This change adds a new driver for AMD I2S machine device. Currently,
this device is added as part of `acp_fill_ssdt()` in Picasso, but with
addition of this driver, this device can be added just like any other
device in the devicetree.
BUG=b:157708581
Change-Id: I49d1a867d7941397acca1054632b6ad855a021de
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43541
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This change increases the maximum length of device path string to 40
characters to accommodate growing hierarchy of devices.
TEST=Ensured that "\_SB.PCI0.LPCB.EC0.CREC.TUN0.RT58" is correctly
added to SSDT.
Change-Id: Id2ef71a32b26e366b56c652942a247de4889544a
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43540
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This change drops _INI and OSFL methods under \_SB since they are not doing
anything useful. _INI only calls OSFL and OSFL initializes OSVR if not
already initialized and returns OSVR value. However, OSVR is not used
anywhere and hence both these functions can be dropped.
BUG=b:153879530
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: I4f3e1c93a855006cc115087fded20bfb76c1133e
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43515
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Global variable `PMOD` that stores the interrupt mode used by OS is
required by all mainboards. This change moves the variable definition to
globalnvs.asl under picasso.
Additionally, ACPI spec says that BIOS should assume interrupt mode as PIC
until _PIC() method is called by OS. Thus, this change also updates the
default value of PMOD as 0 i.e. PIC mode.
BUG=b:153879530
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: I731c03d965882281a7a23f55894451210ba72274
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43514
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
GPIO_9 is associated with gevent 22. Correct all the misconfigurations
and use macros for clarity as to what bit offset is being used instead
of open coding things.
BUG=b:161205804
Change-Id: Ic4cfd62763d72d12a55f89585f24e07df6af0f4f
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43516
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Currently, the code and data in psp_verstage is ~59K. Adding the code
to save vbnv to the SPI rom increases that to 66K.
Getting rid of VERSTAGE_SIZE allows verstage to grow as it needs to.
BUG=b:161366241
TEST=Build & Boot Morphius with VBOOT_VBNV_CMOS_BACKUP_TO_FLASH enabled
Signed-off-by: Martin Roth <martinroth@chromium.org>
Change-Id: Ic6853b70073f9e781fc10402a2a47c9c8e0d49d3
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43486
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
On zork, bootblock is part of RW firmware in non-recovery mode, so PCIe
GPIOs can be configured early on in bootblock rather than waiting until
romstage. This change moves the call to variant_pcie_gpio_configure() to
happen in bootblock and drops romstage.c file.
BUG=b:154351731
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: Ic515304f35fe5623d58d6000efcb11fb9039e137
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43476
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
gpio_set_stage_rom table is now configuring only PCIe related GPIOs in
romstage. This change moves the configuration of PCIe related GPIOs to
variant_pcie_gpio_configure() to keep all the configuration for WiFi and
non-WiFi PCIe pads in one place. It also drops the function
variant_romstage_gpio_table() as it is unused.
BUG=b:154351731
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: Ib1c41ba141dce6b52b6e0a250a3aa07c296068aa
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43475
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Now that the power and reset GPIO configuration for non-PCIe devices is
dropped from romstage GPIO table, the tables for pre-v3 and v3 version of
schematics are exactly same. So, this change drops the duplicate table and
also removes the check for v3 schematics when configuring the pads in
romstage.
BUG=b:154351731
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: I67ca9f587c3f47912393ebaf38badcc9d76cc393
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43474
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Currently the common soc code automatically selects
POWER_STATE_DEFAULT_ON_AFTER_FAILURE which making other mainboard
options unselectable.
However, there're some cases that power state should change to different
states after reapplying power.
Make POWER_STATE_DEFAULT_ON_AFTER_FAILURE default y but do not select it
in soc code so that we can disable it and select other options in the
mainboard code.
Tested on OCP Delta Lake.
Signed-off-by: Tim Chu <Tim.Chu@quantatw.com>
Change-Id: Ifa853d81ee9477d2440ceaa67b2bd6b863ee52c2
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43338
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Implement sending POST start/end command to BMC.
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 May 28 13:21:22 bmc-oob. user.info fby3-v2020.20.2:
ipmid: POST Start Event for Payload#1
2020 May 28 13:21:25 bmc-oob. user.info fby3-v2020.20.2:
ipmid: POST End Event for Payload#1
root@bmc-oob:~#
Signed-off-by: TimChu <Tim.Chu@quantatw.com>
Change-Id: I38b512ee97c0eda6ba54482a448ef9ffc27b4ddb
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41993
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Jonathan Zhang <jonzhang@fb.com>
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 pin has been removed from recent revisions (or at least moved to a
pin that's not controlled like a normal GPIO). It has also never been
used on older ones (I think?). The same GPIO is now used as a SKU ID
pin, so make sure we're not incorrectly configuring it as an output
(although sku_id() overrides that later anyway, but it's still
incorrect).
