25 Jan, 2021
1 commit
-
HAVE_COMPAT_IOCTL define is removed from include/linux/fs.h
on 5.10 kernel by below kernel patch, driver should remove
HAVE_COMPAT_IOCTL check, other drivers don’t have such check.
Kernel patch:
https://lkml.org/lkml/2020/6/22/218Change-Id: I428f762f929d1689d558b18349dd4ce536ac51f6
Signed-off-by: Richard Liu
18 Jan, 2021
2 commits
-
GPU clock will switch to minClock/64, 32/64 64/64 of max GPU frequency
according to thermal event.Change-Id: Ibcfa98ec8976215119acecbd92f2e130f0a39a73
Signed-off-by: Zhang Bo -
change "ifdef CONFIG_DEVICE_THERMAL" to "if IS_ENABLED(CONFIG_DEVICE_THERMAL)"
to enable the module with IS_BUILTIN(option) or IS_MODULE(option)).Change-Id: I116ed60acae56629713163888cb4a7d5a2d6acb3
Signed-off-by: Dandan Sun
13 Jan, 2021
3 commits
-
driver will use the data to report the skipped frame.
but the vpu driver of 8q can't get the skipped frame id,
so set it to an invalid value.Signed-off-by: Ming Qian
Reviewed-by: Shijie Qin
(cherry picked from commit f9782fd6e516647ad73476bb9d39eb2dd892ae05) -
…_G and V4L2_PIX_FMT_VC1_ANNEX_L
Due to historical reasons,
we have made a mistake abort vc1_l and vc1_g,
they are reversed.
so we need to exchange them in driver,
and the userspace also needs to be modified accordinglySigned-off-by: Ming Qian <ming.qian@nxp.com>
Reviewed-by: Shijie Qin <shijie.qin@nxp.com>
(cherry picked from commit 9e4f4f44257312c2b4df47d0005762e8068e1bd0) -
we have add a header file imx_vpu.h
to define the custom interface based on v4l2 driver.
to unify the interface,
we need to align the 8qm/qxp v4l2 driver to this header.Signed-off-by: Ming Qian
Reviewed-by: Shijie Qin
(cherry picked from commit 8cfb286daa653c5876c157d20bed6a44420bab27)
22 Dec, 2020
1 commit
-
include more bug-fixings for vulkan and opencl3.0
Signed-off-by: Xianzhong
18 Dec, 2020
6 commits
-
* mxc/next: (279 commits)
LF-2111-2: mxc: vpu_windsor: Avoid hide parameter
MLK-24960: [8QM_MEK/8QXP_MEK]mxc:vpu_malone: move timestamp manager to firmware
MLK-24959: [8QM_MEK/8QXP_MEK]mxc:vpu_windsor: move timestamp handler to firmware
MLK-24906: [8QM_MEK/8QXP_MEK]mxc:vpu_malone: report V4L2_COLORSPACE_DEFAULT before seq hdr found
MLK-24593:[8QM_MEK/8QXP_MEK]mxc:vpu_windsor: support change bitrate dynamically
... -
* origin/mxc/vpu: (123 commits)
LF-2111-2: mxc: vpu_windsor: Avoid hide parameter
MLK-24960: [8QM_MEK/8QXP_MEK]mxc:vpu_malone: move timestamp manager to firmware
MLK-24959: [8QM_MEK/8QXP_MEK]mxc:vpu_windsor: move timestamp handler to firmware
MLK-24906: [8QM_MEK/8QXP_MEK]mxc:vpu_malone: report V4L2_COLORSPACE_DEFAULT before seq hdr found
MLK-24593:[8QM_MEK/8QXP_MEK]mxc:vpu_windsor: support change bitrate dynamically
... -
* origin/mxc/sim: (33 commits)
LF-2932 mxc: sim/emvsim: fix competition when call pm_runtime_put
MLK-24363-2 mxc: emvsim: fix error report in probe stage
MLK-24363-1 mxc: emvsim: add error check when enable clk
MLK-24364 mxc: sim: add runtime pm support
LF-1762-7 mxc: sim: replace '---help---' in Kconfig files with 'help'
... -
* origin/mxc/mlb: (20 commits)
LF-2594-6 mxc: mlb: remove imx8qm/qxp support
LF-2225-7 mlb: drop the using of smp_read_barrier_depends
LF-665-12: mxc: mlb: fix build break due to un-exist pr_warning
MLK-23000-1 mxc: mlb: fix kernel dump when use gen_pool_dma_alloc
mxc mlb: Allow compilation on imx8
... -
* origin/mxc/ipu: (13 commits)
LF-2092 mxc: IPUv3: Correct di number when enabling DIRECT_ASYNC0 channel
LF-2091 media: platform: mxc_vout: Do not check 'crop->c.width/heightr.width/heightvidioc_s_selection
... -
* origin/mxc/hdmi:
LF-2739: mfd: mxc-hdmi-core: Complete cts value for 32kHz
mxc: hdmi-cec: Add imx6 hdmi cec driver
driver: mfd: hdmi: Add hdmi core driver
14 Dec, 2020
27 commits
-
Fix another declaration hides parameter 'attr'.
