This is usually is handled by Haswell mrc.bin, disabling VGA
decode on the iGPU when a dGPU is installed. However, Broadwell
mrc.bin does not, so the iGPU and dGPU are both enabled.
This patch disables legacy VGA cycles for iGPU, under such
conditions. It has been tested on Broadwell mrc.bin when
using a graphics card on Dell OptiPlex 9020 SFF (currently
under review at this time of writing, submitted by Mate
Kukri).
This patch has also been tested when Haswell mrc.bin is used,
and there are seemingly no breaking changes caused by it.
Change-Id: I1df0a3aa42f8475b7741007bf3e28c2e089d916b
Signed-off-by: Leah Rowe <info@minifree.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/80717
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Use uintptr_t for the IOAPIC base parameter of the various IOAPIC-
related functions to avoid needing type casts in the callers. This also
allows dropping the VIO_APIC_VADDR define and consistently use the
IO_APIC_ADDR define instead.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I912943e923ff092708e90138caa5e1daf269a69f
Reviewed-on: https://review.coreboot.org/c/coreboot/+/80358
Reviewed-by: Elyes Haouas <ehaouas@noos.fr>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Reviewed-by: Jérémy Compostella <jeremy.compostella@intel.com>
It seems that we have some applications where we need to calculate a GCD
in 64 bits. Now, we could instantiate the algorithm multiple times for
different bit width combinations to be able to use the most efficient
one for each problem... but considering that the function usually only
gets called once per callsite per stage, and that software emulation of
64-bit division on 32-bit systems doesn't take *that* long either, we
would probably usually be paying more time loading the second instance
of the function than we save with faster divisions. So let's just make
things easy and always do it in 64-bit and then nobody has to spend time
thinking on which version to call.
Change-Id: I028361444c4048a0d76ba4f80c7334a9d9983c87
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/80319
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Yidi Lin <yidilin@google.com>
This patch moves the IP checksum algorithm into commonlib to prepare for
it being shared with libpayload. The current implementation is ancient
and pretty hard to read (and does some unnecessary questionable things
like the type-punning stuff which leads to suboptimal code generation),
so this reimplements it from scratch (that also helps with the
licensing).
This algorithm is prepared to take in a pre-calculated "wide" checksum
in a machine-register-sized data type which is then narrowed down to 16
bits (see RFC 1071 for why that's valid). This isn't used yet (and the
code will get optimized out), but will be used later in this patch
series for architecture-specific optimization.
Change-Id: Ic04c714c00439a17fc04a8a6e730cc2aa19b8e68
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/80251
Reviewed-by: Yidi Lin <yidilin@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Jakub Czapiga <czapiga@google.com>
This renames bus to upstream and link_list to downstream.
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Change-Id: I80a81b6b8606e450ff180add9439481ec28c2420
Reviewed-on: https://review.coreboot.org/c/coreboot/+/78330
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Macros can be confusing on their own; hiding commas make things worse.
This can sometimes be downright misleading. A "good" example would be
the code in soc/intel/xeon_sp/spr/chip.c:
CHIP_NAME("Intel SapphireRapids-SP").enable_dev = chip_enable_dev,
This appears as CHIP_NAME() being some struct when in fact these are
defining 2 separate members of the same struct.
It was decided to remove this macro altogether, as it does not do
anything special and incurs a maintenance burden.
Change-Id: Iaed6dfb144bddcf5c43634b0c955c19afce388f0
Signed-off-by: Nicholas Sudsgaard <devel+coreboot@nsudsgaard.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/80239
Reviewed-by: Yidi Lin <yidilin@google.com>
Reviewed-by: Felix Singer <service+coreboot-gerrit@felixsinger.de>
Reviewed-by: Jakub Czapiga <czapiga@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Multiple links are unused throughout the tree and make the code more
confusing as an iteration over all busses is needed to get downstream
devices. This also not done consistently e.g. the allocator does not
care about multiple links on busses. A better way of dealing multiple
links below a device is to feature dummy devices with each their
respective bus.
This drops the sconfig capability to declare the same device multiple
times which was previously used to declare multiple links.
