Variable length arrays were a feature added in C99 that allows the
length of an array to be determined at runtime. Eg.
int sum(size_t n) {
int arr[n];
...
}
This adds a small amount of runtime overhead, but is also very
dangerous, since it allows use of an unlimited amount of stack memory,
potentially leading to stack overflow. This is only worsened in
coreboot, which often has very little stack space to begin with. Citing
concerns like this, all instances of VLA's were recently removed from the
Linux kernel. In the immortal words of Linus Torvalds [0],
AND USING VLA'S IS ACTIVELY STUPID! It generates much more code, and
much _slower_ code (and more fragile code), than just using a fixed
key size would have done. [...] Anyway, some of these are definitely
easy to just fix, and using VLA's is actively bad not just for
security worries, but simply because VLA's are a really horribly bad
idea in general in the kernel.
This patch follows suit and zaps all VLA's in coreboot. Some of the
existing VLA's are accidental ones, and all but one can be replaced with
small fixed-size buffers. The single tricky exception is in the SPI
controller interface, which will require a rewrite of old drivers
to remove [1].
[0] https://lkml.org/lkml/2018/3/7/621
[1] https://ticket.coreboot.org/issues/217
Change-Id: I7d9d1ddadbf1cee5f695165bbe3f0effb7bd32b9
Signed-off-by: Jacob Garber <jgarber1@ualberta.ca>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33821
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
In addition to zero IccMax specified by mainboard with socketed CPU, allow
a zero LoadLine default.
The SoC code will fill in the default AC/DC LoadLine values are per
datasheets:
* "7th Generation Intel® Processor Families for H Platforms, Vol 1"
Document Number: 335190-003
* "7th Generation Intel® Processor Families for S Platforms and
Intel ®Core™ X-Series Processor Family, Vol 1"
Document Number: 335195-003
The AC/DC LoadLine is CPU and board specific.
TODO: Find out how to get the LoadLine from vendor firmware and find out
how to map those to different CPU LoadLines.
Change-Id: I849845ced094697e8700470b4af95ad0afb98e3e
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34938
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Reviewed-by: Maxim Polyakov <max.senia.poliak@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Datasheets used:
* "7th Generation Intel® Processor Families for H Platforms, Vol 1"
Document Number: 335190-003
* "7th Generation Intel® Processor Families for S Platforms and
Intel ®Core™ X-Series Processor Family, Vol 1"
Document Number: 335195-003
This allows mainboards to specify a zero IccMax, which all mainboards with
socketed CPU should do.
Change-Id: I303c5dc8ed03e9a98a834a2acfb400022dfc2fde
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34937
Reviewed-by: Maxim Polyakov <max.senia.poliak@gmail.com>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Use a switch case to find the correct VR config.
The following commit will add more entries for which a lookup table
isn't the best solution.
Change-Id: Ib11c3d6e1eb339a0c7358c312a32731d835e7c73
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34936
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Reviewed-by: Maxim Polyakov <max.senia.poliak@gmail.com>
Get rid of defines and hardcode values directly.
Just a cosmetic cleanup to make it more readable.
Change-Id: I3eec44b38af356c3d87235740c65e2c2f6fc5876
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34935
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Maxim Polyakov <max.senia.poliak@gmail.com>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
When entry to romstage is via cpu/intel/car/romstage.c
BIST has not been passed down the path for sometime.
Change-Id: I345975c53014902269cee21fc393331d33a84dce
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34908
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
This is needed for the AST2500 to work, because it uses 4E/4F.
Change-Id: Ie47474e9bf1edfe98555a148469c41283e9a4ea6
Signed-off-by: Christian Walter <christian.walter@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34862
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Frans Hendriks <fhendriks@eltan.com>
This change ports some previous work for Skylake:
cb58683ef5 soc/intel/skylake: Add support for mode-aware DPTF
...to common DPTF code so that we can support mode-aware DPTF for other
Intel platforms.
