linux-kernel - Re: Re: [PATCH] sched/fair: Correct CPU selection from isolated domain

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <tencent_16253196C5C7F0141593B633CA21A0150505@qq.com>
Date: Tue, 30 Jul 2024 18:11:06 +0800
From: wujing <realwujing@...com>
To: gregkh@...uxfoundation.org
Cc: dongml2@...natelecom.cn,
	linux-kernel@...r.kernel.org,
	menglong8.dong@...il.com,
	mingo@...hat.com,
	peterz@...radead.org,
	realwujing@...com,
	stable@...r.kernel.org,
	yuanql9@...natelecom.cn
Subject: Re: Re: [PATCH] sched/fair: Correct CPU selection from isolated domain

> On Tue, Jul 30, 2024 at 05:40:17PM +0800, wujing wrote:
> > > What "current patch"?
> > >
> > > confused,
> > >
> > > greg k-h
> >
> > The current patch is in my first email.
>
> What message exactly?  I don't see any such message on the stable list.
>
> > Please ignore the previous two emails.
> > The "current patch" mentioned in the earlier emails refers to the upstream
> > status, but the latest upstream patch can no longer be applied to linux-4.19.y.
>
> Again, please read:
>     https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
> for how to do this properly.
>
> thanks,
>
> greg k-h

The email you just replied to is correct.

I reviewed the link in the email, and according to the link,
the patch I submitted meets the third criterion. I have noted
Upstream commit <8aeaffef8c6e> in the patch log.



>From 9d4ecc9314088c2b0aa39c2248fb5e64042f1eef Mon Sep 17 00:00:00 2001
From: wujing <realwujing@...il.com>
Date: Tue, 30 Jul 2024 15:35:53 +0800
Subject: [PATCH] sched/fair: Correct CPU selection from isolated domain

We encountered an issue where the kernel thread `ksmd` runs on the PMD
dedicated isolated core, leading to high latency in OVS packets.

Upon analysis, we discovered that this is caused by the current
select_idle_smt() function not taking the sched_domain mask into account.

Upstream commit <8aeaffef8c6e>

Kernel version: linux-4.19.y

Signed-off-by: wujing <realwujing@...com>
Signed-off-by: QiLiang Yuan <yuanql9@...natelecom.cn>
---
 kernel/sched/fair.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 09f82c84474b..0950cabfc1d0 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6171,7 +6171,8 @@ static int select_idle_smt(struct task_struct *p, struct sched_domain *sd, int t
 		return -1;

 	for_each_cpu(cpu, cpu_smt_mask(target)) {
-		if (!cpumask_test_cpu(cpu, &p->cpus_allowed))
+		if (!cpumask_test_cpu(cpu, &p->cpus_allowed) ||
+			!cpumask_test_cpu(cpu, sched_domain_span(sd)))
 			continue;
 		if (available_idle_cpu(cpu))
 			return cpu;
--
2.45.2