snappy: support lid accel matrix by board version.
As the new form-factor has the lid accelerometer on the
reversed side facing the B-cover, the matrix setting
depending on board version; in such matter, it should be
able to compatible with old version of boards.
We create a new hook function for board specific tweaks,
this is because since the commit of 0c57824
("reef: Re-factor PP5000 and PP3300 enable/disable"),
the board_init() is no longer a good place for tweaks,
because ADC read should come after adc_init();
such that, new hook ensures robust ADC reading which is
the source of board version.
Also, we fix an arithmetic error for version-3 workaround,
i.e. patch the commit of ca99f38 ("snappy: BMI160 is
powered down on board v3 and older in S3"), else it could
trigger unexpected EC panic like this:
[89.770776 chipset -> S3]
[89.771222 power state 2 = S3, in 0x006d]
[89.772428 I2C unwedge failed, SCL is being held low]
[89.773775 TCPC p0 Low Power Mode]
[89.812962 Reset i2c 01 fail!]
...snip...
[91.816415 Unexpected i2c state machine! 1]
Time: 0x00000000057a7d9c us, 91.913628 s
Deadline: 0x00000000057a8a1d -> 0.003201 s from now
...snip...
Rebooting...
--- UART initialized after reboot ---
[Reset cause: soft]
...snip...
BUG=chrome-os-partner:62676
BRANCH=reef
TEST=check the DVT1 and DVT2 unit rotate normally.
Change-Id: Ic53e67e0c97e57056587adb6b260e81c0f99437a
Signed-off-by: Bruce.Wan <Bruce.Wan@quantatw.com>
Reviewed-on: https://chromium-review.googlesource.com/442252
Commit-Ready: Chen Wisley <wisley.chen@quantatw.com>
Tested-by: Bruce Wan <Bruce.Wan@quantatw.com>
Tested-by: Harry Pan <harry.pan@intel.com>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
1 file changed
