To fix problem in redmine #297983, make some rules about fiq_target_cpu:
The default value of fiq_target_cpu is zero and
can only be updated in sip_fiq_debugger_switch_cpu
or sip_fiq_debugger_enable_fiq.
View redmine#297983 for details.
Change-Id: I947a73c3ffc0c818a611e108a343f05b8465645b
Signed-off-by: XiaoDong Huang <derrick.huang@rock-chips.com>
When using 24MHz reference clock, some devices can't identify
the SATA PM chip, And the signal quality is not as good as 100MHz.
so change the reference clock to 100MHz.
Signed-off-by: Yifeng Zhao <yifeng.zhao@rock-chips.com>
Change-Id: If7d951a0b77d503f9faf1c1f88c78a9e07471e47
EPD_A2 mode use part update to save panel power cost
Signed-off-by: Zorro Liu <lyx@rock-chips.com>
Change-Id: I65181c3ef51034b9302532f9f1276ebd2724d0ee
usage:
csi has 4 available channels, generally id0->vc0, id1->vc1, id2->vc2, id3->vc3.
A channel can only takes one data type. When you need to collect spd data or embedded data,
you can use an idle channel and change its vc to be consistent with active data.
dts configuration:
imx258: imx258@10 {
rockchip,spd-id = <3>;//use id3 to capture spd data
rockchip,ebd-id = <2>;//use id2 to capture embedded data
};
Signed-off-by: Zefa Chen <zefa.chen@rock-chips.com>
Change-Id: I89a0f9472bbe475c9da368b09eaad0cd00fc69c6
Test on RK3568S EVB1, the otg device data eye test fail with
far end template when use 1.2 meter long cable. So tuning
the hs eye height from 400mv(default) to 437.5mv. And we
test on RK3568 EVB1, it can also benefit from this patch.
Signed-off-by: William Wu <william.wu@rock-chips.com>
Change-Id: Ie2342aba5546990838fdd6faf27a007a8843fd0d
To more precisely follow the spec, we should make sure
refclk is available and stable for device only after
stable training. In previous way, if the refclk is provided
by external chip, enabling power means refclk is ready before
anything, and then we reset the device to hope the chip back
to the initial state. But we find some devices are not really
meant to do that, that being said, the spec is vague here so
we can't make any promise from the vendor that #PERST will take
everything back.
Move resetting device before enabling power to restrictly follow
the spec, no matter whether #PERST is work or not, we just rely
on the power control.
Fixes: c84a4aa411 ("PCI: rockchip: dw: Move deassert #PERST after enabling LTSSM")
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Change-Id: Ie1f4aa1a3fb7b6de813512bf4b2c025328e0c17f
MEDIA_BUS_FMT_EBD_1X8: embedded data
MEDIA_BUS_FMT_SPD_2X8: shield pixels data
Change-Id: I5224dd860058f0d5d78a9038c53de9c9e048cb8f
Signed-off-by: Cai YiWei <cyw@rock-chips.com>
tips:
1. If fd has imported, then just copy mem_regiony struct.
2. Task is the unit for running, and mem_region is the
internal element of task.
3. In mem_regions, it can only kref_get once with the same fd,
instead of all fds which the same fd.
Change-Id: I7236803a5a263d6e79256036caf580b5ac2c35e9
Signed-off-by: Ding Wei <leo.ding@rock-chips.com>
The cpll clk gate bits had an ordering issue. This led to the loss of
the boot sdmmc controller when the gmac was shut down with:
`ip link set eth0 down`
as the cpll_100m was shut off instead of the cpll_62p5.
cpll_62p5, cpll_50m, cpll_25m were all off by one with cpll_100m
misplaced.
Fixes: e9ac850b88 ("clk: rockchip: add clock controller for rk3568")
Signed-off-by: Peter Geis <pgwipeout@gmail.com>
Link: https://patchwork.kernel.org/project/linux-clk/patch/20210519174149.3691335-1-pgwipeout@gmail.com/
Signed-off-by: Elaine Zhang <zhangqing@rock-chips.com>
Change-Id: I012bdbdc44c4e8de1b42a00c2a9bffb7bd66faef
Many TCON devices include an embedded LCD panel self-test mode.
This mode is designed to help system integrators identify
the root cause of abnormal display operation, without the use of
complicated debug tools.
Signed-off-by: Wyon Bi <bivvy.bi@rock-chips.com>
Change-Id: I19770c7488d43e2486c5fde5cc0a5b345e5be0eb
reason: mem_regions are within a task, and the differ time between
mem_region can be ignore. Thus, it can be remove when attach task fds.
