When reading pinconf-pins from debugfs it fails to get the configured pull
type on RK3568, "unsupported pinctrl type" error messages is also reported.
Fix this by adding support for RK3568 in rockchip_get_pull, including a
reverse of the pull-up value swap applied in rockchip_set_pull so that
pull-up is correctly reported in pinconf-pins.
Also update the workaround comment to reflect affected pins, GPIO0_D3-D6.
Fixes: c0dadc0e47 ("pinctrl: rockchip: add support for rk3568")
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Reviewed-by: Jianqun Xu <jay.xu@rock-chips.com>
Link: https://lore.kernel.org/r/20230110172955.1258840-1-jonas@kwiboo.se
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit f8ac5467e1)
Signed-off-by: Steven Liu <steven.liu@rock-chips.com>
Change-Id: I8c6a115f1c6c62b734f76487a0ace63f4bc43743
This patch Use DMC-DVFS-SCENE policy for High Bitrate Audio.
Ref: commit 5f0eb1c57240 ("ASoC: rockchip: Introduce rockchip utils common API")
Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
Change-Id: I58a34020fabbddd7f47a5831b3a33c6f81118fe1
This patch Use DMC-DVFS-SCENE policy for High Bitrate Audio.
Ref: commit 5f0eb1c57240 ("ASoC: rockchip: Introduce rockchip utils common API")
Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
Change-Id: I9492fa993b6095f53f253615304ec445ac9cb53f
This patch Use DMC-DVFS-SCENE policy for High Bitrate Audio.
Ref: commit 5f0eb1c57240 ("ASoC: rockchip: Introduce rockchip utils common API")
Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
Change-Id: I9b919a8b2ebc85400eafb4229cd9f70b2df2bd9b
At the monment, we add utils_get/put_performance for
DMC-DVFS-SCENE policy.
Ref: commit cae65d78f0 ("ASoC: rockchip: i2s-tdm: Add support for FIFO-XRUN detection")
Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
Change-Id: I188e5cfff960aa3679db5c3a8ee847d9c7b685ee
Cannot use android_device to get multiple android instances.
The cdev will not be freed by the set_gadget_data(gadget, NULL),
so we can do not use the spinlock of gadget item to prevent
the condition.
Bug: 298603473
Fixes: dd139186ef ("ANDROID: usb: gadget: fix NULL pointer dereference in android_setup")
Signed-off-by: William Wu <william.wu@rock-chips.com>
Signed-off-by: Bin Yang <yangbin@rock-chips.com>
Change-Id: Ic0c72971051401d218e2eba75c40194933290e16
When three threads run on three different cpu, and write
60 byte to the same uart at the same time, which causes
the data loss. After test, it is caused by DMA bursts 16 byte
when the fifo is more than 48 byte(fifo total 64 byte).
It is solved by clearing the uart dma request before start
dma tx.
Signed-off-by: Huibin Hong <huibin.hong@rock-chips.com>
Change-Id: I4e2099337e25e41a34ceeb9ea774a04cc7c71f07
This patch add support for hw_free which is paired to
hw_params.
Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
Change-Id: Iaddffe89212bd734dcd3fc8de38c7ff3b2abe652
Should skip the dais which mapped as zero channel.
Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
Change-Id: I875f6dc6795b8cfed5abffcaf849fb52b729d5a3
This patch fix error path handle and also clean the
log message.
Change-Id: Id361bc2b0765371b80937ef1437f649c70e948ed
Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
The vup_gpio is in output high by default when
system enter suspend, this may cause gpio leakage
current 0.7mA. So we need to set the vup_gpio to
output low when enter pm suspend, and set it to
output high on pm resume.
Note that if the usb gadget needs to support
remote wakeup on suspend, we should add the
property "wakeup-source" in the u2phy0 node
to avoid manage the vup_gpio in suspend and
resume.
Signed-off-by: William Wu <william.wu@rock-chips.com>
Change-Id: Id7cbcc100d4e946d561b85eba1defe51f115745d
Add binding documentation for the power domains
found on Rockchip RK3562 SoCs.
Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>
Change-Id: I755b753bb08aab5a9c2b1ee8784f762a8bdf368a
In RK3588 VOP, HDMI1/eDP1 MUX1 module's reset signal should be
released when PD_VOP turn on. If this reset signal is not be
released, the HDMI1 or eDP1 output interface can't work normally.
However, If the deassert signal want to transfer to HDMI1/eDP1
MUX1 and take effect, it need the video port0 dclk's source clk
work a few moment.
In some cases, the video port0 dclk's source clk is disabled(now
only the hdmi0/1 phy pll as the dclk source parent will appear)
after PD_VOP turn on, for example, vidoe port0 dclk source select
hdmi phy pll. To fix this issue, enable video port0 dclk for a
few monent when active a video port which attach to eDP1/HDMI1.
Signed-off-by: Zhang Yubing <yubing.zhang@rock-chips.com>
Change-Id: I45d69eb06d8336caf92debb70d17f12b06c1eccf
1. Support ioctl PCIE_EP_DMA_XFER_BLOCK
2. Support dma_debug
Change-Id: I1d3e19bafcc9daf4d36e15cd089b2fd0ef45c3c2
Signed-off-by: Jon Lin <jon.lin@rock-chips.com>
1.Export DMA transmission interface from dmatest to kernel
2.Export ioctl DMA transmission interface from kernel to userspace
Change-Id: Icd129e23747d5f3ec386c6de9371711a07ee7965
Signed-off-by: Jon Lin <jon.lin@rock-chips.com>
This patch handle for err path and throw readable log
message for better debug.
Change-Id: I02aaf3fbedb56dfa0ff72966fa7489f28b50e436
Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
In order to use dw_pcie_writel_dbi2() and standard macro
to disable unused BARs.
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Change-Id: Ibe26abfb319f3f75899dd2f8c4f7b0a9a733bfa7
[W=1] sound/soc/rockchip/rockchip_sai.c:883:29:
warning: ‘tsft_enum’ defined but not used [-Wunused-const-variable=]
Change-Id: Ifd7fa236cf8fecc9dd570c9aa5f1a812bc4a64ea
Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
When different processes import buffers, the same virtual address may
exist, so current needs to be used for judgment.
Change-Id: If522075de57d4e4fb02dbf5dfcaca4f23fb22de9
Signed-off-by: Yu Qiaowei <cerf.yu@rock-chips.com>
