Skip to contentSkip to conversation

Fix possible NULL Pointer Dereference in 'asoc_qcom_lpass_cpu_platform_probe'

PATCH In review Gax-c 5 participants 8 messages Sep 30, 2024 – Oct 1, 2024

Subsystems (5)
lore
Share & tools
lore

Gax-c

author
#0 lore raw

PATCH 1/1 Fix possible NULL Pointer Dereference in 'asoc_qcom_lpass_cpu_platform_probe'

A 'devm_kzalloc' in 'asoc_qcom_lpass_cpu_platform_probe' could possibly return NULL pointer. NULL Pointer Dereference may be triggerred in 'asoc_qcom_lpass_cpu_platform_probe' without addtional check. Add a null check for the returned pointer.

Fixes b5022a36d28f ("ASoC: qcom: lpass: Use regmap_field for i2sctl and dmactl registers")
---
 sound/soc/qcom/lpass-cpu.c | 3 +++
 1 file changed, 3 insertions(+)
sound/soc/qcom/lpass-cpu.c +3 -0
index 5a47f661e0c6..a8e56f47f237 100644
@@ -1243,6 +1243,9 @@ int asoc_qcom_lpass_cpu_platform_probe(struct platform_device *pdev)
 	drvdata->i2sctl = devm_kzalloc(&pdev->dev, sizeof(struct lpaif_i2sctl),
 					GFP_KERNEL);
 
+	if (!drvdata->i2sctl)
+		return -ENOMEM;
+
 	/* Initialize bitfields for dai I2SCTL register */
 	ret = lpass_cpu_init_i2sctl_bitfields(dev, drvdata->i2sctl,
 						drvdata->lpaif_map);
Signature · 1 line
2.25.1
#1 lore raw
Quoted 3 lines

A 'devm_kzalloc' in 'asoc_qcom_lpass_cpu_platform_probe' could possibly return NULL pointer. NULL Pointer Dereference may be triggerred in 'asoc_qcom_lpass_cpu_platform_probe' without addtional check. Add a null check for the returned pointer.

How do you think about a wording variant like the following?

The result from a call of the function “devm_kzalloc” was passed to a subsequent function call without checking for a null pointer before (according to a memory allocation failure).
Thus return directly after a failed devm_kzalloc() call.

…

Quoted 2 lines

…

How good does such a tag combination fit together for the same person?
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/submitting-patches.rst?h=v6.12-rc1#n525

Can a subject like “[PATCH] ASoC: qcom: lpass-cpu: Return directly after a failed devm_kzalloc() call in asoc_qcom_lpass_cpu_platform_probe()” be more appropriate?
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/submitting-patches.rst?h=v6.12-rc1#n613

…

sound/soc/qcom/lpass-cpu.c int asoc_qcom_lpass_cpu_platform_probe(struct platform_device *pdev) · line 1247

Quoted 7 lines

+++ b/sound/soc/qcom/lpass-cpu.c

sound/soc/qcom/lpass-cpu.c +2 -0
@@ -1243,6 +1243,9 @@ int asoc_qcom_lpass_cpu_platform_probe(struct platform_device *pdev)
 	drvdata->i2sctl = devm_kzalloc(&pdev->dev, sizeof(struct lpaif_i2sctl),
 					GFP_KERNEL);
 
+	if (!drvdata->i2sctl)
+		return -ENOMEM;

…

I suggest to omit a blank line here.

By the way:
Would you become interested to omit the label “err” from this function implementation finally?

Regards,
Markus

maintainer
#2 lore raw
Quoted 6 lines · Markus Elfring

How do you think about a wording variant like the following?

The result from a call of the function “devm_kzalloc” was passed to a subsequent function call without checking for a null pointer before (according to a memory allocation failure).
Thus return directly after a failed devm_kzalloc() call.

Feel free to ignore Markus, he has a long history of sending unhelpful review comments and continues to ignore repeated requests to stop.

#3 lore raw
Quoted 10 lines
Quoted 6 lines

How do you think about a wording variant like the following?

The result from a call of the function “devm_kzalloc” was passed to a subsequent function call without checking for a null pointer before (according to a memory allocation failure).
Thus return directly after a failed devm_kzalloc() call.

Feel free to ignore Markus, he has a long history of sending unhelpful review comments and continues to ignore repeated requests to stop.

