; } else if (err == -ENOENT) { - /* direct node is not exist */ + /* direct node does not exists */ if (whence == SEEK_DATA) { pgofs = PGOFS_OF_NEXT_DNODE(pgofs, F2FS_I(inode)); @@ -417,7 +430,7 @@ out: f2fs_put_page(page, 1); } -int truncate_blocks(struct inode *inode, u64 from) +int truncate_blocks(struct inode *inode, u64 from, bool lock) { struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); unsigned int blocksize = inode->i_sb->s_blocksize; @@ -433,14 +446,16 @@ int truncate_blocks(struct inode *inode, u64 from) free_from = (pgoff_t) ((from + blocksize - 1) >> (sbi->log_blocksize)); - f2fs_lock_op(sbi); + if (lock) + f2fs_lock_op(sbi); set_new_dnode(&dn, inode, NULL, NULL, 0); err = get_dnode_of_data(&dn, free_from, LOOKUP_NODE); if (err) { if (err == -ENOENT) goto free_next; - f2fs_unlock_op(sbi); + if (lock) + f2fs_unlock_op(sbi); trace_f2fs_truncate_blocks_exit(inode, err); return err; } @@ -458,7 +473,8 @@ int truncate_blocks(struct inode *inode, u64 from) f2fs_put_dnode(&dn); free_next: err = truncate_inode_blocks(inode, free_from); - f2fs_unlock_op(sbi); + if (lock) + f2fs_unlock_op(sbi); done: /* lastly zero out the first data page */ truncate_partial_data_page(inode, from); @@ -475,7 +491,7 @@ void f2fs_truncate(struct inode *inode) trace_f2fs_truncate(inode); - if (!truncate_blocks(inode, i_size_read(inode))) { + if (!truncate_blocks(inode, i_size_read(inode), true)) { inode->i_mtime = inode->i_ctime = CURRENT_TIME; mark_inode_dirty(inode); } @@ -533,7 +549,7 @@ int f2fs_setattr(struct dentry *dentry, struct iattr *attr) if ((attr->ia_valid & ATTR_SIZE) && attr->ia_size != i_size_read(inode)) { - err = f2fs_convert_inline_data(inode, attr->ia_size); + err = f2fs_convert_inline_data(inode, attr->ia_size, NULL); if (err) return err; @@ -622,7 +638,7 @@ static int punch_hole(struct inode *inode, loff_t offset, loff_t len) loff_t off_start, off_end; int ret = 0; - ret = f2fs_convert_inline_data(inode, MAX_INLINE_DATA + 1); + ret = f2fs_convert_inline_data(inode, MAX_INLINE_DATA + 1, NULL); if (ret) return ret; @@ -678,7 +694,7 @@ static int expand_inode_data(struct inode *inode, loff_t offset, if (ret) return ret; - ret = f2fs_convert_inline_data(inode, offset + len); + ret = f2fs_convert_inline_data(inode, offset + len, NULL); if (ret) return ret; diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index d7947d9..943a31d 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -58,7 +58,7 @@ static int gc_thread_func(void *data) * 3. IO subsystem is idle by checking the # of requests in * bdev's request list. * - * Note) We have to avoid triggering GCs too much frequently. + * Note) We have to avoid triggering GCs frequently. * Because it is possible that some segments can be * invalidated soon after by user update or deletion. * So, I'd like to wait some time to collect dirty segments. @@ -222,7 +222,7 @@ static unsigned int get_cb_cost(struct f2fs_sb_info *sbi, unsigned int segno) u = (vblocks * 100) >> sbi->log_blocks_per_seg; - /* Handle if the system time is changed by user */ + /* Handle if the system time has changed by the user */ if (mtime < sit_i->min_mtime) sit_i->min_mtime = mtime; if (mtime > sit_i->max_mtime) @@ -593,7 +593,7 @@ next_step: if (phase == 2) { inode = f2fs_iget(sb, dni.ino); - if (IS_ERR(inode)) + if (IS_ERR(inode) || is_bad_inode(inode)) continue; start_bidx = start_bidx_of_node(nofs, F2FS_I(inode)); @@ -693,7 +693,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi) gc_more: if (unlikely(!(sbi->sb->s_flags & MS_ACTIVE))) goto stop; - if (unlikely(is_set_ckpt_flags(F2FS_CKPT(sbi), CP_ERROR_FLAG))) + if (unlikely(f2fs_cp_error(sbi))) goto stop; if (gc_type == BG_GC && has_not_enough_free_secs(sbi, nfree)) { diff --git a/fs/f2fs/gc.h b/fs/f2fs/gc.h index 5d5eb60..16f0b2b 100644 --- a/fs/f2fs/gc.h +++ b/fs/f2fs/gc.h @@ -91,7 +91,7 @@ static inline bool has_enough_invalid_blocks(struct f2fs_sb_info *sbi) block_t invalid_user_blocks = sbi->user_block_count - written_block_count(sbi); /* - * Background GC is triggered with the following condition. + * Background GC is triggered with the following conditions. * 1. There are a number of invalid blocks. * 2. There is not enough free space. */ diff --git a/fs/f2fs/hash.c b/fs/f2fs/hash.c index 948d17bf..a844fcf 100644 --- a/fs/f2fs/hash.c +++ b/fs/f2fs/hash.c @@ -42,7 +42,8 @@ static void TEA_transform(unsigned int buf[4], unsigned int const in[]) buf[1] += b1; } -static void str2hashbuf(const char *msg, size_t len, unsigned int *buf, int num) +static void str2hashbuf(const unsigned char *msg, size_t len, + unsigned int *buf, int num) { unsigned pad, val; int i; @@ -73,9 +74,9 @@ f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info) { __u32 hash; f2fs_hash_t f2fs_hash; - const char *p; + const unsigned char *p; __u32 in[8], buf[4]; - const char *name = name_info->name; + const unsigned char *name = name_info->name; size_t len = name_info->len; if ((len <= 2) && (name[0] == '.') && diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c index 5beecce..3e8ecdf 100644 --- a/fs/f2fs/inline.c +++ b/fs/f2fs/inline.c @@ -68,7 +68,7 @@ out: static int __f2fs_convert_inline_data(struct inode *inode, struct page *page) { - int err; + int err = 0; struct page *ipage; struct dnode_of_data dn; void *src_addr, *dst_addr; @@ -86,6 +86,10 @@ static int __f2fs_convert_inline_data(struct inode *inode, struct page *page) goto out; } + /* someone else converted inline_data already */ + if (!f2fs_has_inline_data(inode)) + goto out; + /* * i_addr[0] is not used for inline data, * so reserving new block will not destroy inline data @@ -124,9 +128,10 @@ out: return err; } -int f2fs_convert_inline_data(struct inode *inode, pgoff_t to_size) +int f2fs_convert_inline_data(struct inode *inode, pgoff_t to_size, + struct page *page) { - struct page *page; + struct page *new_page = page; int err; if (!f2fs_has_inline_data(inode)) @@ -134,17 +139,20 @@ int f2fs_convert_inline_data(struct inode *inode, pgoff_t to_size) else if (to_size <= MAX_INLINE_DATA) return 0; - page = grab_cache_page(inode->i_mapping, 0); - if (!page) - return -ENOMEM; + if (!page || page->index != 0) { + new_page = grab_cache_page(inode->i_mapping, 0); + if (!new_page) + return -ENOMEM; + } - err = __f2fs_convert_inline_data(inode, page); - f2fs_put_page(page, 1); + err = __f2fs_convert_inline_data(inode, new_page); + if (!page || page->index != 0) + f2fs_put_page(new_page, 1); return err; } int f2fs_write_inline_data(struct inode *inode, - struct page *page, unsigned size) + struct page *page, unsigned size) { void *src_addr, *dst_addr; struct page *ipage; @@ -199,7 +207,7 @@ void truncate_inline_data(struct inode *inode, u64 from) f2fs_put_page(ipage, 1); } -int recover_inline_data(struct inode *inode, struct page *npage) +bool recover_inline_data(struct inode *inode, struct page *npage) { struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); struct f2fs_inode *ri = NULL; @@ -218,7 +226,7 @@ int recover_inline_data(struct inode *inode, struct page *npage) ri = F2FS_INODE(npage); if (f2fs_has_inline_data(inode) && - ri && ri->i_inline & F2FS_INLINE_DATA) { + ri && (ri->i_inline & F2FS_INLINE_DATA)) { process_inline: ipage = get_node_page(sbi, inode->i_ino); f2fs_bug_on(IS_ERR(ipage)); @@ -230,7 +238,7 @@ process_inline: memcpy(dst_addr, src_addr, MAX_INLINE_DATA); update_inode(inode, ipage); f2fs_put_page(ipage, 1); - return -1; + return true; } if (f2fs_has_inline_data(inode)) { @@ -242,10 +250,10 @@ process_inline: clear_inode_flag(F2FS_I(inode), FI_INLINE_DATA); update_inode(inode, ipage); f2fs_put_page(ipage, 1); - } else if (ri && ri->i_inline & F2FS_INLINE_DATA) { - truncate_blocks(inode, 0); + } else if (ri && (ri->i_inline & F2FS_INLINE_DATA)) { + truncate_blocks(inode, 0, false); set_inode_flag(F2FS_I(inode), FI_INLINE_DATA); goto process_inline; } - return 0; + return false; } diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c index 27b0377..ee103fd 100644 --- a/fs/f2fs/namei.c +++ b/fs/f2fs/namei.c @@ -134,9 +134,7 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode, return 0; out: clear_nlink(inode); - unlock_new_inode(inode); - make_bad_inode(inode); - iput(inode); + iget_failed(inode); alloc_nid_failed(sbi, ino); return err; } @@ -229,7 +227,7 @@ static int f2fs_unlink(struct inode *dir, struct dentry *dentry) f2fs_delete_entry(de, page, inode); f2fs_unlock_op(sbi); - /* In order to evict this inode, we set it dirty */ + /* In order to evict this inode, we set it dirty */ mark_inode_dirty(inode); fail: trace_f2fs_unlink_exit(inode, err); @@ -267,9 +265,7 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry, return