Hi Nicolas,
On Wed, Sep 03, 2025 at 01:13:00PM -0400, Nicolas Dufresne wrote:
Le mardi 29 juillet 2025 à 17:52 -0700, Nathan Chancellor a écrit :
Hi Nicolas,
On Tue, Jul 29, 2025 at 10:24:22AM -0400, Nicolas Dufresne wrote:
Le mardi 15 juillet 2025 à 15:13 -0700, Nathan Chancellor a écrit :
From what I can tell, it seems like ->cmd_host2risc() is only ever called from v6 code, which always passes NULL? It seems like it should be possible to just drop .cmd_host2risc on the v5 side, then update .cmd_host2risc to only take two parameters? If so, I can send a follow up as a clean up, so that this can go back relatively conflict free.
It seems so yes. For this specific patch, I would probably rename "args" to "__unused" to make the reading faster. But does not matter so much if you later remove it.
Yes, after this change is picked up in a maintainer's tree, I do plan to send a patch to remove the "args" parameter altogether. If you really care, I can certainly rename the parameter in this change to "__unused" as suggested but if you don't, I will just leave it as is to make backporting this a little easier.
Reviewed-by: Nicolas Dufresne nicolas.dufresne@collabora.com
Thanks a lot for taking a look!
While applying your patch, I realized the Hans merged another version of this fix, but made by Arnd. It covers the remaining too, so I will mark yours as superseded now.
https://gitlab.freedesktop.org/linux-media/media-committers/-/commit/7fa37ba...
Thanks, Arnd's patch is obviously equally correct and the stable tags remained so everything looks good from my end. Appreciate you doubling back to this.
Cheers, Nathan