BUG=b:138702459
BRANCH=none
TEST=Manually test on hatch:
(1)Add DPTF_TSR0_TABLET_PASSIVE and DPTF_TSR1_TABLET_PASSIVE
to hatch baseboard dptf.asl
(2)Flash custom EC FW code which updates DPTF profile number when
entering/exiting tablet mode
(3)On DUT, see /sys/class/thermal/thermal_zone2/trip_point_{1,2}_temp
updated when device mode is switched (tablet/clamshell)
Signed-off-by: Philip Chen <philipchen@google.com>
Change-Id: I5e7b97d23b8567c96a7d60f7a434e98dd9c69544
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34785
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Paul Fagerburg <pfagerburg@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Some timeouts given were too small when serial console is enabled due to
its spinlock making code runtime worse with every AP present.
In addition we usually don't know how long specific code runs and how
long ago it was sent to the APs.
Remove the timeout argument from mp_run_on_all_cpus and instead wait up
to 1 second, to prevent possible crashing of secondary APs still
processing the old job.
Tested on Supermicro X11SSH-TF.
Change-Id: I456be647b159f7a2ea7d94986a24424e56dcc8c4
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34587
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
There was no code present to call wrmsr with the data we
prepared in the structs. The MSRS are already set up by FSP,
just reference with the more recent names of PRMRR and UNCORE_PRMRR.
Change-Id: Ib49e7af52e1170a1304975ff0ae63f99e106dffe
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34824
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
As most platforms will share the subset of enabling
both low RAM WB and high ROM WP MTRRs, provide them
with a single function.
Add possibility for the platform to skip these if
required.
Change-Id: Id1f8b7682035e654231f6133a42909a36e3e15a1
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34809
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
There are benefits in placing the postcar_frame structure
in .bss and returning control to romstage_main().
Change-Id: I0418a2abc74f749203c587b2763c5f8a5960e4f9
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34808
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
At the moment we only have two splitting of TSEG,
one with and one without IED. They can all use
same implementation.
Make configuration problems of TSEG region assertion
failures.
Rename file from stage_cache.c to tseg_region.c to
reflect it's purpose.
Change-Id: I9daf0dec8fbaaa1f4e6004ea034869f43412d7d5
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34776
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: David Guckian
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Add code to read SPD data, parse it and save into SMBIOS table. This is
implemented for socketed DDR4 chips only. For soldered-down memory this
is not implemented and probably won't be ever needed.
TEST=tested on OCP Monolake mainboard, and found dmidecode -t memory to
work. The stack has also been tested on an out-of-tree board.
Signed-off-by: Andrey Petrov <anpetrov@fb.com>
Change-Id: I1162eb4484dab46f1ab9fe3426eecc4d9378e8e2
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34681
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Matt DeVillier <matt.devillier@gmail.com>
Add read/write functions to hook it up with existing SPD retrieval code.
Signed-off-by: Andrey Petrov <anpetrov@fb.com>
Change-Id: I9f5993dc795badf72751a4e6c9d974119a653e30
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34679
Reviewed-by: David Hendricks <david.hendricks@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
IMC is found on certain Xeon processors. On such platforms SPDs are not
connected to SMBus on PCH but to dedicated IMC-owned pins. The purpose
of this driver is to expose access to the i2c/smbus controller associated
with IMC.
Datasheet used: Intel Xeon Processor D-1500 Product Family, Volume 2,
reference 332051-001
This driver is largely based on i2c-imc.c Linux driver.
https://lwn.net/Articles/685475/
TEST=single/double reads and single writes on Xeon-D1500.
Hardware: Open Compute Project Monolake platform.
Signed-off-by: Andrey Petrov <anpetrov@fb.com>
Change-Id: Idbcda1c2273b9a5721fcd9470b4de182192779e7
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34678
Reviewed-by: David Hendricks <david.hendricks@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
In order to use internal UART it needs to 'enabled'. This is normally done
by FSP. However sometimes internal UART is needed before FSP is invoked.
TEST=check if printk() show up in early romstage. Tested on OCP Monolake.
Tested on out-of-tree mainboard to see if UART on LPC still works.
Signed-off-by: Andrey Petrov <anpetrov@fb.com>
Change-Id: I88a7b1a38abf9a09137f6dd75a5a9dee104daaca
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34683
Reviewed-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Any platform should need just one of these.
Change-Id: Ia0ff8eff152cbd3d82e8b372ec662d3737078d35
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34820
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
It is easier to track CAR_GLOBAL_MIGRATION which is
the approach to be deprecated with the next release.
This change enforces new policy; POSTCAR_STAGE=y is
not allowed together with CAR_GLOBAL_MIGRATION=y.
