I think it makes the code a bit cleaner.
BUG=b:170595019
BRANCH=zork
TEST=boot zork with pci=nomsi and verify /proc/interrupts didn't change
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: Ib5e8e5b690d9612e8ae257f5d15c25122e1c91e6
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49842
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Each bridge can only have one device.
BUG=b:170595019
BRANCH=zork
TEST=none
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: I7e476221dfcabc841cc1ed4bc4b1175c0652dcfe
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49841
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
The PCI interrupt line registers are used as a last resort if routing
can't be fetched from either ACPI or the MPTable. This change correctly
sets the registers. It overrides the pirq_data set by the mainboards
since the routing is fixed in AGESA.
BUG=b:170595019
TEST=Boot ezkinil with `pci=nomsi,noacpi amd_iommu=off noapic`
Verified all PCI peripherals are still functional.
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: If5d4d8f613c8d0fa9b43cefa804824681c3410d6
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48665
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nikolai Vyssotski <nikolai.vyssotski@amd.corp-partner.google.com>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
The original routing table did not handle all 8 INTx interrupts.
Additionally it also didn't take the swizzling into account.
Now that we know how AGESA programs the routing table we can correctly
generate it.
We still route the PCI interrupts through the FCH IOAPIC. A follow up
will have the GNB IOAPIC handle the PCI interrupts.
There is still work to be done to fix the legacy PCI_IRQ register for
each PCI device. We can then remove the mainboard_pirq_data from each
mainboard.
BUG=b:170595019
TEST=Used ezkinil
Boot kernel with `pci=nomsi amd_iommu=off noapic` and
`pci=nomsi amd_iommu=off` then verified system
was usable and verified /proc/interrupts looked correct.
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: I2b2cce9913081d5cd456043ba619a79c1dfd4a8e
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48632
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nikolai Vyssotski <nikolai.vyssotski@amd.corp-partner.google.com>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Different models within family 17h have different PCI IDs for their PCIe
GPP port and internal bus devices.
Change-Id: I386df908ce5451b4484be2a2e4a9018c3d47d030
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/47677
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
This change adds a driver pcie_gpp.c which provides device_operations
for external and internal PCIe GPP bridges. These device operations
include standard PCI bridge operations as well as operations for
generating ACPI node for the device and returning appropriate ACPI
name for it.
Signed-off-by: Furquan Shaikh <furquan@google.com>
Change-Id: I9f8809c2735bdc09435deda91a570c89e71e8062
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43312
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
