Re: [PATCH RESEND] perf record: Add --buildid-all option

[Adrian Hunter]

On 12/01/16 16:35, Arnaldo Carvalho de Melo wrote:
> Em Tue, Jan 12, 2016 at 02:28:22PM +0200, Adrian Hunter escreveu:
>> On 11/01/16 15:38, Namhyung Kim wrote:
>>> The --buildid-all option is to record build-id of all DSOs in the file.
>>> It might be very costly to postprocess samples to find which DSO hits.
>>>
>>> Cc: Adrian Hunter <adrian.hunter@xxxxxxxxx>
>>> Cc: Andi Kleen <andi@xxxxxxxxxxxxxx>
>>> Cc: Stephane Eranian <eranian@xxxxxxxxxx>
>>> Signed-off-by: Namhyung Kim <namhyung@xxxxxxxxxx>
>>> ---
>>>  tools/perf/Documentation/perf-record.txt |  3 +++
>>>  tools/perf/builtin-record.c              | 32 ++++++++++++++++++++++++++------
>>>  2 files changed, 29 insertions(+), 6 deletions(-)
>>>
>>> diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
>>> index 3a1a32f5479f..fbceb631387c 100644
>>> --- a/tools/perf/Documentation/perf-record.txt
>>> +++ b/tools/perf/Documentation/perf-record.txt
>>> @@ -338,6 +338,9 @@ Options passed to clang when compiling BPF scriptlets.
>>>  Specify vmlinux path which has debuginfo.
>>>  (enabled when BPF prologue is on)
>>>  
>>> +--buildid-all::
>>> +Record build-id of all DSOs regardless whether it's actually hit or not.
>>> +
>>>  SEE ALSO
>>>  --------
>>>  linkperf:perf-stat[1], linkperf:perf-list[1]
>>> diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
>>> index dc4e0adf5c5b..a42cb2955697 100644
>>> --- a/tools/perf/builtin-record.c
>>> +++ b/tools/perf/builtin-record.c
>>> @@ -50,6 +50,7 @@ struct record {
>>>  	int			realtime_prio;
>>>  	bool			no_buildid;
>>>  	bool			no_buildid_cache;
>>> +	bool			buildid_all;
>>>  	unsigned long long	samples;
>>>  };
>>>  
>>> @@ -362,6 +363,13 @@ static int process_buildids(struct record *rec)
>>>  	 */
>>>  	symbol_conf.ignore_vmlinux_buildid = true;
>>>  
>>> +	/*
>>> +	 * If --buildid-all is given, it marks all DSO regardless of hits,
>>> +	 * so no need to process samples.
>>> +	 */
>>> +	if (rec->buildid_all)
>>> +		rec->tool.sample = NULL;
>>
>> I wonder, if we are not processing samples, could the processing could be
>> much simpler?
>> All we need to do is read all the MMAP events, in any order, and create DSOs
>> - no need to
>> create threads or map-groups etc etc.
> 
> Right, if that was the overhead, have you ever measured this?

No, it was just a thought.