From: Jiri Olsa <jolsa@redhat.com>
To: Aaron Tomlin <atomlin@redhat.com>
Cc: acme@kernel.org, jolsa@redhat.com, mingo@redhat.com,
alexander.shishkin@linux.intel.com, namhyung@kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH/RFC] perf report: Add option to change offset format when address is specified as a sort_key
Date: Sat, 23 Dec 2017 00:59:41 +0100 [thread overview]
Message-ID: <20171222235941.GA22304@krava> (raw)
In-Reply-To: <20171215153510.14499-1-atomlin@redhat.com>
On Fri, Dec 15, 2017 at 03:35:10PM +0000, Aaron Tomlin wrote:
> With --call-graph option, a user can choose to display call chains with a
> specific sort_key. The sort_key can be either: function, address or srcline.
> By default, when the address is specified as the sort_key, the offset (i.e.
> symbol + offset) is provided in decimal (base 10) form.
> This is because in __get_srcline() we use PRIu64 to display the offset.
> This may or may not be desirable.
>
> This patch adds a new option --hex, to change the offset format to
> hexidecimal. Note, this can only be used with the -g, --call-graph option
> when address is specified as a sort_key. When this option is not
> specified, the default behavior is used, which is to display the offset
> in decimal form.
>
> Before:
>
> $ ./perf report -n --fields=overhead,sample,period,pid,symbol --stdio \
> > -g graph,callee,address --percent-limit 10 --no-children
> [ ... ]
> 10.87% 1 506940477 15:migration/3 [k] _spin_trylock
> |
> ---_spin_trylock +13
> double_lock_balance +48
> schedule +2007
> migration_thread +613
> kthread +158
> child_rip +10
>
> After:
>
> $ ./perf report -n --fields=overhead,sample,period,pid,symbol --stdio \
> > -g graph,callee,address --percent-limit 10 --hex --no-children
> [ ... ]
> 10.87% 1 506940477 15:migration/3 [k] _spin_trylock
> |
> ---_spin_trylock +0xd
> double_lock_balance +0x30
> schedule +0x7d7
> migration_thread +0x265
> kthread +0x9e
> child_rip +0xa
not sure we've already discussed that, but this could be default?
if not, I think the hex option should be part of -g option arg, maybe like:
-g graph,callee,xaddress
but default hex seems better to me, thoughts?
jirka
next prev parent reply other threads:[~2017-12-22 23:59 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-12-15 15:35 [PATCH/RFC] perf report: Add option to change offset format when address is specified as a sort_key Aaron Tomlin
2017-12-22 23:59 ` Jiri Olsa [this message]
2017-12-24 18:23 ` Aaron Tomlin
2017-12-25 21:46 ` Jiri Olsa
2017-12-26 14:51 ` Arnaldo Carvalho de Melo
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20171222235941.GA22304@krava \
--to=jolsa@redhat.com \
--cc=acme@kernel.org \
--cc=alexander.shishkin@linux.intel.com \
--cc=atomlin@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox