Temp sensors, I2C devices, ADC inputs, and keyboard GPIOs are all different.
This removes the things we don't need, connecting some of the things we
will.
BUG=chrome-os-partner:18825
BRANCH=slippy
TEST=none
Doesn't break Link.
Change-Id: Ifb9775d921f6e9952497bffa98d405353ed4df71
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49964
Reviewed-by: Randall Spangler <rspangler@chromium.org>
I'll still need to decide what to do differently for Slippy, but for now
let's just identify the places where there will likely be a difference.
BUG=chrome-os-partner:18825
BRANCH=slippy
TEST=manual
Link still works.
Change-Id: I950f0e5356ccf9838f2140d853122235f884e34f
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49931
1) Port needs to be open-drain. Missed this when porting from STM32F
because open-drain and alternate function are set in the same register
on STM32F and are different regs on STM32L.
2) Queue a stop condition if a transaction failed, so the clock goes
back high.
BUG=chrome-os-partner:18969
BRANCH=none
TEST=i2cscan finds both the PMU at 0x90 and the battery at 0x16
Change-Id: I708b925e4e30da9d5864b74641b1cbe90c9313fe
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49898
Reviewed-by: Bill Richardson <wfrichar@chromium.org>
Also renaming to avoid confusion as to what's being charged.
BUG=chrome-os-partner:18825
BRANCH=slippy
TEST=manual
Build everything, Link still works.
Change-Id: I4205a1210c7dfe57cfbbdd740970ef57e6a011b8
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49775
Reviewed-by: Randall Spangler <rspangler@chromium.org>
Now that I2C works, we can re-enable the charging task to debug
battery charging.
BUG=chrome-os-partner:18657
BRANCH=none
TEST=in EC log, should see battery state messages
Change-Id: I600d0152ba416c13420ce853b55206afba9f6cec
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49649
The state REINIT in TPS65090 charge state machine is more like IDLE0
state in charge_state.h. Rename it so that it's less confusing and
easier to merge the two state machines in the future. Also move the
state name definition to the header file.
BUG=chrome-os-partner:18914
TEST=Boot Spring
BRANCH=None
Change-Id: I116438fedc46ff188dfb6a3964795715b5af4d1f
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49732
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
This eliminates a per-second hook and removes a duplicated ADC read per
second. Also, TSU6721 is now reset after every detachment. This way, we
don't suffer from TSU6721 dirty state (most commonly seen after OTG
dongle detached.)
BUG=chrome-os-partner:17928
TEST=1. Test plugging/unplugging video dongle.
2. Test Toad cable mode switching.
3. Test charging with 200K charger.
BRANCH=spring
Change-Id: Ic035b7332e07ca385d766c735ce39efd31e46034
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49578
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
This fixes a bug in ADC watchdog that sample sequence is not set before
enabling ADC watchdog. Also, current sample time isn't long enough for
some weakly driven net. This extends ADC sample time to 13.5 cycles.
BUG=chrome-os-partner:17928
TEST=Test with Toad cable and video dongle on Spring
BRANCH=spring
Change-Id: Iecdfd4aada4e974225a41144e213e92897f4797b
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49680
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
The voltage rails, inputs, and sequencing is completely different. Easiest
to just handle it separately for each chipset.
BUG=chrome-os-partner:18825
BRANCH=slippy
TEST=manual
Built Link, still works.
Change-Id: Ibf26ef47cdf2284b7bfb3a2e5ccfb6841aba5ac6
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49559
Reviewed-by: Randall Spangler <rspangler@chromium.org>
This puts a copy of test binaries at build/${BOARD}/test-*.bin. This way
it is easier for us to get all test binaries.
BUG=chrome-os-partner:19057
TEST='make tests' and check build/${BOARD}/test-*.bin
BRANCH=None
Change-Id: I8394cea6bee95f00fc01d725d2494ca5e222f47d
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49608
Reviewed-by: Randall Spangler <rspangler@chromium.org>
STM32L doesn't need the DMA-based workarounds needed by STM32F, since
the STM32L I2C block isn't broken. DMA adds a lot of code overhead
when transferring 2-3 bytes, and is implemented differently on STM32F
vs STM32L so it doesn't even work on STM32L
Add a simple polled I2C implementation for STM32L. This is not the
final implementation, which will use interrupts, but for now it works,
unlike the DMA-based version.
BUG=chrome-os-partner:18969
BRANCH=none
TEST=i2cscan on pit finds a device at 0x90
Change-Id: Ie2a6c9ac6f62b7fd3c35e313b4015e080d9f937a
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49555
With a Toad cable, the user can access the EC serial console through the
micro-B connector.
