Re: [PATCH v2] sched: Store restrict_cpus_allowed_ptr() call state

[Waiman Long <longman@redhat.com>]

On 1/26/23 15:58, Waiman Long wrote:
> On 1/26/23 15:49, Waiman Long wrote:
>> On 1/26/23 11:11, Will Deacon wrote:
>>> On Tue, Jan 24, 2023 at 03:24:36PM -0500, Waiman Long wrote:
>>>> On 1/24/23 14:48, Will Deacon wrote:
>>>>> 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?
>>> Yes, as in the original report. If, on a 4-CPU system, I do the 
>>> following
>>> with v6.1 and "allow_mismatched_32bit_el0" on the kernel cmdline:
>>>
>>> # for c in `seq 1 3`; do echo 0 > 
>>> /sys/devices/system/cpu/cpu$c/online; done
>>> # yes > /dev/null &
>>> [1] 334
>>> # taskset -p 334
>>> pid 334's current affinity mask: 1
>>> # for c in `seq 1 3`; do echo 1 > 
>>> /sys/devices/system/cpu/cpu$c/online; done
>>> # taskset -p 334
>>> pid 334's current affinity mask: f
>>>
>>> but with v6.2-rc5 that last taskset invocation gives:
>>>
>>> pid 334's current affinity mask: 1
>>>
>>> so, yes, the performance definitely regresses, but that's because the
>>> affinity mask is wrong!
>>
>> I see what you mean now. Hotplug doesn't work quite well now because 
>> user_cpus_ptr has been repurposed to store the value set of 
>> sched_setaffinity() but not the previous cpus_mask before 
>> force_compatible_cpus_allowed_ptr().
>>
>> One possible solution is to modify the hotplug related code to check 
>> for the cpus_allowed_restricted, and if set, check 
>> task_cpu_possible_mask() to see if the cpu can be added back to its 
>> cpus_mask. I will take a further look at that later.
>
> Wait, I think the cpuset hotplug code should be able to restore the 
> right cpumask since task_cpu_possible_mask() is used there. Is cpuset 
> enabled? Does the test works without allow_mismatched_32bit_el0?

BTW, if the test result is from running on a kernel built with the v2 
patch, it is the unexpected result. That should be fixed in the v3 patch.

Cheers,
Longman