Move i2c SoC related code from early_fch.c to i2c.c
TEST=build boards for each SoC
Signed-off-by: Fred Reitberger <reitbergerfred@gmail.com>
Change-Id: I69d4b32cf95ce74586bd8971c7ee4b56c1c2fc04
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68499
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Jason Glenesk <jason.glenesk@gmail.com>
Reviewed-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Rename soc/amd/common/block/cpu/smm/smi_ampc_helper.c to smi_apmc.c and
add the fch_apmc_smi_handler function.
Remove the duplicated function from picasso, cezanne, mendocino, and
morgana SoC.
The stoneyridge soc does not implement the APM_CNT_SMMINFO handler, so
give the handler a unique name that does not conflict with the common
handler name.
Signed-off-by: Fred Reitberger <reitbergerfred@gmail.com>
Change-Id: I2e6fb59a1ee15b075ee3bbb5f95debe884b66789
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68441
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martin.roth@amd.corp-partner.google.com>
Cezanne has two SATA controllers, but doesn't select
SOC_AMD_COMMON_BLOCK_SATA, so it's not added to the SATA devices in the
Cezanne chipset devicetree.
Change-Id: If7f0a9638151cf981d891464a2c3a0ec5fc9c780
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68142
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martin.roth@amd.corp-partner.google.com>
Reviewed-by: Fred Reitberger <reitbergerfred@gmail.com>
This removed the need to maintain a PCI driver.
Change-Id: I43def81d615749008fcc9de8734fa2aca752aa9d
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68146
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martin.roth@amd.corp-partner.google.com>
This removes the need for a PCI driver.
Change-Id: I6674d13f434cfa27fa6514623ba305af6681f70d
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68144
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martin.roth@amd.corp-partner.google.com>
This removes the need for a PCI driver.
Change-Id: Iab75f8c28a247f1370f4425e19cc215678bfa3e5
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68140
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martin.roth@amd.corp-partner.google.com>
The northbridge ops should be added to the actual northbridge and not
the first HT device. Neither of the devices has BARs on it, so
read_resources implementation will still work correctly.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I2e5f21bfe5fff043d7d9afafa360764203dd61f6
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68409
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Stoneyridge is a SoC so it makes sense to statically use ops instead of
matching them to PCI DID/VID at runtime. In contrast to the other AMD
SoCs in the coreboot tree the PC driver used the PCI ID of the first HT
PCI device function, so add the ops to the device 0x18 function 0
devicetree entry in this patch.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I500521701479aa271ebd61e22a1494c8bfaf87fb
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68408
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This removes the need for a lot of boilerplate code in the soc code to
hook up device_operations to devices.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Id668587e1b747c28207b213b985204b7a961a631
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68410
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Add chipset devicetrees for Stoneyridge and Carrizo, which is also
supported by the Stoneyridge code, but has more external PCIe ports and
devices. The mainboard's devicetrees will be changed to use the aliases
defined in the chipset devicetree in follow-up patches. This is a
preparation to statically assign the ops for the internal devices
statically in the SoC devicetree instead of dynamically adding them in
ramstage.
BKDG #55072 Rev 3.04 was used to check the PCI devices and functions and
the MMIO addresses.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Ia45260b1168ed1d99993adfb98475da5b5c90d11
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68316
Reviewed-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
apu/amdfw should be restricted to the RO region only when building with
VBOOT + any RW region (RW_A or RW_A + RW_B); it is not tied to ChromeOS
in any way. Fix guarding to match newer AMD platforms (eg, CZN/MDN).
TEST=build google/zork without CHROMEOS, with VBOOT_SLOTS_RW_A
Change-Id: I32d7fa7a4b3d41107cfdba96128a4a75f7066c6f
Signed-off-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68125
Reviewed-by: Fred Reitberger <reitbergerfred@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
This is the same for all supported AMD hardware.
Change-Id: Ic6b954308dbb4c5a2050f1eb8f15acb41d0b81bd
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/67617
Reviewed-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Definition of FIRMWARE_LOCATION, POUND_SIGN, DEP_FILES,
amd_microcode_bins are moved to common Makefile.inc.
Change-Id: I5a0ea27002e09d0b879bafad37a5d418ddb4e644
Signed-off-by: Zheng Bao <fishbaozi@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/62658
Reviewed-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Reviewed-by: Fred Reitberger <reitbergerfred@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Jon Murphy <jpmurphy@google.com>
Only 16 MByte of the SPI flash can be mapped right below the 4 GB
boundary.
