linux-kernel - Re: sched_core_balance() releasing interrupts with pi

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: <20220316203524.GK8939@worktop.programming.kicks-ass.net>
Date:   Wed, 16 Mar 2022 21:35:24 +0100
From:   Peter Zijlstra <peterz@...radead.org>
To:     Sebastian Andrzej Siewior <bigeasy@...utronix.de>
Cc:     Steven Rostedt <rostedt@...dmis.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Thomas Gleixner <tglx@...utronix.de>
Subject: Re: sched_core_balance() releasing interrupts with pi_lock held

On Wed, Mar 16, 2022 at 05:18:40PM +0100, Sebastian Andrzej Siewior wrote:

> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> index 33ce5cd113d8..f4675bd8f878 100644
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -5968,6 +5967,9 @@ static bool try_steal_cookie(int this, int that)
>  		if (p == src->core_pick || p == src->curr)
>  			goto next;
>  
> +		if (p->migration_disabled)
> +			goto next;
> +
>  		if (!cpumask_test_cpu(this, &p->cpus_mask))
>  			goto next;
>  
> on top my problems are gone. Let me do some testing and then I would
> patch unless PeterZ does the yelling :)

The previous thing in wrong because it tries to solve the wrong thing,
the above makes sense, except I would write it like so:

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 83872f95a1ea..04c05bc4062e 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -5914,7 +5914,7 @@ static bool try_steal_cookie(int this, int that)
 		if (p == src->core_pick || p == src->curr)
 			goto next;
 
-		if (!cpumask_test_cpu(this, &p->cpus_mask))
+		if (!is_cpu_allowed(p, this))
 			goto next;
 
 		if (p->core_occupation > dst->idle->core_occupation)