From: Baokun Li <libaokun1@huawei.com>
To: <linux-ext4@vger.kernel.org>
Cc: <tytso@mit.edu>, <adilger.kernel@dilger.ca>, <jack@suse.cz>,
<linux-kernel@vger.kernel.org>, <ojaswin@linux.ibm.com>,
<julia.lawall@inria.fr>, <yi.zhang@huawei.com>,
<yangerkun@huawei.com>, <libaokun1@huawei.com>,
<libaokun@huaweicloud.com>
Subject: [PATCH v3 04/17] ext4: remove unnecessary s_md_lock on update s_mb_last_group
Date: Mon, 14 Jul 2025 21:03:14 +0800 [thread overview]
Message-ID: <20250714130327.1830534-5-libaokun1@huawei.com> (raw)
In-Reply-To: <20250714130327.1830534-1-libaokun1@huawei.com>
After we optimized the block group lock, we found another lock
contention issue when running will-it-scale/fallocate2 with multiple
processes. The fallocate's block allocation and the truncate's block
release were fighting over the s_md_lock. The problem is, this lock
protects totally different things in those two processes: the list of
freed data blocks (s_freed_data_list) when releasing, and where to start
looking for new blocks (mb_last_group) when allocating.
Now we only need to track s_mb_last_group and no longer need to track
s_mb_last_start, so we don't need the s_md_lock lock to ensure that the
two are consistent. Since s_mb_last_group is merely a hint and doesn't
require strong synchronization, READ_ONCE/WRITE_ONCE is sufficient.
Besides, the s_mb_last_group data type only requires ext4_group_t
(i.e., unsigned int), rendering unsigned long superfluous.
Performance test data follows:
Test: Running will-it-scale/fallocate2 on CPU-bound containers.
Observation: Average fallocate operations per container per second.
|CPU: Kunpeng 920 | P80 | P1 |
|Memory: 512GB |------------------------|-------------------------|
|960GB SSD (0.5GB/s)| base | patched | base | patched |
|-------------------|-------|----------------|--------|----------------|
|mb_optimize_scan=0 | 4821 | 9636 (+99.8%) | 314065 | 337597 (+7.4%) |
|mb_optimize_scan=1 | 4784 | 4834 (+1.04%) | 316344 | 341440 (+7.9%) |
|CPU: AMD 9654 * 2 | P96 | P1 |
|Memory: 1536GB |------------------------|-------------------------|
|960GB SSD (1GB/s) | base | patched | base | patched |
|-------------------|-------|----------------|--------|----------------|
|mb_optimize_scan=0 | 15371 | 22341 (+45.3%) | 205851 | 219707 (+6.7%) |
|mb_optimize_scan=1 | 6101 | 9177 (+50.4%) | 207373 | 215732 (+4.0%) |
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Baokun Li <libaokun1@huawei.com>
---
fs/ext4/ext4.h | 2 +-
fs/ext4/mballoc.c | 12 +++---------
2 files changed, 4 insertions(+), 10 deletions(-)
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index b83095541c98..7f5c070de0fb 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -1630,7 +1630,7 @@ struct ext4_sb_info {
unsigned int s_mb_group_prealloc;
unsigned int s_max_dir_size_kb;
/* where last allocation was done - for stream allocation */
- unsigned long s_mb_last_group;
+ ext4_group_t s_mb_last_group;
unsigned int s_mb_prefetch;
unsigned int s_mb_prefetch_limit;
unsigned int s_mb_best_avail_max_trim_order;
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index e3a5103e1620..025b759ca643 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -2168,11 +2168,8 @@ static void ext4_mb_use_best_found(struct ext4_allocation_context *ac,
ac->ac_buddy_folio = e4b->bd_buddy_folio;
folio_get(ac->ac_buddy_folio);
/* store last allocated for subsequent stream allocation */
- if (ac->ac_flags & EXT4_MB_STREAM_ALLOC) {
- spin_lock(&sbi->s_md_lock);
- sbi->s_mb_last_group = ac->ac_f_ex.fe_group;
- spin_unlock(&sbi->s_md_lock);
- }
+ if (ac->ac_flags & EXT4_MB_STREAM_ALLOC)
+ WRITE_ONCE(sbi->s_mb_last_group, ac->ac_f_ex.fe_group);
/*
* As we've just preallocated more space than
