revert: ASoC: Fail card instantiation if DAI format setup fails

revert: ASoC: Fail card instantiation if DAI format setup fails

[Ricard Wanderlof]

Sorry for the repost, I relized I stupidly got Greg's email adress wrong 
first time around.

> > On Tue, Aug 27, 2019 at 12:00:14PM +0100, Mark Brown wrote:
> > > On Sun, Aug 25, 2019 at 09:35:15PM -0400, Sasha Levin wrote:
> > > > On Wed, Aug 14, 2019 at 10:22:13AM +0100, Mark Brown wrote:
> > > 
> > > > > > If the DAI format setup fails, there is no valid communication format
> > > > > > between CPU and CODEC, so fail card instantiation, rather than
> > > > continue
> > > > > > with a card that will most likely not function properly.
> > > 
> > > > > This is another one where if nobody noticed a problem already and things
> > > > > just happened to be working this might break things, it's vanishingly
> > > > > unlikely to fix anything that was broken.
> > > 
> > > > Same as the other patch: this patch suggests it fixes a real bug, and if
> > > > this patch is broken let's fix it.
> > > 
> > > If anyone ran into this on the older kernel and fixed or worked
> > > around it locally there's a reasonable chance this will then
> > > break what they're doing.  The patch itself is perfectly fine but

(Sorry about the mangled subject line, I'd accidentally deleted the 
original message from my inbox.)

I'm a bit bewildered here. As the author of the original patch I'm of 
course biased, and I can certainly understand the patch being dropped from 
existing release branches, since as Mark correctly states, it does not fix 
any broken behavior and might even break things that happen to work by 
chance.

But is this being dropped from the master branch as well? To me it makes 
the kernel behave in an inconsistent way, first reporting a failure to 
instantiate a specific sound card in the kernel log, but then seemingly 
bringing it up anyway.

/Ricard
-- 
Ricard Wolf Wanderlof                           ricardw(at)axis.com
Axis Communications AB, Lund, Sweden            www.axis.com
Phone +46 46 272 2016                           Fax +46 46 13 61 30

Re: revert: ASoC: Fail card instantiation if DAI format setup fails

[Sasha Levin]

On Fri, Sep 06, 2019 at 05:00:28PM +0200, Ricard Wanderlof wrote:
>
>Sorry for the repost, I relized I stupidly got Greg's email adress wrong
>first time around.
>
>> > On Tue, Aug 27, 2019 at 12:00:14PM +0100, Mark Brown wrote:
>> > > On Sun, Aug 25, 2019 at 09:35:15PM -0400, Sasha Levin wrote:
>> > > > On Wed, Aug 14, 2019 at 10:22:13AM +0100, Mark Brown wrote:
>> > >
>> > > > > > If the DAI format setup fails, there is no valid communication format
>> > > > > > between CPU and CODEC, so fail card instantiation, rather than
>> > > > continue
>> > > > > > with a card that will most likely not function properly.
>> > >
>> > > > > This is another one where if nobody noticed a problem already and things
>> > > > > just happened to be working this might break things, it's vanishingly
>> > > > > unlikely to fix anything that was broken.
>> > >
>> > > > Same as the other patch: this patch suggests it fixes a real bug, and if
>> > > > this patch is broken let's fix it.
>> > >
>> > > If anyone ran into this on the older kernel and fixed or worked
>> > > around it locally there's a reasonable chance this will then
>> > > break what they're doing.  The patch itself is perfectly fine but
>
>(Sorry about the mangled subject line, I'd accidentally deleted the
>original message from my inbox.)
>
>I'm a bit bewildered here. As the author of the original patch I'm of
>course biased, and I can certainly understand the patch being dropped from
>existing release branches, since as Mark correctly states, it does not fix
>any broken behavior and might even break things that happen to work by
>chance.
>
>But is this being dropped from the master branch as well? To me it makes
>the kernel behave in an inconsistent way, first reporting a failure to
>instantiate a specific sound card in the kernel log, but then seemingly
>bringing it up anyway.

Hi Richard,

This patch is only dropped from the stable branches, it still remains in
the mainline branch.

--
Thanks,
Sasha

[PATCH AUTOSEL 5.2 001/123] ASoC: simple_card_utils.h: care NULL dai at asoc_simple_debug_dai()

[Sasha Levin]

From: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>

[ Upstream commit 52db6685932e326ed607644ab7ebdae8c194adda ]

props->xxx_dai might be NULL when DPCM.
This patch cares it for debug.

