This patch enriches coreboot FSP2.0 driver to extract the FSP timestamp
from FPDT (Firmware Performance Data Table) and display right after
FSP-S exits (from `fsp_silicon_init()` function), based on SoC user
selects the required `DISPLAY_FSP_TIMESTAMPS` config.
The prerequisite to this implementation is to have FSP binary built with
`PcdFspPerformanceEnable` PCD set to `TRUE` to allow FSP to populate
the FPDT HOB.
BUG=b:216635831
TEST=Able to dump FSP performance data with DISPLAY_FSP_TIMESTAMPS
Kconfig selected and met the FSP prerequisites.
+--------------------------------------------------+
|------ FSP Performance Timestamp Table Dump ------|
+--------------------------------------------------+
| Perf-ID Timestamp(ms) String/GUID |
+--------------------------------------------------+
0 460253 SEC/52c05b14-0b98-496c-bc3b04b50211d680
50 460263 PEI/52c05b14-0b98-496c-bc3b04b50211d680
40 460274 PreMem/52c05b14-0b98-496c-bc3b04b50211d680
1 495803 9b3ada4f-ae56-4c24-8deaf03b7558ae50
2 508959 9b3ada4f-ae56-4c24-8deaf03b7558ae50
1 515253 6141e486-7543-4f1a-a579ff532ed78e75
2 525453 6141e486-7543-4f1a-a579ff532ed78e75
1 532059 baeb5bee-5b33-480a-8ab7b29c85e7ceab
2 546806 baeb5bee-5b33-480a-8ab7b29c85e7ceab
1 553302 1b04374d-fa9c-420f-ac62fee6d45e8443
2 563859 1b04374d-fa9c-420f-ac62fee6d45e8443
1 569955 88c17e54-ebfe-4531-a992581029f58126
2 575753 88c17e54-ebfe-4531-a992581029f58126
1 582099 a8499e65-a6f6-48b0-96db45c266030d83
50f0 599599 unknown name/3112356f-cc77-4e82-86d53e25ee8192a4
50f1 716649 unknown name/3112356f-cc77-4e82-86d53e25ee8192a4
2 728507 a8499e65-a6f6-48b0-96db45c266030d83
1 734755 9e1cc850-6731-4848-87526673c7005eee
....
Signed-off-by: Subrata Banik <subratabanik@google.com>
Change-Id: Ia1b7f6b98bafeec0afe843f0f78c99c2f34f50b3
Reviewed-on: https://review.coreboot.org/c/coreboot/+/62942
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Kangheui Won <khwon@chromium.org>
Reviewed-by: Eric Lai <eric_lai@quanta.corp-partner.google.com>
Reviewed-by: Michael Niewöhner <foss@mniewoehner.de>
This patch implements coreboot native debug handler to manage the FSP
event messages.
`FSP Event Handlers` feature introduced in FSP to generate event
messages to aid in the debugging of firmware issues. This eliminates
the need for FSP to directly write debug messages to the UART and FSP
might not need to know the board related UART port configuration.
Instead FSP signals the bootloader to inform it of a new debug message.
This allows the coreboot to provide board specific methods of reporting
debug messages, example: legacy UART or LPSS UART etc.
This implementation has several advantages as:
1. FSP relies on XIP `DebugLib` driver even while printing FSP-S debug
messages, hence, without ROM being cached, post `romstage` would
results into sluggish boot with FSP debug enabled.
This patch utilities coreboot native debug implementation which is
XIP during FSP-M and relocatable to DRAM based resource for FSP-S.
2. This patch simplifies the FSP DebugLib implementation and remove the
need to have serial port library. Instead coreboot `printk` can be
used for display FSP serial messages. Additionally, unifies the debug
library between coreboot and FSP.
3. This patch is also useful to get debug prints even with FSP
non-serial image (refer to `Note` below) as FSP PEIMs are now
leveraging coreboot debug library instead FSP `NULL` DebugLib
reference for release build.
4. Can optimize the FSP binary size by removing the DebugLib dependency
from most of FSP PEIMs, for example: on Alder Lake FSP-M debug binary
size is reduced by ~100KB+ and FSP-S debug library size is also
reduced by ~300KB+ (FSP-S debug and release binary size is exactly
same with this code changes). The total savings is ~400KB for each
FSP copy, and in case of Chrome AP firmware with 3 copies, the total
savings would be 400KB * 3 = ~1.2MB.