err; out: clear_nlink(inode); - unlock_new_inode(inode); - make_bad_inode(inode); - iput(inode); + iget_failed(inode); alloc_nid_failed(sbi, inode->i_ino); return err; } @@ -308,9 +304,7 @@ static int f2fs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) out_fail: clear_inode_flag(F2FS_I(inode), FI_INC_LINK); clear_nlink(inode); - unlock_new_inode(inode); - make_bad_inode(inode); - iput(inode); + iget_failed(inode); alloc_nid_failed(sbi, inode->i_ino); return err; } @@ -354,9 +348,7 @@ static int f2fs_mknod(struct inode *dir, struct dentry *dentry, return 0; out: clear_nlink(inode); - unlock_new_inode(inode); - make_bad_inode(inode); - iput(inode); + iget_failed(inode); alloc_nid_failed(sbi, inode->i_ino); return err; } @@ -688,9 +680,7 @@ release_out: out: f2fs_unlock_op(sbi); clear_nlink(inode); - unlock_new_inode(inode); - make_bad_inode(inode); - iput(inode); + iget_failed(inode); alloc_nid_failed(sbi, inode->i_ino); return err; } @@ -704,7 +694,6 @@ const struct inode_operations f2fs_dir_inode_operations = { .mkdir = f2fs_mkdir, .rmdir = f2fs_rmdir, .mknod = f2fs_mknod, - .rename = f2fs_rename, .rename2 = f2fs_rename2, .tmpfile = f2fs_tmpfile, .getattr = f2fs_getattr, diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index d3d90d2..4537819 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -237,7 +237,7 @@ retry: nat_get_blkaddr(e) != NULL_ADDR && new_blkaddr == NEW_ADDR); - /* increament version no as node is removed */ + /* increment version no as node is removed */ if (nat_get_blkaddr(e) != NEW_ADDR && new_blkaddr == NULL_ADDR) { unsigned char version = nat_get_version(e); nat_set_version(e, inc_node_version(version)); @@ -274,7 +274,7 @@ int try_to_free_nats(struct f2fs_sb_info *sbi, int nr_shrink) } /* - * This function returns always success + * This function always returns success */ void get_node_info(struct f2fs_sb_info *sbi, nid_t nid, struct node_info *ni) { @@ -650,7 +650,7 @@ static int truncate_partial_nodes(struct dnode_of_data *dn, /* get indirect nodes in the path */ for (i = 0; i < idx + 1; i++) { - /* refernece count'll be increased */ + /* reference count'll be increased */ pages[i] = get_node_page(sbi, nid[i]); if (IS_ERR(pages[i])) { err = PTR_ERR(pages[i]); @@ -823,22 +823,26 @@ int truncate_xattr_node(struct inode *inode, struct page *page) */ void remove_inode_page(struct inode *inode) { - struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); - struct page *page; - nid_t ino = inode->i_ino; struct dnode_of_data dn; - page = get_node_page(sbi, ino); - if (IS_ERR(page)) + set_new_dnode(&dn, inode, NULL, NULL, inode->i_ino); + if (get_dnode_of_data(&dn, 0, LOOKUP_NODE)) return; - if (truncate_xattr_node(inode, page)) { - f2fs_put_page(page, 1); + if (truncate_xattr_node(inode, dn.inode_page)) { + f2fs_put_dnode(&dn); return; } - /* 0 is possible, after f2fs_new_inode() is failed */ + + /* remove potential inline_data blocks */ + if (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || + S_ISLNK(inode->i_mode)) + truncate_data_blocks_range(&dn, 1); + + /* 0 is possible, after f2fs_new_inode() has failed */ f2fs_bug_on(inode->i_blocks != 0 && inode->i_blocks != 1); - set_new_dnode(&dn, inode, page, page, ino); + + /* will put inode & node pages */ truncate_node(&dn); } @@ -1129,8 +1133,11 @@ continue_unlock: set_fsync_mark(page, 0); set_dentry_mark(page, 0); } - NODE_MAPPING(sbi)->a_ops->writepage(page, wbc); - wrote++; + + if (NODE_MAPPING(sbi)->a_ops->writepage(page, wbc)) + unlock_page(page); + else + wrote++; if (--wbc->nr_to_write == 0) break; @@ -1212,6 +1219,8 @@ static int f2fs_write_node_page(struct page *page, if (unlikely(sbi->por_doing)) goto redirty_out; + if (unlikely(f2fs_cp_error(sbi))) + goto redirty_out; f2fs_wait_on_page_writeback(page, NODE); @@ -1540,15 +1549,6 @@ void alloc_nid_failed(struct f2fs_sb_info *sbi, nid_t nid) kmem_cache_free(free_nid_slab, i); } -void recover_node_page(struct f2fs_sb_info *sbi, struct page *page, - struct f2fs_summary *sum, struct node_info *ni, - block_t new_blkaddr) -{ - rewrite_node_page(sbi, page, sum, ni->blk_addr, new_blkaddr); - set_node_addr(sbi, ni, new_blkaddr, false); - clear_node_page_dirty(page); -} - void recover_inline_xattr(struct inode *inode, struct page *page) { struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); @@ -1557,40 +1557,33 @@ void recover_inline_xattr(struct inode *inode, struct page *page) struct page *ipage; struct f2fs_inode *ri; - if (!f2fs_has_inline_xattr(inode)) - return; - - if (!IS_INODE(page)) - return; - - ri = F2FS_INODE(page); - if (!(ri->i_inline & F2FS_INLINE_XATTR)) - return; - ipage = get_node_page(sbi, inode->i_ino); f2fs_bug_on(IS_ERR(ipage)); + ri = F2FS_INODE(page); + if (!(ri->i_inline & F2FS_INLINE_XATTR)) { + clear_inode_flag(F2FS_I(inode), FI_INLINE_XATTR); + goto update_inode; + } + dst_addr = inline_xattr_addr(ipage); src_addr = inline_xattr_addr(page); inline_size = inline_xattr_size(inode); f2fs_wait_on_page_writeback(ipage, NODE); memcpy(dst_addr, src_addr, inline_size); - +update_inode: update_inode(inode, ipage); f2fs_put_page(ipage, 1); } -bool recover_xattr_data(struct inode *inode, struct page *page, block_t blkaddr) +void recover_xattr_data(struct inode *inode, struct page *page, block_t blkaddr) { struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); nid_t prev_xnid = F2FS_I(inode)->i_xattr_nid; nid_t new_xnid = nid_of_node(page); struct node_info ni; - if (!f2fs_has_xattr_block(ofs_of_node(page))) - return false; - /* 1: invalidate the previous xattr nid */ if (!prev_xnid) goto recover_xnid; @@ -1618,7 +1611,6 @@ recover_xnid: set_node_addr(sbi, &ni, blkaddr, false); update_inode_page(inode); - return true; } int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page) @@ -1637,7 +1629,7 @@ int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page) if (!ipage) return -ENOMEM; - /* Should not use this inode from free nid list */ + /* Should not use this inode from free nid list */ remove_free_nid(NM_I(sbi), ino); SetPageUptodate(ipage); @@ -1651,6 +1643,7 @@ int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page) dst->i_blocks = cpu_to_le64(1); dst->i_links = cpu_to_le32(1); dst->i_xattr_nid = 0; + dst->i_inline = src->i_inline & F2FS_INLINE_XATTR; new_ni = old_ni; new_ni.ino = ino; @@ -1659,13 +1652,14 @@ int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page) WARN_ON(1); set_node_addr(sbi, &new_ni, NEW_ADDR, false); inc_valid_inode_count(sbi); + set_page_dirty(ipage); f2fs_put_page(ipage, 1); return 0; } /* * ra_sum_pages() merge contiguous pages into one bio and submit. - * these pre-readed pages are alloced in bd_inode's mapping tree. + * these pre-read pages are allocated in bd_inode's mapping tree. */ static int ra_sum_pages(struct f2fs_sb_info *sbi, struct page **pages, int start, int nrpages) @@ -1709,7 +1703,7 @@ int restore_node_summary(struct f2fs_sb_info *sbi, for (i = 0; !err && i < last_offset; i += nrpages, addr += nrpages) { nrpages = min(last_offset - i, bio_blocks); - /* read ahead node pages */ + /* readahead node pages */ nrpages = ra_sum_pages(sbi, pages, addr, nrpages); if (!nrpages) return -ENOMEM; @@ -1967,7 +1961,7 @@ static int init_node_manager(struct f2fs_sb_info *sbi) nm_i->max_nid = NAT_ENTRY_PER_BLOCK * nat_blocks; /* not used nids: 0, node, meta, (and root counted as valid node) */ - nm_i->available_nids = nm_i->max_nid - 3; + nm_i->available_nids = nm_i->max_nid - F2FS_RESERVED_NODE_NUM; nm_i->fcnt = 0; nm_i->nat_cnt = 0; nm_i->ram_thresh = DEF_RAM_THRESHOLD; diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c index fe1c6d9..756c41c 100644 --- a/fs/f2fs/recovery.c +++ b/fs/f2fs/recovery.c @@ -62,8 +62,10 @@ static int recover_dentry(struct page *ipage, struct inode *inode) } retry: de = f2fs_find_entry(dir, &name, &page); - if (de && inode->i_ino == le32_to_cpu(de->ino)) + if (de && inode->i_ino == le32_to_cpu(de->ino)) { + clear_inode_flag(F2FS_I(inode), FI_INC_LINK); goto out_unmap_put; + } if (de) { einode = f2fs_iget(inode->i_sb, le32_to_cpu(de->ino)); if (IS_ERR(einode)) { @@ -300,14 +302,19 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode, struct node_info ni; int err = 0, recovered = 0; - recover_inline_xattr(inode, page); - - if (recover_inline_data(inode, page)) + /* step 1: recover xattr */ + if (IS_INODE(page)) { + recover_inline_xattr(inode, page); + } else if (f2fs_has_xattr_block(ofs_of_node(page))) { + recover_xattr_data(inode, page, blkaddr); goto out; + } - if (recover_xattr_data(inode, page, blkaddr)) + /* step 2: recover inline data */ + if (recover_inline_data(inode, page)) goto