Based on voema schematics, two USB2 ports 3 and 5 are assigned to type C
connectors on Voema board.
BUG=b:177483061 b:172535001
TEST=Build and boot Voema.
Signed-off-by: David Wu <david_wu@quanta.corp-partner.google.com>
Change-Id: I12cef85595e511801ab9c563ae4aa26e25875679
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49451
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Fagerburg <pfagerburg@chromium.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Do LPC/eSPI pad configuration at board-level to match other platforms.
Early gpio configuration was done in romstage, while LPC pads were
configured in bootblock. Instead of adding another dedicated gpio table
for bootblock, move early gpio configuration completely to bootblock on
these boards. This won't hurt, since there is no code touching the pads
in between.
The soc code gets dropped in CB:49410.
Change-Id: I2a614afb305036b0581eac8ed6a723a3f80747b3
Tested-by: Mario Scheithauer <mario.scheithauer@siemens.com>
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49413
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Mario Scheithauer <mario.scheithauer@siemens.com>
Command timing is the absolute value of the most negative `pi_coding`
value across all ranks, or zero if there are no negative values. Use the
MAX() macro to ease proving that `cmd_delay` can never be negative, and
then drop the always-false underflow check.
The variable type for `cmd_delay` still needs to be signed because of
the comparisons with `pi_coding`, which is a signed value. Using an
unsigned type would result in undefined and also undesired behavior.
Change-Id: I714d3cf57d0f62376a1107af63bcd761f952bc3a
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49320
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Clock is a differential signal and propagates faster than command and
control, therefore its timing needs to be offset with `pi_code_offset`.
It is also a periodic signal, so it can safely wrap around.
To avoid potential undefined behavior, make `clk_delay` signed. It makes
no difference with valid values, because the initial value can be proven
to never be negative and `pi_code_offset` is always positive. With this
change, it is possible to add an underflow check, for additional sanity.
Change-Id: I375adf84142079f341b060fba5e79ce4dcb002be
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49319
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Commit 7584e550cc (nb/intel/sandybridge: Clean up program_timings)
introduced this condition along with a comment that says the opposite.
Command and clock timings always need to be computed, so drop both the
nonsensical condition and the equally-worthless corresponding comment.
Change-Id: I509f0f6304bfb3e033c0c3ecd1dd5c9645e004b2
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49318
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
There's no need to have them in the devicetree. ACPI generation can now
be simplified even further, and is done in subsequent commits.
Change-Id: I3a788423aee9be279797a1f7c60ab892a0af37e7
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/46908
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Basking Ridge is not ULT, thus does not support C-states deeper than C7.
Replace them with the values used by all other Haswell non-ULT boards to
allow subsequent commits to cleanly factor them out of the devicetree.
Change-Id: Ife34f7828f9ef19c8fccb3ac7b60146960112a81
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/46907
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
LTE module is not expected to be powered off during warm reset. Hence
configure the LTE_PWR_OFF_ODL (GPP_A10) gpio pad reset configuration to
PWROK and set the TX state to 1.
BUG=b:163100335
BRANCH=dedede
TEST=Verified through the waveforms that power sequence is meeting the LTE module requirements.
Change-Id: I8676da6186559288aabe078b6158fc01075c7b41
Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48623
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Henry Sun <henrysun@google.com>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Exit early if the chipset power state info isn't in CBMEM. Return -1 in
order to ensure the one caller of this function exits early as well.
Found-by: Coverity CID 1442304
Change-Id: Ifa42ba3024d3144de486d90ed7752820482549bf
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49359
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Lance Zhao
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
LTE module used in metaknight has a specific power on/off sequence.
GPIOs related to power sequence are:
* GPP_A10 - LTE_PWR_OFF_R_ODL
* GPP_H17 - LTE_RESET_R_ODL
1. Power on: GPP_A10 -> 20ms -> GPP_H17
2. Power off: GPP_H17 -> 10ms -> GPP_A10
3. Warm reset: GPP_A10 keeps high, GPP_H17 goes low at least 2ms
Configure the GPIOs based on these requirements.
BUG=b:173671094
TEST=Build and boot Metaknight to OS. Ensure that the LTE module power
sequence requirements are met.
Change-Id: Ibff16129dfe2f1de2b1519049244aba4b3123e52
Signed-off-by: Tim Chen <tim-chen@quanta.corp-partner.google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48195
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Use C-style comments everywhere, and follow the coding style.
Tested with BUILD_TIMELESS=1, Asus P5QL PRO remains identical.
Change-Id: I3ef96c5f6553ad50cee7d7f5614128b62a89e4ea
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49387
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
The caches have already been enabled during MP-init,
so these function calls are redundant. Remove them.
Change-Id: Ia9be1a3388d8e7c73c35a1c68b3dd5bc488658c2
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49383
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Fix C code to match comment and assembly implementation.
Tested on Prodrive hermes:
The microcode spinlock is no longer used.
Change-Id: I21441299f538783551d4d5ba2b2e7567e152d718
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49304
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change affects Intel CPUs only. As most platforms are doing
uCode update using FIT, they aren't affected by this code either.