Note: Need to modify FSP source code to remove `MDEPKG_NDEBUG` as
compilation flag for release build and generate FSP binary with non-NULL
FSP debug wrapper module injected (to allow FSP event handler to execute
even with FSP non-serial image) in the final FSP.fd.
BUG=b:225544587
TEST=Able to build and boot brya. Also, verified the FSP debug log is
exactly same before and with this code change.
Signed-off-by: Subrata Banik <subratabanik@google.com>
Change-Id: I1018e67d70492b18c76531f9e78d3b58fa435cd4
Reviewed-on: https://review.coreboot.org/c/coreboot/+/63007
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
FSP 2.3 spec introduced new version of NV storage HOB
FSP_NON_VOLATILE_STORAGE_HOB2. This new HOB addresses the limitation of
FSP_NON_VOLATILE_STORAGE_HOB which can support data length
upto 64KB. FSP_NON_VOLATILE_STORAGE_HOB2 allows >64KB of NVS data to be
stored by specifying a pointer to the NVS data.
FSP_NON_VOLATILE_STORAGE_HOB HOB is deprecated
from FSP 2.3 onwards and is maintained for backward compatibility only.
This patch implements the parsing method for
FSP_NON_VOLATILE_STORAGE_HOB2 HOB structure .The HOB list is first
searched for FSP_NON_VOLATILE_STORAGE_HOB2. If not found we continue
to search for FSP_NON_VOLATILE_STORAGE_HOB HOB.
BUG=b:200113959
TEST=Verified on sapphire rapids and meteor lake FSP platform that
introduces FSP_NON_VOLATILE_STORAGE_HOB2 for retrieving MRC cached data.
Signed-off-by: Anil Kumar <anil.kumar.k@intel.com>
Signed-off-by: Subrata Banik <subratabanik@google.com>
Change-Id: I27647e9ac1a4902256b3f1c34b60e1f0b787a06e
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59638
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Wonkyu Kim <wonkyu.kim@intel.com>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
This reverts a change that was causing hangs and exceptions during boot
on an ADL brya4es.
The hang (or APIC exception) occurs at what appears to be the FSP MP
initialization sequence, prior to the "Display FSP Version Info HOB"
log being displayed :
[DEBUG] Detected 10 core, 12 thread CPU.
[DEBUG] Display FSP Version Info HOB
This reverts commit 40ca79714a.
BUG=b:224873032
TEST=`emerge-brya coreboot chromeos-bootimage`, flash and verify brya4es
is able to successfully reboot 200 times without any issues.
Change-Id: I88c15a51c5d27fbd243478c923e75962d3f8d67d
Signed-off-by: Nick Vaccaro <nvaccaro@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/62907
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Subrata Banik <subratabanik@google.com>
Reviewed-by: Ronak Kanabar <ronak.kanabar@intel.com>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
This patch aims to make timestamps more consistent in naming,
to follow one pattern. Until now there were many naming patterns:
- TS_START_*/TS_END_*
- TS_BEFORE_*/TS_AFTER_*
- TS_*_START/TS_*_END
This change also aims to indicate, that these timestamps can be used
to create time-ranges, e.g. from TS_BOOTBLOCK_START to TS_BOOTBLOCK_END.
Signed-off-by: Jakub Czapiga <jacz@semihalf.com>
Change-Id: I533e32392224d9b67c37e6a67987b09bf1cf51c6
Reviewed-on: https://review.coreboot.org/c/coreboot/+/62019
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Yu-Ping Wu <yupingso@google.com>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Set the log level to BIOS_NOTICE for the case where the mainboard can
not provide a MAC address since this can be a valid case. Showing this
message with log level BIOS_ERR is not appropriate.
In addition, rephrase the message to make clear that if the mainboard
does not provide a MAC address the one stored in the MAC will be used.
Change-Id: Ibfc58845f0ea47ced048b446e685c4860a29f075
Signed-off-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/62008
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Mario Scheithauer <mario.scheithauer@siemens.com>
Shorten define names containing PCI_{DEVICE,VENDOR}_ID_ with
PCI_{DID,VID}_ using the commands below, which also take care of some
spacing issues. An additional clean up of pci_ids.h is done in
CB:61531.
