Add minimal MP init. No SMM, no turbo, not c/p states.
TEST=boot linux kernel, observe CPUs are online, schedule tasks
and perform useful work. Tested on Cedar Island CRB with only 1
socket populated
Change-Id: I0af374ab3956009e9208917d911d29eb21db6069
Signed-off-by: Andrey Petrov <anpetrov@fb.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/40035
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Maxim Polyakov <max.senia.poliak@gmail.com>
Done with sed and God Lines. Only done for C-like code for now.
Change-Id: I29e746115e3b0630238176a0f913a3b5340962eb
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/40047
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Done with sed and God Lines. Only done for C-like code for now.
Change-Id: Ib8a40e5633faf840e19a35bcdc8edc7e7cdd0ad9
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/40048
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Our realloc() works (somewhat suboptimally) by free()ing the existing
allocation and then reallocating it wherever it fits. If there was free
space before the old location, this means the new allocation may be
before the old one, and if the free space block is smaller than the old
allocation it may overlap. Thus, we should be moving memmove() instead
of memcpy() to move the block over.
This is not a problem in practice since all our existing memcpy()s are
simple iterate and copy front to back implementations which are safe for
overlaps when the destination is in front of the source. but it's still
the more correct thing to do (in case we ever change our memcpy()s to do
something more advanced or whatever).
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: I35f77a94b7a72c01364ee7eecb5c3ff5ecde57f6
Reviewed-on: https://review.coreboot.org/c/coreboot/+/40028
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Disable SoC serial output by default, since no production devices
have this exposed, but leave it as a user option so it can be selected
as needed (eg, for use with a Google debug servo).
Same change as made for google/cyan in CB:39872
Change-Id: Id6b2c28658aca03d8c5042d719a0f6f504c29288
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39997
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Since the variants' devicetrees are almost identical, convert to
using an overridetree setup for simplicity.
Change-Id: I52b71cf12a4e0b67135cfb106c3e89b00205d3bc
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39996
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The jack detect GPIOs are initialized as dual edge-triggered GPIs,
and Linux doesn't care if they are set to ActiveLow, ActiveHigh, or
ActiveBoth -- a single interrupt is detected on jack insertion or
removal.
The Windows drivers on the other hand, will not function unless the
codec and LPE ACPI interrupts entries are set as in the Intel
Cherry Trail Tianocore platform reference code.
So adjust the ACPI interrupt triggers to make Windows happy, since
Linux doesn't care either way.
Test: boot Linux (GalliumOS 3.1) and Windows 10 on google/edgar,
observe functional audio output for both built-in speakers and
headphones.
Change-Id: Ic1dd8ece610d761791c060ece2d0aa51addf97ad
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/24989
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Adjust CID to allow for Realtek's Windows drivers to attach
without breaking functionality under Linux.
Both Linux and Windows use ACPI HID/CID matching for driver attachment.
Since the Realtek 5650 isn't used in standard Windows laptops, the
'10EC5650' HID/CID isn't contained in the Windows drivers' lookup file
(.inf), but a catch-all 'INTCCFFD' entry does exist, so concatenate
that to the existing CID to allow the drivers to attach.
Test: build/boot google/edgar, verify working audio under
both Windows 10 (with Realtek drivers 10.0.10586.4393) and Linux
(GalliumOS 3.1 / kernel 4.16.18, Manjaro 18.1 / kernel 5.1.x)
Change-Id: Idca5cc86ba1f5ef3978cfba291a0c06e56ef5958
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/40003
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Braswell boards don't work well with the eMMC and SD controller
in ACPI in payloads other than depthcharge - SeaBIOS requires
an onerous workaround (manually determining the PCI BAR0 address
for each eMMC and SD controller, then adding adding etc/sdcard
entries to the CBFS), and Tianocore can't see the devices at all.
To make the common use-case work better, switch to PCI mode.
Test: build/boot cyan variants with SeaBIOS and Tianocore
payloads, verify eMMC and SD card visible and bootable to
both payloads and OSes.
Change-Id: I71947603e22a37fe2c8ef4eaac8a3aa0d0ed1cec
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/40002
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Simply cyan variants by converting to overridetree format.