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Change-Id: Iab6fe269faef46ae77ed1ea425440cf5c7dbd49b
Reviewed-on: https://review.coreboot.org/c/coreboot/+/78328
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Jincheng Li <jincheng.li@intel.com>
Reviewed-by: Lean Sheng Tan <sheng.tan@9elements.com>
The .inc suffix is confusing to various tools as it's not specific to
Makefiles. This means that editors don't recognize the files, and don't
open them with highlighting and any other specific editor functionality.
This issue is also seen in the release notes generation script where
Makefiles get renamed before running cloc.
Signed-off-by: Martin Roth <gaumless@gmail.com>
Change-Id: Iddac15cc42532f44dda44032be0f8525f6347abd
Reviewed-on: https://review.coreboot.org/c/coreboot/+/80070
Reviewed-by: Felix Singer <service+coreboot-gerrit@felixsinger.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-by: Maximilian Brune <maximilian.brune@9elements.com>
This reverts commit acbc491237.
Reason for revert: CB:79525 fixes the issue that led to the revert
by not maintaining the heap in the SMM-stored copy of ramstage at all.
Change-Id: I3c8ef785486d275c9341859d34fce12253bd2bb9
Signed-off-by: Patrick Georgi <patrick@coreboot.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/80023
Reviewed-by: Sean Rhodes <sean@starlabs.systems>
Reviewed-by: Subrata Banik <subratabanik@google.com>
Reviewed-by: Julius Werner <jwerner@chromium.org>
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>
Add initial support for multiple PCI segment groups. Instead of
modifying secondary in the bus struct introduce a new segment_group
struct element and keep existing common code.
Since all platforms currently only use 1 segment this is not a
functional change. On platforms that support more than 1 segment the
segment has to be set when creating the PCI domain.
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Ied3313c41896362dd989ee2ab1b1bcdced840aa8
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79927
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Martin Roth <martin.roth@amd.corp-partner.google.com>
Built roda/rk9 with BUILD_TIMELESS=1 and the resulting coreboot.rom
remains identical.
Change-Id: Ib1e7144eebf8148c4eb5cc0e7bc03ae3d7281092
Signed-off-by: Elyes Haouas <ehaouas@noos.fr>
Signed-off-by: Felix Singer <felixsinger@posteo.net>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/77971
Reviewed-by: Felix Singer <service+coreboot-gerrit@felixsinger.de>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Enable x86_64 support for MRC.bin:
- Add a wrapper function for console printing that calls into
long mode to call native do_putchar
- Remove Kconfig guard for x86_64 when MRC is being used
Tested: Booted Lenovo X220 using mrc.bin under x86_64 and
MRC is able to print to the console.
Change-Id: I21ffcb5f5d4bf155593e8111531bdf0ed7071dfc
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79754
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
In DDR3 DLL-Off mode is an optional feature advertised by SPD.
Honor the SPD and only use DLL-Off mode when all DIMMs on the
same channel indicate support for it.
The same is done on MRC.bin.
Tested on Lenovo X220: Still boots fine.
Change-Id: Ief4bfb9e045cad7ff9953f6fda248586ea951a52
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79758
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
- Remove pointers in argument list passed to MRC to make sure the struct
has the same size on x86_64 as on x86_32.
- Add assembly wrapper to call the MRC with argument in EAX.
- Wrap calling MRC in protected_mode_call_2arg, which is a stub on x86_32
Tested: Boots on Lenovo X220 using MRC in x86_32 and x86_64 mode.
Change-Id: Id755e7381c5a94360e3511c53432d68b7687df67
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79751
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Jérémy Compostella <jeremy.compostella@intel.com>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Use same indent levels for switch/case in order to comply with the
linter.
Change-Id: I64361262e5b16419351fa139c8fdf04c5c07662d
Signed-off-by: Felix Singer <felixsinger@posteo.net>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79444
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Eric Lai <ericllai@google.com>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
SATA_IDE_DEVID, AHCI_DEVID_MS and AHCI_DEVID_AMD are still kept even
though they're unused at the moment, but those might still be useful to
keep around, since the SATA controller can have different PCI device IDs
depending on in which mode it is in.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Ia05683b732d9748d9198225acaecbd4dc196733a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79577
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Eric Lai <ericllai@google.com>
There's neither need to remove get_hw_mem_hole_info from the code if the
Kconfig option was set to 0 nor the actual value didn't make any
difference in the behavior of the code: When node_id has still its
initial value of -1, domain_read_resources won't use the value of
hole_startk, and when node_id is set to 0, get_hw_mem_hole_info also
sets hole_startk to the actual value that then gets used by
domain_read_resources.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Ieffab695a3151ed7f6bf9d6c880bbb43eecf7893
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79609
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This APU is always a single-node, so the nodeid parameter of
get_node_pci is always 0. Since this SoC has a chipset devicetree, we
can just use DEV_PTR(ht_X) instead of the pcidev_on_root call.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I1bf9d214b4c2e5d995976fb79fef6fe43a6e9fa0
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79608
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This APU is always a single-node and since we're in ramstage when
domain_read_resources gets called, there's DRAM on this node, so no need
to check for this. To be extra sure, also initialize basek and limitk
before calling get_dram_base_limit with pointers to those as arguments.
