linux-kernel - Re: BUG in 2.6.22-rc2-mm1: Parts of Alsa sound architecture broken

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] [thread-next>] [day] [month] [year] [list]

Message-Id: <1180190293.21547.46.camel@localhost>
Date:	Sat, 26 May 2007 11:38:13 -0300
From:	Mauro Carvalho Chehab <mchehab@...radead.org>
To:	Jan Engelhardt <jengelh@...ux01.gwdg.de>
Cc:	Takashi Iwai <tiwai@...e.de>,
	Andrew Morton <akpm@...ux-foundation.org>,
	linux-kernel@...r.kernel.org, Jaroslav Kysela <perex@...e.cz>
Subject: Re: BUG in 2.6.22-rc2-mm1: Parts of Alsa sound architecture broken

Em Sáb, 2007-05-26 às 12:51 +0200, Jan Engelhardt escreveu:
> On May 25 2007 23:33, Takashi Iwai wrote:
> >
> >Yeah, I'll check it again if reposted.  Jan, could you split ALSA
> >portins at the next time?
> 
> How much split? This time, I made four out of it.
> 
> >This will make testing and merging much
> >easier for me...
> >
> >But, above all, I'm not convinced much by that patch, especially
> >because it introduces new kconfigs just for menuconfig.
> >For example, CONFIG_SND_PCI_DRIVERS doesn't appear in any Makefiles.
> 
> Neither is/was CONFIG_DVB IIRC. The world does not stop turning just
> because there are now extra options. If you want the functionality
> of menuconfigs without introducing some variable to store their state,
> well, I think that's going to be a bigger kconf patch. (Just think of
> compatibility of the .config format.)

I suspect that kconf is not properly handing the newer way for
menuconfig. Maybe this is the same stuff that happened on ALSA.

We have some stuff like this:

menuconfig VIDEO_CAPTURE_DRIVERS
        bool "Video capture adapters"
        depends on VIDEO_DEV

if VIDEO_CAPTURE_DRIVERS

config VIDEO_VIVI
        tristate "Virtual Video Driver"
        depends on VIDEO_V4L2 && !SPARC32 && !SPARC64 && PCI
        select VIDEO_BUF

endif #VIDEO_CAPTURE_DRIVERS

config VIDEO_BUF
        depends on PCI
        tristate

Before adding menuconfig, VIDEO_VIVI were dependent on VIDEO_DEV. After
the patch, what happens is that:
if 
	VIDEO_DEV='m' and VIDEO_CAPTURE_DRIVERS='y' and VIDEO_VIVI='m'
then
	VIDEO_BUF='y'

But, as video-buf is dependent on video-core (compiled as a module,
since VIDEO_DEV='m'), it is generating compilation errors.

To fix it, I needed to add an explicit dependency on VIDEO_VIVI (and
also on VIDEO_SAA7146_VV):

http://linuxtv.org/hg/v4l-dvb?cmd=changeset;node=5cd49ffd9004;style=gitweb

I didn't looked inside kconf, but it seems that it is not checking that,
as VIDEO_CAPTURE_DRIVERS depends on VIDEO_DEV, all drivers inside the
"if" should also depend on VIDEO_DEV.
 
Cheers,
Mauro

-
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/