Update microcode in MP-init using a single spinlock when running on
a Hyper-Threading enabled CPU on pre FIT platforms.
This will slow down the MP-init boot flow.
Intel SDM and various BWGs specify to use a semaphore to update
microcode on one thread per core on Hyper-Threading enabled CPUs.
Due to this complex code would be necessary to determine the core #ID,
initializing and picking the right semaphore out of CONFIG_MAX_CPUS / 2.
Instead use the existing global spinlock already present in MPinit code.
Assuming that only pre-FIT platforms with Hyper-Threading enabled and at
most 8 threads will ever run into this condition, the boot delay is
negligible.
This change is a counterproposal to the previous published patch series
being much more unsophisticated.
Change-Id: I27bf5177859c12e92d6ce7a2966c965d7262b472
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49303
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Eaglelake MRC 2.55 does this, and also stalls for less time.
Change-Id: Iaaefd32c341a490e5c129df865407ec3f8da8212
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49385
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
To allow other platforms to reuse this code, extract it into a separate
compilation unit. Since HPET is enabled through the southbridge, place
the code in the southbridge scope. Finally, select the newly-added
Kconfig option from i82801gx and replace lpc.c `enable_hpet` function.
Change-Id: I7a28cc4d12c6d79cd8ec45dfc8100f15e6eac303
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49365
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
On Coffee Lake systems prog_locate_hook() is called with PROG_POSTCAR.
For this reason the early check is not executed.
Add check for prog->type == PROG_POSTCAR, but execute
verified_boot_early_check() once.
BUG = N/A
TEST = Build and boot on Facebook FBG1701 and Intel CoffeeLake system
Change-Id: Ia3bd36064bcc8176302834c1e46a225937d61c20
Signed-off-by: Frans Hendriks <fhendriks@eltan.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48852
Reviewed-by: Wim Vervoorn <wvervoorn@eltan.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
make olddefconfig on other projects using USE_VENDORCODE_ELTAN results in error.
USE_VENDORCODE_ELTAN unmet direct dependencies.
Remove dependency on VBOOT for USE_VENDORCODE_ELTAN.
TEST = Build and boot on Facebook FBG1701
Change-Id: I5881c334955c73ae0f1a693f95ceb1aee62ee898
Signed-off-by: Frans Hendriks <fhendriks@eltan.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48690
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Wim Vervoorn <wvervoorn@eltan.com>
This struct isn't used anywhere else, so there's no need to name it.
Change-Id: I22eda07f14096d2b7400e6ab715641ffd68fbc08
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reported-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49444
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The default value is not sufficient to correctly configure the Type-C
ports as it has all ports disabled by default. On Volteer ports 0
and 1 are enabled so setting this value to 0x3 and correctly
keeping the IomPortPadCfg values at 0 for ports that have a
retimer and ports that are not configured. These values were set
to 0x90000000 to avoid s0ix issues which arose from the UsbTcPortEn
value being incorrect.
BUG=b:159151238
BRANCH=firmware-volteer-13672.B
TEST=Built image for Voxel and verified that s0ix cycles complete
without any issues
Change-Id: Ib4f2bd0f68debd4e97ccaab9e1d8a873dc4e4d9f
Signed-off-by: Brandon Breitenstein <brandon.breitenstein@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48814
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Caveh Jalali <caveh@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
As a requirement of TCSS this setting needs to be correctly set
to determine what Type-C ports are enabled on the platform. Without
this value correctly set there can be adverse effects on the other
TCSS specific values.
BUG=b:159151238
BRANCH=firmware-volteer-13672.B
TEST=Built image for Voxel and verified that S0ix cycles no longer
fail when the IomPortPad is set to 0
Change-Id: I6c5260cda71041439fe89d15bd3cafd4052ef1e7
Signed-off-by: Brandon Breitenstein <brandon.breitenstein@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48813
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Caveh Jalali <caveh@chromium.org>
These are no longer really useful. We can also enable Power Resource
ACPI debug in the kernel if we want these messages.
BUG=none
BRANCH=zork
TEST=emerge-zork and verify debug messages are no longer posted
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: I936e816266825f1c59377c2e079ffe1a5188838c
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49366
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Target added to INTERMEDIATE all operate on coreboot.pre, each modifying
the file in some way. When running them in parallel, coreboot.pre can be
read from and written to in parallel which can corrupt the result.
Add a function to create those rules that also adds existing
INTERMEDIATE targets to enforce an order (as established by evaluation
order of Makefile.inc files).
While at it, also add the addition to the PHONY target so we don't
forget it.
BUG=chromium:1154313, b:174585424
TEST=Built a configuration with SeaBIOS + SeaBIOS config files (ps2
timeout and sercon) and saw that they were executed.
Change-Id: Ia5803806e6c33083dfe5dec8904a65c46436e756
Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49358
Reviewed-by: Martin Roth <martinroth@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The ACPI part still needs some more code to be in place, so add that
later.
TEST=Together with the currently not merged rest of the amdfw patch
train applied this results in working serial console in bootblock in
Majolica.
