Message-ID: <fdf76fb2-1da4-2d72-7eb3-21137a7d6845@redhat.com>
Date:   Tue, 24 Jan 2023 15:24:36 -0500
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.6.0
Subject: Re: [PATCH v2] sched: Store restrict_cpus_allowed_ptr() call state
Content-Language: en-US
To:     Will Deacon <will@kernel.org>
Cc:     Ingo Molnar <mingo@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Juri Lelli <juri.lelli@redhat.com>,
        Vincent Guittot <vincent.guittot@linaro.org>,
        Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
        Daniel Bristot de Oliveira <bristot@redhat.com>,
        Valentin Schneider <vschneid@redhat.com>,
        Phil Auld <pauld@redhat.com>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        linux-kernel@vger.kernel.org, regressions@lists.linux.dev,
        regressions@leemhuis.info
References: <20230121021749.55313-1-longman@redhat.com>
 <20230124194805.GA27257@willie-the-truck>
From:   Waiman Long <longman@redhat.com>
In-Reply-To: <20230124194805.GA27257@willie-the-truck>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk


On 1/24/23 14:48, Will Deacon wrote:
> Hi Waiman,
>
> [+Thorsten given where we are in the release cycle]
>
> On Fri, Jan 20, 2023 at 09:17:49PM -0500, Waiman Long wrote:
>> The user_cpus_ptr field was originally added by commit b90ca8badbd1
>> ("sched: Introduce task_struct::user_cpus_ptr to track requested
>> affinity"). It was used only by arm64 arch due to possible asymmetric
>> CPU setup.
>>
>> Since commit 8f9ea86fdf99 ("sched: Always preserve the user requested
>> cpumask"), task_struct::user_cpus_ptr is repurposed to store user
>> requested cpu affinity specified in the sched_setaffinity().
>>
>> This results in a performance regression in an arm64 system when booted
>> with "allow_mismatched_32bit_el0" on the command-line. The arch code will
>> (amongst other things) calls force_compatible_cpus_allowed_ptr() and
>> relax_compatible_cpus_allowed_ptr() when exec()'ing a 32-bit or a 64-bit
>> task respectively. Now a call to relax_compatible_cpus_allowed_ptr()
>> will always result in a __sched_setaffinity() call whether there is a
>> previous force_compatible_cpus_allowed_ptr() call or not.
> I'd argue it's more than just a performance regression -- the affinity
> masks are set incorrectly, which is a user visible thing
> (i.e. sched_getaffinity() gives unexpected values).

Can your elaborate a bit more on what you mean by getting unexpected 
sched_getaffinity() results? You mean the result is wrong after a 
relax_compatible_cpus_allowed_ptr(). Right?

sched_getaffinity() just return whatever is in cpus_mask. Normally, it 
should be whatever cpus are allowed by the current cpuset unless 
sched_setaffinity() has been called before. So after a call to 
relax_compatible_cpus_allowed_ptr(), it should revert back to the 
cpu_allowed set in the cpuset. If sched_setaffinity() has been called, 
it should revert back to the intersection of the current cpuset and 
user_cpus_ptr.

Cheers,
Longman