This won't be necessary for the code to work as intended, but will
probably keep the compiler from complaining. Also move the declaration
of basek, limitk and sizek to the beginning of the function.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I4ef8011eb57b16218b8f5fea295900b855c3014b
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79611
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Start with the resource index 0 and increment it after reporting each
resource.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I6fb59ff3d371b744b53093d17392d1c3510bef82
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79610
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Nico Huber <nico.h@gmx.de>
This APU is always a single-node and also only has one DRAM controller,
so there is only one valid DRAM base and limit register. It's also worth
mentioning that the assumption made in get_dram_base_limit that the n-th
node is using the n-tn DRAM range register was valid for K8, but not
necessarily on newer generations than that.
TEST=PC Engines APU2 still boots and doesn't show any new problems
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Id0529c66e8d0e6c8eb42eec2c6d9d2e892287865
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79607
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
This APU is always a single-node and also only has one DRAM controller,
so we don't need to loop over the different nodes to find the memory
hole below 4GB. We also don't need to check for the special case where
the memory hole is non-DRAM address space between the parts of the
address space decoded by different DRAM controllers.
TEST=PC Engines APU2 still boots and doesn't show any new problems
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I9793d911d2d496be49168c06d83ceb802bc2b647
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79606
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Nico Huber <nico.h@gmx.de>
This APU is always a single-node, so domain_read_resources only needs to
handle exactly one node and doesn't need to loop over the nodes.
TEST=PC Engines APU2 still boots and doesn't show any new problems
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I4218077cb4e11b762ce0e8694a97bdec33eaa056
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79605
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
This SoC only supports a single-node configuration, so all the code
related to multi-node support can be removed. In this commit only the
get_fx_devs function and related code are removed for better
reviewability. In f1_write_config32 it's no longer needed to loop over
the different devices of the different nodes, so only a single PCI
config space write remains.
TEST=PC Engines APU2 still boots and doesn't show any new problems
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I5dc7324d3fcd0d07ac7a3a246a740fd9e91c3840
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79604
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This APU is always a single-node system and bits 4..6 of the node ID
register D18F0x60 are also marked as reserved in BKDG #52740 Rev 3.05.
On an APU2 board with quad-core APU, this register reads back 0x00030000
which results in a value of 1 to be returned from get_node_nums, so this
patch doesn't change behavior, but stops using reserved bits.
TEST=PC Engines APU2 still boots and doesn't show any new problems
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I65ed1124c0ca8e7eba54ff53dc626d35cd5e2e58
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79603
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
This system only has one northbridge and amd_initcpuio has already set
up the routing of the legacy VGA IO and MMIO ranges to it. Since only
the pci_dev_set_resources call remains in nb_set_resources, use
pci_dev_set_resources directly as set_resources function.
TEST=PC Engines APU2 still boots and doesn't show any new problems
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Ib3835db9fd83221ac2b8e34d998f938812d24413
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79582
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Since the IOMMU is always function 2 of device 0 on bus 0, the device
operations can be statically assigned in the devicetree and there's no
need to bind the IOMMU device operations to the PCI device during
runtime via a list of PCI IDs.
TEST=PC Engines APU2 still boots and doesn't show any new problems
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I84e949500ee86e0fcb2d15791502f5e3e7127703
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79105
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Since the northbridge is always function 0 of device 0 on bus 0, the
device operations can be statically assigned in the devicetree and
there's no need to bind the northbridge device operations to the PCI
device during runtime via a list of PCI IDs.
