Pull cgroup fix from Tejun Heo: "One low-risk patch to fix a cpuset bug where it over-eagerly tries to modify CPU affinity of kernel threads" * tag 'cgroup-for-6.15-rc6-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup: cgroup/cpuset: Extend kthread_is_per_cpu() check to all PF_NO_SETAFFINITY tasks
Unified · Split
| @@ -1116,9 +1116,11 @@ void cpuset_update_tasks_cpumask(struct cpuset *cs, struct cpumask *new_cpus) · 1115 unmodified lines | ||
| 1116 | 1116 | |
| 1117 | 1117 | if (top_cs) { |
| 1118 | 1118 | /* |
| 1119 | − * Percpu kthreads in top_cpuset are ignored | |
| 1119 | + * PF_NO_SETAFFINITY tasks are ignored. | |
| 1120 | + * All per cpu kthreads should have PF_NO_SETAFFINITY | |
| 1121 | + * flag set, see kthread_set_per_cpu(). | |
| 1120 | 1122 | */ |
| 1121 | − if (kthread_is_per_cpu(task)) | |
| 1123 | + if (task->flags & PF_NO_SETAFFINITY) | |
| 1122 | 1124 | continue; |
| 1123 | 1125 | cpumask_andnot(new_cpus, possible_mask, subpartitions_cpus); |
| 1124 | 1126 | } else { |