(coverity ID 5572378)Signed-off-by: Shijie Qin
Reviewed-by: ming_qian -
move timestamp to firmware, it will make our driver more easy to
upstreamSigned-off-by: Ming Qian
-
move timestamp to firmware, it will make our driver more easy to
upstreamSigned-off-by: Ming Qian
-
the new gst need driver report V4L2_COLORSPACE_DEFAULT
before seq hdr is found,
otherwise the gpaly may fail to playback some stream.
To be compatible with the new version of gst,
change the behavior of try_fmt.and the gst is adapted for kernel 5.10,
so this patch will only be pushed to 5.10,
won't be included in 5.4Signed-off-by: Ming Qian
Reviewed-by: Shijie Qin -
update the encoder parameters when bitrate is changed
Signed-off-by: Ming Qian
Reviewed-by: Shijie Qin -
use bframe to control low latency mode by default
deprecate the ctrl V4L2_CID_MPEG_VIDEO_H264_ASOSigned-off-by: Ming Qian
Reviewed-by: Shijie Qin -
gst will depend on the try_fmt api
if driver don't report the correct color space in try_fmt
gst may return failSigned-off-by: Ming Qian
Reviewed-by: Shijie Qin -
Fix declaration hides parameter 'attr'.
(coverity ID 5572378)Signed-off-by: Shijie Qin
Reviewed-by: ming_qian -
Bit field of 'valid_ts_received' is 1, change its typedef to
unsigned int. (coverity ID 9000734)Signed-off-by: Shijie Qin
Reviewed-by: ming_qian -
use the v4l2 api instead of custom api
Signed-off-by: Ming Qian
Reviewed-by: Shijie Qin -
1. enable color description in vui when encoding.
2. improve the get/set color description apiSigned-off-by: Ming Qian
Reviewed-by: Shijie Qin -
Update Kconfig to cope with upstream change
commit 84af7a6194e4 ("checkpatch: kconfig: prefer 'help' over
'---help---'").Signed-off-by: Dong Aisheng
-
Replace spin_lock() with spin_lock_irqsave() to avoid potential deadlock
Signed-off-by: Zhou Peng
(cherry picked from commit 85b5d239b0e2fb6c534064ff203f74a2e2b0e7e8) -
HW enable and disable are shared with the same function, driver need to
check the parameter to identify whether HW is enabled or notSigned-off-by: Zhou Peng
(cherry picked from commit 742c7adea8b19b047b6a5bce1c8d7da6895a3b71) -
If 'pm_runtime_get_sync()' fail, need to reset device power usage_count
to '0'. Otherwise, the usage_count always bigger than '0' and not do
'runtime_suspend()' actually.Signed-off-by: Shijie Qin
Reviewed-by: Anson Huang
Reviewed-by: ming_qian -
If 'pm_runtime_get_sync()' fail, need to reset device power usage_count
to '0'. Otherwise, the usage_count always bigger than '0' and not do
'runtime_suspend()' actually.Signed-off-by: Shijie Qin
Reviewed-by: Anson Huang
Reviewed-by: ming_qian -
there may be some redundant data before the encoded frame,
generally it won't bring any problem.
but on android, the redundant data may affect the mux,
can cause the mp4 can't be decoded.
so delete the redundant data in the driver.Signed-off-by: Ming Qian
Reviewed-by: Shijie Qin -
For more readable, use a handler table to insert scode for
each format.Signed-off-by: Shijie Qin
Reviewed-by: ming_qian -
The payload header of sequence / picture / slice are the
same operation, could unify them in a function.
payload size = data_size - payload_header_size - start_code_sizeSigned-off-by: Shijie Qin
Reviewed-by: ming_qian -
Correct VP8 ivf frame header
firmware just parse 64-bit timestamp(8 bytes).
As not transfer timestamp to firmware, use default value(ZERO).
No need to do anything here.
Refine the function of set vp8 ivf sequence header.Signed-off-by: Shijie Qin
Reviewed-by: ming_qian -
The seq payload size should is 20+12.
Refine the function of set rcv header.
Refine the function of create nal header.Signed-off-by: Shijie Qin
Reviewed-by: ming_qian -
1. the mu won't be really power off,
so we need restore the mu but not initialize it
2. the vpu is shared with vpu windsor,
so we don't do vpu reset when remove module.Signed-off-by: Ming Qian
-
delay the creation of video device node in order to avoid inmod failure
in probe function, if create video node in a early stage,
the video node may be created before initialize the vpu dev,
and in such case, open video node,
the mu mailbox may be not requested,
then it will led to download firmware fail.so create the video device node at the end of the probe function.
Signed-off-by: Ming Qian
-
1. the mu won't be really power off,
so we need restore the mu but not initialize it
2. the vpu is shared with vpu malone,
so we don't do vpu reset when remove module.Signed-off-by: Ming Qian
-
delay the creation of video device node in order to avoid inmod failure
in probe function, if create video node in a early stage,
the video node may be created before initialize the vpu dev,
and in such case, open video node,
the mu mailbox may be not requested,
then it will led to download firmware fail.so create the video device node at the end of the probe function.
Signed-off-by: Ming Qian
-
fix clk use-after-free issue
Signed-off-by: Zhou Peng
(cherry picked from commit 0d1b3814c976cb2148958f48c4c4f4701ced6b4e) -
release clk to avoid memory leak
Signed-off-by: Zhou Peng
(cherry picked from commit 420a357a1b43c57383c74a548a9a3714b6da7de2)
