[PATCH] wifi: brcmfmac: Add null pointer check to brcmf_acpi_probe

[PATCH] wifi: brcmfmac: Add null pointer check to brcmf_acpi_probe

[Li Qiang]

devm_kasprintf() returns a pointer to dynamically allocated memory
which can be NULL upon failure.

Fixes: 0f485805d008a ("wifi: brcmfmac: acpi: Add support for fetching Apple ACPI properties")
Signed-off-by: Li Qiang <liqiang01@kylinos.cn>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c
index c4a54861bfb4..f8f6433bdf69 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c
@@ -25,6 +25,8 @@ void brcmf_acpi_probe(struct device *dev, enum brcmf_bus_type bus_type,
 		settings->board_type = devm_kasprintf(dev, GFP_KERNEL,
 						      "apple,%s",
 						      o->string.pointer);
+		if (!settings->board_type)
+			return;
 	} else {
 		brcmf_dbg(INFO, "No ACPI module-instance\n");
 		return;
-- 
2.25.1

Re: [PATCH] wifi: brcmfmac: Add null pointer check to brcmf_acpi_probe

[Markus Elfring]

> devm_kasprintf() returns a pointer to dynamically allocated memory
> which can be NULL upon failure.

Will another imperative wording approach become more helpful for an improved
change description?
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/submitting-patches.rst?h=v6.17#n94


Can a summary phrase like “Prevent null pointer dereference in brcmf_acpi_probe()”
be nicer?

Regards,
Markus

Re: [PATCH] wifi: brcmfmac: Add null pointer check to brcmf_acpi_probe

[Arend van Spriel]

On 10/15/2025 9:16 AM, Li Qiang wrote:
> devm_kasprintf() returns a pointer to dynamically allocated memory
> which can be NULL upon failure.

Thanks. However, it does not make sense to just bail out. 
settings->board_type will still be NULL and apparently that is not an 
issue. It is used to fill the firmware request (see struct 
brcmf_fw_request [1]) which is done for PCIe devices in 
brcmf_pcie_prepare_fw_request() [2].

So struct brcmf_fw_request::board_types array may hold NULL pointers 
which is not an issue.

So this patch does not solve a problem.

Regards,
Arend

[1] 
https://elixir.bootlin.com/linux/v6.17.5/source/drivers/net/wireless/broadcom/brcm80211/brcmfmac/firmware.h#L67
[2] 
https://elixir.bootlin.com/linux/v6.17.5/source/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c#L2264

> Fixes: 0f485805d008a ("wifi: brcmfmac: acpi: Add support for fetching Apple ACPI properties")
> Signed-off-by: Li Qiang <liqiang01@kylinos.cn>
> ---
>   drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c | 2 ++
>   1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c
> index c4a54861bfb4..f8f6433bdf69 100644
> --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c
> +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/acpi.c
> @@ -25,6 +25,8 @@ void brcmf_acpi_probe(struct device *dev, enum brcmf_bus_type bus_type,
>   		settings->board_type = devm_kasprintf(dev, GFP_KERNEL,
>   						      "apple,%s",
>   						      o->string.pointer);
> +		if (!settings->board_type)
> +			return;
>   	} else {
>   		brcmf_dbg(INFO, "No ACPI module-instance\n");
>   		return;