2010-01-18 13:24:17

by Hagen Paul Pfeifer

[permalink] [raw]
Subject: [PATCH 1/1] thinkpad-acpi: fix link error to include sound arch

Thinkpad ACPI depends on sound for resume/suspend/shutdown
events. This fix corrects the Kconfig dependency between
in-kernel build and module build. The preprocessor
directive #ifdef CONFIG_THINKPAD_ACPI_ALSA_SUPPORT
isn't sufficient to handle this build dependency.

Signed-off-by: Hagen Paul Pfeifer <[email protected]>
---
drivers/platform/x86/Kconfig | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig
index db32c25..a4a31b5 100644
--- a/drivers/platform/x86/Kconfig
+++ b/drivers/platform/x86/Kconfig
@@ -214,6 +214,7 @@ config THINKPAD_ACPI
tristate "ThinkPad ACPI Laptop Extras"
depends on ACPI
depends on INPUT
+ depends on SND
depends on RFKILL || RFKILL = n
select BACKLIGHT_LCD_SUPPORT
select BACKLIGHT_CLASS_DEVICE
--
1.6.5


Subject: Re: [PATCH 1/1] thinkpad-acpi: fix link error to include sound arch

On Mon, 18 Jan 2010, Hagen Paul Pfeifer wrote:
> Thinkpad ACPI depends on sound for resume/suspend/shutdown
> events. This fix corrects the Kconfig dependency between

Does it? I'd rather fix the real problem, it shouldn't depend on sound for
resume/suspend/shutdown. If it does, the fix is to break the dependency.

> in-kernel build and module build. The preprocessor
> directive #ifdef CONFIG_THINKPAD_ACPI_ALSA_SUPPORT
> isn't sufficient to handle this build dependency.

NAK.

Can you describe the problem in more detail? That will save me precious
time, as I am rather overworked right now...

--
"One disk to rule them all, One disk to find them. One disk to bring
them all and in the darkness grind them. In the Land of Redmond
where the shadows lie." -- The Silicon Valley Tarot
Henrique Holschuh

2010-01-19 15:14:18

by Hagen Paul Pfeifer

[permalink] [raw]
Subject: Re: [PATCH 1/1] thinkpad-acpi: fix link error to include sound arch


On Tue, 19 Jan 2010 12:51:08 -0200, <[email protected]> wrote:

> Can you describe the problem in more detail? That will save me precious
> time, as I am rather overworked right now...

If you build the sound as a module and thinkpad as a build-in component
the linker will complain about missing symbols. So the right fix may
be a patch that 1) enforces the sound component to be build static or
to enforces thinkpad apic to be build as a module. Another way 2) is
to separate the thinkpad sound component in a autonomous separate
module.

Any other ideas Henrique?

Hagen

Subject: Re: [PATCH 1/1] thinkpad-acpi: fix link error to include sound arch

On Tue, 19 Jan 2010, Hagen Paul Pfeifer wrote:
> On Tue, 19 Jan 2010 12:51:08 -0200, <[email protected]> wrote:
> > Can you describe the problem in more detail? That will save me precious
> > time, as I am rather overworked right now...
>
> If you build the sound as a module and thinkpad as a build-in component
> the linker will complain about missing symbols. So the right fix may

This is not supposed to be possible anymore, and in fact, I can't force
that breakage to happen in 2.6.33-rc6.

> be a patch that 1) enforces the sound component to be build static or
> to enforces thinkpad apic to be build as a module. Another way 2) is

Currently, the Kconfig disables ALSA support for thinkpad-acpi static
and ALSA as module, but allows static-static or module-module.

> Any other ideas Henrique?

No, you're correct, but the fix was already in its way to mainline when
you sent your report. I took some time to try to break it today, just
in case there was something I missed.

What kernel was giving you trouble? I still need to fix the backports in
the thinkpad-acpi git tree, maybe it was one of them?

--
"One disk to rule them all, One disk to find them. One disk to bring
them all and in the darkness grind them. In the Land of Redmond
where the shadows lie." -- The Silicon Valley Tarot
Henrique Holschuh

2010-02-20 09:47:10

by Hagen Paul Pfeifer

[permalink] [raw]
Subject: Re: [PATCH 1/1] thinkpad-acpi: fix link error to include sound arch

* Henrique de Moraes Holschuh | 2010-02-07 18:05:06 [-0200]:

>No, you're correct, but the fix was already in its way to mainline when
>you sent your report. I took some time to try to break it today, just
>in case there was something I missed.

OK, I was not aware of any efforts. I had vacation so I am little bit
outdated. I saw your patches and was impressed - nice work!

>What kernel was giving you trouble? I still need to fix the backports in
>the thinkpad-acpi git tree, maybe it was one of them?

No sure, somewhere in the area of 9db2f1bec36805e57a003f7bb90e003815d96de8
My posted patch applied to the HEAD of linus at this time.


Best regards, Hagen

--
Hagen Paul Pfeifer <[email protected]> || http://jauu.net/
Telephone: +49 174 5455209 || Key Id: 0x98350C22
Key Fingerprint: 490F 557B 6C48 6D7E 5706 2EA2 4A22 8D45 9835 0C22