30 non-merge commits were published with related subjects so far.
Can you get a more constructive view from parts of public software development history?
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/log/?h=v6.12-rc1&qt=grep&q=Return+directly+after+a+failed

Regards,
Markus

#4 lore raw
In reply to Gax-c
Quoted 4 lines · Gax-c

A 'devm_kzalloc' in 'asoc_qcom_lpass_cpu_platform_probe' could possibly return NULL pointer. NULL Pointer Dereference may be triggerred in 'asoc_qcom_lpass_cpu_platform_probe' without addtional check. Add a null check for the returned pointer.

Your description and patch looks good to me.

But please run git log on the changed file and add a prefix to your subject to match other changes to this file, and please wrap your commit message at 75 characters.

A good resource to read about this is:
https://www.kernel.org/doc/html/latest/process/submitting-patches.html#the-canonical-patch-format

Please also use the form devm_kzalloc() instead of 'devm_kzalloc' when referring to kernel functions.

It seems reasonable to mark this for backporting to stable, so I'd also suggest adding the following tag:

Quoted 14 lines
Fixes b5022a36d28f ("ASoC: qcom: lpass: Use regmap_field for i2sctl and dmactl registers")
---
 sound/soc/qcom/lpass-cpu.c | 3 +++
 1 file changed, 3 insertions(+)
sound/soc/qcom/lpass-cpu.c +0 -0
index 5a47f661e0c6..a8e56f47f237 100644

sound/soc/qcom/lpass-cpu.c int asoc_qcom_lpass_cpu_platform_probe(struct platform_device *pdev) · line 1245

Quoted 3 lines
sound/soc/qcom/lpass-cpu.c +0 -0
@@ -1243,6 +1243,9 @@ int asoc_qcom_lpass_cpu_platform_probe(struct platform_device *pdev)
 	drvdata->i2sctl = devm_kzalloc(&pdev->dev, sizeof(struct lpaif_i2sctl),
 					GFP_KERNEL);
 

Please drop this empty line.

Regards,
Bjorn

Quoted 13 lines
sound/soc/qcom/lpass-cpu.c +3 -0
+	if (!drvdata->i2sctl)
+		return -ENOMEM;
+
 	/* Initialize bitfields for dai I2SCTL register */
 	ret = lpass_cpu_init_i2sctl_bitfields(dev, drvdata->i2sctl,
 						drvdata->lpaif_map);
Signature · 1 line
2.25.1
#5 lore raw
Quoted 5 lines
Quoted 3 lines

A 'devm_kzalloc' in 'asoc_qcom_lpass_cpu_platform_probe' could possibly return NULL pointer. NULL Pointer Dereference may be triggerred in 'asoc_qcom_lpass_cpu_platform_probe' without addtional check. Add a null check for the returned pointer.

Your description and patch looks good to me.

Interesting “view” …

Quoted 3 lines

But please run git log on the changed file and add a prefix to your subject to match other changes to this file, and please wrap your commit message at 75 characters.

* How does your initial information fit to this patch review advice?

* Would you like to “tolerate” any typos (at the first glance)?

Regards,
Markus

maintainer
#6 lore raw
Quoted 3 lines · Markus Elfring
Quoted 1 line

Your description and patch looks good to me.

Interesting “view” …

Feel free to ignore Markus, he has a long history of sending unhelpful review comments and continues to ignore repeated requests to stop.

#7 lore raw
In reply to Gax-c
Quoted 3 lines · Gax-c

A 'devm_kzalloc' in 'asoc_qcom_lpass_cpu_platform_probe' could possibly return NULL pointer. NULL Pointer Dereference may be triggerred in 'asoc_qcom_lpass_cpu_platform_probe' without addtional check. Add a null check for the returned pointer.

Please wrap commit message according to Linux coding style / submission process (neither too early nor over the limit):
https://elixir.bootlin.com/linux/v6.4-rc1/source/Documentation/process/submitting-patches.rst#L597

Quoted 4 lines
Fixes b5022a36d28f ("ASoC: qcom: lpass: Use regmap_field for i2sctl and dmactl registers")

Drop the unnecessary tag. You are the author.

Quoted 2 lines

Why two people reported it? And where?

Quoted 3 lines
---
 sound/soc/qcom/lpass-cpu.c | 3 +++
 1 file changed, 3 insertions(+)

Best regards,
Krzysztof