PD#147089: watchdog: meson_wdt: fix register and callback of reboot notify
The function register_reboot_notifier() use the wrong notifier_block
variable. So the container_of in the callback can't get the right pointer
of the struct.
The watchdog is disable in case of system down. If the kernel panic, the
reboot notify should ignore that.
Change-Id: I3fd21590808f4b380ddd54883db576e5c13c04a1
Signed-off-by: Bo Yang <bo.yang@amlogic.com>
PD#147071: dtsi: add label for the ee watchdog
So the node can be referrenced for changing.
Change-Id: I5198949ce550b7e8260cdfca3ae6164dcf5ad9c4
Signed-off-by: Bo Yang <bo.yang@amlogic.com>
PD#146748: audio capture noise on analog and digital mic
Change-Id: Id9820999399d3d933fc055fa45188fc702bba211
Signed-off-by: Peipeng Zhao <peipeng.zhao@amlogic.com>
PD#138714: remove the following error when checkpatch
Remove Gerrit Change-Id's before submitting upstream
Change-Id: I7276a44cef37d7cc4906be996d6eb9a19ae1102a
Signed-off-by: jianxin.pan <jianxin.pan@amlogic.com>
PD#146614: adc_keypad: update interface to access adc and add sysfs interface
1.use sysfs interface to set dynamic key
inquire key map:
cat /sys/devices/platform/adc_keypad/table
set key:
echo [name]:[code]:[channel]:[value]:[tolerance] >
/sys/devices/platform/adc_keypad/table
2. use the consumer interface of the IIO to access sar adc
Change-Id: I0357e84720340b0fee4baa882f41f0aa7f36a9a1
Signed-off-by: xingyu.chen <xingyu.chen@amlogic.com>
PD#146653: fix tlv320adc3101 support S24_LE and 16k
Change-Id: I54daeb69125f83e53164c03b39cfb5c8af742e95
Signed-off-by: Alex Deng <alex.deng@amlogic.com>
PD#146619: AXG fail or hard lock up when idle enable.
Change-Id: I688e134f22a1fed67aae98a7a99242c54db5aab7
Signed-off-by: Yan Wang <yan.wang@amlogic.com>
PD#146592: audio: clock only enable in runtime for power consumption
1. clocks for tdm in & out/spdif in & out/pdm in, only enable when runtime
2. remove unused codes
Change-Id: Ic75ab4557b9e1c26fa032a5b14a7a39ae53f6a06
Signed-off-by: Xing Wang <xing.wang@amlogic.com>
PD#145715: new audio features:
1. mute gpio
2. tas575x driver
3. multi-codec prefix name
4. tlv320adc3101 can revert bclk
5. set clkmsr to check mclk for AXG tdm
Change-Id: Ibd3d9ed8086715439c4bebeb574b998c1ffd1e45
Signed-off-by: Shuai Li <shuai.li@amlogic.com>
PD#146222: saradc: refactor the driver based on the iio sub-system
the sar adc driver has been moved from directory input/saradc
into directory iio/adc, and is refactored base on the standard
iio sub-system.
the iio sub-system provide two ways to access sar adc resources:
1. standard api for consumer driver
it can be found at the file "include/linux/iio/consumer.h".
2. sysfs interface for user space(X: 0-7)
- obtain 10-bit sample value:
cat sys/bus/iio/devices/iio:device0/in_voltage[X]_input
- obtain raw sample value(10-bit or 12-bit):
cat sys/bus/iio/devices/iio:device0/in_voltage[X]_raw
- obtain mean raw sample value:
cat sys/bus/iio/devices/iio:device0/in_voltage[X]_mean_raw
test pass on the gxl skt, axg skt, m200.
Change-Id: I6c8877c782c51a01993557cabc0d43f212a8e524
Signed-off-by: xingyu.chen <xingyu.chen@amlogic.com>
PD#146539: usb: fix adb reboot panic.
Avoid kernel panic caused by race condition. For example,
1. In the ffs_epfile_io function, data buffer is allocated
for non-halt requests and the address of this buffer is
writed to usb controller registers.
