FAILED: patch "[PATCH] io_uring/sqpoll: do not allow pinning outside of cpuset" failed to apply to 5.15-stable tree

The patch below does not apply to the 5.15-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y
git checkout FETCH_HEAD
git cherry-pick -x f011c9cf04c06f16b24f583d313d3c012e589e50
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<[email protected]>' --in-reply-to '2024100131-number-deface-36a6@gregkh' --subject-prefix 'PATCH 5.15.y' HEAD^..

Possible dependencies:

f011c9cf04c0 ("io_uring/sqpoll: do not allow pinning outside of cpuset") 17437f311490 ("io_uring: move SQPOLL related handling into its own file") 59915143e89f ("io_uring: move timeout opcodes and handling into its own file") e418bbc97bff ("io_uring: move our reference counting into a header") 36404b09aa60 ("io_uring: move msg_ring into its own file") f9ead18c1058 ("io_uring: split network related opcodes into its own file") e0da14def1ee ("io_uring: move statx handling to its own file") a9c210cebe13 ("io_uring: move epoll handler to its own file") 4cf90495281b ("io_uring: add a dummy -EOPNOTSUPP prep handler") 99f15d8d6136 ("io_uring: move uring_cmd handling to its own file") cd40cae29ef8 ("io_uring: split out open/close operations") 453b329be5ea ("io_uring: separate out file table handling code") f4c163dd7d4b ("io_uring: split out fadvise/madvise operations") 0d5847274037 ("io_uring: split out fs related sync/fallocate functions") 531113bbd5bf ("io_uring: split out splice related operations") 11aeb71406dd ("io_uring: split out filesystem related operations") e28683bdfc2f ("io_uring: move nop into its own file") 5e2a18d93fec ("io_uring: move xattr related opcodes to its own file") 97b388d70b53 ("io_uring: handle completions in the core") de23077eda61 ("io_uring: set completion results upfront")

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From f011c9cf04c06f16b24f583d313d3c012e589e50 Mon Sep 17 00:00:00 2001
From: Felix Moessbauer <[email protected]>
Date: Mon, 9 Sep 2024 17:00:36 +0200
Subject: [PATCH] io_uring/sqpoll: do not allow pinning outside of cpuset

The submit queue polling threads are userland threads that just never exit to the userland. When creating the thread with IORING_SETUP_SQ_AFF, the affinity of the poller thread is set to the cpu specified in sq_thread_cpu. However, this CPU can be outside of the cpuset defined by the cgroup cpuset controller. This violates the rules defined by the cpuset controller and is a potential issue for realtime applications.

In b7ed6d8ffd6 we fixed the default affinity of the poller thread, in case no explicit pinning is required by inheriting the one of the creating task. In case of explicit pinning, the check is more complicated, as also a cpu outside of the parent cpumask is allowed. We implemented this by using cpuset_cpus_allowed (that has support for cgroup cpusets) and testing if the requested cpu is in the set.

index e545bf240d35..272df9d00f45 100644
@@ -10,6 +10,7 @@
 #include <linux/slab.h>
 #include <linux/audit.h>
 #include <linux/security.h>
+#include <linux/cpuset.h>
 #include <linux/io_uring.h>
 
 #include <uapi/linux/io_uring.h>

@@ -460,10 +461,12 @@ __cold int io_sq_offload_create(struct io_ring_ctx *ctx,
 			return 0;
 
 		if (p->flags & IORING_SETUP_SQ_AFF) {
+			struct cpumask allowed_mask;
 			int cpu = p->sq_thread_cpu;
 
 			ret = -EINVAL;
-			if (cpu >= nr_cpu_ids || !cpu_online(cpu))
+			cpuset_cpus_allowed(current, &allowed_mask);
+			if (!cpumask_test_cpu(cpu, &allowed_mask))
 				goto err_sqpoll;
 			sqd->sq_cpu = cpu;
 		} else {

Here's one for 5.15-stable AND 5.10-stable, remember that any stable io_uring backport for 5.15-stable also applies to 5.10-stable as they use the same codebase.

Jens Axboe

From 0ce44d871d55dc7067d6ef85553ba52510a04337 Mon Sep 17 00:00:00 2001
From: Felix Moessbauer <[email protected]>
Date: Mon, 9 Sep 2024 17:00:36 +0200
Subject: [PATCH] io_uring/sqpoll: do not allow pinning outside of cpuset

The submit queue polling threads are userland threads that just never exit to the userland. When creating the thread with IORING_SETUP_SQ_AFF, the affinity of the poller thread is set to the cpu specified in sq_thread_cpu. However, this CPU can be outside of the cpuset defined by the cgroup cpuset controller. This violates the rules defined by the cpuset controller and is a potential issue for realtime applications.

In b7ed6d8ffd6 we fixed the default affinity of the poller thread, in case no explicit pinning is required by inheriting the one of the creating task. In case of explicit pinning, the check is more complicated, as also a cpu outside of the parent cpumask is allowed. We implemented this by using cpuset_cpus_allowed (that has support for cgroup cpusets) and testing if the requested cpu is in the set.

---
 io_uring/io_uring.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

index 8ed2c6552971..6b6fd244233f 100644
@@ -56,6 +56,7 @@
 #include <linux/mm.h>
 #include <linux/mman.h>
 #include <linux/percpu.h>
+#include <linux/cpuset.h>
 #include <linux/slab.h>
 #include <linux/blkdev.h>
 #include <linux/bvec.h>

@@ -8746,10 +8747,12 @@ static int io_sq_offload_create(struct io_ring_ctx *ctx,
 			return 0;
 
 		if (p->flags & IORING_SETUP_SQ_AFF) {
+			struct cpumask allowed_mask;
 			int cpu = p->sq_thread_cpu;
 
 			ret = -EINVAL;
-			if (cpu >= nr_cpu_ids || !cpu_online(cpu))
+			cpuset_cpus_allowed(current, &allowed_mask);
+			if (!cpumask_test_cpu(cpu, &allowed_mask))
 				goto err_sqpoll;
 			sqd->sq_cpu = cpu;
 		} else {

2.45.2

Both now queued up, thanks.

greg k-h