[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <075169b4-3c6a-2379-0dc9-34b89fb83ef1@collabora.com>
Date: Mon, 16 Jan 2023 11:00:02 +0100
From: Andrzej Pietrasiewicz <andrzej.p@...labora.com>
To: Ezequiel Garcia <ezequiel@...guardiasur.com.ar>,
Sebastian Fricke <sebastian.fricke@...labora.com>
Cc: Mauro Carvalho Chehab <mchehab@...nel.org>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
linux-media@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-rockchip@...ts.infradead.org, linux-staging@...ts.linux.dev,
Jonas Karlman <jonas@...boo.se>,
Alex Bee <knaerzche@...il.com>,
Nicolas Dufresne <nicolas.dufresne@...labora.com>,
Collabora Kernel-domain <kernel@...labora.com>,
Robert Beckett <bob.beckett@...labora.com>,
Laurent Pinchart <laurent.pinchart@...asonboard.com>,
Benjamin Gaignard <benjamin.gaignard@...labora.com>
Subject: Re: [PATCH v2 05/12] staging: media: rkvdec: Add SPS structure to
internal context
Hi Sebastian,
W dniu 12.01.2023 o 16:02, Ezequiel Garcia pisze:
> Hi Sebastian,
>
> On Thu, Jan 12, 2023 at 9:56 AM Sebastian Fricke
> <sebastian.fricke@...labora.com> wrote:
>>
>> Prepare storing the SPS structure for HEVC & H264 in the internal
>> context of the rkvdec instance. This structure is used to figure out
>> which capture queue format is appropriate for decoding.
>>
>> Signed-off-by: Sebastian Fricke <sebastian.fricke@...labora.com>
>> ---
>> drivers/staging/media/rkvdec/rkvdec.h | 1 +
>> 1 file changed, 1 insertion(+)
>>
>> diff --git a/drivers/staging/media/rkvdec/rkvdec.h b/drivers/staging/media/rkvdec/rkvdec.h
>> index 633335ebb9c4..332126e7b812 100644
>> --- a/drivers/staging/media/rkvdec/rkvdec.h
>> +++ b/drivers/staging/media/rkvdec/rkvdec.h
>> @@ -105,6 +105,7 @@ struct rkvdec_ctx {
>> struct v4l2_ctrl_handler ctrl_hdl;
>> struct rkvdec_dev *dev;
>> void *priv;
>> + void *sps;
>
> I don't really like re-caching the SPS in the context,
> since all the controls are already stored in the context,
> via the ctrl_handler.
>
> See hantro_get_ctrl().
>
> Duplicating state can lead to problems and even if we get it
> right this time, will be hard to maintain.
>
Also, just looking at this patch is seems a bit too fine-grained
a change to add a member to a driver-internal struct and have no
users at the same time. OTOH Ezequiel has easily spotted it :D
Andrzej
> Thanks,
> Ezequiel
>
>> };
>>
>> static inline struct rkvdec_ctx *fh_to_rkvdec_ctx(struct v4l2_fh *fh)
>>
>> --
>> 2.25.1
Powered by blists - more mailing lists