There are only two lines in mtk_vcodec_release_enc_pm, using
pm_runtime_disable and put_device instead directly.

Move pm_runtime_enable outside mtk_vcodec_release_enc_pm to symmetry with
pm_runtime_disable, after that, rename mtk_vcodec_init_enc_pm to *_clk since
it only has clock operations now.

Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Co-developed-by: Yong Wu <yong.wu@mediatek.com>

There are only two lines in mtk_vcodec_release_dec_pm, using
pm_runtime_disable and put_device instead directly.

Move pm_runtime_enable outside mtk_vcodec_init_dec_pm to symmetry with
pm_runtime_disable, after that, rename mtk_vcodec_init_dec_pm to *_clk since
it only has clock operations now.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Co-developed-by: Yong Wu <yong.wu@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>

There is just one core thread, in order to separate different
hardware, using codec type to separeate it in scp driver.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>

Add core dec and dec end ipi msg: AP_IPIMSG_DEC_CORE/AP_IPIMSG_DEC_CORE_END.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>

Adds decoder dt-bindings for mt8192.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: Rob Herring <robh@kernel.org>

Use the dma_set_mask_and_coherent helper to set vdec
DMA bit mask to support 34bits iova space(16GB) that
the mt8192 iommu HW support.

Whole the iova range separate to 0~4G/4G~8G/8G~12G/12G~16G,
regarding which iova range VDEC actually locate, it
depends on the dma-ranges property of vdec dtsi node.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>

Add work queue to process core hardware information.
First, get lat_buf from message queue, then call core
hardware of each codec(H264/VP9/AV1) to decode, finally
puts lat_buf back to the message.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: Steve Cho <stevecho@chromium.org>

For add new hardware, not only need to lock lat hardware, also
need to lock core hardware in case of different instance start
to decoder at the same time.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>

Generalizes power and clock on/off interfaces to support different hardware.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>

For lat and core architecture, lat thread will send message to core
thread when lat decode done. Core hardware will use the message
from lat to decode, then free message to lat thread when decode done.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>

Adds irq interface for multi hardware.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>

Separates different architecture for hardware: pure_sin_core
and lat_sin_core. MT8183 is pure single core. Uses .hw_arch to
distinguish.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>

…-bindings

Separate decoder and encoder document for the dts are big difference.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: Rob Herring <robh@kernel.org>

There are more than two hardwares for decoder: LAT0, LAT1 and CORE. In order to
manage these hardwares, register each hardware as independent platform device
for the larbs are different.

Each hardware module controls its own information which includes interrupt/power/
clocks/registers.

Calling of_platform_populate in parent device, and use subdev_bitmap to record
whether the hardwares are registered.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: Steve Cho <stevecho@chromium.org>

Adds MT8192's compatible "mediatek,mt8192-vcodec-dec".
Adds MT8192's device private data mtk_lat_sig_core_pdata.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>

When mtk vcodec decoder is build as a module, we need to export
mtk-vcodec-dec pm functions to make them visible by the other components.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Benjamin Gaignard <benjamin.gaignard@collabora.com>

Using the needed params for pm init/release function and remove unused
param mtkdev in 'struct mtk_vcodec_pm'.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: Tzung-Bi Shih <tzungbi@google.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Steve Cho <stevecho@chromium.org>

Vdec and venc can use the same function to wake up interrupt event.

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: Tzung-Bi Shih <tzungbi@google.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Steve Cho <stevecho@chromium.org>

Different platforms may have different numbers of register bases. Gets the
numbers of register bases from dts (sizeof(u32) * 4 bytes for each).

Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Reviewed-by: Tzung-Bi Shih <tzungbi@google.com>

Use DEVICE_ATTR_RO() helper macro instead of DEVICE_ATTR(), which is
simpler and more readable.

Due to the name of the read function of the sysfs attribute is normalized,
there is a natural association.

