Re: [PATCH 3/3] cpufrequtils aperf: Make data collection cpu hotplug proof

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



From: Dominik Brodowski <linux@xxxxxxxxxxxxxxxxxxxx>
Date: Thu, Apr 29, 2010 at 05:01:41PM +0200

> 
> On Thu, Apr 29, 2010 at 01:46:45PM +0200, Thomas Renninger wrote:
> > From: Borislav Petkov <borislav.petkov@xxxxxxx>
> > 
> > Add cpuid() helpers blatantly stolen from the kernel sources and
> > use that for checking for aperf/mperf support instead of accessing
> > "/dev/cpu/%d/cpuid" which requires the module to be enabled on the
> > target system.
> > Signed-off-by: Borislav Petkov <borislav.petkov@xxxxxxx>
> > Signed-off-by: Thomas Renninger <trenn@xxxxxxx>
> > CC: cpufreq@xxxxxxxxxxxxxxx
> > CC: linux@xxxxxxxxxxxxxxxxxxxx
> 
> Hm, the only thing I do not like about this patch is the following:
> 
> brodo@comet$ sudo ./cpufreq-aperf 
> CPU	Average freq(KHz)	Time in C0	Time in Cx	C0 percentage
> 000	1220610			00 sec 282 ms	00 sec 717 ms	28
> 001	1160580			00 sec 316 ms	00 sec 683 ms	31
> 
> brodo@comet$ ./cpufreq-aperf 
> CPU	Average freq(KHz)	Time in C0	Time in Cx	C0 percentage
> 	[offline]
> 	[offline]
> 
> We should abort if we can't read the MSR, not loop with reporting "all cores
> are offline".

This shouldn't be happening actually, since you can't take cpu 0
offline, IOW, it should be still showing CPU0 online.

> Also, should we still print the CPU no for offline CPUs?

IMHO it would be more informative if we print the cpu number even if the
cpu is offline.

-- 
Regards/Gruss,
Boris.

--
Advanced Micro Devices, Inc.
Operating Systems Research Center
--
To unsubscribe from this list: send the line "unsubscribe cpufreq" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[Index of Archives]     [Linux Kernel Devel]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Forum]     [Linux SCSI]

  Powered by Linux