[PATCH 1/6] ASoC: topology: Constify an argument of snd_soc_tplg_component_load()

[PATCH 1/6] ASoC: topology: Constify an argument of snd_soc_tplg_component_load()

[Christophe JAILLET]

snd_soc_tplg_component_load() does not modify its "*ops" argument. It
only read some values and stores it in "soc_tplg.ops".

This argument and the ops field in "struct soc_tplg" can be made const.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
---
 include/sound/soc-topology.h | 2 +-
 sound/soc/soc-topology.c     | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/sound/soc-topology.h b/include/sound/soc-topology.h
index f055c6917f6c..1eedd203ac29 100644
--- a/include/sound/soc-topology.h
+++ b/include/sound/soc-topology.h
@@ -178,7 +178,7 @@ static inline const void *snd_soc_tplg_get_data(struct snd_soc_tplg_hdr *hdr)
 
 /* Dynamic Object loading and removal for component drivers */
 int snd_soc_tplg_component_load(struct snd_soc_component *comp,
-	struct snd_soc_tplg_ops *ops, const struct firmware *fw);
+	const struct snd_soc_tplg_ops *ops, const struct firmware *fw);
 int snd_soc_tplg_component_remove(struct snd_soc_component *comp);
 
 /* Binds event handlers to dynamic widgets */
diff --git a/sound/soc/soc-topology.c b/sound/soc/soc-topology.c
index 90ca37e008b3..b00ec01361c2 100644
--- a/sound/soc/soc-topology.c
+++ b/sound/soc/soc-topology.c
@@ -73,7 +73,7 @@ struct soc_tplg {
 	int bytes_ext_ops_count;
 
 	/* optional fw loading callbacks to component drivers */
-	struct snd_soc_tplg_ops *ops;
+	const struct snd_soc_tplg_ops *ops;
 };
 
 /* check we dont overflow the data for this control chunk */
@@ -2334,7 +2334,7 @@ static int soc_tplg_load(struct soc_tplg *tplg)
 
 /* load audio component topology from "firmware" file */
 int snd_soc_tplg_component_load(struct snd_soc_component *comp,
-	struct snd_soc_tplg_ops *ops, const struct firmware *fw)
+	const struct snd_soc_tplg_ops *ops, const struct firmware *fw)
 {
 	struct soc_tplg tplg;
 	int ret;
-- 
2.45.0

[PATCH 2/6] ASoC: Intel: avs: Constify struct snd_soc_tplg_ops

[Christophe JAILLET]

Constifying "struct snd_soc_tplg_ops" moves some data to a read-only
section, so increase overall security.

On a x86_64, with allmodconfig:
Before:
   text	   data	    bss	    dec	    hex	filename
  28046	    794	      0	  28840	   70a8	sound/soc/intel/avs/topology.o

After:
   text	   data	    bss	    dec	    hex	filename
  28206	    614	      0	  28820	   7094	sound/soc/intel/avs/topology.o

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
---
Compile tested-only.
---
 sound/soc/intel/avs/topology.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sound/soc/intel/avs/topology.c b/sound/soc/intel/avs/topology.c
index 02bae207f6ec..35381a835c93 100644
--- a/sound/soc/intel/avs/topology.c
+++ b/sound/soc/intel/avs/topology.c
@@ -1889,7 +1889,7 @@ avs_control_load(struct snd_soc_component *comp, int index, struct snd_kcontrol_
 	return 0;
 }
 
-static struct snd_soc_tplg_ops avs_tplg_ops = {
+static const struct snd_soc_tplg_ops avs_tplg_ops = {
 	.io_ops			= avs_control_ops,
 	.io_ops_count		= ARRAY_SIZE(avs_control_ops),
 	.control_load		= avs_control_load,
-- 
2.45.0

Re: [PATCH 2/6] ASoC: Intel: avs: Constify struct snd_soc_tplg_ops

[Amadeusz Sławiński]

