Rename psp_fill_head() and call it with the cookie to populate the
header. The combo header and PSP directory header are similar and
should be calculated the same way.
Change-Id: I7e634542de65576addadbe683596cbe572de3dcd
Signed-off-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31732
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
There are effectively two unique sets of arguments for the utility,
causing one of two tables to be constructed. Both tables are
identical, however, and therefore the only practical difference is
the offset in the Embedded Firmware Structure which holds the pointer
to the table.
This patch is part 2 of 2 to reduce the number of command-line options
to amdfwtool. Part 1 added the --combo-capable option that helps
put the PSP directory pointer in the correct location. Part 2
removes the duplicated table, the support code, options, and updates
the usage text.
TEST=Build before/after images for grunt, bettong, apu2, and diff
hexdumps of the amdfw.rom files. Built/ran grunt with PSP_COMBO
defined as 1.
BUG=b:126691068
Change-Id: I542a7f5023137f30fbe00533452d4448117df487
Signed-off-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31731
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martinroth@google.com>
There are effectively two unique sets of arguments for the utility,
causing one of two tables to be constructed. Both tables are
identical, however, and therefore the only practical difference is
the offset in the Embedded Firmware Structure which holds the pointer
to the table.
This patch is part 1 of 2 to reduce the number of command-line options
to amdfwtool. Create a new option that is used as an indicator for
which Embedded Firmware offset to use. Part 2 will be added once
makefiles no longer use the duplicated options.
This patch also adds two new options for fanless SMU firmware to be
used instead of the ones that will be removed in part 2.
TEST=Verify no difference in amdfw.rom for google/grunt before
and after the patch is applied
BUG=b:126691068
Change-Id: I249700c6addad1c0ecb495a406ffe7a022dd920b
Signed-off-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31729
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martinroth@google.com>
The fletcher32 algorithm generates a sum over a range of 16-bit
WORDs. Change the function's interface to be more generic,
accepting a more intuitive size in BYTEs. Don't require the
caller to understand the nature of the algorithm and convert to
WORDs prior to calling.
TEST=Verify no difference in amdfw.rom for google/grunt before
and after the patch is applied
Change-Id: Iad70558347cbdb3c51bd598479ee4484219c0869
Signed-off-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31728
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Replace the use of multiples of DWORDs with structures that
describe the Embedded Firmware Table, and PSP directory
headers & entries.
TEST=Verify no difference in amdfw.rom for google/grunt build
(Family 15h Models 70h-7Fh), amd/bettong (F15h 60h-6Fh),
and pcengines/apu2 (F16h 30h-3Fh). PSP_COMBO builds but
was not verified.
Change-Id: If05952d9282a0fa5a397984eaae671fb33f6134a
Signed-off-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31727
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
Correct an oversight in the utility that attempts to match up eligible
PSP directory table entries with blob names passed on the command
line. A 1:1 matchup of items shouldn't be assumed, so the i iterator
shouldn't be used to walk both lists.
This change has no effect on google/grunt (Family 15h Models 70h-7Fh),
but eliminates blank entries of all FF's on builds of amd/bettong
(F15h 60h-6Fh) and pcengines/apu2 (F16h 30h-3Fh). Removal of entries
also affects the checksum accordingly.
TEST=Build before/after images for grunt, bettong, apu2, and diff
hexdumps of the amdfw.rom files
Change-Id: I13e359d3cc6f5ce408bbf077feec3707ee2b3838
Signed-off-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31726
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Until now the TCPA log wasn't working correctly.
* Refactor TCPA log code.
* Add TCPA log dump fucntion.
* Make TCPA log available in bootblock.
* Fix TCPA log formatting.
* Add x86 and Cavium memory for early log.
Change-Id: Ic93133531b84318f48940d34bded48cbae739c44
Signed-off-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/29563
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
The __pci_driver and __cpu_driver uses variable length arrays which are
constructed by the linker at build-time.
The linker always place the structs at 16-byte boundary, as per
"System V ABI". That's not a problem on x86, as the struct is exactly
16 Bytes in size. On other platforms, like x86_64 it breaks, because the
default data alignment isn't SysV compatible.
