1 files changed, 100 insertions, 34 deletions

diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 4c878476bb91..8036d3a84853 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -45,6 +45,7 @@ struct extent_page_data {
         struct bio *bio;
         struct extent_io_tree *tree;
         get_extent_t *get_extent;
+        unsigned long bio_flags;
 
         /* tells writepage not to lock the state bits for this range
          * it still does the unlocking
@@ -64,13 +65,13 @@ tree_fs_info(struct extent_io_tree *tree)
 
 int __init extent_io_init(void)
 {
-        extent_state_cache = kmem_cache_create("extent_state",
+        extent_state_cache = kmem_cache_create("btrfs_extent_state",
                         sizeof(struct extent_state), 0,
                         SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
         if (!extent_state_cache)
                 return -ENOMEM;
 
-        extent_buffer_cache = kmem_cache_create("extent_buffers",
+        extent_buffer_cache = kmem_cache_create("btrfs_extent_buffer",
                         sizeof(struct extent_buffer), 0,
                         SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
         if (!extent_buffer_cache)
@@ -107,6 +108,12 @@ void extent_io_exit(void)
                 list_del(&eb->leak_list);
                 kmem_cache_free(extent_buffer_cache, eb);
         }
+
+        /*
+         * Make sure all delayed rcu free are flushed before we
+         * destroy caches.
+         */
+        rcu_barrier();
         if (extent_state_cache)
                 kmem_cache_destroy(extent_state_cache);
         if (extent_buffer_cache)
@@ -936,6 +943,7 @@ int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, int bits,
  * @end:        the end offset in bytes (inclusive)
  * @bits:       the bits to set in this range
  * @clear_bits: the bits to clear in this range
+ * @cached_state:       state that we're going to cache
  * @mask:       the allocation mask
  *
  * This will go through and set bits for the given range.  If any states exist
@@ -945,7 +953,8 @@ int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, int bits,
  * boundary bits like LOCK.
  */
 int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
-                       int bits, int clear_bits, gfp_t mask)
+                       int bits, int clear_bits,
+                       struct extent_state **cached_state, gfp_t mask)
 {
         struct extent_state *state;
         struct extent_state *prealloc = NULL;
@@ -962,6 +971,15 @@ again:
         }
 
         spin_lock(&tree->lock);
+        if (cached_state && *cached_state) {
+                state = *cached_state;
+                if (state->start <= start && state->end > start &&
+                    state->tree) {
+                        node = &state->rb_node;
+                        goto hit_next;
+                }
+        }
+
         /*
          * this search will find all the extents that end after
          * our range starts.
@@ -992,6 +1010,7 @@ hit_next:
          */
         if (state->start == start && state->end <= end) {
                 set_state_bits(tree, state, &bits);
+                cache_state(state, cached_state);
                 state = clear_state_bit(tree, state, &clear_bits, 0);
                 if (last_end == (u64)-1)
                         goto out;
@@ -1032,6 +1051,7 @@ hit_next:
                         goto out;
                 if (state->end <= end) {
                         set_state_bits(tree, state, &bits);
+                        cache_state(state, cached_state);
                         state = clear_state_bit(tree, state, &clear_bits, 0);
                         if (last_end == (u64)-1)
                                 goto out;
@@ -1070,6 +1090,7 @@ hit_next:
                                    &bits);
                 if (err)
                         extent_io_tree_panic(tree, err);
+                cache_state(prealloc, cached_state);
                 prealloc = NULL;
                 start = this_end + 1;
                 goto search_again;
@@ -1092,6 +1113,7 @@ hit_next:
                         extent_io_tree_panic(tree, err);
 
                 set_state_bits(tree, prealloc, &bits);
+                cache_state(prealloc, cached_state);
                 clear_state_bit(tree, prealloc, &clear_bits, 0);
                 prealloc = NULL;
                 goto out;
@@ -1144,6 +1166,14 @@ int set_extent_delalloc(struct extent_io_tree *tree, u64 start, u64 end,
                               NULL, cached_state, mask);
 }
 