A few differences were ignored as there appears to be no
reason behind them:
- cyan had PCIe RP2 enabled, but nothing is attached to it
- kefka had the SPI 1 device disabled
- reks, relm, and ultima had HSUART 1 disabled
- edgar had I2C1 UPD disabled
Test: build/boot cyan and edgar variants, verify everything
still works
Change-Id: I9928cc59adcfda4661ddfdfa95f53a7820053b4a
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39964
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The VR config values used in the variants' devicetrees is
identical to the domain defaults in vr_config.c, with the
exception of the icc_max value, which is calculated dynamically
based on SKU, and again matches the default values for
each domain and each varaint.
Test: add a print function to dump the VR config values for each
domain from the UPDs after setting, verify same output before/after.
Change-Id: I4307f6e19ae6f99d4f5e475b181fd66c5b92f28c
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39982
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
In preparation for conversion to overridetree format, clean up
the variant devicetrees in order to minimize the differences
across glados variants. This entails:
- minor reformatting and reordering of devicetree entries
- addition of setting default values on boards which skipped them
- disabling unused I2C2 on boards which left it enabled
- ensuring TCC offset set for all SKL-Y boards
- setting VR mailbox command 1 for caroline
- skipping init for UART2 on cave and glados
- dropping unused PCIe RP5 for sentry
Change-Id: I628b20a69fab187e67901c9eb98c0e2ddcb76b0d
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39981
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
There are boards that do not need a specific domain_vr_config because
the defaults provided by the soc code are sufficient. Currently, this
means that these boards can't benefit from lower power states (PSI 3
and 4) because the settings default to being disabled since at the time
the defaults have been defined (2015) there were bugs in FSP in this
regard.
Set the default values of psiXenable to 1 for boards that do not have a
domain_vr_config setting in their devicetree, just like Cannon Lake
does.
Boards that have a domain_vr_config and set their specific settings are
not affected at all. Currently, there are only three boards that have
no domain_vr_config:
- supermicro/x11-lga1151-series
These boards have a MPS MP2955 which we can assume support for PS3
(the MP2965 and MP2935 support it, too).
S-series CPUs with a 1151 socket do not have C9/C10 but only C8 and
since only C10 makes use of PS4, those CPUs won't ever request PS4.
That means we do not need to disable it explicitly for these boards.
- 51nb/x210:
Needs testing and/or VR datasheet check for PS3/PS4 support
Change-Id: I5b5fd9fb3b9b89e80c47f15d706e2dd62dcc0748
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39980
Reviewed-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-by: Benjamin Doron <benjamin.doron00@gmail.com>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Since the variants' devicetrees are almost identical, convert to
using an overridetree setup for simplicity.
As part of the cleanup, drop unused PCIe RP5 for buddy as well.
Test: build all auron variants, compare generated static.c to ensure
resulting generated contents unchanged (although layout will)
Change-Id: I290e7243335a64afdcfc629db7b8ce18f5aa993c
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39940
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Add hook to generate ACPI methods in SSDT for screen backlight control.
To make use of this, individual boards will need to
include default_brightness_levels.asl in their dsdt, as
well as add 'register "gfx" = "GMA_STATIC_DISPLAYS(0)"' to
their devicetree.
Change-Id: I0adccc6c8bee71d3c1b7840518308c8dc8ea2d81
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39943
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Add hook to generate ACPI methods in SSDT for screen backlight control.
To make use of this, individual boards will need to
include default_brightness_levels.asl in their dsdt, as
well as add 'register "gfx" = "GMA_STATIC_DISPLAYS(0)"' to
their devicetree.
Change-Id: I0b7fc45bda3aaf89306bedb579fb1e9f8ce07926
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39942
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
beltino variants are Chromeboxes without built-in displays,
so now that default_brightness_levels.asl is no longer required
for all boards in a platform, drop it.
Change-Id: Ie8147763fc9fdf4f184d3d000bffd6794e134d9e
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39946
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Drop function prototype, since all implementations have been removed.
Change-Id: I2e7b5ac7352a1434652b5e6d37bb3744c68b2328
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39962
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
No longer used by southbridge, no longer needed since pineview
doesn't utilize drivers_intel_gma_displays_ssdt_generate()
Change-Id: Ia386f8fcd208e201fb8bc2a37cdbecd6f45a044b
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39960
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Simplify generation of GMA SSDT, using updated naming convention.
If acpi_fill_ssdt is being invoked, then we know the IGD device is
present and enabled, so we can skip those checks. And the SSDT
generator now checks that the gfx struct is populated, so we can
skip that too.
