sound/soc/sdca/sdca_interrupts.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
Hi all,
After merging the tip tree, today's linux-next build (x86_64 allmodconfig)
failed like this:
In file included from include/linux/kernfs.h:11,
from include/linux/sysfs.h:16,
from include/linux/kobject.h:20,
from include/linux/energy_model.h:7,
from include/linux/device.h:16,
from sound/soc/sdca/sdca_interrupts.c:13:
sound/soc/sdca/sdca_interrupts.c: In function 'sdca_irq_allocate':
include/linux/mutex.h:153:9: error: ignoring return value of '__devm_mutex_init' declared with attribute 'warn_unused_result' [-Werror=unused-result]
153 | __devm_mutex_init(dev, __mutex_init_ret(mutex))
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sound/soc/sdca/sdca_interrupts.c:422:9: note: in expansion of macro 'devm_mutex_init'
422 | devm_mutex_init(dev, &info->irq_lock);
| ^~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
Caused by commit
daec29dcc873 ("locking/mutex: Mark devm_mutex_init() as __must_check")
interacting with commit
b126394d9ec6 ("ASoC: SDCA: Generic interrupt support")
from the sound-asoc tree.
I have applied the following merge fix patch (just a hack, something
better needs to be done).
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 22 Jul 2025 12:26:10 +1000
Subject: [PATCH] fix up for "ASoC: SDCA: Generic interrupt support"
interacting with "locking/mutex: Mark devm_mutex_init() as
__must_check".
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
sound/soc/sdca/sdca_interrupts.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/sound/soc/sdca/sdca_interrupts.c b/sound/soc/sdca/sdca_interrupts.c
index b76512732af8..b393471e9760 100644
--- a/sound/soc/sdca/sdca_interrupts.c
+++ b/sound/soc/sdca/sdca_interrupts.c
@@ -419,7 +419,11 @@ struct sdca_interrupt_info *sdca_irq_allocate(struct device *dev,
info->irq_chip = sdca_irq_chip;
- devm_mutex_init(dev, &info->irq_lock);
+ ret = devm_mutex_init(dev, &info->irq_lock);
+ if (ret) {
+ /* do some error recovery */
+ return ERR_PTR(ret);
+ }
ret = devm_regmap_add_irq_chip(dev, regmap, irq, IRQF_ONESHOT, 0,
&info->irq_chip, &info->irq_data);
--
2.50.1
--
Cheers,
Stephen Rothwell
On Tue, Jul 22, 2025 at 01:01:40PM +1000, Stephen Rothwell wrote: > Hi all, > Hi Stephen, > After merging the tip tree, today's linux-next build (x86_64 allmodconfig) > failed like this: > > In file included from include/linux/kernfs.h:11, > from include/linux/sysfs.h:16, > from include/linux/kobject.h:20, > from include/linux/energy_model.h:7, > from include/linux/device.h:16, > from sound/soc/sdca/sdca_interrupts.c:13: > sound/soc/sdca/sdca_interrupts.c: In function 'sdca_irq_allocate': > include/linux/mutex.h:153:9: error: ignoring return value of '__devm_mutex_init' declared with attribute 'warn_unused_result' [-Werror=unused-result] > 153 | __devm_mutex_init(dev, __mutex_init_ret(mutex)) > | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > sound/soc/sdca/sdca_interrupts.c:422:9: note: in expansion of macro 'devm_mutex_init' > 422 | devm_mutex_init(dev, &info->irq_lock); > | ^~~~~~~~~~~~~~~ > cc1: all warnings being treated as errors > > Caused by commit > > daec29dcc873 ("locking/mutex: Mark devm_mutex_init() as __must_check") > > interacting with commit > > b126394d9ec6 ("ASoC: SDCA: Generic interrupt support") > > from the sound-asoc tree. > > I have applied the following merge fix patch (just a hack, something > better needs to be done). > Thanks for spotting this, I think it'll be easier if sound-asoc tree could apply this change (with proper error handling). Maciej, Charles and Mark, does it sound good? Apologies that I was a bit late for locking changes in this cycle. Thanks! Regards, Boqun > From: Stephen Rothwell <sfr@canb.auug.org.au> > Date: Tue, 22 Jul 2025 12:26:10 +1000 > Subject: [PATCH] fix up for "ASoC: SDCA: Generic interrupt support" > > interacting with "locking/mutex: Mark devm_mutex_init() as > __must_check". > > Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au> > --- > sound/soc/sdca/sdca_interrupts.c | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) > > diff --git a/sound/soc/sdca/sdca_interrupts.c b/sound/soc/sdca/sdca_interrupts.c > index b76512732af8..b393471e9760 100644 > --- a/sound/soc/sdca/sdca_interrupts.c > +++ b/sound/soc/sdca/sdca_interrupts.c > @@ -419,7 +419,11 @@ struct sdca_interrupt_info *sdca_irq_allocate(struct device *dev, > > info->irq_chip = sdca_irq_chip; > > - devm_mutex_init(dev, &info->irq_lock); > + ret = devm_mutex_init(dev, &info->irq_lock); > + if (ret) { > + /* do some error recovery */ > + return ERR_PTR(ret); > + } > > ret = devm_regmap_add_irq_chip(dev, regmap, irq, IRQF_ONESHOT, 0, > &info->irq_chip, &info->irq_data); > -- > 2.50.1 > > -- > Cheers, > Stephen Rothwell
On Mon, Jul 21, 2025 at 08:15:50PM -0700, Boqun Feng wrote: > On Tue, Jul 22, 2025 at 01:01:40PM +1000, Stephen Rothwell wrote: > > Hi all, > > > > Hi Stephen, > > > After merging the tip tree, today's linux-next build (x86_64 allmodconfig) > > failed like this: > > > > In file included from include/linux/kernfs.h:11, > > from include/linux/sysfs.h:16, > > from include/linux/kobject.h:20, > > from include/linux/energy_model.h:7, > > from include/linux/device.h:16, > > from sound/soc/sdca/sdca_interrupts.c:13: > > sound/soc/sdca/sdca_interrupts.c: In function 'sdca_irq_allocate': > > include/linux/mutex.h:153:9: error: ignoring return value of '__devm_mutex_init' declared with attribute 'warn_unused_result' [-Werror=unused-result] > > 153 | __devm_mutex_init(dev, __mutex_init_ret(mutex)) > > | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > > sound/soc/sdca/sdca_interrupts.c:422:9: note: in expansion of macro 'devm_mutex_init' > > 422 | devm_mutex_init(dev, &info->irq_lock); > > | ^~~~~~~~~~~~~~~ > > cc1: all warnings being treated as errors > > > > Caused by commit > > > > daec29dcc873 ("locking/mutex: Mark devm_mutex_init() as __must_check") > > > > interacting with commit > > > > b126394d9ec6 ("ASoC: SDCA: Generic interrupt support") > > > > from the sound-asoc tree. > > > > I have applied the following merge fix patch (just a hack, something > > better needs to be done). > > > > Thanks for spotting this, I think it'll be easier if sound-asoc tree > could apply this change (with proper error handling). Maciej, Charles > and Mark, does it sound good? Apologies that I was a bit late for > locking changes in this cycle. Thanks! > I am happy enough to do a patch for this, will look at it today. Thanks, Charles
© 2016 - 2025 Red Hat, Inc.