diff options
author | Matt DeVillier <matt.devillier@gmail.com> | 2019-08-02 20:16:36 -0500 |
---|---|---|
committer | Martin Roth <martinroth@google.com> | 2019-08-04 15:13:19 +0000 |
commit | 4183312cec55f00fe22c4dbfd682376e521fc6d3 (patch) | |
tree | 080d79ca4a7b0fdbb88f522501c687deff243fde /src/soc/intel/cannonlake | |
parent | bccd2b6c492ee597d6cfebc4b6ec21ebef7252c6 (diff) |
drivers/fsp1_1/raminit: fix use of mrc_hob
Commit 509f469 [drivers/fsp1_1/raminit.c: Always check FSP HOBs]
inadvertently made use of the mrc_hob conditional on
CONFIG_DISPLAY_HOBS, when there is no relation between the two,
leading to MRC cache data being corrupted. On some devices this
caused RAM training to be redone, on others it resulted in a
bricked device.
Fix this by removing the condition on CONFIG_DISPLAY_HOBS.
Test: boot google/{cyan,edgar}, observe third boot and onward do not
brick device, properly use mrc_hob via cbmem console and timestamps.
Change-Id: I01f6d1d6dfd10297b30de638301c5e0b6545da9c
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34685
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Diffstat (limited to 'src/soc/intel/cannonlake')
0 files changed, 0 insertions, 0 deletions