Note that due to UNKNOWN_TSC_RATE, each stage will have
a slow run of calibrate_tsc_with_pit(). This is easy enough
to fix with followup implementation of tsc_freq_mhz() for
the cpu.
Change-Id: I0f5e16993e19342dfc4801663e0025bb4cee022a
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36525
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Dead increment spotted out using clang-tools.
Value stored to 'some_delay_3_halfcycles' is never read.
Change-Id: I8133f9e8786006bd278d281a132b6a2bd863a967
Signed-off-by: Elyes HAOUAS <ehaouas@noos.fr>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36135
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
This is needed for the definition of the MCHBARx() macros.
Change-Id: I654344451c0568f306f39365fd07371b2d6cff04
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36555
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Currently all stages that need cbmem need an implementation of a
cbmem_top function. On FSP and AGESA platforms this proves to be
painful and a pointer to the top of lower memory if often passed via
lower memory (e.g. EBDA) or via a PCI scratchpad register.
The problem with writing to lower memory is that also need to be
written on S3 as one cannot assume it to be still there. Writing
things on S3 is always a fragile thing to do.
A very generic solution is to pass cbmem_top via the program argument.
It should be possible to implement this solution on every
architecture.
Instead trying to figure out which files can be removed from stages
and which cbmem_top implementations need with preprocessor, rename all
cbmem_top implementation to cbmem_top_romstage.
TESTED on qemu-x86.
Change-Id: I6d5a366d6f1bc76f26d459628237e6b2c8ae03ea
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36144
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
The x86 timers are a bit of a mess. Cases where different stages use
different counters and timestamps use different counters from udelays.
The original intention was to only flip TSC_CONSTANT_RATE Kconfig
to NOT_CONSTANT_TSC_RATE. The name would be incorrect though, those
counters do run with a constant rate but we just lack tsc_freq_mhz()
implementation for three platforms.
Note that for boards with UNKNOWN_TSC_RATE=y, each stage will have a
slow run of calibrate_tsc_with_pit(). This is easy enough to fix with
followup implementation of tsc_freq_mhz() for the platforms.
Implementations with LAPIC_MONOTONIC_TIMER typically will not have
tsc_freq_mhz() implemented and default to UNKNOWN_TSC_RATE. However,
as they don't use TSC for udelay() the slow calibrate_tsc_with_pit()
is avoided.
Because x86/tsc_delay.tsc was using two different guards and nb/via/vx900
claimed UDELAY_TSC, but pulled UDELAY_IO implementation, we also switch
that romstage to use UDELAY_TSC.
Change-Id: I1690cb80295d6b006b75ed69edea28899b674b68
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33928
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This adds a common cbmem_top implementation to all coreboot target.
In romstage a static variable will be used to cache the result of
cbmem_top_romstage.
In ramstage if CONFIG_RAMSTAGE_CBMEM_TOP_ARG is set a global variable
needs to be populated by the stage entry with the value passed via the
calling arguments. if CONFIG_RAMSTAGE_CBMEM_TOP_ARG is not set the
same implementation as will be used as in romstage.
Change-Id: Ie767542ee25483acc9a56785ce20a885e9a63098
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36273
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.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>
During the raminit the CPU gets reset but the platform does not. To
deal with TPM init failure (a TPM can only be started up once) ignore
the invalid POSTINIT return code.
Change-Id: Ib15e796914d3e6d5f01b35fa46b3ead40f56122b
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36055
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This is already done in nb/intel/nehalem/romstage.c.
Change-Id: I2dc34af874ac96af31ca9ebe6a43b2805da32d18
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36039
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
During the raminit the CPU gets reset, so reprogram those bits in
PM1_CNT such that the CPU remains aware that this is a S3 resume path
after the reset.
Change-Id: I8f5cafa235c8ab0d0a59fbeeee3465ebca4cc5d0
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36038
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
This decreases the output verbosity of RAM_DEBUG to be useful.
Change-Id: I9fa681303da0e063dd2ca789d7711151b2365f16
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/36036
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This assumes ACPI aware OS also have a driver for this particular
hardware, which is the case for at least Linux. This saves ~60ms on S3
resume.
