linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [RFC] perf record: missing buildid for callstack modules -- again
@ 2016-06-25 15:46 Alexander Lochmann
  0 siblings, 0 replies; only message in thread
From: Alexander Lochmann @ 2016-06-25 15:46 UTC (permalink / raw)
  To: linux-kernel


[-- Attachment #1.1: Type: text/plain, Size: 1275 bytes --]

Hi folks!

As mentioned by this discussion https://lkml.org/lkml/2016/1/7/661, perf
does not record the build id for modules, which are hit on the way down
the callstack.
Sadly that issue is not completely fixed by the cmd argument --buildid-all.
In my case, I have my own kernel module starting a kernel thread. It
does some stuff, and calls schedule from time to time.
If I start perf like this

	perf record -e sched:sched_stat_sleep -e sched:sched_switch -a -g -o
perf.data.raw -p <pid> -- sleep 10

, it records the callstacks for those three events.
Afterwards I want to do some post processing to generate a cold graph:

	perf inject -v -s -i perf.data.raw -o perf.data

That step fails ('failed to write feature 2') due to the lack of a
build-id for my module, which isn't included in perf.data.raw.

Can anyone confirm that this is a bug?
Does anyone have a workaround for it?

Thanks in advance!

Cheers,
Alex

P.S.: Pls put me in CC, because I'm not subscribed to the mailinglist.
-- 
Technische Universität Dortmund
Alexander Lochmann                PGP key: 0xBC3EF6FD
Otto-Hahn-Str. 16                 phone:  +49.231.7556141
D-44227 Dortmund                  fax:    +49.231.7556116
http://ess.cs.tu-dortmund.de/Staff/al


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2016-06-25 16:30 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-06-25 15:46 [RFC] perf record: missing buildid for callstack modules -- again Alexander Lochmann

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).