On Tue, 2024-10-01 at 09:57 +0200, gregkh@xxxxxxxxxxxxxxxxxxx wrote: > > This is a note to let you know that I've just added the patch titled > > io_uring/sqpoll: do not allow pinning outside of cpuset > > to the 6.6-stable tree which can be found at: > > http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary > > The filename of the patch is: > io_uring-sqpoll-do-not-allow-pinning-outside-of-cpuset.patch > and it can be found in the queue-6.6 subdirectory. > > If you, or anyone else, feels it should not be added to the stable > tree, > please let <stable@xxxxxxxxxxxxxxx> know about it. > > > From f011c9cf04c06f16b24f583d313d3c012e589e50 Mon Sep 17 00:00:00 > 2001 > From: Felix Moessbauer <felix.moessbauer@xxxxxxxxxxx> > Date: Mon, 9 Sep 2024 17:00:36 +0200 > Subject: io_uring/sqpoll: do not allow pinning outside of cpuset > > From: Felix Moessbauer <felix.moessbauer@xxxxxxxxxxx> > > commit f011c9cf04c06f16b24f583d313d3c012e589e50 upstream. > > 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. > > Fixes: 37d1e2e3642e ("io_uring: move SQPOLL thread io-wq forked > worker") > Cc: stable@xxxxxxxxxxxxxxx # 6.1+ > Signed-off-by: Felix Moessbauer <felix.moessbauer@xxxxxxxxxxx> > Link: > https://lore.kernel.org/r/20240909150036.55921-1-felix.moessbauer@xxxxxxxxxxx > Signed-off-by: Jens Axboe <axboe@xxxxxxxxx> > Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> > --- > io_uring/sqpoll.c | 5 ++++- > 1 file changed, 4 insertions(+), 1 deletion(-) > > --- a/io_uring/sqpoll.c > +++ b/io_uring/sqpoll.c > @@ -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> > @@ -401,10 +402,12 @@ __cold int io_sq_offload_create(struct i > 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)) This patch contains a bug which is fixed in "io_uring/sqpoll: retain test for whether the CPU is valid" (upstream a09c17240bdf2e9fa6d0591afa9448b59785f7d4). I don't know why the fix was not tagged stable. Shall I send a09c17240 to stable with a cc stable tag, or do you want to pick that directly for all stable kernels? I'm really sorry for this mess. Felix > + cpuset_cpus_allowed(current, &allowed_mask); > + if (!cpumask_test_cpu(cpu, &allowed_mask)) > goto err_sqpoll; > sqd->sq_cpu = cpu; > } else { > > > Patches currently in stable-queue which might be from > felix.moessbauer@xxxxxxxxxxx are > > queue-6.6/io_uring-io-wq-do-not-allow-pinning-outside-of-cpuse.patch > queue-6.6/io_uring-sqpoll-do-not-allow-pinning-outside-of- > cpuset.patch > queue-6.6/io_uring-io-wq-inherit-cpuset-of-cgroup-in-io-worker.patch -- Siemens AG, Technology Linux Expert Center