out; + /* step 3: recover data indices */ start = start_bidx_of_node(ofs_of_node(page), fi); end = start + ADDRS_PER_PAGE(page, fi); @@ -364,8 +371,6 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode, fill_node_footer(dn.node_page, dn.nid, ni.ino, ofs_of_node(page), false); set_page_dirty(dn.node_page); - - recover_node_page(sbi, dn.node_page, &sum, &ni, blkaddr); err: f2fs_put_dnode(&dn); f2fs_unlock_op(sbi); @@ -452,6 +457,9 @@ int recover_fsync_data(struct f2fs_sb_info *sbi) /* step #1: find fsynced inode numbers */ sbi->por_doing = true; + /* prevent checkpoint */ + mutex_lock(&sbi->cp_mutex); + blkaddr = NEXT_FREE_BLKADDR(sbi, curseg); err = find_fsync_dnodes(sbi, &inode_list); @@ -465,7 +473,8 @@ int recover_fsync_data(struct f2fs_sb_info *sbi) /* step #2: recover data */ err = recover_data(sbi, &inode_list, CURSEG_WARM_NODE); - f2fs_bug_on(!list_empty(&inode_list)); + if (!err) + f2fs_bug_on(!list_empty(&inode_list)); out: destroy_fsync_dnodes(&inode_list); kmem_cache_destroy(fsync_entry_slab); @@ -482,8 +491,13 @@ out: /* Flush all the NAT/SIT pages */ while (get_pages(sbi, F2FS_DIRTY_META)) sync_meta_pages(sbi, META, LONG_MAX); + set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG); + mutex_unlock(&sbi->cp_mutex); } else if (need_writecp) { + mutex_unlock(&sbi->cp_mutex); write_checkpoint(sbi, false); + } else { + mutex_unlock(&sbi->cp_mutex); } return err; } diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 0dfeeba..0aa337c 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -62,7 +62,7 @@ static inline unsigned long __reverse_ffs(unsigned long word) } /* - * __find_rev_next(_zero)_bit is copied from lib/find_next_bit.c becasue + * __find_rev_next(_zero)_bit is copied from lib/find_next_bit.c because * f2fs_set_bit makes MSB and LSB reversed in a byte. * Example: * LSB <--> MSB @@ -808,7 +808,7 @@ static void __refresh_next_blkoff(struct f2fs_sb_info *sbi, } /* - * This function always allocates a used segment (from dirty seglist) by SSR + * This function always allocates a used segment(from dirty seglist) by SSR * manner, so it should recover the existing segment information of valid blocks */ static void change_curseg(struct f2fs_sb_info *sbi, int type, bool reuse) @@ -1103,55 +1103,6 @@ void recover_data_page(struct f2fs_sb_info *sbi, mutex_unlock(&curseg->curseg_mutex); } -void rewrite_node_page(struct f2fs_sb_info *sbi, - struct page *page, struct f2fs_summary *sum, - block_t old_blkaddr, block_t new_blkaddr) -{ - struct sit_info *sit_i = SIT_I(sbi); - int type = CURSEG_WARM_NODE; - struct curseg_info *curseg; - unsigned int segno, old_cursegno; - block_t next_blkaddr = next_blkaddr_of_node(page); - unsigned int next_segno = GET_SEGNO(sbi, next_blkaddr); - struct f2fs_io_info fio = { - .type = NODE, - .rw = WRITE_SYNC, - }; - - curseg = CURSEG_I(sbi, type); - - mutex_lock(&curseg->curseg_mutex); - mutex_lock(&sit_i->sentry_lock); - - segno = GET_SEGNO(sbi, new_blkaddr); - old_cursegno = curseg->segno; - - /* change the current segment */ - if (segno != curseg->segno) { - curseg->next_segno = segno; - change_curseg(sbi, type, true); - } - curseg->next_blkoff = GET_BLKOFF_FROM_SEG0(sbi, new_blkaddr); - __add_sum_entry(sbi, type, sum); - - /* change the current log to the next block addr in advance */ - if (next_segno != segno) { - curseg->next_segno = next_segno; - change_curseg(sbi, type, true); - } - curseg->next_blkoff = GET_BLKOFF_FROM_SEG0(sbi, next_blkaddr); - - /* rewrite node page */ - set_page_writeback(page); - f2fs_submit_page_mbio(sbi, page, new_blkaddr, &fio); - f2fs_submit_merged_bio(sbi, NODE, WRITE); - refresh_sit_entry(sbi, old_blkaddr, new_blkaddr); - locate_dirty_segment(sbi, old_cursegno); - - mutex_unlock(&sit_i->sentry_lock); - mutex_unlock(&curseg->curseg_mutex); -} - static inline bool is_merged_page(struct f2fs_sb_info *sbi, struct page *page, enum page_type type) { diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index 55973f7..ff48325 100644 --- a/fs/f2fs/segment.h +++ b/fs/f2fs/segment.h @@ -549,7 +549,7 @@ static inline void verify_block_addr(struct f2fs_sb_info *sbi, block_t blk_addr) } /* - * Summary block is always treated as invalid block + * Summary block is always treated as an invalid block */ static inline void check_block_count(struct f2fs_sb_info *sbi, int segno, struct f2fs_sit_entry *raw_sit) diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 657582f..41bdf51 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -432,9 +432,15 @@ static void f2fs_put_super(struct super_block *sb) stop_gc_thread(sbi); /* We don't need to do checkpoint when it's clean */ - if (sbi->s_dirty && get_pages(sbi, F2FS_DIRTY_NODES)) + if (sbi->s_dirty) write_checkpoint(sbi, true); + /* + * normally superblock is clean, so we need to release this. + * In addition, EIO will skip do checkpoint, we need this as well. + */ + release_dirty_inode(sbi); + iput(sbi->node_inode); iput(sbi->meta_inode); @@ -457,9 +463,6 @@ int f2fs_sync_fs(struct super_block *sb, int sync) trace_f2fs_sync_fs(sb, sync); - if (!sbi->s_dirty && !get_pages(sbi, F2FS_DIRTY_NODES)) - return 0; - if (sync) { mutex_lock(&sbi->gc_mutex); write_checkpoint(sbi, false); @@ -505,8 +508,8 @@ static int f2fs_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_bfree = buf->f_blocks - valid_user_blocks(sbi) - ovp_count; buf->f_bavail = user_block_count - valid_user_blocks(sbi); - buf->f_files = sbi->total_node_count; - buf->f_ffree = sbi->total_node_count - valid_inode_count(sbi); + buf->f_files = sbi->total_node_count - F2FS_RESERVED_NODE_NUM; + buf->f_ffree = buf->f_files - valid_inode_count(sbi); buf->f_namelen = F2FS_NAME_LEN; buf->f_fsid.val[0] = (u32)id; @@ -663,7 +666,7 @@ restore_gc: if (need_restart_gc) { if (start_gc_thread(sbi)) f2fs_msg(sbi->sb, KERN_WARNING, - "background gc thread is stop"); + "background gc thread has stopped"); } else if (need_stop_gc) { stop_gc_thread(sbi); } @@ -812,7 +815,7 @@ static int sanity_check_ckpt(struct f2fs_sb_info *sbi) if (unlikely(fsmeta >= total)) return 1; - if (unlikely(is_set_ckpt_flags(ckpt, CP_ERROR_FLAG))) { + if (unlikely(f2fs_cp_error(sbi))) { f2fs_msg(sbi->sb, KERN_ERR, "A bug case: need to run fsck"); return 1; } @@ -899,8 +902,10 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent) struct buffer_head *raw_super_buf; struct inode *root; long err = -EINVAL; + bool retry = true; int i; +try_onemore: /* allocate memory for f2fs-specific super block info */ sbi = kzalloc(sizeof(struct f2fs_sb_info), GFP_KERNEL); if (!sbi) @@ -1080,9 +1085,11 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent) /* recover fsynced data */ if (!test_opt(sbi, DISABLE_ROLL_FORWARD)) { err = recover_fsync_data(sbi); - if (err) + if (err) { f2fs_msg(sb, KERN_ERR, "Cannot recover all fsync data errno=%ld", err); + goto free_kobj; + } } /* @@ -1123,6 +1130,13 @@ free_sb_buf: brelse(raw_super_buf); free_sbi: kfree(sbi); + + /* give only one another chance */ + if (retry) { + retry = 0; + shrink_dcache_sb(sb); + goto try_onemore; + } return err; } diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c index 8bea941..728a5dc 100644 --- a/fs/f2fs/xattr.c +++ b/fs/f2fs/xattr.c @@ -528,7 +528,7 @@ static int __f2fs_setxattr(struct inode *inode, int index, int free; /* * If value is NULL, it is remove operation. - * In case of update operation, we caculate free. + * In case of update operation, we calculate free. */ free = MIN_OFFSET(inode) - ((char *)last - (char *)base_addr); if (found) diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 6fac743..b73e021 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -97,7 +97,7 @@ static void jbd2_commit_block_csum_set(journal_t *j, struct buffer_head *bh) struct commit_header *h; __u32 csum; - if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (!jbd2_journal_has_csum_v2or3(j)) return; h = (struct commit_header *)(bh->b_data); @@ -313,11 +313,11 @@ static __u32 jbd2_checksum_data(__u32 crc32_sum, struct buffer_head *bh) return checksum; } -static void write_tag_block(int tag_bytes, journal_block_tag_t *tag, +static void write_tag_block(journal_t *j, journal_block_tag_t *tag, unsigned long long block) { tag->t_blocknr = cpu_to_be32(block & (u32)~0); - if (tag_bytes > JBD2_TAG_SIZE32) + if (JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_64BIT)) tag->t_blocknr_high = cpu_to_be32((block >> 31) >> 1); } @@ -327,7 +327,7 @@ static void jbd2_descr_block_csum_set(journal_t *j, struct jbd2_journal_block_tail *tail; __u32 csum; - if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (!jbd2_journal_has_csum_v2or3(j)) return; tail = (struct jbd2_journal_block_tail *)(bh->b_data + j->j_blocksize - @@ -340,12 +340,13 @@ static void jbd2_descr_block_csum_set(journal_t *j, static void jbd2_block_tag_csum_set(journal_t *j, journal_block_tag_t *tag, struct buffer_head *bh, __u32 sequence) { + journal_block_tag3_t *tag3 = (journal_block_tag3_t *)tag; struct page *page = bh->b_page; __u8 *addr; __u32 csum32; __be32 seq; - if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (!jbd2_journal_has_csum_v2or3(j)) return; seq = cpu_to_be32(sequence); @@ -355,8 +356,10 @@ static void jbd2_block_tag_csum_set(journal_t *j, journal_block_tag_t *tag, bh->b_size); kunmap_atomic(addr); - /* We only have space to store the lower 16 bits of the crc32c. */ - tag->t_checksum = cpu_to_be16(csum32); + if (JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V3)) + tag3->t_checksum = cpu_to_be32(csum32); + else + tag->t_checksum = cpu_to_be16(csum32); } /* * jbd2_journal_commit_transaction @@ -396,7 +399,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) LIST_HEAD(io_bufs); LIST_HEAD(log_bufs); - if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (jbd2_journal_has_csum_v2or3(journal)) csum_size = sizeof(struct jbd2_journal_block_tail); /* @@ -690,7 +693,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) tag_flag |= JBD2_FLAG_SAME_UUID; tag = (journal_block_tag_t *) tagp; - write_tag_block(tag_bytes, tag, jh2bh(jh)->b_blocknr); + write_tag_block(journal, tag, jh2bh(jh)->b_blocknr); tag->t_flags = cpu_to_be16(tag_flag); jbd2_block_tag_csum_set(journal, tag, wbuf[bufs], commit_transaction->t_tid); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 67b8e30..19d74d8 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -124,7 +124,7 @@ EXPORT_SYMBOL(__jbd2_debug); /* Checksumming functions */ static int jbd2_verify_csum_type(journal_t *j, journal_superblock_t *sb) { - if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (!jbd2_journal_has_csum_v2or3(j)) return 1; return sb->s_checksum_type == JBD2_CRC32C_CHKSUM; @@ -145,7 +145,7 @@ static __be32 jbd2_superblock_csum(journal_t *j, journal_superblock_t *sb) static int jbd2_superblock_csum_verify(journal_t *j, journal_superblock_t *sb) { - if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (!jbd2_journal_has_csum_v2or3(j)) return 1; return sb->s_checksum == jbd2_superblock_csum(j, sb); @@ -153,7 +153,7 @@ static int jbd2_superblock_csum_verify(journal_t *j, journal_superblock_t *sb) static void jbd2_superblock_csum_set(journal_t *j, journal_superblock_t *sb) { - if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (!jbd2_journal_has_csum_v2or3(j)) return; sb->s_checksum = jbd2_superblock_csum(j, sb); @@ -1522,21 +1522,29 @@ static int journal_get_superblock(journal_t *journal) goto out; } - if (JBD2_HAS_COMPAT_FEATURE(journal, JBD2_FEATURE_COMPAT_CHECKSUM) && - JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2)) { + if (jbd2_journal_has_csum_v2or3(journal) && + JBD2_HAS_COMPAT_FEATURE(journal, JBD2_FEATURE_COMPAT_CHECKSUM)) { /* Can't have checksum v1 and v2 on at the same time! */ printk(KERN_ERR "JBD2: Can't enable checksumming v1 and v2 " "at the same time!\n"); goto out; } + if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2) && + JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V3)) { + /* Can't have checksum v2 and v3 at the same time! */ + printk(KERN_ERR "JBD2: Can't enable checksumming v2 and v3 " + "at the same time!\n"); + goto out; + } + if (!jbd2_verify_csum_type(journal, sb)) { printk(KERN_ERR "JBD2: Unknown checksum type\n"); goto out; } /* Load the checksum driver */ - if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2)) { + if (jbd2_journal_has_csum_v2or3(journal)) { journal->j_chksum_driver = crypto_alloc_shash("crc32c", 0, 0); if (IS_ERR(journal->j_chksum_driver)) { printk(KERN_ERR "JBD2: Cannot load crc32c driver.\n"); @@ -1553,7 +1561,7 @@ static int journal_get_superblock(journal_t *journal) } /* Precompute checksum seed for all metadata */ - if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (jbd2_journal_has_csum_v2or3(journal)) journal->j_csum_seed = jbd2_chksum(journal, ~0, sb->s_uuid, sizeof(sb->s_uuid)); @@ -1813,8 +1821,14 @@ int jbd2_journal_set_features (journal_t *journal, unsigned long compat, if (!jbd2_journal_check_available_features(journal, compat, ro, incompat)) return 0; - /* Asking for checksumming v2 and v1? Only give them v2. */ - if (incompat & JBD2_FEATURE_INCOMPAT_CSUM_V2 && + /* If enabling v2 checksums, turn on v3 instead */ + if (incompat & JBD2_FEATURE_INCOMPAT_CSUM_V2) { + incompat &= ~JBD2_FEATURE_INCOMPAT_CSUM_V2; + incompat |= JBD2_FEATURE_INCOMPAT_CSUM_V3; + } + + /* Asking for checksumming v3 and v1? Only give them v3. */ + if (incompat & JBD2_FEATURE_INCOMPAT_CSUM_V3 && compat & JBD2_FEATURE_COMPAT_CHECKSUM) compat &= ~JBD2_FEATURE_COMPAT_CHECKSUM; @@ -1823,8 +1837,8 @@ int jbd2_journal_set_features (journal_t *journal, unsigned long compat, sb = journal->j_superblock; - /* If enabling v2 checksums, update superblock */ - if (INCOMPAT_FEATURE_ON(JBD2_FEATURE_INCOMPAT_CSUM_V2)) { + /* If enabling v3 checksums, update superblock */ + if (INCOMPAT_FEATURE_ON(JBD2_FEATURE_INCOMPAT_CSUM_V3)) { sb->s_checksum_type = JBD2_CRC32C_CHKSUM; sb->s_feature_compat &= ~cpu_to_be32(JBD2_FEATURE_COMPAT_CHECKSUM); @@ -1842,8 +1856,7 @@ int jbd2_journal_set_features (journal_t *journal, unsigned long compat, } /* Precompute checksum seed for all metadata */ - if (JBD2_HAS_INCOMPAT_FEATURE(journal, - JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (jbd2_journal_has_csum_v2or3(journal)) journal->j_csum_seed = jbd2_chksum(journal, ~0, sb->s_uuid, sizeof(sb->s_uuid)); @@ -1852,7 +1865,8 @@ int jbd2_journal_set_features (journal_t *journal, unsigned long compat, /* If enabling v1 checksums, downgrade superblock */ if (COMPAT_FEATURE_ON(JBD2_FEATURE_COMPAT_CHECKSUM)) sb->s_feature_incompat &= - ~cpu_to_be32(JBD2_FEATURE_INCOMPAT_CSUM_V2); + ~cpu_to_be32(JBD2_FEATURE_INCOMPAT_CSUM_V2 | + JBD2_FEATURE_INCOMPAT_CSUM_V3); sb->s_feature_compat |= cpu_to_be32(compat); sb->s_feature_ro_compat |= cpu_to_be32(ro); @@ -2165,16 +2179,20 @@ int jbd2_journal_blocks_per_page(struct inode *inode) */ size_t journal_tag_bytes(journal_t *journal) { - journal_block_tag_t tag; - size_t x = 0; + size_t sz; + + if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V3)) + return sizeof(journal_block_tag3_t); + + sz = sizeof(journal_block_tag_t); if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2)) - x += sizeof(tag.t_checksum); + sz += sizeof(__u16); if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT)) - return x + JBD2_TAG_SIZE64; + return sz; else - return x + JBD2_TAG_SIZE32; + return sz - sizeof(__u32); } /* diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index 3b6bb19..9b329b5 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -181,7 +181,7 @@ static int jbd2_descr_block_csum_verify(journal_t *j, __be32 provided; __u32 calculated; - if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (!jbd2_journal_has_csum_v2or3(j)) return 1; tail = (struct jbd2_journal_block_tail *)(buf + j->j_blocksize - @@ -205,7 +205,7 @@ static int count_tags(journal_t *journal, struct buffer_head *bh) int nr = 0, size = journal->j_blocksize; int tag_bytes = journal_tag_bytes(journal); - if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (jbd2_journal_has_csum_v2or3(journal)) size -= sizeof(struct jbd2_journal_block_tail); tagp = &bh->b_data[sizeof(journal_header_t)]; @@ -338,10 +338,11 @@ int jbd2_journal_skip_recovery(journal_t *journal) return err; } -static inline unsigned long long read_tag_block(int tag_bytes, journal_block_tag_t *tag) +static inline unsigned long long read_tag_block(journal_t *journal, + journal_block_tag_t *tag) { unsigned long long block = be32_to_cpu(tag->t_blocknr); - if (tag_bytes > JBD2_TAG_SIZE32) + if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT)) block |= (u64)be32_to_cpu(tag->t_blocknr_high) << 32; return block; } @@ -384,7 +385,7 @@ static int jbd2_commit_block_csum_verify(journal_t *j, void *buf) __be32 provided; __u32 calculated; - if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (!jbd2_journal_has_csum_v2or3(j)) return 1; h = buf; @@ -399,17 +400,21 @@ static int jbd2_commit_block_csum_verify(journal_t *j, void *buf) static int jbd2_block_tag_csum_verify(journal_t *j, journal_block_tag_t *tag, void *buf, __u32 sequence) { + journal_block_tag3_t *tag3 = (journal_block_tag3_t *)tag; __u32 csum32; __be32 seq; - if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (!jbd2_journal_has_csum_v2or3(j)) return 1; seq = cpu_to_be32(sequence); csum32 = jbd2_chksum(j, j->j_csum_seed, (__u8 *)&seq, sizeof(seq)); csum32 = jbd2_chksum(j, csum32, buf, j->j_blocksize); - return tag->t_checksum == cpu_to_be16(csum32); + if (JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V3)) + return tag3->t_checksum == cpu_to_be32(csum32); + else + return tag->t_checksum == cpu_to_be16(csum32); } static int do_one_pass(journal_t *journal, @@ -426,6 +431,7 @@ static int do_one_pass(journal_t *journal, int tag_bytes = journal_tag_bytes(journal); __u32 crc32_sum = ~0; /* Transactional Checksums */ int descr_csum_size = 0; + int block_error = 0; /* * First thing is to establish what we expect to find in the log @@ -512,8 +518,7 @@ static int do_one_pass(journal_t *journal, switch(blocktype) { case JBD2_DESCRIPTOR_BLOCK: /* Verify checksum first */ - if (JBD2_HAS_INCOMPAT_FEATURE(journal, - JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (jbd2_journal_has_csum_v2or3(journal)) descr_csum_size = sizeof(struct jbd2_journal_block_tail); if (descr_csum_size > 0 && @@ -574,7 +579,7 @@ static int do_one_pass(journal_t *journal, unsigned long long blocknr; J_ASSERT(obh != NULL); - blocknr = read_tag_block(tag_bytes, + blocknr = read_tag_block(journal, tag); /* If the block has been @@ -598,7 +603,8 @@ static int do_one_pass(journal_t *journal, "checksum recovering " "block %llu in log\n", blocknr); - continue; + block_error = 1; + goto skip_write; } /* Find a buffer for the new @@ -797,7 +803,8 @@ static int do_one_pass(journal_t *journal, success = -EIO; } } - + if (block_error && success == 0) + success = -EIO; return success; failed: @@ -811,7 +818,7 @@ static int jbd2_revoke_block_csum_verify(journal_t *j, __be32 provided; __u32 calculated; - if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (!jbd2_journal_has_csum_v2or3(j)) return 1; tail = (struct jbd2_journal_revoke_tail *)(buf + j->j_blocksize - diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 198c9c1..d5e95a1 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -91,8 +91,8 @@ #include <linux/list.h> #include <linux/init.h> #include <linux/bio.h> -#endif #include <linux/log2.h> +#endif static struct kmem_cache *jbd2_revoke_record_cache; static struct kmem_cache *jbd2_revoke_table_cache; @@ -597,7 +597,7 @@ static void write_one_revoke_record(journal_t *journal, offset = *offsetp; /* Do we need to leave space at the end for a checksum? */ - if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (jbd2_journal_has_csum_v2or3(journal)) csum_size = sizeof(struct jbd2_journal_revoke_tail); /* Make sure we have a descriptor with space left for the record */ @@ -644,7 +644,7 @@ static void jbd2_revoke_csum_set(journal_t *j, struct buffer_head *bh) struct jbd2_journal_revoke_tail *tail; __u32 csum; - if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + if (!jbd2_journal_has_csum_v2or3(j)) return; tail = (struct jbd2_journal_revoke_tail *)(bh->b_data + j->j_blocksize - diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index 8f27c93..ec9e082f 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -253,13 +253,11 @@ static int lockd_up_net(struct svc_serv *serv, struct net *net) error = make_socks(serv, net); if (error < 0) - goto err_socks; + goto err_bind; set_grace_period(net); dprintk("lockd_up_net: per-net data created; net=%p\n", net); return 0; -err_socks: - svc_rpcb_cleanup(serv, net); err_bind: ln->nlmsvc_users--; return error; @@ -1619,7 +1619,7 @@ static int generic_add_lease(struct file *filp, long arg, struct file_lock **flp smp_mb(); error = check_conflicting_open(dentry, arg); if (error) - locks_unlink_lock(flp); + locks_unlink_lock(before); out: if (is_deleg) mutex_unlock(&inode->i_mutex); diff --git a/fs/namespace.c b/fs/namespace.c index a01c773..ef42d9b 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1217,6 +1217,11 @@ static void namespace_unlock(void) head.first->pprev = &head.first; INIT_HLIST_HEAD(&unmounted); + /* undo decrements we'd done in umount_tree() */ + hlist_for_each_entry(mnt, &head, mnt_hash) + if (mnt->mnt_ex_mountpoint.mnt) + mntget(mnt->mnt_ex_mountpoint.mnt); + up_write(&namespace_sem); synchronize_rcu(); @@ -1253,6 +1258,9 @@ void umount_tree(struct mount *mnt, int how) hlist_add_head(&p->mnt_hash, &tmp_list); } + hlist_for_each_entry(p, &tmp_list, mnt_hash) + list_del_init(&p->mnt_child); + if (how) propagate_umount(&tmp_list); @@ -1263,9 +1271,9 @@ void umount_tree(struct mount *mnt, int how) p->mnt_ns = NULL; if (how < 2) p->mnt.mnt_flags |= MNT_SYNC_UMOUNT; - list_del_init(&p->mnt_child); if (mnt_has_parent(p)) { put_mountpoint(p->mnt_mp); + mnt_add_count(p->mnt_parent, -1); /* move the reference to mountpoint into ->mnt_ex_mountpoint */ p->mnt_ex_mountpoint.dentry = p->mnt_mountpoint; p->mnt_ex_mountpoint.mnt = &p->mnt_parent->mnt; diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c index d0fec26..24c6898 100644 --- a/fs/nfs/nfs3acl.c +++ b/fs/nfs/nfs3acl.c @@ -129,7 +129,10 @@ static int __nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, .rpc_argp = &args, .rpc_resp = &fattr, }; - int status; + int status = 0; + + if (acl == NULL && (!S_ISDIR(inode->i_mode) || dfacl == NULL)) + goto out; status = -EOPNOTSUPP; if (!nfs_server_capable(inode, NFS_CAP_ACLS)) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 75ae8d2..7dd8aca 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -2560,6 +2560,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data) struct nfs4_closedata *calldata = data; struct nfs4_state *state = calldata->state; struct nfs_server *server = NFS_SERVER(calldata->inode); + nfs4_stateid *res_stateid = NULL; dprintk("%s: begin!\n", __func__); if (!nfs4_sequence_done(task, &calldata->res.seq_res)) @@ -2570,12 +2571,12 @@ static void nfs4_close_done(struct rpc_task *task, void *data) */ switch (task->tk_status) { case 0: - if (calldata->roc) + res_stateid = &calldata->res.stateid; + if (calldata->arg.fmode == 0 && calldata->roc) pnfs_roc_set_barrier(state->inode, calldata->roc_barrier); - nfs_clear_open_stateid(state, &calldata->res.stateid, 0); renew_lease(server, calldata->timestamp); - goto out_release; + break; case -NFS4ERR_ADMIN_REVOKED: case -NFS4ERR_STALE_STATEID: case -NFS4ERR_OLD_STATEID: @@ -2589,7 +2590,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data) goto out_release; } } - nfs_clear_open_stateid(state, NULL, calldata->arg.fmode); + nfs_clear_open_stateid(state, res_stateid, calldata->arg.fmode); out_release: nfs_release_seqid(calldata->arg.seqid); nfs_refresh_inode(calldata->inode, calldata->res.fattr); @@ -2601,6 +2602,7 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data) struct nfs4_closedata *calldata = data; struct nfs4_state *state = calldata->state; struct inode *inode = calldata->inode; + bool is_rdonly, is_wronly, is_rdwr; int call_close = 0; dprintk("%s: begin!\n", __func__); @@ -2608,18 +2610,24 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data) goto out_wait; task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE]; - calldata->arg.fmode = FMODE_READ|FMODE_WRITE; spin_lock(&state->owner->so_lock); + is_rdwr = test_bit(NFS_O_RDWR_STATE, &state->flags); + is_rdonly = test_bit(NFS_O_RDONLY_STATE, &state->flags); + is_wronly = test_bit(NFS_O_WRONLY_STATE, &state->flags); + /* Calculate the current open share mode */ + calldata->arg.fmode = 0; + if (is_rdonly || is_rdwr) + calldata->arg.fmode |= FMODE_READ; + if (is_wronly || is_rdwr) + calldata->arg.fmode |= FMODE_WRITE; /* Calculate the change in open mode */ if (state->n_rdwr == 0) { if (state->n_rdonly == 0) { - call_close |= test_bit(NFS_O_RDONLY_STATE, &state->flags); - call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags); + call_close |= is_rdonly || is_rdwr; calldata->arg.fmode &= ~FMODE_READ; } if (state->n_wronly == 0) { - call_close |= test_bit(NFS_O_WRONLY_STATE, &state->flags); - call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags); + call_close |= is_wronly || is_rdwr; calldata->arg.fmode &= ~FMODE_WRITE; } } diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index ba49192..be7cbce 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -116,7 +116,7 @@ __nfs_iocounter_wait(struct nfs_io_counter *c) if (atomic_read(&c->io_count) == 0) break; ret = nfs_wait_bit_killable(&q.key); - } while (atomic_read(&c->io_count) != 0); + } while (atomic_read(&c->io_count) != 0 && !ret); finish_wait(wq, &q.wait); return ret; } @@ -139,26 +139,49 @@ nfs_iocounter_wait(struct nfs_io_counter *c) /* * nfs_page_group_lock - lock the head of the page group * @req - request in group that is to be locked + * @nonblock - if true don't block waiting for lock * * this lock must be held if modifying the page group list * - * returns result