Re: [PATCH v2] mm/shrinker: simplify shrinker_memcg_alloc() using guard()
From: wangxuewen
Date: Thu May 14 2026 - 04:03:30 EST
Hi SeongJae,
Thank you for pointing out the patch changelog convention.
I appreciate the feedback and will make sure to follow the rule of placing version changelogs after the --- separator in all my future submissions, as well as adding links to previous versions.
Thanks,
Wang Xuewen
在 2026/5/14 9:12, SeongJae Park 写道:
On Wed, 13 May 2026 15:52:14 +0800 wangxuewen <18810879172@xxxxxxx> wrote:
Use guard(mutex) to automatically handle shrinker_mutex locking and
unlocking in shrinker_memcg_alloc(). This removes the explicit
mutex_unlock() call, the goto-based error path, and the redundant
ret variable, resulting in cleaner and more concise code.
v2:
- Drop unused 'ret' variable to avoid compiler warning (Muchun Song).
Patch changelog should go to commentary area [1]. Also adding links to
previous versions would be helpful.
Signed-off-by: wangxuewen <wangxuewen@xxxxxxxxxx>
---
mm/shrinker.c | 13 +++++--------
1 file changed, 5 insertions(+), 8 deletions(-)
diff --git a/mm/shrinker.c b/mm/shrinker.c
index 76b3f750cf65..cb03fbecc75d 100644
--- a/mm/shrinker.c
+++ b/mm/shrinker.c
@@ -215,29 +215,26 @@ static DEFINE_IDR(shrinker_idr);
static int shrinker_memcg_alloc(struct shrinker *shrinker)
{
- int id, ret = -ENOMEM;
+ int id;
if (mem_cgroup_disabled())
return -ENOSYS;
if (mem_cgroup_kmem_disabled() && !(shrinker->flags & SHRINKER_NONSLAB))
return -ENOSYS;
- mutex_lock(&shrinker_mutex);
+ guard(mutex)(&shrinker_mutex);
id = idr_alloc(&shrinker_idr, shrinker, 0, 0, GFP_KERNEL);
if (id < 0)
- goto unlock;
+ return id;
Could this change the return value? This code was always returning -ENOMEM
before, but seems it can now return -ENOSPC? Seems that doesn't matter,
though.
if (id >= shrinker_nr_max) {
if (expand_shrinker_info(id)) {
idr_remove(&shrinker_idr, id);
- goto unlock;
+ return -ENOMEM;
}
}
shrinker->id = id;
- ret = 0;
-unlock:
- mutex_unlock(&shrinker_mutex);
- return ret;
+ return 0;
}
[1] https://docs.kernel.org/process/submitting-patches.html#commentary
Thanks,
SJ
[...]