On 2015/12/22 13:20, Chao Yu wrote:
Hi Jaegeuk, We should update &total_zombie_tree whenever removing unreferenced extent tree during shrinking: - f2fs_shrink_extent_tree if (!atomic_read(&et->refcount)) { ... atomic_dec(&sbi->total_ext_tree); atomic_dec(&sbi->total_zombie_tree); ... } Other parts look good to me. :) Reviewed-by: Chao Yu <chao2.yu@xxxxxxxxxxx> Thanks,-----Original Message----- From: Jaegeuk Kim [mailto:jaegeuk@xxxxxxxxxx] Sent: Tuesday, December 22, 2015 11:39 AM To: linux-kernel@xxxxxxxxxxxxxxx; linux-fsdevel@xxxxxxxxxxxxxxx; linux-f2fs-devel@xxxxxxxxxxxxxxxxxxxxx Cc: Jaegeuk Kim Subject: [f2fs-dev] [PATCH 2/2] f2fs: speed up shrinking extent tree entries If there is no candidates for shrinking slab entries, we don't need to traverse any trees at all. Signed-off-by: Jaegeuk Kim <jaegeuk@xxxxxxxxxx> --- fs/f2fs/extent_cache.c | 12 ++++++++++++ fs/f2fs/f2fs.h | 1 + fs/f2fs/shrinker.c | 2 +- 3 files changed, 14 insertions(+), 1 deletion(-) diff --git a/fs/f2fs/extent_cache.c b/fs/f2fs/extent_cache.c index 0e97d6af..32693af 100644 --- a/fs/f2fs/extent_cache.c +++ b/fs/f2fs/extent_cache.c @@ -71,6 +71,8 @@ static struct extent_tree *__grab_extent_tree(struct inode *inode) atomic_set(&et->refcount, 0); et->count = 0; atomic_inc(&sbi->total_ext_tree); + } else { + atomic_dec(&sbi->total_zombie_tree); } atomic_inc(&et->refcount); up_write(&sbi->extent_tree_lock); @@ -547,10 +549,14 @@ unsigned int f2fs_shrink_extent_tree(struct f2fs_sb_info *sbi, int nr_shrink) unsigned int found; unsigned int node_cnt = 0, tree_cnt = 0; int remained; + bool do_free = false; if (!test_opt(sbi, EXTENT_CACHE)) return 0; + if (!atomic_read(&sbi->total_zombie_tree)) + goto free_node; + if (!down_write_trylock(&sbi->extent_tree_lock)) goto out; @@ -580,6 +586,7 @@ unsigned int f2fs_shrink_extent_tree(struct f2fs_sb_info *sbi, int nr_shrink) } up_write(&sbi->extent_tree_lock); +free_node: /* 2. remove LRU extent entries */ if (!down_write_trylock(&sbi->extent_tree_lock)) goto out; @@ -591,9 +598,13 @@ unsigned int f2fs_shrink_extent_tree(struct f2fs_sb_info *sbi, int nr_shrink) if (!remained--) break; list_del_init(&en->list); + do_free = true; } spin_unlock(&sbi->extent_lock); + if (do_free == false) + goto unlock_out; + /* * reset ino for searching victims from beginning of global extent tree. */ @@ -651,6 +662,7 @@ void f2fs_destroy_extent_tree(struct inode *inode) if (inode->i_nlink && !is_bad_inode(inode) && et->count) { atomic_dec(&et->refcount); + atomic_dec(&sbi->total_zombie_tree); return; }
Hi,all here, sbi->total_ext_tree-- also should change to atomic_dec(&sbi->total_ext_tree); Thanks,
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index a7f6191..90fb970 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -763,6 +763,7 @@ struct f2fs_sb_info { struct list_head extent_list; /* lru list for shrinker */ spinlock_t extent_lock; /* locking extent lru list */ atomic_t total_ext_tree; /* extent tree count */ + atomic_t total_zombie_tree; /* extent zombie tree count */ atomic_t total_ext_node; /* extent info count */ /* basic filesystem units */ diff --git a/fs/f2fs/shrinker.c b/fs/f2fs/shrinker.c index a11e099..93606f2 100644 --- a/fs/f2fs/shrinker.c +++ b/fs/f2fs/shrinker.c @@ -32,7 +32,7 @@ static unsigned long __count_free_nids(struct f2fs_sb_info *sbi) static unsigned long __count_extent_cache(struct f2fs_sb_info *sbi) { - return atomic_read(&sbi->total_ext_tree) + + return atomic_read(&sbi->total_zombie_tree) + atomic_read(&sbi->total_ext_node); } -- 2.5.4 (Apple Git-61) ------------------------------------------------------------------------------ _______________________________________________ Linux-f2fs-devel mailing list Linux-f2fs-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel------------------------------------------------------------------------------ _______________________________________________ Linux-f2fs-devel mailing list Linux-f2fs-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel .
-- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
