diff options
| author | Martin Roth <martinroth@chromium.org> | 2020-10-28 11:52:30 -0600 |
|---|---|---|
| committer | Felix Held <felix-coreboot@felixheld.de> | 2020-11-06 18:53:05 +0000 |
| commit | 5632c6bc020b02868e21714ea990d550e44e91d3 (patch) | |
| tree | 3997e0102f6449c38cdd27a14ba04346b22e42fd /src/soc/amd/picasso | |
| parent | 03d06d3bcd368a8e635e3279d4f17336f09adef8 (diff) | |
| download | coreboot-5632c6bc020b02868e21714ea990d550e44e91d3.tar.xz | |
soc/amd/picasso: Set vboot hashing block size to 36k
On picasso's psp_verstage, the vboot hash is being calculated by
hardware using relatively expensive system calls. By increasing the
block size, we can save roughly 150ms of boot and S3 resume time.
TEST=Build & boot see that boot time has decreased.
BRANCH=Zork
BUG=b:169217270 - Zork: SHA calculation in vboot takes too long
Signed-off-by: Martin Roth <martinroth@chromium.org>
Change-Id: I6642073357327811b415dcbcad6930ac6d2598f9
Reviewed-on: https://review.coreboot.org/c/coreboot/+/46903
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Kangheui Won <khwon@chromium.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Diffstat (limited to 'src/soc/amd/picasso')
| -rw-r--r-- | src/soc/amd/picasso/Kconfig | 16 |
1 files changed, 16 insertions, 0 deletions
diff --git a/src/soc/amd/picasso/Kconfig b/src/soc/amd/picasso/Kconfig index bae2d2868a..5ae0a2a58f 100644 --- a/src/soc/amd/picasso/Kconfig +++ b/src/soc/amd/picasso/Kconfig @@ -535,6 +535,22 @@ config VBOOT_STARTS_BEFORE_BOOTBLOCK Runs verstage on the PSP. Only available on certain Chrome OS branded parts from AMD. +config VBOOT_HASH_BLOCK_SIZE + hex + default 0x9000 + depends on VBOOT_STARTS_BEFORE_BOOTBLOCK + help + Because the bulk of the time in psp_verstage to hash the RO cbfs is + spent in the overhead of doing svc calls, increasing the hash block + size significantly cuts the verstage hashing time as seen below. + + 4k takes 180ms + 16k takes 44ms + 32k takes 33.7ms + 36k takes 32.5ms + There's actually still room for an even bigger stack, but we've + reached a point of diminishing returns. + config CMOS_RECOVERY_BYTE hex default 0x51 |