Now that when ID_MUX=1, DP_SNS can be connected to either ID_OUT or
CABLE_DET, we need to handle the case where a video dongle is not
reflected by DP_SNS going low. This is done by leaving ID_MUX=1 for USB
host and monitor its detachment by sensing VBUS. As for unpowered
dongle, we just ignore it when it's not recognized.
Note that due to the polarity of CABLE_DET, we now sense dongle
detachement by DP_SNS < 0.25V. To support older boards with ID_OUT
connected, we also disconnect video dongle on system suspend and
shutdown.
BUG=chrome-os-partner:19911
TEST=Powered/unpowered video dongle detected correctly.
TEST=Add a console command to simulate CABLE_DET being high. With that,
check battery charges with powered video dongle. Check nothing happens
with unpowered video dongle.
TEST=Check video dongle considered disconnected when OTG dongle plugged
in or system shutdown.
BRANCH=spring
Change-Id: If7b530b67c98c85017ca663d43c5148f0eb9163c
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58070
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Partners are adamant that this needs doing. Whatever...
BUG=chrome-os-partner:19868
BRANCH=none
TEST=none
You can run the "charger" command on the EC console to see that the option
bits have been changed. I couldn't reproduce the original complaint, so I
don't know if this solves it.
Change-Id: I428697f69a7ba1b360e2acb123b42ed41244ebc5
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58576
We've been providing the battery's max voltage and the charger's minimum
current to try to awaken a deep-discharged battery. For some batteries, that
doesn't appear to be enough. This change will use the battery's
preconfigured precharge_current value instead.
BUG=chrome-os-partner:20142
BRANCH=none
TEST=manual
I don't have a deep-discharged battery to try it on. I've been using these
values manually in order to get the battery away. With no battery connected,
you can run the "charger" command on the EC console to see what values it's
using. They should match what the battery wants.
Change-Id: I16d06011103ba70682397859d9844a37938d3e90
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58575
Frank's reworks turn AC_PWRBTN_L from a signal which co-mingled
AC-detect and pwower button press to a straight AC detect signal.
Switch from the hacky snow version of extpower to the simple gpio
version.
BUG=chrome-os-partner:20034
BRANCH=none
TEST=manual, using gpioget command
1. With AC plugged in, AC_PRESENT=1
2. Press power button. Still AC_PRESENT=1
3. Unplug AC, AC_PRESENT=0
4. Press power button. Still AC_PRESENT=0
Change-Id: I35df4ba80d0003310f3e01d9815158cfa3d0cb51
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58550
Reviewed-by: Bill Richardson <wfrichar@chromium.org>
We've been asked to disable the IFAULT_HI protection for the Falco battery
charger, with no explanation. This change adds the code to do that, but
leaves it ifdef'd out. If/when we get confirmation that we really want to do
that, we can.
BUG=chrome-os-partner:19868
BRANCH=none
TEST=none
No functional change yet, nothing to test.
Change-Id: I02a71461f55849a2fdb5ec220fe00e3c812ddf0b
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58467
This adds the BQ24738 smart battery charger, and a placeholder for the Falco
battery pack. I don't have either documentation or a battery to use to test,
so the battery pack stuff is just a guess (see crosbug.com/p/20142).
BUG=chrome-os-partner:20098
BRANCH=none
TEST=none
Well, if you like, from the EC console, run "charger". It should say
something like this:
> charger
Name : bq24738
Option: 1111100100010010 (0xf912)
Man id: 0x0040
Dev id: 0x000f
V_batt: 0 (1024 - 19200, 16)
I_batt: 0 ( 128 - 8128, 64)
I_in : 3968 ( 128 - 8064, 128)
>
But since I don't have either a battery or a spec, I had to guess at the
battery configuration. To test the charger, we kind of need a battery.
Change-Id: I6e63d6b5aa8be4ba15e2c427d2e86364ef6251b3
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58466
This measures the analog voltage from the AC adapater's ID pin. The values
don't seem to match what we expected, so there's more work to come. Still,
here it is.
BUG=chrome-os-partner:19594
BRANCH=none
TEST=manual
From the EC console, run "adc". You should see a line like this:
ADC channel "AdapterIDVoltage" = 553
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Change-Id: Idf84ceb6cd85d66d879f07bf4f26881ec078f638
Reviewed-on: https://gerrit.chromium.org/gerrit/58465
I'm still not convinced that the Slippy battery voltages are right for all
cases, but these are closer to what seems to be working. It matches what the
battery we've tested with seems to want.
