ARM: kirkwood: fix a not initialized variable in the sound subsystem
PATCH 1/2 ARM: kirkwood: fix a not initialized variable in the sound subsystem
In the function kirkwood_set_rate, in case of a non dco supported rate and no external clock, the clock source was set to an undefined value. This patch just displays a message without changing the clock source.
--- sound/soc/kirkwood/kirkwood-i2s.c | 3 +++ 1 file changed, 3 insertions(+)
sound/soc/kirkwood/kirkwood-i2s.c +3 -0
@@ -118,6 +118,9 @@ static void kirkwood_set_rate(struct snd_soc_dai *dai,clk_set_rate (priv->extclk,256 * rate); clks_ctrl =KIRKWOOD_MCLK_SOURCE_EXTCLK ; + }else { +dev_err (dai->dev,"%s: no clock\n" , __func__); +return ; }writel (clks_ctrl, priv->io +KIRKWOOD_CLOCKS_CTRL ); }
Signature · 5 lines
1.7.10.4 -- Ken ar c'henta? | ** Breizh ha Linux atav! ** Jef | http://moinejf.free.fr/
PATCH 1/2 ARM: kirkwood: fix a not initialized variable in the sound subsystem
Quoted 14 lines · Jean-Francois Moine
In the function
kirkwood_set_rate, in case of a non dco supported rate and no external clock, the clock source was set to an undefined value. This patch just displays a message without changing the clock source.Signed-off-by Jean-Francois Moine<[email protected]>--- sound/soc/kirkwood/kirkwood-i2s.c | 3 +++ 1 file changed, 3 insertions(+)sound/soc/kirkwood/kirkwood-i2s.c +0
-0
sound/soc/kirkwood/kirkwood-i2s.c static void kirkwood_set_rate(struct snd_soc_dai *dai, · line 126
Quoted 9 lines
sound/soc/kirkwood/kirkwood-i2s.c +3
-0@@ -118,6 +118,9 @@ static void kirkwood_set_rate(struct snd_soc_dai *dai,clk_set_rate (priv->extclk,256 * rate); clks_ctrl =KIRKWOOD_MCLK_SOURCE_EXTCLK ; + }else { +dev_err (dai->dev,"%s: no clock\n" , __func__); +return ; }writel (clks_ctrl, priv->io +KIRKWOOD_CLOCKS_CTRL ); }
NACK.
Having no clock at all should be catched during _probe. Moreover, not having the internal clock enabled will lead to system hang due to clock gating. You should rather pass an optional (DT-only) extclk phandle on the second clocks property.
Sebastian
P.S. as this is alsa stuff you should have some alsa maintainers on your Cc list. Please run ./scripts/get_maintainer.pl on your patches next time. It will give you a list of people you have to to Cc.
PATCH 1/2 ARM: kirkwood: fix a not initialized variable in the sound subsystem
Quoted 32 lines · On Tue, 26 Mar 2013 20:41:40 +0100 Sebastian Hesselbarth…
Quoted 14 lines · Jean-Francois Moine
In the function
kirkwood_set_rate, in case of a non dco supported rate and no external clock, the clock source was set to an undefined value. This patch just displays a message without changing the clock source.Signed-off-by Jean-Francois Moine<[email protected]>--- sound/soc/kirkwood/kirkwood-i2s.c | 3 +++ 1 file changed, 3 insertions(+)sound/soc/kirkwood/kirkwood-i2s.c +0
-0sound/soc/kirkwood/kirkwood-i2s.c static void kirkwood_set_rate(struct snd_soc_dai *dai, · line 126
Quoted 9 lines
sound/soc/kirkwood/kirkwood-i2s.c +3
-0@@ -118,6 +118,9 @@ static void kirkwood_set_rate(struct snd_soc_dai *dai,clk_set_rate (priv->extclk,256 * rate); clks_ctrl =KIRKWOOD_MCLK_SOURCE_EXTCLK ; + }else { +dev_err (dai->dev,"%s: no clock\n" , __func__); +return ; }writel (clks_ctrl, priv->io +KIRKWOOD_CLOCKS_CTRL ); }NACK.
Having no clock at all should be catched during
_probe. Moreover, not having the internal clock enabled will lead to system hang due to clock gating. You should rather pass an optional (DT-only) extclk phandle on the second clocks property.
Quoted 5 lines
Sebastian
P.S. as this is alsa stuff you should have some alsa maintainers on your Cc list. Please run./scripts/get_maintainer.plon your patches next time. It will give you a list of people you have to to Cc.
It is a compilation error: the variable clks_ctrl is not initialized.
The sequence has been created by the commit 363589bf110aa0352a2031
"ASoC: kirkwood-i2s: add support for external clock rates"
Russell is in the Cc list.
