[PATCH] alsa: use subsys_initcall for sound core instead of module_init

[PATCH] alsa: use subsys_initcall for sound core instead of module_init

[Thadeu Lima de Souza Cascardo]

This is needed for built-in drivers which are built before the sound directory,
like thinkpad_acpi.

Otherwise, registering a card fails.

Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
---
 sound/core/sound.c |    4 ++--
 sound/sound_core.c |    2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/sound/core/sound.c b/sound/core/sound.c
index 7872a02..563d196 100644
--- a/sound/core/sound.c
+++ b/sound/core/sound.c
@@ -468,5 +468,5 @@ static void __exit alsa_sound_exit(void)
 	unregister_chrdev(major, "alsa");
 }
 
-module_init(alsa_sound_init)
-module_exit(alsa_sound_exit)
+subsys_initcall(alsa_sound_init);
+module_exit(alsa_sound_exit);
diff --git a/sound/sound_core.c b/sound/sound_core.c
index dbca7c9..7c2d677 100644
--- a/sound/sound_core.c
+++ b/sound/sound_core.c
@@ -61,7 +61,7 @@ static void __exit cleanup_soundcore(void)
 	class_destroy(sound_class);
 }
 
-module_init(init_soundcore);
+subsys_initcall(init_soundcore);
 module_exit(cleanup_soundcore);
 
 
-- 
1.6.6

Re: [PATCH] alsa: use subsys_initcall for sound core instead of module_init

[Takashi Iwai]

At Tue, 12 Jan 2010 20:20:39 -0200,
Thadeu Lima de Souza Cascardo wrote:
> 
> This is needed for built-in drivers which are built before the sound directory,
> like thinkpad_acpi.
> 
> Otherwise, registering a card fails.

Oh well, this kind of problem hits again.

I don't remember why subsys_init() wasn't taken before...  Maybe some
dependency between sound and V4L or so.  But, it's worth to check again.

I'll apply this after some testing.


thanks,

Takashi

> 
> Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
> ---
>  sound/core/sound.c |    4 ++--
>  sound/sound_core.c |    2 +-
>  2 files changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/sound/core/sound.c b/sound/core/sound.c
> index 7872a02..563d196 100644
> --- a/sound/core/sound.c
> +++ b/sound/core/sound.c
> @@ -468,5 +468,5 @@ static void __exit alsa_sound_exit(void)
>  	unregister_chrdev(major, "alsa");
>  }
>  
> -module_init(alsa_sound_init)
> -module_exit(alsa_sound_exit)
> +subsys_initcall(alsa_sound_init);
> +module_exit(alsa_sound_exit);
> diff --git a/sound/sound_core.c b/sound/sound_core.c
> index dbca7c9..7c2d677 100644
> --- a/sound/sound_core.c
> +++ b/sound/sound_core.c
> @@ -61,7 +61,7 @@ static void __exit cleanup_soundcore(void)
>  	class_destroy(sound_class);
>  }
>  
> -module_init(init_soundcore);
> +subsys_initcall(init_soundcore);
>  module_exit(cleanup_soundcore);
>  
>  
> -- 
> 1.6.6
> 

Re: [PATCH] alsa: use subsys_initcall for sound core instead of module_init

[Thadeu Lima de Souza Cascardo]

[-- Attachment #1: Type: text/plain, Size: 1948 bytes --]

On Wed, Jan 13, 2010 at 07:48:07AM +0100, Takashi Iwai wrote:
> At Tue, 12 Jan 2010 20:20:39 -0200,
> Thadeu Lima de Souza Cascardo wrote:
> > 
> > This is needed for built-in drivers which are built before the sound directory,
> > like thinkpad_acpi.
> > 
> > Otherwise, registering a card fails.
> 
> Oh well, this kind of problem hits again.
> 
> I don't remember why subsys_init() wasn't taken before...  Maybe some
> dependency between sound and V4L or so.  But, it's worth to check again.
> 

Any references to this dependency problem, so I can try to help and investigate
about it and do some tests?

> I'll apply this after some testing.
> 

Built-in thinkpad-acpi started working for me after these changes. I can't tell
you about any subsystem that stopped working, like some V4L driver.

> 
> thanks,
> 
> Takashi
> 

Regards,
Cascardo.

> > 
> > Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
> > ---
> >  sound/core/sound.c |    4 ++--
> >  sound/sound_core.c |    2 +-
> >  2 files changed, 3 insertions(+), 3 deletions(-)
> > 
> > diff --git a/sound/core/sound.c b/sound/core/sound.c
> > index 7872a02..563d196 100644
> > --- a/sound/core/sound.c
> > +++ b/sound/core/sound.c
> > @@ -468,5 +468,5 @@ static void __exit alsa_sound_exit(void)
> >  	unregister_chrdev(major, "alsa");
> >  }
> >  
> > -module_init(alsa_sound_init)
> > -module_exit(alsa_sound_exit)
> > +subsys_initcall(alsa_sound_init);
> > +module_exit(alsa_sound_exit);
> > diff --git a/sound/sound_core.c b/sound/sound_core.c
> > index dbca7c9..7c2d677 100644
> > --- a/sound/sound_core.c
> > +++ b/sound/sound_core.c
> > @@ -61,7 +61,7 @@ static void __exit cleanup_soundcore(void)
> >  	class_destroy(sound_class);
> >  }
> >  
> > -module_init(init_soundcore);
> > +subsys_initcall(init_soundcore);
> >  module_exit(cleanup_soundcore);
> >  
> >  
> > -- 
> > 1.6.6
> > 