Link: https://lore.kernel.org/linux-media/20210603071009.11438-1-thunder.leizhen@huawei.com

Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

use SPDX-License-Identifier instead of a verbose license text

Link: https://lore.kernel.org/linux-media/20210916020006.8497-1-caihuoqing@baidu.com

CC: Mauro Carvalho Chehab <mchehab@kernel.org>, <linux-media@vger.kernel.org>, <linux-kernel@vger.kernel.org>
Signed-off-by: Cai Huoqing <caihuoqing@baidu.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

The devices 0x187f,0x0202 and 0x187f,0x0301 are already on the list.

Link: https://lore.kernel.org/linux-media/20210917092132.19576-1-krzysztof.kozlowski@canonical.com

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Using stack-allocated pointers for USB message data don't work.
This driver is almost OK with that, except for the I2C read
logic.

Fix it by using a temporary read buffer, just like on all other
calls to m920x_read().

Link: https://lore.kernel.org/all/ccc99e48-de4f-045e-0fe4-61e3118e3f74@mida.se/
Reported-by: rkardell@mida.se
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

An spi remove callback is never called with an spi_device pointer that
is NULL. Also it is only called for devices that probed successfully. As
cxd2880_spi_probe() always sets driver data, spi_get_drvdata() cannot be
NULL.

Also the return value of spi remove callbacks is ignored anyway and not
freeing resources in .remove() is a bad idea.

Link: https://lore.kernel.org/linux-media/20211012153945.2651412-8-u.kleine-koenig@pengutronix.de

Cc: Mark Brown <broonie@kernel.org>, kernel@pengutronix.de, linux-media@vger.kernel.org, linux-spi@vger.kernel.org
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

If disabling the hardware fails the driver propagates the error code to
the i2c core. However this only results in a generic error message; the
device still disappears.

So instead emit a message that better describes the actual problem than
the i2c core's "remove failed (ESOMETHING), will be ignored" and return
0 to suppress the generic message.

Link: https://lore.kernel.org/linux-media/20211026194010.109029-1-u.kleine-koenig@pengutronix.de

Cc: linux-media@vger.kernel.org, kernel@pengutronix.de
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

… module

The module currently prints only "Rafael Micro r820t successfully
identified" when successfully loaded, which might be misleading as the
module actually supports various chip types.

Link: https://lore.kernel.org/linux-media/20211119234401.271193-1-dkk089@gmail.com

Cc: linux-kernel@vger.kernel.org, Daniel Kamil Kozar <dkk089@gmail.com>
Signed-off-by: Daniel Kamil Kozar <dkk089@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

There is a spelling mistake in a dev_dbg message. Fix it.

Link: https://lore.kernel.org/linux-media/20211123090645.165299-1-colin.i.king@gmail.com

Cc: kernel-janitors@vger.kernel.org, linux-kernel@vger.kernel.org
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Commit e955f95 ("media: si2157: Better check for running tuner in
init") completely broke the "warm" tuner detection of the si2157 driver
due to a simple endian error: The Si2157 CRYSTAL_TRIM property code is
0x0402 and needs to be transmitted LSB first. However, it was inserted
MSB first, causing the warm detection to always fail and spam the kernel
log with tuner initialization messages each time the DVB frontend
device was closed and reopened:

[  312.215682] si2157 16-0060: found a 'Silicon Labs Si2157-A30'
[  312.264334] si2157 16-0060: firmware version: 3.0.5
[  342.248593] si2157 16-0060: found a 'Silicon Labs Si2157-A30'
[  342.295743] si2157 16-0060: firmware version: 3.0.5
[  372.328574] si2157 16-0060: found a 'Silicon Labs Si2157-A30'
[  372.385035] si2157 16-0060: firmware version: 3.0.5

Also, the reinitializations were observed disturb _other_ tuners on
multi-tuner cards such as the Hauppauge WinTV-QuadHD, leading to missed
or errored packets when one of the other DVB frontend devices on that
card was opened.

