The io worker threads are userland threads that just never exit to the userland. By that, they are also assigned to a cgroup (the group of the creating task). When changing the affinity of the io_wq thread via syscall, we must only allow cpumasks within the limits defined by the cpuset controller of the cgroup (if enabled). Fixes: da64d6db3bd3 ("io_uring: One wqe per wq") Signed-off-by: Felix Moessbauer <felix.moessbauer@xxxxxxxxxxx> --- io_uring/io-wq.c | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-) diff --git a/io_uring/io-wq.c b/io_uring/io-wq.c index f1e7c670add8..c7055a8895d7 100644 --- a/io_uring/io-wq.c +++ b/io_uring/io-wq.c @@ -13,6 +13,7 @@ #include <linux/slab.h> #include <linux/rculist_nulls.h> #include <linux/cpu.h> +#include <linux/cpuset.h> #include <linux/task_work.h> #include <linux/audit.h> #include <linux/mmu_context.h> @@ -1322,17 +1323,29 @@ static int io_wq_cpu_offline(unsigned int cpu, struct hlist_node *node) int io_wq_cpu_affinity(struct io_uring_task *tctx, cpumask_var_t mask) { + cpumask_var_t allowed_mask; + int ret = 0; + if (!tctx || !tctx->io_wq) return -EINVAL; + if (!alloc_cpumask_var(&allowed_mask, GFP_KERNEL)) + return -ENOMEM; + rcu_read_lock(); - if (mask) - cpumask_copy(tctx->io_wq->cpu_mask, mask); - else - cpumask_copy(tctx->io_wq->cpu_mask, cpu_possible_mask); + cpuset_cpus_allowed(tctx->io_wq->task, allowed_mask); + if (mask) { + if (cpumask_subset(mask, allowed_mask)) + cpumask_copy(tctx->io_wq->cpu_mask, mask); + else + ret = -EINVAL; + } else { + cpumask_copy(tctx->io_wq->cpu_mask, allowed_mask); + } rcu_read_unlock(); - return 0; + free_cpumask_var(allowed_mask); + return ret; } /* -- 2.39.2