All of lore.kernel.org
 help / color / mirror / Atom feed
From: kyrie.wu <kyrie.wu@mediatek.com>
To: Hans Verkuil <hverkuil-cisco@xs4all.nl>,
	Mauro Carvalho Chehab <mchehab@kernel.org>,
	Rob Herring <robh+dt@kernel.org>,
	Tomasz Figa <tfiga@chromium.org>,
	Matthias Brugger <matthias.bgg@gmail.com>,
	"Tzung-Bi Shih" <tzungbi@chromium.org>
Cc: <Project_Global_Chrome_Upstream_Group@mediatek.com>,
	<linux-media@vger.kernel.org>, <devicetree@vger.kernel.org>,
	<linux-kernel@vger.kernel.org>,
	<linux-arm-kernel@lists.infradead.org>,
	<linux-mediatek@lists.infradead.org>, <xia.jiang@mediatek.com>,
	<maoguang.meng@mediatek.com>, <srv_heupstream@mediatek.com>,
	<kyrie.wu@mediatek.com>, <irui.wang@mediatek.com>
Subject: [PATCH V4,0/5] Support jpeg encoder for MT8195
Date: Fri, 24 Sep 2021 15:45:38 +0800	[thread overview]
Message-ID: <1632469543-27345-1-git-send-email-kyrie.wu@mediatek.com> (raw)

MT8195 has two jpeg encoding HW, each one has its own
power-domain, clock, interrupt, register base.
The two jpeg encoding HW can work together to
achieve higher performance.

Generally, one HW needs to register one device node,
and the APP operates the HW through the device node.
If there are multiple HWs, the multiple device node
needs to be registered. At this time, the app needs
to design software logic to use multiple HWs in parallel,
which will undoubtedly increase the difficulty of
app development and weaken its compatibility.

These serial patches use the component framework
to achieve the purpose of using multiple HWs
through only one device node. About the component
framework Block Diagram, please check below:

		Master(device node)
	________|_________________________
	|	               |             |
component0(HW0)	component1(HW1) component2(HW2) ......

As above, the component framework is similar to a
master-component two level architecture.
Both master and component have the compatible
attribute in the device tree, and the corresponding
platform driver structure is defined in the driver.
Therefore, when the compatible attribute matches,
the corresponding probe function will be called.

In the probe function of component, operations
such as initialization clock, remmap register base,
registration interrupt will be performed.
And add the component to a component linked
list to find the corresponding master.

In the probe function of the master, the device
node will be registered, and the master will be
added to a master linked list to find the
corresponding component.

After the master matches the component,
it will obtain the clock, interrupt, register base
and other resources in the component,
and operate the HW through these resources.
At the same time, multiple HWs software function
interfaces are designed in the master driver.

kyrie.wu (5):
  dt-bindings: mediatek: Add mediatek, mt8195-jpgenc compatible
  media: mtk-jpegenc: Use component framework to manage each hardware
  media: mtk-jpegenc: add jpegenc timeout func interface
  media: mtk-jpegenc: add jpeg encode worker interface
  media: mtk-jpegenc: add output pic reorder interface

 .../bindings/media/mediatek-jpeg-encoder.yaml      |   3 +
 drivers/media/platform/mtk-jpeg/mtk_jpeg_core.c    | 395 ++++++++++++++++++---
 drivers/media/platform/mtk-jpeg/mtk_jpeg_core.h    |  79 +++++
 drivers/media/platform/mtk-jpeg/mtk_jpeg_enc_hw.c  | 298 ++++++++++++++++
 4 files changed, 730 insertions(+), 45 deletions(-)

--
2.6.4


_______________________________________________
Linux-mediatek mailing list
Linux-mediatek@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-mediatek

WARNING: multiple messages have this Message-ID (diff)
From: kyrie.wu <kyrie.wu@mediatek.com>
To: Hans Verkuil <hverkuil-cisco@xs4all.nl>,
	Mauro Carvalho Chehab <mchehab@kernel.org>,
	Rob Herring <robh+dt@kernel.org>,
	Tomasz Figa <tfiga@chromium.org>,
	Matthias Brugger <matthias.bgg@gmail.com>,
	"Tzung-Bi Shih" <tzungbi@chromium.org>
Cc: <Project_Global_Chrome_Upstream_Group@mediatek.com>,
	<linux-media@vger.kernel.org>, <devicetree@vger.kernel.org>,
	<linux-kernel@vger.kernel.org>,
	<linux-arm-kernel@lists.infradead.org>,
	<linux-mediatek@lists.infradead.org>, <xia.jiang@mediatek.com>,
	<maoguang.meng@mediatek.com>, <srv_heupstream@mediatek.com>,
	<kyrie.wu@mediatek.com>, <irui.wang@mediatek.com>
Subject: [PATCH V4,0/5] Support jpeg encoder for MT8195
Date: Fri, 24 Sep 2021 15:45:38 +0800	[thread overview]
Message-ID: <1632469543-27345-1-git-send-email-kyrie.wu@mediatek.com> (raw)

MT8195 has two jpeg encoding HW, each one has its own
power-domain, clock, interrupt, register base.
The two jpeg encoding HW can work together to
achieve higher performance.