Change-Id: I2dcd399fee8e2d1cd1b70e60e1669a49c7aa8cb4
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35995
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
This does the following:
- Add gen[1-4]_dec options to the devicetree to set up generic LPC
decode ranges in the southbridge code.
- Move setting up some default decode ranges to a common place. If
somehow a board needs to override this behavior it can happen in the
mb_setup_superio() hook (that will be renamed when moving to
C_ENVIRONMENT_BOOTBLOCK).
Change-Id: I3d904b1125bc410c11aa73a89b1969284e88dac1
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35991
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
A few notable changes:
- Microcode init is done in assembly during the CAR init.
- The DCACHE_BSP_STACK_SIZE is set to 0x2000, which is the same size
against which the romstage stack guards protected.
- The romstage mainboard_lpc_init() hook is removed in favor of the
existing bootblock_mainboard_early_init().
Change-Id: Iccd7ceaa35db49e170bfb901bbff1c1a11223c63
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35951
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Some of the initialization isn't necessary before console INIT is
done.
EHCI debug still works fine on the Lenovo Thinkpad X201.
Change-Id: I0c33efd98844f7188e0258cf9f90049d45145e7c
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35949
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Some boards with the G41 chipset lacked programming CIR, so this
change add that to those boards too.
Change-Id: Ia10c050785170fc743f7aef918f4849dbdd6840e
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35795
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Devices behind LPC can expose more buses (e.g. I2C on a super-i/o).
So we should scan buses on LPC devices, too.
Change-Id: I0eb005e41b9168fffc344ee8e666d43b605a30ba
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/29474
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
RCBA_HPTC needs to be read back to consistently enable HPET.
This ought to fix raminit failing sometimes and SeaBIOS endlessly
waiting for user input.
TESTED on Intel D510MO, Fixes SeaBIOS waiting for input, without a
timeout.
Change-Id: I20a25fd97cd09fedb70469262c64d8d3828bb684
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35758
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The assumption is made that an ACPI aware an OS does not rely on
firmware to initialize the display.
TESTED on a Lenovo Thinkpad X201 with Linux 5.2, display still works
after S3, more than 200ms in time saved (dropped from 411ms to 182ms
in total in one test).
Change-Id: I36219e6d04db561d4f2ddb6e962166c598d5bc4f
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35773
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
This change does the following:
- Move PCH init code from the common romstage to sb code, this allows
for easier reuse in bootblock
- Provide a common minimal LPC io decode setup, mainboards can
override this in the mainboard_lpc_init if required
- Set up LPC generic IO decode up in romstage based on devicetree
settings
- Remove the ramstage LPC generic IO decode from ramstage as this is
now done in romstage.c
- Get rid of unneeded setup of spi_read configuration in BIOS_CNTL as
this is already done in the bootblock.
Change-Id: I3f448ad1fdc445c4c1fedbc8497e1025af111412
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35772
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Move the mainboard_romstage_entry to a common location and provide
mainboard specific callbacks.
Change-Id: Ia827053617cead5d2cf8e9f06cb68c2cbb668ca9
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35771
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
This was simply copied from sandybridge/gma.c.
All these registers read back 0xffffffff or 0 or don't respond to
reads.
Change-Id: I094e7caa889a3175477aa78b91545ca804d423c8
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35746
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
On these CPUs the MCHBAR window is 16KiB large. This code was just
copied from SNB.
Change-Id: I263cfc678a2eb8eeee8ab9157c749359064a9be8
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35743
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
The extra PCI bus RST# and 200ms delay there was workaround
for custom add-on hardware.
Change-Id: I38c4677cfb41d620498be8e0c257b517995bad5c
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35765
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Intel adopted xx_DEVFN_xx naming for macros expanding to
PCI_DEVFN() starting with apollolake. The ones named
xx_DEV_FUNC are being renamed, or dropped, if they
were generally not used at all for a platform.
Change-Id: I6ead2bc5e41a86c9aeef730f5664a30406414c8c
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35730
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
This is a PCI standard register, no need to alias its
definitions under different names.
Change-Id: Iea6b198dd70fe1e49b5dc0824dba62628dedc69a
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/35521
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
The implementation of udelay() with LAPIC timers
existed first, as we did not have calculations
implemented for TSC frequency.
Change-Id: If510bcaadee67e3a5792b3fc7389353b672712f9
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/34200
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
