Our target is to get rid of backup_top_of_ram() and get_top_of_ram()
entirely so only declare these with LATE_CBMEM_INIT=y.
Change-Id: I54f549fe774996f4d803f9ec527e0fac46f6576f
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/8749
Tested-by: build bot (Jenkins)
Reviewed-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
The CBMEM memory segment is always placed at TOM - UMASIZE when GFXUMA
is enabled, however when GFXUMA is disabled an attempt was made to locate
the CBMEM memory segment above the I/O hole in certain rare cases.
Removing this special case does not impact functionality, and paves
the way for early CBMEM support.
Change-Id: I98d29ab9d601a4e20f58e2cd0a66abb13b494e74
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/8664
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Tested-by: Aaron Durbin <adurbin@google.com>
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
The AMD RS780 early initialization code originally used the
CF8/CFC I/O method for PCI configuration space access. After
the default configuration access method was changed to MMIO
(http://review.coreboot.org/#q,aad07472), booting would hang
at "PCI: pci_scan_bus for bus 01". Fix the problem by changing
function rs780_nb_gfx_dev_table() so that it no longer borrows
the BAR3 address needed for PCIe MMIO config usage.
Change-Id: I8816b94c848e1b50f8c880e5867a96ca2a33a8a7
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/8394
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
We use paths relative to that in the buildgcc script.
Change-Id: I2b79c3d2c75088af7e8e362d18a38274352eb965
Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-on: http://review.coreboot.org/8713
Reviewed-by: Aaron Durbin <adurbin@google.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
The GCC 4.9.2 update showed that the boot_state_init_entry
structures were being padded and assumed to be aligned in to an
increased size. The bootstate scheduler for static entries,
boot_state_schedule_static_entries(), was then calculating the
wrong values within the array. To fix this just use a pointer to
the boot_state_init_entry structure that needs to be scheduled.
In addition to the previous issue noted above, the .bs_init
section was sitting in the read only portion of the image while
the fields within it need to be writable. Also, the
boot_state_schedule_static_entries() was using symbol comparison
to terminate a loop which in C can lead the compiler to always
evaluate the loop at least once since the language spec indicates
no 2 symbols can be the same value.
Change-Id: I6dc5331c2979d508dde3cd5c3332903d40d8048b
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/8699
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
SERIRQ_CONTINUOUS_MODE is specific feature of LPC busses.
This fixes a KCONFIG unmet dependency warning on ARM mainboards with
chromeec.
Change-Id: Iae61986219585dcb1124cf3b24fa32a8596d56c8
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Reviewed-on: http://review.coreboot.org/8665
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Add license header with copyright of the original authors.
Change-Id: I8c55bb38a2a2a387ad2461e11d402c7392fa2497
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: http://review.coreboot.org/8691
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
The decode of UART addresses down to the LPC bus needs
to occur early to allow romstage console messages to
be seen. This enables the decode of most of the I/O
ports typically seen in a system.
Change-Id: I6636946af4ad5320a5a46c2920b4f06345b5f806
Signed-off-by: Dave Frodin <dave.frodin@se-eng.com>
Reviewed-on: http://review.coreboot.org/8661
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
The current spi_xfer() function sets the count in hardware and then
loops while waiting for the requested number of bytes to be sent or
received. However, the number of bytes to be transferred may exceed
the maximum count that can be programmed into the controller.
This patch re-factors spi_xfer() to split the low-level FIFO handling
portions for transmit/receive into their own functions to be called
by loops in spi_xfer() which will break large transfers into smaller
ones.
BUG=chrome-os-partner:30904
BRANCH=storm
TEST=built and booted with a >64KB payload on Storm
Original-Change-Id: I70743487996cf08cfc602449f2181a7fcd99bfa4
Original-Signed-off-by: David Hendricks <dhendrix@chromium.org>
Original-Signed-off-by: Vadim Bendebury <vbendeb@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/209838
Original-Reviewed-by: Trevor Bourget <tbourget@codeaurora.org>
Original-Tested-by: Trevor Bourget <tbourget@codeaurora.org>
(cherry picked from commit 5ec28de11f12c2438356f45ce978a17fbb603bf7)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I0033e0dd96006cfd30a7a4f5e5a052f677e05108
Reviewed-on: http://review.coreboot.org/8676
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Since CCLK_BURST_POLICY and SUPER_CCLK_DIVIDER are not accesible
from AVP, the first place that can change CPU clock is after CPU
has been brought up, ie, ramstage in this case.
