Skip to contentSkip to conversation

ASoC: two trivial fixes

PATCH In review Daniel Mack 4 participants 6 messages Oct 19, 2014 – Oct 22, 2014

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

cover ASoC: two trivial fixes

Just two random fixes spotted by Coverity.

Daniel Mack (2):
ASoC: soc-compress: consolidate two identical branches
ASoC: fsl: use strncpy() to prevent copying of over-long names

 sound/soc/fsl/fsl_asrc.c |  2 +-
 sound/soc/fsl/fsl_esai.c |  2 +-
 sound/soc/soc-compress.c | 11 ++---------
 3 files changed, 4 insertions(+), 11 deletions(-)
Signature · 1 line
2.1.0
#1 lore raw

PATCH 1/2 ASoC: soc-compress: consolidate two identical branches

The actions taken in both branches are identical, so we can simplify the code. Spotted by Coverity.

---
 sound/soc/soc-compress.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)
sound/soc/soc-compress.c +2 -9
index cecfab3..590a82f 100644
@@ -258,10 +258,7 @@ static int soc_compr_free_fe(struct snd_compr_stream *cstream)
 	list_for_each_entry(dpcm, &fe->dpcm[stream].be_clients, list_be)
 		dpcm->state = SND_SOC_DPCM_LINK_STATE_FREE;
 
-	if (stream == SNDRV_PCM_STREAM_PLAYBACK)
-		dpcm_dapm_stream_event(fe, stream, SND_SOC_DAPM_STREAM_STOP);
-	else
-		dpcm_dapm_stream_event(fe, stream, SND_SOC_DAPM_STREAM_STOP);
+	dpcm_dapm_stream_event(fe, stream, SND_SOC_DAPM_STREAM_STOP);
 
 	fe->dpcm[stream].state = SND_SOC_DPCM_STATE_CLOSE;
 	fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_NO;
@@ -456,11 +453,7 @@ static int soc_compr_set_params_fe(struct snd_compr_stream *cstream,
 	if (ret < 0)
 		goto out;
 
-	if (stream == SNDRV_PCM_STREAM_PLAYBACK)
-		dpcm_dapm_stream_event(fe, stream, SND_SOC_DAPM_STREAM_START);
-	else
-		dpcm_dapm_stream_event(fe, stream, SND_SOC_DAPM_STREAM_START);
-
+	dpcm_dapm_stream_event(fe, stream, SND_SOC_DAPM_STREAM_START);
 	fe->dpcm[stream].state = SND_SOC_DPCM_STATE_PREPARE;
 
 out:
Signature · 1 line
2.1.0
#2 lore raw

PATCH 2/2 ASoC: fsl: use strncpy() to prevent copying of over-long names

In reply to Daniel Mack

Use strncpy() instead of strcpy(). That's not a security issue, as the source buffer is taken from DT nodes, but we should still enforce bound checks. Spotted by Coverity.

---
 sound/soc/fsl/fsl_asrc.c | 2 +-
 sound/soc/fsl/fsl_esai.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
sound/soc/fsl/fsl_asrc.c +1 -1
index 8221104..dd00b9d 100644
@@ -792,7 +792,7 @@ static int fsl_asrc_probe(struct platform_device *pdev)
 		return -ENOMEM;
 
 	asrc_priv->pdev = pdev;
-	strcpy(asrc_priv->name, np->name);
+	strncpy(asrc_priv->name, np->name, sizeof(asrc_priv->name) - 1);
 
 	/* Get the addresses and IRQ */
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
sound/soc/fsl/fsl_esai.c +1 -1
index a3b29ed..fda1d46 100644
@@ -729,7 +729,7 @@ static int fsl_esai_probe(struct platform_device *pdev)
 		return -ENOMEM;
 
 	esai_priv->pdev = pdev;
-	strcpy(esai_priv->name, np->name);
+	strncpy(esai_priv->name, np->name, sizeof(esai_priv->name) - 1);
 
 	if (of_property_read_bool(np, "big-endian"))
 		fsl_esai_regmap_config.val_format_endian = REGMAP_ENDIAN_BIG;
Signature · 1 line
2.1.0
#3 lore raw

Re: [PATCH 2/2] ASoC: fsl: use strncpy() to prevent copying of over-long names

This hardly improves matters. When the source string is larger than the buffer, the destination may not be nul-terminated. Also, strncpy ALWAYS writes the full buffer so it may be wasting cycles when the destination buffer is large.

I'm sure the kernel offers a better alternative. Even "snprintf" is a better alternative.

Mike.

Quoted 29 lines · Daniel Mack

Use strncpy() instead of strcpy(). That's not a security issue, as the source buffer is taken from DT nodes, but we should still enforce bound checks. Spotted by Coverity.