Used commands:
* find -type f -exec sed -i 's/PCI_\([DV]\)\(EVICE\|ENDOR\)_ID_\([_0-9A-Za-z]\{2\}\([_0-9A-Za-z]\{8\}\)*[_0-9A-Za-z]\{0,5\}\)\t/PCI_\1ID_\3\t\t/g'
* find -type f -exec sed -i 's/PCI_\([DV]\)\(EVICE\|ENDOR\)_ID_\([_0-9A-Za-z]*\)/PCI_\1ID_\3/g'
Change-Id: If9027700f53b6d0d3964c26a41a1f9b8f62be178
Signed-off-by: Felix Singer <felixsinger@posteo.net>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39331
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Michael Niewöhner <foss@mniewoehner.de>
This patch renames all FSP Notify Phase API configs to primarily remove
"SKIP_" prefix.
1. SKIP_FSP_NOTIFY_PHASE_AFTER_PCI_ENUM ->
USE_FSP_NOTIFY_PHASE_POST_PCI_ENUM
2. SKIP_FSP_NOTIFY_PHASE_READY_TO_BOOT ->
USE_FSP_NOTIFY_PHASE_READY_TO_BOOT
3. SKIP_FSP_NOTIFY_PHASE_END_OF_FIRMWARE ->
USE_FSP_NOTIFY_PHASE_END_OF_FIRMWARE
The idea here is to let SoC selects all required FSP configs to execute
FSP Notify Phase APIs unless SoC deselects those configs to run native
coreboot implementation as part of the `.final` ops.
For now all SoC that uses FSP APIs have selected all required configs
to let FSP to execute Notify Phase APIs.
Note: coreboot native implementation to skip FSP notify phase API (post
pci enumeration) is still WIP.
Additionally, fixed SoC configs inclusion order alphabetically.
BUG=b:211954778
TEST=Able to build and boot brya.
Signed-off-by: Subrata Banik <subratabanik@google.com>
Change-Id: Ib95368872acfa3c49dad4eb7d0d73fca04b4a1fb
Reviewed-on: https://review.coreboot.org/c/coreboot/+/61792
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Lean Sheng Tan <sheng.tan@9elements.com>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Guard macro via CONFIG_INTEL_GMA_ADD_VBT, rather than guarding
each of the calls to it (most of which are currently unguarded).
Test: build google/coral w/ and w/o CONFIG_INTEL_GMA_ADD_VBT selected,
verify VBTs added (or not) to CBFS based on Kconfig selection.
Change-Id: Ic25554cb2c61b81bdb4b0987094c3558e0bbcbd8
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/61768
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Now that the console system itself will clearly differentiate loglevels,
it is no longer necessary to explicitly add "ERROR: " in front of every
BIOS_ERR message to help it stand out more (and allow automated tooling
to grep for it). Removing all these extra .rodata characters should save
us a nice little amount of binary size.
This patch was created by running
find src/ -type f -exec perl -0777 -pi -e 's/printk\(\s*BIOS_ERR,\s*"ERROR: /printk\(BIOS_ERR, "/gi' '{}' ';'
and doing some cursory review/cleanup on the result. Then doing the same
thing for BIOS_WARN with
's/printk\(\s*BIOS_WARNING,\s*"WARN(ING)?: /printk\(BIOS_WARNING, "/gi'
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: I3d0573acb23d2df53db6813cb1a5fc31b5357db8
Reviewed-on: https://review.coreboot.org/c/coreboot/+/61309
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Reviewed-by: Lance Zhao
Reviewed-by: Jason Glenesk <jason.glenesk@gmail.com>
This patch aligns fsp_header with the Intel specification 2.0 and 2.3.
The main impetus for this change is to make the fsp_info_header fully
accessible in soc/vendor code. Here items such as image_revision can be
checked.
TEST=verify image revision output in the coreboot serial log.
compare to FSP version shown in serial debug output.
verify Google Guybrush machine boots into OS.
Signed-off-by: Julian Schroeder <julianmarcusschroeder@gmail.com>
Change-Id: Ibf50f16b5e9793d946a95970fcdabc4c07289646
Reviewed-on: https://review.coreboot.org/c/coreboot/+/58869
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Current max count for camera power ops is 5 which is not sufficient.