BUG=chrome-os-partner:18825
BRANCH=none
TEST=none
Change-Id: I2d86c5ef39a70d3826fec31207250617596baa33
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58415
BUG=chrome-os-partner:18788
BRANCH=none
TEST=none
The connection to the LVDS controller is unstuffed, so there's nothing to
test without some soldering.
Change-Id: I5e6d9a3705554335512611b6b463568251b848b6
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58268
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
I don't think there are any power-sequencing changes.
BUG=chrome-os-partner:18788
BRANCH=falco
TEST=none
Nothing to test yet.
Change-Id: I2d0b7085fb5a5b853385049b61fe7796709702ae
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/56137
The new constants used to convert PWM duty cycle to input current,
based on a linear regression done on Aaron's characterization data
measured on DVT1 PCB.
The data points are linear enough to use the linear relationship
to set the current limitation.
Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BRANCH=spring
BUG=chrome-os-partner:19281
TEST=none
Original-Change-Id: I8378aaea21d5b3229d505caf4849257ded77e1ad
Reviewed-on: https://gerrit.chromium.org/gerrit/58143
Reviewed-by: Vic Yang <victoryang@chromium.org>
Commit-Queue: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>
(cherry picked from commit d7b9b2a088dcf129739c7a6b8a30bc292ea7234d)
Change-Id: I11962991d6d7ba7b9d437fc56eb23974d30930a8
Reviewed-on: https://gerrit.chromium.org/gerrit/58198
Commit-Queue: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>
Reviewed-by: Vic Yang <victoryang@chromium.org>
Implement the EC hibernate mode by using the stm32f100 standby low power
mode.
As we cannot de-activate the watchdog during long hibernation, the
following workaround is implemented:
we are woken-up once by the watchdog and go back to hibernate if we
detect that condition.
Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BRANCH=spring
BUG=chrome-os-partner:19595
TEST=on Spring with rework on the EC wake-up pin,
type "hibernate 10" and see the EC console going blank for 10s, then
booting with reset cause equals to "hibernate".
Press Alt+VolUp+H, then wake-up the system by pressing power key.
Change-Id: I28150e69817ae80314f52977ec6b62750017c2c4
Reviewed-on: https://gerrit.chromium.org/gerrit/58086
Tested-by: Vincent Palatin <vpalatin@chromium.org>
Reviewed-by: Randall Spangler <rspangler@chromium.org>
Reviewed-by: Vic Yang <victoryang@chromium.org>
Commit-Queue: Vincent Palatin <vpalatin@chromium.org>
If the I2C transaction fails, we shouldn't cache the status because that
status is not set correctly in LP5562.
BUG=chrome-os-partner:20020
TEST=Boot and check battery LED still works.
BRANCH=spring
Change-Id: I3f40c2d5c85db41e4ba4b80dc5252e5d86100823
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58072
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Sometimes initialization may fail due to bad I2C bus status. Let's allow
for maximum 3 tries of initialization 500ms apart from previous attempt.
BUG=chrome-os-partner:20020
TEST=Boot and check device type detection still works.
BRANCH=spring
Change-Id: I6ccedf77c92c4b6014ca24c7a63534316eaa7b6a
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58071
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
This creates a separate thread that keeps reading characters from stdin
and feed to UART process.
BUG=chrome-os-partner:19235
TEST=Start up a emulator and type 'help'. See command list.
TEST=Pass all tests for 300 times.
BRANCH=None
Change-Id: I190c1d939b0b4ad0f8f0517d8d7b06f2f3df3832
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57866
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
This reverts commit c848e32161.
Too noisy. It was supposed to only show it once, but it spews constantly
instead. Debug it later - just remove it for now.
BUG=chrome-os-partner:20057
BRANCH=none
TEST=none
Change-Id: Ie6dfd4ccb84a612bb72a0d2b758361a13644e4d9
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58111
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Currently, we need to have #ifdefs everywhere watchdog_reload() is
called. With this fix we don't.
Also don't bother including unused hardware timer watchdog code if the
watchdog isn't defined.
No change when CONFIG_WATCHDOG is defined (which it is for all normal builds).
BUG=chrome-os-partner:20056
BRANCH=none
TEST=build all platforms with CONFIG_WATCHDOG commented out in config.h
Change-Id: Id3ce33af1a497eda127a4892e13651d9d2534d92
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58094
Reviewed-by: Simon Glass <sjg@chromium.org>
For simplicity of the code, we don't allow (head == tail) when the queue
is full. But currently we are wasting the size of a single unit, while
we can actually just waste 1 byte. Let's fix this.
BUG=None
TEST=Pass the unit test.
