linux-kernel - Re: [RFC/PATCH] perf report: Show random usage tip on the help line

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] [thread-next>] [day] [month] [year] [list]

Message-ID: <20160105233753.GB18895@sejong>
Date:	Wed, 6 Jan 2016 08:37:53 +0900
From:	Namhyung Kim <namhyung@...nel.org>
To:	Andi Kleen <andi@...stfloor.org>
CC:	"Wangnan (F)" <wangnan0@...wei.com>,
	Arnaldo Carvalho de Melo <acme@...nel.org>,
	Ingo Molnar <mingo@...nel.org>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	Jiri Olsa <jolsa@...hat.com>,
	LKML <linux-kernel@...r.kernel.org>,
	David Ahern <dsahern@...il.com>,
	Stephane Eranian <eranian@...gle.com>,
	Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>
Subject: Re: [RFC/PATCH] perf report: Show random usage tip on the help line

On Tue, Jan 05, 2016 at 09:52:46PM +0100, Andi Kleen wrote:
> > Btw, does anyone have some tips to add? :)
> 
> Use perf script to show individual samples
> Use perf report --percent-limit 5 to only show entires above 5%
> Use perf record -b -g / perf report --branch-history to show branch events
> Use perf record -b / perf report to show branch mispredictions
> Use perf report -M intel to show Intel style assembler 
> Use perf report --sort srcfile to sort by source file
> Use perf report --sort symbol,srcline to sort by source line
> Use perf record -e '{cycles,cache-misses}:S' / perf report --group to associate events

We should enable --group by default if recorded data file has groups IMHO.

Arnaldo, could you please take this?

  https://lkml.org/lkml/2015/11/29/76


> Use perf mem record / perf mem report for memory address profiling
> Use perf stat -I 1000 to count events in intervals
> Use perf stat -I 1000 -x, to count events and output as CSV
> Use perf timechart record / report to generate high level time line
> Use perf probe --add kfunc / perf record -e kfunc to trace a custom trace point

Thanks for sharing!

> 
> You may also want to adapt some of Brendan's one liners:
> 
> http://www.brendangregg.com/perf.html

Will take a look.

Thanks,
Namhyung
--
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/