linux-kernel - Re: [GIT PULL 0/1] perf/core ui browser improvement

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] [day] [month] [year] [list]

Message-ID: <20100914130753.GA5341@nowhere>
Date:	Tue, 14 Sep 2010 15:07:57 +0200
From:	Frederic Weisbecker <fweisbec@...il.com>
To:	Ingo Molnar <mingo@...e.hu>
Cc:	Peter Zijlstra <peterz@...radead.org>,
	Arnaldo Carvalho de Melo <acme@...radead.org>,
	linux-kernel@...r.kernel.org,
	Christoph Hellwig <hch@...radead.org>,
	Mike Galbraith <efault@....de>,
	Paul Mackerras <paulus@...ba.org>,
	Stephane Eranian <eranian@...gle.com>,
	Tom Zanussi <tzanussi@...il.com>
Subject: Re: [GIT PULL 0/1] perf/core ui browser improvement

On Tue, Sep 14, 2010 at 02:16:11PM +0200, Ingo Molnar wrote:
> 
> * Peter Zijlstra <peterz@...radead.org> wrote:
> 
> > On Mon, 2010-09-13 at 18:29 -0300, Arnaldo Carvalho de Melo wrote:
> > > +      3.00%         find  [kernel.kallsyms]   [k] n_tty_write
> > 
> > > +      3.00%         find  kernel       k n_tty_write
> > 
> > I find that [k] is visually easier to separate from the symbol name.
> 
> me too. Something like:
> 
> before:
> 
>      1.08%           cc1  libc-2.12.so                  [.] _int_malloc
>      0.94%           cc1  [kernel.kallsyms]             [k] clear_page_c
>      0.74%           cc1  libc-2.12.so                  [.] _int_free
>      0.69%           cc1  [kernel.kallsyms]             [k] page_fault
>      0.46%           cc1  libc-2.12.so                  [.] malloc_consolidate
> 
> after:
> 
>      1.08%           cc1  # libc-2.12.so                [.] _int_malloc
>      0.94%           cc1  # kernel                      [k] clear_page_c
>      0.74%           cc1  # libc-2.12.so                [.] _int_free
>      0.69%           cc1  # kernel                      [k] page_fault
>      0.46%           cc1  # libc-2.12.so                [.] malloc_consolidate
> 
> 
> Would largely do the trick i think. The comments delineate the DSO 
> portion from the task portion nicely - while also being a familar 
> pattern. The column of [.] and [k] gives kernel developers a quick 
> glance wrt. whether it's kernel or user-space - and it also delineates 
> the symbol column from the DSO column.
> 
> Thanks,
> 
> 	Ingo


<random ideas for the future>

BTW, having [u] instead of [.] would perhaps be more clearer?
And one day we might want to extend that with more granularity,
like major contexts in uppercase: K=kernel, U=user, H=hypervisor
and some minor in lowercase: h=hardirq, s=softirq, t=task.

So you can get [Kh], [Ks], etc...

User could then sort by context, which could be useful without
callchains for example.


</random ideas for the future>

--
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/