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] [day] [month] [year] [list]
Message-ID: <32d7df47-6d7e-6b7c-f444-9f31e6425ff2@linux.intel.com>
Date:   Mon, 6 Feb 2023 11:25:03 -0500
From:   "Liang, Kan" <kan.liang@...ux.intel.com>
To:     Arnaldo Carvalho de Melo <acme@...nel.org>
Cc:     peterz@...radead.org, mingo@...hat.com,
        linux-kernel@...r.kernel.org, ak@...ux.intel.com,
        eranian@...gle.com, irogers@...gle.com
Subject: Re: [PATCH V3 3/3] perf test: Support the retire_lat check



On 2023-02-06 10:49 a.m., Liang, Kan wrote:
> 
> 
> On 2023-02-06 10:34 a.m., Arnaldo Carvalho de Melo wrote:
>> Em Mon, Feb 06, 2023 at 12:32:54PM -0300, Arnaldo Carvalho de Melo escreveu:
>>> Em Mon, Feb 06, 2023 at 10:17:46AM -0500, Liang, Kan escreveu:
>>>>
>>>>
>>>> On 2023-02-06 10:01 a.m., Arnaldo Carvalho de Melo wrote:
>>>>> Em Thu, Feb 02, 2023 at 11:22:09AM -0800, kan.liang@...ux.intel.com escreveu:
>>>>>> From: Kan Liang <kan.liang@...ux.intel.com>
>>>>>>
>>>>>> Add test for the new field for Retire Latency in the X86 specific test.
>>>>>
>>>>> Is this passing 'perf test' for you?
>>>>
>>>> Ah, it should be the original V2 missed the below change.
>>>
>>> Can you please send this as a separate patch as I already merged
>>> torvalds/master and added more csets on top, so to just fix it and
>>> force push now would be bad.
>>>
>>> Please use what is in my perf/core branch and add a Fixes for that v2
>>> patch.
>>
>> BTW, the 3rd patch with the test is already on the tmp.perf/core branch,
>> that will move to perf/core after the next round of container build
>> tests.
>>
> 
> Thanks. I will sent a V4 to fix the 'perf test' issue.
> 

The V4 is here. It includes a fix for the 'perf test' issue and a fix
for the perf script document.
https://lore.kernel.org/lkml/20230206162100.3329395-1-kan.liang@linux.intel.com/

Sorry again for the inconvenience.

Thanks,
Kan

> Thanks,
> Kan
>> - Arnaldo
>>  
>>> Thanks,
>>>
>>> - Arnaldo
>>>  
>>>> @@ -100,5 +101,25 @@ void arch_perf_synthesize_sample_weight(const
>>>> struct perf_sample *data,
>>>>  	if (type & PERF_SAMPLE_WEIGHT_STRUCT) {
>>>>  		*array &= 0xffffffff;
>>>>  		*array |= ((u64)data->ins_lat << 32);
>>>> +		*array |= ((u64)data->retire_lat << 48);
>>>>  	}
>>>>  }
>>>>
>>>> Could you please remove the V2 and re-apply the V3?
>>>  
>>>> $ sudo ./perf test -v "x86 sample parsing"
>>>>  74: x86 Sample parsing                                              :
>>>> --- start ---
>>>> test child forked, pid 3316797
>>>> test child finished with 0
>>>> ---- end ----
>>>> x86 Sample parsing: Ok
>>>>
>>>>
>>>> Thanks,
>>>> Kan
>>>>
>>>>>
>>>>> [root@...co ~]# perf test -v "x86 sample parsing"
>>>>>  74: x86 Sample parsing                                              :
>>>>> --- start ---
>>>>> test child forked, pid 72526
>>>>> Samples differ at 'retire_lat'
>>>>> parsing failed for sample_type 0x1000000
>>>>> test child finished with -1
>>>>> ---- end ----
>>>>> x86 Sample parsing: FAILED!
>>>>> [root@...co ~]#
>>>>>
>>>>> - Arnaldo
>>>>>  
>>>>>> Signed-off-by: Kan Liang <kan.liang@...ux.intel.com>
>>>>>> ---
>>>>>>
>>>>>> New patch since V2
>>>>>>
>>>>>>  tools/perf/arch/x86/tests/sample-parsing.c | 5 ++++-
>>>>>>  1 file changed, 4 insertions(+), 1 deletion(-)
>>>>>>
>>>>>> diff --git a/tools/perf/arch/x86/tests/sample-parsing.c b/tools/perf/arch/x86/tests/sample-parsing.c
>>>>>> index 690c7c07e90d..a061e8619267 100644
>>>>>> --- a/tools/perf/arch/x86/tests/sample-parsing.c
>>>>>> +++ b/tools/perf/arch/x86/tests/sample-parsing.c
>>>>>> @@ -27,8 +27,10 @@ static bool samples_same(const struct perf_sample *s1,
>>>>>>  			 const struct perf_sample *s2,
>>>>>>  			 u64 type)
>>>>>>  {
>>>>>> -	if (type & PERF_SAMPLE_WEIGHT_STRUCT)
>>>>>> +	if (type & PERF_SAMPLE_WEIGHT_STRUCT) {
>>>>>>  		COMP(ins_lat);
>>>>>> +		COMP(retire_lat);
>>>>>> +	}
>>>>>>  
>>>>>>  	return true;
>>>>>>  }
>>>>>> @@ -48,6 +50,7 @@ static int do_test(u64 sample_type)
>>>>>>  	struct perf_sample sample = {
>>>>>>  		.weight		= 101,
>>>>>>  		.ins_lat        = 102,
>>>>>> +		.retire_lat     = 103,
>>>>>>  	};
>>>>>>  	struct perf_sample sample_out;
>>>>>>  	size_t i, sz, bufsz;
>>>>>> -- 
>>>>>> 2.35.1
>>>>>>
>>>>>
>>>
>>> -- 
>>>
>>> - Arnaldo
>>

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