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [PATCH] alsa: use subsys_initcall for sound core instead of module_init

[Takashi Iwai]

At Thu, 14 Jan 2010 13:53:43 -0200,
Thadeu Lima de Souza Cascardo wrote:
> 
> On Wed, Jan 13, 2010 at 07:48:07AM +0100, Takashi Iwai wrote:
> > At Tue, 12 Jan 2010 20:20:39 -0200,
> > Thadeu Lima de Souza Cascardo wrote:
> > > 
> > > This is needed for built-in drivers which are built before the sound directory,
> > > like thinkpad_acpi.
> > > 
> > > Otherwise, registering a card fails.
> > 
> > Oh well, this kind of problem hits again.
> > 
> > I don't remember why subsys_init() wasn't taken before...  Maybe some
> > dependency between sound and V4L or so.  But, it's worth to check again.
> > 
> 
> Any references to this dependency problem, so I can try to help and investigate
> about it and do some tests?

I also don't remember.  That was a long long ago... :)

> > I'll apply this after some testing.
> > 
> 
> Built-in thinkpad-acpi started working for me after these changes. I can't tell
> you about any subsystem that stopped working, like some V4L driver.

Did you test whether the sound subsystem still works as modules?


Takashi

Re: [PATCH] alsa: use subsys_initcall for sound core instead of module_init

[Thadeu Lima de Souza Cascardo]

[-- Attachment #1: Type: text/plain, Size: 1366 bytes --]

On Thu, Jan 14, 2010 at 05:18:26PM +0100, Takashi Iwai wrote:
> At Thu, 14 Jan 2010 13:53:43 -0200,
> Thadeu Lima de Souza Cascardo wrote:
> > 
> > On Wed, Jan 13, 2010 at 07:48:07AM +0100, Takashi Iwai wrote:
> > > At Tue, 12 Jan 2010 20:20:39 -0200,
> > > Thadeu Lima de Souza Cascardo wrote:
> > > > 
> > > > This is needed for built-in drivers which are built before the sound directory,
> > > > like thinkpad_acpi.
> > > > 
> > > > Otherwise, registering a card fails.
> > > 
> > > Oh well, this kind of problem hits again.
> > > 
> > > I don't remember why subsys_init() wasn't taken before...  Maybe some
> > > dependency between sound and V4L or so.  But, it's worth to check again.
> > > 
> > 
> > Any references to this dependency problem, so I can try to help and investigate
> > about it and do some tests?
> 
> I also don't remember.  That was a long long ago... :)
> 
> > > I'll apply this after some testing.
> > > 
> > 
> > Built-in thinkpad-acpi started working for me after these changes. I can't tell
> > you about any subsystem that stopped working, like some V4L driver.
> 
> Did you test whether the sound subsystem still works as modules?
> 

It does. I've just reapplied those patches on top of rc4, and I'm currently
running it with CONFIG_SOUND=m, and my sound card works nicely, thanks.

> 
> Takashi

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [PATCH] alsa: use subsys_initcall for sound core instead of module_init

[Takashi Iwai]

At Thu, 14 Jan 2010 14:22:31 -0200,
Thadeu Lima de Souza Cascardo wrote:
> 
> On Thu, Jan 14, 2010 at 05:18:26PM +0100, Takashi Iwai wrote:
> > At Thu, 14 Jan 2010 13:53:43 -0200,
> > Thadeu Lima de Souza Cascardo wrote:
> > > 
> > > On Wed, Jan 13, 2010 at 07:48:07AM +0100, Takashi Iwai wrote:
> > > > At Tue, 12 Jan 2010 20:20:39 -0200,
> > > > Thadeu Lima de Souza Cascardo wrote:
> > > > > 
> > > > > This is needed for built-in drivers which are built before the sound directory,
> > > > > like thinkpad_acpi.
> > > > > 
> > > > > Otherwise, registering a card fails.
> > > > 
> > > > Oh well, this kind of problem hits again.
> > > > 
> > > > I don't remember why subsys_init() wasn't taken before...  Maybe some
> > > > dependency between sound and V4L or so.  But, it's worth to check again.
> > > > 
> > > 
> > > Any references to this dependency problem, so I can try to help and investigate
> > > about it and do some tests?
> > 
> > I also don't remember.  That was a long long ago... :)
> > 
> > > > I'll apply this after some testing.
> > > > 
> > > 
> > > Built-in thinkpad-acpi started working for me after these changes. I can't tell
> > > you about any subsystem that stopped working, like some V4L driver.
> > 
> > Did you test whether the sound subsystem still works as modules?
> > 
> 
> It does. I've just reapplied those patches on top of rc4, and I'm currently
> running it with CONFIG_SOUND=m, and my sound card works nicely, thanks.

Good, it's clear it must, but just to be sure :)


thanks,

Takashi

Re: [PATCH] alsa: use subsys_initcall for sound core instead of module_init

[Takashi Iwai]

At Wed, 13 Jan 2010 07:48:07 +0100,
I wrote:
> 
> At Tue, 12 Jan 2010 20:20:39 -0200,
> Thadeu Lima de Souza Cascardo wrote:
> > 
> > This is needed for built-in drivers which are built before the sound directory,
> > like thinkpad_acpi.
> > 
> > Otherwise, registering a card fails.
> 
> Oh well, this kind of problem hits again.
> 
> I don't remember why subsys_init() wasn't taken before...  Maybe some
> dependency between sound and V4L or so.  But, it's worth to check again.
> 
> I'll apply this after some testing.

The patch is merged now to sound git tree.


thanks,

Takashi