In case of a larger SPI flash size, still only the 16 MByte region
starting at 0xff000000 can be configured as WRPROT and be reserved for
the MMIO mapped SPI flash region. The next 16 MByte MMIO region starting
at address 0xfe000000 contain for example the LAPIC MMIO region, the
ACPIMMIO region and the UART/I2C controller MMIO regions which shouldn't
be configured as WRPROT. Reserving this region for the MMIO mapped SPI
flash would also result in an overlap with the MMIO resources mentioned
above.
In the case of a smaller SPI flash, reserving the full 16 MByte flash
MMIO region makes sure that the resource allocator won't try to put
anything else in the lower parts of the 16 MByte SPI mapping region.
To avoid the issues described above, always reserve/cache the maximum
amount of 16 MBytes of flash that can be mapped below 4 GB.
TEST=On boards with 16 MByte SPI flash chips, the resulting image of a
timeless build doesn't change with this patch. Verified this on Chausie
(Mendocino), Majolica (Cezanne), Cereme (Picasso) and Google/Careena
(Stoneyridge). On Mandolin (Picasso) with an 8 MByte flash, the
resulting image of a timeless build is different, but neither the
coreboot console output nor the Linux dmesg output shows any errors that
might be related to this change.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Ie12bd48e48e267a84dc494f67e8e0c7a4a01a320
Reviewed-on: https://review.coreboot.org/c/coreboot/+/66700
Reviewed-by: Martin Roth <martin.roth@amd.corp-partner.google.com>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Fred Reitberger <reitbergerfred@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Since the I2C controller is part of the FCH, move the early
initialization from bootblock.c to early_fch.c which also matches what
the newer AMD SoCs do.
TEST=Successfully boots on google/liara and all I2C/cr50/TPM functions
appear to work properly
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I22d3a8888eaa34ea612da719c408c0083769e806
Reviewed-on: https://review.coreboot.org/c/coreboot/+/66866
Reviewed-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The functionality of sb_enable_lpc is implemented in the common LPC
support code as lpc_enable_controller. This gets called by the common
lpc_early_init which also calls lpc_disable_decodes and lpc_set_spibase.
The lpc_set_spibase call was already done in bootblock_fch_early_init,
so the main change in code behavior is that now lpc_disable_decodes gets
called during early FCH initialization. The lpc_enable_port80 and
sb_lpc_decode calls after the lpc_early_init code will reenable some of
the decodes.
TEST=Successfully boots on google/liara, cbmem and dmesg logs look clean
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Ia58a6f609fa149a6c09ed99f08bdc4f05eb56f96
Reviewed-on: https://review.coreboot.org/c/coreboot/+/66841
Reviewed-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Since bootblock_soc_early_init gets called before
bootblock_mainboard_early_init which does the early GPIO setup, external
I2C level shifters that are controlled by GPIOs might not be enabled yet.
Moving the reset_i2c_peripherals call to bootblock_soc_init makes sure
that the early GPIO setup is already done when reset_i2c_peripherals is
called.
Haven't probed any SCL signal on the non-SoC side of the I2C level
shifters yet, but the waveform on the SCL pin of I2C3 on the SoC of a
barla/careena Chromebook doesn't have the longer than expected SCL
pulses any more.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: If02140aef56ed6db7ecee24811724b5b24e54a91
Reviewed-on: https://review.coreboot.org/c/coreboot/+/57291
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
This moves the die() statement to a common place.
Change-Id: I24c9f00bfee169b4ca57b469c089188ec62ddada
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/65812
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Julius Werner <jwerner@chromium.org>
There is a lot of going back-and-forth with the KiB arguments, start
the work to migrate away from this.
Change-Id: I329864d36137e9a99b5640f4f504c45a02060a40
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/64658
Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
All AMD SoCs which select SOC_AMD_COMMON_BLOCK_I2C also select
DRIVERS_I2C_DESIGNWARE, so make the pairing explicit by moving the
selection into SOC_AMD_COMMON_BLOCK_I2C. This will facilitating adding
the Designware I2C bus ops handler in a subsequent commit.