Fixes: commit 0580dde59438 ("ASoC: simple-card-utils: add asoc_simple_debug_info()")
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87o922gw4u.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 include/sound/simple_card_utils.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/include/sound/simple_card_utils.h b/include/sound/simple_card_utils.h
index 3429888347e7c..b3609e4c46e0f 100644
--- a/include/sound/simple_card_utils.h
+++ b/include/sound/simple_card_utils.h
@@ -149,6 +149,10 @@ inline void asoc_simple_debug_dai(struct asoc_simple_priv *priv,
 {
 	struct device *dev = simple_priv_to_dev(priv);
 
+	/* dai might be NULL */
+	if (!dai)
+		return;
+
 	if (dai->name)
 		dev_dbg(dev, "%s dai name = %s\n",
 			name, dai->name);
-- 
2.20.1

[PATCH AUTOSEL 5.2 040/123] ASoC: Fail card instantiation if DAI format setup fails

[Sasha Levin]

From: Ricard Wanderlof <ricard.wanderlof@axis.com>

[ Upstream commit 40aa5383e393d72f6aa3943a4e7b1aae25a1e43b ]

If the DAI format setup fails, there is no valid communication format
between CPU and CODEC, so fail card instantiation, rather than continue
with a card that will most likely not function properly.

Signed-off-by: Ricard Wanderlof <ricardw@axis.com>
Link: https://lore.kernel.org/r/alpine.DEB.2.20.1907241132350.6338@lnxricardw1.se.axis.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 sound/soc/soc-core.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c
index 6aeba0d66ec50..dd0f43a1c5e14 100644
--- a/sound/soc/soc-core.c
+++ b/sound/soc/soc-core.c
@@ -1605,8 +1605,11 @@ static int soc_probe_link_dais(struct snd_soc_card *card,
 		}
 	}
 
-	if (dai_link->dai_fmt)
-		snd_soc_runtime_set_dai_fmt(rtd, dai_link->dai_fmt);
+	if (dai_link->dai_fmt) {
+		ret = snd_soc_runtime_set_dai_fmt(rtd, dai_link->dai_fmt);
+		if (ret)
+			return ret;
+	}
 
 	ret = soc_post_component_init(rtd, dai_link->name);
 	if (ret)
-- 
2.20.1

Re: [PATCH AUTOSEL 5.2 040/123] ASoC: Fail card instantiation if DAI format setup fails

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 586 bytes --]

On Tue, Aug 13, 2019 at 10:09:24PM -0400, Sasha Levin wrote:
> From: Ricard Wanderlof <ricard.wanderlof@axis.com>
> 
> [ Upstream commit 40aa5383e393d72f6aa3943a4e7b1aae25a1e43b ]
> 
> If the DAI format setup fails, there is no valid communication format
> between CPU and CODEC, so fail card instantiation, rather than continue
> with a card that will most likely not function properly.

This is another one where if nobody noticed a problem already and things
just happened to be working this might break things, it's vanishingly
unlikely to fix anything that was broken.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH AUTOSEL 5.2 040/123] ASoC: Fail card instantiation if DAI format setup fails

[Sasha Levin]

On Wed, Aug 14, 2019 at 10:22:13AM +0100, Mark Brown wrote:
>On Tue, Aug 13, 2019 at 10:09:24PM -0400, Sasha Levin wrote:
>> From: Ricard Wanderlof <ricard.wanderlof@axis.com>
>>
>> [ Upstream commit 40aa5383e393d72f6aa3943a4e7b1aae25a1e43b ]
>>
>> If the DAI format setup fails, there is no valid communication format
>> between CPU and CODEC, so fail card instantiation, rather than continue
>> with a card that will most likely not function properly.
>
>This is another one where if nobody noticed a problem already and things
>just happened to be working this might break things, it's vanishingly
>unlikely to fix anything that was broken.

Same as the other patch: this patch suggests it fixes a real bug, and if
this patch is broken let's fix it.

--
Thanks,
Sasha

Re: [PATCH AUTOSEL 5.2 040/123] ASoC: Fail card instantiation if DAI format setup fails

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 928 bytes --]

On Sun, Aug 25, 2019 at 09:35:15PM -0400, Sasha Levin wrote:
> On Wed, Aug 14, 2019 at 10:22:13AM +0100, Mark Brown wrote:

> > > If the DAI format setup fails, there is no valid communication format
> > > between CPU and CODEC, so fail card instantiation, rather than continue
> > > with a card that will most likely not function properly.

> > This is another one where if nobody noticed a problem already and things
> > just happened to be working this might break things, it's vanishingly
> > unlikely to fix anything that was broken.

