public inbox for linux-media@vger.kernel.org
 help / color / mirror / Atom feed
From: Mauro Carvalho Chehab <mchehab@redhat.com>
To: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: Helmut Auer <helmut@helmutauer.de>,
	linux-media@vger.kernel.org, Oliver Endriss <o.endriss@gmx.de>
Subject: Re: Bug: media_build always compiles with '-DDEBUG'
Date: Mon, 20 Jun 2011 11:05:39 -0300	[thread overview]
Message-ID: <4DFF53B3.3040608@redhat.com> (raw)
In-Reply-To: <201106201538.15214.laurent.pinchart@ideasonboard.com>

Em 20-06-2011 10:38, Laurent Pinchart escreveu:
> Hi Mauro,
> 
> On Monday 20 June 2011 14:50:28 Mauro Carvalho Chehab wrote:
>> Em 20-06-2011 09:35, Laurent Pinchart escreveu:
>>> On Sunday 19 June 2011 13:47:16 Mauro Carvalho Chehab wrote:
>>>> Em 19-06-2011 02:00, Helmut Auer escreveu:
>>>>> Am 18.06.2011 23:38, schrieb Oliver Endriss:
>>>>>> On Saturday 18 June 2011 23:11:21 Helmut Auer wrote:
>>>>>>> Hi
>>>>>>>
>>>>>>>> Replacing
>>>>>>>>
>>>>>>>>       ifdef CONFIG_VIDEO_OMAP3_DEBUG
>>>>>>>>
>>>>>>>> by
>>>>>>>>
>>>>>>>>       ifeq ($(CONFIG_VIDEO_OMAP3_DEBUG),y)
>>>>>>>>
>>>>>>>> would do the trick.
>>>>>>>
>>>>>>> I guess that would not ive the intended result.
>>>>>>> Setting CONFIG_VIDEO_OMAP3_DEBUG to yes should not lead to debug
>>>>>>> messages in all media modules,
>>>>>>
>>>>>> True, but it will happen only if you manually enable
>>>>>> CONFIG_VIDEO_OMAP3_DEBUG in Kconfig.
>>>>>>
>>>>>> You cannot avoid this without major changes of the
>>>>>> media_build system - imho not worth the effort.
>>>>>
>>>>> Then imho it would be better to drop the  CONFIG_VIDEO_OMAP3_DEBUG
>>>>> variable completely, you can set CONFIG_DEBUG which would give the same
>>>>> results.
>>>>
>>>> Good catch!
>>>>
>>>> Yes, I agree that the better is to just drop CONFIG_VIDEO_OMAP3_DEBUG
>>>> variable completely. If someone wants to build with -DDEBUG, he can just
>>>> use CONFIG_DEBUG.
>>>>
>>>> Laurent,
>>>>
>>>> Any comments?
>>>
>>> CONFIG_VIDEO_OMAP3_DEBUG is used to build the OMAP3 ISP driver in debug
>>> mode, without having to compile the whole kernel with debugging enabled.
>>> I'd like to keep that feature if possible.
>>
>> If you want that, build it using media_build. I don't care of having such
>> hacks there, but having it upstream is not the right thing to do.
> 
> It's not a hack. Lots of drivers have debugging Kconfig options.
> 
> $ find linux-2.6 -type f -name Kconfig* -exec grep '^config.*DEBUG' {} \; | wc
>     243     486    5826

Your query is wrong. The proper query is:
$ find . -type f -name Makefile -exec grep -rH 'EXTRA_CFLAGS.*\-DDEBUG$' {} \; 
./arch/x86/pci/Makefile:EXTRA_CFLAGS += -DDEBUG
./drivers/pps/generators/Makefile:EXTRA_CFLAGS += -DDEBUG
./drivers/media/video/omap3isp/Makefile:EXTRA_CFLAGS += -DDEBUG

There's nothing wrong with a debug Kconfig option
for omap3. What's wrong is:

1) It is badly implemented: it is just enabling -DDEBUG for the entire
subsystem, as the test is wrong;

2) Even if you fix, you'll be enabling debug to the entire subsystem, if you
keep adding things to EXTRA_CFLAGS. 

It should be noticed that several places use a different syntax for enabling
per-driver/per-subsystem -D flag:

find . -type f -name Makefile -exec grep -rH '\-DDEBUG$' {} \; 
...
./drivers/mmc/Makefile:subdir-ccflags-$(CONFIG_MMC_DEBUG) := -DDEBUG
...
./drivers/infiniband/hw/cxgb3/Makefile:ccflags-$(CONFIG_INFINIBAND_CXGB3_DEBUG) += -DDEBUG

I _suspect_ that using ccflags-$(foo_DEBUG) may work.

Thanks,
Mauro

  reply	other threads:[~2011-06-20 14:05 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-18 20:46 Bug: media_build always compiles with '-DDEBUG' Oliver Endriss
2011-06-18 21:11 ` Helmut Auer
2011-06-18 21:38   ` Oliver Endriss
2011-06-19  5:00     ` Helmut Auer
2011-06-19 11:47       ` Mauro Carvalho Chehab
2011-06-20 12:35         ` Laurent Pinchart
2011-06-20 12:50           ` Mauro Carvalho Chehab
2011-06-20 13:38             ` Laurent Pinchart
2011-06-20 14:05               ` Mauro Carvalho Chehab [this message]
2011-06-20 14:25                 ` Laurent Pinchart

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=4DFF53B3.3040608@redhat.com \
    --to=mchehab@redhat.com \
    --cc=helmut@helmutauer.de \
    --cc=laurent.pinchart@ideasonboard.com \
    --cc=linux-media@vger.kernel.org \
    --cc=o.endriss@gmx.de \
    /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