---
  sound/soc/fsl/fsl_asrc.c | 2 +-
  sound/soc/fsl/fsl_esai.c | 2 +-
  2 files changed, 2 insertions(+), 2 deletions(-)
sound/soc/fsl/fsl_asrc.c +1 -1
index 8221104..dd00b9d 100644
@@ -792,7 +792,7 @@ static int fsl_asrc_probe(struct platform_device *pdev)
  		return -ENOMEM;
 
  	asrc_priv->pdev = pdev;
-	strcpy(asrc_priv->name, np->name);
+	strncpy(asrc_priv->name, np->name, sizeof(asrc_priv->name) - 1);
 
  	/* Get the addresses and IRQ */
  	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
sound/soc/fsl/fsl_esai.c +0 -0
index a3b29ed..fda1d46 100644

sound/soc/fsl/fsl_esai.c static int fsl_esai_probe(struct platform_device *pdev) · line 735

Quoted 8 lines
sound/soc/fsl/fsl_esai.c +1 -1
@@ -729,7 +729,7 @@ static int fsl_esai_probe(struct platform_device *pdev)
  		return -ENOMEM;
 
  	esai_priv->pdev = pdev;
-	strcpy(esai_priv->name, np->name);
+	strncpy(esai_priv->name, np->name, sizeof(esai_priv->name) - 1);
 
  	if (of_property_read_bool(np, "big-endian"))
  		fsl_esai_regmap_config.val_format_endian = REGMAP_ENDIAN_BIG;

Met vriendelijke groet / kind regards,

Mike Looijmans

TOPIC Embedded Systems Eindhovenseweg 32-C, NL-5683 KH Best Postbus 440, NL-5680 AK Best
Telefoon: (+31) (0) 499 33 69 79
Telefax: (+31) (0) 499 33 69 70
E-mail: [email protected]
Website: www.topic.nl

Please consider the environment before printing this e-mail

Topic zoekt gedreven (embedded) software specialisten!
http://topic.nl/vacatures/topic-zoekt-software-engineers/

_______________________________________________
Alsa-devel mailing list [email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

#4 lore raw

Re: [PATCH 2/2] ASoC: fsl: use strncpy() to prevent copying of over-long names

At Mon, 20 Oct 2014 07:55:22 +0200,

Quoted 5 lines · Mike Looijmans

This hardly improves matters. When the source string is larger than the buffer, the destination may not be nul-terminated.
Also, strncpy ALWAYS writes the full buffer so it may be wasting cycles when the destination buffer is large.

Indeed, strncpy() should be avoided.

Quoted 2 lines

I'm sure the kernel offers a better alternative. Even "snprintf" is a better alternative.

strlcpy() is the best choice.

Takashi

Quoted 59 lines

Mike.

Quoted 29 lines · Daniel Mack

Use strncpy() instead of strcpy(). That's not a security issue, as the source buffer is taken from DT nodes, but we should still enforce bound checks. Spotted by Coverity.

---
  sound/soc/fsl/fsl_asrc.c | 2 +-
  sound/soc/fsl/fsl_esai.c | 2 +-
  2 files changed, 2 insertions(+), 2 deletions(-)
sound/soc/fsl/fsl_asrc.c +1 -1
index 8221104..dd00b9d 100644
@@ -792,7 +792,7 @@ static int fsl_asrc_probe(struct platform_device *pdev)
  		return -ENOMEM;
 
  	asrc_priv->pdev = pdev;
-	strcpy(asrc_priv->name, np->name);
+	strncpy(asrc_priv->name, np->name, sizeof(asrc_priv->name) - 1);
 
  	/* Get the addresses and IRQ */
  	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
sound/soc/fsl/fsl_esai.c +0 -0
index a3b29ed..fda1d46 100644

sound/soc/fsl/fsl_esai.c static int fsl_esai_probe(struct platform_device *pdev) · line 735

Quoted 8 lines
sound/soc/fsl/fsl_esai.c +1 -1
@@ -729,7 +729,7 @@ static int fsl_esai_probe(struct platform_device *pdev)
  		return -ENOMEM;
 
  	esai_priv->pdev = pdev;
-	strcpy(esai_priv->name, np->name);
+	strncpy(esai_priv->name, np->name, sizeof(esai_priv->name) - 1);
 
  	if (of_property_read_bool(np, "big-endian"))
  		fsl_esai_regmap_config.val_format_endian = REGMAP_ENDIAN_BIG;

Met vriendelijke groet / kind regards,

Mike Looijmans

TOPIC Embedded Systems
Eindhovenseweg 32-C, NL-5683 KH Best
Postbus 440, NL-5680 AK Best
Telefoon: (+31) (0) 499 33 69 79
Telefax: (+31) (0) 499 33 69 70
E-mail: [email protected]
Website: www.topic.nl

Please consider the environment before printing this e-mail

Quoted 7 lines

Topic zoekt gedreven (embedded) software specialisten!
http://topic.nl/vacatures/topic-zoekt-software-engineers/

_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

maintainer
#5 lore raw
In reply to Daniel Mack
Quoted 1 line · Daniel Mack

Just two random fixes spotted by Coverity.

Applied both. I'd been holding off on these waiting for the developers of the relevant code to review but I now notice that you didn't CC them which is probably why they didn't respond!