From: Thomas Renninger <trenn@xxxxxxx> Date: Thu, Apr 29, 2010 at 05:54:16PM +0200 > On Thursday 29 April 2010 17:36:47 Dominik Brodowski wrote: > > Hey, > > > > On Thu, Apr 29, 2010 at 05:13:13PM +0200, Borislav Petkov wrote: > > > > brodo@comet$ sudo ./cpufreq-aperf > > > > brodo@comet$ ./cpufreq-aperf > > > > > > > > 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. > > > > No, it now reports a CPU to be offline even if the call to MSR fails for > > other reasons. > > > > > Right. > > > If we cannot read MSR on CPU 0, something is wrong... > > > > Well, the reason for this is only root can read this MSR, it seems. > This is one letter: > + ret = access(msr_path, F_OK); > must be: > + ret = access(msr_path, R_OK); Right, this should do it. You can only run aperf as root since accessing the MSRs is done over /dev/cpu/0/msr which is (root:root) here. Making it more permissive could open the floodgates on normal users crashing the system by simply fumbling with hardware settings controlled over MSRs. -- 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
