On 04/29/2018 01:12 PM, Jeffrin Jose T wrote:
> Testing script for Intel P-State driver crashes during
> the run. this patch fixes that crash and produce typical results.
>
> Signed-off-by: Jeffrin Jose T <[email protected]>
>
> ---
> tools/testing/selftests/intel_pstate/run.sh | 7 +++++--
> 1 file changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/tools/testing/selftests/intel_pstate/run.sh b/tools/testing/selftests/intel_pstate/run.sh
> index c670359becc6..cd32aca8806f 100755
> --- a/tools/testing/selftests/intel_pstate/run.sh
> +++ b/tools/testing/selftests/intel_pstate/run.sh
> @@ -27,6 +27,7 @@
> # and the test will error out in those cases. The result.X file can be checked
> # for consistency and modified to remove the extra MHz values. The result.X
> # files can be re-evaluated by setting EVALUATE_ONLY to 1 below.
> +#set -xv
>
> EVALUATE_ONLY=0
>
> @@ -101,12 +102,13 @@ cpupower frequency-set -g powersave --max=${max_freq}MHz >& /dev/null
> echo "Target Actual Difference MSR(0x199) max_perf_pct"
> for freq in `seq $max_freq -100 $min_freq`
> do
> - result_freq=$(cat /tmp/result.${freq} | grep "cpu MHz" | awk ' { print $4 } ' | awk -F "." ' { print $1 } ')
> + result_freq=$(cat /tmp/result.${freq} | grep "cpu MHz" \
> | awk '{print $4}' | awk -F "." '{print $1}')
> + result_freq=$(echo $result_freq | head -n1 | cut -d " " -f1)
> msr=$(cat /tmp/result.${freq} | grep "msr" | awk ' { print $3 } ')
> max_perf_pct=$(cat /tmp/result.${freq} | grep "max_perf_pct" | awk ' { print $2 } ' )
> - if [ $result_freq -eq $freq ]; then
> + if (( "$result_freq" == "$freq" ))
> + then
> echo " $freq $result_freq 0 $msr $(($max_perf_pct*3300))"
> else
> echo " $freq $result_freq $(($result_freq-$freq)) $msr $(($max_perf_pct*$max_freq))"
>
Thanks. I will get this into 4.17-rc5
-- Shuah
On 05/04/2018 01:47 PM, Shuah Khan wrote:
> On 04/29/2018 01:12 PM, Jeffrin Jose T wrote:
>> Testing script for Intel P-State driver crashes during
>> the run. this patch fixes that crash and produce typical results.
>>
>> Signed-off-by: Jeffrin Jose T <[email protected]>
>>
>> ---
>> tools/testing/selftests/intel_pstate/run.sh | 7 +++++--
>> 1 file changed, 5 insertions(+), 2 deletions(-)
>>
>> diff --git a/tools/testing/selftests/intel_pstate/run.sh b/tools/testing/selftests/intel_pstate/run.sh
>> index c670359becc6..cd32aca8806f 100755
>> --- a/tools/testing/selftests/intel_pstate/run.sh
>> +++ b/tools/testing/selftests/intel_pstate/run.sh
>> @@ -27,6 +27,7 @@
>> # and the test will error out in those cases. The result.X file can be checked
>> # for consistency and modified to remove the extra MHz values. The result.X
>> # files can be re-evaluated by setting EVALUATE_ONLY to 1 below.
>> +#set -xv
>>
>> EVALUATE_ONLY=0
>>
>> @@ -101,12 +102,13 @@ cpupower frequency-set -g powersave --max=${max_freq}MHz >& /dev/null
>> echo "Target Actual Difference MSR(0x199) max_perf_pct"
>> for freq in `seq $max_freq -100 $min_freq`
>> do
>> - result_freq=$(cat /tmp/result.${freq} | grep "cpu MHz" | awk ' { print $4 } ' | awk -F "." ' { print $1 } ')
>> + result_freq=$(cat /tmp/result.${freq} | grep "cpu MHz" \
>> | awk '{print $4}' | awk -F "." '{print $1}')
>> + result_freq=$(echo $result_freq | head -n1 | cut -d " " -f1)
>> msr=$(cat /tmp/result.${freq} | grep "msr" | awk ' { print $3 } ')
>> max_perf_pct=$(cat /tmp/result.${freq} | grep "max_perf_pct" | awk ' { print $2 } ' )
>> - if [ $result_freq -eq $freq ]; then
>> + if (( "$result_freq" == "$freq" ))
>> + then
>> echo " $freq $result_freq 0 $msr $(($max_perf_pct*3300))"
>> else
>> echo " $freq $result_freq $(($result_freq-$freq)) $msr $(($max_perf_pct*$max_freq))"
>>
>
> Thanks. I will get this into 4.17-rc5
>
> -- Shuah
>
This patch didn't apply. In any case Daniel Diaz sent in a patch a coupld of weeks ago
which did apply and fixes the same problem.
