diff mbox series

ASoC: topology: Fix route memory corruption

Message ID 20240613090126.841189-1-amadeuszx.slawinski@linux.intel.com (mailing list archive)
State Accepted
Commit 0298f51652be47b79780833e0b63194e1231fa34
Headers show
Series ASoC: topology: Fix route memory corruption | expand

Commit Message

Amadeusz Sławiński June 13, 2024, 9:01 a.m. UTC
It was reported that recent fix for memory corruption during topology
load, causes corruption in other cases. Instead of being overeager with
checking topology, assume that it is properly formatted and just
duplicate strings.

Reported-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Closes: https://lore.kernel.org/linux-sound/171812236450.201359.3019210915105428447.b4-ty@kernel.org/T/#m8c4bd5abf453960fde6f826c4b7f84881da63e9d
Suggested-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
---
 sound/soc/soc-topology.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

Comments

Peter Ujfalusi June 13, 2024, 11:11 a.m. UTC | #1
On 13/06/2024 12:01, Amadeusz Sławiński wrote:
> It was reported that recent fix for memory corruption during topology
> load, causes corruption in other cases. Instead of being overeager with
> checking topology, assume that it is properly formatted and just
> duplicate strings.

Reviewed-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>

> Reported-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
> Closes: https://lore.kernel.org/linux-sound/171812236450.201359.3019210915105428447.b4-ty@kernel.org/T/#m8c4bd5abf453960fde6f826c4b7f84881da63e9d
> Suggested-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
> Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
> ---
>  sound/soc/soc-topology.c | 12 +++---------
>  1 file changed, 3 insertions(+), 9 deletions(-)
> 
> diff --git a/sound/soc/soc-topology.c b/sound/soc/soc-topology.c
> index 0225bc5fc425a..4b166294602fa 100644
> --- a/sound/soc/soc-topology.c
> +++ b/sound/soc/soc-topology.c
> @@ -1052,21 +1052,15 @@ static int soc_tplg_dapm_graph_elems_load(struct soc_tplg *tplg,
>  			break;
>  		}
>  
> -		route->source = devm_kmemdup(tplg->dev, elem->source,
> -					     min(strlen(elem->source), maxlen),
> -					     GFP_KERNEL);
> -		route->sink = devm_kmemdup(tplg->dev, elem->sink,
> -					   min(strlen(elem->sink), maxlen),
> -					   GFP_KERNEL);
> +		route->source = devm_kstrdup(tplg->dev, elem->source, GFP_KERNEL);
> +		route->sink = devm_kstrdup(tplg->dev, elem->sink, GFP_KERNEL);
>  		if (!route->source || !route->sink) {
>  			ret = -ENOMEM;
>  			break;
>  		}
>  
>  		if (strnlen(elem->control, maxlen) != 0) {
> -			route->control = devm_kmemdup(tplg->dev, elem->control,
> -						      min(strlen(elem->control), maxlen),
> -						      GFP_KERNEL);
> +			route->control = devm_kstrdup(tplg->dev, elem->control, GFP_KERNEL);
>  			if (!route->control) {
>  				ret = -ENOMEM;
>  				break;
Pierre-Louis Bossart June 13, 2024, 2:13 p.m. UTC | #2
On 6/13/24 13:11, Péter Ujfalusi wrote:
> 
> 
> On 13/06/2024 12:01, Amadeusz Sławiński wrote:
>> It was reported that recent fix for memory corruption during topology
>> load, causes corruption in other cases. Instead of being overeager with
>> checking topology, assume that it is properly formatted and just
>> duplicate strings.
> 
> Reviewed-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>

Test results look much better indeed with
https://github.com/thesofproject/linux/pull/5057


Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>

> 
>> Reported-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
>> Closes: https://lore.kernel.org/linux-sound/171812236450.201359.3019210915105428447.b4-ty@kernel.org/T/#m8c4bd5abf453960fde6f826c4b7f84881da63e9d
>> Suggested-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
>> Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
>> ---
>>  sound/soc/soc-topology.c | 12 +++---------
>>  1 file changed, 3 insertions(+), 9 deletions(-)
>>
>> diff --git a/sound/soc/soc-topology.c b/sound/soc/soc-topology.c
>> index 0225bc5fc425a..4b166294602fa 100644
>> --- a/sound/soc/soc-topology.c
>> +++ b/sound/soc/soc-topology.c
>> @@ -1052,21 +1052,15 @@ static int soc_tplg_dapm_graph_elems_load(struct soc_tplg *tplg,
>>  			break;
>>  		}
>>  
>> -		route->source = devm_kmemdup(tplg->dev, elem->source,
>> -					     min(strlen(elem->source), maxlen),
>> -					     GFP_KERNEL);
>> -		route->sink = devm_kmemdup(tplg->dev, elem->sink,
>> -					   min(strlen(elem->sink), maxlen),
>> -					   GFP_KERNEL);
>> +		route->source = devm_kstrdup(tplg->dev, elem->source, GFP_KERNEL);
>> +		route->sink = devm_kstrdup(tplg->dev, elem->sink, GFP_KERNEL);
>>  		if (!route->source || !route->sink) {
>>  			ret = -ENOMEM;
>>  			break;
>>  		}
>>  
>>  		if (strnlen(elem->control, maxlen) != 0) {
>> -			route->control = devm_kmemdup(tplg->dev, elem->control,
>> -						      min(strlen(elem->control), maxlen),
>> -						      GFP_KERNEL);
>> +			route->control = devm_kstrdup(tplg->dev, elem->control, GFP_KERNEL);
>>  			if (!route->control) {
>>  				ret = -ENOMEM;
>>  				break;
>
Mark Brown June 14, 2024, 9:24 a.m. UTC | #3
On Thu, 13 Jun 2024 11:01:26 +0200, Amadeusz Sławiński wrote:
> It was reported that recent fix for memory corruption during topology
> load, causes corruption in other cases. Instead of being overeager with
> checking topology, assume that it is properly formatted and just
> duplicate strings.
> 
> 

Applied to

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-next

Thanks!

[1/1] ASoC: topology: Fix route memory corruption
      commit: b9dd212b14d27a53b63fc6621c452c0b3a01f61d

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark
Mark Brown June 14, 2024, 2:16 p.m. UTC | #4
On Thu, 13 Jun 2024 11:01:26 +0200, Amadeusz Sławiński wrote:
> It was reported that recent fix for memory corruption during topology
> load, causes corruption in other cases. Instead of being overeager with
> checking topology, assume that it is properly formatted and just
> duplicate strings.
> 
> 

Applied to

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-next

Thanks!

[1/1] ASoC: topology: Fix route memory corruption
      commit: 0298f51652be47b79780833e0b63194e1231fa34

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark
Joakim Tjernlund Aug. 13, 2024, 5:11 p.m. UTC | #5
On Thu, 2024-06-13 at 11:01 +0200, Amadeusz Sławiński wrote:
> It was reported that recent fix for memory corruption during topology
> load, causes corruption in other cases. Instead of being overeager with
> checking topology, assume that it is properly formatted and just
> duplicate strings.
> 
> Reported-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
> Closes: https://lore.kernel.org/linux-sound/171812236450.201359.3019210915105428447.b4-ty@kernel.org/T/#m8c4bd5abf453960fde6f826c4b7f84881da63e9d
> Suggested-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
> Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
> ---
>  sound/soc/soc-topology.c | 12 +++---------
>  1 file changed, 3 insertions(+), 9 deletions(-)
> 
> diff --git a/sound/soc/soc-topology.c b/sound/soc/soc-topology.c
> index 0225bc5fc425a..4b166294602fa 100644
> --- a/sound/soc/soc-topology.c
> +++ b/sound/soc/soc-topology.c
> @@ -1052,21 +1052,15 @@ static int soc_tplg_dapm_graph_elems_load(struct soc_tplg *tplg,
>  			break;
>  		}
>  
> -		route->source = devm_kmemdup(tplg->dev, elem->source,
> -					     min(strlen(elem->source), maxlen),
> -					     GFP_KERNEL);
> -		route->sink = devm_kmemdup(tplg->dev, elem->sink,
> -					   min(strlen(elem->sink), maxlen),
> -					   GFP_KERNEL);
> +		route->source = devm_kstrdup(tplg->dev, elem->source, GFP_KERNEL);
> +		route->sink = devm_kstrdup(tplg->dev, elem->sink, GFP_KERNEL);
>  		if (!route->source || !route->sink) {
>  			ret = -ENOMEM;
>  			break;
>  		}
>  
>  		if (strnlen(elem->control, maxlen) != 0) {
> -			route->control = devm_kmemdup(tplg->dev, elem->control,
> -						      min(strlen(elem->control), maxlen),
> -						      GFP_KERNEL);
> +			route->control = devm_kstrdup(tplg->dev, elem->control, GFP_KERNEL);
>  			if (!route->control) {
>  				ret = -ENOMEM;
>  				break;

