Disable the unused PCIe root ports that are disabled in the PCIe port
corresponding descriptor list passed to AGESA/binaryPI. This descriptor
list is in src/mainboard/google/kahlee/variants/baseboard/OemCustomize.c
and it only has B0D2F2 (gpp_bridge_1) and B0D2F4 (gpp_bridge_3) enabled.
Since the PCIe engines marked as unused in the port descriptor list
won't show up as PCI devices, don't enable those PCI devices in the
devicetree so that coreboot won't complain about static PCI devices not
being found on the PCI bus.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: If8378e343a2eb13de66171cf4f38d77ae3401016
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68382
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
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>
When using a Merlin Falcon APU, explicitly enable the PCIe root port at
B0D3F1. B0D3F0 is only a dummy PCI device function, but needs to also be
enabled in order for the actually used function to be usable. Prairie
Falcon doesn't have and PCI device 3 on bus 0, so remove D3F0 from the
common mainboard devicetree.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I01f9b9ac2a9ebd5899a093d97eb5b2d76d309f66
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68315
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>
PCI devices that aren't present in the devicetree will be treated as
enabled. Since the chipset devicetree that will be added in a follow-up
patch disables this device by default, explicitly enable the IOMMU
device on the Stoneyridge mainboards that don't disable it to keep the
same behavior.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I4a2cdd00abe8309244829dc633dd8a9ca0038dfa
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68313
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>
Since the devicetree files are passed to util/sconfig without being
processed by the C preprocessor, using #if in the devicetree won't give
the behavior that might be expected. Instead sconfig treats the #if as a
comment, but still processes all other lines. To get the intended
behavior, replace the C preprocessor usage in the devicetree by moving
the APU-specific parts to override devicetrees that get selected
according to the selected APU type.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Iddd317b27a838849fa40c0fb77d942609104cf04
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68312
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>
Switch from gpio_configure_pads() to gpio_configure_pads_with_override()
so variants can override romstage GPIO defaults. Rename baseboard
function and add an weak empty override function to be used by variants.
Will be used for touchscreen power sequencing in a follow-on commit.
Change-Id: I45586237919cd07a171beac57f3510e26338f67f
Signed-off-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/67811
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Create the zombie variant of the herobrine reference board by
copying the template files to a new directory named for the variant.
(Auto-Generated by create_coreboot_variant.sh version 4.5.0).
BUG=b:249180463
BRANCH=None
TEST=util/abuild/abuild -p none -t google/herobrine -x -a
make sure the build includes GOOGLE_ZOMBIE
Signed-off-by: Maulik Vaghela <maulikvaghela@google.com>
Change-Id: Ifecf0a6323b20012defbf14bd16ce2f1f41f4714
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68303
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Bob Moragues <moragues@google.com>
Reviewed-by: Shelley Chen <shchen@google.com>
HP Z220 series has PCI slot(s) but Interrupt Routing Table in ACPI
used to be missing, so one is added.
Note that the values within the added one are obtained from my own SFF
variant. If other variants have different values, please add them in a
manner similar to mb/gigabyte/ga-b75m-d3h/acpi/pci.asl.
Test result:
Log lines like
pci 0000:00:1e.0: can't derive routing for PCI INT A
ath9k 0000:04:00.0: PCI INT A: no GSI
disappeared from dmesg.
Signed-off-by: Bill XIE <persmule@hardenedlinux.org>
Change-Id: I8522b25ac46db2054302c8f2418927c722b157e0
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68334
Reviewed-by: Felix Singer <felixsinger@posteo.net>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Add a missing period, use "SoC" with lowercase 'o' to refer to a "System
on a Chip", fix up the redundant "CRB board" expression (that stands for
"Customer Reference Board board"), switch the position of a verb and its
adverb ("never was" ---> "was never") to sound more natural, and replace
"depreciate" with "deprecate" for semantic correctness.
Change-Id: Ic821a9030d4ff32c76765f51f1feb0f5503d4cc0
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68330
Reviewed-by: Felix Singer <felixsinger@posteo.net>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The `eeprom_read_serial()` function could return a non-NULL terminated
string if the serial in EEPROM has `HERMES_SN_PN_LENGTH` (32) non-NULL
characters. Make this impossible by adding an additional character for
a NULL byte in the static buffer, which always gets set to 0 (NULL).
Change-Id: I306fe1b6dd3836156afca786e352d2a7dca0d77c
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68322
Reviewed-by: Patrick Georgi <patrick@coreboot.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Singer <felixsinger@posteo.net>
This reverts commit 4b5ba94363.
Reason for revert: This optimization is causing the non-serial enabled
tot BIOS to not boot. To get tot back into good shape, will revert
for now and reevalute this fix and resubmit at a later time.
BUG=b:218406702
BRANCH=None
TEST=reboot from AP console (on herobrine) after flashing
image-herobrine.bin.
prior to fix the device would never boot to login prompt.
after rever the device would boot to login prompt again.
Change-Id: Iaac5f2fb2120f6aa41a0ce9a763d50fd7b9a3ec7
Signed-off-by: Shelley Chen <shchen@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/68339
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
The decompression is critical for speed of boot. So we sacrifice some
generated code size to optimize for speed.
This change speeds up the LZMA decompression between 3% and 6% at a
cost of just over 2k of additional code space.
BUG=b:223985641
TEST=Majolica
The test is done on Majolica and the result is listed below.
Time saved:
We tested the boot time with each flag for 10 times. The duration of
each decompression process is listed as below.
Load FSP-M Load ramstage Load payload
Ofast Os Ofast Os Ofast Os
------------------------------------------
62543 62959 20585 22458 9945 10626
62548 62967 20587 22461 9951 10637
62560 62980 20588 22478 9951 10641
62561 62988 20596 22478 9954 10643
62569 62993 20596 22479 9954 10643
62574 63000 20605 22492 9958 10647
62575 63026 20615 22495 9959 10647
62576 63038 20743 22614 9960 10647
62587 63044 20758 22625 9961 10647
62592 63045 20769 22637 9961 10647
-----------------------------------------
average 62568 63004 20644 22521 9955 10642
(unit: microseconds)
Size sacrificed:
The size of object file with -Os:
./build/ramstage/lib/lzmadecode.o: file format elf32-i386
4 .text.LzmaDecode 00000d84 00000000 00000000 00000076 2**0
CONTENTS, ALLOC, LOAD, READONLY, CODE
The size of object file with -Ofast:
./build/ramstage/lib/lzmadecode.o: file format elf32-i386
4 .text.LzmaDecode 00001719 00000000 00000000 00000080 2**4
CONTENTS, ALLOC, LOAD, READONLY, CODE
(Output by running "objdump -h ./build/ramstage/lib/lzmadecode.o")
We can see that size is increased from 3460 bytes to 5913 bytes, a
change of 2453 bytes or 171%.
Change-Id: Ie003164e2e93ba8ed3ccd207f3af31c6acf1c5e2
Signed-off-by: Zheng Bao <fishbaozi@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/66392
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martin.roth@amd.corp-partner.google.com>
