From: Stanimir Varbanov <stanimir.varbanov@linaro.org>
To: Philipp Zabel <p.zabel@pengutronix.de>, linux-media@vger.kernel.org
Cc: Mauro Carvalho Chehab <mchehab@kernel.org>,
Hans Verkuil <hans.verkuil@cisco.com>,
Stanimir Varbanov <stanimir.varbanov@linaro.org>,
Kyungmin Park <kyungmin.park@samsung.com>,
Kamil Debski <kamil@wypas.org>,
Jeongtae Park <jtp.park@samsung.com>,
Andrzej Hajda <a.hajda@samsung.com>
Subject: Re: [PATCH 0/2] Clarify H.264 loop filter offset controls and fix them for coda
Date: Wed, 28 Nov 2018 18:25:40 +0200 [thread overview]
Message-ID: <d6e60d32-d4eb-1d51-1f08-2f3547bb9524@linaro.org> (raw)
In-Reply-To: <20181128130122.4916-1-p.zabel@pengutronix.de>
Hi Philipp,
On 11/28/18 3:01 PM, Philipp Zabel wrote:
> Hi,
>
> the coda driver handles the H.264 loop filter alpha/beta offset controls
> incorrectly. When trying to fix them, I noticed that the documentation
> is not clear about what these values actually are.
>
> From the value range of -6 to +6 used in the existing drivers (s5p-mfc,
> venus), it looks like they currently correspond directly to the values
> stored into the slice headers: slice_alpha_c0_offset_div2 and
> slice_beta_offset_div2. These are only half of the actual alpha/beta
> filter offsets.
>
> The ITU-T Rec. H.264 (02/2016) states:
>
> slice_alpha_c0_offset_div2 specifies the offset used in accessing the
> α [...] deblocking filter tables for filtering operations controlled
> by the macroblocks within the slice. From this value, the offset that
> shall be applied when addressing these tables shall be computed as
>
> FilterOffsetA = slice_alpha_c0_offset_div2 << 1 (7-32)
>
> The value of slice_alpha_c0_offset_div2 shall be in the range of −6 to
> +6, inclusive. When slice_alpha_c0_offset_div2 is not present in the
> slice header, the value of slice_alpha_c0_offset_div2 shall be inferred
> to be equal to 0.
>
> And the same for slice_beta_offset_div2 / FilterOffsetB.
>
> Do the s5p-mfc and venus drivers use the controls
> V4L2_MPEG_VIDEO_H264_LOOP_FILTER_ALPHA and _BETA directly as slice
> header fields, and thus their values are to be interpreted as half of
> FilterOffsetA/B defined in the H.264 spec, respectively?
That is correct for Venus encoder, it uses slice header fields directly
[-6, 6].
--
regards,
Stan
prev parent reply other threads:[~2018-11-29 3:27 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-11-28 13:01 [PATCH 0/2] Clarify H.264 loop filter offset controls and fix them for coda Philipp Zabel
2018-11-28 13:01 ` [PATCH 1/2] media: v4l2: clarify H.264 loop filter offset controls Philipp Zabel
2018-11-28 13:01 ` [PATCH 2/2] media: coda: fix H.264 deblocking filter controls Philipp Zabel
2018-11-28 16:25 ` Stanimir Varbanov [this message]
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=d6e60d32-d4eb-1d51-1f08-2f3547bb9524@linaro.org \
--to=stanimir.varbanov@linaro.org \
--cc=a.hajda@samsung.com \
--cc=hans.verkuil@cisco.com \
--cc=jtp.park@samsung.com \
--cc=kamil@wypas.org \
--cc=kyungmin.park@samsung.com \
--cc=linux-media@vger.kernel.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;
as well as URLs for NNTP newsgroup(s).