Re: [PATCH rdma-rc 4/4] RDMA/bnxt_re: Fix null pointer dereference in bnxt_re_suspend

[Leon Romanovsky <leon@kernel.org>]

On Tue, May 20, 2025 at 09:29:10AM +0530, Kalesh AP wrote:
> There is a possibility that the Ethernet driver invokes bnxt_ulp_stop()
> twice during reset recovery scenario. This will result in a NULL pointer
> dereference in bnxt_re_suspend() as en_info->rdev has been set to NULL
> during the first invocation.

Please fix bnxt_ulp_stop() to do not do that. The fix should be there
and not in bnxt_re.

Thanks

> 
> Fixes: dee3da3422d5 ("RDMA/bnxt_re: Change aux driver data to en_info to hold more information")
> Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
> ---
>  drivers/infiniband/hw/bnxt_re/main.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/infiniband/hw/bnxt_re/main.c b/drivers/infiniband/hw/bnxt_re/main.c
> index 293b0a96c8e3..bc0b40073461 100644
> --- a/drivers/infiniband/hw/bnxt_re/main.c
> +++ b/drivers/infiniband/hw/bnxt_re/main.c
> @@ -2410,6 +2410,8 @@ static int bnxt_re_suspend(struct auxiliary_device *adev, pm_message_t state)
>  	struct bnxt_re_dev *rdev;
>  
>  	rdev = en_info->rdev;
> +	if (!rdev)
> +		return 0;
>  	en_dev = en_info->en_dev;
>  	mutex_lock(&bnxt_re_mutex);
>  
> -- 
> 2.43.5
> 
>