This patch adds error handling to the at91_shdwc_probe function
in drivers/power/reset/at91-sama5d2_shdwc.c. The function
previously did not handle the case where of_match_node could fail,
potentially leading to unexpected behavior if the device tree match
was unsuccessful.
Signed-off-by: Haoran Liu <[email protected]>
---
drivers/power/reset/at91-sama5d2_shdwc.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/drivers/power/reset/at91-sama5d2_shdwc.c b/drivers/power/reset/at91-sama5d2_shdwc.c
index e76b102b57b1..2ac566c83aec 100644
--- a/drivers/power/reset/at91-sama5d2_shdwc.c
+++ b/drivers/power/reset/at91-sama5d2_shdwc.c
@@ -353,6 +353,11 @@ static int __init at91_shdwc_probe(struct platform_device *pdev)
return PTR_ERR(at91_shdwc->shdwc_base);
match = of_match_node(at91_shdwc_of_match, pdev->dev.of_node);
+ if (!match) {
+ dev_err(&pdev->dev, "No matching device found\n");
+ return -ENODEV;
+ }
+
at91_shdwc->rcfg = match->data;
at91_shdwc->sclk = devm_clk_get(&pdev->dev, NULL);
--
2.17.1
On 29/11/2023 at 14:29, Haoran Liu wrote:
> This patch adds error handling to the at91_shdwc_probe function
> in drivers/power/reset/at91-sama5d2_shdwc.c. The function
As the platforms are DT only and that this driver was instantiated
precisely because a "match" existed, I don't think this error handling
is needed because it cannot happen.
> previously did not handle the case where of_match_node could fail,
> potentially leading to unexpected behavior if the device tree match
> was unsuccessful.
You didn't say how you reached this "unexpected behavior": is it a real
case (I'm interested in the boot log then) or found running a static
analysis tool?
For now, it's a NACK, sorry. Best regards,
Nicolas
> Signed-off-by: Haoran Liu <[email protected]>
> ---
> drivers/power/reset/at91-sama5d2_shdwc.c | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/drivers/power/reset/at91-sama5d2_shdwc.c b/drivers/power/reset/at91-sama5d2_shdwc.c
> index e76b102b57b1..2ac566c83aec 100644
> --- a/drivers/power/reset/at91-sama5d2_shdwc.c
> +++ b/drivers/power/reset/at91-sama5d2_shdwc.c
> @@ -353,6 +353,11 @@ static int __init at91_shdwc_probe(struct platform_device *pdev)
> return PTR_ERR(at91_shdwc->shdwc_base);
>
> match = of_match_node(at91_shdwc_of_match, pdev->dev.of_node);
> + if (!match) {
> + dev_err(&pdev->dev, "No matching device found\n");
> + return -ENODEV;
> + }
> +
> at91_shdwc->rcfg = match->data;
>
> at91_shdwc->sclk = devm_clk_get(&pdev->dev, NULL);
> --
> 2.17.1
>
On 29/11/2023 05:29:39-0800, Haoran Liu wrote:
> This patch adds error handling to the at91_shdwc_probe function
> in drivers/power/reset/at91-sama5d2_shdwc.c. The function
> previously did not handle the case where of_match_node could fail,
> potentially leading to unexpected behavior if the device tree match
> was unsuccessful.
>
> Signed-off-by: Haoran Liu <[email protected]>
> ---
> drivers/power/reset/at91-sama5d2_shdwc.c | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/drivers/power/reset/at91-sama5d2_shdwc.c b/drivers/power/reset/at91-sama5d2_shdwc.c
> index e76b102b57b1..2ac566c83aec 100644
> --- a/drivers/power/reset/at91-sama5d2_shdwc.c
> +++ b/drivers/power/reset/at91-sama5d2_shdwc.c
> @@ -353,6 +353,11 @@ static int __init at91_shdwc_probe(struct platform_device *pdev)
> return PTR_ERR(at91_shdwc->shdwc_base);
>
> match = of_match_node(at91_shdwc_of_match, pdev->dev.of_node);
> + if (!match) {
Can you elaborate how this will ever happen?
> + dev_err(&pdev->dev, "No matching device found\n");
> + return -ENODEV;
> + }
> +
> at91_shdwc->rcfg = match->data;
>
> at91_shdwc->sclk = devm_clk_get(&pdev->dev, NULL);
> --
> 2.17.1
>
--
Alexandre Belloni, co-owner and COO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com