Change-Id: Ice30c8806766deb9a6ba617c3e633ab069af3b46
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/65231
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Fred Reitberger <reitbergerfred@gmail.com>
The CPUID function to get the number of cores on a package is common
across multiple generations of AMD cpus.
Change-Id: I28bff875ea2df7837e4495787cf8a4c2d522d43d
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/64869
Reviewed-by: Paul Menzel <paulepanter@mailbox.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
The syscfg has to option to automatically mark the range between 4G and
TOM2, which contains DRAM, as WB. Making it generally not necessary to
allocate MTRRs for memory above 4G if no PCI BARs are placed up there.
Change-Id: Ifbacae28e272ab2f39f268ad034354a9c590d035
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/64868
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
It might be possible to have this used for more than x86, but that
will be for a later commit.
Change-Id: I4968364a95b5c69c21d3915d302d23e6f1ca182f
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55067
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Elyes Haouas <ehaouas@noos.fr>
All targets now use cbmem for the BERT region, so the implementation can
be common.
This also drops the obsolete comment about the need to have bert in a
reserved region (cbmem gets fixed to be in a reserved region).
Change-Id: I6f33d9e05a02492a1c91fb7af94aadaa9acd2931
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/64602
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@mailbox.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
This removes the need to align BERT so that TSEG remains aligned.
Change-Id: I21b55a87838dcb4bd4099f051ba0a011a4d41eea
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/64601
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@mailbox.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
There are efforts to have bootflows that do not follow a traditional
bootblock-romstage-postcar-ramstage model. As part of that CBMEM
initialisation hooks will need to move from romstage to bootblock.
The interface towards platforms and drivers will change to use one of
CBMEM_CREATION_HOOK() or CBMEM_READY_HOOK(). Former will only be called
in the first stage with CBMEM available.
Change-Id: Ie24bf4e818ca69f539196c3a814f3c52d4103d7e
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/63375
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
The postcar frame can now be a local variable to that function.
Change-Id: I873298970fff76b9ee1cae7da156613eb557ffbc
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/61964
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Setting up postcar MTRRs is done when invd is already called so there
is no reason to do this in assembly anymore.
This also drops the custom code for Quark to set up MTRRs.
TESTED on foxconn/g41m and hermes/prodrive that MTRR are properly set
in postcar & ramstage.
Change-Id: I5ec10e84118197a04de0a5194336ef8bb049bba4
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/54299
Reviewed-by: Elyes Haouas <ehaouas@noos.fr>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The first target for the add_intermediate targets is always
$(obj)/coreboot.pre.
Change-Id: Iea2322ca1abd43900f3631b7965f07fed4235ca0
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/56117
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Martin L Roth <gaumless@tutanota.com>
Now the bootblock is not limited to 64K so integrating vboot into the
bootblock reduces the binary size. intel/apl is an exception since the
bootblock size is limited to 32K.
Change-Id: I5e02961183b5bcc37365458a3b10342e5bc2b525
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/52788
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Rename SPIROM_BASE_ADDRESS_REGISTER to SPI_BASE_ADDRESS_REGISTER to
clarify that this isn't the address the SPI flash gets mapped, but the
address of the SPI controller MMIO region. This also aligns the register
name with the PPR.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Ifd9f98bd01b1c7197b80d642a45657c97f708bcd
Reviewed-on: https://review.coreboot.org/c/coreboot/+/62578
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Jason Glenesk <jason.glenesk@gmail.com>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Fred Reitberger <reitbergerfred@gmail.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>
Add a comment to point out that the read_resources functions aren't
missing a pci_dev_read_resources call that would add the resources for
the BARs of the PC device.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Ie480832e0d7954135d2171dda986e477ef7b6c09
Reviewed-on: https://review.coreboot.org/c/coreboot/+/62547
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
In the northbridge's and root complex' read_resources function, the
GNB IOAPIC resource used MMIO base address of the GNB IOAPIC as index
which might be misleading. Instead use idx++ as a unique index for this
resource.
TEST=Resource allocator doesn't complain and no related warnings or
errors in dmesg. The update_constraints console output changes like
expected:
Before: PCI: 00:00.0 fec01000 base fec01000 limit fec01fff mem (fixed)
After: PCI: 00:00.0 0d base fec01000 limit fec01fff mem (fixed)
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I8061364879d772469882fc060f92676de6f600a9
Reviewed-on: https://review.coreboot.org/c/coreboot/+/62546
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