Fix the order of the property code bytes to make the warm detection work
again, also reducing the tuner initialization message in the kernel log
to once per power-on, as well as fixing the interference with other
tuners.

Link: https://lore.kernel.org/linux-media/trinity-2a86eb9d-6264-4387-95e1-ba7b79a4050f-1638392923493@3c-app-gmx-bap03

Fixes: e955f95 ("media: si2157: Better check for running tuner in init")
Reported-by: Robert Schlabbach <robert_s@gmx.net>
Signed-off-by: Robert Schlabbach <robert_s@gmx.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Pointer tsin is being assigned a value that is never read. The assignment
is redundant and can be removed.

Link: https://lore.kernel.org/linux-media/20211205003745.227491-1-colin.i.king@gmail.com

Cc: kernel-janitors@vger.kernel.org
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

The beginning part of the "existing HSV/HSL formats" table (line 7742)
reads:

  .. raw:: latex

      \begingroup
      \tiny
      \setlength{\tabcolsep}{2pt}

However, the ending part (line 7834) reads:

  .. raw:: latex

      \normalsize

Fix the imbalance by replacing the \normalsize with \endgroup.

Note:
    Actually, the imbalance is harmless and just results in an
    informative message near the bottom of userspace-api.log:

      (\end occurred inside a group at level 1)

      ### semi simple group (level 1) entered at line 70696 (\begingroup)
      ### bottom level

Link: https://lore.kernel.org/linux-media/b3eeec4d-1a34-0a1a-3097-1ddea3b5f1c8@gmail.com
Signed-off-by: Akira Yokosawa <akiyks@gmail.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

There are two occurrences where the variable 'asd' is dereferenced
before check. Fix this issue by using the variable after the check.

Link: https://lore.kernel.org/linux-media/20211122074122.GA6581@kili/

Link: https://lore.kernel.org/linux-media/20211201141904.47231-1-kitakar@gmail.com
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Tsuchiya Yuto <kitakar@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Don't populate the array idx_map on the stack but instead make it
static const. Also makes the object code smaller by 16 bytes.

Link: https://lore.kernel.org/linux-media/20211129230626.549090-1-colin.i.king@gmail.com
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

… length

In function vb2_set_plane_payload, report if the given bytesused is
bigger than the buffer size, and clamp it to the buffer size.

Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

In many places the bytesused field of struct vb2_buffer is set
directly. Replace that with the function call
vb2_set_plane_payload

Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

The cec_devnode struct has a lock meant to serialize access
to the fields of this struct. This lock is taken during
device node (un)registration and when opening or releasing a
filehandle to the device node. When the last open filehandle
is closed the cec adapter might be disabled by calling the
adap_enable driver callback with the devnode.lock held.

However, if during that callback a message or event arrives
then the driver will call one of the cec_queue_event()
variants in cec-adap.c, and those will take the same devnode.lock
to walk the open filehandle list.

This obviously causes a deadlock.

This is quite easy to reproduce with the cec-gpio driver since that
uses the cec-pin framework which generated lots of events and uses
a kernel thread for the processing, so when adap_enable is called
the thread is still running and can generate events.

But I suspect that it might also happen with other drivers if an
interrupt arrives signaling e.g. a received message before adap_enable
had a chance to disable the interrupts.

This patch adds a new mutex to serialize access to the fhs list.
When adap_enable() is called the devnode.lock mutex is held, but
not devnode.lock_fhs. The event functions in cec-adap.c will now
use devnode.lock_fhs instead of devnode.lock, ensuring that it is
safe to call those functions from the adap_enable callback.

This specific issue only happens if the last open filehandle is closed
and the physical address is invalid. This is not something that
happens during normal operation, but it does happen when monitoring
CEC traffic (e.g. cec-ctl --monitor) with an unconfigured CEC adapter.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Cc: <stable@vger.kernel.org>  # for v5.13 and up
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>