From: Benjamin Gaignard <benjamin.gaignard@collabora.com>
To: mchehab@kernel.org, tfiga@chromium.org, m.szyprowski@samsung.com,
ezequiel@vanguardiasur.com.ar, p.zabel@pengutronix.de,
hverkuil-cisco@xs4all.nl, nicolas@ndufresne.ca
Cc: linux-media@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-rockchip@lists.infradead.org, kernel@collabora.com,
Benjamin Gaignard <benjamin.gaignard@collabora.com>
Subject: [PATCH v21 0/9] Add REMOVE_BUF ioctl
Date: Thu, 14 Mar 2024 16:32:17 +0100 [thread overview]
Message-ID: <20240314153226.197445-1-benjamin.gaignard@collabora.com> (raw)
Unlike when resolution change on keyframes, dynamic resolution change
on inter frames doesn't allow to do a stream off/on sequence because
it is need to keep all previous references alive to decode inter frames.
This constraint have two main problems:
- more memory consumption.
- more buffers in use.
To solve these issue this series introduce REMOVE_BUFS ioctl.
VP9 conformance tests using fluster give a score of 210/305.
The 24 of the 24 resize inter tests (vp90-2-21-resize_inter_* files) are ok
but require to use postprocessor.
Kernel branch is available here:
https://gitlab.collabora.com/benjamin.gaignard/for-upstream/-/commits/remove_vb2_queue_limit_v21
GStreamer branch to use DELETE_BUF ioctl and testing dynamic resolution
change is here:
https://gitlab.freedesktop.org/benjamin.gaignard1/gstreamer/-/commits/VP9_drc
changes in version 21:
- Be more careful about checking remove_bufs type field vs queue type.
- Add documentation about type checking error.
- Always set capabilities flags field.
- Do not set vidioc_remove_bufs for vim2m driver.
changes in version 20:
- Rename DELETE_BUFS into REMOVE_BUFS
- Change documention, structure and variables name to use 'remove'
changes in version 19:
- Fix typo in commit message.
- Fix ioctl domentation
- Rework q->is_busy patch following Hans's comments
- Change where DELETE_BUFS is enabled
Benjamin Gaignard (9):
media: videobuf2: Update vb2_is_busy() logic
videobuf2: Add min_reqbufs_allocation field to vb2_queue structure
media: test-drivers: Set REQBUFS minimum number of buffers
media: core: Rework how create_buf index returned value is computed
media: core: Add bitmap manage bufs array entries
media: core: Free range of buffers
media: v4l2: Add REMOVE_BUFS ioctl
media: v4l2: Add mem2mem helpers for REMOVE_BUFS ioctl
media: verisilicon: Support removing buffers on capture queue
.../userspace-api/media/v4l/user-func.rst | 1 +
.../media/v4l/vidioc-remove-bufs.rst | 85 +++++++
.../media/v4l/vidioc-reqbufs.rst | 1 +
.../media/common/videobuf2/videobuf2-core.c | 223 ++++++++++++------
.../media/common/videobuf2/videobuf2-v4l2.c | 34 ++-
.../media/platform/verisilicon/hantro_v4l2.c | 1 +
.../media/test-drivers/vicodec/vicodec-core.c | 1 +
.../media/test-drivers/vimc/vimc-capture.c | 3 +-
drivers/media/test-drivers/visl/visl-video.c | 1 +
drivers/media/test-drivers/vivid/vivid-core.c | 5 +-
drivers/media/v4l2-core/v4l2-dev.c | 3 +
drivers/media/v4l2-core/v4l2-ioctl.c | 30 +++
drivers/media/v4l2-core/v4l2-mem2mem.c | 15 ++
include/media/v4l2-ioctl.h | 4 +
include/media/v4l2-mem2mem.h | 2 +
include/media/videobuf2-core.h | 52 +++-
include/media/videobuf2-v4l2.h | 2 +
include/uapi/linux/videodev2.h | 17 ++
18 files changed, 394 insertions(+), 86 deletions(-)
create mode 100644 Documentation/userspace-api/media/v4l/vidioc-remove-bufs.rst
--
2.40.1
next reply other threads:[~2024-03-14 15:32 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-03-14 15:32 Benjamin Gaignard [this message]
2024-03-14 15:32 ` [PATCH v21 1/9] media: videobuf2: Update vb2_is_busy() logic Benjamin Gaignard
2024-03-21 13:49 ` Mauro Carvalho Chehab
2024-03-14 15:32 ` [PATCH v21 2/9] videobuf2: Add min_reqbufs_allocation field to vb2_queue structure Benjamin Gaignard
2024-03-21 13:52 ` Mauro Carvalho Chehab
2024-03-14 15:32 ` [PATCH v21 3/9] media: test-drivers: Set REQBUFS minimum number of buffers Benjamin Gaignard
2024-03-18 10:06 ` Hans Verkuil
2024-03-18 13:06 ` Hans Verkuil
2024-03-21 13:53 ` Mauro Carvalho Chehab
2024-03-14 15:32 ` [PATCH v21 4/9] media: core: Rework how create_buf index returned value is computed Benjamin Gaignard
2024-03-14 15:32 ` [PATCH v21 5/9] media: core: Add bitmap manage bufs array entries Benjamin Gaignard
2024-03-14 15:32 ` [PATCH v21 6/9] media: core: Free range of buffers Benjamin Gaignard
2024-03-21 14:18 ` Mauro Carvalho Chehab
2024-03-14 15:32 ` [PATCH v21 7/9] media: v4l2: Add REMOVE_BUFS ioctl Benjamin Gaignard
2024-03-21 14:14 ` Mauro Carvalho Chehab
2024-03-21 14:24 ` Hans Verkuil
2024-03-21 15:03 ` Mauro Carvalho Chehab
2024-03-22 7:54 ` Hans Verkuil
2024-03-22 9:51 ` Benjamin Gaignard
2024-03-14 15:32 ` [PATCH v21 8/9] media: v4l2: Add mem2mem helpers for " Benjamin Gaignard
2024-03-21 14:15 ` Mauro Carvalho Chehab
2024-03-14 15:32 ` [PATCH v21 9/9] media: verisilicon: Support removing buffers on capture queue Benjamin Gaignard
2024-03-21 14:16 ` Mauro Carvalho Chehab
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=20240314153226.197445-1-benjamin.gaignard@collabora.com \
--to=benjamin.gaignard@collabora.com \
--cc=ezequiel@vanguardiasur.com.ar \
--cc=hverkuil-cisco@xs4all.nl \
--cc=kernel@collabora.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-rockchip@lists.infradead.org \
--cc=m.szyprowski@samsung.com \
--cc=mchehab@kernel.org \
--cc=nicolas@ndufresne.ca \
--cc=p.zabel@pengutronix.de \
--cc=tfiga@chromium.org \
/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).