Re: [PATCH 2/8] staging: vchiq_core: Properly log dev_err()

[Stefan Wahren]

Hi Umang,

[add Raspberry Pi guys to the loop]

Am 11.10.24 um 09:22 schrieb Umang Jain:
> Properly log a dev_err() message when the msgid is not of
> VCHIQ_MSG_PADDING type. Drop 'oldmsgid' scoped variable and improve
> on the error string as well.
>
> Signed-off-by: Umang Jain <umang.jain@xxxxxxxxxxxxxxxx>
> ---
>   .../vc04_services/interface/vchiq_arm/vchiq_core.c    | 11 ++++-------
>   1 file changed, 4 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_core.c b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_core.c
> index e9b60dd8d419..1dca676186b6 100644
> --- a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_core.c
> +++ b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_core.c
> @@ -1188,13 +1188,10 @@ queue_message_sync(struct vchiq_state *state, struct vchiq_service *service,
>   	header = (struct vchiq_header *)SLOT_DATA_FROM_INDEX(state,
>   		local->slot_sync);
>
> -	{
> -		int oldmsgid = header->msgid;
> -
> -		if (oldmsgid != VCHIQ_MSGID_PADDING)
> -			dev_err(state->dev, "core: %d: qms - msgid %x, not PADDING\n",
> -				state->id, oldmsgid);
> -	}
> +	if (header->msgid != VCHIQ_MSGID_PADDING)
> +		dev_err(state->dev,
> +			"core: %d: qms - msgid %x, is not a PADDING message\n",
> +			state->id, header->msgid);
I'm fine with this change, but the behavior looks strange to me.

Either this is a real error, I would expect the function would return
with something like EINVAL here or this should be a warning?

Sorry, no idea what's correct here.

Best regards
>   	dev_dbg(state->dev, "sync: %d: qms %s@%pK,%x (%d->%d)\n",
>   		state->id, msg_type_str(VCHIQ_MSG_TYPE(msgid)), header, size,