* user requested originally, we store allocated
@@ -2845,10 +2842,7 @@ ext4_mb_regular_allocator(struct ext4_allocation_context *ac)
/* if stream allocation is enabled, use global goal */
if (ac->ac_flags & EXT4_MB_STREAM_ALLOC) {
- /* TBD: may be hot point */
- spin_lock(&sbi->s_md_lock);
- ac->ac_g_ex.fe_group = sbi->s_mb_last_group;
- spin_unlock(&sbi->s_md_lock);
+ ac->ac_g_ex.fe_group = READ_ONCE(sbi->s_mb_last_group);
ac->ac_g_ex.fe_start = -1;
ac->ac_flags &= ~EXT4_MB_HINT_TRY_GOAL;
}
--
2.46.1
next prev parent reply other threads:[~2025-07-14 13:18 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-07-14 13:03 [PATCH v3 00/17] ext4: better scalability for ext4 block allocation Baokun Li
2025-07-14 13:03 ` [PATCH v3 01/17] ext4: add ext4_try_lock_group() to skip busy groups Baokun Li
2025-07-17 10:09 ` Ojaswin Mujoo
2025-07-19 0:37 ` Baokun Li
2025-07-17 22:28 ` Andi Kleen
2025-07-18 3:09 ` Theodore Ts'o
2025-07-19 0:29 ` Baokun Li
2025-07-22 20:59 ` Andi Kleen
2025-07-14 13:03 ` [PATCH v3 02/17] ext4: separate stream goal hits from s_bal_goals for better tracking Baokun Li
2025-07-17 10:29 ` Ojaswin Mujoo
2025-07-19 1:37 ` Baokun Li
2025-07-14 13:03 ` [PATCH v3 03/17] ext4: remove unnecessary s_mb_last_start Baokun Li
2025-07-17 10:31 ` Ojaswin Mujoo
2025-07-14 13:03 ` Baokun Li [this message]
2025-07-17 13:36 ` [PATCH v3 04/17] ext4: remove unnecessary s_md_lock on update s_mb_last_group Ojaswin Mujoo
2025-07-19 1:54 ` Baokun Li
2025-07-14 13:03 ` [PATCH v3 05/17] ext4: utilize multiple global goals to reduce contention Baokun Li
2025-07-14 13:03 ` [PATCH v3 06/17] ext4: get rid of some obsolete EXT4_MB_HINT flags Baokun Li
2025-07-14 13:03 ` [PATCH v3 07/17] ext4: fix typo in CR_GOAL_LEN_SLOW comment Baokun Li
2025-07-14 13:03 ` [PATCH v3 08/17] ext4: convert sbi->s_mb_free_pending to atomic_t Baokun Li
2025-07-14 13:03 ` [PATCH v3 09/17] ext4: merge freed extent with existing extents before insertion Baokun Li
2025-07-14 13:03 ` [PATCH v3 10/17] ext4: fix zombie groups in average fragment size lists Baokun Li
2025-07-14 13:03 ` [PATCH v3 11/17] ext4: fix largest free orders lists corruption on mb_optimize_scan switch Baokun Li
2025-07-14 13:03 ` [PATCH v3 12/17] ext4: factor out __ext4_mb_scan_group() Baokun Li
2025-07-14 13:03 ` [PATCH v3 13/17] ext4: factor out ext4_mb_might_prefetch() Baokun Li
2025-07-14 13:03 ` [PATCH v3 14/17] ext4: factor out ext4_mb_scan_group() Baokun Li
2025-07-14 13:03 ` [PATCH v3 15/17] ext4: convert free groups order lists to xarrays Baokun Li
2025-07-21 11:07 ` Jan Kara
2025-07-21 12:33 ` Baokun Li
2025-07-21 13:45 ` Baokun Li
2025-07-21 18:01 ` Theodore Ts'o
2025-07-22 5:58 ` Baokun Li
2025-07-24 3:55 ` Guenter Roeck
2025-07-24 4:54 ` Theodore Ts'o
2025-07-24 5:20 ` Guenter Roeck
2025-07-24 11:14 ` Zhang Yi
2025-07-24 14:30 ` Guenter Roeck
2025-07-24 14:54 ` Theodore Ts'o
2025-07-25 2:28 ` Zhang Yi
2025-07-26 0:50 ` Baokun Li
2025-07-14 13:03 ` [PATCH v3 16/17] ext4: refactor choose group to scan group Baokun Li
2025-07-14 13:03 ` [PATCH v3 17/17] ext4: implement linear-like traversal across order xarrays Baokun Li
2025-07-15 1:11 ` [PATCH v3 00/17] ext4: better scalability for ext4 block allocation Zhang Yi
2025-07-19 21:45 ` Theodore Ts'o
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20250714130327.1830534-5-libaokun1@huawei.com \
--to=libaokun1@huawei.com \
--cc=adilger.kernel@dilger.ca \
--cc=jack@suse.cz \
--cc=julia.lawall@inria.fr \
--cc=libaokun@huaweicloud.com \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=ojaswin@linux.ibm.com \
--cc=tytso@mit.edu \
--cc=yangerkun@huawei.com \
--cc=yi.zhang@huawei.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®