Set -malign-data=abi to make x86_64 gcc use the SysV psABI.
Fixes broken __pci_driver and __cpu_driver on x86_64.
Change-Id: I2491d47ed03dcfd8db110dfb181b2c5281449591
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/30116
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Based on not publicly available IT8613E Preliminary Specification V0.3.
Change-Id: Iec99d4d998f645dbad9c803d6d5477580b0bccc4
Signed-off-by: Krystian Hebel <krystian.hebel@3mdeb.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31620
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
This is a great check, but unfortunately it's currently not effective
because most uses of IS_ENABLED() do not have whitespace in front of
them (they're mostly used as part of an if (IS_ENABLED(...)) condition).
This patch makes the linter a little more generous in what it considers
in scope to avoid these false negatives in the future.
Change-Id: I2296410c73cd6e918465c90db33e782936bec0f9
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31746
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
When updating firmware, it is very often that we may want to preserve
few sections, for example vital product data (VPD) including serial
number, calibration data and cache. A firmware updater has to hard-code
the section names that need to be preserved and is hard to maintain.
A better approach is to specify that in FMAP area flags (the `area_flag`
field) using FMAP_AREA_PRESERVE. With this patchset, a FMD parser flag
"PRESERVE" is introduced and will be converted to FMAP_AREA_PRESERVE
when generating FMAP data (by fmap_from_fmd.c).
For example, The FMD statement:
RO_VPD(PRESERVE)@0x0 16k
will generate an FMAP firmware section that:
area_name = "RO_VPD"
area_offset = 0
area_size = 16384
area_flags = FMAP_AREA_PRESERVE
BUG=chromium:936768
TEST=make; boots on x86 "google/eve" and arm "google/kukui" devices
Manually added 'PRESERVE' to some FMD files, and verify (by running
fmap.py) the output coreboot.rom has FMAP_AREA_PRESERVE set
Change-Id: I51e7d31029b98868a1cab0d26bf04a14db01b1c0
Signed-off-by: Hung-Te Lin <hungte@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31707
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
The recent toolchain update also updated binutils, which has a new
relocation type, introduced with commit bd7ab16b
(x86-64: Generate branch with PLT32 relocation).
Add support for R_X86_64_PLT32, which is handled as R_X86_64_PC32.
Add comment explaining the situation.
Fixes build error on x86_64.
Change-Id: I81350d2728c20ac72cc865e7ba92319858352632
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31468
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The idea of "annotation" for firmware sections was pretty flexible, but
in future we will want multiple attributes applied to same area. For
example, indicate the section must be preserved when updating firmware
so serial number or MAC address can be preserved.
The solution here is to extend annotation so it can take multiple
identifiers (flags) in a row. For example, to declare a 64KB COREBOOT
section as CBFS using annotation:
COREBOOT(CBFS)@0x0 64k
If there's a new flag "PRESERVE" indicating the section must be
preserved before update, we can declare it following CBFS flag:
COREBOOT(CBFS PRESERVE)@0x0 64k
The flags are directly parsed in fmd_parser, and stored in an union
flashmap_flags. Output modules can choose to ignore or process the
flags.
Currently the only supported flag is "CBFS" (for backward compatible
with annotation). There will be more new flags in follow up patches.
BUG=chromium:936768
TEST=make; boots on x86 "google/eve" and arm "google/kukui" devices
Change-Id: Ie2d99f570e6faff6ed3a4344d6af7526a4515fae
Signed-off-by: Hung-Te Lin <hungte@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31706
Reviewed-by: Julius Werner <jwerner@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
git diff needed to emit diffs without prefix (e.g. a/ and b/) for
clang-format-diff to be able to work.
Also require that the test succeeds, but note that it only runs on
trees whitelisted in $(top)/.clang-format-scope.
Change-Id: I7e9a32eb9281b5cb0b45506a206500fd1d315372
Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31654
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martinroth@google.com>
80 chars + 2 tabs was the compromise we got to in the last round of
discussion.
Change-Id: I9293a69d1bea900da36501cde512004d0695ad37
Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31651
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Apollo Lake has four GPIO communities each with a single group named
after the physical location of the pads (I guess): North West, North,
West and South West.