+int set_extent_defrag(struct extent_io_tree *tree, u64 start, u64 end,
+                      struct extent_state **cached_state, gfp_t mask)
+{
+        return set_extent_bit(tree, start, end,
+                              EXTENT_DELALLOC | EXTENT_UPTODATE | EXTENT_DEFRAG,
+                              NULL, cached_state, mask);
+}
+
 int clear_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
                        gfp_t mask)
 {
@@ -1288,18 +1318,42 @@ out:
  * If nothing was found, 1 is returned. If found something, return 0.
  */
 int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
-                          u64 *start_ret, u64 *end_ret, int bits)
+                          u64 *start_ret, u64 *end_ret, int bits,
+                          struct extent_state **cached_state)
 {
         struct extent_state *state;
+        struct rb_node *n;
         int ret = 1;
 
         spin_lock(&tree->lock);
+        if (cached_state && *cached_state) {
+                state = *cached_state;
+                if (state->end == start - 1 && state->tree) {
+                        n = rb_next(&state->rb_node);
+                        while (n) {
+                                state = rb_entry(n, struct extent_state,
+                                                 rb_node);
+                                if (state->state & bits)
+                                        goto got_it;
+                                n = rb_next(n);
+                        }
+                        free_extent_state(*cached_state);
+                        *cached_state = NULL;
+                        goto out;
+                }
+                free_extent_state(*cached_state);
+                *cached_state = NULL;
+        }
+
         state = find_first_extent_bit_state(tree, start, bits);
+got_it:
         if (state) {
+                cache_state(state, cached_state);
                 *start_ret = state->start;
                 *end_ret = state->end;
                 ret = 0;
         }
+out:
         spin_unlock(&tree->lock);
         return ret;
 }
@@ -2062,7 +2116,7 @@ static int bio_readpage_error(struct bio *failed_bio, struct page *page,
                 }
                 read_unlock(&em_tree->lock);
 
-                if (!em || IS_ERR(em)) {
+                if (!em) {
                         kfree(failrec);
                         return -EIO;
                 }
@@ -2298,8 +2352,8 @@ static void end_bio_extent_readpage(struct bio *bio, int err)
                 struct extent_state *cached = NULL;
                 struct extent_state *state;
 
-                pr_debug("end_bio_extent_readpage: bi_vcnt=%d, idx=%d, err=%d, "
-                         "mirror=%ld\n", bio->bi_vcnt, bio->bi_idx, err,
+                pr_debug("end_bio_extent_readpage: bi_sector=%llu, err=%d, "
+                         "mirror=%ld\n", (u64)bio->bi_sector, err,
                          (long int)bio->bi_bdev);
                 tree = &BTRFS_I(page->mapping->host)->io_tree;
 
@@ -2703,12 +2757,15 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
                                          end_bio_extent_readpage, mirror_num,
                                          *bio_flags,
                                          this_bio_flag);
-                        BUG_ON(ret == -ENOMEM);
-                        nr++;
-                        *bio_flags = this_bio_flag;
+                        if (!ret) {
+                                nr++;
+                                *bio_flags = this_bio_flag;
+                        }
                 }
-                if (ret)
+                if (ret) {
                         SetPageError(page);
+                        unlock_extent(tree, cur, cur + iosize - 1);
+                }
                 cur = cur + iosize;
                 pg_offset += iosize;
         }
@@ -3155,12 +3212,16 @@ static int write_one_eb(struct extent_buffer *eb,
         struct block_device *bdev = fs_info->fs_devices->latest_bdev;
         u64 offset = eb->start;
         unsigned long i, num_pages;
+        unsigned long bio_flags = 0;
         int rw = (epd->sync_io ? WRITE_SYNC : WRITE);
         int ret = 0;
 
         clear_bit(EXTENT_BUFFER_IOERR, &eb->bflags);
         num_pages = num_extent_pages(eb->start, eb->len);
         atomic_set(&eb->io_pages, num_pages);
+        if (btrfs_header_owner(eb) == BTRFS_TREE_LOG_OBJECTID)
+                bio_flags = EXTENT_BIO_TREE_LOG;
+
         for (i = 0; i < num_pages; i++) {
                 struct page *p = extent_buffer_page(eb, i);
 
@@ -3169,7 +3230,8 @@ static int write_one_eb(struct extent_buffer *eb,
                 ret = submit_extent_page(rw, eb->tree, p, offset >> 9,
                                          PAGE_CACHE_SIZE, 0, bdev, &epd->bio,
                                          -1, end_bio_extent_buffer_writepage,
-                                         0, 0, 0);
+                                         0, epd->bio_flags, bio_flags);
+                epd->bio_flags = bio_flags;
                 if (ret) {
                         set_bit(EXTENT_BUFFER_IOERR, &eb->bflags);
                         SetPageError(p);
@@ -3204,6 +3266,7 @@ int btree_write_cache_pages(struct address_space *mapping,
                 .tree = tree,
                 .extent_locked = 0,
                 .sync_io = wbc->sync_mode == WB_SYNC_ALL,
+                .bio_flags = 0,
         };
         int ret = 0;
         int done = 0;
@@ -3248,19 +3311,34 @@ retry:
                                 break;
                         }
 
