linux-kernel - Re: [PATCH] base: arch_topology: Use policy->max to calculate freq

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: <37941182-4f0f-0d3c-1831-83ca9765feb6@arm.com>
Date:   Wed, 24 Nov 2021 16:22:51 +0000
From:   Lukasz Luba <lukasz.luba@....com>
To:     Thara Gopinath <thara.gopinath@...aro.org>, rafael@...nel.org
Cc:     sudeep.holla@....com, gregkh@...uxfoundation.org,
        bjorn.andersson@...aro.org, linux-kernel@...r.kernel.org,
        linux-arm-msm@...r.kernel.org
Subject: Re: [PATCH] base: arch_topology: Use policy->max to calculate
 freq_factor



On 11/17/21 10:12 AM, Lukasz Luba wrote:
> 
> 
> On 11/15/21 8:10 PM, Thara Gopinath wrote:
>> cpuinfo.max_freq can reflect boost frequency if enabled during boot.  
>> Since
>> we don't consider boost frequencies while calculating cpu capacities, use
>> policy->max to populate the freq_factor during boot up.
>>
>> Signed-off-by: Thara Gopinath <thara.gopinath@...aro.org>
>> ---
>>   drivers/base/arch_topology.c | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/base/arch_topology.c b/drivers/base/arch_topology.c
>> index 43407665918f..df818b439bc3 100644
>> --- a/drivers/base/arch_topology.c
>> +++ b/drivers/base/arch_topology.c
>> @@ -334,7 +334,7 @@ init_cpu_capacity_callback(struct notifier_block *nb,
>>       cpumask_andnot(cpus_to_visit, cpus_to_visit, policy->related_cpus);
>>       for_each_cpu(cpu, policy->related_cpus)
>> -        per_cpu(freq_factor, cpu) = policy->cpuinfo.max_freq / 1000;
>> +        per_cpu(freq_factor, cpu) = policy->max / 1000;
>>       if (cpumask_empty(cpus_to_visit)) {
>>           topology_normalize_cpu_scale();
>>
> 
> LGTM
> 
> Reviewed-by: Lukasz Luba <lukasz.luba@...c.com>

Rafael, Thara, please ignore for a while this review.
We are going to do full investigation of this boost frequency,
capacity, schedutil util-to-freq mapping with cpuinfo.max_freq.
The code pointed by Rafael in that sched_util function
already has issue [1]. We have to figure out the consistent
solution for all platforms.

Regards,
Lukasz

[1] 
https://elixir.bootlin.com/linux/v5.16-rc2/source/kernel/sched/cpufreq_schedutil.c#L152