I am getting very similar corruption/SOF errors in 6.6.45 but there is no similar
fix in 6.6.x that I can see. Hand hacked this patch and now the errors are gone:

--- ./sound/soc/soc-topology.c.org	2024-08-13 17:58:49.837295484 +0200
+++ ./sound/soc/soc-topology.c	2024-08-13 18:20:44.564107024 +0200
@@ -1060,23 +1060,15 @@
 			break;
 		}
 
-		route->source = devm_kmemdup(tplg->dev, elem->source,
-					     min(strlen(elem->source),
-						 SNDRV_CTL_ELEM_ID_NAME_MAXLEN),
-					     GFP_KERNEL);
-		route->sink = devm_kmemdup(tplg->dev, elem->sink,
-					   min(strlen(elem->sink), SNDRV_CTL_ELEM_ID_NAME_MAXLEN),
-					   GFP_KERNEL);
+		route->source = devm_kstrdup(tplg->dev, elem->source, GFP_KERNEL);
+		route->sink = devm_kstrdup(tplg->dev, elem->sink, GFP_KERNEL);
 		if (!route->source || !route->sink) {
 			ret = -ENOMEM;
 			break;
 		}
 
 		if (strnlen(elem->control, SNDRV_CTL_ELEM_ID_NAME_MAXLEN) != 0) {
-			route->control = devm_kmemdup(tplg->dev, elem->control,
-						      min(strlen(elem->control),
-							  SNDRV_CTL_ELEM_ID_NAME_MAXLEN),
-						      GFP_KERNEL);
+			route->control = devm_kstrdup(tplg->dev, elem->control, GFP_KERNEL);
 			if (!route->control) {
 				ret = -ENOMEM;
 				break;

     
Just luck?

