Usually we think rk3568 max resolution is 4096x2160, in face rk3568 can
also support 2160x4096.
Signed-off-by: Sandy Huang <hjc@rock-chips.com>
Change-Id: I96358f481f9e26d1196e2118b3ce3b82c4a6ce82
API snd_soc_register_component will try to rebind card per
each component register. and the ASoC allow no-pcm card instance.
devm_snd_soc_register_component
snd_soc_try_rebind_card
snd_soc_bind_card
snd_soc_add_pcm_runtime
devm_snd_dmaengine_pcm_register
So, we should register PCM before DAI component.
Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
Change-Id: I578b7c54b96cfa89fd2f8cf346832e2b68a1d355
This patch do full hw params re-configuation instead of the reset one.
Fixes: 0c2c6c9430 ("drm/bridge: synopsys: Fix wrong FIFO reset on AUD_CONF0")
Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
Change-Id: I1e259797ce22cd9a3be7086e1f21bd02c440ac57
If the DRAM frequency conversion jitters during the transmission process,
it will cause the DMA to be unable to transport SPI FIFO data in a timely
manner, resulting in FIFO overflow/underflow.
Clear performance status for short cmd packet and Set the performance
status for data packet.
Change-Id: I65532ba309677a8d98c8277875a3bd358ca44e44
Signed-off-by: Jon Lin <jon.lin@rock-chips.com>
At following commit, we update connectors_changed flag for writeback
enable lead to other display dev do disable and enable:
c7e2a3387e ("drm/rockchip: vop2: No need for a full modested when the
only writeback connector changed")
And revert the drm framework update for GKI:
82a56746d9 ("Revert "drm: Not mark crtc state as connectors_changed
when a writeback connector attatch to a crtc"")
But when writeback disable, the DRM framework still thought crtc
connectors is changed and set connectors_changed to true, this will lead
other display dev do disable and enable, so we add this commit to fix
it.
Signed-off-by: Sandy Huang <hjc@rock-chips.com>
Change-Id: Ie69d63ea19110251273eb8969fc40a4fceb5e309
The edp controller can not support display timings whose
front porch or back porch is 0.
Signed-off-by: Damon Ding <damon.ding@rock-chips.com>
Change-Id: I689f9f23154cffd7199e65767e2a977cd3404ea9
If using split mode with edp0 and edp1, the node of which
displays in the left area should have "split-mode" property
and config panel node, and the other does not.
Signed-off-by: Damon Ding <damon.ding@rock-chips.com>
Change-Id: I0533df43d5fd5d9ea3fd2fb94c9a8cdab1fd59e2
The enc did not stop working when the watch dog timeout was triggered,
and still check timeout and trigger watch dog irq. It will cause watch
dog interrupt storm.
Solution:
1. Mask watch dog irq when first watch dog irq trigger.
2. Set IRQF_ONESHOT to ensure the irq handling is completed.
Signed-off-by: Yandong Lin <yandong.lin@rock-chips.com>
Change-Id: I741fe9fd96486641b1ee5d337ecc144e29778157
This patch adds support to parse pvtpll config from devicetree, and
change config through sip smc interface.
Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>
Change-Id: I74f2127d862db005433d0a9da2703629f80b0faa
This patch adds support to parse pvtpll config from devicetree, and
change config through sip smc interface.
Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>
Change-Id: I5f23b9eb1d358a95b690ca214e35927e74451af1
When there is a small probability of signal abnormality, try resetting
and enumerating again.
Change-Id: Id37950dc62a497dd1c4703c42de036ce50aaec68
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Jon Lin <jon.lin@rock-chips.com>
We can get the binding relationship between plane and vop from the standard
prop possible crtcs, so we remove the unused PLANE_MASK prop from vop.
Signed-off-by: Sandy Huang <hjc@rock-chips.com>
Change-Id: I2c68ed60447024f4f5f379fa39b28ae0da065eae
If the old plane state is invalid, no printing should be performed,
otherwise panic will occur.
Fixes: 2e409456c1 ("drm/rockchip: add dynamic debug for vop")
Change-Id: I0687f9231c49bc068322f555f95af96a214b049d
Signed-off-by: Chaoyi Chen <chaoyi.chen@rock-chips.com>
The tty_port struct is part of the rpmsg_tty_port structure.
The issue is that the rpmsg_tty_port structure is freed on
rpmsg_tty_remove while it is still referenced in the tty_struct.
Its release is not predictable due to workqueues.
For instance following ftrace shows that rpmsg_tty_close is called after
rpmsg_tty_release_cport:
nr_test.sh-389 [000] ..... 212.093752: rpmsg_tty_remove <-rpmsg_dev_
remove
cat-1191 [001] ..... 212.095697: tty_release <-__fput
nr_test.sh-389 [000] ..... 212.099166: rpmsg_tty_release_cport <-rpm
sg_tty_remove
cat-1191 [001] ..... 212.115352: rpmsg_tty_close <-tty_release
cat-1191 [001] ..... 212.115371: release_tty <-tty_release_str
As consequence, the port must be free only when user has released the TTY
interface.
This path :
- Introduce the .destruct port tty ops function to release the allocated
rpmsg_tty_port structure.
- Introduce the .hangup tty ops function to call tty_port_hangup.
- Manages the tty port refcounting to trig the .destruct port ops,
- Introduces the rpmsg_tty_cleanup function to ensure that the TTY is
removed before decreasing the port refcount.
Fixes: 7c0408d805 ("tty: add rpmsg driver")
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Arnaud Pouliquen <arnaud.pouliquen@foss.st.com>
Link: https://lore.kernel.org/r/20220104163545.34710-1-arnaud.pouliquen@foss.st.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Change-Id: I5616481fa97807c954f03cd1dfdff206605b65d3
(cherry picked from commit db7f19c0aa)
[Jiahang: Resolving type conflict in drivers/tty/rpmsg_tty.c ]
Signed-off-by: Jiahang Zheng <jiahang.zheng@rock-chips.com>