Signature · 2 lines
Ken ar c'henta? | ** Breizh ha Linux atav! ** Jef | http://moinejf.free.fr/
PATCH 1/2 ARM: kirkwood: fix a not initialized variable in the sound subsystem
Quoted 35 lines · Jean-Francois Moine
Quoted 32 lines · On Tue, 26 Mar 2013 20:41:40 +0100 Sebastian…
Quoted 14 lines · Jean-Francois Moine
In the function
kirkwood_set_rate, in case of a non dco supported rate and no external clock, the clock source was set to an undefined value. This patch just displays a message without changing the clock source.Signed-off-by Jean-Francois Moine<[email protected]>--- sound/soc/kirkwood/kirkwood-i2s.c | 3 +++ 1 file changed, 3 insertions(+)sound/soc/kirkwood/kirkwood-i2s.c +0
-0sound/soc/kirkwood/kirkwood-i2s.c static void kirkwood_set_rate(struct snd_soc_dai *dai, · line 126
Quoted 9 lines
sound/soc/kirkwood/kirkwood-i2s.c +3
-0@@ -118,6 +118,9 @@ static void kirkwood_set_rate(struct snd_soc_dai *dai,clk_set_rate (priv->extclk,256 * rate); clks_ctrl =KIRKWOOD_MCLK_SOURCE_EXTCLK ; + }else { +dev_err (dai->dev,"%s: no clock\n" , __func__); +return ; }writel (clks_ctrl, priv->io +KIRKWOOD_CLOCKS_CTRL ); }NACK.
Having no clock at all should be catched during
_probe. Moreover, not having the internal clock enabled will lead to system hang due to clock gating. You should rather pass an optional (DT-only) extclk phandle on the second clocks property.
Quoted 7 lines
Quoted 5 lines
Sebastian
P.S. as this is alsa stuff you should have some alsa maintainers on your Cc list. Please run./scripts/get_maintainer.plon your patches next time. It will give you a list of people you have to to Cc.It is a compilation error: the variable
clks_ctrlis not initialized.
If it is about non-initialized clks_ctrl, I suggest to remove it completely and clone the writel to both if and else branch.
Quoted 3 lines
The sequence has been created by the commit
363589bf110aa0352a2031
"ASoC: kirkwood-i2s: add support for external clock rates"
Russell is in the Cc list.
Russell was the commit author and also should be added because of his ARM maintainer status. But kirkwood-i2s is alsa and needs some sound maintainers to be Cc'd:
$ ./scripts/get_maintainer.pl -f sound/soc/kirkwood/kirkwood-i2s.c Liam Girdwood <[email protected]> (supporter:SOUND - SOC LAYER...) Mark Brown <[email protected]> (supporter:SOUND - SOC LAYER...,commit_signer:9/11=82%) Jaroslav Kysela <[email protected]> (maintainer:SOUND) Takashi Iwai <[email protected]> (maintainer:SOUND) Russell King <[email protected]> (commit_signer:6/11=55%) Greg Kroah-Hartman <[email protected]> (commit_signer:2/11=18%) Andrew Lumm <[email protected]> (commit_signer:2/11=18%) Thierry Reding <[email protected]> (commit_signer:1/11=9%) alsa-devel at alsa-project.org (moderated list:SOUND - SOC LAYER...) linux-kernel at vger.kernel.org (open list)
Sebastian
PATCH 1/2 ARM: kirkwood: fix a not initialized variable in the sound subsystem
In reply to Jean-Francois MoineQuoted 30 lines · Jean-Francois Moine
Quoted 27 lines · On Tue, 26 Mar 2013 20:41:40 +0100 Sebastian…
Quoted 14 lines · Jean-Francois Moine
In the function
kirkwood_set_rate, in case of a non dco supported rate and no external clock, the clock source was set to an undefined value. This patch just displays a message without changing the clock source.Signed-off-by Jean-Francois Moine<[email protected]>--- sound/soc/kirkwood/kirkwood-i2s.c | 3 +++ 1 file changed, 3 insertions(+)sound/soc/kirkwood/kirkwood-i2s.c +0
-0sound/soc/kirkwood/kirkwood-i2s.c static void kirkwood_set_rate(struct snd_soc_dai *dai, · line 126
Quoted 9 lines
sound/soc/kirkwood/kirkwood-i2s.c +3
-0@@ -118,6 +118,9 @@ static void kirkwood_set_rate(struct snd_soc_dai *dai,clk_set_rate (priv->extclk,256 * rate); clks_ctrl =KIRKWOOD_MCLK_SOURCE_EXTCLK ; + }else { +dev_err (dai->dev,"%s: no clock\n" , __func__); +return ; }writel (clks_ctrl, priv->io +KIRKWOOD_CLOCKS_CTRL ); }NACK.
