linux-ext4 - [PATCH 0/4] ext4: better scalability for ext4 block allocation

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: <20250523085821.1329392-1-libaokun@huaweicloud.com>
Date: Fri, 23 May 2025 16:58:17 +0800
From: libaokun@...weicloud.com
To: linux-ext4@...r.kernel.org
Cc: tytso@....edu,
	adilger.kernel@...ger.ca,
	jack@...e.cz,
	linux-kernel@...r.kernel.org,
	yi.zhang@...wei.com,
	yangerkun@...wei.com,
	libaokun1@...wei.com,
	libaokun@...weicloud.com
Subject: [PATCH 0/4] ext4: better scalability for ext4 block allocation

From: Baokun Li <libaokun1@...wei.com>

Since servers have more and more CPUs, and we're running more containers
on them, we've been using will-it-scale to test how well ext4 scales. The
fallocate2 test (append 8KB to 1MB, truncate to 0, repeat) run concurrently
on 64 containers revealed significant contention in block allocation/free,
leading to much lower aggregate fallocate OPS compared to a single
container (see below).

   1   |    2   |    4   |    8   |   16   |   32   |   64
-------|--------|--------|--------|--------|--------|-------
295287 | 70665  | 33865  | 19387  | 10104  |  5588  |  3588

The main bottleneck was the ext4_lock_group(), which both block allocation
and free fought over. While the block group for block free is fixed and
unoptimizable, the block group for allocation is selectable. Consequently,
the ext4_try_lock_group() helper function was added to avoid contention on
busy groups, and you can see more in Patch 1.

After we fixed the ext4_lock_group bottleneck, another one showed up:
s_md_lock. This lock protects different data when allocating and freeing
blocks. We got rid of the s_md_lock call in block allocation by making
stream allocation work per inode instead of globally. You can find more
details in Patch 2.

Patches 3 and 4 are just some minor cleanups.

Performance test data follows:

CPU: HUAWEI Kunpeng 920
Memory: 480GB
Disk: 480GB SSD SATA 3.2
Test: Running will-it-scale/fallocate2 on 64 CPU-bound containers.
Observation: Average fallocate operations per container per second.

|--------|--------|--------|--------|--------|--------|--------|--------|
|    -   |    1   |    2   |    4   |    8   |   16   |   32   |   64   |
|--------|--------|--------|--------|--------|--------|--------|--------|
|  base  | 295287 | 70665  | 33865  | 19387  | 10104  |  5588  |  3588  |
|--------|--------|--------|--------|--------|--------|--------|--------|
| linear | 286328 | 123102 | 119542 | 90653  | 60344  | 35302  | 23280  |
|        | -3.0%  | 74.20% | 252.9% | 367.5% | 497.2% | 531.6% | 548.7% |
|--------|--------|--------|--------|--------|--------|--------|--------|
|mb_optim| 292498 | 133305 | 103069 | 61727  | 29702  | 16845  | 10430  |
|ize_scan| -0.9%  | 88.64% | 204.3% | 218.3% | 193.9% | 201.4% | 190.6% |
|--------|--------|--------|--------|--------|--------|--------|--------|

Running "kvm-xfstests -c ext4/all -g auto" showed that 1k/generic/347 often
fails. The test seems to think that a dm-thin device with a virtual size of
5000M but a real size of 500M, after being formatted as ext4, would have
500M free.

But it doesn't – we run out of space after making about 430 1M
files. Since the block size is 1k, making so many files turns on dir_index,
and dm-thin waits a minute, sees no free space, and then throws IO error.
This can cause a directory index block to fail to write and abort journal.

What's worse is that _dmthin_check_fs doesn't replay the journal, so fsck
finds inconsistencies and the test failed. I think the problem is with the
test itself, and I'll send a patch to fix it soon.

Comments and questions are, as always, welcome.

Thanks,
Baokun

Baokun Li (4):
  ext4: add ext4_try_lock_group() to skip busy groups
  ext4: move mb_last_[group|start] to ext4_inode_info
  ext4: get rid of some obsolete EXT4_MB_HINT flags
  ext4: fix typo in CR_GOAL_LEN_SLOW comment

 fs/ext4/ext4.h              | 38 ++++++++++++++++++-------------------
 fs/ext4/mballoc.c           | 34 +++++++++++++++++++--------------
 fs/ext4/super.c             |  2 ++
 include/trace/events/ext4.h |  3 ---
 4 files changed, 41 insertions(+), 36 deletions(-)

-- 
2.46.1