CPU initial clock source is set to PLLP by MTS.
BUG=None
TEST=Norrin64 and A44
Original-Change-Id: I525bb2fa2be0afba52837bc0178950541535fd22
Original-Signed-off-by: Jimmy Zhang <jimmzhang@nvidia.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/209698
Original-Reviewed-by: Tom Warren <twarren@nvidia.com>
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
(cherry picked from commit ba77e26508bb4a50a08d07ad15632ff1ba501bfa)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Icf2458c491b4b3a553d3e01f88c6f25b25639e89
Reviewed-on: http://review.coreboot.org/8677
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Currently, the rmodules inclusion for vboot is dependent on ramstage_arch.
This change adds dependency on romstage_arch, since vboot is associated with
romstage. Inclusion based on ramstage_arch is left as is in case someone needs
it in ramstage.
BUG=chrome-os-partner:30784
BRANCH=None
TEST=Compiles successfully for link, rush and nyan
Original-Change-Id: Ib62415671c26a4a18c7133d98e8c683414def32b
Original-Signed-off-by: Furquan Shaikh <furquan@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/209568
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Original-Tested-by: Furquan Shaikh <furquan@chromium.org>
Original-Commit-Queue: Aaron Durbin <adurbin@chromium.org>
(cherry picked from commit 00da67cc02c81d7a6160f7336b33bf53b00e1875)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I9df02134af4e396c7257a2db2e2c371cfd1a02bc
Reviewed-on: http://review.coreboot.org/8673
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Provide functionality to create dynamic classes based on program name and the
architecture for which the program needs to be compiled/linked. define_class
takes program_name and arch as its arguments and adds the program_name to
classes-y to create dynamic class and compiler toolset is created for the
specified arch. All the files for this program can then be added to
program_name-y += .. Ensure that define_class is called before any files are
added to the class. Check subdirs-y for order of directory inclusion.
One such example of dynamic class is rmodules. Multiple rmodules can be used
which need to be compiled for different architectures. With dynamic classes,
this is possible.
BUG=chrome-os-partner:30784
BRANCH=None
TEST=Compiles successfully for nyan, rush and link.
Original-Change-Id: I3e3aadbe723d432b9b3500c44bcff578c98f5643
Original-Signed-off-by: Furquan Shaikh <furquan@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/209379
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Original-Tested-by: Furquan Shaikh <furquan@chromium.org>
Original-Commit-Queue: Aaron Durbin <adurbin@chromium.org>
(cherry picked from commit 242bb90d7476c2ee47d60c50ee18785edeb1a295)
Some of this cherry-pick had already been committed here:
commit 133096b6dc
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I9f5868d704c4b3251ca6f54afa634588108a788c
Reviewed-on: http://review.coreboot.org/8672
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
In some previous attempt to enable monotonic timers on all platforms,
the LAPIC monotonic timer was selected for Haswell devices, despite
the fact that LAPIC timers are not used in coreboot on Haswell
(See haswell Kconfig) and there already was a monotonic timer
implementation enabled that just needed to be added for SMM as well.
Change-Id: I6beb2977864e507956636860ed463e1991cea1ed
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/8702
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
When the Intel SPI drivers were refactored, compilation for Chrome OS
devices broke, because ELOG uses the SPI driver in SMM.
Change-Id: If2b2da5d526196ed742e17409b01a381417d0ce8
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/8701
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
On ChromeOS devices the ELOG section size and offset are
provided by the FMAP, rather than KConfig. Some upstream
refactoring broke compilation in that case.
Change-Id: I8b08daa327726218815855c7c2be45f44fcffeed
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: http://review.coreboot.org/8700
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@google.com>
Under certain conditions, e.g. automated testing, it is useful
to have the payload (e.g. Linux) reset the reboot_bits CMOS
value. This allows automated recovery in the case of coreboot
starting properly but the payload failing to start due to bad
configuration data provided by the coreboot image under test.
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Change-Id: Ifc8f565f8292941d90b2e520cc9c5993b41e9cdd
Reviewed-on: http://review.coreboot.org/8698
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
From GCC's documentation:
Optimize debugging experience. -Og enables optimizations that do not interfere
with debugging. It should be the optimization level of choice for the standard
edit-compile-debug cycle, offering a reasonable level of optimization while
maintaining fast compilation and a good debugging experience.
