With the updated FSP UPD headers there are enough DXIO descriptor slots
in the UPD, so we can now add asserts to make sure that the mainboard
doesn't pass more DXIO/DDI descriptors than the UPD has slots for. This
is part of the DXIO/DDI descriptor handling cleanup.
BUG=b:158695393
Change-Id: Ia220d5a9d4ff11707b795b04662ff7eead4e2888
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43435
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
A new version of UPD headers generated from the FSP tree. This adds UPDs
for downcoring and increases the number of DXIO descriptor slots.
BUG=b:161152720
TEST=SATA on Mandolin works now.
Cq-Depend: chrome-internal:3175393
Change-Id: I1e27597e22af4df65d206a38b67c4920298b30b2
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43659
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Use macros to configure each of the IIO ports instead of an array
of some unknown parameters. This will clean up the code and make
it easier to read.
Tested with BUILD_TIMELESS=1, Tioga Pass, remains identical.
Change-Id: I2911992435a6c93624525426d56212f821abb866
Signed-off-by: Maxim Polyakov <max.senia.poliak@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43502
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Jonathan Zhang <jonzhang@fb.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
When RTC failure is detected, send IPMI OEM command to issue CMOS clear.
This is to let the payload (LinuxBoot) handle the IPMI OEM CMOS
clear command by resetting RTC data, erasing RW_VPD (TODO) and add a
SEL, then reboot the system.
Tested=on OCP Delta Lake, after removing RTC battery we can see the above
flow can be executed correctly.
Signed-off-by: Jingle Hsu <jingle_hsu@wiwynn.com>
Change-Id: I27428c02e99040754e15e07782ec1ad8524def2f
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43005
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
1. Populate SMBIOS data from OCP_DMI driver read from FRU
2. Set the read PPIN MSR for CPU0 and CPU1 to BMC, selecting
PARALLEL_MP_AP_WORK to enable OCP DMI driver to read remote socket PPIN.
Tested on OCP Tioga Pass.
Signed-off-by: Johnny Lin <johnny_lin@wiwynn.com>
Change-Id: Ie11ab68267438ea9c669c809985c0c2d7578280e
Reviewed-on: https://review.coreboot.org/c/coreboot/+/40524
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The Raydium ACPI entry currently provides a reset GPIO and an _ON/_OFF
method to the kernel. These are contradictory. The ownership of the GPIO
should be mutually exclusive between either the OS or the FW. Since we
have two methods exposed this causes the OS to reset the TS twice. Once
using the _ON method, and once using the GPIO. Additionally the _ON
method is waiting for 20ms after reset while the OS driver uses a 50ms
delay. The Raydium TS datasheet specifies 20ms for FW ready time, so the
OS driver is adding additional padding.
The reference design has a 32ms rise time on the reset line. So without
this patch, the OS tries to reset the TS using the _ON method and it
waits for 20ms. This is not enough time for the reset line to reach
high, let alone account for the FW ready time. The OS driver then tries
to reset the device by toggling the GPIO. It waits 50ms which is still
2ms less than required.
This CL removes the GPIO from being exported in the _CRS so the OS
driver won't try and reset the device. It also increases the reset delay
by 32ms to account for the rise time.
This isn't a complete fix. I think that the slow rise time is causing
some kind of metastability in the TS reset hardware. Using a script to
bind and unbind the TS driver, the TS device becomes unresponsive after
~200 iterations. The only way to reset the device is to power cycle.
The TS power is also not currently controlled by the power resource.
This means that we have no guarantee over when the reset line is
toggled. This will lead to issues while spending and resuming.
BUG=b:160854397
TEST=Boot trembyle and make sure TS works. Suspend/Resume trembyle 300+
times.
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: I23131be5d7109eed660a8bd6e2c156c015aa3c4e
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43467
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Earlier versions of Dalboz did not correctly handle HS400. One fix was
to add stitching vias, but these boards did not have them. b/156539551
Another possible fix is to add tuning parameters including drive
strength, but that is still a WIP. b/158959725
This should correct OS load failures in the meantime by running the bus
slower.
BUG=b:158845662
TEST=build, flash, boot sku 0x5a80000c to OS
BRANCH=None
Signed-off-by: Eric Peers <epeers@google.com>
Change-Id: Ia3e7a641bde04c5a7be29bf91c38dd8c110ed17a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43572
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Create the dirinboz variant of the dalboz reference board by copying
the template files to a new directory named for the variant.
