Let's not rely on the type to get the correct result,
casting 0 to 0ull made the result wrong.
Change-Id: I6dfba3800170fdd4267e3bb74c55b05533c101fc
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/65266
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
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>
This adds delivery of PIC/i8259 interrupts via ExtNMI on the
affected platfoms.
Change-Id: If99e321fd9b153101d71e1b995b43dba48d8763f
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/58406
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
The number of redirection table entries (aka interrupt vectors) inside
an I/O APIC may depend of the SKU, with the related register being of
type read/write-once. Provide support utilities to either lock or set
this registers value.
Change-Id: I8da869ba390dd821b43032e4ccbc9291c39e6bab
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55289
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
This file only contains the `broadwell_run_reference_code()` function
prototype (either a declaration or an inline stub definition). Rename
this file to refcode.h and only include it where necessary.
Tested with BUILD_TIMELESS=1, Purism Librem 13 v1 remains identical.
Change-Id: I6513f45b8914a84312b27ef4860870a89fd0aab3
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55582
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
APIC Serial Bus pins were removed with ICH5 already, so a choice
'irq_on_fsb = 0' would not take effect. The related register BOOT_CONFIG
0x3 is also not documented since ICH5.
For emulation/qemu-q35 with ICH9 the choice INTERRUPT_ON_APIC_BUS was
wrong and ignored as BOOT_CONFIG register emulation was never implemented.
For ICH4 and earlier, the choice to use FSB can be made based on the
installed CPU model but this is now just hardwired to match P4 CPUs of
aopen/dxplplusu.
For sb/intel/i82371eb register BOOT_CONFIG 0x3 is also not defined
and the only possible operation mode there is APIC Serial Bus, which
requires no configuration.
Change-Id: Id433e0e67cb83b44a3041250481f307b2ed1ad18
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55257
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The same functionality can be provided through a runtime-generated SSDT.
The remaining parts of device NVS are removed in a follow-up.
Since the SSDTs are only loaded after the DSDT (if loaded at all), using
SSDT-provided objects outside method bodies is not possible: the objects
are not yet in OSPM's ACPI namespace, which causes in ACPI errors. Owing
to this, the operation regions used by the _PS0 and _PS3 methods need to
be moved into the SSDT, as they depend on the SSDT-provided BAR1 values.
Tested on out-of-tree Compal LA-A992P, generated SSDT disassembles with
no errors and contains expected values. Linux does not complain either.
Change-Id: I89fb658fbb10a8769ebea2e6535c45cd7c212d06
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/52520
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Use the same code from Lynx Point on Broadwell, and adjust as needed.
Also add a config file to ensure the code gets build-tested.
Tested on out-of-tree Compal LA-A992P (Haswell ULT), UART 0 works.
Change-Id: I527024098738700d5fbaf3e27cf4db331a0322bd
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/37553
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
The CMOS option system does not support negative integers. Thus, retype
and rename the option API functions to reflect this.
Change-Id: Id3480e5cfc0ec90674def7ef0919e0b7ac5b19b3
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/52672
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Felix Singer <felixsinger@posteo.net>
With chipset_power_state filled in romstage CBMEM hooks and
GNVS allocated early in ramstage, GNVS wake source is now
also filled for normal boot path.
Change-Id: I2d44770392d14d2d6e22cc98df9d1751c8717ff3
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/50004
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Use `lp_gpio.h` from Lynxpoint instead. Subsequent commits will update
the mainboards and then drop all GPIO code from Broadwell.
Tested with BUILD_TIMELESS=1, Google Tidus remains identical.
Change-Id: Idef89037c2ca781ac3e921abb4b3dc3f7c4b3b5f
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/50079
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This is what Lynxpoint does. It is equivalent, but simpler.
Change-Id: Ifdbb291a6cea0bb29b4e46c7a33c5abe61dbe86b
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/47045
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
There's no need to die here. Also simplifies merging with Haswell.
Change-Id: I3d4bc79b32279180442dbc82126e297f11f1fb80
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/46890
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Only amd/picasso and amd/stoneyridge have reference to
PCNT and that could be replaced with acpigen.
Remove the PCNT name from GNVS OperationRegion elsewhere.
Change-Id: I7dd45a840b3585fd24c31fd923b991c34ab4d783
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49272
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Name the common part of GNVS initialisation as soc_fill_gnvs().
It is also moved before the call to acpi_create_gnvs(), which
followup will rename to mainbord_fill_gnvs() to reflect that
implementation is under mb/.
Change-Id: Ic4cf1548b65a86212d6e45d460fcd23bb8036365
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48706
Reviewed-by: Lance Zhao
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Already done in common gnvs_get_or_create() implementation
once gnvs_chromeos_ptr() is defined for platforms.
Change-Id: I90fa2bc28ae76da734b3f88be057435aed9fe374
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48703
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Allocation now happens prior to device enumeration. The
step cbmem_add() is a no-op here, if reached for some
boards. The memset() here is also redundant and becomes
harmful with followup works, as it would wipe out the
CBMEM console and ChromeOS related fields without them
being set again.
Change-Id: I9b2625af15cae90b9c1eb601e606d0430336609f
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48701
Reviewed-by: Lance Zhao
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Instead of relying on mainboards to call it, do like Lynx Point.
Change-Id: Idb7457e0734e19d0a26f0762079e273b6e740475
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/46793
Reviewed-by: Michael Niewöhner <foss@mniewoehner.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