Also add some logic to be able to tag the default function of a pad
(with an asterisk before its name). This seems easier to review in the
tables, but we could also encode the number of the default explicitly
instead.
Used Intel documents:
- 334817-001 (datasheet vol. 1)
- 334819-001 (datasheet vol. 3)
Change-Id: I5cd687fdc1d2ae81f2e948178bf319897b47f031
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Signed-off-by: Felix Singer <migy@darmstadt.ccc.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/29897
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
When updating firmware, we may need to preserve some sections like VPD,
calibration data, ... etc. The logic can be hard-coded in updater as a
list of known names, but a better solution is to have that directly
declared inside FMAP area flags.
To do that, the first step is to apply the changes in flash map
(http://crosreview.com/1493767). A new FMAP_AREA_PRESERVE is now
defined and will be set in future with new syntax in FMD parser.
BUG=chromium:936768
TEST=make; boots an x86 image.
Change-Id: Idba5c8d4a4c5d272f22be85d2054c6c0ce020b1b
Signed-off-by: Hung-Te Lin <hungte@chromium.org>
Reviewed-on: https://review.coreboot.org/c/31676
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Relocations for PC relative instructions must not emitted.
As PC64 are unlikely with current code, it never was an issue.
Change-Id: Ife472a287ff15b1c04a516e25ff13221441fd122
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/c/31469
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The name was wrong. mFCPGA478 is actually a pseudonym for mPGA478MN,
the successor of the socket that was meant.
The official name of this socket is mPGA478MT. But "Socket M" is much
easier to distinguish.
Change-Id: I4efeaca69acddfcdc5e957b0b521544314d46eeb
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/31642
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Because coreboot's asserts aren't fatal by default, scan-build finds
problems in code that is actually protected by an assert. This
change fixes that and allows us to add asserts to protect
against other failures.
Change-Id: I9fa605d6309bb40a9cef33b434c9256bf731f457
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/c/31650
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
* Introduce a measured boot mode into vboot.
* Add hook for stage measurements in prog_loader and cbfs.
* Implement and hook-up CRTM in vboot and check for suspend.
Change-Id: I339a2f1051e44f36aba9f99828f130592a09355e
Signed-off-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Signed-off-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-on: https://review.coreboot.org/c/29547
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Upstream intelmetool is out of date; I suggest I archive it
instead of trying to merge coreboot's changes into it.
However I would like to preserve the licensing of files in the tool
as GPLv2+ where possible instead of GPLv2-only.
Change-Id: I47b1ff2734f54c65f4214b39244bd868ef44b83c
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: https://review.coreboot.org/c/31587
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
VBT on Intel(R) systems is available via sysfs as
/sys/kernel/debug/dri/0/i915_vbt
However the size of this file reads as 0 causing
intelvbttool to fail. This patch implements incremental reads
with realloc for such cases or whenever the file size is not
available (e.g. reading from stdin).
After this patch is applied, intelvbttool can be used as follows:
sudo intelvbttool -f /sys/kernel/debug/dri/0/i915_vbt -d
Change-Id: I5d17095a5747550b7115a54a7619b7294a846196
Signed-off-by: Alex Feinman <alexfeinman@hotmail.com>
Reviewed-on: https://review.coreboot.org/c/31531
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Allow EC region to be readable by BIOS/CPU so that flashrom
can read it.
BUG=b:123199222
TEST=Build coreboot with CONFIG_LOCK_MANAGEMENT_ENGINE set,
run firmware_LockedME test.
Change-Id: I306c74a0893355e57632a22a712b1f4fdaa19306
Signed-off-by: Bora Guvendik <bora.guvendik@intel.com>
Reviewed-on: https://review.coreboot.org/c/31377
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Put the FMAP FMAP region right above the coreboot CBFS region.
The other regions like RW_MRC_CACHE and CONSOLE often have alignment
requirements so it makes sense to put those on top. This also
simplifies the code the generate the default fmap a little.
