linux-ext4 - [PATCH v3] ext4: improve trim efficiency

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20230725121848.26865-1-changfengnan@bytedance.com>
Date:   Tue, 25 Jul 2023 20:18:48 +0800
From:   Fengnan Chang <changfengnan@...edance.com>
To:     adilger.kernel@...ger.ca, tytso@....edu, guoqing.jiang@...ux.dev
Cc:     linux-ext4@...r.kernel.org,
        Fengnan Chang <changfengnan@...edance.com>,
        kernel test robot <oliver.sang@...el.com>
Subject: [PATCH v3] ext4: improve trim efficiency

In commit a015434480dc("ext4: send parallel discards on commit
completions"), issue all discard commands in parallel make all
bios could merged into one request, so lowlevel drive can issue
multi segments in one time which is more efficiency, but commit
55cdd0af2bc5 ("ext4: get discard out of jbd2 commit kthread contex")
seems broke this way, let's fix it.
In my test:
1. create 10 normal files, each file size is 10G.
2. deallocate file, punch a 16k holes every 32k.
3. trim all fs.

the time of fstrim fs reduce from 6.7s to 1.3s.

Reported-by: kernel test robot <oliver.sang@...el.com>
Closes: https://lore.kernel.org/oe-lkp/202307171455.ee68ef8b-oliver.sang@intel.com
Signed-off-by: Fengnan Chang <changfengnan@...edance.com>
---
 fs/ext4/mballoc.c | 49 +++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 43 insertions(+), 6 deletions(-)

diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index a2475b8c9fb5..b75ca1df0d30 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -6790,7 +6790,8 @@ int ext4_group_add_blocks(handle_t *handle, struct super_block *sb,
  * be called with under the group lock.
  */
 static int ext4_trim_extent(struct super_block *sb,
-		int start, int count, struct ext4_buddy *e4b)
+		int start, int count, bool noalloc, struct ext4_buddy *e4b,
+		struct bio **biop, struct ext4_free_data **entryp)
 __releases(bitlock)
 __acquires(bitlock)
 {
@@ -6812,9 +6813,16 @@ __acquires(bitlock)
 	 */
 	mb_mark_used(e4b, &ex);
 	ext4_unlock_group(sb, group);
-	ret = ext4_issue_discard(sb, group, start, count, NULL);
+	ret = ext4_issue_discard(sb, group, start, count, biop);
+	if (!ret && !noalloc) {
+		struct ext4_free_data *entry = kmem_cache_alloc(ext4_free_data_cachep,
+				GFP_NOFS|__GFP_NOFAIL);
+		entry->efd_start_cluster = start;
+		entry->efd_count = count;
+		*entryp  = entry;
+	}
+
 	ext4_lock_group(sb, group);
-	mb_free_blocks(NULL, e4b, start, ex.fe_len);
 	return ret;
 }
 
@@ -6824,26 +6832,40 @@ static int ext4_try_to_trim_range(struct super_block *sb,
 __acquires(ext4_group_lock_ptr(sb, e4b->bd_group))
 __releases(ext4_group_lock_ptr(sb, e4b->bd_group))
 {
-	ext4_grpblk_t next, count, free_count;
+	ext4_grpblk_t next, count, free_count, bak;
 	void *bitmap;
+	struct ext4_free_data *entry = NULL, *fd, *nfd;
+	struct list_head discard_data_list;
+	struct bio *discard_bio = NULL;
+	struct blk_plug plug;
+	bool noalloc = false;
+
+	INIT_LIST_HEAD(&discard_data_list);
 
 	bitmap = e4b->bd_bitmap;
 	start = (e4b->bd_info->bb_first_free > start) ?
 		e4b->bd_info->bb_first_free : start;
 	count = 0;
 	free_count = 0;
+	bak = start;
 
+	blk_start_plug(&plug);
 	while (start <= max) {
 		start = mb_find_next_zero_bit(bitmap, max + 1, start);
 		if (start > max)
 			break;
 		next = mb_find_next_bit(bitmap, max + 1, start);
+		/* when only one segment, there is no need to alloc entry */
+		noalloc = (free_count == 0) && (next >= max);
 
 		if ((next - start) >= minblocks) {
-			int ret = ext4_trim_extent(sb, start, next - start, e4b);
+			int ret = ext4_trim_extent(sb, start, next - start, noalloc, e4b,
+							&discard_bio, &entry);
 
-			if (ret && ret != -EOPNOTSUPP)
+			if (ret < 0)
 				break;
+			if (entry)
+				list_add_tail(&entry->efd_list, &discard_data_list);
 			count += next - start;
 		}
 		free_count += next - start;
@@ -6863,6 +6885,21 @@ __releases(ext4_group_lock_ptr(sb, e4b->bd_group))
 		if ((e4b->bd_info->bb_free - free_count) < minblocks)
 			break;
 	}
+	if (discard_bio) {
+		ext4_unlock_group(sb, e4b->bd_group);
+		submit_bio_wait(discard_bio);
+		bio_put(discard_bio);
+		ext4_lock_group(sb, e4b->bd_group);
+	}
+	blk_finish_plug(&plug);
+
+	if (noalloc)
+		mb_free_blocks(NULL, e4b, bak, free_count);
+
+	list_for_each_entry_safe(fd, nfd, &discard_data_list, efd_list) {
+		mb_free_blocks(NULL, e4b, fd->efd_start_cluster, fd->efd_count);
+		kmem_cache_free(ext4_free_data_cachep, fd);
+	}
 
 	return count;
 }
-- 
2.37.1 (Apple Git-137.1)