BRANCH=None
Change-Id: Id09c20a9345ebb3ec0cad659afe36e25b422e688
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58058
Reviewed-by: Yung-Chieh Lo <yjlou@chromium.org>
Current implementation of queue_has_space doesn't handle the case where
queue tail has wrapped around the end. This CL fixes the bug.
BUG=None
TEST=Pass the test in the following CL.
BRANCH=link
Change-Id: I774f388081af50f0e930a6cbc3a723da1c8283b0
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/58031
Reviewed-by: Yung-Chieh Lo <yjlou@chromium.org>
The IOUT pin of the smart battery charger can be used to monitor the AC
adapter current (default) or the battery charging current.
BUG=none
BRANCH=none
TEST=manual
Discharge the battery a bit, and connect to the EC console. With the AC
power plugged in, the "battery" command should show charging status,
including current.
The "adc" command will display the A-D converters, including the current
measurement. For example:
> battery
Temp: 0x0b88 = 295.2 K (22.1 C)
Manuf: SMP-COS20
Device: OC2
Chem: LION
Serial: 0x0005
V: 0x4130 = 16688 mV
V-desired: 0x41a0 = 16800 mV
V-design: 0x39d0 = 14800 mV
I: 0x008e = 142 mA(CHG)
I-desired: 0x0080 = 128 mA
Mode: 0x6001
Charge: 98 %
Abs: 94 %
Remaining: 1871 mAh
Cap-full: 1923 mAh
Design: 2000 mAh
Time-full: 0h:23
Empty: 0h:0
>
> adc
ADC channel "ECTemp" = 317
ADC channel "ChargerCurrent" = 455
>
That current is significantly higher than the "I:" reported by the "battery"
command. But look at the charger options:
> sbc 0x12
0x7904 (30980)
>
Bit 5 controls the IOUT Selection. When clear, it monitors the current from
the AC adapter. Set bit 5 to monitor the current provided to the battery:
> sbc 0x12 0x7924
> adc
ADC channel "ECTemp" = 318
ADC channel "ChargerCurrent" = 128
>
That matches what the smart battery sees.
Change-Id: I2fe351304421dfb22d83ef13d416aa44c9f56e8a
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57940
Reviewed-by: Randall Spangler <rspangler@chromium.org>
The ADC pins were being misconfigured for quite a while now. Let's fix 'em.
BUG=none
BRANCH=link
TEST=none
The effect was that the adc values that measure GPIO pins were less
accurate. Since no one noticed, there's probably not much that can
be done to test it.
Change-Id: I9def4eb95c89f72e5df0721e6153c8caa3ef8e2a
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57939
Reviewed-by: Randall Spangler <rspangler@chromium.org>
This adds the initial support for Slippy's battery. The data I have is
unclear and incomplete, so this is NOT the final form. It seems to work
right now, and hasn't caught fire or anything, but it will need futher
tweaks.
BUG=chrome-os-partner:19976
BRANCH=none
TEST=manual (and watch it!)
Connect the EC console and watch what happens. You should see the battery
charging, discharging, etc. Keep an eye on it, though, and never leave it
unattended when on AC - we don't have the full data sheets available yet.
Change-Id: Id9bf93dc04a1399a9cdbc2156b3fac74be62038f
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57814
Reviewed-by: Randall Spangler <rspangler@chromium.org>
The sbc command can be used to get/set registers from the smart battery
controller. The sb command already does that for the smart battery itself.
BUG=none
BRANCH=none
TEST=manual
Try it out.
Change-Id: Idaea451e58988ab2d6bc40164721cb5577d903af
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57813
Reviewed-by: Randall Spangler <rspangler@chromium.org>
This is very similar to the BQ24725. There are just enough differences to
require a separate file.
BUG=chrome-os-partner:19976
BRANCH=none
TEST=none
Nothing to test until it's enabled.
Change-Id: I3247fcfde93ac75f5f9790acadc7feca28038608
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57811
Reviewed-by: Randall Spangler <rspangler@chromium.org>
Just tweaking the build.mk file so that other batteries can be specified
for use by the same charging logic.
BUG=none
BRANCH=none
TEST=none
Change-Id: I01b39a6ffc4091b9b7824cf1759b36168f6f5701
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57810
Reviewed-by: Randall Spangler <rspangler@chromium.org>
Reading from a nonexistant/nonresponsive smart battery fails (as it should).
But then it leaves the i2c clock line low, so that subsequent i2cscan
commands fail too.
This change just sends a STOP to clear the bus when an i2c error occurs.