Change-Id: I24fa6c89ecf85fb9002c0357f14aa970ee51b1df
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/30419
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The coreboot table entry containing the memory entries can have
fields unnaturally aligned in memory. Therefore one needs to perform
an aligned_memcpy() so that it doesn't cause faults on certain
architectures that assume naturally aligned accesses.
BUG=chromium:925961
Change-Id: I28365b204962ac89d65d046076d862b6f9374c06
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/c/31181
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Manoj Gupta <manojgupta@google.com>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Now running 1.8.3, with a fix to the theme so search still works, and
a recommonmark version that properly rewrites links to .md files.
Change-Id: Ice25554c77a398a71782c8d1cb9e205debd80d67
Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-on: https://review.coreboot.org/c/31129
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Version 2 IFD will have flmstr5 as EC region access control, consider it
during descriptor lock/unlock process.
BUG=N/A
TEST=Build coreboot with CONFIG_LOCK_MANAGEMENT_ENGINE set, and check
flmstr5 value by hexdump the SPI image at offset FMBA+0x90.
Signed-off-by: Lijian Zhao <lijian.zhao@intel.com>
Change-Id: I970064dcf6114a15f054ab7c44349841deb99dc8
Reviewed-on: https://review.coreboot.org/c/31111
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Cannonlake and Icelake have same read/write region permission settings
with skylake and kabylake, so add it here as well.
BUG=b:123199222
TEST=Turn on CONFIG_LOCK_MANAGEMENT_ENGINE and build image, check the
setting matches 0x0D for read and 0x04 for write.
Signed-off-by: Lijian Zhao <lijian.zhao@intel.com>
Change-Id: I71d8b815c7dff7dcbcff2bf77c85ebf80b8df6d2
Reviewed-on: https://review.coreboot.org/c/31104
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
This reverts commit 48c24ce5ee.
Reason for revert: Commit broke bincfg, and sym_table as a global
variable is less bad than passing it around in function calls.
Change-Id: Ib8d64a1dc201d17a4e278ab0114958b6807a45ac
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/31105
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
F81866 detection tested with the iBASE SI-613:
superiotool r4.9-420-g034e5e6
Found Fintek F81866 (vid=0x3419, id=0x1010) at 0x4e
F8196* detection is based on chip IDs provided by iBASE, but untested.
Change-Id: I7210e1523a188a8593cd03547bb0c95cd3e7aa39
Signed-off-by: Kyle Stevenson <kstevenson@comqi.com>
Reviewed-on: https://review.coreboot.org/c/31052
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Converts `auto.conf` to an Ada spec file. Write to
$(obj)/cb-config.ads and set the package name to
`CB.Config`.
Change-Id: I97c060d8a613c74a82a18aff9524ad4b01f9df56
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/c/31053
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
With the memory controller the separate sockets becomes a useless
distinction. They all used the same code anyway.
UNTESTED: This also updates autoport.
Change-Id: I044d434a5b8fca75db9eb193c7ffc60f3c78212b
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/31031
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
On FreeBSD, this test was failing with the error:
"grep: Argument list too long"
- Remove support for testing coreboot not in a git repo. Many of
the other linters already don't support this.
- Use git grep to find offending files, then xargs to print out
the lines.
Change-Id: Ic017dc3465fd9a46ff4e6ec5ef16396e963483cd
Signed-off-by: Martin Roth <martinr@coreboot.org>
Reviewed-on: https://review.coreboot.org/c/28448
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
After merging util/crossgcc: derive date and version from latest commit
(https://review.coreboot.org/c/coreboot/+/30804),
crossgcc build is broken in internal repository due to long version
name;coreboot.org repository is ok because it uses short tag name.
The patch uses "git describe" which is dependent on git tag name.
If tag name is little bit long, it can cause crossgcc build failed.
To avoid this issue, use only short version of hash string
which is fixed length. And it's enough as version string,
because we also use date(CROSSGCC_DATE) together.
TEST=Build crossgcc in both coreboot.org and internal repository
which uses longer tag name and check version string in build log.
Change-Id: I405b2e4e5c05831c25aebf1c73a281adab8ef452
Signed-off-by: Wonkyu Kim <wonkyu.kim@intel.com>
Reviewed-on: https://review.coreboot.org/c/31001
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Marshall Dawson