diff options
author | Bill Richardson <wfrichar@chromium.org> | 2016-09-21 16:58:26 -0700 |
---|---|---|
committer | chrome-bot <chrome-bot@chromium.org> | 2016-09-24 16:22:14 -0700 |
commit | 4250da3ba977fd6bfb80b1bf6086cd142c461e5a (patch) | |
tree | 486186f63516d68ef1f7738523a640727fb306b7 /PRESUBMIT.cfg | |
parent | e29005abe7497988e14e0ef87f602a5417d003e6 (diff) | |
download | chrome-ec-4250da3ba977fd6bfb80b1bf6086cd142c461e5a.tar.gz |
Cr50: Add mostly-synchronous tpm_reset() function.
To reset the TPM task, we send it an event so that it will reset
only when it's not busy doing actual TPM stuff that might fiddle
with the stack or shared memory. But that means that we can't
always know when the task finally gets around to resetting
itself.
This CL adds a tpm_reset() function that blocks until the reset
actually occurs. Obviously it can't do that if it's being called
in interrupt context or from the TPM task itself, but otherwise
it does.
BUG=chrome-os-partner:52366
BRANCH=none
CQ-DEPEND=CL:361680
TEST=make buildall, test on Gru, manual tests
In addition to the normal rebooting, logging in/out, and so
forth. I added a temporary console command to call tpm_reset()
and scattered a bunch of ccprintfs around it. When called due to
SYS_RST_L, it didn't block. When invoked with the console
command, it did.
Change-Id: I51e8b1299dbdcd1a12273cf48a890e93ed32a8c8
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/388125
Reviewed-by: Randall Spangler <rspangler@chromium.org>
Diffstat (limited to 'PRESUBMIT.cfg')
0 files changed, 0 insertions, 0 deletions