BUG=b:160754995
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: I6e0d56e230d0bef172d7b78cc48263e9b6f059de
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43471
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Philip Chen <philipchen@google.com>
Reviewed-by: Bob Moragues <moragues@google.com>
Reviewed-by: EricR Lai <ericr_lai@compal.corp-partner.google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This information is redundant since it's already specified in
baseboard/devicetree_trembyle.cb or baseboard/devicetree_dalboz.cb
domain 0 is still required because sconfig uses it as an identity anchor
to match devicetree and overridetree.
BUG=b:157580724
TEST=Boot zork, usb functional
Change-Id: I3c3c1c2410166b99599d7343fae3ee756f4da321
Signed-off-by: Rob Barnes <robbarnes@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43437
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
If the AP actually needs to write to the TPM, then it is important and
the TPM should commit those changes to NVMEM immediately in case there
is an unexpected power loss (e.g. from a USB-C port partner reset upon
cold reboot request).
BRANCH=none
BUG=b:160913048
TEST=Verify that puff will no longer reboot loop when coreboot writes a
new Hmir (Hash mirror) in the TPM
Change-Id: I9597a55891d11bdf040d70f38b4c5a59c7888b8a
Signed-off-by: Jett Rink <jettrink@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43414
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
SataPortsEnableDitoConfig Enable DEVSLP Idle Timeout settings DmVal and
DitoVal. SataPortsDmVal and SataPortsDitoVal helps to determine when to
enter Device Sleep. Device Sleep enables the host bus adapter (HBA) to
assert the DEVSLP signal as soon as there are no commands outstanding
to the device and the port specific Device Sleep idle timer has expired.
Device Sleep Idle Timeout values (PxDEVSLP.DITO and PxDEVSLP.DM) are
port specific timeout values used by the HBA for determining when to
assert the DEVSLP signal. They provides a mechanism for the HBA to apply
a programmable amount of hysteresis so as to prevent the HBA from
asserting the DEVSLP signal too quickly which may result in undesirable
latencies. This patch is created based on Intel Tiger Lake Processor
PCH Datasheet with Document number:575857 and Chapter number:12.
* PxDEVSLP.DM -> SataPortsDmVal: Enable SATA Port DmVal DITO multiplier.
Default is 15.
* PxDEVSLP.DITO -> SataPortsDitoVal: Enable SATA Port DmVal DEVSLP Idle Timeout
(DITO), Default is 625ms.
BUG=b:151163106
BRANCH=None
TEST=Build and boot volteer and TGL RVP.
Signed-off-by: Shaunak Saha <shaunak.saha@intel.com>
Change-Id: I6a824524738a9e0609f54bea9d892b4a42a1d3db
Reviewed-on: https://review.coreboot.org/c/coreboot/+/42214
Reviewed-by: Wonkyu Kim <wonkyu.kim@intel.com>
Reviewed-by: Caveh Jalali <caveh@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
0xc50, 0xc52, 0xc6f don't exist on Picasso. The PCI config space
registers define SATA and OHCI which are at the wrong bus locations.
I just remove the whole section since it's not used. We never access the
PCIe Error region, or the PM2 region either.
BUG=b:153001807, b:154756391
TEST=Build Trembyle
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: I98aee09770f1df9f553c94580c1ee00c06a9cec1
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43427
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Furquan Shaikh <furquan@google.com>
The ACPI specification, version 2.0 says:
_BFS is an optional control method.
So, remove them. They have been copy-pasted around quite a bit, and do
not do anything useful. Plus, it's deprecated in later ACPI versions.
Change-Id: I9ef21f231dd6051d410ac3a0fe554908409c2fa7
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43443
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Forgetting to add the #pragma pack() at the end of the header file can
lead to hard to debug breakage, so get rid of the #pragma pack usage and
add a __packed to the structs that need to be packed which has less
possibly unwanted side effects.
Since commit d44221f9c8 coreboot always
includes commonlib/compiler.h which provides __packed.
TEST=Timeless build results in identical binary.
Change-Id: Icc53168f4fbc3a63a859f686b18e7023d225f8d2
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43433
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
