| 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
| ||
|
Message-Id: <20230822015135.50579-1-changfengnan@bytedance.com>
Date: Tue, 22 Aug 2023 09:51:35 +0800
From: Fengnan Chang <changfengnan@...edance.com>
To: tytso@....edu, adilger.kernel@...ger.ca
Cc: linux-ext4@...r.kernel.org,
Fengnan Chang <changfengnan@...edance.com>
Subject: [PATCH v4] 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.
Signed-off-by: Fengnan Chang <changfengnan@...edance.com>
---
fs/ext4/mballoc.c | 95 +++++++++++++++++++++++++----------------------
1 file changed, 51 insertions(+), 44 deletions(-)
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 1e4c667812a9..4fe832ff0e4b 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -6874,70 +6874,61 @@ int ext4_group_add_blocks(handle_t *handle, struct super_block *sb,
return err;
}
-/**
- * ext4_trim_extent -- function to TRIM one single free extent in the group
- * @sb: super block for the file system
- * @start: starting block of the free extent in the alloc. group
- * @count: number of blocks to TRIM
- * @e4b: ext4 buddy for the group
- *
- * Trim "count" blocks starting at "start" in the "group". To assure that no
- * one will allocate those blocks, mark it as used in buddy bitmap. This must
- * be called with under the group lock.
- */
-static int ext4_trim_extent(struct super_block *sb,
- int start, int count, struct ext4_buddy *e4b)
-__releases(bitlock)
-__acquires(bitlock)
-{
- struct ext4_free_extent ex;
- ext4_group_t group = e4b->bd_group;
- int ret = 0;
-
- trace_ext4_trim_extent(sb, group, start, count);
-
- assert_spin_locked(ext4_group_lock_ptr(sb, group));
-
- ex.fe_start = start;
- ex.fe_group = group;
- ex.fe_len = count;
-
- /*
- * Mark blocks used, so no one can reuse them while
- * being trimmed.
- */
- mb_mark_used(e4b, &ex);
- ext4_unlock_group(sb, group);
- ret = ext4_issue_discard(sb, group, start, count, NULL);
- ext4_lock_group(sb, group);
- mb_free_blocks(NULL, e4b, start, ex.fe_len);
- return ret;
-}
-
static int ext4_try_to_trim_range(struct super_block *sb,
struct ext4_buddy *e4b, ext4_grpblk_t start,
ext4_grpblk_t max, ext4_grpblk_t minblocks)
__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;
+ ext4_group_t group = e4b->bd_group;
+ struct ext4_free_extent ex;
+ bool noalloc = false;
+ int ret = 0;
+
+ INIT_LIST_HEAD(&discard_data_list);
bitmap = e4b->bd_bitmap;
start = max(e4b->bd_info->bb_first_free, start);
count = 0;
free_count = 0;
+ blk_start_plug(&plug);
while (start <= max) {
start = mb_find_next_zero_bit(bitmap, max + 1, start);
if (start > max)
break;
+ bak = start;
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);
-
- if (ret && ret != -EOPNOTSUPP)
+ trace_ext4_trim_extent(sb, group, start, next - start);
+ ex.fe_start = start;
+ ex.fe_group = group;
+ ex.fe_len = next - start;
+ /*
+ * Mark blocks used, so no one can reuse them while
+ * being trimmed.
+ */
+ mb_mark_used(e4b, &ex);
+ ext4_unlock_group(sb, group);
+ ret = ext4_issue_discard(sb, group, start, next - start, &discard_bio);
+ if (!noalloc) {
+ entry = kmem_cache_alloc(ext4_free_data_cachep,
+ GFP_NOFS|__GFP_NOFAIL);
+ entry->efd_start_cluster = start;
+ entry->efd_count = next - start;
+ list_add_tail(&entry->efd_list, &discard_data_list);
+ }
+ ext4_lock_group(sb, group);
+ if (ret < 0)
break;
count += next - start;
}
@@ -6959,6 +6950,22 @@ __releases(ext4_group_lock_ptr(sb, e4b->bd_group))
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;
}
--
2.20.1
Powered by blists - more mailing lists