On Tue, Jun 21, 2022 at 03:28:40PM +0200, Jean Delvare wrote:
> Unlike release_mem_region(), a call to release_resource() does not
> free the resource, so it has to be freed explicitly to avoid a memory
> leak.
>
> Signed-off-by: Jean Delvare <jdelvare@xxxxxxx>
> Fixes: 0578fff4aae5 ("Watchdog: sp5100_tco: Add initialization using EFCH MMIO")
> Cc: Terry Bowman <terry.bowman@xxxxxxx>
> Cc: Wim Van Sebroeck <wim@xxxxxxxxxxxxxxxxxx>
> Cc: Guenter Roeck <linux@xxxxxxxxxxxx>
Reviewed-by: Guenter Roeck <linux@xxxxxxxxxxxx>
Unless I am missing something, it looks like every caller of
request_mem_region_muxed() has that problem.
Guenter
> ---
> drivers/watchdog/sp5100_tco.c | 1 +
> 1 file changed, 1 insertion(+)
>
> --- linux-5.18.orig/drivers/watchdog/sp5100_tco.c 2022-05-22 21:52:31.000000000 +0200
> +++ linux-5.18/drivers/watchdog/sp5100_tco.c 2022-06-21 15:07:31.868814828 +0200
> @@ -402,6 +402,7 @@ static int sp5100_tco_setupdevice_mmio(s
> iounmap(addr);
>
> release_resource(res);
> + kfree(res);
>
> return ret;
> }
>
>
> --
> Jean Delvare
> SUSE L3 Support