Change-Id: Ideddfc3d327c4421faffb6583e347cd2b094e155
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39959
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Simplify generation of GMA SSDT, using updated naming convention.
If acpi_fill_ssdt is being invoked, then we know the IGD device is
present and enabled, so we can skip those checks. And the SSDT
generator now checks that the gfx struct is populated, so we can
skip that too.
Change-Id: I68848516fab2058d4aa96ac0342c883fd1df2d6d
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39958
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Simplify generation of GMA SSDT, using updated naming convention.
If acpi_fill_ssdt is being invoked, then we know the IGD device is
present and enabled, so we can skip those checks. And the SSDT
generator now checks that the gfx struct is populated, so we can
skip that too.
Change-Id: Iacce01ab7d6c220779e84c2b695fbb597b493586
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39957
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Simplify generation of GMA SSDT, using updated naming convention.
If acpi_fill_ssdt is being invoked, then we know the IGD device is
present and enabled, so we can skip those checks. And the SSDT
generator now checks that the gfx struct is populated, so we can
skip that too.
Change-Id: I1b6d57c091441aa7431061b1f16135d54cc97b47
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39950
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Simplify generation of GMA SSDT, using updated naming convention.
If acpi_fill_ssdt is being invoked, then we know the IGD device is
present and enabled, so we can skip those checks. And the SSDT
generator now checks that the gfx struct is populated, so we can
skip that too.
Change-Id: If34ebe0edc46674244c9d5afc7ed165c2ad685ba
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39949
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Simplify generation of GMA SSDT, using updated naming convention.
If acpi_fill_ssdt is being invoked, then we know the IGD device is
present and enabled, so we can skip those checks. And the SSDT
generator now checks that the gfx struct is populated, so we can
skip that too.
Change-Id: Icd9caf622dd4c46b13589ebb772138b25888752f
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39948
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
NDID/DID entries are no longer used by the GMA SSDT generator, so
drop them. SSDT generation will be simplified in a subsequent commit.
Change-Id: Ibb7898ab4fbbbfcd29b6ba72367cc55a02cd4b1e
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39956
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
NDID/DID entries are no longer used by the GMA SSDT generator, so
drop them. SSDT generation will be simplified in a subsequent commit.
Change-Id: Iecd8559f660cc748c417ec94b7a822e16603cbd8
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39955
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
NDID/DID entries are no longer used by the GMA SSDT generator, so
drop them. SSDT generation will be simplified in a subsequent commit.
Remove direct setting of gnvs->ndid in qemu-q35 board since build
will otherwise break.
Change-Id: Ifbf08f43291c1fff7ccbc85272dc97334207983b
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39954
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
NDID/DID entries are no longer used by the GMA SSDT generator, so
drop them. SSDT generation will be simplified in a subsequent commit.
Change-Id: I51a11e7ed6686ab67dac3f02097457ea9c6a7e6a
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39953
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
NDID/DID entries are no longer used by the GMA SSDT generator, so
drop them. SSDT generation will be simplified in a subsequent commit.
Change-Id: Ie7491409681d8c2721dd6d6a16a8d5004cd0cf8a
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39952
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
NDID/DID entries are no longer used by the GMA SSDT generator, so
drop them. SSDT generation will be simplified in a subsequent commit.
Change-Id: Iec3a18871725fd5f5c4c568c2bd771bb56245bc7
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39951
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Add framework to generate ACPI methods in SSDT for
screen backlight control. Adjust params for gtt_ methods
to match prototypes in i915.h and avoid conflicts.
To make use of this, individual boards will need to
include default_brightness_levels.asl in their dsdt, as
well as add 'register "gfx" = "GMA_STATIC_DISPLAYS(0)"' to
their devicetree.
Change-Id: If93b7690ef36b5d19ca43957e8a1bef91ec5821d
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39941
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Whether the GMA is used depends on the mainboard, so we shouldn't rely
on the presence of the static ACPI code around `GFX0`.
Change-Id: I4d20b459b8361e43435b535b2b395f51ce1704e6
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39978
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Matt DeVillier <matt.devillier@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
These two identifiers were always very confusing. We're not filling and
injecting generators. We are filling SSDTs and injecting into the DSDT.
So drop the `_generator` suffix. Hopefully, this also makes ACPI look a
little less scary.
Change-Id: I6f0e79632c9c855f38fe24c0186388a25990c44d
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39977
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: David Guckian
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change adds new memory topologies (SODIMM, MIXED) that are
supported by DDR4 and macros required for DDR4 support.
