[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <E19FF97A-F5A0-4E75-B835-C651AD15AE8C@gmail.com>
Date: Thu, 22 Oct 2015 22:57:35 +0900
From: Taeung Song <treeze.taeung@...il.com>
To: Namhyung Kim <namhyung@...nel.org>
Cc: Ingo Molnar <mingo@...nel.org>,
Arnaldo Carvalho de Melo <acme@...nel.org>,
Peter Zijlstra <a.p.zijlstra@...llo.nl>,
Jiri Olsa <jolsa@...hat.com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
David Ahern <dsahern@...il.com>,
Adrian Hunter <adrian.hunter@...el.com>,
Borislav Petkov <bp@...e.de>,
Chandler Carruth <chandlerc@...il.com>,
Frederic Weisbecker <fweisbec@...il.com>,
Stephane Eranian <eranian@...gle.com>,
Wang Nan <wangnan0@...wei.com>,
Brendan Gregg <brendan.d.gregg@...il.com>
Subject: Re: [PATCH 2/3] perf top: Support call-graph display options also
Hi, Namhyung and Ingo
> On Oct 22, 2015, at 9:20 PM, Namhyung Kim <namhyung@...nel.org> wrote:
>
> On Thu, Oct 22, 2015 at 5:23 PM, Ingo Molnar <mingo@...nel.org> wrote:
>>
>> * Namhyung Kim <namhyung@...nel.org> wrote:
>>
>>> Currently 'perf top --call-graph' option is same as 'perf record'. But
>>> 'perf top' also need to receive display options in 'perf report'. To do
>>> that, change parse_callchain_report_opt() to allow record options too.
>>>
>>> Now perf top can receive display options like below:
>>>
>>> $ perf top --call-graph
>>> Error: option `call-graph' requires a value
>>>
>>> Usage: perf top [<options>]
>>>
>>> --call-graph
>>> <mode[,dump_size],output_type,min_percent[,print_limit],call_order[,branch]>
>>> setup and enables call-graph (stack chain/backtrace)
>>> recording: fp dwarf lbr, output_type (graph, flat,
>>> fractal, or none), min percent threshold, optional
>>> print limit, callchain order, key (function or
>>> address), add branches
>>
>> Yeah, so this fix is nice, and I think we should also do another patch to fix the
>> help text output to be the following:
>>
>> --call-graph
>> <record_mode[,record_size],print_style,limit[,print_limit],call_order[,sort_key]>
>>
>> Set up and enable call graph (call chain, stack backtrace) recording:
>>
>> record_mode: call graph recording mode (fp|dwarf|lbr)
>> record_size: if rec_mode == dwarf, maximum depth of stack recording (bytes),
>> default: 8192 bytes.
>> print_style: call graph printing style (graph|flat|fractal|none)
>> limit: minimum call graph inclusion threshold (percent)
>> print_limit: printing threshold (percent)
>> call_order: call graph order (caller|callee)
>> sort_key: sorting key (function|address|branch)
>>
>> Default: fp,graph,0.5,0.0,caller,function
>>
>> Note that this text evolved a bit over what I sent in my previous mail.
>
> OK. I'll cook a patch to improve help message like this as well as man page.
>
>
>>
>> Also note that I think we should sync up the perf config options to be the same as
>> the option name shortcuts used in this help text.
>
> Ah, good idea.
>
>>
>> Side note: to improve perf config usability, we should probably also recognize
>> underscores in perf config entries, i.e. the following variants should both work:
>>
>> print_percent = 1
>> print-percent = 1
>>
>> Right now only the second one will match.
>
> Taeung, could you consider fixing this in your config patchiest?
>
Sure, no problem.
I thought as below.
Even if one use a print_percent variable, it will be handled like print-percent.
Also if a user change print_percent variable by perf-config command,
print-percent variable will remain in perfconfig file.
(Because perf-config command will rewrite perfconfig file.)
Is anything wrong ?
Thanks,
Taeung
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists