diff options
| author | Randall Spangler <rspangler@chromium.org> | 2012-12-06 15:46:28 -0800 |
|---|---|---|
| committer | Gerrit <chrome-bot@google.com> | 2012-12-06 17:58:32 -0800 |
| commit | c26a242e947a110ea3cf0c43d8274eebb9aaeb5a (patch) | |
| tree | 14fb9861e92901091174eeb54e8d7ce51e3f6278 | |
| parent | 3d0d2adb63b9ca148216d866c772ea0b6e5e2598 (diff) | |
| download | chrome-ec-stabilize2.tar.gz | |
Increase cold reset hibernate time to 200msstabilize2
The hardware fix for issue 16600 adds RC delays to ENABLE_5VALW, so
that +5VALW stays on during a warm reset of the EC. In the worst
case, +5VALW will drop around 150ms, which could then move the +3VALW
glitch right into the time frame where the EC is booting following
hibernate.
Increase the cold reset hibernate time from 150ms to 200ms. This
ensures that +5VALW has dropped before the EC comes out of hibernate.
BUG=chrome-os-partner:16600
BRANCH=link
TEST=manual
From the EC console, 'reboot cold' a bunch of times. The system
shouldn't hang. (Alternately, you can 'ectool reboot_ec cold' a bunch
of times)
Change-Id: I4bebdb552b8e917c6345badd6efb68b10d7d1f86
Signed-off-by: Randall Spangler <rspangler@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/39340
Reviewed-by: Bill Richardson <wfrichar@chromium.org>
| -rw-r--r-- | chip/lm4/system.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/chip/lm4/system.c b/chip/lm4/system.c index b6e37ca0ec..bbe70e6793 100644 --- a/chip/lm4/system.c +++ b/chip/lm4/system.c @@ -35,9 +35,9 @@ enum hibdata_index { /* * Time to hibernate to trigger a power-on reset. 50 ms is sufficient for the * EC itself, but we need a longer delay to ensure the rest of the components - * on the same power rail are reset. + * on the same power rail are reset and 5VALW has dropped. */ -#define HIB_RESET_USEC 150000 +#define HIB_RESET_USEC 200000 /** * Wait for a write to commit to a hibernate register. |