linux-kernel - Re: [PATCH] cpufreq: intel_pstate: Fix rounding of core

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: <539862DB.9060905@semaphore.gr>
Date:	Wed, 11 Jun 2014 17:08:27 +0300
From:	Stratos Karafotis <stratosk@...aphore.gr>
To:	Doug Smythies <dsmythies@...us.net>, rjw@...ysocki.net,
	viresh.kumar@...aro.org, dirk.j.brandewie@...el.com
CC:	linux-pm@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] cpufreq: intel_pstate: Fix rounding of core_pct

On 11/06/2014 04:41 μμ, Doug Smythies wrote:
> 
> On 2014.06.11 05:34 Stratos Karafotis wrote:
> 
>> Local variable core_pct holds fixed point values.
>> When we round it we add "1" to core_pct. This has almost
>> no effect.
>>
>> So, add int_toftp(1) to core_pct when rounding.
>>
>> For example, in a given sample point (values taken from
>> tracepoint) with:
>> aperf = 5024
>> mperf = 10619
>>
>> the core_pct is (before rounding):
>> core_pct = 12111
>> fp_toint(core_pct) = 47
>>
>> After rounding:
>> core_pct = 12112
>> fp_toint(core_pct) = 47
>>
>> After rounding with int_toftp(1):
>> core_pct = 12367
>> fp_toint(core_pct) = 48
>>
>> Signed-off-by: Stratos Karafotis <stratosk@...aphore.gr>
>> ---
>>
>> Hi Rafael,
>>
>> I'm sorry for submitting again in merge window, but
>> I thought that maybe we need this fix for 3.16.
>>
>>
>> drivers/cpufreq/intel_pstate.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
>> index 4e7f492..dd80aa2 100644
>> --- a/drivers/cpufreq/intel_pstate.c
>> +++ b/drivers/cpufreq/intel_pstate.c
>> @@ -564,7 +564,7 @@ static inline void intel_pstate_calc_busy(struct cpudata *cpu)
>> 	core_pct = div_u64_rem(core_pct, int_tofp(sample->mperf), &rem);
>>
>> 	if ((rem << 1) >= int_tofp(sample->mperf))
>> -		core_pct += 1;
>> +		core_pct += int_tofp(1);
>>
>> 	sample->freq = fp_toint(
>> 		mul_fp(int_tofp(cpu->pstate.max_pstate * 1000), core_pct));
>> -- 
>> 1.9.3
> 
> No.
> 
> The intent was only ever to round properly the pseudo floating point result of the divide.
> It was much more important (ugh, well 4 times more) when FRACBITS was still 6, which also got changed to 8 in a recent patch.
> 

Are you sure? This rounding was very recently added.
As far as I can understand, I don't see the meaning of this rounding, as is.
Even if FRAC_BITS was 6, I think it would have almost no improvement in
calculations.


Stratos

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/