> Same as the other patch: this patch suggests it fixes a real bug, and if
> this patch is broken let's fix it.

If anyone ran into this on the older kernel and fixed or worked
around it locally there's a reasonable chance this will then
break what they're doing.  The patch itself is perfectly fine but
that doesn't mean the rest of the changes it's being backported
into are also fine.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH AUTOSEL 5.2 040/123] ASoC: Fail card instantiation if DAI format setup fails

[Sasha Levin]

On Tue, Aug 27, 2019 at 12:00:14PM +0100, Mark Brown wrote:
>On Sun, Aug 25, 2019 at 09:35:15PM -0400, Sasha Levin wrote:
>> On Wed, Aug 14, 2019 at 10:22:13AM +0100, Mark Brown wrote:
>
>> > > If the DAI format setup fails, there is no valid communication format
>> > > between CPU and CODEC, so fail card instantiation, rather than continue
>> > > with a card that will most likely not function properly.
>
>> > This is another one where if nobody noticed a problem already and things
>> > just happened to be working this might break things, it's vanishingly
>> > unlikely to fix anything that was broken.
>
>> Same as the other patch: this patch suggests it fixes a real bug, and if
>> this patch is broken let's fix it.
>
>If anyone ran into this on the older kernel and fixed or worked
>around it locally there's a reasonable chance this will then
>break what they're doing.  The patch itself is perfectly fine but

But there's not much we can do here. We can't hold off on fixing
breakage such as this because existing users have workarounds for this.
Are we breaking kernel ABI with this patch then?

And what about new users? We'll let them get hit by the issue and
develop their own workarounds?

>that doesn't mean the rest of the changes it's being backported
>into are also fine.

This is fair, and we can always hold off on patches if you want more
time for them to be tested/reviewed. Is it the case here?

--
Thanks,
Sasha

Re: [PATCH AUTOSEL 5.2 040/123] ASoC: Fail card instantiation if DAI format setup fails

[Ricard Wanderlof]

On Wed, 28 Aug 2019, Sasha Levin wrote:

> On Tue, Aug 27, 2019 at 12:00:14PM +0100, Mark Brown wrote:
> > On Sun, Aug 25, 2019 at 09:35:15PM -0400, Sasha Levin wrote:
> > > On Wed, Aug 14, 2019 at 10:22:13AM +0100, Mark Brown wrote:
> > 
> > > > > If the DAI format setup fails, there is no valid communication format
> > > > > between CPU and CODEC, so fail card instantiation, rather than
> > > continue
> > > > > with a card that will most likely not function properly.
> > 
> > > > This is another one where if nobody noticed a problem already and things
> > > > just happened to be working this might break things, it's vanishingly
> > > > unlikely to fix anything that was broken.
> > 
> > > Same as the other patch: this patch suggests it fixes a real bug, and if
> > > this patch is broken let's fix it.
> > 
> > If anyone ran into this on the older kernel and fixed or worked
> > around it locally there's a reasonable chance this will then
> > break what they're doing.  The patch itself is perfectly fine but
> 
> But there's not much we can do here. We can't hold off on fixing
> breakage such as this because existing users have workarounds for this.
> Are we breaking kernel ABI with this patch then?
> 
> And what about new users? We'll let them get hit by the issue and
> develop their own workarounds?

My $0.02 here: In my specific case, we noticed the problem because there 
was an unexpected left shift in the captured audio data, since the codec 
and CPU DAIs were using different formats when the DAI format was not 
explicitly set. The fix for that was to add

simple-audio-card,format= "i2s";

to the devicetree audio card section which of course should have been 
there all the time. The fact that the kernel failed halt the 
initialization of the audio card lengthened the debug time, but did not 
provoke me to attempt a workaround, since the information (the error 
printout from the ALSA framework when an invalid daifmt setting was made) 
was actually right there in the kernel log.

Possibly there might be other usecases, but in our case, if the kernel had 
stopped the audio initialization it would then have been more obvious 
where to start looking.