TEST=PC Engines APU2 still boots and doesn't show any new problems
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Ia7faaa468ff77e05c378c5555622c3584cfe3f81
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79104
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Since the EHCI controllers in the PCH are always on the same device
functions, the device operations can be statically assigned in the
devicetree and there's no need to bind the EHCI device operations to the
PCI devices during runtime via a list of PCI IDs.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I92ecc3607216fb2f31639db9628898c9ce81770d
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79171
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Since the XHCI controller in the PCH is always on the same device
function, the device operations can be statically assigned in the
devicetree and there's no need to bind the XHCI device operations to the
PCI device during runtime via a list of PCI IDs.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I8685bec734415346a53330c9bd1aa82986995f1a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79170
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Since the PCI bridge in the PCH is always on the same device function,
the device operations can be statically assigned in the devicetree and
there's no need to bind the PCI bridge device operations to the PCI
device during runtime via a list of PCI IDs.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Ic9ca925a12e64c9a5b3bf295653bf032572ff29a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79169
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Since the SMBus controller in the PCH is always on the same device
function, the device operations can be statically assigned in the
devicetree and there's no need to bind the SMBus device operations to
the PCI device during runtime via a list of PCI IDs.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I3d3745ba5aefa30efbe705155d216aa7eadd26a7
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79168
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Function 0 of the device that has the bridges to other buses is a dummy
function that can be left enabled to not have to shuffle around the
device function numbers when the first PCI bridge on that device isn't
enabled. That dummy device function is however not a PCI host bridge, so
change the comment from 'Dummy Host Bridge' to 'Dummy device function'.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Suggested-by: Nico Huber <nico.h@gmx.de>
Change-Id: I6069205bd2e1cb0f75025e9f330afc50462e742a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79397
Reviewed-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Since the LPC bridge in the PCH is always on the same device function,
the device operations can be statically assigned in the devicetree and
there's no need to bind the LPC bridge device operations to the PCI
device during runtime via a list of PCI IDs.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I366226be4aba75b98e45e4832bfe129fac14dbfa
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79167
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
This is likely a copy-paste leftover, since this SoC neither has a PCI
device with the device ID 0x1200 nor is family 10h.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I7095f208a7503545ea012241d058692a510109f3
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79094
Reviewed-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Instead of using MSR IA32_PLATFORM_ID read the SystemAgent device id
to figure out the PC type. This follows the BWG which suggest to not
use MSR IA32_PLATFORM_ID for system identification.
Tested: Lenovo X220 still boots.
Change-Id: Ibddf6c75d15ca7a99758c377ed956d483abe7ec1
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/78826
Reviewed-by: Lean Sheng Tan <sheng.tan@9elements.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Since the HD audio controller in the PCH are always on the same device
functions, the device operations can be statically assigned in the
devicetree and there's no need to bind the host bridge device operations
to the PCI device during runtime via a list of PCI IDs.
TEST=Lenovo X220 still boots to Linux and audio still works
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Tested-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Change-Id: I9bbbe9f4490dc6fb21174d63d1c8906d69ea3ee0
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79118
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Since the PCIe root ports in the PCH are always on the same device
functions, the device operations can be statically assigned in the
devicetree and there's no need to bind the host bridge device operations
to the PCI device during runtime via a list of PCI IDs.
TEST=Lenovo X220 still boots to Linux and all PCIe devices on PCH are
visible and working.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Tested-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Change-Id: I05bfe8db88fd54415f320f32ea147636ca4e0df8
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79117
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Since the integrated GPU is always function 0 of device 2 on bus 0, the
device operations can be statically assigned in the devicetree and
there's no need to bind the host bridge device operations to the PCI
device during runtime via a list of PCI IDs.
TEST=Lenovo X220 still boots to Linux and graphics works in UEFI
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Tested-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Change-Id: I20e387e626e19dc441aceda18451186d1e86cd5f
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79114
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Since the host bridge is always function 0 of device 0 on bus 0, the
device operations can be statically assigned in the devicetree and
there's no need to bind the host bridge device operations to the PCI
device during runtime via a list of PCI IDs.
TEST=Lenovo X220 still boots to Linux
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Tested-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Change-Id: Icf3d9f8cd2be2f8ef71fd9fdb5f005f3b683332e
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79113
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Add the CPU and PCI domain operation bindings statically in the chipset
devicetree instead of adding them during runtime.