If we increase the ops by 1 in current variants the compiler
will not throw error for intel mipi camera driver.
Hence increase current max count for camera power ops to 6 from 5.
BUG=b:214665783
TEST=Build and boot to OS
Signed-off-by: Varshit B Pandya <varshit.b.pandya@intel.com>
Change-Id: I4f4c090f2275616816dfc697f27520cd1cbc1a80
Reviewed-on: https://review.coreboot.org/c/coreboot/+/61146
Reviewed-by: Subrata Banik <subratabanik@google.com>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Without acpi name, acpi_device_path will return NULL.
<NULL>: Intel USB4 Retimer at GENERIC: 0.0
Replace with usb4_retimer_scope for the identify.
BUG=b:215742472
TEST=show below meaasge in coreboot log
\_SB.PCI0.TMD0.HR : Intel USB4 Retimer at GENERIC: 0.0
\_SB.PCI0.TMD1.HR : Intel USB4 Retimer at GENERIC: 0.0
Signed-off-by: Eric Lai <ericr_lai@compal.corp-partner.google.com>
Change-Id: Idfa8b204894409b11936e5f221c218daa206cc02
Reviewed-on: https://review.coreboot.org/c/coreboot/+/61315
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
The FSP API is used to notify the FSP about different phases in the
boot process. The current FSP specification supports three notify
phases:
- Post PCI enumeration
- Ready to Boot
- End of Firmware
This patch attempts to make calling into the FSP Notify Phase APIs
optional by using native coreboot implementations to perform the
required lock down and chipset register configuration prior boot to
payload.
BUG=b:211954778
TEST=Able to build brya without any compilation issue and coreboot
log with this code changes when SKIP_FSP_NOTIFY_PHASE_READY_TO_BOOT
and SKIP_FSP_NOTIFY_PHASE_END_OF_FIRMWARE config enabled.
coreboot skipped calling FSP notify phase: 00000040.
coreboot skipped calling FSP notify phase: 000000f0.
Signed-off-by: Subrata Banik <subratabanik@google.com>
Change-Id: Ia95e9ec25ae797f2ac8e1c74145cf21e59867d64
Reviewed-on: https://review.coreboot.org/c/coreboot/+/60402
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Lean Sheng Tan <sheng.tan@9elements.com>
Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-by: EricR Lai <ericr_lai@compal.corp-partner.google.com>
Reviewed-by: Paul Menzel <paulepanter@mailbox.org>
FSP 2.3 specification introduces following changes:
1. FSP_INFO_HEADER changes
Updated SpecVersion from 0x22 to 0x23
Updated HeaderRevision from 5 to 6
Added ExtendedImageRevision
FSP_INFO_HEADER length changed to 0x50
2. Added FSP_NON_VOLATILE_STORAGE_HOB2
Following changes are implemented in the patch to support FSP 2.3:
- Add Kconfig option
- Update FSP build binary version info based on ExtendedImageRevision
field in header
- New NV HOB related changes will be pushed as part of another patch
Signed-off-by: Anil Kumar <anil.kumar.k@intel.com>
Change-Id: Ica1bd004286c785aa8a431f39d8efc69982874c1
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59324
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-by: Subrata Banik <subratabanik@google.com>
Currently, the pmc_mux/conn driver uses integer fields to store the
USB-2 and USB-3 port numbers from the SoC's point of view. Specifying
these as integers in the devicetree is error-prone, and this
information can instead be represented using pointers to the USB-2 and
USB-3 devices. The port numbers can then be obtained from the paths of
the linked devices, i.e. dev->path.usb.port_id.
Modify the driver to store device pointers instead of integer port
numbers, and update all devicetrees using the driver. These are the
mainboards affected (all are Intel TGL or ADL based):
google/brya
google/volteer
intel/adlrvp
intel/shadowmountain
intel/tglrvp
system76/darp7
system76/galp5
system76/lemp10
Command used to update the devicetrees:
git grep -l "usb._port_number" src/mainboard/ | \
xargs sed -i \
-e 's/register "usb2_port_number" = "\(.*\)"/use usb2_port\1 as usb2_port/g' \
-e 's/register "usb3_port_number" = "\(.*\)"/use tcss_usb3_port\1 as usb3_port/g'
BUG=b:208502191
TEST=Build test all affected boards. On brya0, boot device and check
that the ACPI tables generated with and without the change are the same.
