This includes a huge set of refactorings to support Core Display Clock
(CDClk) frequency switching based on the current mode requirements.
The CDClk is configurable since Haswell and runtime switching is suppor-
ted since Broadwell. Always using the lowest possible frequency setting
should allow some power-savings. While, on the upper end, we can support
higher resolution panels now, without having to change the static confi-
guration.
There have also been some smaller changes and fixes, including:
o Parsing of eDP 1.4+ DPCD link rates, enables panels that don't
advertise a maximum link rate but only individual ones.
o DP support for Ibex Peak.
o Corrected limit for HDMI on G45 to 165MHz.
o Reworked GMBUS reset handling and timeouts, should help with
stalled GMBUS controllers when unimplemented ports were probed
by accident.
Tested on various boards from GM45 to KBL-R.
Change-Id: I0a90bd4afe2091699a46a5a1323af9723ff43018
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35898
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
As stated in CB:36334 cbmem_top() should not be called before memory is
initialized. Therefore drop the check to see if MRC finished.
Change-Id: I964a20a5e9aa69fdb75413c36a17d34b7ba00098
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36386
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Determine maximum speed by looking at either turbo flex limit or
uncore ratio limit.
Signed-off-by: Andrey Petrov <anpetrov@fb.com>
Change-Id: I0f3a64a40cb1d28d8eb9380c2071ec748e345b88
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36284
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: David Hendricks <david.hendricks@gmail.com>
If CPUID leaf 0x16 is available (Skylake and later) use it to obtain
current and maximum speed. Otherwise call weak function that can be
provided elsewhere (cpu/soc/mainboard). Also, populate "core enabled"
with the same value as "core count".
TEST=tested on OCP Monolake with dmidecode -t processor
Change-Id: Ie5d88dacae6623dfa0ceb3ca1bb5eeff2adda103
Signed-off-by: Andrey Petrov <anpetrov@fb.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36283
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: David Hendricks <david.hendricks@gmail.com>
This cpuid function (0) is same across Intel and AMD so having it in
arch/x86 seems like a good idea.
Change-Id: I38f1c40bceac38ed6428f74b08bf60b971644f5a
Signed-off-by: Andrey Petrov <anpetrov@fb.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36282
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: David Hendricks <david.hendricks@gmail.com>
To get a better idea what this code does, this patch adds
a new method called 'print_system_info'.
Change-Id: I16f1c9cdc402b1a816fac65d1490432e39c07baf
Signed-off-by: Felix Singer <felix.singer@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36315
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Michael Niewöhner
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The i82801ix_early_init is now called both in the bootblock and
romstage. The rationale behind setting this up twice is to ensure
bootblock-romstage compatibility in the future if for instance VBOOT
is used.
This moves the console init to the bootblock.
The romstage now runs uncached. Adding a prog_run hooks to set up an
MTRR to cache the romstage will be done in a followup patch.
The default size of 64KiB is not modified for the bootblock as trying
to fit both EHCI and SPI flash debugging needs a more space and 64KiB
is the next power of 2 size that fits it.
TESTED on Thinkpad X200.
Change-Id: I8f59736cb54377973215f35e35d2cbcd1d82c374
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35992
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
With CONFIG_C_ENVIRONMENT_BOOTBLOCK it makes more sense to rely on the
size of the bootblock over CONFIG_XIP_ROM_SIZE. To make this work,
only powers of 2 are allowed as bootblock size.
Change-Id: Ic8104ca9c51e4d2eccdb277e4c2111d2da662f3e
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35990
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
To make data flow more explicit, global variables 'MR01Value' and
'MR13Value' are replaced with local variables, which are passed as
function arguments.
BRANCH=kukui
BUG=none
TEST=1. emerge-kukui coreboot
2. Fast calibration succeeded
Change-Id: Id21483092c86c3ae7dbb1173a2b943defe41a379
Signed-off-by: Yu-Ping Wu <yupingso@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36286
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Hung-Te Lin <hungte@chromium.org>
Also:
add some missing includes spotted by Jenkins.
Including <types.h>, is supposed to provide stdint and stddef.
Change-Id: I7bd999b59d1c0bdfa5999bf5805576f94c9a2390
Signed-off-by: Elyes HAOUAS <ehaouas@noos.fr>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33680
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martinroth@google.com>
Also, including <types.h>, is supposed to provide stdint and stddef.
Change-Id: I7e1a3483aae0e3f0a7bc2eb2c1862ae7a325fd9c
Signed-off-by: Elyes HAOUAS <ehaouas@noos.fr>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33683
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martinroth@google.com>
If a device node should be enabled on some variants, but disabled on
others, it had better be declared as disabled (rather than absent) in
base device tree (rather than override tree for the variant disabling
it), and enabled in override tree for the variant needing it, so that
it does not need to be declared once more when adding another variant
with such node disabled.
Change-Id: I4b28360905ae38149ace9ac5d21cd6d5045b7584
Signed-off-by: Bill XIE <persmule@hardenedlinux.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36347
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
This should improve boot times as it skips unnecessary read and
writes.
Change-Id: I8cf36613903783e741b2e3bfeeee5bc29ab26d5b
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36323
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
There is no posibility to prevent loading images from cbfs at this stage
For security features prog_locate_hook() is added. This hook can be used
to prevent loading the image.
BUG=N/A
TEST=Created verified binary and verify logging on Facebook FBG-1701
Change-Id: I12207fc8f2e9ca45d048cf8c8d9c057f53e5c2c7
Signed-off-by: Frans Hendriks <fhendriks@eltan.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/30811
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Using docker to build to documentation eases the process of building
the documentation. Given that some versions of sphinx are
incompatible, the option to use docker is presented first.
Change-Id: I6c18f81a829364ada1859c04ba2dc4f886934bcc
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36105
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change removes an inversion of GPP_D17 that caused the
device to get stuck in a reboot loop because the kernel was crashing
within the first couple seconds of kernel boot.
BUG=b:142515200
BRANCH=none
TEST=Flash and boot nocturne, verify boot is stable and that device
doesn't reboot after jumping into kernel, and that it passes the
'tast -verbose run <ip> hardware.SensorRing' test.
Change-Id: Ia1408ef6ea92f6b31a9f3eee8720954af3a7c382
Signed-off-by: Nick Vaccaro <nvaccaro@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35967
Reviewed-by: Yicheng Li <yichengli@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The board does not have any graphics port connected to the SoC. Hence,
use the new Kconfig to hide GOP initialization.
Change-Id: Ia88e062bea243369da27b94608f89f0808257688
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36349
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
There are mainboards that do not have any graphics ports connected to
the SoC. It would be senseless to initialize the iGD, thus add a new
mainboard Kconfig to hide the GOP option.
Change-Id: Ica3b3a7a0c8120c95412369a24d8d669fb59fded
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36348
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Also, including <types.h>, is supposed to provide stdint and stddef.
Change-Id: I99918a5a77e759bc7d4192d2c3fd6ad493c70248
Signed-off-by: Elyes HAOUAS <ehaouas@noos.fr>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33681
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martinroth@google.com>
cbmem_top() should simply not be called before memory is initialed,
in order for the implementation to return something meaningful.
Change-Id: I8fe32844af290626a0f91279143fda4d3442680f
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36334
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Michael Niewöhner
So, this is odd in multiple ways. First of all, we fix something:
We work around a weirdness in `make oldconfig` that adds spurious
entries into the `auto.conf` for choices that were given a symbol
name.
When introducing the Ada config package, it seemed reasonable to
use `auto.conf` as source, but it turned out that we didn't use it
as input, only `config.h` and the original `.config` were used. As
the syntax for `.config` is the same as for `auto.conf` we use the
former now as input for Ada, too. One question remains: If `.config`
already contains all required information, what is this `auto.conf`
and what does it want?
Alternatively, we could try to fix `oldconfig` or add a linter to
forbid named choices. I thought, our build test would reject the
latter already. But the `oldconfig` behaviour is too subtle.
We keep a dependency on the `oldconfig` step, to make sure it runs
first.
Change-Id: If3fe6bc782251cdbd696395d3069a1c0bb0ae802
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36320
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
New changes in the latest binutils 2.32 lead to assembler errors causes
ipxe build failure. IPXE uses the divide test which requires /dev/null as
input as well as the output file name.
This patch facilitates the /dev/null as an exception to the current
changes in binutils package while building crossgcc for coreboot leads to
successful build of ipxe and further tests to pass based on /dev/null and
applies automatically during the crossgcc rebuild.
Also, this can be reverted once binutils/ipxe provides an updated release
in this respect.
Fixes: https://ticket.coreboot.org/issues/204
Change-Id: I9f664829b8c42420c0b2ab1f2316150f86ac0b1a
Signed-off-by: Himanshu Sahdev <himanshusah@hcl.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35098
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Since there are no calls where we wouldn't die(), move die() calls into
the fsp_find_* functions.
Change-Id: I750a225999688137421bbc560d9d1f5fdf68fd01
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36314
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Michael Niewöhner
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Remove the calculation of the Reserved Intel MMIO Memory size from
systemagent and memmap, since it is not needed.
The size is used in SA to calculate the space between cbmem_top and TSEG
without DPR and Chipset Reserved Memory. Since this will always be equal
to 0, the reservation will be skipped and TSEG, DPR and Chipset Reserved
Memory will get reserved alltogether.
By reading the code and pratical testing we figured out that:
- TSEG - DPR - reserved - top_of_memory == 0
- TSEG - DPR - reserved == top_of_memory
This means the whole block will never reserve anything because it is
always 0. Hence the code can be removed for simplification.
Tested successfully on X11SSM-F
Change-Id: I0cc730551eb3a79c78a971b40056de8d029f4b82
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36216
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Michael Niewöhner
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Follow-up commit where only files are moved and paths adapted to make
review of the previous commit easier.
Change-Id: Iff1acbd286c2ba8e6613e866d4e2f893562e8973
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35868
Reviewed-by: Michael Niewöhner
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Remove orphaned functionality from the FSP1.1 driver that only has been
used by skylake, which is now FSP2.0-only.
Change-Id: I732f2d6846788d5c03647c6fb620e45b3b66de5f
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36266
Reviewed-by: Michael Niewöhner
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This drops support for FSP 1.1 in soc/intel/skylake, after all boards
have been migrated to FSP 2.0, which is backwards compatible.
Any moving of files happens in a follow-up commit to make review easier.
Change-Id: I0dd2eab0edfda0545ff94c3908b8574d5ad830bd
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35813
Reviewed-by: Michael Niewöhner
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Nico Huber