TEST=PC Engines APU2 still boots and doesn't show any new problems
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I44fa57458c408e74a6341643620c5e9ac1817557
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79085
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Since this chip is a SoC and also to bring the chipset devicetree more
in line with the chipset devicetree of Sandy Bridge, merge the chip
operations of the northbridge's root complex and the northbridge itself
into one chip operations structure and use it at the top level of the
devicetree.
TEST=PC Engines APU2 still boots and doesn't show any new problems
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I8b42bac07b1409bbc797bc4428cf9f84a40e94c2
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79084
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Matt DeVillier <matt.devillier@gmail.com>
BKDG #52740 Rev 3.05 was used as a reference for the SoC's various PCI
devices. The HDA controller in the FCH at function 2 of device 0x14 on
bus 0 was missing in the mainboard's devicetrees.
TEST=PC Engines APU2 still boots and doesn't show any new problems
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I6970c2f6e6d661d40406586f4e6eeb05bcd07979
Reviewed-on: https://review.coreboot.org/c/coreboot/+/79083
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Matt DeVillier <matt.devillier@gmail.com>
After commit adaeb11021 (nb/intel/sandybridge: Clean up post Haswell SPD mapping API migration), no boards use this header anymore and it
no longer offers original content.
Adjust northbridge code #includes as needed and drop it.
Change-Id: I2785e920bd6188dbfc1a6157351083ec4a2526d0
Signed-off-by: Keith Hui <buurin@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/78785
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin L Roth <gaumless@gmail.com>
With migration to Haswell SPD mapping interface complete:
1. Remove weak stubs meant to ensure smooth transition and
internalizes mainboard_get_spd() within raminit.c.
2. Remove post-mainboard SPD data sanitization code in raminit_mrc.c,
now that it fills its own SPD data.
3. Remove old prototypes from raminit_native.h
4a. Drops raminit_native.h from raminit.c, as individual headers
therein are already included.
4b. Drop another header from raminit.c IWYU identified as unneeded.
asus/p8z77-m still builds afterwards.
(sandybridge to receive a full IWYU cleanup later.)
Change-Id: Ie073c1386cd0a645069f0e1416263b4fa359b74b
Signed-off-by: Keith Hui <buurin@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/76991
Reviewed-by: Martin L Roth <gaumless@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Changes both MRC and native raminit code path to get SPD mapping
from one place.
Boards with all memory socketed specify their mappings in a
devicetree setting introduced in commit 5709e03613
("nb/intel/sandybridge: Migrate MRC settings to devicetree") back in
May 2019 but remains unused as of this patch. This setting
will now hold raw SMBus addresses, and MRC raminit gets code to
translate them into a representation MRC expects.
Boards with soldered down memory (specifically with HAVE_SPD_IN_CBFS
in their board Kconfig), with or without socketed memory, specify
their layouts in mb_get_spd_map() as used by Haswell boards, where
they access hardware GPIO straps to select which SPD data to use.
This harmonizes the way boards specify their SPD layouts across
Haswell/SNB/IVB boards whether using MRC or native raminit. Going
forward they only need to specify the layout in one place. (Going
forward the devicetree setting should be backported to Haswell,
once we get native raminit working there.)
With this, northbridge code is now fully responsible for loading
all SPD data, be it from CBFS or SMBus.
To avoid breakage, transition will happen in stages:
1. This patch gets all the code in, and implements weak stubs that
maintain existing code and data flow (i.e. mainboards still populate
final SPD layout data). At this point devicetree already uses new
representation, but is still unused meaning no breakage.
2. Follow-up patch(es) remove mainboard_get_spd() from mainboards, and
replace it with mb_get_spd_map() or devicetree values (as appropriate)
with converted SPD info. The "weak" mainboard_get_spd() with new logic
takes over. Boards go Haswell Style at this point. Boards with MRC
raminit also lose code to fill in SPD data, allowing new data to take
hold.
3. Clean-up patch removes the weak functions and public prototypes re
mainboard_get_spd(), making it internal to northbridge. Changeover is
complete.
Change-Id: I1a75279d981e46505930a9ce1aae894ccc4e1f24
Signed-off-by: Keith Hui <buurin@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/76965
Reviewed-by: Martin L Roth <gaumless@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