Change-Id: I0dbad6a14e68bf566ac0f151dc8ea259e5ae2250
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34804
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Almost all platforms force it on. Make it enabled by
default but under user control to optionally disable it.
Change-Id: I6b0f19c8bfd6ffed93023d57a1d28ca6acc06835
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34803
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: Martin Roth <martinroth@google.com>
Get rid of this function and its dangerous, weak implementation.
Instead, call pmc_set_power_failure_state() directly from the SMI
handler.
Change-Id: I0718afc5db66447c93289643f9097a4257b10934
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34727
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Needed some Makefile changes to be able to compile for SMM.
Change-Id: Ibf218b90088a45349c54f4b881e895bb852e88bb
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31352
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
pmc_soc_restore_power_failure() is only called from SMM, so add
`pmc.c` to the `smm` class. Once all platforms moved to the new
API, it can be implemented in a central place, avoiding the weak-
function trap.
Change-Id: Ib13eac00002232d4377f683ad92b04a0907529f3
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34726
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Also move pmc_soc_restore_power_failure() which was guarded twice to
not be included in SMM, where the only call lives. Once all platforms
moved to the new API, it can be implemented in a central place, avoi-
ding the weak-function trap.
Change-Id: Ie72753764ecd876e6cb999fa0074d1114ae5efcf
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34725
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This is a consolidation of the respective feature in `soc/intel/*lake/`,
including additional support for MAINBOARD_POWER_STATE_PREVIOUS.
For the latter, firmware has to keep track of the `previous` state. The
feature was already advertised in Kconfig long ago, but not implemented.
SoC code has to call pmc_set_power_failure_state() at least once during
boot and needs to implement pmc_soc_set_afterg3_en() for the actual
register write.
Change-Id: Ic6970a79d9b95373c2855f4c92232d2aa05963bb
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34724
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Currently SPI bus id is used to map to the controller in order to set
the controller state. In certain platforms SPI bus id might not be
exactly the same as GSPI bus id. For example, in Intel platforms SPI bus
id 0 maps to fast spi i.e. SPI going to the flash and SPI bus id 1 .. n
map to GSPI bus id 0 .. n-1. Hence using SPI bus id leads to mapping to the
GSPI controller that is not enabled. Use the GSPI id bus so that the right
controller is set to active state. This fixes the regression introduced
by CB:34449
BUG=b:135941367
TEST=Boot to ChromeOS.
Change-Id: I792ab1fa6529f5317218896ad05321f8f17cedcd
Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34761
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Subrata Banik <subrata.banik@intel.com>
Previous setting was correct but assumed SMI handler is
always located at the beginning of TSEG. Break the assumption.
Change-Id: I5da1a36fc95f76fa3225498bbac41b2dd4d1dfec
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34730
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The callers don't necessarily check return value of
function. Make sure the parameters are not left
uninitialised in that case.
Change-Id: Ic02db2d35b2ec88506320e7df609940de4aef005
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34708
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Do this to avoid some amount of explicit typecasting
that would be required otherwise.
Change-Id: I5bc2c3c1dd579f7c6c3d3354c0691e4ba3c778e1
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34706
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
This is declared weak so that platforms that do not
have smm_subregion() can provide their own implementation.
Change-Id: Ide815b45cbc21a295b8e58434644e82920e84e31
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34704
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Add explicit CBMEM_STAGE_CACHE option. Rename
CACHE_RELOCATED_RAMSTAGE_OUTSIDE_CBMEM to TSEG_STAGE_CACHE.
Platforms with SMM_TSEG=y always need to implement
stage_cache_external_region(). It is allowed to return with a
region of size 0 to effectively disable the cache.
There are no provisions in Kconfig to degrade from
TSEG_STAGE_CACHE to CBMEM_STAGE_CACHE.
As a security measure CBMEM_STAGE_CACHE default is changed to
disabled. AGESA platforms without TSEG will experience slower
S3 resume speed unless they explicitly select the option.
Change-Id: Ibbdc701ea85b5a3208ca4e98c428b05b6d4e5340
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34664
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
No need to limit these declarations to FSP. Both
PARALLEL_MP_INIT smm_relocate() and TSEG_STAGE_CACHE
can be built on top of this.
Change-Id: I7b0b9b8c8bee03aabe251c50c47dc42f6596e169
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34701
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