Change-Id: I9a3dadbf8e894cb28e29d7b2f4e9add252e7bbb3
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Signed-off-by: Scott Duplichan <scott@notabs.org>
Reviewed-on: http://review.coreboot.org/8689
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
You can build your new toolchain with:
$ cd util/crossgcc/
$ ./buildgcc -d /opt/cross -p x86_64-elf -j 16
or
$ make crossgcc-x64
Change-Id: I8eb584166294578d2b33c63e94ed3aca9b5de4f4
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: http://review.coreboot.org/8668
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
This removes the mainboard agesawrapper.c file from binarypi
based boards and creates a common one.
Change-Id: I900dba914f1c401e4ac732eb93d94b98216e629a
Signed-off-by: Dave Frodin <dave.frodin@se-eng.com>
Reviewed-on: http://review.coreboot.org/8671
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
This makes the change to the cpu/amd/pi/00730F01 that was
made for the cpu/amd/agesa based boards in:
commit 48518f0d
AGESA: Add amd_initcpuio() and amd_initmmio()
These are not wrappers for AGESA as they do not enter vendorcode at all.
We expect most of the added fixme.c file to be written without use of AMDLIB.h
and parts relocated as northbridge enable_resources().
The equivalent change has already been made for cpu/amd/pi/00630F01.
Change-Id: I591b50ee807436f5a1dee14d2c88a77462024744
Signed-off-by: Dave Frodin <dave.frodin@se-eng.com>
Reviewed-on: http://review.coreboot.org/8670
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
Tested on an X60, Native graphics init still works perfectly.
Change-Id: I91be3baa658e0332028c512c5a4cb0aee07d540a
Signed-off-by: Francis Rowe <info@gluglug.org.uk>
Reviewed-on: http://review.coreboot.org/8696
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Tested-by: build bot (Jenkins)
Most things still needs to be filled in, but this will allow us to build boards which use this SOC.
BUG=chrome-os-partner:29778
TEST=emerge-veyron coreboot
Original-Change-Id: If643d620c5fb8951faaf1ccde400a8e9ed7db3bc
Original-Signed-off-by: jinkun.hong <jinkun.hong@rock-chips.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/205069
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
Original-Reviewed-by: David Hendricks <dhendrix@chromium.org>
Original-Commit-Queue: David Hendricks <dhendrix@chromium.org>
Original-Tested-by: David Hendricks <dhendrix@chromium.org>
(cherry picked from commit 2f72473a8c2b3fe21d77b351338e6209035878fb)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I53fd0ced42f6ef191d7bf80d8b823bb880344239
Reviewed-on: http://review.coreboot.org/8653
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
I thought this wasn't going to work, and observing the timC detection
failure of early tests, I was getting somewhat discouraged; however,
this works. I've tried it with all possible permutations of the
following memory modules:
* 2 GiB single-rank DDR3-1600
* 4 GiB single-rank DDR3-1600
* 4 GiB dual-rank DDR3-1600
I did notice a limited number of memtest errors during one of the
runs, but they were in an address range that is otherwise marked as
reserved. I wrote that off as "maybe something was doing MMIO there
just when memtest was poking the address range". I was not able to
reproduce that error.
Change-Id: Ibd52e1d52fc8d900591d6a488f9a5b4d1e5e4fd3
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/8477
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@gmail.com>
This brings the KFSN4-DRE in line with other boards in the tree.
Change-Id: I9216130f51ed0576871fd27ca6ae4610c5f5810e
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/8683
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
This was missed in commit bde6d309 as the driver is not enabled
in any configuration by default.
Change-Id: I3d886531f5bcf013fc22ee0a1e8fa250d7c4c1a4
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/8660
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
There is no point in duplicating boardid.h per board - they are all
the same. Let's keep a single instance in the common include directory
and let the linker report a problem if one tries using this function
on a board where it is not supported.
BUG=chrome-os-partner:30489
TEST=verified that coreboot builds fine for nyan_big and nyan_blaze.
Original-Change-Id: Ifbe9c2287a1d828d4db74c637d1d02047ac4da25
Original-Signed-off-by: Vadim Bendebury <vbendeb@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/209699
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Original-Reviewed-by: Furquan Shaikh <furquan@chromium.org>
(cherry picked from commit 589e6415faf18ca6aaf44da343dd33eadc8a53d3)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I8eef89cb822611a0050e5a50fc4b970eebd8d962
Reviewed-on: http://review.coreboot.org/8666
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
The serial driver hangs in cases when FIFO has more than single word to be
processed. Easiest way to reproduce is to paste a string of greater than 4
characters in cli.