from wait_on_bit_lock: 0 on success, < 0 on error + * return 0 on success, < 0 on error: -EDELAY if nonblocking or the + * result from wait_on_bit_lock + * + * NOTE: calling with nonblock=false should always have set the + * lock bit (see fs/buffer.c and other uses of wait_on_bit_lock + * with TASK_UNINTERRUPTIBLE), so there is no need to check the result. */ int -nfs_page_group_lock(struct nfs_page *req, bool wait) +nfs_page_group_lock(struct nfs_page *req, bool nonblock) { struct nfs_page *head = req->wb_head; - int ret; WARN_ON_ONCE(head != head->wb_head); - do { - ret = wait_on_bit_lock(&head->wb_flags, PG_HEADLOCK, - TASK_UNINTERRUPTIBLE); - } while (wait && ret != 0); + if (!test_and_set_bit(PG_HEADLOCK, &head->wb_flags)) + return 0; - WARN_ON_ONCE(ret > 0); - return ret; + if (!nonblock) + return wait_on_bit_lock(&head->wb_flags, PG_HEADLOCK, + TASK_UNINTERRUPTIBLE); + + return -EAGAIN; +} + +/* + * nfs_page_group_lock_wait - wait for the lock to clear, but don't grab it + * @req - a request in the group + * + * This is a blocking call to wait for the group lock to be cleared. + */ +void +nfs_page_group_lock_wait(struct nfs_page *req) +{ + struct nfs_page *head = req->wb_head; + + WARN_ON_ONCE(head != head->wb_head); + + wait_on_bit(&head->wb_flags, PG_HEADLOCK, + TASK_UNINTERRUPTIBLE); } /* @@ -219,7 +242,7 @@ bool nfs_page_group_sync_on_bit(struct nfs_page *req, unsigned int bit) { bool ret; - nfs_page_group_lock(req, true); + nfs_page_group_lock(req, false); ret = nfs_page_group_sync_on_bit_locked(req, bit); nfs_page_group_unlock(req); @@ -701,10 +724,11 @@ int nfs_generic_pgio(struct nfs_pageio_descriptor *desc, struct nfs_pgio_header *hdr) { struct nfs_page *req; - struct page **pages; + struct page **pages, + *last_page; struct list_head *head = &desc->pg_list; struct nfs_commit_info cinfo; - unsigned int pagecount; + unsigned int pagecount, pageused; pagecount = nfs_page_array_len(desc->pg_base, desc->pg_count); if (!nfs_pgarray_set(&hdr->page_array, pagecount)) @@ -712,12 +736,23 @@ int nfs_generic_pgio(struct nfs_pageio_descriptor *desc, nfs_init_cinfo(&cinfo, desc->pg_inode, desc->pg_dreq); pages = hdr->page_array.pagevec; + last_page = NULL; + pageused = 0; while (!list_empty(head)) { req = nfs_list_entry(head->next); nfs_list_remove_request(req); nfs_list_add_request(req, &hdr->pages); - *pages++ = req->wb_page; + + if (WARN_ON_ONCE(pageused >= pagecount)) + return nfs_pgio_error(desc, hdr); + + if (!last_page || last_page != req->wb_page) { + *pages++ = last_page = req->wb_page; + pageused++; + } } + if (WARN_ON_ONCE(pageused != pagecount)) + return nfs_pgio_error(desc, hdr); if ((desc->pg_ioflags & FLUSH_COND_STABLE) && (desc->pg_moreio || nfs_reqs_to_commit(&cinfo))) @@ -788,6 +823,14 @@ static bool nfs_can_coalesce_requests(struct nfs_page *prev, return false; if (req_offset(req) != req_offset(prev) + prev->wb_bytes) return false; + if (req->wb_page == prev->wb_page) { + if (req->wb_pgbase != prev->wb_pgbase + prev->wb_bytes) + return false; + } else { + if (req->wb_pgbase != 0 || + prev->wb_pgbase + prev->wb_bytes != PAGE_CACHE_SIZE) + return false; + } } size = pgio->pg_ops->pg_test(pgio, prev, req); WARN_ON_ONCE(size > req->wb_bytes); @@ -858,13 +901,8 @@ static int __nfs_pageio_add_request(struct nfs_pageio_descriptor *desc, struct nfs_page *subreq; unsigned int bytes_left = 0; unsigned int offset, pgbase; - int ret; - ret = nfs_page_group_lock(req, false); - if (ret < 0) { - desc->pg_error = ret; - return 0; - } + nfs_page_group_lock(req, false); subreq = req; bytes_left = subreq->wb_bytes; @@ -886,11 +924,7 @@ static int __nfs_pageio_add_request(struct nfs_pageio_descriptor *desc, if (desc->pg_recoalesce) return 0; /* retry add_request for this subreq */ - ret = nfs_page_group_lock(req, false); - if (ret < 0) { - desc->pg_error = ret; - return 0; - } + nfs_page_group_lock(req, false); continue; } diff --git a/fs/nfs/write.c b/fs/nfs/write.c index e3b5cf2..175d5d0 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -241,7 +241,7 @@ static bool nfs_page_group_covers_page(struct nfs_page *req) unsigned int pos = 0; unsigned int len = nfs_page_length(req->wb_page); - nfs_page_group_lock(req, true); + nfs_page_group_lock(req, false); do { tmp = nfs_page_group_search_locked(req->wb_head, pos); @@ -478,10 +478,23 @@ try_again: return NULL; } - /* lock each request in the page group */ - ret = nfs_page_group_lock(head, false); - if (ret < 0) + /* holding inode lock, so always make a non-blocking call to try the + * page group lock */ + ret = nfs_page_group_lock(head, true); + if (ret < 0) { + spin_unlock(&inode->i_lock); + + if (!nonblock && ret == -EAGAIN) { + nfs_page_group_lock_wait(head); + nfs_release_request(head); + goto try_again; + } + + nfs_release_request(head); return ERR_PTR(ret); + } + + /* lock each request in the page group */ subreq = head; do { /* diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index f9821ce..e94457c 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -2657,6 +2657,7 @@ nfsd4_encode_dirent(void *ccdv, const char *name, int namlen, struct xdr_stream *xdr = cd->xdr; int start_offset = xdr->buf->len; int cookie_offset; + u32 name_and_cookie; int entry_bytes; __be32 nfserr = nfserr_toosmall; __be64 wire_offset; @@ -2718,7 +2719,14 @@ nfsd4_encode_dirent(void *ccdv, const char *name, int namlen, cd->rd_maxcount -= entry_bytes; if (!cd->rd_dircount) goto fail; - cd->rd_dircount--; + /* + * RFC 3530 14.2.24 describes rd_dircount as only a "hint", so + * let's always let through the first entry, at least: + */ + name_and_cookie = 4 * XDR_QUADLEN(namlen) + 8; + if (name_and_cookie > cd->rd_dircount && cd->cookie_offset) + goto fail; + cd->rd_dircount -= min(cd->rd_dircount, name_and_cookie); cd->cookie_offset = cookie_offset; skip_entry: cd->common.err = nfs_ok; @@ -3321,6 +3329,10 @@ nfsd4_encode_readdir(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4 } maxcount = min_t(int, maxcount-16, bytes_left); + /* RFC 3530 14.2.24 allows us to ignore dircount when it's 0: */ + if (!readdir->rd_dircount) + readdir->rd_dircount = INT_MAX; + readdir->xdr = xdr; readdir->rd_maxcount = maxcount; readdir->common.err = 0; diff --git a/fs/ocfs2/cluster/quorum.c b/fs/ocfs2/cluster/quorum.c index 1ec141e..62e8ec6 100644 --- a/fs/ocfs2/cluster/quorum.c +++ b/fs/ocfs2/cluster/quorum.c @@ -160,9 +160,18 @@ static void o2quo_make_decision(struct work_struct *work) } out: - spin_unlock(&qs->qs_lock); - if (fence) + if (fence) { + spin_unlock(&qs->qs_lock); o2quo_fence_self(); + } else { + mlog(ML_NOTICE, "not fencing this node, heartbeating: %d, " + "connected: %d, lowest: %d (%sreachable)\n", + qs->qs_heartbeating, qs->qs_connected, lowest_hb, + lowest_reachable ? "" : "un"); + spin_unlock(&qs->qs_lock); + + } + } static void o2quo_set_hold(struct o2quo_state *qs, u8 node) diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index 681691b..ea34952 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -1480,6 +1480,14 @@ static int o2net_set_nodelay(struct socket *sock) return ret; } +static int o2net_set_usertimeout(struct socket *sock) +{ + int user_timeout = O2NET_TCP_USER_TIMEOUT; + + return kernel_setsockopt(sock, SOL_TCP, TCP_USER_TIMEOUT, + (char *)&user_timeout, sizeof(user_timeout)); +} + static void o2net_initialize_handshake(void) { o2net_hand->o2hb_heartbeat_timeout_ms = cpu_to_be32( @@ -1536,16 +1544,20 @@ static void o2net_idle_timer(unsigned long data) #endif printk(KERN_NOTICE "o2net: Connection to " SC_NODEF_FMT " has been " - "idle for %lu.%lu secs, shutting it down.\n", SC_NODEF_ARGS(sc), - msecs / 1000, msecs % 1000); + "idle for %lu.%lu secs.