On 5/13/2024 7:37 PM, Christophe JAILLET wrote:
> Constifying "struct snd_soc_tplg_ops" moves some data to a read-only
> section, so increase overall security.
> 
> On a x86_64, with allmodconfig:
> Before:
>     text	   data	    bss	    dec	    hex	filename
>    28046	    794	      0	  28840	   70a8	sound/soc/intel/avs/topology.o
> 
> After:
>     text	   data	    bss	    dec	    hex	filename
>    28206	    614	      0	  28820	   7094	sound/soc/intel/avs/topology.o
> 
> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> ---

Reviewed-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>

[PATCH 3/6] ASoC: qdsp6: audioreach: Constify struct snd_soc_tplg_ops

[Christophe JAILLET]

Constifying "struct snd_soc_tplg_ops" moves some data to a read-only
section, so increase overall security.

On a x86_64, with allmodconfig:
Before:
   text	   data	    bss	    dec	    hex	filename
  19942	    832	      0	  20774	   5126	sound/soc/qcom/qdsp6/topology.o

After:
   text	   data	    bss	    dec	    hex	filename
  20102	    652	      0	  20754	   5112	sound/soc/qcom/qdsp6/topology.o

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
---
Compile tested-only.
---
 sound/soc/qcom/qdsp6/topology.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sound/soc/qcom/qdsp6/topology.c b/sound/soc/qcom/qdsp6/topology.c
index 70572c83e101..c15d1a2b6dbf 100644
--- a/sound/soc/qcom/qdsp6/topology.c
+++ b/sound/soc/qcom/qdsp6/topology.c
@@ -1240,7 +1240,7 @@ static const struct snd_soc_tplg_kcontrol_ops audioreach_io_ops[] = {
 		audioreach_put_vol_ctrl_audio_mixer, snd_soc_info_volsw},
 };
 
-static struct snd_soc_tplg_ops audioreach_tplg_ops  = {
+static const struct snd_soc_tplg_ops audioreach_tplg_ops = {
 	.io_ops = audioreach_io_ops,
 	.io_ops_count = ARRAY_SIZE(audioreach_io_ops),
 
-- 
2.45.0

[PATCH 4/6] ASoC: Intel: Skylake: Constify struct snd_soc_tplg_ops

[Christophe JAILLET]

Constifying "struct snd_soc_tplg_ops" moves some data to a read-only
section, so increase overall security.

On a x86_64, with allmodconfig:
Before:
   text	   data	    bss	    dec	    hex	filename
  58844	   5282	     56	  64182	   fab6	sound/soc/intel/skylake/skl-topology.o

After:
   text	   data	    bss	    dec	    hex	filename
  59004	   5122	     56	  64182	   fab6	sound/soc/intel/skylake/skl-topology.o

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
---
Compile tested-only.
---
 sound/soc/intel/skylake/skl-topology.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sound/soc/intel/skylake/skl-topology.c b/sound/soc/intel/skylake/skl-topology.c
index e27f0fc3d897..602ef4321122 100644
--- a/sound/soc/intel/skylake/skl-topology.c
+++ b/sound/soc/intel/skylake/skl-topology.c
@@ -3470,7 +3470,7 @@ static int skl_tplg_complete(struct snd_soc_component *component)
 	return 0;
 }
 
-static struct snd_soc_tplg_ops skl_tplg_ops  = {
+static const struct snd_soc_tplg_ops skl_tplg_ops = {
 	.widget_load = skl_tplg_widget_load,
 	.control_load = skl_tplg_control_load,
 	.bytes_ext_ops = skl_tlv_ops,
-- 
2.45.0

Re: [PATCH 4/6] ASoC: Intel: Skylake: Constify struct snd_soc_tplg_ops

[Amadeusz Sławiński]

On 5/13/2024 7:37 PM, Christophe JAILLET wrote:
> Constifying "struct snd_soc_tplg_ops" moves some data to a read-only
> section, so increase overall security.
> 
> On a x86_64, with allmodconfig:
> Before:
>     text	   data	    bss	    dec	    hex	filename
>    58844	   5282	     56	  64182	   fab6	sound/soc/intel/skylake/skl-topology.o
> 
> After:
>     text	   data	    bss	    dec	    hex	filename
>    59004	   5122	     56	  64182	   fab6	sound/soc/intel/skylake/skl-topology.o
> 
> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> ---

Reviewed-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>

[PATCH 5/6] ASoC: SOF: topology: Constify struct snd_soc_tplg_ops

[Christophe JAILLET]

Constifying "struct snd_soc_tplg_ops" moves some data to a read-only
section, so increase overall security.