Clearing the RXSTALE interrupt without draining all the characters from FIFO
leads to the issue as the driver is dependent on msm_boot_uart_dm_read
function to reinitialize for next transfer.
Logically the driver is organized in such a manner that next transfer never
gets initiated till rx_data_read < total_rx_data. Clearing the RXSTALE without
consideration of total number of characters (or words) unprocessed makes the
msm_boot_uart_dm_read to return on the first if conditional. Thus the driver is
stuck forever.
A quick fix is to avoid clearing the stale interrupt. Reset is handled whenever
a new transfer is initialized in msm_boot_uart_dm_init_rx_transfer.
BUG=chrome-os-partner:29542
TEST=manual
-Paste a string greater than 4 characters in cli.
Original-Change-Id: I016afb01a77cd14764f0176f6bf144fb29796c2f
Original-Signed-off-by: Yogesh Lal <ylal@codeaurora.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/209512
Original-Reviewed-by: Vadim Bendebury <vbendeb@chromium.org>
Original-Commit-Queue: Vadim Bendebury <vbendeb@chromium.org>
Original-Tested-by: Vadim Bendebury <vbendeb@chromium.org>
(cherry picked from commit 61528884ad2c0a8e146054bbfeb01a3bc73b9692)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I936af5daa52a25f62133bdf9fb44f0b68cf34e88
Reviewed-on: http://review.coreboot.org/8667
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
This is useful when the PCB layout of a mainboard does not allow
stable operation at the increased HyperTransport speeds of newer
processors.
Change-Id: Idc93a1294608178ddf38ca72d40e6bad7deb9004
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/8464
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
This patch removes a chunk of romstage code from Tegra and all Nyan
boards that was supposed to enable some LCD power rails early, but never
really worked. The dev_find_slot() function can only find PCI devices,
which the CPU cluster is not. Since we're done with Nyan-RO and the
ramstage display code is fine as it is, there is no point in trying to
fix this... but we should remove it from ToT lest someone uses it as a
blueprint to add more dead code to future boards.
BRANCH=None
BUG=None
TEST=None
Original-Change-Id: I6eee256873299429d4e3934fe7d454120390f34d
Original-Signed-off-by: Julius Werner <jwerner@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/207720
Original-Reviewed-by: David Hendricks <dhendrix@chromium.org>
(cherry picked from commit a3df62a3bcefcc20ae59648f5d1f0a01db3c02c6)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I8deedea5e9787848aae3064509c611bc349313cc
Reviewed-on: http://review.coreboot.org/8638
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: David Hendricks <dhendrix@chromium.org>
This capability means:
FERR messages are sent out on system detected an
unmasked floating point x87 FPU error.
Even though this capability is supported on nehalem it doesn't
make sense to set it in early stage. This MSR
has a core scope which results in an unsync MSR because
it's not set on other cores than the BSP.
Found-by: BITS
Tested-on: lenovo thinkpad x201t
Change-Id: Ief3c04f57ac69e7289fbd37dbc3fd239f9098155
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
Reviewed-on: http://review.coreboot.org/8659
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
BIOS Writer's Guide, rev 1.6.0, June 2012:
This MSR controls whether and FERR message is sent over the system bus
when unmasked x87 exceptions are generated.
This feature is not supported from Sandy Bridge processor onwards.
Change-Id: I19b260ca4b62f57c26989430693b00b9853bc441
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
Reviewed-on: http://review.coreboot.org/8658
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Stop running AVP at the end of romstage until event conditions are met (JTAG,
GIC_IRQ or LIC_IRQ).
BUG=chrome-os-partner:30831
BRANCH=None
TEST=Compiles successfully and boots till last known good checkpoint.
Original-Change-Id: Ia221f08b27ac0c60a66d588e351677144cc6a322
Original-Signed-off-by: Furquan Shaikh <furquan@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/209424
Original-Tested-by: Furquan Shaikh <furquan@chromium.org>
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Original-Commit-Queue: Furquan Shaikh <furquan@chromium.org>
(cherry picked from commit df4e8b4c8a1002443a936bd0563fbc9e0710f489)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I59f7702bd50a1039b8723e9cb12b8d714e353d37
Reviewed-on: http://review.coreboot.org/8651
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Timothy Pearson