This commit adds a coreboot patch which enables
CONFIG_USE_WATCHDOG_ON_BOOT for kgpe-d16; in upstream coreboot this
feature is found on only a few intel cpu platforms.
When enabled, this feature starts the hardware watchdog very early
in the boot process -- before PNP enumeration or DRAM
initialization. This ensures that any hangs or freezes due to
transient conditions (flakiness, temperature, electrical noise)
won't prevent the machine from eventually booting. This is very
useful for unattended servers.
On kgpe-d16 the watchdog is cancelled immediately before jumping to
the payload (i.e. Linux kernel). I found that if I left the
watchdog enabled, any attempt to use it (for example, to cancel it
or extend it) from Linux resulted in the machine resetting. Perhaps
this can be fixed, but for now I am content to simply re-enable the
watchdog from Linux rather than leaving it running.
Upstream describes this option as "Enable this option if coreboot
shall read options from the CMOS NVRAM instead of using hard-coded
values."
The RTC_BOOT_BYTE (which controls normal/fallback) indicator is
always taken from CMOS NVRAM, regardless of how this option is set.
When set to `no`, no other parts of the CMOS NVRAM are read or
written.
On KGPE-D16 I have found that USE_OPTION_TABLE=yes is frustratingly
flaky and unstable; about 1 in 20 boots will hang in the PNP device
enumeration. Apparently during SMP boot, multiple cores attempt to
access the (single) CMOS NVRAM concurrently, causing massive
headaches. Please don't turn this option on unless you are willing
to deal with extreme frustration.
This commit drops the verbatim `.config` files for coreboot that
were previously in the repository and converts them to NixOS-style
module configurations.
The big benefit of doing this is that the perl script in nixpkgs
that handles this will check to make sure that coreboot's Kconfig
machinery isn't silently ignoring any of our settings. It also
makes this configuration easier for end-users to customize.
This commit cleans up the selection of the serial console device and
gives it sensible default values (usually, the DB9 header on the
back I/O panel) on each platform.
This commit generalizes the selection and activation of the serial
console by promoting it to a top-level package-set member.
The top-level package set is starting to attract clutter, so I will
soon be moving all the non-package constitutents into a `config`
attrset.
This commit makes `hostPlatform` part of the packageset for more
consistent handling. A top-level `hostPlatform` argument is exposed
to the caller, in case they want to customize the `hostPlatform`
(compiler flags, etc).
The `hostPlatform` argument is inherited into the ownerboot package
set, where it will be overridden by `src/platform/*.nix` if it has
not been set explicitly.
Making `hostPlatform` part of the package set allows for more
sophisticated overriding schemes, for example adding additional
compiler flags or sanity-checking the flags that the user has
provided.
On amd64 platforms, booting ownerboot with the recovery jumper
installed will wipe the battery-backed nvram (aka "cmos" aka "rtc
nvram") and overwrite it with known-safe values taken from the
coreboot source code (`src/mainboard/asus/kgpe-d16/cmos.default`).
You should always do this when flashing a motherboard with ownerboot
for the first time.
This commit allows the user to customize the set of known-safe
values which are written when the recovery jumper is installed. To
do so, copy `src/mainboard/asus/kgpe-d16/cmos.default` out of
coreboot, edit to suit your tastes, and then override
`cmos-defaults` with the path to your customized `cmos.default`
file.
The microcode blob is only needed for Opteron 63xx chips. I have a
few of these, so I add the blob in a local overlay.
If other people are interested in this I will publish the overlay.
The 63xx chips are kind of rare and more expensive than the 62xx
chips -- their only real benefit is lower power draw. I ended up
receiving some by accident due to an incorrect eBay listing.
Since a6cd35, ownerboot includes a patch to flashrom which allows
nested (but non-overlapping) fmap regions, so the flashrom.layout
file is no longer necessary.
Adam Joseph