We probably need to de-activate the input on the EC serial console when
the Write-Protect is on, since we have fairly "powerful" commands on the
EC command-line.
Add a new CONFIG_CONSOLE_RESTRICTED_INPUT on platforms with externally
accessible EC serial port.
Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BRANCH=spring
BUG=chrome-os-partner:18716
TEST=on Spring with CONFIG_CONSOLE_RESTRICTED_INPUT set, try with and without
write-protect, use successfully the EC console in the former case, and see
"Console is DISABLED" in the latter case.
Change-Id: Ic9646d5468183f4d8f94b5e5e1d2a727941d7bbe
Reviewed-on: https://gerrit.chromium.org/gerrit/49537
Reviewed-by: Randall Spangler <rspangler@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Vincent Palatin <vpalatin@chromium.org>
Also moves the handy i2cscan command to i2c_common. The
platform-dependent interface is now i2c_xfer().
Still more to do in follow-up CLs; for example, i2c_read_string() has
platform-dependent implementation, and the i2c/i2cread console
commands aren't common yet.
BUG=chrome-os-partner:18969
BRANCH=none
TEST=i2cscan on link, spring
Change-Id: Ia53d57beaa157bece293a4262257e20b4107589e
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49492
Reviewed-by: Simon Glass <sjg@chromium.org>
Commit-Queue: Daniel Erat <derat@chromium.org>
Commit-Queue: Simon Glass <sjg@chromium.org>
LM4 doesn't have implementation of adc_read_all_channels(). Let's use
adc_read_channel() in this case.
BUG=chrome-os-partner:18598
TEST=Build stress test for link. See no error about
adc_read_all_channels() undefined.
BRANCH=None
Change-Id: I5b13384468667cbd17b83faab9f9d3fdc48de91d
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49589
Reviewed-by: Randall Spangler <rspangler@chromium.org>
When battery flags TERMINATE_CHARGE or OVER_CHARGED alarm, we should
treat them as a signal of battery fully charged.
BUG=chrome-os-partner:18914
TEST=On Spring:
1. Plug in adapter when battery if full, see green LED.
2. Plug in adapter when battery is not full, see yellow LED.
BRANCH=spring
Change-Id: Ica414a0e1667b8f30a0cc9a5d66dba1b119a59ba
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49456
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Should use CPRINTF("[%T...\n]") so the messages are timestamped; this
is really helpful to see when things are going wrong.
No functional changes; just changing debug output.
BUG=none
BRANCH=none
TEST=build pit and spring; see prettier debug output
Change-Id: I9c658385b836a184a3ebb84856b844cbfc3224a7
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49551
Reviewed-by: Simon Glass <sjg@chromium.org>
This fixes pmu_init() failing on pit, where the charger task isn't
enabled yet (and thus the charger interrupt is NULL - which can't be
enabled).
BUG=chrome-os-partner:18657
BRANCH=none
TEST=build all platforms; on pit, check that pmu_init() no longer fails
Change-Id: I191bbaeb4df10241e3508ccf7ef5ea83f42c5697
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49550
Reviewed-by: Simon Glass <sjg@chromium.org>
This will let us check the size of static array initializers.
Also moved this macro definition and ARRAY_SIZE into a new "tricks.h"
header, so that userspace utils can use it too.
BUG=none
BRANCH=none
TEST=manual
Built everything, tested on Link. Tried various assertions.
Change-Id: I612891108ea37dbca3572e0f25ab54a7bc0ed860
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49417
Reviewed-by: Randall Spangler <rspangler@chromium.org>
Two states were added, but the descriptions weren't. This caused a crash.
BUG=chrome-os-partner:18914
TEST=boot spring with no battery, wait a few secs; shouldn't crash
BRANCH=spring
Change-Id: I10f9280232259a1f467ea3b02f3b1b61cee57471
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49391
We are moving video power from VFET2 to VFET4. To support old boards, we
need to enable both of them. When new boards are in place, we can then
drop VFET2.
BUG=chrome-os-partner:18186
TEST=Build spring
BRANCH=spring
Change-Id: If0cbc1ac49affc1e3c7ec9650a661f80be826f97
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49431
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Some tests are not applicable to all boards. This change makes test-list
a per-board parameter so a test can be enabled/disabled for individual
boards. Also disable all the tests that don't compile now.
BUG=chrome-os-partner:18598
TEST=make tests for all boards
BRANCH=None
Change-Id: Id2d18e23856f5c64dbdc7c6ca5949f8ad61b5cc0
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49452
We are moving to a per-function mock mechanism, which is less likely to
break when someone adds a function to a file with a mock.
