Re: [PATCH v5 4/4] media: chips-media: wave5: Improve performance of decoder

[Brandon Brnich]

Hi Nicolas and Jackson,

On 9/24/2025 8:20 AM, Nicolas Dufresne wrote:
> Hi Jackson,
>
> Le mercredi 24 septembre 2025 à 01:14 +0000, jackson.lee a écrit :
> > Hi Nicolas
> >
> > > -----Original Message-----
> > > From: Nicolas Dufresne <nicolas.dufresne@xxxxxxxxxxxxx>
> > > Sent: Tuesday, September 23, 2025 3:00 AM
> > > To: Brandon Brnich <b-brnich@xxxxxx>; jackson.lee
> > > <jackson.lee@xxxxxxxxxxxxxxx>; mchehab@xxxxxxxxxx; hverkuil-
> > > cisco@xxxxxxxxx; bob.beckett@xxxxxxxxxxxxx
> > > Cc: linux-media@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; lafley.kim
> > > <lafley.kim@xxxxxxxxxxxxxxx>; hverkuil@xxxxxxxxx; Nas Chung
> > > <nas.chung@xxxxxxxxxxxxxxx>
> > > Subject: Re: [PATCH v5 4/4] media: chips-media: wave5: Improve performance
> > > of decoder
> > >
> > > Hi Brandon,
> > >
> > > Le lundi 22 septembre 2025 à 12:32 -0500, Brandon Brnich a écrit :
> > > > > -	/*
> > > > > -	 * During a resolution change and while draining, the firmware
> > > > > may
> > > > > flush
> > > > > -	 * the reorder queue regardless of having a matching decoding
> > > > > operation
> > > > > -	 * pending. Only terminate the job if there are no more IRQ
> > > > > coming.
> > > > > -	 */
> > > > > -	wave5_vpu_dec_give_command(inst, DEC_GET_QUEUE_STATUS,
> > > > > &q_status);
> > > > > -	if (q_status.report_queue_count == 0 &&
> > > > > -	    (q_status.instance_queue_count == 0 ||
> > > > > dec_info.sequence_changed)) {
> > > > > -		dev_dbg(inst->dev->dev, "%s: finishing job.\n",
> > > > > __func__);
> > > > > -		pm_runtime_mark_last_busy(inst->dev->dev);
> > > >
> > > > Patch is failing to apply here to linux-next because these redundant
> > > > calls have already been removed[0].
> > >
> > > Which have not been merged back from the RC into media-committers/next,
> > > forcing to skip a cycle. Jackson, feel free to rebase on linux-next like
> > > Brandon suggest.
> >
> > Then should I make v6 patch series based on Linux-next ?
>
> I've asked advises from the other maintainers, and the answer is no. Basing it
> on our next branch for linux-media submission is the correct thing to do. Its
> too late for this cycle, but be reassured we will improve our process in future
> iterations to reduce the risk of this happening.
>
> Feel free to send a rebased patch to Brandon, having more testing is always
> good.

Sorry for the confusion here, I was unaware that the patches weren't 
present in -next on linux-media tree. Typically I just do all my testing 
on the linux-next branch. I will make sure to use linux-media in the future.

No need to share the patch with me, I already rebased it to build and 
start my own testing.

Thanks,
Brandon

> cheers,
> Nicolas
>
> > Thanks
> > Jackson
> >
> > > regards,
> >
> > > Nicolas