2. After adb process be killed, data buffer is freed and
this memory is allocated for the other. But the address
is hold by the controller.
3. Adbd in PC is running. So, the controller receive the
data and write to this memory.
4. The value of this memory is modified by the controller.
This could cause the kernel panic.
To avoid this, during FunctionFS mount, we allocated the
data buffer for requests. And the memory resources has
been released in kill_sb.
Change-Id: I494988e48bb40a21d57adcf38d41c126f9b71978
Signed-off-by: Yue Wang <yue.wang@amlogic.com>
PD#144946: Makefile: fix generation dependency for utsrelease.h
we need to explicitly generate kernel.release_full first, before
creating utsrelease.h
Change-Id: I9e25d2511026dba8a5d3135468a53340ee3a59de
Signed-off-by: Yixun Lan <yixun.lan@amlogic.com>
PD#144946: Makefile: generate kernel_full in seperate command
in certain cases/system, echo command will refuse to interpret backslash
without the '-e' option, so it will end up with one line
"4.9.11\n4.9.11-1234-g1234"
this patch will generate KERNELRELEASE & KERNELRELEASE_FULL in seperate file
Change-Id: I51699f193e893f9ea566251d4440edecf3de7ad8
Signed-off-by: Yixun Lan <yixun.lan@amlogic.com>
PD#146107: osd: add cma heap for ion on Android 8.0
Change-Id: I8ebb0c9068a93b9e1bae114a6d5231de5a2ee78f
Signed-off-by: binqi.zhang <binqi.zhang@amlogic.com>
PD#146170: pwm: revert pwm e and f order
1.please ask me to get a patch when you are using pwm on axg.
2.new pwm driver is rebuilding.
3.add pwm e and f for axg,actually the hardware not support.
Change-Id: I3222380b700fbe9668c3d2da1dc8d878f22b30aa
Signed-off-by: Jian Hu <jian.hu@amlogic.com>
PD#146107: osd: add cma heap for ion on Android 8.0
Change-Id: Idf4371fd0dceecc883fd8402a656ed7724be1efd
Signed-off-by: binqi.zhang <binqi.zhang@amlogic.com>
smatch warns about the suspicious formatting in the last line of
open_flags_to_access_mode(). It turns out the only caller was deleted
over a year ago by "ANDROID: sdcardfs: Bring up to date with Android M
permissions:", so we can "fix" the function's formatting by deleting it.
Change-Id: Id85946f3eb01722eef35b1815f405a6fda3aa4ff
Signed-off-by: Greg Hackmann <ghackmann@google.com>
PD#144476: pwm: fix pwm driver errors
1.change pwm e and f order
2.fix several errors
you can request like this:
PWM A 0
PWM B 1
PWM C 2
PWM D 3
PWM AO A 4
PWM AO B 5
PWM AO C 6
PWM AO D 7
Change-Id: Ic1ca7d1f0450db748569447140feab9b9d0c5c34
Signed-off-by: Jian Hu <jian.hu@amlogic.com>
PD#142470: add i2c auto test on axg.
1. if uboot parm(i2c_auto_test) is "open",
open the i2c auto test function.
2. before the test,we should connect i2c slave and master.
Change-Id: Ia37e080dc7995d31e46c9122b389be1ed21acc63
Signed-off-by: Xuhua Zhang <xuhua.zhang@amlogic.com>
These config options are specific to arm64 so should not be universally
required.
Bug: 62523096
Change-Id: Ic5f35db71d73919f2958120f45dd717f5d05f4c5
Signed-off-by: Steve Muckle <smuckle@google.com>
PD#145741: wifi: bcm: fix unable to enter deep suspend mode issue
for the wowlan (wake on wireless lan) function under Linux platfrom,
we have problem to configure it via wpa_supplicant as we do it at Android
platform. so here when it enter into suspend mode, we try to skip calling
cfg80211_leave_all() function at the wire core path.
since this modify the wifi core code, it will also affect other
wifi drivers (not only bcm wifi).
Change-Id: I51eb9b2864afe1bb478ab9411eb7c590f995cd6e
Signed-off-by: Lianghu Su <lianghu.su@amlogic.com>
