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]
Date:   Thu, 18 May 2017 18:17:05 -0700
From:   Joel Fernandes <joelaf@...gle.com>
To:     "Rafael J. Wysocki" <rafael@...nel.org>
Cc:     Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 2/2] sched: Use iowait boost policy option in schedutil

Hi Rafael,

On Thu, May 18, 2017 at 6:08 PM, Rafael J. Wysocki <rafael@...nel.org> wrote:
[..]
>>  static struct governor_attr rate_limit_us = __ATTR_RW(rate_limit_us);
>> +static struct governor_attr iowait_boost_enable = __ATTR_RW(iowait_boost_enable);
>>
>>  static struct attribute *sugov_attributes[] = {
>>         &rate_limit_us.attr,
>> +       &iowait_boost_enable.attr,
>>         NULL
>>  };
>>
>> @@ -543,6 +574,9 @@ static int sugov_init(struct cpufreq_policy *policy)
>>                         tunables->rate_limit_us *= lat;
>>         }
>>
>> +       if (policy->iowait_boost_enable)
>> +               tunables->iowait_boost_enable = policy->iowait_boost_enable;
>
> Well, that's just
>
> tunables->iowait_boost_enable = policy->iowait_boost_enable;
>
> so I'm not sure about the role of the if ().

Yes you're right, will fix.

>
> Also, do we only need the policy field as an initial value here?
> That's sort of confusing, because intel_pstate does iowait boosting in
> the active mode too and then it is unconditional.
>

I didn't get your comment. Could you clarify what you meant by
'intel_pstate does iowait boosting in the active mode' ? Is there
another path outside the governor where intel_pstate does iowait
boosting, or are you referring to the hardware behavior?

thanks,

-Joel

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