linux-kernel - Re: [PATCH v3 6/9] sched: Kill select_task_rq()'s sd

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: <jhj5zdzpw3b.mognet@arm.com>
Date:   Thu, 16 Apr 2020 11:24:08 +0100
From:   Valentin Schneider <valentin.schneider@....com>
To:     Vincent Guittot <vincent.guittot@...aro.org>
Cc:     linux-kernel <linux-kernel@...r.kernel.org>,
        Ingo Molnar <mingo@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Ingo Molnar <mingo@...hat.com>,
        Juri Lelli <juri.lelli@...hat.com>,
        Steven Rostedt <rostedt@...dmis.org>
Subject: Re: [PATCH v3 6/9] sched: Kill select_task_rq()'s sd_flag parameter


Hi Vincent,

Thanks for taking a look at this.

On 16/04/20 08:42, Vincent Guittot wrote:
>> @@ -6622,13 +6622,25 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
>>   * preempt must be disabled.
>>   */
>>  static int
>> -select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_flags)
>> +select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
>>  {
>> +       int sync = (wake_flags & WF_SYNC) && !(current->flags & PF_EXITING);
>>         struct sched_domain *tmp, *sd = NULL;
>>         int cpu = smp_processor_id();
>>         int new_cpu = prev_cpu;
>>         int want_affine = 0;
>> -       int sync = (wake_flags & WF_SYNC) && !(current->flags & PF_EXITING);
>> +       int sd_flag;
>> +
>> +       switch (wake_flags & (WF_TTWU | WF_FORK | WF_EXEC)) {
>
> You remove a function parameter, which was directly set with the right
> flag, but then you add a switch case to recreate this sd_flag
> internally. Not sure we can say that it's real benefit
>

It is indeed the contentious point of this series (IMO). Still, it has a
few things going for it:

1) only CFS is helped by that extra parameter

2) with patch 9, I need a control flow to pick up the right cached pointer
   anyway; the alternative would be to do something like the unsavoury:

DEFINE_PER_CPU(struct sched_domain __rcu *, sd_balance_flags[3]);
...
update_top_cache_domain()
{
        per_cpu(sd_balance_flags[0], cpu) = highest_flag_domain(cpu, SD_BALANCE_EXEC);
        per_cpu(sd_balance_flags[1], cpu) = highest_flag_domain(cpu, SD_BALANCE_FORK);
        per_cpu(sd_balance_flags[2], cpu) = highest_flag_domain(cpu, SD_BALANCE_WAKE);
}
...
select_task_rq_fair()
{
        // Whatever sort of shady constant time conversion you can think of
        int index = !!(wake_flags & WF_FORK) + 2 * !!(wake_flags & WF_TTWU)
        sd_flag = SD_BALANCE_EXEC << index;
        sd = per_cpu(sd_balance_flags[index], cpu);
}

>> +       case WF_TTWU:
>> +               sd_flag = SD_BALANCE_WAKE;
>> +               break;
>> +       case WF_FORK:
>> +               sd_flag = SD_BALANCE_FORK;
>> +               break;
>> +       default:
>> +               sd_flag = SD_BALANCE_EXEC;
>> +       }
>>
>>         if (sd_flag & SD_BALANCE_WAKE) {
>>                 record_wakee(p);