(Auto-Generated by create_coreboot_variant.sh version 4.1.1).
BUG=b:161579679
BRANCH=master
TEST=util/abuild/abuild -p none -t google/zork -x -a
make sure the build includes GOOGLE_DIRINBOZ
Change-Id: I33c03080ffbe0bca61acf4144417b9f5fff6389f
Signed-off-by: Kevin Chiu <kevin.chiu@quantatw.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43601
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martinroth@google.com>
Reset the Touchscreen (TS) and disable the stop GPIO (report switch) at
the same time. Add a delay of 100 ms after disabling the stop GPIO. This
will ensure the required delay is inserted for both reset and stop
disable GPIOs simultaneously.
BUG=b:152936541
TEST=Build and boot the waddledoo mainboard. Ensure that the SiS
Touchscreen is functional.
Change-Id: Icbfb5e07a28ab72b1ff696ad1183a6c2173dcaac
Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43453
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Justin TerAvest <teravest@chromium.org>
Reviewed-by: Marco Chen <marcochen@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change does the following:
a. USI_REPORT_EN is no longer set to high in coreboot. Instead
GPIO_144 is exposed as stop_gpio in ACPI to allow OS to control this
pad as required.
b. Appropriate delays are added for power-down sequencing:
- Delay after REPORT_EN is disabled - 1ms
- Delay after RESET is asserted - 1ms
BUG=b:159501288
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: If4d12fa0d4f4e5123d8fdccdabda996dcafa4523
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43701
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This change configures all missing pads in ramstage for dalboz
reference. This ensures that the state of all pads is set correctly
for the payload/OS. Also, all the pads for the platform are configured
in baseboard gpio table in ramstage to ensure that variants can
override any pads if required.
BUG=b:154351731
Change-Id: Ia30da908d3827177a7b3594ffba38bff81018ab9
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43698
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This change configures all missing pads in ramstage for trembyle
reference. This ensures that the state of all pads is set correctly
for the payload/OS. Also, all the pads for the platform are configured
in baseboard gpio table in ramstage to ensure that variants can
override any pads if required.
BUG=b:154351731
Change-Id: Idd827b6a4f995546493596f22249f8699bdf526b
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43697
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Add function to send POST start command to BMC. This function is
used in romstage and the POST end command will be sent in u-root.
TEST=Read POST command log in OpenBMC,
if command received successfully, message may show as below,
root@bmc-oob:~# cat /var/log/messages |grep -i "POST"
2020 Jul 15 16:36:11 bmc-oob. user.info fby3-v2020.23.1:
ipmid: POST Start Event for Payload#2
root@bmc-oob:~#
Signed-off-by: TimChu <Tim.Chu@quantatw.com>
Change-Id: Ide0e2a52876db555ed8b5e919215e85731fd80ed
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41605
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
SystemMemoryMapHob is necessary for SMBIOS type 17 among other things.
It is a fairly large structure, so the pointer to the data instead of
the structure itself, is included in the HOB. Use pointer to
SystemMemoryMapHob structure to interpret SystemMemoryHob HOB body.
Adjust the structure definition to match with CPX-SP ww28 release.
Display more fields to ensure the structure definition is correct.
TEST=Boot DeltaLake server, and check field values of SystemMemoryMapHob
to make sure they are correct:
0x7590a090, 0x00000020 bytes: HOB_TYPE_GUID_EXTENSION
f8870015-6994-4b98-95a2bd56da91c07f: FSP_SYSTEM_MEMORYMAP_HOB_GUID
================== MEMORY MAP HOB DATA ==================
hob: 0x777f7000, structure size: 0x6c88
lowMemBase: 0x0, lowMemSize: 0x20, highMemBase: 0x40, highMemSize: 0x5d0
memSize: 0x600, memFreq: 0xb76
NumChPerMC: 3
SystemMemoryMapElement Entries: 2, entry size: 16
memory_map 0 BaseAddress: 0x0, ElementSize: 0x20, Type: 0x1
memory_map 1 BaseAddress: 0x40, ElementSize: 0x5d0, Type: 0x1
BiosFisVersion: 0x0
MmiohBase: 0x80000
0x777f7000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
...