Quoted 4 lines
Quoted 4 lines
Having no clock at all should be catched during
_probe. Moreover, not having the internal clock enabled will lead to system hang due to clock gating. You should rather pass an optional (DT-only) extclk phandle on the second clocks property.
Jean-Francois,
I had a close look at the code and your patch. From a driver point-of-view kirkwood_set_rate should never been called with an unsupported rate (see KIRKWOOD_I2S_RATES) when no extclk is available. With extclk available, the else-if branch will be taken on rates != KIRKWOOD_I2S_RATES. Actually, your added else branch will never be taken at all.
I suggest (again) to remove clks_ctrl and move the writel inside if and else-if branch to cure the compiler warning.
Sebastian
PATCH 1/2 ARM: kirkwood: fix a not initialized variable in the sound subsystem
Quoted 45 lines · On Tue, 26 Mar 2013 21:39:40 +0100 Sebastian Hesselbarth…
Quoted 30 lines · Jean-Francois Moine
Quoted 27 lines · On Tue, 26 Mar 2013 20:41:40 +0100 Sebastian…
Quoted 14 lines · Jean-Francois Moine
In the function
kirkwood_set_rate, in case of a non dco supported rate and no external clock, the clock source was set to an undefined value. This patch just displays a message without changing the clock source.Signed-off-by Jean-Francois Moine<[email protected]>--- sound/soc/kirkwood/kirkwood-i2s.c | 3 +++ 1 file changed, 3 insertions(+)sound/soc/kirkwood/kirkwood-i2s.c +0
-0sound/soc/kirkwood/kirkwood-i2s.c static void kirkwood_set_rate(struct snd_soc_dai *dai, · line 126
Quoted 9 lines
sound/soc/kirkwood/kirkwood-i2s.c +3
-0@@ -118,6 +118,9 @@ static void kirkwood_set_rate(struct snd_soc_dai *dai,clk_set_rate (priv->extclk,256 * rate); clks_ctrl =KIRKWOOD_MCLK_SOURCE_EXTCLK ; + }else { +dev_err (dai->dev,"%s: no clock\n" , __func__); +return ; }writel (clks_ctrl, priv->io +KIRKWOOD_CLOCKS_CTRL ); }NACK.
Quoted 4 lines
Quoted 4 lines
Having no clock at all should be catched during
_probe. Moreover, not having the internal clock enabled will lead to system hang due to clock gating. You should rather pass an optional (DT-only) extclk phandle on the second clocks property.Jean-Francois,
I had a close look at the code and your patch. From a driver point-of-view
kirkwood_set_rateshould never been called with an unsupported rate (seeKIRKWOOD_I2S_RATES) when no extclk is available. With extclk available, the else-if branch will be taken on rates !=KIRKWOOD_I2S_RATES. Actually, your added else branch will never be taken at all.
Quoted 4 lines
I suggest (again) to remove
clks_ctrland move the writel inside if and else-if branch to cure the compiler warning.Sebastian
That's what there was in the original patch from Rabeeh, but maybe it is the opportunity to use WARN_ON. Russell?
Signature · 2 lines
Ken ar c'henta? | ** Breizh ha Linux atav! ** Jef | http://moinejf.free.fr/
PATCH 1/2 ARM: kirkwood: fix a not initialized variable in the sound subsystem
Quoted 9 lines · Jean-Francois Moine
Quoted 4 lines · On Tue, 26 Mar 2013 21:39:40 +0100 Sebastian Hesselbarth…
I suggest (again) to remove
clks_ctrland move the writel inside if and else-if branch to cure the compiler warning.Sebastian
That's what there was in the original patch from Rabeeh, but maybe it is the opportunity to use
WARN_ON. Russell?
Sebastian is correct in that such a path should _never_ be reached because ALSA will reject anything but 44.1, 48 or 96kHz rates if we don't have an extclk.
However, I disagree with Sebastian's solution - doing that is likely to generate more code because gcc tends not to optimise:
if (foo) {
writel(some_value, register);
} else {
writel(some_other_value, register);
}very well. It will generate two separate writel()s when in fact, it could just do:
if (foo) {
val = some_value;
} else {
val = some_other_value;
}
writel(val, register);One solution to this would be to just get rid of "if (!IS_ERR(priv->extclk))" entirely, so that the "else" clause always assumes that if we hit that, there will be an external clock. If it does, the clk API gets to deal with being passed an error pointer for a clock, and we switch to extclk mode. Either that'll cause a nice backtrace from the kernel (which is good in this case) or audio will just not work.
Remember, though - if there isn't an extclk set, then we should never get there in the first place. If it makes people feel happier, also put a WARN_ON(IS_ERR(priv->extclk)) there to guarantee a backtrace if it happens.