All of lore.kernel.org
 help / color / mirror / Atom feed
From: Philipp Zabel <p.zabel@pengutronix.de>
To: Hans Verkuil <hverkuil@xs4all.nl>
Cc: Mauro Carvalho Chehab <mchehab@osg.samsung.com>,
	Sylwester Nawrocki <s.nawrocki@samsung.com>,
	Pawel Osciak <pawel@osciak.com>,
	Marek Szyprowski <m.szyprowski@samsung.com>,
	Kyungmin Park <kyungmin.park@samsung.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Hans Verkuil <hans.verkuil@cisco.com>,
	Kamil Debski <kamil@wypas.org>,
	linux-media@vger.kernel.org, kernel@pengutronix.de
Subject: Re: [PATCH v3 3/3] [media] videobuf2: add trace events
Date: Tue, 28 Jul 2015 09:53:20 +0200	[thread overview]
Message-ID: <1438070000.3193.2.camel@pengutronix.de> (raw)
In-Reply-To: <1437995577.3239.20.camel@pengutronix.de>

Am Montag, den 27.07.2015, 13:12 +0200 schrieb Philipp Zabel:
> Hi Hans,
> 
> Am Sonntag, den 26.07.2015, 13:18 +0200 schrieb Hans Verkuil:
> > Hi Philipp,
> [...]
> > > diff --git a/drivers/media/v4l2-core/videobuf2-core.c b/drivers/media/v4l2-core/videobuf2-core.c
> > > index 93b3154..b866a6b 100644
> > > --- a/drivers/media/v4l2-core/videobuf2-core.c
> > > +++ b/drivers/media/v4l2-core/videobuf2-core.c
> > > @@ -30,6 +30,8 @@
> > >  #include <media/v4l2-common.h>
> > >  #include <media/videobuf2-core.h>
> > > 
> > 
> > Shouldn't there be a #define CREATE_TRACE_POINTS added before the include? That's
> > what is done in v4l2-ioctl.c as well.
> 
> Documentation/trace/tracepoints.txt says
>    "if you use the header in multiple source files,
>     #define CREATE_TRACE_POINTS should appear only in one source file."
> 
> > I updated my kernel on my laptop to the latest media master and without this line it
> > gives me link errors:
> > 
> > ERROR: "__tracepoint_vb2_qbuf" [drivers/media/v4l2-core/videobuf2-core.ko] undefined!
> > ERROR: "__tracepoint_vb2_buf_done" [drivers/media/v4l2-core/videobuf2-core.ko] undefined!
> > ERROR: "__tracepoint_vb2_buf_queue" [drivers/media/v4l2-core/videobuf2-core.ko] undefined!
> > ERROR: "__tracepoint_vb2_dqbuf" [drivers/media/v4l2-core/videobuf2-core.ko] undefined!
> > scripts/Makefile.modpost:90: recipe for target '__modpost' failed
> 
> Since drivers/media/v4l2-core/v4l2-ioctl.c is built with the trace
> points whenever CONFIG_VIDEO_V4L2 is enabled, the symbols will currently
> end up in drivers/media/v4l2-core/v4l2-ioctl.o, but they are not
> exported.
> 
> > I'm not sure why I didn't see this anywhere else, but can you take a look at this?
> 
> I didn't notice because I hadn't built kernels with VIDEO_V4L2 and
> VIDEOBUF2_CORE as modules on x86. I'm not sure why I don't get these
> errors on ARM, but I suppose this issue is a reason to split the vb2
> tracepoints out of include/trace/event/v4l2.h into their own header
> anyway. The vb2 tracepoint implementation should reside in the
> videobuf2-core module.

I tried this yesterday and failed to figure out a satisfactory way to do
it since the vb2 trace point macros reuse the v4l2 enum definitions and
__print_symbolic/flags macros. The alternative would be to just export
the vb2 trace points from videodev.

regards
Philipp


  reply	other threads:[~2015-07-28  7:53 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-10 13:49 [PATCH v3 1/3] [media] v4l2-dev: use event class to deduplicate v4l2 trace events Philipp Zabel
2015-07-10 13:49 ` [PATCH v3 2/3] [media] v4l2-mem2mem: set the queue owner field just as vb2_ioctl_reqbufs does Philipp Zabel
2015-07-10 13:49 ` [PATCH v3 3/3] [media] videobuf2: add trace events Philipp Zabel
2015-07-26 11:18   ` Hans Verkuil
2015-07-27 11:12     ` Philipp Zabel
2015-07-28  7:53       ` Philipp Zabel [this message]
2015-08-03 15:50         ` Steven Rostedt

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=1438070000.3193.2.camel@pengutronix.de \
    --to=p.zabel@pengutronix.de \
    --cc=hans.verkuil@cisco.com \
    --cc=hverkuil@xs4all.nl \
    --cc=kamil@wypas.org \
    --cc=kernel@pengutronix.de \
    --cc=kyungmin.park@samsung.com \
    --cc=linux-media@vger.kernel.org \
    --cc=m.szyprowski@samsung.com \
    --cc=mchehab@osg.samsung.com \
    --cc=pawel@osciak.com \
    --cc=rostedt@goodmis.org \
    --cc=s.nawrocki@samsung.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.