On a x86_64, with allmodconfig:
Before:
   text	   data	    bss	    dec	    hex	filename
  44519	   2888	     48	  47455	   b95f	sound/soc/sof/topology.o

After:
   text	   data	    bss	    dec	    hex	filename
  44839	   2552	     48	  47439	   b94f	sound/soc/sof/topology.o

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
---
Compile tested-only.
---
 sound/soc/sof/topology.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/sound/soc/sof/topology.c b/sound/soc/sof/topology.c
index da182314aa87..b54382131991 100644
--- a/sound/soc/sof/topology.c
+++ b/sound/soc/sof/topology.c
@@ -2278,7 +2278,7 @@ static const struct snd_soc_tplg_bytes_ext_ops sof_bytes_ext_ops[] = {
 	{SOF_TPLG_KCTL_BYTES_VOLATILE_RO, snd_sof_bytes_ext_volatile_get},
 };
 
-static struct snd_soc_tplg_ops sof_tplg_ops = {
+static const struct snd_soc_tplg_ops sof_tplg_ops = {
 	/* external kcontrol init - used for any driver specific init */
 	.control_load	= sof_control_load,
 	.control_unload	= sof_control_unload,
@@ -2433,7 +2433,7 @@ static int sof_dspless_link_load(struct snd_soc_component *scomp, int index,
 	return 0;
 }
 
-static struct snd_soc_tplg_ops sof_dspless_tplg_ops = {
+static const struct snd_soc_tplg_ops sof_dspless_tplg_ops = {
 	/* external widget init - used for any driver specific init */
 	.widget_ready	= sof_dspless_widget_ready,
 	.widget_unload	= sof_dspless_widget_unload,
-- 
2.45.0

[PATCH 6/6] const_structs.checkpatch: add snd_soc_tplg_ops

[Christophe JAILLET]

Now that the soc-topology can handle "const struct snd_soc_tplg_ops" make
sure that new usages of the struct already enter the tree as const.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
---
 scripts/const_structs.checkpatch | 1 +
 1 file changed, 1 insertion(+)

diff --git a/scripts/const_structs.checkpatch b/scripts/const_structs.checkpatch
index fa96cfd16e99..bae41780bd39 100644
--- a/scripts/const_structs.checkpatch
+++ b/scripts/const_structs.checkpatch
@@ -78,6 +78,7 @@ snd_rawmidi_ops
 snd_soc_component_driver
 snd_soc_dai_ops
 snd_soc_ops
+snd_soc_tplg_ops
 soc_pcmcia_socket_ops
 stacktrace_ops
 sysfs_ops
-- 
2.45.0

Re: [PATCH 1/6] ASoC: topology: Constify an argument of snd_soc_tplg_component_load()

[Amadeusz Sławiński]

