linux-kernel - Re: [PATCH v3 -next] cgroup/rstat: add force idle show helper

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] [day] [month] [year] [list]

Message-ID: <73b9ebd0-74ac-4c2c-a85c-5bd66d4c2f7c@huawei.com>
Date: Thu, 4 Jul 2024 09:22:52 +0800
From: chenridong <chenridong@...wei.com>
To: Waiman Long <longman@...hat.com>, <tj@...nel.org>,
	<lizefan.x@...edance.com>, <hannes@...xchg.org>
CC: <cgroups@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v3 -next] cgroup/rstat: add force idle show helper



On 2024/7/4 9:14, Waiman Long wrote:
> 
> On 7/3/24 20:50, Chen Ridong wrote:
>> In the function cgroup_base_stat_cputime_show, there are five
>> instances of #ifdef, which makes the code not concise.
>> To address this, add the function cgroup_force_idle_show
>> to make the code more succinct.
>>
>> Signed-off-by: Chen Ridong <chenridong@...wei.com>
>> ---
>>   kernel/cgroup/rstat.c | 36 ++++++++++++++++--------------------
>>   1 file changed, 16 insertions(+), 20 deletions(-)
>>
>> diff --git a/kernel/cgroup/rstat.c b/kernel/cgroup/rstat.c
>> index fb8b49437573..1568558e36e5 100644
>> --- a/kernel/cgroup/rstat.c
>> +++ b/kernel/cgroup/rstat.c
>> @@ -594,49 +594,45 @@ static void root_cgroup_cputime(struct 
>> cgroup_base_stat *bstat)
>>       }
>>   }
>> +
>> +static void cgroup_force_idle_show(struct seq_file *seq, struct 
>> cgroup_base_stat *bstat)
>> +{
>> +#ifdef CONFIG_SCHED_CORE
>> +    u64 forceidle_time = bstat->forceidle_sum;
>> +
>> +    do_div(forceidle_time, NSEC_PER_USEC);
>> +    seq_printf(seq, "core_sched.force_idle_usec %llu\n", 
>> forceidle_time);
>> +#endif
>> +}
>> +
>>   void cgroup_base_stat_cputime_show(struct seq_file *seq)
>>   {
>>       struct cgroup *cgrp = seq_css(seq)->cgroup;
>>       u64 usage, utime, stime;
>> -    struct cgroup_base_stat bstat;
>> -#ifdef CONFIG_SCHED_CORE
>> -    u64 forceidle_time;
>> -#endif
>>       if (cgroup_parent(cgrp)) {
>>           cgroup_rstat_flush_hold(cgrp);
>>           usage = cgrp->bstat.cputime.sum_exec_runtime;
>>           cputime_adjust(&cgrp->bstat.cputime, &cgrp->prev_cputime,
>>                      &utime, &stime);
>> -#ifdef CONFIG_SCHED_CORE
>> -        forceidle_time = cgrp->bstat.forceidle_sum;
>> -#endif
>>           cgroup_rstat_flush_release(cgrp);
>>       } else {
>> -        root_cgroup_cputime(&bstat);
>> -        usage = bstat.cputime.sum_exec_runtime;
>> -        utime = bstat.cputime.utime;
>> -        stime = bstat.cputime.stime;
>> -#ifdef CONFIG_SCHED_CORE
>> -        forceidle_time = bstat.forceidle_sum;
>> -#endif
>> +        root_cgroup_cputime(&cgrp->bstat);
>> +        usage = cgrp->bstat.cputime.sum_exec_runtime;
>> +        utime = cgrp->bstat.cputime.utime;
>> +        stime = cgrp->bstat.cputime.stime;
> 
> Just a simple comment saying that you are reusing the root cgroup bstat 
> which is otherwise not being used will help readability.
> 
> Other than that, the change LGTM.
Thank you, Longman. I will do that.

Regards,
Ridong
> 
> Cheers,
> Longman
> 
>>       }
>>       do_div(usage, NSEC_PER_USEC);
>>       do_div(utime, NSEC_PER_USEC);
>>       do_div(stime, NSEC_PER_USEC);
>> -#ifdef CONFIG_SCHED_CORE
>> -    do_div(forceidle_time, NSEC_PER_USEC);
>> -#endif
>>       seq_printf(seq, "usage_usec %llu\n"
>>              "user_usec %llu\n"
>>              "system_usec %llu\n",
>>              usage, utime, stime);
>> -#ifdef CONFIG_SCHED_CORE
>> -    seq_printf(seq, "core_sched.force_idle_usec %llu\n", 
>> forceidle_time);
>> -#endif
>> +    cgroup_force_idle_show(seq, &cgrp->bstat);
>>   }
>>   /* Add bpf kfuncs for cgroup_rstat_updated() and 
>> cgroup_rstat_flush() */
> 
>