BUG=chrome-os-partner:18598
TEST=None
BRANCH=None
Change-Id: Ie625064730d2224accbaf1febbdc4750e793d047
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49451
If we don't have lid switch, don't test it.
BUG=chrome-os-partner:18598
TEST=Build on mccroskey and see errors about GPIO_LID_OPEN go away.
BRANCH=None
Change-Id: I1084e1bfc2e6bd611dc393c1b2d501e786d2a0e3
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49450
This fills in test parameters for link, pit, snow, and daisy. Also fixes
compilation error due to CONFIG_ADC not defined.
BUG=chrome-os-partner:18598
TEST=Build test-stress. (Link failed due to unrelated issue.)
BRANCH=None
Change-Id: I8a5cfcf4acdae607c23dbe5cf3378945f91c7807
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49449
This creates cscope.files and an emacs TAGS file for the specified BOARD. If
you don't have etags installed, it just creates cscope.files.
BUG=none
BRANCH=none
TEST=manual
make BOARD=link tags
ls
Change-Id: I9b27ba42ac4e85822c5a5a6e8cb5b5148a3435c3
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49356
Reviewed-by: Randall Spangler <rspangler@chromium.org>
Sadly, the existence of fans may not always imply the existence of keyboard
backlights.
BUG=chrome-os-partner:18825
BRANCH=slippy
TEST=manual
Use the Link EC console to make sure that both functions still behave.
faninfo
fanset 4400
faninfo
fanset 9999
faninfo
autofan
faninfo
fanduty 50
faninfo
fanduty 100
faninfo
autofan
kblight 0
kblight 100
kblight 50
kbligth 100
Change-Id: I2e07cd46c21bce2d0d4162275a8ea6ae40135e96
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49355
It will soon have cross-platform code moved to it.
This CL just does the renaming; no functional changes.
BUG=chrome-os-partner:18969
BRANCH=none
TEST=build spring,link
Change-Id: I8cc475f9dd1422a0a9c586c93eb0d7aaf21a001d
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49359
Reviewed-by: Bill Richardson <wfrichar@chromium.org>
I2C communication doesn't work on pit yet, so the charger task is
spewing errors to the console. This change allows the task to be
disabled cleanly on pit, and has no effect on other platforms.
BUG=chrome-os-partner:18657
BRANCH=none
TEST=build all platforms and see that charger task is still compiled
everywhere but pit
Change-Id: I788e817d5630fb1a28694819f4ef9948a503a744
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49344
Reviewed-by: Bill Richardson <wfrichar@chromium.org>
No code changes; this is just moving the irq to the module which
actually handles it. This makes it simpler to disable the charger
task separately from the PMU module.
BUG=chrome-os-partner:18343
BRANCH=none
TEST=build pit, spring
Change-Id: Ife44a17f69bf21aa000d162318a459cea6fdf661
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49334
Reviewed-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
STM32L has 16KB RAM, so can use a bigger output buffer.
BUG=chrome-os-partner:18657
BRANCH=none
TEST=boot pit, look at EC console output; shouldn't drop output
Change-Id: I8a3c7ff2923fd815eff748a898dc873f5d86b46c
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49332
Reviewed-by: Bill Richardson <wfrichar@chromium.org>
This includes two situations:
1. When charge state machine is still trying to decide the next state.
2. When we are waiting for acceptable condition to start charging.
For now, we show yellow for both.
BUG=chrome-os-partner:18914
TEST=none
BRANCH=spring
Change-Id: I5dd97c735a1ecfee29d506452695014c85d2da0b
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49327
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Currently INIT state can mean different things. This change introduces
two new states to demultiplex INIT. Now, REINIT means charge state
machine is deciding the next state. BAD_COND means it's waiting for
acceptable charging condition. Note that the three states are doing the
same thing so there is no functional change.
BUG=chrome-os-partner:18914
TEST=none
BRANCH=spring
CQ-DEPEND=CL:49327
Change-Id: Ia783e76cf27e28103ef44fe7b8a43674dadccc54
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49326
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
We don't have any interrupt handler for I2C1, so enabling of the I2C1
interrupts will just end up triggering a panic everytime the interrupt
fires.
By the way, I2C1 is used as master and all communications happen without
using the I2C interrupt.
Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BRANCH=spring
BUG=chrome-os-partner:18499
TEST=On Spring, run EC flashing in a loop from the AP
while true
do
flashrom --fast-verify -p internal:bus=lpc -w ec.bin
flashrom --fast-verify -p internal:bus=lpc -w ec_autest_image.bin
done
and check that we always succeed and have no panic.
