Re: [PATCH v6.1-rt 00/15] net: Backport backlog-NAPI threads to resolve PREEMPT_RT softirq warning in SMP flush

Re: [PATCH v6.1-rt 00/15] net: Backport backlog-NAPI threads to resolve PREEMPT_RT softirq warning in SMP flush

[Sebastian Andrzej Siewior]

-net folks and lkml
+ linux-rt-devel@lists.linux.dev

On 2026-10-14 17:08:55 [+0200], To wen.yang@linux.dev wrote:
> Please keep the networking folks out of the loop. This is a backport
> request which goes entirely via the RT-queue.
> 
> On 2026-10-14 01:58:34 [+0800], wen.yang@linux.dev wrote:
> > From: Wen Yang <wen.yang@linux.dev>
> > 
> > *** Overview ***
> > Backport the backlog-NAPI smpboot-thread mechanism and its prerequisite
> > NAPI/RPS/defer_list work from upstream (v6.4 - v6.9) to v6.1-rt-next.
> > 
> > *** Problem ***
> > On PREEMPT_RT, RPS steers packets to a remote CPU's backlog by sending an
> > IPI. The IPI handler (rps_trigger_softirq) calls ____napi_schedule(), which
> > raises NET_RX_SOFTIRQ. do_softirq_post_smp_call_flush() then sees a softirq
> > pending that was not pending before the flush, and NET_RX_SOFTIRQ is not
> > masked by SCHED_SOFTIRQ_MASK, so it fires:
> 
> I was about to point out that v6.1-RT should have duct tape for this but
> from reading 1/15 is see that this duct is not complete and is missing
> RPS. Indeed.
> 
> A backport of what is merged is my preference, too. However, we should
> get v6.6-rt fixed first then v6.1-rt. Otherwise users upgrading from
> v6.1 to v6.6 will have this case broken.
> 
> Clark, any objections here from your side?
> 
Sebastian

Re: [PATCH v6.1-rt 00/15] net: Backport backlog-NAPI threads to resolve PREEMPT_RT softirq warning in SMP flush

[Sebastian Andrzej Siewior]

- NET + proper list

On 2026-10-15 08:34:28 [-0500], Clark Williams wrote:
> I don't have a fundamental objection to the backport and agree we should
> do v6.6-rt first, then move that back to v6.1-rt.

Okay.

> So it sounds like we need to pull in the RPS code first?

Yes. Wen, do you mind preparing a series for v6.6-RT?

> Clark

Sebastian

Re: [PATCH v6.1-rt 00/15] net: Backport backlog-NAPI threads to resolve PREEMPT_RT softirq warning in SMP flush

[Wen Yang]

On 10/15/26 21:38, Sebastian Andrzej Siewior wrote:
> - NET + proper list
> 
> On 2026-10-15 08:34:28 [-0500], Clark Williams wrote:
>> I don't have a fundamental objection to the backport and agree we should
>> do v6.6-rt first, then move that back to v6.1-rt.
> 
> Okay.
> 
>> So it sounds like we need to pull in the RPS code first?
> 
> Yes. Wen, do you mind preparing a series for v6.6-RT?
> 

Hi Sebastian, Clark,

Thanks for the feedback.
Agreed, v6.6-rt should be fixed first to avoid reintroducing the issue
when users upgrade from v6.1-rt. Regarding the RPS prerequisites, I will
check which upstream changes are already present in v6.6-rt and identify
the remaining dependencies for the backlog-NAPI threads and defer_list
cleanup. I will prepare and test a series for v6.6-RT soon.

Thanks,
Wen

Re: [PATCH v6.1-rt 00/15] net: Backport backlog-NAPI threads to resolve PREEMPT_RT softirq warning in SMP flush

[Clark Williams]

On Sat, Oct 17, 2026 at 02:16:02AM +0800, Wen Yang wrote:
> 
> 
> On 10/15/26 21:38, Sebastian Andrzej Siewior wrote:
> > - NET + proper list
> > 
> > On 2026-10-15 08:34:28 [-0500], Clark Williams wrote:
> > > I don't have a fundamental objection to the backport and agree we should
> > > do v6.6-rt first, then move that back to v6.1-rt.
> > 
> > Okay.
> > 
> > > So it sounds like we need to pull in the RPS code first?
> > 
> > Yes. Wen, do you mind preparing a series for v6.6-RT?
> > 
> 
> Hi Sebastian, Clark,
> 
> Thanks for the feedback.
> Agreed, v6.6-rt should be fixed first to avoid reintroducing the issue
> when users upgrade from v6.1-rt. Regarding the RPS prerequisites, I will
> check which upstream changes are already present in v6.6-rt and identify
> the remaining dependencies for the backlog-NAPI threads and defer_list
> cleanup. I will prepare and test a series for v6.6-RT soon.
> 
> Thanks,
> Wen

Perfect. I expect you'll have a small series for v6.6-rt and then a
pretty large one for v6.1-rt.

Thanks Wen, I'll be ready to apply/test when you send it. 

Clark

-- 
The United States Coast Guard
Ruining Natural Selection since 1790