Memory initialization support for DDR4 will be added in a follow-up
change.
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: I4b565c3d71bbf437da64ac29597cc19e58f1b98a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39866
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: EricR Lai <ericr_lai@compal.corp-partner.google.com>
This change reorganizes memory initialization code for LPDDR4x on
TGL to allow sharing of code when adding support for other memory
types. In follow-up changes, support for DDR4 will be added.
1. It adds configuration for memory topology which is currently only
MEMORY_DOWN, however DDR4 requires more topologies to be
supported.
2. spd_info structure is organized to allow mixed topologies as well.
3. DQ/DQS maps are organized to reflect hardware configuration.
TEST=Verified that volteer still boots and memory initialization is
successful.
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: Ib625f2ab30a6e1362a310d9abb3f2051f85c3013
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39865
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: EricR Lai <ericr_lai@compal.corp-partner.google.com>
The initialization code in common//psp is very specific to Family 15h.
Move this to the stoneyridge directory.
BUG=b:130660285
TEST: Verify PSP functionality on google/grunt
Change-Id: Ice3d06d6437f59a529c26fc2359565c940d39482
Signed-off-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Reviewed-on: https://chromium-review.googlesource.com/2020365
Reviewed-by: Eric Peers <epeers@google.com>
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/40000
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
The commands used in Family 15h for loading the SMU FW blobs out of
flash had already been defined differently in Family 17h. To begin
removing Family 15h dependencies from the common/psp, change the
definitions of blob type to no longer match the Family 15h commands.
Consolidate the two functions used for interpreting the command and
applying the command into a single one.
BUG=b:130660285
TEST: Verify PSP functionality on google/grunt
Change-Id: Ic5a4926175d50c01b70ff9b10908c38b3cbe8f35
Signed-off-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Reviewed-on: https://chromium-review.googlesource.com/2020364
Reviewed-by: Eric Peers <epeers@google.com>
Reviewed-by: Raul E Rangel <rrangel@chromium.org>
Tested-by: Eric Peers <epeers@google.com>
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39999
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
This commit creates a malefor variant for Volteer. The initial settings
override the baseboard was copied from variant ripto. Fine tune GPIO
and memory DQ based on malefor schematics.
BUG=b:150653745
BRANCH=volteer
TEST=emerge-volteer coreboot
Signed-off-by: William Wei <wenxu.wei@bitland.corp-partner.google.com>
Change-Id: Idbeebb13e537287686344740211143df35b7863a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39857
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This patch adds GOOGLE_LAZOR which is just a copy of GOOGLE_TROGDOR for
now.
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: I0dca8e1c29bdd91625d58b3cb583b530ed925e9f
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39784
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Bob Moragues <moragues@chromium.org>
Reviewed-by: Philip Chen <philipchen@google.com>
CB:35077 pulled TPM measurement code into the bootblock, with the catch
that we'll only cache PCR extensions and not actually write them to the
TPM until it gets initialized in a later stage. The goal of this was to
keep the heavy TPM driver code out of the size-constrained bootblock.
Unfortunately, a small mistake in the tspi_tpm_is_setup() function
prevents the compiler from eliminating references to the TPM driver
code in the bootblock on platforms with CONFIG_VBOOT and
CONFIG_SEPARATE_VERSTAGE. In those cases vboot_logic_executed() is known
at compile-time to be 0, but that still makes the final expression
`return 0 || tpm_is_setup;`. We know that tpm_is_setup can never be set
to 1 in the bootblock, but the compiler doesn't.
This patch rewrites the logic slightly to achieve the same effect in a
way that the compiler can follow (because we only really need to check
tpm_is_setup in the stage that actually runs the vboot code).
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: Idc25acf1e6c02d929639e83d529cc14af80e0870
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39993
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Bill XIE <persmule@hardenedlinux.org>
Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-by: Christian Walter <christian.walter@9elements.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Switch to using Jasper Lake SoC code from soc/intel/jasperlake and stop
referring from soc/intel/tigerlake.
Addtionally mainboard changes are done to support build.
BUG=b:150217037
TEST=Build and boot waddledoo. Build jasperlake_rvp and volteer board.
Change-Id: I39f117bd66cb610a305bcdb8ea65332fd0ff4814
Signed-off-by: Aamir Bohra <aamir.bohra@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/39825
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Andrey Petrov