lkml.org 
[lkml]   [2021]   [Jan]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH v2] perf script: Fix overrun issue for dynamically-allocated pmu type number
From
Date
Hi Arnaldo, Jiri,

On 12/25/2020 9:10 AM, Jin, Yao wrote:
> Hi Arnaldo, Jiri,
>
> On 12/11/2020 2:10 PM, Adrian Hunter wrote:
>> On 9/12/20 2:58 am, Jin Yao wrote:
>>> When unpacking the event which is from dynamic pmu, the array
>>> output[OUTPUT_TYPE_MAX] may be overrun. For example, type number of
>>> SKL uncore_imc is 10, but OUTPUT_TYPE_MAX is 7 now (OUTPUT_TYPE_MAX =
>>> PERF_TYPE_MAX + 1).
>>>
>>> /* In builtin-script.c */
>>> process_event()
>>> {
>>>          unsigned int type = output_type(attr->type);
>>>
>>>          if (output[type].fields == 0)
>>>                  return;
>>> }
>>>
>>> output[10] is overrun.
>>>
>>> Create a type OUTPUT_TYPE_OTHER for dynamic pmu events, then
>>> output_type(attr->type) will return OUTPUT_TYPE_OTHER here.
>>>
>>> Note that if PERF_TYPE_MAX ever changed, then there would be a conflict
>>> between old perf.data files that had a dynamicaliy allocated PMU number
>>> that would then be the same as a fixed PERF_TYPE.
>>>
>>> Example:
>>>
>>> perf record --switch-events -C 0 -e
>>> "{cpu-clock,uncore_imc/data_reads/,uncore_imc/data_writes/}:SD" -a -- sleep 1
>>> perf script
>>>
>>> Before:
>>>           swapper     0 [000] 1479253.987551:     277766               cpu-clock:
>>> ffffffff9d4ddb6f cpuidle_enter_state+0xdf ([kernel.kallsyms])
>>>           swapper     0 [000] 1479253.987797:     246709               cpu-clock:
>>> ffffffff9d4ddb6f cpuidle_enter_state+0xdf ([kernel.kallsyms])
>>>           swapper     0 [000] 1479253.988127:     329883               cpu-clock:
>>> ffffffff9d4ddb6f cpuidle_enter_state+0xdf ([kernel.kallsyms])
>>>           swapper     0 [000] 1479253.988273:     146393               cpu-clock:
>>> ffffffff9d4ddb6f cpuidle_enter_state+0xdf ([kernel.kallsyms])
>>>           swapper     0 [000] 1479253.988523:     249977               cpu-clock:
>>> ffffffff9d4ddb6f cpuidle_enter_state+0xdf ([kernel.kallsyms])
>>>           swapper     0 [000] 1479253.988877:     354090               cpu-clock:
>>> ffffffff9d4ddb6f cpuidle_enter_state+0xdf ([kernel.kallsyms])
>>>           swapper     0 [000] 1479253.989023:     145940               cpu-clock:
>>> ffffffff9d4ddb6f cpuidle_enter_state+0xdf ([kernel.kallsyms])
>>>           swapper     0 [000] 1479253.989383:     359856               cpu-clock:
>>> ffffffff9d4ddb6f cpuidle_enter_state+0xdf ([kernel.kallsyms])
>>>           swapper     0 [000] 1479253.989523:     140082               cpu-clock:
>>> ffffffff9d4ddb6f cpuidle_enter_state+0xdf ([kernel.kallsyms])
>>>
>>> After:
>>>           swapper     0 [000] 1397040.402011:     272384               cpu-clock:
>>> ffffffff9d4ddb6f cpuidle_enter_state+0xdf ([kernel.kallsyms])
>>>           swapper     0 [000] 1397040.402011:       5396  uncore_imc/data_reads/:
>>>           swapper     0 [000] 1397040.402011:        967 uncore_imc/data_writes/:
>>>           swapper     0 [000] 1397040.402259:     249153               cpu-clock:
>>> ffffffff9d4ddb6f cpuidle_enter_state+0xdf ([kernel.kallsyms])
>>>           swapper     0 [000] 1397040.402259:       7231  uncore_imc/data_reads/:
>>>           swapper     0 [000] 1397040.402259:       1297 uncore_imc/data_writes/:
>>>           swapper     0 [000] 1397040.402508:     249108               cpu-clock:
>>> ffffffff9d4ddb6f cpuidle_enter_state+0xdf ([kernel.kallsyms])
>>>           swapper     0 [000] 1397040.402508:       5333  uncore_imc/data_reads/:
>>>           swapper     0 [000] 1397040.402508:       1008 uncore_imc/data_writes/:
>>>
>>> Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
>>
>> Acked-by: Adrian Hunter <adrian.hunter@intel.com>
>>
>
> Can this patch be accepted? :)
>
> Thanks
> Jin Yao
>

Can this bug fix be accepted or anything else I need to improve?

Thanks
Jin Yao

\
 
 \ /
  Last update: 2021-01-19 01:52    [W:0.054 / U:1.380 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site