\n", + SC_NODEF_ARGS(sc), msecs / 1000, msecs % 1000); - /* - * Initialize the nn_timeout so that the next connection attempt - * will continue in o2net_start_connect. + /* idle timerout happen, don't shutdown the connection, but + * make fence decision. Maybe the connection can recover before + * the decision is made. */ atomic_set(&nn->nn_timeout, 1); + o2quo_conn_err(o2net_num_from_nn(nn)); + queue_delayed_work(o2net_wq, &nn->nn_still_up, + msecs_to_jiffies(O2NET_QUORUM_DELAY_MS)); + + o2net_sc_reset_idle_timer(sc); - o2net_sc_queue_work(sc, &sc->sc_shutdown_work); } static void o2net_sc_reset_idle_timer(struct o2net_sock_container *sc) @@ -1560,6 +1572,15 @@ static void o2net_sc_reset_idle_timer(struct o2net_sock_container *sc) static void o2net_sc_postpone_idle(struct o2net_sock_container *sc) { + struct o2net_node *nn = o2net_nn_from_num(sc->sc_node->nd_num); + + /* clear fence decision since the connection recover from timeout*/ + if (atomic_read(&nn->nn_timeout)) { + o2quo_conn_up(o2net_num_from_nn(nn)); + cancel_delayed_work(&nn->nn_still_up); + atomic_set(&nn->nn_timeout, 0); + } + /* Only push out an existing timer */ if (timer_pending(&sc->sc_idle_timeout)) o2net_sc_reset_idle_timer(sc); @@ -1650,6 +1671,12 @@ static void o2net_start_connect(struct work_struct *work) goto out; } + ret = o2net_set_usertimeout(sock); + if (ret) { + mlog(ML_ERROR, "set TCP_USER_TIMEOUT failed with %d\n", ret); + goto out; + } + o2net_register_callbacks(sc->sc_sock->sk, sc); spin_lock(&nn->nn_lock); @@ -1831,6 +1858,12 @@ static int o2net_accept_one(struct socket *sock, int *more) goto out; } + ret = o2net_set_usertimeout(new_sock); + if (ret) { + mlog(ML_ERROR, "set TCP_USER_TIMEOUT failed with %d\n", ret); + goto out; + } + slen = sizeof(sin); ret = new_sock->ops->getname(new_sock, (struct sockaddr *) &sin, &slen, 1); diff --git a/fs/ocfs2/cluster/tcp.h b/fs/ocfs2/cluster/tcp.h index 5bada2a..c571e84 100644 --- a/fs/ocfs2/cluster/tcp.h +++ b/fs/ocfs2/cluster/tcp.h @@ -63,6 +63,7 @@ typedef void (o2net_post_msg_handler_func)(int status, void *data, #define O2NET_KEEPALIVE_DELAY_MS_DEFAULT 2000 #define O2NET_IDLE_TIMEOUT_MS_DEFAULT 30000 +#define O2NET_TCP_USER_TIMEOUT 0x7fffffff /* TODO: figure this out.... */ static inline int o2net_link_down(int err, struct socket *sock) diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c index 6f66b37..53e6c40 100644 --- a/fs/ocfs2/ioctl.c +++ b/fs/ocfs2/ioctl.c @@ -35,9 +35,8 @@ copy_to_user((typeof(a) __user *)b, &(a), sizeof(a)) /* - * This call is void because we are already reporting an error that may - * be -EFAULT. The error will be returned from the ioctl(2) call. It's - * just a best-effort to tell userspace that this request caused the error. + * This is just a best-effort to tell userspace that this request + * caused the error. */ static inline void o2info_set_request_error(struct ocfs2_info_request *kreq, struct ocfs2_info_request __user *req) @@ -146,136 +145,105 @@ bail: static int ocfs2_info_handle_blocksize(struct inode *inode, struct ocfs2_info_request __user *req) { - int status = -EFAULT; struct ocfs2_info_blocksize oib; if (o2info_from_user(oib, req)) - goto bail; + return -EFAULT; oib.ib_blocksize = inode->i_sb->s_blocksize; o2info_set_request_filled(&oib.ib_req); if (o2info_to_user(oib, req)) - goto bail; - - status = 0; -bail: - if (status) - o2info_set_request_error(&oib.ib_req, req); + return -EFAULT; - return status; + return 0; } static int ocfs2_info_handle_clustersize(struct inode *inode, struct ocfs2_info_request __user *req) { - int status = -EFAULT; struct ocfs2_info_clustersize oic; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); if (o2info_from_user(oic, req)) - goto bail; + return -EFAULT; oic.ic_clustersize = osb->s_clustersize; o2info_set_request_filled(&oic.ic_req); if (o2info_to_user(oic, req)) - goto bail; - - status = 0; -bail: - if (status) - o2info_set_request_error(&oic.ic_req, req); + return -EFAULT; - return status; + return 0; } static int ocfs2_info_handle_maxslots(struct inode *inode, struct ocfs2_info_request __user *req) { - int status = -EFAULT; struct ocfs2_info_maxslots oim; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); if (o2info_from_user(oim, req)) - goto bail; + return -EFAULT; oim.im_max_slots = osb->max_slots; o2info_set_request_filled(&oim.im_req); if (o2info_to_user(oim, req)) - goto bail; + return -EFAULT; - status = 0; -bail: - if (status) - o2info_set_request_error(&oim.im_req, req); - - return status; + return 0; } static int ocfs2_info_handle_label(struct inode *inode, struct ocfs2_info_request __user *req) { - int status = -EFAULT; struct ocfs2_info_label oil; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); if (o2info_from_user(oil, req)) - goto bail; + return -EFAULT; memcpy(oil.il_label, osb->vol_label, OCFS2_MAX_VOL_LABEL_LEN); o2info_set_request_filled(&oil.il_req); if (o2info_to_user(oil, req)) - goto bail; + return -EFAULT; - status = 0; -bail: - if (status) - o2info_set_request_error(&oil.il_req, req); - - return status; + return 0; } static int ocfs2_info_handle_uuid(struct inode *inode, struct ocfs2_info_request __user *req) { - int status = -EFAULT; struct ocfs2_info_uuid oiu; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); if (o2info_from_user(oiu, req)) - goto bail; + return -EFAULT; memcpy(oiu.iu_uuid_str, osb->uuid_str, OCFS2_TEXT_UUID_LEN + 1); o2info_set_request_filled(&oiu.iu_req); if (o2info_to_user(oiu, req)) - goto bail; - - status = 0; -bail: - if (status) - o2info_set_request_error(&oiu.iu_req, req); + return -EFAULT; - return status; + return 0; } static int ocfs2_info_handle_fs_features(struct inode *inode, struct ocfs2_info_request __user *req) { - int status = -EFAULT; struct ocfs2_info_fs_features oif; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); if (o2info_from_user(oif, req)) - goto bail; + return -EFAULT; oif.if_compat_features = osb->s_feature_compat; oif.if_incompat_features = osb->s_feature_incompat; @@ -284,39 +252,28 @@ static int ocfs2_info_handle_fs_features(struct inode *inode, o2info_set_request_filled(&oif.if_req); if (o2info_to_user(oif, req)) - goto bail; + return -EFAULT; - status = 0; -bail: - if (status) - o2info_set_request_error(&oif.if_req, req); - - return status; + return 0; } static int ocfs2_info_handle_journal_size(struct inode *inode, struct ocfs2_info_request __user *req) { - int status = -EFAULT; struct ocfs2_info_journal_size oij; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); if (o2info_from_user(oij, req)) - goto bail; + return -EFAULT; oij.ij_journal_size = i_size_read(osb->journal->j_inode); o2info_set_request_filled(&oij.ij_req); if (o2info_to_user(oij, req)) - goto bail; + return -EFAULT; - status = 0; -bail: - if (status) - o2info_set_request_error(&oij.ij_req, req); - - return status; + return 0; } static int ocfs2_info_scan_inode_alloc(struct ocfs2_super *osb, @@ -373,7 +330,7 @@ static int ocfs2_info_handle_freeinode(struct inode *inode, u32 i; u64 blkno = -1; char namebuf[40]; - int status = -EFAULT, type = INODE_ALLOC_SYSTEM_INODE; + int status, type = INODE_ALLOC_SYSTEM_INODE; struct ocfs2_info_freeinode *oifi = NULL; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); struct inode *inode_alloc = NULL; @@ -385,8 +342,10 @@ static int ocfs2_info_handle_freeinode(struct inode *inode, goto out_err; } - if (o2info_from_user(*oifi, req)) - goto bail; + if (o2info_from_user(*oifi, req)) { + status = -EFAULT; + goto out_free; + } oifi->ifi_slotnum = osb->max_slots; @@ -424,14 +383,16 @@ static int ocfs2_info_handle_freeinode(struct inode *inode, o2info_set_request_filled(&oifi->ifi_req); - if (o2info_to_user(*oifi, req)) - goto bail; + if (o2info_to_user(*oifi, req)) { + status = -EFAULT; + goto out_free; + } status = 0; bail: if (status) o2info_set_request_error(&oifi->ifi_req, req); - +out_free: kfree(oifi); out_err: return status; @@ -658,7 +619,7 @@ static int ocfs2_info_handle_freefrag(struct inode *inode, { u64 blkno = -1; char namebuf[40]; - int status = -EFAULT, type = GLOBAL_BITMAP_SYSTEM_INODE; + int status, type = GLOBAL_BITMAP_SYSTEM_INODE; struct ocfs2_info_freefrag *oiff; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); @@ -671,8 +632,10 @@ static int ocfs2_info_handle_freefrag(struct inode *inode, goto out_err; } - if (o2info_from_user(*oiff, req)) - goto bail; + if (o2info_from_user(*oiff, req)) { + status = -EFAULT; + goto out_free; + } /* * chunksize from userspace should be power of 2. */ @@ -711,14 +674,14 @@ static int ocfs2_info_handle_freefrag(struct inode *inode, if (o2info_to_user(*oiff, req)) { status = -EFAULT; - goto bail; + goto out_free; } status = 0; bail: if (status) o2info_set_request_error(&oiff->iff_req, req); - +out_free: kfree(oiff); out_err: return status; @@ -727,23 +690,17 @@ out_err: static int ocfs2_info_handle_unknown(struct inode *inode, struct ocfs2_info_request __user *req) { - int status = -EFAULT; struct ocfs2_info_request oir; if (o2info_from_user(oir, req)) - goto bail; + return -EFAULT; o2info_clear_request_filled(&oir); if (o2info_to_user(oir, req)) - goto bail; + return -EFAULT; - status = 0; -bail: - if (status) - o2info_set_request_error(&oir, req); - - return status; + return 0; } /* @@ -381,6 +381,7 @@ static void __propagate_umount(struct mount *mnt) * other children */ if (child && list_empty(&child->mnt_mounts)) { + list_del_init(&child->mnt_child); hlist_del_init_rcu(&child->mnt_hash); hlist_add_before_rcu(&child->mnt_hash, &mnt->mnt_hash); } @@ -65,7 +65,7 @@ int sync_filesystem(struct super_block *sb) return ret; return __sync_filesystem(sb, 