 Jocke
Joakim Tjernlund Aug. 15, 2024, 7:23 a.m. UTC | #6
On Tue, 2024-08-13 at 19:11 +0200, Joakim Tjernlund wrote:
> On Thu, 2024-06-13 at 11:01 +0200, Amadeusz Sławiński wrote:
> > It was reported that recent fix for memory corruption during topology
> > load, causes corruption in other cases. Instead of being overeager with
> > checking topology, assume that it is properly formatted and just
> > duplicate strings.
> > 
> > Reported-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
> > Closes: https://lore.kernel.org/linux-sound/171812236450.201359.3019210915105428447.b4-ty@kernel.org/T/#m8c4bd5abf453960fde6f826c4b7f84881da63e9d
> > Suggested-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
> > Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
> > ---
> >  sound/soc/soc-topology.c | 12 +++---------
> >  1 file changed, 3 insertions(+), 9 deletions(-)
> > 
> > diff --git a/sound/soc/soc-topology.c b/sound/soc/soc-topology.c
> > index 0225bc5fc425a..4b166294602fa 100644
> > --- a/sound/soc/soc-topology.c
> > +++ b/sound/soc/soc-topology.c
> > @@ -1052,21 +1052,15 @@ static int soc_tplg_dapm_graph_elems_load(struct soc_tplg *tplg,
> >  			break;
> >  		}
> >  
> > -		route->source = devm_kmemdup(tplg->dev, elem->source,
> > -					     min(strlen(elem->source), maxlen),
> > -					     GFP_KERNEL);
> > -		route->sink = devm_kmemdup(tplg->dev, elem->sink,
> > -					   min(strlen(elem->sink), maxlen),
> > -					   GFP_KERNEL);
> > +		route->source = devm_kstrdup(tplg->dev, elem->source, GFP_KERNEL);
> > +		route->sink = devm_kstrdup(tplg->dev, elem->sink, GFP_KERNEL);
> >  		if (!route->source || !route->sink) {
> >  			ret = -ENOMEM;
> >  			break;
> >  		}
> >  
> >  		if (strnlen(elem->control, maxlen) != 0) {
> > -			route->control = devm_kmemdup(tplg->dev, elem->control,
> > -						      min(strlen(elem->control), maxlen),
> > -						      GFP_KERNEL);
> > +			route->control = devm_kstrdup(tplg->dev, elem->control, GFP_KERNEL);
> >  			if (!route->control) {
> >  				ret = -ENOMEM;
> >  				break;
> 
> I am getting very similar corruption/SOF errors in 6.6.45 but there is no similar
> fix in 6.6.x that I can see. Hand hacked this patch and now the errors are gone:
> 
> --- ./sound/soc/soc-topology.c.org	2024-08-13 17:58:49.837295484 +0200
> +++ ./sound/soc/soc-topology.c	2024-08-13 18:20:44.564107024 +0200
> @@ -1060,23 +1060,15 @@
>  			break;
>  		}
>  
> -		route->source = devm_kmemdup(tplg->dev, elem->source,
> -					     min(strlen(elem->source),
> -						 SNDRV_CTL_ELEM_ID_NAME_MAXLEN),
> -					     GFP_KERNEL);
> -		route->sink = devm_kmemdup(tplg->dev, elem->sink,
> -					   min(strlen(elem->sink), SNDRV_CTL_ELEM_ID_NAME_MAXLEN),
> -					   GFP_KERNEL);
> +		route->source = devm_kstrdup(tplg->dev, elem->source, GFP_KERNEL);
> +		route->sink = devm_kstrdup(tplg->dev, elem->sink, GFP_KERNEL);
>  		if (!route->source || !route->sink) {
>  			ret = -ENOMEM;
>  			break;
>  		}
>  
>  		if (strnlen(elem->control, SNDRV_CTL_ELEM_ID_NAME_MAXLEN) != 0) {
> -			route->control = devm_kmemdup(tplg->dev, elem->control,
> -						      min(strlen(elem->control),
> -							  SNDRV_CTL_ELEM_ID_NAME_MAXLEN),
> -						      GFP_KERNEL);
> +			route->control = devm_kstrdup(tplg->dev, elem->control, GFP_KERNEL);
>  			if (!route->control) {
>  				ret = -ENOMEM;
>  				break;
> 
>      
> Just luck?
> 
>  Jocke

Never mind, it was just added to 6.6.x stable queue.

 Jocke
diff mbox series

Patch

diff --git a/sound/soc/soc-topology.c b/sound/soc/soc-topology.c
index 0225bc5fc425a..4b166294602fa 100644
--- a/sound/soc/soc-topology.c
+++ b/sound/soc/soc-topology.c
@@ -1052,21 +1052,15 @@  static int soc_tplg_dapm_graph_elems_load(struct soc_tplg *tplg,
 			break;
 		}
 
-		route->source = devm_kmemdup(tplg->dev, elem->source,
-					     min(strlen(elem->source), maxlen),
-					     GFP_KERNEL);
-		route->sink = devm_kmemdup(tplg->dev, elem->sink,
-					   min(strlen(elem->sink), maxlen),
-					   GFP_KERNEL);
+		route->source = devm_kstrdup(tplg->dev, elem->source, GFP_KERNEL);
+		route->sink = devm_kstrdup(tplg->dev, elem->sink, GFP_KERNEL);
 		if (!route->source || !route->sink) {
 			ret = -ENOMEM;
 			break;
 		}
 
 		if (strnlen(elem->control, maxlen) != 0) {
-			route->control = devm_kmemdup(tplg->dev, elem->control,
-						      min(strlen(elem->control), maxlen),
-						      GFP_KERNEL);
+			route->control = devm_kstrdup(tplg->dev, elem->control, GFP_KERNEL);
 			if (!route->control) {
 				ret = -ENOMEM;
 				break;