mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Nicolas Dufresne <nicolas.dufresne@collabora.com>
To: Benjamin Gaignard <benjamin.gaignard@collabora.com>,
	 p.zabel@pengutronix.de, mchehab@kernel.org
Cc: linux-media@vger.kernel.org, linux-rockchip@lists.infradead.org,
	 linux-kernel@vger.kernel.org
Subject: Re: [PATCH] media: verisilicon: HEVC: Initialize start_bit field
Date: Mon, 20 Jan 2025 09:10:15 -0500	[thread overview]
Message-ID: <95a8e3c6d28444750bd312ff61a3f6b0daea177a.camel@collabora.com> (raw)
In-Reply-To: <cc3b22c9-ff98-4328-9e70-4d22c29a6b0d@collabora.com>

Le lundi 20 janvier 2025 à 11:00 +0100, Benjamin Gaignard a écrit :
> Le 20/01/2025 à 09:10, Benjamin Gaignard a écrit :
> > Always set start_bit field to 0, if not it could lead to corrupted frames
> > specially when decoding VP9 bitstreams at the same time since VP9 driver
> > set it for it own purpose.

               its

> > 
> > Signed-off-by: Benjamin Gaignard <benjamin.gaignard@collabora.com>
> 
> I could add this tag:
> 
> Fixes: cb5dd5a0fa518 ("media: hantro: Introduce G2/HEVC decoder")

I have tested this on IMX8MQ board using the following GStreamer pipeline.
Before the change, the HEVC window was entirely corrupted. The streams don't
matter as long as they use both HEVC and VP9 codec.

gst-launch-1.0 \
  filesrc location=hevc.mp4 ! parsebin ! v4l2slh265dec ! fakevideosink \
  filesrc location=vp9.mkv ! parsebin ! v4l2slvp9dec ! fakevideosink

Tested-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>


> 
> > ---
> >   drivers/media/platform/verisilicon/hantro_g2_hevc_dec.c | 1 +
> >   1 file changed, 1 insertion(+)
> > 
> > diff --git a/drivers/media/platform/verisilicon/hantro_g2_hevc_dec.c b/drivers/media/platform/verisilicon/hantro_g2_hevc_dec.c
> > index 85a44143b378..0e212198dd65 100644
> > --- a/drivers/media/platform/verisilicon/hantro_g2_hevc_dec.c
> > +++ b/drivers/media/platform/verisilicon/hantro_g2_hevc_dec.c
> > @@ -518,6 +518,7 @@ static void set_buffers(struct hantro_ctx *ctx)
> >   	hantro_reg_write(vpu, &g2_stream_len, src_len);
> >   	hantro_reg_write(vpu, &g2_strm_buffer_len, src_buf_len);
> >   	hantro_reg_write(vpu, &g2_strm_start_offset, 0);
> > +	hantro_reg_write(vpu, &g2_start_bit, 0);
> >   	hantro_reg_write(vpu, &g2_write_mvs_e, 1);

I've also crossed against "decoder_swreg_map_g2.xlsx" documentation, if you are
lucky to have access to that, and within swreg5 there is only g2_tempor_mvp_e
that is also shared, and its already being set in both drivers.

Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>


> >   
> >   	hantro_write_addr(vpu, G2_TILE_SIZES_ADDR, ctx->hevc_dec.tile_sizes.dma);
> 


  reply	other threads:[~2025-01-20 14:10 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-01-20  8:10 Benjamin Gaignard
2025-01-20 10:00 ` Benjamin Gaignard
2025-01-20 14:10   ` Nicolas Dufresne [this message]
2025-01-20 14:15     ` Adam Ford
2025-01-20 14:19       ` Benjamin Gaignard
2025-01-20 14:43       ` Nicolas Dufresne

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=95a8e3c6d28444750bd312ff61a3f6b0daea177a.camel@collabora.com \
    --to=nicolas.dufresne@collabora.com \
    --cc=benjamin.gaignard@collabora.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-rockchip@lists.infradead.org \
    --cc=mchehab@kernel.org \
    --cc=p.zabel@pengutronix.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®