Change-Id: Ia844e86a80c19026ac5b47a5a1e91c2553ea5cca
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49378
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Since the functions that get called by the coreboot console
initialization code aren't in the SOC-specific code anymore, the SOC's
uart.c can be included unconditionally in the build now. This also
replaces the STONEYRIDGE_UART Kconfig option with the common
AMD_SOC_CONSOLE_UART one.
Change-Id: I09c15566a402895d6388715e8e5a802dc3c94fdd
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49375
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This will result in less code duplication when the common AMD SoC UART
support gets used for more AMD SoCs.
Change-Id: Id1786f32324de3e3947d792c599e2019705c5a85
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49373
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This partially reverts commit 6f8f9c969b
by moving CONSOLE_UART_BASE_ADDRESS back to the SoC-specific code, since
the number and base addresses of UARTs turned out to be rather SoC-
specific. The help text for the AMD_SOC_CONSOLE_UART option also
contained those base addresses, so remove that as well.
Change-Id: I01211ec62421c56f22ed611313d6245a05bdea67
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49372
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The UARTs in the Picasso SoC are memory mapped, but there is also some
hardware support that isn't used by any board to make the UARTs behave
like the ones found on legacy x86 machines from the 90s.
In the MMIO mode the MMIO address of the UART controller is passed to
the OS via ACPI. The OS expects the base clock of the UART controller to
be 48MHz (see the cz_uart_desc struct in drivers/acpi/acpi_apd.c and
drivers/tty/serial/8250/8250_dw.c in the Linux kernel) in this case. It
is also possible to enable additional decodes from four 8 byte legacy
I/O locations used for serial ports to the different UART controllers,
which doesn't disable the MMIO access though. The legacy I/O-mapped
serial ports are usually expected to have a base clock of 16*115200Hz
which the hardware can also provide to the UART's baud rate generator.
So there are two possible valid configurations to use the UARTs; either
MMIO access in combination with a 48MHz base clock or the legacy I/O
decode with a ~1.8MHz base clock.
The existing code unconditionally generates ACPI objects for all enabled
UARTs, so those shouldn't be put into legacy mode and switching the base
clock to ~1.8MHz was only done in the case that the UART was used as
coreboot console UART which still used the MMIO access, but the lower
base clock. Since no board even selects this option and it's rather
invasive to properly implement this feature, just drop the corresponding
broken code.
TEST=SoC UART console still works on Mandolin.
Change-Id: I26fa8fdfc781b583ba56ac4dbcbbfb6100e84852
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reported-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49371
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This is causing boot errors on zork:
coreboot-v1.9308_26_0.0.22-18590-g4598a7bed945 Wed Dec 16 17:32:25 UTC 2020 bootblock starting (log level: 8)...
Family_Model: 00820f01
PSP boot mode: Development
Silicon level: Pre-Production
PMxC0 STATUS: 0x800 BIT11
I2C bus 3 version 0x3132322a
DW I2C bus 3 at 0xfedc5000 (400 KHz)
FMAP: area FW_MAIN_B found @ 312000 (3137280 bytes)
ASSERTION ERROR: file 'src/commonlib/bsd/cbfs_mcache.c', line 106
BUG=b:177323348
TEST=Boot ezkinil to OS
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: I1f2bbdd9c87c4efdfb0042e90a20b489fa0efced
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49128
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
While unused, this allows use of a common initialisation
code for GNVS allocation.
Change-Id: Ie84b5a3e16d3baa12bcd5dadac0b1f7edb323272
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49343
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Update PL1 min and max values for Delbin systems
BUG=b:168958222
BRANCH=None
TEST=Build and verify on delbin system
Signed-off-by: Deepika Punyamurtula <deepika.punyamurtula@intel.com>
Change-Id: I2152f0dbeb0ae463b78464571b6c434830f0082a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49364
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Zhuohao Lee <zhuohao@google.com>
Reviewed-by: Zhuohao Lee <zhuohao@chromium.org>
The S3/S4 workaround is specific to Panther Point stepping A0, and it is
wrongly implemented. Rewrite the whole function as per reference code.
Since this runs in SMM, be overly cautious and double-check everything.
Do not rely on GNVS to determine if xHCI is enabled. Instead, check
whether the corresponding bit in the Function Disable register is set.
Only Panther Point has xHCI, so exit early if this is not the case.
Change-Id: Iabce6c52fac781dc694f5b589fab2e9fe438f3f5
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49130
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
This will allow CSE RW FW updates and also fixes the problem where no sound is emitted from the speakers.
BUG=b:174338903
BRANCH=firmware-volteer-13672.B
TEST=emerge-volteer coreboot
Signed-off-by: hao_chou <hao_chou@pegatron.corp-partner.google.com>
Change-Id: I875f6b32c4053ef6d23ad7606cd35a129a78c306
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49290
Reviewed-by: Zhuohao Lee <zhuohao@google.com>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Compilation has been broken in commit I022468f6957415ae68a7a7e70428ae6f82d23b06
Adding a missing define solved this. See https://cgit.freebsd.org/src/tree/sys/sys/fcntl.h#n319
Signed-off-by: Idwer Vollering <vidwer@gmail.com>
Change-Id: I3433e4c9269880d3202dd494e5b2e962757a6b87
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49314
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
David Wu