Generally, one HW needs to register one device node,
and the APP operates the HW through the device node.
If there are multiple HWs, the multiple device node
needs to be registered. At this time, the app needs
to design software logic to use multiple HWs in parallel,
which will undoubtedly increase the difficulty of
app development and weaken its compatibility.

These serial patches use the component framework
to achieve the purpose of using multiple HWs
through only one device node. About the component
framework Block Diagram, please check below:

		Master(device node)
	________|_________________________
	|	               |             |
component0(HW0)	component1(HW1) component2(HW2) ......

As above, the component framework is similar to a
master-component two level architecture.
Both master and component have the compatible
attribute in the device tree, and the corresponding
platform driver structure is defined in the driver.
Therefore, when the compatible attribute matches,
the corresponding probe function will be called.

In the probe function of component, operations
such as initialization clock, remmap register base,
registration interrupt will be performed.
And add the component to a component linked
list to find the corresponding master.

In the probe function of the master, the device
node will be registered, and the master will be
added to a master linked list to find the
corresponding component.

After the master matches the component,
it will obtain the clock, interrupt, register base
and other resources in the component,
and operate the HW through these resources.
At the same time, multiple HWs software function
interfaces are designed in the master driver.

kyrie.wu (5):
  dt-bindings: mediatek: Add mediatek, mt8195-jpgenc compatible
  media: mtk-jpegenc: Use component framework to manage each hardware
  media: mtk-jpegenc: add jpegenc timeout func interface
  media: mtk-jpegenc: add jpeg encode worker interface
  media: mtk-jpegenc: add output pic reorder interface

 .../bindings/media/mediatek-jpeg-encoder.yaml      |   3 +
 drivers/media/platform/mtk-jpeg/mtk_jpeg_core.c    | 395 ++++++++++++++++++---
 drivers/media/platform/mtk-jpeg/mtk_jpeg_core.h    |  79 +++++
 drivers/media/platform/mtk-jpeg/mtk_jpeg_enc_hw.c  | 298 ++++++++++++++++
 4 files changed, 730 insertions(+), 45 deletions(-)

--
2.6.4


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

             reply	other threads:[~2021-09-24  7:46 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-24  7:45 kyrie.wu [this message]
2021-09-24  7:45 ` [PATCH V4,0/5] Support jpeg encoder for MT8195 kyrie.wu
2021-09-24  7:45 ` [PATCH V4, 1/5] dt-bindings: mediatek: Add mediatek, mt8195-jpgenc compatible kyrie.wu
2021-09-24  7:45   ` kyrie.wu
2021-09-24  7:45 ` [PATCH V4, 2/5] media: mtk-jpegenc: Use component framework to manage each hardware kyrie.wu
2021-09-24  7:45   ` kyrie.wu
2021-09-24  7:45 ` [PATCH V4, 3/5] media: mtk-jpegenc: add jpegenc timeout func interface kyrie.wu
2021-09-24  7:45   ` kyrie.wu
2021-09-24  7:45 ` [PATCH V4,4/5] media: mtk-jpegenc: add jpeg encode worker interface kyrie.wu
2021-09-24  7:45   ` kyrie.wu
2021-09-24  7:45 ` [PATCH V4,5/5] media: mtk-jpegenc: add output pic reorder interface kyrie.wu
2021-09-24  7:45   ` kyrie.wu
2021-09-24  7:55 ` [PATCH V4,0/5] Support jpeg encoder for MT8195 Chen-Yu Tsai
2021-09-24  7:55   ` Chen-Yu Tsai
2021-09-24  7:55   ` Chen-Yu Tsai

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=1632469543-27345-1-git-send-email-kyrie.wu@mediatek.com \
    --to=kyrie.wu@mediatek.com \
    --cc=Project_Global_Chrome_Upstream_Group@mediatek.com \
    --cc=devicetree@vger.kernel.org \
    --cc=hverkuil-cisco@xs4all.nl \
    --cc=irui.wang@mediatek.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-mediatek@lists.infradead.org \
    --cc=maoguang.meng@mediatek.com \
    --cc=matthias.bgg@gmail.com \
    --cc=mchehab@kernel.org \
    --cc=robh+dt@kernel.org \
    --cc=srv_heupstream@mediatek.com \
    --cc=tfiga@chromium.org \
    --cc=tzungbi@chromium.org \
    --cc=xia.jiang@mediatek.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.