On 5/13/2024 7:37 PM, Christophe JAILLET wrote:
> snd_soc_tplg_component_load() does not modify its "*ops" argument. It
> only read some values and stores it in "soc_tplg.ops".
> 
> This argument and the ops field in "struct soc_tplg" can be made const.
> 
> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> ---
>   include/sound/soc-topology.h | 2 +-
>   sound/soc/soc-topology.c     | 4 ++--
>   2 files changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/include/sound/soc-topology.h b/include/sound/soc-topology.h
> index f055c6917f6c..1eedd203ac29 100644
> --- a/include/sound/soc-topology.h
> +++ b/include/sound/soc-topology.h
> @@ -178,7 +178,7 @@ static inline const void *snd_soc_tplg_get_data(struct snd_soc_tplg_hdr *hdr)
>   
>   /* Dynamic Object loading and removal for component drivers */
>   int snd_soc_tplg_component_load(struct snd_soc_component *comp,
> -	struct snd_soc_tplg_ops *ops, const struct firmware *fw);
> +	const struct snd_soc_tplg_ops *ops, const struct firmware *fw);
>   int snd_soc_tplg_component_remove(struct snd_soc_component *comp);
>   
>   /* Binds event handlers to dynamic widgets */
> diff --git a/sound/soc/soc-topology.c b/sound/soc/soc-topology.c
> index 90ca37e008b3..b00ec01361c2 100644
> --- a/sound/soc/soc-topology.c
> +++ b/sound/soc/soc-topology.c
> @@ -73,7 +73,7 @@ struct soc_tplg {
>   	int bytes_ext_ops_count;
>   
>   	/* optional fw loading callbacks to component drivers */
> -	struct snd_soc_tplg_ops *ops;
> +	const struct snd_soc_tplg_ops *ops;
>   };
>   
>   /* check we dont overflow the data for this control chunk */
> @@ -2334,7 +2334,7 @@ static int soc_tplg_load(struct soc_tplg *tplg)
>   
>   /* load audio component topology from "firmware" file */
>   int snd_soc_tplg_component_load(struct snd_soc_component *comp,
> -	struct snd_soc_tplg_ops *ops, const struct firmware *fw)
> +	const struct snd_soc_tplg_ops *ops, const struct firmware *fw)
>   {
>   	struct soc_tplg tplg;
>   	int ret;

Yes, makes sense to me.

Reviewed-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>

Re: [PATCH 1/6] ASoC: topology: Constify an argument of snd_soc_tplg_component_load()

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 458 bytes --]

On Mon, May 13, 2024 at 07:37:20PM +0200, Christophe JAILLET wrote:
> snd_soc_tplg_component_load() does not modify its "*ops" argument. It
> only read some values and stores it in "soc_tplg.ops".

As mentioned in submitting-patches.rst when submitting a patch series
you should supply a cover letter for that patch series which describes
the overall content of the series.  This helps people understand what
they are looking at and how things fit together.

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

Re: [PATCH 1/6] ASoC: topology: Constify an argument of snd_soc_tplg_component_load()

[Christophe JAILLET]

Le 14/05/2024 à 12:21, Mark Brown a écrit :
> On Mon, May 13, 2024 at 07:37:20PM +0200, Christophe JAILLET wrote:
>> snd_soc_tplg_component_load() does not modify its "*ops" argument. It
>> only read some values and stores it in "soc_tplg.ops".
> 
> As mentioned in submitting-patches.rst when submitting a patch series
> you should supply a cover letter for that patch series which describes
> the overall content of the series.  This helps people understand what
> they are looking at and how things fit together.

Ok.
I usually do, but I thought that the subjects were self-explanatory 
enough in this case.

Do you want me to resend?

CJ

Re: [PATCH 1/6] ASoC: topology: Constify an argument of snd_soc_tplg_component_load()

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 557 bytes --]

On Sat, May 18, 2024 at 10:34:33AM +0200, Christophe JAILLET wrote:
> Le 14/05/2024 à 12:21, Mark Brown a écrit :

> > As mentioned in submitting-patches.rst when submitting a patch series
> > you should supply a cover letter for that patch series which describes
> > the overall content of the series.  This helps people understand what
> > they are looking at and how things fit together.

> Ok.
> I usually do, but I thought that the subjects were self-explanatory enough
> in this case.

> Do you want me to resend?

It's fine this time.

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

Re: [PATCH 1/6] ASoC: topology: Constify an argument of snd_soc_tplg_component_load()

[Pierre-Louis Bossart]

On 5/20/24 08:24, Mark Brown wrote:
> On Sat, May 18, 2024 at 10:34:33AM +0200, Christophe JAILLET wrote:
>> Le 14/05/2024 à 12:21, Mark Brown a écrit :
> 
>>> As mentioned in submitting-patches.rst when submitting a patch series
>>> you should supply a cover letter for that patch series which describes
>>> the overall content of the series.  This helps people understand what
>>> they are looking at and how things fit together.
> 
>> Ok.
>> I usually do, but I thought that the subjects were self-explanatory enough
>> in this case.
> 
>> Do you want me to resend?
> 
> It's fine this time.

no issues with
https://github.com/thesofproject/linux/pull/4993, so

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