On Sat, Jan 11, 2025 at 06:34:21AM +0800, Zijun Hu wrote:

On 2025/1/11 04:26, Rob Herring wrote:

quoted

On Thu, Jan 09, 2025 at 09:26:57PM +0800, Zijun Hu wrote:

quoted

From: Zijun Hu <redacted>

parse_interrupt_map() will use uninitialized variable @imaplen if fails
to get property 'interrupt-map'.

Fix by using the variable after successfully getting the property.

Fixes: e7985f43609c ("of: property: Fix fw_devlink handling of interrupt-map")
Signed-off-by: Zijun Hu <redacted>
---
 drivers/of/property.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/of/property.c b/drivers/of/property.c
index dca1a3ebccae1093b2b11f51e8e692bca854d0e3..6245cbff3527d762c16e7f4b7b7b3d4f2e9ddbe6 100644
--- a/drivers/of/property.c
+++ b/drivers/of/property.c

@@ -1391,9 +1391,9 @@ static struct device_node *parse_interrupt_map(struct device_node *np,
 	addrcells = of_bus_n_addr_cells(np);
 
 	imap = of_get_property(np, "interrupt-map", &imaplen);
-	imaplen /= sizeof(*imap);
 	if (!imap)
 		return NULL;
+	imaplen /= sizeof(*imap);

sizeof() is a compile time constant, there's not an actual dereference 
here.

the uninitialized variable is @imaplen, and not sizeof(*imap).

Correct. I think error is harmless, because whatever stack/random value
of imaplen we use in 'imaplen / =sizeof', we immediately return.

Anyway, for code correctness and silencing whatever warnings there are
(if there are you should actually paste them in commit msg):

Reviewed-by: Krzysztof Kozlowski <krzk@kernel.org>

Best regards,
Krzysztof