Change-Id: I5045b8ea57e8ca6f9ebd7d68a19486736b7e2809
Signed-off-by: Reka Norman <rekanorman@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/60143
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Tim Crawford <tcrawford@system76.com>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Currently coreboot interprets TCSS port number as per physical port
number while EC abstracts port number and provides indices as port
number. For example, if TCSS port 1 and 3 are enabled on the board,
coreboot will interpret port numbers as 0 and 2, but since only 2 ports
are enabled in the system EC will assign port numbers as 0 and 1.
This creates a port number mismatch while communicating between EC and
coreboot. This patch addresses issue where SoC can implement function
to map correct EC port as per port enabled in mainboard.
BUG=b:207057940
BRANCH=None
TEST=Check if code compiles successfully. Functionality will work once
function is implemented in SoC code.
Change-Id: Ia7a5e63838e6529196bd211516e4d665b084f79e
Signed-off-by: MAULIK V VAGHELA <maulik.v.vaghela@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59665
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Add entry in ACPI table under IPU device to provide silicon type
information to IPU driver. IPU kernel driver can decide the type of
firmware to load based on this information.
BUG=b:207721978
BRANCH=none
TEST=Check for the ACPI entry in the SSDT after booting to kernel
Change-Id: I4e0af1dd50b9c014cae5454fcd4f9f76d0e0a85f
Cq-Depend: chromium:3319905
Signed-off-by: Sugnan Prabhu S <sugnan.prabhu.s@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59869
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Sridhar Siricilla <sridhar.siricilla@intel.com>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
The _DSC (Device State for Configuration) object evaluates to an integer
may be used to tell Linux the highest allowed D state for a device
during probe. The support for _DSC requires support from the kernel
bus type if the bus driver normally sets the device in D0 state for
probe.
The D states and thus also the allowed values for _DSC are listed below.
Number State Description
0 D0 Device fully powered on
1 D1
2 D2
3 D3hot
4 D3cold Off
More details can be found here https://lkml.org/lkml/2021/10/25/397
BUG=none
BRANCH=none
TEST=Add corresponding field in brya, boot and dump SSDT to check if
_DSC field is as per expectation.
Name (_ADR, Zero) // _ADR: Address
Name (_HID, "OVTI8856") // _HID: Hardware ID
Name (_UID, Zero) // _UID: Unique ID
Name (_DDN, "Ov 8856 Camera") // _DDN: DOS Device Name
Method (_STA, 0, NotSerialized) // _STA: Status
{
Return (0x0F)
}
Method (_DSC, 0, NotSerialized)
{
Return (0x04)
}
Signed-off-by: Varshit B Pandya <varshit.b.pandya@intel.com>
Change-Id: I5471f144918413a2982f86beaf3dbf7e4e66cc9b
Reviewed-on: https://review.coreboot.org/c/coreboot/+/58767
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Maulik V Vaghela <maulik.v.vaghela@intel.com>
Reviewed-by: Subrata Banik <subrata.banik@intel.com>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Currently, the MMCONF Kconfigs only support the Enhanced Configuration
Access mechanism (ECAM) method for accessing the PCI config address
space. Some platforms have a different way of mapping the PCI config
space to memory. This patch renames the following configs to
make it clear that these configs are ECAM-specific:
- NO_MMCONF_SUPPORT --> NO_ECAM_MMCONF_SUPPORT
- MMCONF_SUPPORT --> ECAM_MMCONF_SUPPORT
- MMCONF_BASE_ADDRESS --> ECAM_MMCONF_BASE_ADDRESS
- MMCONF_BUS_NUMBER --> ECAM_MMCONF_BUS_NUMBER
- MMCONF_LENGTH --> ECAM_MMCONF_LENGTH
Please refer to CB:57861 "Proposed coreboot Changes" for more
details.
BUG=b:181098581
BRANCH=None
TEST=./util/abuild/abuild -p none -t GOOGLE_KOHAKU -x -a -c max
Make sure Jenkins verifies that builds on other boards
Change-Id: I1e196a1ed52d131a71f00cba1d93a23e54aca3e2
Signed-off-by: Shelley Chen <shchen@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/57333
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
