linux-kernel - Re: [PATCH V2 2/2] cpufreq: Optimize cpufreq_frequency_table

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, 2 Jun 2016 11:28:35 -0700
From:	Steve Muckle <steve.muckle@...aro.org>
To:	Viresh Kumar <viresh.kumar@...aro.org>
Cc:	Steve Muckle <steve.muckle@...aro.org>,
	Rafael Wysocki <rjw@...ysocki.net>,
	linaro-kernel@...ts.linaro.org, linux-pm@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...hat.com>
Subject: Re: [PATCH V2 2/2] cpufreq: Optimize cpufreq_frequency_table_target()

On Thu, Jun 02, 2016 at 06:59:04AM +0530, Viresh Kumar wrote:
> On 01-06-16, 12:46, Steve Muckle wrote:
> > >  	/*
> > >  	 * Find the closest frequency above target_freq.
> > > -	 *
> > > -	 * The table is sorted in the reverse order with respect to the
> > > -	 * frequency and all of the entries are valid (see the initialization).
> > >  	 */
> > > -	entry = policy->freq_table;
> > > -	do {
> > > -		entry++;
> > > -		freq = entry->frequency;
> > > -	} while (freq >= target_freq && freq != CPUFREQ_TABLE_END);
> > > -	entry--;
> > > +	index = cpufreq_frequency_table_target(policy, target_freq,
> > > +					       CPUFREQ_RELATION_L);
> > 
> > This adds a function call to the fast path...
> 
> I understand that, but I am not sure how far should we go to avoid
> that. Open coding routines to save on that isn't a good idea surely.
> 
> I have at least kept this routine in cpufreq.h to avoid a call, but
> eventually we will have at least a call somewhere within it. :(

Shouldn't we be able to avoid extra function calls through the use of
macros/inlines? Otherwise this is making things slower for schedutil
than it is today. 

Actually cpufreq_driver_resolve_freq() shouldn't require any calls from
schedutil when a freq_table is available - the whole thing could be run
inline.