[PATCH] ASoC: core: don't try to resume uninstantiated a soc-audio devices

[PATCH] ASoC: core: don't try to resume uninstantiated a soc-audio devices

[Daniel Mack]

The code tries to schedule a work item from the resume context, which
will crash the kernel if snd_soc_instantiate_card() has not been called
for the card before.

The bug I triggered is related to 82e14e8b ("ASoC: core: Don't schedule
deferred_resume_work twice"), hence a backport to 3.1 is necessary.

Signed-off-by: Daniel Mack <zonque@gmail.com>
Cc: Liam Girdwood <lrg@ti.com>
Cc: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: Stephen Warren <swarren@nvidia.com>
Cc: stable@kernel.org [v3.1]
---
 sound/soc/soc-core.c |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c
index a5d3685..721f16e 100644
--- a/sound/soc/soc-core.c
+++ b/sound/soc/soc-core.c
@@ -709,6 +709,9 @@ int snd_soc_resume(struct device *dev)
 	struct snd_soc_card *card = dev_get_drvdata(dev);
 	int i, ac97_control = 0;
 
+	if (!card->instantiated)
+		return 0;
+
 	/* AC97 devices might have other drivers hanging off them so
 	 * need to resume immediately.  Other drivers don't have that
 	 * problem and may take a substantial amount of time to resume
-- 
1.7.7.3

Re: [PATCH] ASoC: core: don't try to resume uninstantiated a soc-audio devices

[Mark Brown]

On Wed, Nov 23, 2011 at 08:24:10PM +0100, Daniel Mack wrote:
> The code tries to schedule a work item from the resume context, which
> will crash the kernel if snd_soc_instantiate_card() has not been called
> for the card before.

Eric already sent a slightly different fix for this which is applied.

Re: [PATCH] ASoC: core: don't try to resume uninstantiated a soc-audio devices

[Daniel Mack]

On 11/23/2011 10:26 PM, Mark Brown wrote:
> On Wed, Nov 23, 2011 at 08:24:10PM +0100, Daniel Mack wrote:
>> The code tries to schedule a work item from the resume context, which
>> will crash the kernel if snd_soc_instantiate_card() has not been called
>> for the card before.
> 
> Eric already sent a slightly different fix for this which is applied.

Could you give me a pointer to this patch? I had this problem today with
Linus' latest git tip.

Re: [PATCH] ASoC: core: don't try to resume uninstantiated a soc-audio devices

[Mark Brown]

On Wed, Nov 23, 2011 at 10:37:43PM +0100, Daniel Mack wrote:
> On 11/23/2011 10:26 PM, Mark Brown wrote:
> > On Wed, Nov 23, 2011 at 08:24:10PM +0100, Daniel Mack wrote:
> >> The code tries to schedule a work item from the resume context, which
> >> will crash the kernel if snd_soc_instantiate_card() has not been called
> >> for the card before.

> > Eric already sent a slightly different fix for this which is applied.

> Could you give me a pointer to this patch? I had this problem today with
> Linus' latest git tip.

It's the tip of my for-3.2.  Slightly different fix but the effect
should be the same.

Re: [PATCH] ASoC: core: don't try to resume uninstantiated a soc-audio devices

[Daniel Mack]

On 11/23/2011 10:51 PM, Mark Brown wrote:
> On Wed, Nov 23, 2011 at 10:37:43PM +0100, Daniel Mack wrote:
>> On 11/23/2011 10:26 PM, Mark Brown wrote:
>>> On Wed, Nov 23, 2011 at 08:24:10PM +0100, Daniel Mack wrote:
>>>> The code tries to schedule a work item from the resume context, which
>>>> will crash the kernel if snd_soc_instantiate_card() has not been called
>>>> for the card before.
> 
>>> Eric already sent a slightly different fix for this which is applied.
> 
>> Could you give me a pointer to this patch? I had this problem today with
>> Linus' latest git tip.
> 
> It's the tip of my for-3.2.  Slightly different fix but the effect
> should be the same.

I see, yes. But note that 3.1 also has this problem, so we need the
patch to get backported.