I applied Daniel's patch to linux-kselftest fixes for 4.17-rc5
https://lists.linaro.org/pipermail/linux-kselftest-mirror/2018-April/000858.html
thank you both.
-- Shuah
hello Shuah,
i had later send another version of that patch.
anyway thank you daniel for your effort
thank you both.
On Wed, May 9, 2018 at 1:51 AM, Shuah Khan <[email protected]> wrote:
> On 05/04/2018 01:47 PM, Shuah Khan wrote:
>> On 04/29/2018 01:12 PM, Jeffrin Jose T wrote:
>>> Testing script for Intel P-State driver crashes during
>>> the run. this patch fixes that crash and produce typical results.
>>>
>>> Signed-off-by: Jeffrin Jose T <[email protected]>
>>>
>>> ---
>>> tools/testing/selftests/intel_pstate/run.sh | 7 +++++--
>>> 1 file changed, 5 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/tools/testing/selftests/intel_pstate/run.sh b/tools/testing/selftests/intel_pstate/run.sh
>>> index c670359becc6..cd32aca8806f 100755
>>> --- a/tools/testing/selftests/intel_pstate/run.sh
>>> +++ b/tools/testing/selftests/intel_pstate/run.sh
>>> @@ -27,6 +27,7 @@
>>> # and the test will error out in those cases. The result.X file can be checked
>>> # for consistency and modified to remove the extra MHz values. The result.X
>>> # files can be re-evaluated by setting EVALUATE_ONLY to 1 below.
>>> +#set -xv
>>>
>>> EVALUATE_ONLY=0
>>>
>>> @@ -101,12 +102,13 @@ cpupower frequency-set -g powersave --max=${max_freq}MHz >& /dev/null
>>> echo "Target Actual Difference MSR(0x199) max_perf_pct"
>>> for freq in `seq $max_freq -100 $min_freq`
>>> do
>>> - result_freq=$(cat /tmp/result.${freq} | grep "cpu MHz" | awk ' { print $4 } ' | awk -F "." ' { print $1 } ')
>>> + result_freq=$(cat /tmp/result.${freq} | grep "cpu MHz" \
>>> | awk '{print $4}' | awk -F "." '{print $1}')
>>> + result_freq=$(echo $result_freq | head -n1 | cut -d " " -f1)
>>> msr=$(cat /tmp/result.${freq} | grep "msr" | awk ' { print $3 } ')
>>> max_perf_pct=$(cat /tmp/result.${freq} | grep "max_perf_pct" | awk ' { print $2 } ' )
>>> - if [ $result_freq -eq $freq ]; then
>>> + if (( "$result_freq" == "$freq" ))
>>> + then
>>> echo " $freq $result_freq 0 $msr $(($max_perf_pct*3300))"
>>> else
>>> echo " $freq $result_freq $(($result_freq-$freq)) $msr $(($max_perf_pct*$max_freq))"
>>>
>>
>> Thanks. I will get this into 4.17-rc5
>>
>> -- Shuah
>>
>
> This patch didn't apply. In any case Daniel Diaz sent in a patch a coupld of weeks ago
> which did apply and fixes the same problem.
>
> I applied Daniel's patch to linux-kselftest fixes for 4.17-rc5
>
> https://lists.linaro.org/pipermail/linux-kselftest-mirror/2018-April/000858.html
>
> thank you both.
>
> -- Shuah
--
software engineer
rajagiri school of engineering and technology