Change-Id: I5986d2dca6eff41d6c6a8fa3f037e183b102e63c
Signed-off-by: Ding Wei <leo.ding@rock-chips.com>
1.add debug ioctl
2.EPD_A2 mode no need to do part count
Signed-off-by: Zorro Liu <lyx@rock-chips.com>
Change-Id: I6ddfbc5683743759cb2f7be96265df06d3db7de9
if rcb_info not change, then the infos will not ioctl from userspace.
thus, it needs to restore infos in session.
Change-Id: I4ff3b5bcc22b033b1d7549866af7c253bf46945c
Signed-off-by: Ding Wei <leo.ding@rock-chips.com>
By making the Y channel and the UV channel's access address equal,
the function of RGA input grayscale image is realized, without
need to allocate extra UV channel memory.
Signed-off-by: Yu Qiaowei <cerf.yu@rock-chips.com>
Change-Id: I0110ec6935c7233905e724be3df9f4fba9ef8cf0
win_mask is more safe than plane_mask on crtc_state,
because crtc_state may changed by many interface.
Signed-off-by: Andy Yan <andy.yan@rock-chips.com>
Change-Id: I886c8e1e1c0505e46292721de05d9be7c167d956
The f_uac1 use ep address 0x81 and 0x01 to get sample rate,
so it can only support uac1 ep1-in and ep1-out to get playback
and capture sample rate, it has limitation for USB composite
device (e.g. RNDIS & UAC1). This patch use in_ep->address and
out_ep->address instead of the fixed addresses.
Signed-off-by: William Wu <william.wu@rock-chips.com>
Change-Id: Ia1a454d9738324605c00c4889b3fe187fdfcdb7d
Because alt settings in an interface are supported only
for ISOC endpoints, so do not handle alt 0 for bulk
streaming interface.
Change-Id: Iadcf6c058023d5a3d6caa353cd783752f84a77a9
Signed-off-by: William Wu <william.wu@rock-chips.com>
when pixels less then default-max-load, use normal-rates,
otherwise, use advanced-rates instead.
Change-Id: Ie2d58c05a2e8d2f5e5f9d932b3e51024b48b91d2
Signed-off-by: Ding Wei <leo.ding@rock-chips.com>
According to USB Audio Device 1.0 Spec, Ch4.6.1.1:
The wMaxPacketSize of endpoint is defined as follows:
Maximum packet size of endpoint is capable of sending
or receiving when this configuration is selected. This
is determined by the audio bandwidth constraints of
the endpoint.
In current code, the wMaxPacketSize is limited to 200,
and the bInterval is set to 4 (1ms). That is, the maximum
bandwidth over USB bus is 200 * 1000 = 200000 bytes.
We find an issue about bandwidth limitation when we try to
support UAC1 with 8ch * 16bit * 16KHz on RK3308 platform,
which needs more bandwidth than it can support.
This patch sets the wMaxPacketSize dynamically according
to the parameters of UAC1. It is similar to the same thing
done earlier for f_uac2.
Change-Id: I9af3fd7665a80b5eb0cfb5dc91ebe2c20df1dd46
Signed-off-by: William Wu <william.wu@rock-chips.com>
As per UAC2 Audio Data Formats spec (2.3.1.1 USB Packets),
if the sampling rate is a constant, the allowable variation
of number of audio slots per virtual frame is +/- 1 audio slot.
It means that endpoint should be able to accept/send +1 audio
slot.
Previous endpoint max_packet_size calculation code
was adding sometimes +1 audio slot due to DIV_ROUND_UP
behaviour which was rounding up to closest integer.
However this doesn't work if the numbers are divisible.
It had no any impact with Linux hosts which ignore
this issue, but in case of more strict Windows it
caused rejected enumeration
Thus always add +1 audio slot to endpoint's max packet size
Change-Id: I3bef24e77d5d9dfc11438da24664192753e2cac4
Fixes: 913e4a90b6 ("usb: gadget: f_uac2: finalize wMaxPacketSize according to bandwidth")
Cc: Peter Chen <peter.chen@freescale.com>
Cc: <stable@vger.kernel.org> #v4.3+
Signed-off-by: Ruslan Bilovol <ruslan.bilovol@gmail.com>
Link: https://lore.kernel.org/r/1614599375-8803-2-git-send-email-ruslan.bilovol@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: William Wu <william.wu@rock-chips.com>
(cherry picked from commit 789ea77310)
RGA need to access CMA buffer at kernel space, so add this flag to keep kernel
line mapping for RGA.
Change-Id: Ia59acee3c904a495792229a80c42f74ae34200e3
Signed-off-by: Sandy Huang <hjc@rock-chips.com>
