block/blk-cgroup.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-)
Move the validation check for cpd/pd_alloc_fn and cpd/pd_free_fn function
pairs to the start of blkcg_policy_register(). This ensures we immediately
return -EINVAL if the function pairs are not correctly provided, rather
than returning -ENOSPC after locking and unlocking mutexes unnecessarily.
Co-authored-by: Wen Tao <wentao@uniontech.com>
Signed-off-by: Wen Tao <wentao@uniontech.com>
Signed-off-by: Chen Linxuan <chenlinxuan@uniontech.com>
---
block/blk-cgroup.c | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index 9ed93d91d754..81c166ee003b 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -1727,6 +1727,14 @@ int blkcg_policy_register(struct blkcg_policy *pol)
struct blkcg *blkcg;
int i, ret;
+ /*
+ * Make sure cpd/pd_alloc_fn and cpd/pd_free_fn in pairs, and policy
+ * without pd_alloc_fn/pd_free_fn can't be activated.
+ */
+ if ((!pol->cpd_alloc_fn ^ !pol->cpd_free_fn) ||
+ (!pol->pd_alloc_fn ^ !pol->pd_free_fn))
+ return -EINVAL;
+
mutex_lock(&blkcg_pol_register_mutex);
mutex_lock(&blkcg_pol_mutex);
@@ -1740,14 +1748,6 @@ int blkcg_policy_register(struct blkcg_policy *pol)
goto err_unlock;
}
- /*
- * Make sure cpd/pd_alloc_fn and cpd/pd_free_fn in pairs, and policy
- * without pd_alloc_fn/pd_free_fn can't be activated.
- */
- if ((!pol->cpd_alloc_fn ^ !pol->cpd_free_fn) ||
- (!pol->pd_alloc_fn ^ !pol->pd_free_fn))
- goto err_unlock;
-
/* register @pol */
pol->plid = i;
blkcg_policy[pol->plid] = pol;
--
2.43.0
Hello Linxuan.
On Mon, Feb 10, 2025 at 11:18:27AM +0800, Chen Linxuan <chenlinxuan@uniontech.com> wrote:
> Move the validation check for cpd/pd_alloc_fn and cpd/pd_free_fn function
> pairs to the start of blkcg_policy_register(). This ensures we immediately
> return -EINVAL if the function pairs are not correctly provided, rather
> than returning -ENOSPC after locking and unlocking mutexes unnecessarily.
>
> Co-authored-by: Wen Tao <wentao@uniontech.com>
> Signed-off-by: Wen Tao <wentao@uniontech.com>
> Signed-off-by: Chen Linxuan <chenlinxuan@uniontech.com>
If you consider those locks contention a problem (policy registrations
are "only" boot time, possibly module load time), then it's good to refer
Fixes: e84010732225c ("blkcg: add sanity check for blkcg policy operations")
> ---
> block/blk-cgroup.c | 16 ++++++++--------
> 1 file changed, 8 insertions(+), 8 deletions(-)
But it's correct,
Reviewed-by: Michal Koutný <mkoutny@suse.com>
Hi,
在 2025/02/11 21:57, Michal Koutný 写道:
> Hello Linxuan.
>
> On Mon, Feb 10, 2025 at 11:18:27AM +0800, Chen Linxuan <chenlinxuan@uniontech.com> wrote:
>> Move the validation check for cpd/pd_alloc_fn and cpd/pd_free_fn function
>> pairs to the start of blkcg_policy_register(). This ensures we immediately
>> return -EINVAL if the function pairs are not correctly provided, rather
>> than returning -ENOSPC after locking and unlocking mutexes unnecessarily.
>>
>> Co-authored-by: Wen Tao <wentao@uniontech.com>
>> Signed-off-by: Wen Tao <wentao@uniontech.com>
>> Signed-off-by: Chen Linxuan <chenlinxuan@uniontech.com>
>
> If you consider those locks contention a problem (policy registrations
> are "only" boot time, possibly module load time), then it's good to refer
>
> Fixes: e84010732225c ("blkcg: add sanity check for blkcg policy operations")
This is super cold path, so I don't think it's a problem.
>
>> ---
>> block/blk-cgroup.c | 16 ++++++++--------
>> 1 file changed, 8 insertions(+), 8 deletions(-)
>
> But it's correct,
> Reviewed-by: Michal Koutný <mkoutny@suse.com>
>
Since we're here, can you also change the return value to -ENOMEM from
error path err_free_cpds?
Thanks,
Kuai
© 2016 - 2026 Red Hat, Inc.