diff options
author | Wai-Hong Tam <waihong@google.com> | 2019-06-13 13:32:28 -0700 |
---|---|---|
committer | Commit Bot <commit-bot@chromium.org> | 2019-07-19 02:25:54 +0000 |
commit | 1f14229fa7e499dfcee07d17add187598ff0a46c (patch) | |
tree | b8cbecc3cbe7137e00d64a9be83ced76378d9cbe /board/cheza | |
parent | d5a4014a7c9331e099d0eaeb8974e53841395ea3 (diff) | |
download | chrome-ec-1f14229fa7e499dfcee07d17add187598ff0a46c.tar.gz |
servo_v4: Fake CC2 voltage when servo v4 as snk
In servo v4 hardware logic, both CC lines are wired directly
to DUT. When servo v4 as a snk, DUT may source Vconn to CC2
and make the voltage high as vRd-3.0, which makes the PD
state mess up. As the PD state machine doesn't handle this
case. It assumes that CC2 is separated by a Type-C cable,
resulting a voltage lower than the max of vRa.
It fakes the voltage within vRa so the PD state machine
checks the value as expected.
This is an issue only happening on servo v4 as it wires both
CC lines to DUT.
BUG=b:134700685
BRANCH=servo
TEST=Servo v4 as snk, verified the CC2 is sensed vRa, i.e.
the output "CC1:0" (TYPEC_CC_VOLT_OPEN), instead of "CC1:7"
(TYPEC_CC_VOLT_SNK_3_0).
2019-06-18 15:05:47 > cc snk
2019-06-18 15:05:50 cc: on
2019-06-18 15:05:50 dts mode: off
2019-06-18 15:05:50 chg mode: off
2019-06-18 15:05:50 chg allowed: off
2019-06-18 15:05:50 > C1 st2 SNK_DISCONNECTED
2019-06-18 15:05:50 C1 st3 SNK_DISCONNECTED_DEBOUNCE
2019-06-18 15:05:50 C1 st5 SNK_DISCOVERY
2019-06-18 15:05:51 > tcpc 1 state
2019-06-18 15:05:53 Port C1, Dis - CC:2, CC0:6, CC1:0
2019-06-18 15:05:53 Alert: 0x00 Mask: 0x007d
2019-06-18 15:05:53 Power Status: 0x48 Mask: 0x00
TEST=Ran the PD FAFT test firmware_PDConnect passed.
Change-Id: I10f1ffe80768100ee3ed4c374598df7c2f9a8d05
Signed-off-by: Wai-Hong Tam <waihong@google.com>
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/ec/+/1666468
Reviewed-by: Aseda Aboagye <aaboagye@chromium.org>
Diffstat (limited to 'board/cheza')
0 files changed, 0 insertions, 0 deletions