Signed-off-by: Jonathan Zhang <jonzhang@fb.com>
Change-Id: I271bcbd6030276b8fcd99d5b4f2c93f034dd9b52
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43336
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This patch enables the SkipCpuReplacementCheck config for jasperlake rvp
to avoid the forced MRC training with the soldered down SOC.
BUG=b:160201335
BRANCH=None
TEST=Build and verify on jasperlake rvp with CSE Lite SKU.
Cq-Depend: chrome-internal:3142530
Change-Id: I40fb9a25170e8db3c63a71428ba459160a918961
Signed-off-by: Krishna Prasad Bhat <krishna.p.bhat.d@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43146
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Reviewed-by: Sridhar Siricilla <sridhar.siricilla@intel.com>
Reviewed-by: Rizwan Qureshi <rizwan.qureshi@intel.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This patches enables the SkipCpuReplacementCheck config for
the dedede baseboard to avoid the forced MRC training for all
its variants with the soldered down SOC.
BUG=b:160201335
TEST=Build and verify CSE Lite SKU on Waddledoo.
Cq-Depend: chrome-internal:3142530
Change-Id: I611e66f74a3b9b090ab5e0d836231643d3f919dc
Signed-off-by: V Sowmya <v.sowmya@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/42930
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Reviewed-by: Rizwan Qureshi <rizwan.qureshi@intel.com>
Add SkipCpuReplacementCheck config to control the FSPM UPD used
for skipping the CPU replacementment check to avoid the forced
MRC training for the platforms with soldered down SOC.
BUG=b:160201335
TEST=Build and verify CSE Lite SKU on Waddleddo.
Cq-Depend: chrome-internal:3142530
Change-Id: I63fcdab3686322406cf7c24fc26cbb535cc58c8d
Signed-off-by: V Sowmya <v.sowmya@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/42453
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Reviewed-by: Sridhar Siricilla <sridhar.siricilla@intel.com>
Reviewed-by: Rizwan Qureshi <rizwan.qureshi@intel.com>
A recent Coverity scan found an issue with the way the
EC_HOST_EVENT_MASK macro was being used. It was being passed values
between 0 and 63, but since it is doing basically (1ULL << (value - 1)),
this caused a shift of -1 when `i` is 0 and also doesn't reach the 63rd
bit of the mask. This is fixed by incrementing the start and end
conditions of the loop by 1, so the event mask ranges from bits 0 to 63,
instead of -1 to 62.
Found-by: Coverity CID 1430218
Signed-off-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Change-Id: I6a7cfa64545f3d313de24407f0a91b48368f2a8a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43460
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Change PCH power policy. Set default of
POWER_STATE_DEFAULT_ON_AFTER_FAILURE to n in order to change power
state to S5 when power is reapplied after power failure.
TEST=Base on CB:42289, CB:43338 and build for Deltalake.
The following Kconfig options must be selected:
select SOC_INTEL_COMMON_BLOCK_PMC
select ACPI_INTEL_HARDWARE_SLEEP_VALUES
select CPU_INTEL_COMMON_SMM
Boot the system and check the last bit of GEN_PMCON_B is set to 1
through ITP with command: pch.pm_dump
Signed-off-by: Tim Chu <Tim.Chu@quantatw.com>
Change-Id: I4d4f14bdfc18740976171fd5d369b2d79a916dc4
Reviewed-on: https://review.coreboot.org/c/coreboot/+/42976
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Jonathan Zhang <jonzhang@fb.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
1. Read VPD variable 'fsp_log_level' to decide FSP log level.
2. Define the default values when the VPD variables cannot be found,
put all the values to vpd.h for better documentation and maintenance.
Tested=On OCP DeltaLake, the fsp_log_level can be changed from the VPD variable.
Change-Id: I44cd59ed0c942c31aaf95ed0c8ac78eb7d661123
Signed-off-by: Johnny Lin <johnny_lin@wiwynn.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43606
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
The three mainboards using this southbridge are desktop boards, which
are not dockable. The Dell Precision M6400 laptop is dockable, but even
though it has an Eaglelake MCH, it uses an i82801ix southbridge instead.
So, one could still port that laptop to coreboot after this change! :P
Also, drop the now-unnecessary `chip` and `dev` variables.
Change-Id: Ic9ab497c91d66032929190cde22d59a208887f50
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/42657
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Reviewed-by: Nico Huber <nico.h@gmx.de>
