diff options
| author | Julius Werner <jwerner@chromium.org> | 2019-03-28 18:01:26 -0700 |
|---|---|---|
| committer | Patrick Georgi <pgeorgi@google.com> | 2019-04-01 07:56:48 +0000 |
| commit | 9993b6f0b52257b05ccf5d6dab658c3bfe17e372 (patch) | |
| tree | 02c5e759b2318a083398e00c97865f11f0aa431a /src/soc | |
| parent | cd429a8b0c8f46be00e2b147465eec513a2559e9 (diff) | |
| download | coreboot-9993b6f0b52257b05ccf5d6dab658c3bfe17e372.tar.xz | |
vboot: Select CONFIG_VBOOT_OPROM_MATTERS in more cases
This patch enables CONFIG_VBOOT_OPROM_MATTERS in a few more cases where
I think(?) it should be. Haswell, Broadwell and Baytrail Chromebooks
have this enabled in their old depthcharge firmware branches -- we
presumably just forgot to move it over when vboot2 migrated the option
to coreboot. Braswell didn't, but it seems like this requirement was
added when it was migrated to FSP 1.1...? (Not very sure about that one,
but it does call load_vbt() right now which executes things based on
display_init_required().) Additionally, it seems to make sense to enable
it whenever the user explicitly selects VGA_ROM_RUN in menuconfig (like
one of the Intel defconfigs does).
Once we have all this, one could take a step back and ask whether this
option still makes sense at all anymore. It's enabled for almost all
devices (that work with vboot at all), it will presumably be enabled for
all future devices, and it seems that most devices that don't enable it
use libgfxinit, which as far as I can tell isn't gated on
display_init_required() but probably should be. Realistically, whatever
kind of display init a board needs to do (native or option ROM), it's
probably expensive enough that it's worth skipping on a normal mode
vboot boot, and we'd want to have this enabled by default on everything
except boards that actually don't have a display. So maybe we should
flip it around to CONFIG_VBOOT_OPROM_DOESNT_MATTER, but doing that would
probably lead to nobody ever selecting it at all.
Not sure what the best solution there is yet, but I think this patch
at least moves things in the more correct direction.
Change-Id: Id96a88296ddb9cfbb58ea67d93e1638d95570e2c
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32114
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Diffstat (limited to 'src/soc')
| -rw-r--r-- | src/soc/intel/baytrail/Kconfig | 1 | ||||
| -rw-r--r-- | src/soc/intel/braswell/Kconfig | 1 | ||||
| -rw-r--r-- | src/soc/intel/broadwell/Kconfig | 1 |
3 files changed, 3 insertions, 0 deletions
diff --git a/src/soc/intel/baytrail/Kconfig b/src/soc/intel/baytrail/Kconfig index e824ee49a6..a83a7e9ee6 100644 --- a/src/soc/intel/baytrail/Kconfig +++ b/src/soc/intel/baytrail/Kconfig @@ -43,6 +43,7 @@ config CPU_SPECIFIC_OPTIONS select CPU_HAS_L2_ENABLE_MSR config VBOOT + select VBOOT_OPROM_MATTERS select VBOOT_STARTS_IN_ROMSTAGE config BOOTBLOCK_CPU_INIT diff --git a/src/soc/intel/braswell/Kconfig b/src/soc/intel/braswell/Kconfig index fda5a6d687..8db47955a6 100644 --- a/src/soc/intel/braswell/Kconfig +++ b/src/soc/intel/braswell/Kconfig @@ -53,6 +53,7 @@ config CPU_SPECIFIC_OPTIONS select SOUTHBRIDGE_INTEL_COMMON_SMBUS config VBOOT + select VBOOT_OPROM_MATTERS select VBOOT_STARTS_IN_ROMSTAGE config BOOTBLOCK_CPU_INIT diff --git a/src/soc/intel/broadwell/Kconfig b/src/soc/intel/broadwell/Kconfig index 1a8349d08f..5f503da340 100644 --- a/src/soc/intel/broadwell/Kconfig +++ b/src/soc/intel/broadwell/Kconfig @@ -65,6 +65,7 @@ config PCIEXP_L1_SUB_STATE default y config VBOOT + select VBOOT_OPROM_MATTERS select VBOOT_STARTS_IN_ROMSTAGE config BOOTBLOCK_CPU_INIT |