Re: [PATCH 1/5] perf tools: Count filtered entries to total period also

[Namhyung Kim <namhyung@kernel.org>]

On Thu, 23 Jan 2014 14:21:00 +0100, Jiri Olsa wrote:
> On Thu, Jan 23, 2014 at 08:28:49AM +0900, Namhyung Kim wrote:
>> Currently if a sample was filtered by command line option, it just
>> dropped.  But this affects final output in that the percentage can be
>> different since the filtered entries were not included to the total.
>> 
>> But user might want to see the original percentages when filter
>> applied so change the behavior depends on a new symbol_conf.filter_
>> relative value in order to be controlled by user later.
>> 
>> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
>
> SNIP
>
>> +++ b/tools/perf/util/hist.h
>> @@ -14,6 +14,13 @@ struct hist_entry;
>>  struct addr_location;
>>  struct symbol;
>>  
>> +enum hist_filter {
>> +	HIST_FILTER__DSO,
>> +	HIST_FILTER__THREAD,
>> +	HIST_FILTER__PARENT,
>> +	HIST_FILTER__SYMBOL,
>> +};
>
> in thread__find_addr_map we still have:
> 		...
>                 if ((cpumode == PERF_RECORD_MISC_GUEST_USER ||
>                         cpumode == PERF_RECORD_MISC_GUEST_KERNEL) &&
>                         !perf_guest)
>                         al->filtered = true;
>                 if ((cpumode == PERF_RECORD_MISC_USER ||
>                         cpumode == PERF_RECORD_MISC_KERNEL) &&
>                         !perf_host)
>                         al->filtered = true;
> 		...
>
> seems like we need HIST_FILTER__GUEST|HOST
>
> also in machine__resolve_callchain_sample there's old (and superfluos) init:
> 		...
>                 al.filtered = false;
>                 thread__find_addr_location(thread, machine, cpumode,
>                                            MAP__FUNCTION, ip, &al);
> 		...

Both seems to need to be changed - will fix.

Thanks,
Namhyung