On 18/08/2017 08:29, Viresh Kumar wrote:
> On 18 August 2017 at 11:52, Dai Xiang wrote:
[ ... ]
> I can see all of them fail and the reason is quite simple to see. Even
> intel-pstate
> doesn't provide a frequency table and so whatsoever depends on a frequency
> table will fail.
>
> These tests
On 18 August 2017 at 12:03, Dai Xiang wrote:
> On Fri, Aug 18, 2017 at 11:59:47AM +0530, Viresh Kumar wrote:
>> On 18 August 2017 at 11:52, Dai Xiang wrote:
>> > Cool, after use this kernel_cmdline, 4\5\6\8 can pass now, but 1\7\9 still
>> > can not:
>>
>> Why do you say that 4/5/6/8 have passed
On Fri, Aug 18, 2017 at 11:59:47AM +0530, Viresh Kumar wrote:
> On 18 August 2017 at 11:52, Dai Xiang wrote:
> > Cool, after use this kernel_cmdline, 4\5\6\8 can pass now, but 1\7\9 still
> > can not:
>
> Why do you say that 4/5/6/8 have passed ?
###
### cpufreq_04:
### test the change of the fr
On 18 August 2017 at 11:52, Dai Xiang wrote:
> Cool, after use this kernel_cmdline, 4\5\6\8 can pass now, but 1\7\9 still
> can not:
Why do you say that 4/5/6/8 have passed ?
> cpufreq_01.0/cpu0: checking 'scaling_available_frequencies' exists in
> '/sys/devices/system/cpu/cpu0/cpufreq'... Err
On 18/08/2017 08:20, Viresh Kumar wrote:
> On 18 August 2017 at 11:48, Daniel Lezcano wrote:
>> I don't know the passive / active mode but the path changed.
>>
>> /sys/devices/system/cpu/cpufreq/policy[0-9]/scaling_driver
>
> Naah, we wouldn't have broken userspace like that :)
>
> $ ls -l /sys/
On Fri, Aug 18, 2017 at 11:13:35AM +0530, Viresh Kumar wrote:
> On 18 August 2017 at 11:11, Dai Xiang wrote:
>
> > In fact, i has used intel_pstate driver:
> > $ cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_driver
> > intel_pstate
>
> It may be running in *active* mode and you need to put it
On 18 August 2017 at 11:48, Daniel Lezcano wrote:
> I don't know the passive / active mode but the path changed.
>
> /sys/devices/system/cpu/cpufreq/policy[0-9]/scaling_driver
Naah, we wouldn't have broken userspace like that :)
$ ls -l /sys/devices/system/cpu/cpu0/cpufreq
lrwxrwxrwx 1 root root
On 18/08/2017 07:43, Viresh Kumar wrote:
> On 18 August 2017 at 11:11, Dai Xiang wrote:
>
>> In fact, i has used intel_pstate driver:
>> $ cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_driver
>> intel_pstate
>
> It may be running in *active* mode and you need to put it in *passive* mode.
>
>
On 18 August 2017 at 11:11, Dai Xiang wrote:
> In fact, i has used intel_pstate driver:
> $ cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_driver
> intel_pstate
It may be running in *active* mode and you need to put it in *passive* mode.
Try passing following with cmdline:
intel_pstate=passi
On Fri, Aug 18, 2017 at 09:42:19AM +0530, Viresh Kumar wrote:
> On 18 August 2017 at 08:33, Dai Xiang wrote:
> > Hi!
> >
> > I test with below two cpu to test:
> > - Intel(R) Xeon(R) CPU E7-8890 v4 @ 2.20GHz
> > - Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz
> >
> > With below action:
> > make -C cpu
On 18 August 2017 at 08:33, Dai Xiang wrote:
> Hi!
>
> I test with below two cpu to test:
> - Intel(R) Xeon(R) CPU E7-8890 v4 @ 2.20GHz
> - Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz
>
> With below action:
> make -C cpuidle cpuidle_killer
> make -C cpufreq run_tests
>
> And only cpufreq_02 and cpuf
Hi!
I test with below two cpu to test:
- Intel(R) Xeon(R) CPU E7-8890 v4 @ 2.20GHz
- Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz
With below action:
make -C cpuidle cpuidle_killer
make -C cpufreq run_tests
And only cpufreq_02 and cpufreq_03 can pass.
Log as below(ignore pass part):
make: Entering
12 matches
Mail list logo