BUG=chromium:247037
BRANCH=none
TEST=none
Change-Id: I599bed7149ed3dd3748f1a939c2ea8fdf65e3d72
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57808
Reviewed-by: Randall Spangler <rspangler@chromium.org>
This is preparation for the common userspace EC interface. If/when that
appears, this will be ready.
BUG=chromium:239197
BRANCH=all
TEST=manual
Build, install, run it. Shouldn't be any change.
Change-Id: I9fa78515ec5443ba659f10a66bbaadcb7f4802b0
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/56131
Leaving the PLL turned off saves about 0.5 mW (and also presumably
speeds up boot a tiny amount, since we don't need to wait for the PLL
to lock).
BUG=chrome-os-partner:19951
BRANCH=none
TEST=system boots normally. power consumption on P1.8V_MICOM rail is lower.
Change-Id: I93dc1efabbf6c6b0b5a43ffebe1068d18c689bef
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57645
Reviewed-by: Bill Richardson <wfrichar@chromium.org>
This costs 232 bytes RAM; we still have over 6KB free.
BUG=chrome-os-partner:19945
BRANCH=none
TEST=taskinfo shows timing and interrupt counts
Change-Id: I1c4e4eac44b48c66b075b8ea209b4748c55b7fcb
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57511
The two I2C commands for battery cut-off must be sent out back to back.
Thus we need to guard them with I2C port lock to prevent being
preempted.
BUG=chrome-os-partner:19901
TEST=Check battery cutoff still works.
BRANCH=spring
Change-Id: Iac51037432b108d4cac29d5c73cafa9ce2310b12
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57598
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Reviewed-by: Randall Spangler <rspangler@chromium.org>
On Spring, we now have commands 'i2c' and 'i2cscan'. Currently if we
type 'i2c', it's rejected as it's also the prefix of 'i2cscan'. Since
'i2c' is a full match of a legal command, we should accept it.
BUG=None
TEST=On Spring, check 'i2c' invokes 'i2c' command, and 'i2cs'/'i2cscan'
invokes 'i2cscan' command. Also check 'i2' is still rejected.
BRANCH=all
Change-Id: I65c4c148a5a3e9b025554fa8165ba76da7bc312f
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57576
Reviewed-by: Randall Spangler <rspangler@chromium.org>
Sometimes TSU6721 falls into a weird state where video dongle is
recognized as non-standard charger. Let's reset TSU6721 when EC boots,
so that we are guaranteed to have a clean state after initialization.
BUG=None
TEST=Keep doing Power+F3 reset. Doesn't see weird state anymore.
BRANCH=spring
Change-Id: Id09bb1721ae79804dc9b3300a2f3def850c2f70a
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57575
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
This is needed to properly notify kernel of power changes.
BUG=chrome-os-partner:19925
TEST=Attach/remove power from video dongle, and see device type changes.
BRANCH=spring
Change-Id: Ic91ad43ed934be021689c4c4557914e6163e06f8
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57569
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
The WAKE# pin was being driven low when the power button was
pressed and no other events were occuring. This causes a PCIE
wake event to be observed on the host. This is incorrect. Therefore
only assert the WAKE# pin when any other event but the power button
has occured.
The implementation introduces a board-specific callback,
board_process_wake_events(), which handles the specific logic
for the wake events.
BUG=chrome-os-partner:19810
BRANCH=None
TEST=Manual. Both power button wakeup and lid events were tested. The
presence of PCIE Wake no longer exists for S5->S0 transitions.
Change-Id: If1311ccc36629b04d2d9e021c3e103e379836a3a
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/56970
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Hard-reset triggers a pulse on the PMIC_RESET pin but this is hard to
see when debugging. By printing error message explicitly before
resetting the board, it's easier to tell why the board is resetting.
BUG=chrome-os-partner:19778
TEST=Trivial change. Build and boot on Spring. Make sure system is
alive.
BRANCH=spring
Change-Id: I26b749af2f8760339373c3e4db46c59d7b0d039e
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57101
Reviewed-by: Randall Spangler <rspangler@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
If EC-RO protects itself via the WRPx option bytes, flash_ec will fail
to reprogram it. And if EC-RO is broken/unbootable, it won't be able
to clear the WRPx option bytes itself when it sees the write protect
GPIO is not asserted.
Add the -u flag to stm32mon so that it will automatically attempt to
unprotect the flash via the option bytes.
BUG=chrome-os-partner:19723
BRANCH=none
TEST=enable write protect GPIO. Then from console, 'flashwp enable'.
Then from the chroot:
./util/flash_ec --board=pit --ro --image=build/pit/ec.RO.flat
That should successfully reflash the board.