1); } -EXPORT_SYMBOL_GPL(sync_filesystem); +EXPORT_SYMBOL(sync_filesystem); static void sync_inodes_one_sb(struct super_block *sb, void *arg) { diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c index 7c580c9..be7d42c 100644 --- a/fs/ufs/inode.c +++ b/fs/ufs/inode.c @@ -902,9 +902,6 @@ void ufs_evict_inode(struct inode * inode) invalidate_inode_buffers(inode); clear_inode(inode); - if (want_delete) { - lock_ufs(inode->i_sb); - ufs_free_inode (inode); - unlock_ufs(inode->i_sb); - } + if (want_delete) + ufs_free_inode(inode); } diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c index 90d74b8..2df62a7 100644 --- a/fs/ufs/namei.c +++ b/fs/ufs/namei.c @@ -126,12 +126,12 @@ static int ufs_symlink (struct inode * dir, struct dentry * dentry, if (l > sb->s_blocksize) goto out_notlocked; - lock_ufs(dir->i_sb); inode = ufs_new_inode(dir, S_IFLNK | S_IRWXUGO); err = PTR_ERR(inode); if (IS_ERR(inode)) - goto out; + goto out_notlocked; + lock_ufs(dir->i_sb); if (l > UFS_SB(sb)->s_uspi->s_maxsymlinklen) { /* slow symlink */ inode->i_op = &ufs_symlink_inode_operations; @@ -181,13 +181,9 @@ static int ufs_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) struct inode * inode; int err; - lock_ufs(dir->i_sb); - inode_inc_link_count(dir); - inode = ufs_new_inode(dir, S_IFDIR|mode); - err = PTR_ERR(inode); if (IS_ERR(inode)) - goto out_dir; + return PTR_ERR(inode); inode->i_op = &ufs_dir_inode_operations; inode->i_fop = &ufs_dir_operations; @@ -195,6 +191,9 @@ static int ufs_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) inode_inc_link_count(inode); + lock_ufs(dir->i_sb); + inode_inc_link_count(dir); + err = ufs_make_empty(inode, dir); if (err) goto out_fail; @@ -212,7 +211,6 @@ out_fail: inode_dec_link_count(inode); inode_dec_link_count(inode); iput (inode); -out_dir: inode_dec_link_count(dir); unlock_ufs(dir->i_sb); goto out; diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index de2d26d..86df952 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -5424,7 +5424,7 @@ xfs_bmap_shift_extents( struct xfs_bmap_free *flist, int num_exts) { - struct xfs_btree_cur *cur; + struct xfs_btree_cur *cur = NULL; struct xfs_bmbt_rec_host *gotp; struct xfs_bmbt_irec got; struct xfs_bmbt_irec left; @@ -5435,7 +5435,7 @@ xfs_bmap_shift_extents( int error = 0; int i; int whichfork = XFS_DATA_FORK; - int logflags; + int logflags = 0; xfs_filblks_t blockcount = 0; int total_extents; @@ -5478,16 +5478,11 @@ xfs_bmap_shift_extents( } } - /* We are going to change core inode */ - logflags = XFS_ILOG_CORE; if (ifp->if_flags & XFS_IFBROOT) { cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork); cur->bc_private.b.firstblock = *firstblock; cur->bc_private.b.flist = flist; cur->bc_private.b.flags = 0; - } else { - cur = NULL; - logflags |= XFS_ILOG_DEXT; } /* @@ -5545,11 +5540,14 @@ xfs_bmap_shift_extents( blockcount = left.br_blockcount + got.br_blockcount; xfs_iext_remove(ip, *current_ext, 1, 0); + logflags |= XFS_ILOG_CORE; if (cur) { error = xfs_btree_delete(cur, &i); if (error) goto del_cursor; XFS_WANT_CORRUPTED_GOTO(i == 1, del_cursor); + } else { + logflags |= XFS_ILOG_DEXT; } XFS_IFORK_NEXT_SET(ip, whichfork, XFS_IFORK_NEXTENTS(ip, whichfork) - 1); @@ -5575,6 +5573,7 @@ xfs_bmap_shift_extents( got.br_startoff = startoff; } + logflags |= XFS_ILOG_CORE; if (cur) { error = xfs_bmbt_update(cur, got.br_startoff, got.br_startblock, @@ -5582,6 +5581,8 @@ xfs_bmap_shift_extents( got.br_state); if (error) goto del_cursor; + } else { + logflags |= XFS_ILOG_DEXT; } (*current_ext)++; @@ -5597,6 +5598,7 @@ del_cursor: xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR); - xfs_trans_log_inode(tp, ip, logflags); + if (logflags) + xfs_trans_log_inode(tp, ip, logflags); return error; } diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 11e9b4c..b984647 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -1753,11 +1753,72 @@ xfs_vm_readpages( return mpage_readpages(mapping, pages, nr_pages, xfs_get_blocks); } +/* + * This is basically a copy of __set_page_dirty_buffers() with one + * small tweak: buffers beyond EOF do not get marked dirty. If we mark them + * dirty, we'll never be able to clean them because we don't write buffers + * beyond EOF, and that means we can't invalidate pages that span EOF + * that have been marked dirty. Further, the dirty state can leak into + * the file interior if the file is extended, resulting in all sorts of + * bad things happening as the state does not match the underlying data. + * + * XXX: this really indicates that bufferheads in XFS need to die. Warts like + * this only exist because of bufferheads and how the generic code manages them. + */ +STATIC int +xfs_vm_set_page_dirty( + struct page *page) +{ + struct address_space *mapping = page->mapping; + struct inode *inode = mapping->host; + loff_t end_offset; + loff_t offset; + int newly_dirty; + + if (unlikely(!mapping)) + return !TestSetPageDirty(page); + + end_offset = i_size_read(inode); + offset = page_offset(page); + + spin_lock(&mapping->private_lock); + if (page_has_buffers(page)) { + struct buffer_head *head = page_buffers(page); + struct buffer_head *bh = head; + + do { + if (offset < end_offset) + set_buffer_dirty(bh); + bh = bh->b_this_page; + offset += 1 << inode->i_blkbits; + } while (bh != head); + } + newly_dirty = !TestSetPageDirty(page); + spin_unlock(&mapping->private_lock); + + if (newly_dirty) { + /* sigh - __set_page_dirty() is static, so copy it here, too */ + unsigned long flags; + + spin_lock_irqsave(&mapping->tree_lock, flags); + if (page->mapping) { /* Race with truncate? */ + WARN_ON_ONCE(!PageUptodate(page)); + account_page_dirtied(page, mapping); + radix_tree_tag_set(&mapping->page_tree, + page_index(page), PAGECACHE_TAG_DIRTY); + } + spin_unlock_irqrestore(&mapping->tree_lock, flags); + __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); + } + return newly_dirty; +} + const struct address_space_operations xfs_address_space_operations = { .readpage = xfs_vm_readpage, .readpages = xfs_vm_readpages, .writepage = xfs_vm_writepage, .writepages = xfs_vm_writepages, + .set_page_dirty = xfs_vm_set_page_dirty, .releasepage = xfs_vm_releasepage, .invalidatepage = xfs_vm_invalidatepage, .write_begin = xfs_vm_write_begin, diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index 2f1e30d..1707980 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -1470,6 +1470,26 @@ xfs_collapse_file_space( start_fsb = XFS_B_TO_FSB(mp, offset + len); shift_fsb = XFS_B_TO_FSB(mp, len); + /* + * Writeback the entire file and force remove any post-eof blocks. The + * writeback prevents changes to the extent list via concurrent + * writeback and the eofblocks trim prevents the extent shift algorithm + * from running into a post-eof delalloc extent. + * + * XXX: This is a temporary fix until the extent shift loop below is + * converted to use offsets and lookups within the ILOCK rather than + * carrying around the index into the extent list for the next + * iteration. + */ + error = filemap_write_and_wait(VFS_I(ip)->i_mapping); + if (error) + return error; + if (xfs_can_free_eofblocks(ip, true)) { + error = xfs_free_eofblocks(mp, ip, false); + if (error) + return error; + } + error = xfs_free_file_space(ip, offset, len); if (error) return error; diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 076b170..de5368c 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -291,12 +291,22 @@ xfs_file_read_iter( if (inode->i_mapping->nrpages) { ret = filemap_write_and_wait_range( VFS_I(ip)->i_mapping, - pos, -1); + pos, pos + size - 1); if (ret) { xfs_rw_iunlock(ip, XFS_IOLOCK_EXCL); return ret; } - truncate_pagecache_range(VFS_I(ip), pos, -1); + + /* + * Invalidate whole pages. This can return an error if + * we fail to invalidate a page, but this should never + * happen on XFS. Warn if it does fail. + */ + ret = invalidate_inode_pages2_range(VFS_I(ip)->i_mapping, + pos >> PAGE_CACHE_SHIFT, + (pos + size - 1) >> PAGE_CACHE_SHIFT); + WARN_ON_ONCE(ret); + ret = 0; } xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL); } @@ -632,10 +642,19 @@ xfs_file_dio_aio_write( if (mapping->nrpages) { ret = filemap_write_and_wait_range(VFS_I(ip)->i_mapping, - pos, -1); + pos, pos + count - 1); if (ret) goto out; - truncate_pagecache_range(VFS_I(ip), pos, -1); + /* + * Invalidate whole pages. This can return an error if + * we fail to invalidate a page, but this should never + * happen on XFS. Warn if it does fail. + */ + ret = invalidate_inode_pages2_range(VFS_I(ip)->i_mapping, + pos >> PAGE_CACHE_SHIFT, + (pos + count - 1) >> PAGE_CACHE_SHIFT); + WARN_ON_ONCE(ret); + ret = 0; } /* |