Original-Change-Id: I6b071f309218410840707380d296bdf14b10e9f5
Reviewed-on: https://gerrit.chromium.org/gerrit/47749
Tested-by: Vincent Palatin <vpalatin@chromium.org>
Reviewed-by: Vic Yang <victoryang@chromium.org>
Commit-Queue: Vincent Palatin <vpalatin@chromium.org>
(cherry picked from commit 8b3262d644003b3c9727998dc8d4b0c749450aef)
Change-Id: Ie2fc30e5661e7eade024e6893eaf16427f94cac2
Reviewed-on: https://gerrit.chromium.org/gerrit/49192
Commit-Queue: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>
Reviewed-by: Vic Yang <victoryang@chromium.org>
The power button code is platform-independent. This change splits the
code out of the LM4 switch.c module so that a subseqent change to
STM32 platforms can start using it.
BUG=chrome-os-partner:18945
BRANCH=none
TEST=manual
1. Power+refresh+esc goes to recovery mode,
2. Press power button at recovery screen turns off.
3. With system off, power button turns system on.
4. Press power button for a second; screen locks.
5. Press power button while typing; blocks keystrokes while it's pressed.
6. Hold power button down for 8 sec; system forced to shutdown.
7. From EC console, with system on:
hostevent clear
hostevent -> event 0x04 is clear
press power button
hostevent -> event 0x04 is set
8. From EC console, with system off:
powerbtn -> system turns on
powerbtn 5000 -> system turns off, just like power button was held for 5 sec
Change-Id: If2a9b02514a201e1d03c857d128e2ccab51a16ef
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49217
Reviewed-by: Bill Richardson <wfrichar@chromium.org>
This just copies the board/link stuff into board/slippy (with one formatting
tweak to pass the presubmit check). It builds, but of course is useless.
Customization will follow later.
BUG=chrome-os-partner:18825
BRANCH=slippy
TEST=manual
cd src/platform/ec
make BOARD=slippy
Change-Id: I2ba62812d9199eec776f9da6732921c6281762ab
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49130
Reviewed-by: Randall Spangler <rspangler@chromium.org>
This will allow debugging until the I2C passthru command is supported,
and should be removed once it is supported.
BUG=chrome-os-partner:18778
BRANCH=pit
TEST=build pit; can't test until we have boards
Change-Id: I6285e76763a5f3bfd759161115d4905697b56e6c
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49096
Reviewed-by: Doug Anderson <dianders@chromium.org>
These should be specific to the lm4's lpc.c, not the board.
BUG=chrome-os-partner:18343
TEST=build all, run link
BRANCH=none
Change-Id: Ie02bbd3cf90f09035d302f5dcaad4d33f4ef9e7c
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49101
CONFIG_ macros should be set directly. Expanding the task names in the same
way made it difficult to tell what was a configuration choice and what was
due to changes in ec.tasklist
BUG=chrome-os-partner:18343
TEST=build all, run link
BRANCH=none
Change-Id: Ib82e34f974238ee2dd216f33b701b6f4c6a4f1f1
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/49098
Reviewed-by: Randall Spangler <rspangler@chromium.org>
UART0 is the EC console, and it's consistent.
UART1 is the AP console that we export via servo. It was connected to a
different set of GPIOS on the BDS.
BUG=chrome-os-partner:18343
TEST=build link, bds
BRANCH=none
Change-Id: Ib4c10fd4d2b7a8ffb4e41e216528d4760ba50de3
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/48975
Reviewed-by: Randall Spangler <rspangler@chromium.org>
apreset and apshutdown are now the same across all platforms.
BUG=chrome-os-partner:18343
BRANCH=none
TEST=build all platforms; make sure 'apreset warm', 'apreset cold', and
'apshutdown' work.
Change-Id: I4ace1ab2bbda4d400528039ad5f84da63c9c7d0e
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/48942
Previously, we sometimes saw battery LED went off when going into or out
of battery assist mode. This CL adds a state machine to stabilize
battery LED. By controlling the state of lighting engine, the
tranisition from/to battery assist mode is seamless now.
BUG=chrome-os-partner:18844
TEST=On spring, check:
1. LED doesn't go off when going into battery assist mode.
2. LED doesn't suddenly change brightness when coming out of
battery assist mode.
3. LED goes red after few seconds when battery is not present.
4. LED goes green when battery is charged.
BRANCH=spring
Signed-off-by: Vic Yang <victoryang@chromium.org>
Change-Id: I3286b54a76edd77c65033ec3bbf0d81d916f7be0
Reviewed-on: https://gerrit.chromium.org/gerrit/48875
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Vic Yang <victoryang@chromium.org>
Tested-by: Vic Yang <victoryang@chromium.org>
