2022-01-20 05:11:05

by Tianchen Ding

Subject: [PATCH] cpuset: Fix the bug that subpart_cpus updated wrongly in update_cpumask()

subparts_cpus should be limited as a subset of cpus_allowed, but it is
updated wrongly by using cpumask_andnot(). Use cpumask_and() instead to
fix it.

Fixes: ee8dde0cd2ce ("cpuset: Add new v2 cpuset.sched.partition flag")
Signed-off-by: Tianchen Ding <[email protected]>
---
kernel/cgroup/cpuset.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
index bb3531e7fda7..804ff5738c5f 100644
--- a/kernel/cgroup/cpuset.c
+++ b/kernel/cgroup/cpuset.c
@@ -1635,8 +1635,7 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs,
* Make sure that subparts_cpus is a subset of cpus_allowed.
*/
if (cs->nr_subparts_cpus) {
- cpumask_andnot(cs->subparts_cpus, cs->subparts_cpus,
- cs->cpus_allowed);
+ cpumask_and(cs->subparts_cpus, cs->subparts_cpus, cs->cpus_allowed);
cs->nr_subparts_cpus = cpumask_weight(cs->subparts_cpus);
}
spin_unlock_irq(&callback_lock);
--
2.33.0

2022-01-21 07:36:21

by Waiman Long

Subject: Re: [PATCH] cpuset: Fix the bug that subpart_cpus updated wrongly in update_cpumask()

On 1/18/22 05:05, Tianchen Ding wrote:
> subparts_cpus should be limited as a subset of cpus_allowed, but it is
> updated wrongly by using cpumask_andnot(). Use cpumask_and() instead to
> fix it.
>
> Fixes: ee8dde0cd2ce ("cpuset: Add new v2 cpuset.sched.partition flag")
> Signed-off-by: Tianchen Ding <[email protected]>
> ---
> kernel/cgroup/cpuset.c | 3 +--
> 1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
> index bb3531e7fda7..804ff5738c5f 100644
> --- a/kernel/cgroup/cpuset.c
> +++ b/kernel/cgroup/cpuset.c
> @@ -1635,8 +1635,7 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs,
> * Make sure that subparts_cpus is a subset of cpus_allowed.
> */
> if (cs->nr_subparts_cpus) {
> - cpumask_andnot(cs->subparts_cpus, cs->subparts_cpus,
> - cs->cpus_allowed);
> + cpumask_and(cs->subparts_cpus, cs->subparts_cpus, cs->cpus_allowed);
> cs->nr_subparts_cpus = cpumask_weight(cs->subparts_cpus);
> }
> spin_unlock_irq(&callback_lock);

Thanks for the fix.

Reviewed-by: Waiman Long <[email protected]>

2022-01-26 15:23:22

by Tianchen Ding

Subject: Re: [PATCH] cpuset: Fix the bug that subpart_cpus updated wrongly in update_cpumask()

Ping~
Any maintainer take a look?

On 2022/1/18 18:05, Tianchen Ding wrote:
> subparts_cpus should be limited as a subset of cpus_allowed, but it is
> updated wrongly by using cpumask_andnot(). Use cpumask_and() instead to
> fix it.
>
> Fixes: ee8dde0cd2ce ("cpuset: Add new v2 cpuset.sched.partition flag")
> Signed-off-by: Tianchen Ding <[email protected]>
> ---
> kernel/cgroup/cpuset.c | 3 +--
> 1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
> index bb3531e7fda7..804ff5738c5f 100644
> --- a/kernel/cgroup/cpuset.c
> +++ b/kernel/cgroup/cpuset.c
> @@ -1635,8 +1635,7 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs,
> * Make sure that subparts_cpus is a subset of cpus_allowed.
> */
> if (cs->nr_subparts_cpus) {
> - cpumask_andnot(cs->subparts_cpus, cs->subparts_cpus,
> - cs->cpus_allowed);
> + cpumask_and(cs->subparts_cpus, cs->subparts_cpus, cs->cpus_allowed);
> cs->nr_subparts_cpus = cpumask_weight(cs->subparts_cpus);
> }
> spin_unlock_irq(&callback_lock);

2022-01-26 22:19:07

by Tejun Heo

Subject: Re: [PATCH] cpuset: Fix the bug that subpart_cpus updated wrongly in update_cpumask()

On Tue, Jan 18, 2022 at 06:05:18PM +0800, Tianchen Ding wrote:
> subparts_cpus should be limited as a subset of cpus_allowed, but it is
> updated wrongly by using cpumask_andnot(). Use cpumask_and() instead to
> fix it.
>
> Fixes: ee8dde0cd2ce ("cpuset: Add new v2 cpuset.sched.partition flag")
> Signed-off-by: Tianchen Ding <[email protected]>

Applied to cgroup/for-5.17-fixes.

Thanks.

--
tejun