+                        spin_lock(&mapping->private_lock);
+                        if (!PagePrivate(page)) {
+                                spin_unlock(&mapping->private_lock);
+                                continue;
+                        }
+
                         eb = (struct extent_buffer *)page->private;
+
+                        /*
+                         * Shouldn't happen and normally this would be a BUG_ON
+                         * but no sense in crashing the users box for something
+                         * we can survive anyway.
+                         */
                         if (!eb) {
+                                spin_unlock(&mapping->private_lock);
                                 WARN_ON(1);
                                 continue;
                         }
 
-                        if (eb == prev_eb)
+                        if (eb == prev_eb) {
+                                spin_unlock(&mapping->private_lock);
                                 continue;
+                        }
 
-                        if (!atomic_inc_not_zero(&eb->refs)) {
-                                WARN_ON(1);
+                        ret = atomic_inc_not_zero(&eb->refs);
+                        spin_unlock(&mapping->private_lock);
+                        if (!ret)
                                 continue;
-                        }
 
                         prev_eb = eb;
                         ret = lock_extent_buffer_for_io(eb, fs_info, &epd);
@@ -3451,7 +3529,7 @@ static void flush_epd_write_bio(struct extent_page_data *epd)
                 if (epd->sync_io)
                         rw = WRITE_SYNC;
 
-                ret = submit_one_bio(rw, epd->bio, 0, 0);
+                ret = submit_one_bio(rw, epd->bio, 0, epd->bio_flags);
                 BUG_ON(ret < 0); /* -ENOMEM */
                 epd->bio = NULL;
         }
@@ -3474,6 +3552,7 @@ int extent_write_full_page(struct extent_io_tree *tree, struct page *page,
                 .get_extent = get_extent,
                 .extent_locked = 0,
                 .sync_io = wbc->sync_mode == WB_SYNC_ALL,
+                .bio_flags = 0,
         };
 
         ret = __extent_writepage(page, wbc, &epd);
@@ -3498,6 +3577,7 @@ int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode,
                 .get_extent = get_extent,
                 .extent_locked = 1,
                 .sync_io = mode == WB_SYNC_ALL,
+                .bio_flags = 0,
         };
         struct writeback_control wbc_writepages = {
                 .sync_mode      = mode,
@@ -3537,6 +3617,7 @@ int extent_writepages(struct extent_io_tree *tree,
                 .get_extent = get_extent,
                 .extent_locked = 0,
                 .sync_io = wbc->sync_mode == WB_SYNC_ALL,
+                .bio_flags = 0,
         };
 
         ret = extent_write_cache_pages(tree, mapping, wbc,
@@ -3914,18 +3995,6 @@ out:
         return ret;
 }
 
-inline struct page *extent_buffer_page(struct extent_buffer *eb,
-                                              unsigned long i)
-{
-        return eb->pages[i];
-}
-
-inline unsigned long num_extent_pages(u64 start, u64 len)
-{
-        return ((start + len + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT) -
-                (start >> PAGE_CACHE_SHIFT);
-}
-
 static void __free_extent_buffer(struct extent_buffer *eb)
 {
 #if LEAK_DEBUG
@@ -4041,7 +4110,7 @@ struct extent_buffer *alloc_dummy_extent_buffer(u64 start, unsigned long len)
 
         return eb;
 err:
-        for (i--; i > 0; i--)
+        for (i--; i >= 0; i--)
                 __free_page(eb->pages[i]);
         __free_extent_buffer(eb);
         return NULL;
@@ -4186,10 +4255,8 @@ struct extent_buffer *alloc_extent_buffer(struct extent_io_tree *tree,
 
         for (i = 0; i < num_pages; i++, index++) {
                 p = find_or_create_page(mapping, index, GFP_NOFS);
-                if (!p) {
-                        WARN_ON(1);
+                if (!p)
                         goto free_eb;
-                }
 
                 spin_lock(&mapping->private_lock);
                 if (PagePrivate(p)) {
@@ -4332,7 +4399,6 @@ static int release_extent_buffer(struct extent_buffer *eb, gfp_t mask)
 
                 /* Should be safe to release our pages at this point */
                 btrfs_release_extent_buffer_page(eb, 0);
-
                 call_rcu(&eb->rcu_head, btrfs_release_extent_buffer_rcu);
                 return 1;
         }