Change-Id: Ic356d29104294d04d5e2a36dc75fd0206b89bd34
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/56659
The firmware defines had two almost-identical sets. Coalesce into one
consistent set.
Link had 256 KB flash, but only allowed 2 80KB images. Future
LM4-based platforms (slippy/peppy/falco/etc) will now use the entire
flash, with RO=124KB, pstate=4KB, RW=128KB. This matches what the
STM32 platforms do, where pstate is contiguous with the RO firmware.
No functional change to STM32-based platforms.
BUG=chrome-os-partner:19176
BRANCH=none
TEST=build all platforms and dump_fmap ec.bin.
- stm32-based platforms should report RO=61440@0, RW=65536@0x10000
- link should report RO=81920@0, RW=81920@0x14000
- slippy should report RO=129024@0, RW=131072@0x20000
Change-Id: I20b1d95c16250d9a5d228ead06eef03d96548823
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/56655
This adds the option to specify which LED to control as well as the
ability to query the supported LED color on the board.
BUG=chrome-os-partner:19745
TEST=On Spring:
- ectool led 0 query -> See the max value for R, G, Y is 0x80.
- ectool led 1 query -> See error message.
- ectool led 0 yellow -> See LED turns yellow.
- ectool led 0 green=0x40 red=0x40 -> See green and red lit up.
- ectool led 0 auto -> See LED turns off (without charger.)
BRANCH=spring
Change-Id: Ibdde2f7450122f59383dad1030a0a2a985386f73
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/56877
Haswell devices have EC control of the WWAN power rail.
Expose a new wireless switch enable flag for this under
the existing wirless enable command.
This change also abstracts the wireless enable function
to call a per-board handler so the different boards can
do the right thing based on their GPIO setup.
The haswell boards will switch WLAN radio and WWAN power
rails based on the switch inputs. These boards do not have
EC control of bluetooth radio/rail power.
WLAN (power and radio) still defaults to enabled. Disabling
with ectool will turn off the radio but keep the power enabled
in order to prevent the PCIe device from disappearing.
WWAN (power) still defaults to disabled. Disabling with
ectool will turn off the power rail.
BUG=chrome-os-partner:19871
BRANCH=none
TEST=manual: boot on slippy
DEFAULT:
> ectool gpioget pp3300_wlan_en
GPIO pp3300_wlan_en = 1
> ectool gpioget wlan_off_l
GPIO wlan_off_l = 1
> ectool gpioget pp3300_lte_en
GPIO pp3300_lte_en = 0
ENABLE WWAN:
> ectool wireless 0x7
Success.
> ectool gpioget pp3300_lte_en
GPIO pp3300_lte_en = 1
DISABLE WLAN (radio):
> ectool wireless 0x7
Success.
> ectool gpioget pp3300_wlan_en
GPIO pp3300_wlan_en = 1
> ectool gpioget wlan_off_l
GPIO wlan_off_l = 0
Change-Id: I6f760b8cf5ab47d8f7f0dd8cd4d3e6563464043e
Signed-off-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/57215
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
These were tested to make sure they are not responsible for
bricking boards.
The flash code is necessary for software sync, and it is
essential to be able to update boards without servo.
The hibernate setup code was breaking jumping between images
and the passing of data between the images -- needed for
SW SYNC but also needed for things like "reboot ap-off".
BUG=chrome-os-partner:19366
BRANCH=none
TEST=manual: emerge-slippy chromeos-ec, flash+boot, update again via SW Sync
Change-Id: Ib1363b353f18b21e0cad3209783e1e0b4ff24e86
Signed-off-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/56820
If the user plug in the charger really slowly, VBUS and GND will make
contact before ID pin. Currently timer is set to fire in 0.6 seconds,
which means ID pin need to make contact within 0.6 seconds after VBUS
make contact. This is causing chargers to be recognized as non-standard
charger. Let's extend this to 3.6 seconds.
BUG=chrome-os-partner:19765
TEST=Plug in a charger half-way through and see device type 0x20000.
Wait for 3.6 seconds, and see it changes to 0x60000.
TEST=Plug in a charger half-way through, wait for 2 seconds, and then
plug it in all the way. See device type 0x20010.
TEST=Plug in a charger to the end. See device type 0x20010 immediately.
BRANCH=spring
Change-Id: I46c408e7bc733fe4f3655db253f4065a4f12ee1a
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/56772
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
If the AP is already on, the kernel should handle low-power event. We
shouldn't power off the system on sysjump.
BUG=chrome-os-partner:18318
TEST=None
BRANCH=spring
Change-Id: I4e80c61a25d2fa503d0c97e22dc2f4ad9c44f716
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/56706
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