/Ricard
-- 
Ricard Wolf Wanderlof                           ricardw(at)axis.com
Axis Communications AB, Lund, Sweden            www.axis.com
Phone +46 46 272 2016                           Fax +46 46 13 61 30

Re: revert: ASoC: Fail card instantiation if DAI format setup fails

[Ricard Wanderlof]

> > On Tue, Aug 27, 2019 at 12:00:14PM +0100, Mark Brown wrote:
> > > On Sun, Aug 25, 2019 at 09:35:15PM -0400, Sasha Levin wrote:
> > > > On Wed, Aug 14, 2019 at 10:22:13AM +0100, Mark Brown wrote:
> > > 
> > > > > > If the DAI format setup fails, there is no valid communication format
> > > > > > between CPU and CODEC, so fail card instantiation, rather than
> > > > continue
> > > > > > with a card that will most likely not function properly.
> > > 
> > > > > This is another one where if nobody noticed a problem already and things
> > > > > just happened to be working this might break things, it's vanishingly
> > > > > unlikely to fix anything that was broken.
> > > 
> > > > Same as the other patch: this patch suggests it fixes a real bug, and if
> > > > this patch is broken let's fix it.
> > > 
> > > If anyone ran into this on the older kernel and fixed or worked
> > > around it locally there's a reasonable chance this will then
> > > break what they're doing.  The patch itself is perfectly fine but

(Sorry about the mangled subject line, I'd accidentally deleted the 
original message from my inbox.)

I'm a bit bewildered here. As the author of the original patch I'm of 
course biased, and I can certainly understand the patch being dropped from 
existing release branches, since as Mark correctly states, it does not fix 
any broken behavior and might even break things that happen to work by 
chance.

But is this being dropped from the master branch as well? To me it makes 
the kernel behave in an inconsistent way, first reporting a failure to 
instantiate a specific sound card in the kernel log, but then seemingly 
bringing it up anyway.

/Ricard
-- 
Ricard Wolf Wanderlof                           ricardw(at)axis.com
Axis Communications AB, Lund, Sweden            www.axis.com
Phone +46 46 272 2016                           Fax +46 46 13 61 30

Re: revert: ASoC: Fail card instantiation if DAI format setup fails

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 771 bytes --]

On Fri, Sep 06, 2019 at 10:40:01AM +0200, Ricard Wanderlof wrote:

> But is this being dropped from the master branch as well? To me it makes 
> the kernel behave in an inconsistent way, first reporting a failure to 
> instantiate a specific sound card in the kernel log, but then seemingly 
> bringing it up anyway.

No, this is absolutely a good and positive change to have in
master and I'm not suggesting that we should drop it there -
sorry if I sounded like that.  I just want to be conservative for
stable so that we don't have anyone updating their stable kernel
and having their audio blow up on them, we don't want to do
anything that'd discourage people from taking stable updates and
hence missing out on security or critical stability updates.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: revert: ASoC: Fail card instantiation if DAI format setup fails

[Sasha Levin]

On Fri, Sep 06, 2019 at 11:58:24AM +0100, Mark Brown wrote:
>On Fri, Sep 06, 2019 at 10:40:01AM +0200, Ricard Wanderlof wrote:
>
>> But is this being dropped from the master branch as well? To me it makes
>> the kernel behave in an inconsistent way, first reporting a failure to
>> instantiate a specific sound card in the kernel log, but then seemingly
>> bringing it up anyway.
>
>No, this is absolutely a good and positive change to have in
>master and I'm not suggesting that we should drop it there -
>sorry if I sounded like that.  I just want to be conservative for
>stable so that we don't have anyone updating their stable kernel
>and having their audio blow up on them, we don't want to do
>anything that'd discourage people from taking stable updates and
>hence missing out on security or critical stability updates.

Hi Mark,

I'm sorry for not dropping this to begin with: I saw your nack and the
patch ended up still being released because of my fuck up rather than
me purposefuly ignoring your ack, sorry.

However, I'd like to say that I don't agree with it. I understand your
reasoning about keeping the stable trees conservative, but I feel that
going to the extreme with it will just encourage folks to not upgrade
between major versions.

I'd like to think that upgrading major versions should be the same as
upgrading minor ones (because numbers don't matter here). If that's not
the case, let's fix it!

--
Thanks,
Sasha

Re: revert: ASoC: Fail card instantiation if DAI format setup fails

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 837 bytes --]

On Fri, Sep 06, 2019 at 02:38:24PM -0400, Sasha Levin wrote:

> However, I'd like to say that I don't agree with it. I understand your
> reasoning about keeping the stable trees conservative, but I feel that
> going to the extreme with it will just encourage folks to not upgrade
> between major versions.

This is a case where the change can't possibly make anything work
in itself, it can only break things and help with debugging.  If
people are sitting on stable hopefully they're not still
debugging their systems.  I don't understand why you are pushing
so hard for this, there is very little upside on stable.

> I'd like to think that upgrading major versions should be the same as
> upgrading minor ones (because numbers don't matter here). If that's not
> the case, let's fix it!

Backporting this won't help achieve that aim.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]