Instead of having the code for the RAM init time stamps in each
mainboard’s `romstage.c`, factor it out to the northbridge code, done in
commit 771328f7 (intel/i945: add timestamps in romstage).
Change-Id: Ibb699a1fea2f0b1f3c6564d401542d2fb3249f5a
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: https://review.coreboot.org/17994
Reviewed-by: Elyes HAOUAS <ehaouas@noos.fr>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Using a conditional statement to return 0 or 1 depending on a logic
value is unnecessarily complex.
Change-Id: I449ce2b71b72374de5ec4986f9cc9f91a67856ee
Signed-off-by: Elyes HAOUAS <ehaouas@noos.fr>
Reviewed-on: https://review.coreboot.org/29265
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
There are some small mistakes in these recently-added mainboards:
- board_info.txt: Lists board socket incorrectly.
- cmos.default: Loglevel was decreased some time ago.
- devicetree.cb: Spelling mistake.
- Kconfig: Mainboard name does not have a hyphen.
Change-Id: I08d9b06e79683acd3994b84647bce401ed6741e2
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/29446
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Nico Huber <nico.h@gmx.de>
And here comes the mess...
This just renames do_hard_reset() to do_board_reset() and keeps current
behaviour. As these are never called from chipset or board code but only
from common code, it's likely that their implementations are untested
and not what we actually want. Also note, that sometimes implementations
for rom- and ramstage differ considerably.
Change-Id: Icdf55ed1a0e0294933f61749a37da2ced01da61c
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/29058
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Phase 1. Due to the size of the effort, this CL is broken into several
phases.
Change-Id: I0236e0960cd1e79558ea50c814e1de2830aa0550
Signed-off-by: Elyes HAOUAS <ehaouas@noos.fr>
Reviewed-on: https://review.coreboot.org/29065
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Richard Spiegel <richard.spiegel@silverbackltd.com>
This mainboard is quite similar to the p5qc. The main differences being a second
PEG slot, the IDE slot and being DDR2 only.
The following was tested:
- both PEG slots populated (coreboot sets legacy VGA decoding on the GPU in the
black slot)
- USB
- Ethernet NIC
- PS2 Keyboard
- COM1
- S3 resume
Change-Id: I49a4bca4256e2a905aff3252eca76387c81152c1
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/29102
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
A shortcoming of this driver is that if multiple devices with the same PCI ID
are present and don't have an eeprom, they would all get the same macadress set.
The r8168 driver deals with such cases so it should be easy to implement if
needed.
Change-Id: I5c32df00e25453c350a45e7f1ee6834b89c4289f
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/28265
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
SeaBIOS does not seem to like the Marvel IDE controller, so disabled SeaBIOS
support for ATA. It works fine in Linux afterwards.
Working:
- SATA on southbridge port
- SATA on marvel IDE controller ports (only in Linux)
- USB
- COM1
- PS2 Keyboard
- DDR2 DIMMs
- PCIe x16 PEG port
- PCI port
- NIC (needs a driver to set macaddress)
- S3 resume
Not working:
- SeaBIOS with ATA support (long timeout marvel controller so disabled)
- DDR3 fails because the proper clock signal does not get enabled. Even when
fixing this it fails later or during memtest, so it should be considered
unsupported for now
Untested:
- PCIe x1 ports (expected to work)
- sound (expected to work)
TODO:
add documentation
Change-Id: I4a81940707566776bd048904ca1387fea741fece
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/28264
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
These files are being updated to match the prevailing style
of cmos.default files.
Change-Id: I47d31d6fec8c9eb856aed0c63824d9556b7705e4
Signed-off-by: Elyes HAOUAS <ehaouas@noos.fr>
Reviewed-on: https://review.coreboot.org/28051
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Martin Roth <martinroth@google.com>
Used default console log level is 7 in src/console/Kconfig.
So let cmos.default use the same level as default.
Change-Id: Ia39ee457a8985142f6e7a674532995b11cb52198
Signed-off-by: Elyes HAOUAS <ehaouas@noos.fr>
Reviewed-on: https://review.coreboot.org/28006
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Martin Roth <martinroth@google.com>
The file `data.vbt` matches the VBT in the latest version of the vendor
firmware (version 3603). Tested with Linux 4.9 and everything works as
expected.
Change-Id: I8e3b1d274ac0df63989d966f477013e780611fa1
Signed-off-by: Tristan Corrick <tristan@corrick.kiwi>
Reviewed-on: https://review.coreboot.org/28050
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Since only a handful of boards have descriptor blobs in the tree, it makes no
sense to have `HAVE_IFD_BIN` enabled by default then disabled on each mainboard.
This patch flips the default value of said variable, rendering all current
overrides unnecessary. The few boards which have an IFD in the blobs repo use
`select HAVE_IFD_BIN` to enable adding the IFD by default.
Since `HAVE_ME_BIN` depends on `HAVE_IFD_BIN`, the former has been removed
alongside the latter, and has been added to the boards with a ME blob as
`select HAVE_ME_BIN`.
Both `HAVE_IFD_BIN` and `HAVE_ME_BIN` have been removed from autoport as well.
Change-Id: I330c4886f8bea4b1a8ecad6505a0e5cc381654d1
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/27218
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Tested with GRUB 2.02 as a payload, booting Debian GNU/Linux 9.5 with
kernel 4.9. This code is based on the output of autoport.
The file `data.vbt` matches the VBT in the latest version of the vendor
firmware (version 4601).
This board works well under coreboot. A list of what works and what
doesn't can be found in the documentation part of this commit. To
summarise: the only known issues are that S3 suspend/resume doesn't
work, and that there is no automatic fan control via the super I/O.
Change-Id: I2a0579f486d3a44de2dd927fa1e76b90c3b48f62
Signed-off-by: Tristan Corrick <tristan@corrick.kiwi>
Reviewed-on: https://review.coreboot.org/27798
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Logic: If vboot is not used and the tpm is not initialized in the
romstage makes use of the ramstage driver to initialize the TPM
globally without having setup calls in lower SoC level implementations.
* Add TPM driver in ramstage chip init which calls the tpm_setup
function.
* Purge all occurrences of TPM init code and headers.
* Only compile TIS drivers into ramstage except for vboot usage.
* Remove Google Urara/Rotor TPM support because of missing i2c driver
in ramstage.
Change-Id: I7536c9734732aeaa85ccc7916c12eecb9ca26b2e
Signed-off-by: Philipp Deppenwiese <zaolin@das-labor.org>
Reviewed-on: https://review.coreboot.org/24905
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Both southbridges need to be done at once since this southbridge code
is used for different northbridges, which fails to compile when done
separately.
This needs an acpi_name functions in the northbridge code to be
defined.
TESTED on Intel DG43GT: show correct PIRQ ACPI entries in
/sys/firmware/acpi/tables/SSDT.
Change-Id: I286d251ddf8fcae27dd07011a1cd62d8f4847683
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/22981
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested with GRUB 2.02 as a payload, booting Arch Linux as
well as Debian. This code is based on the output of autoport
as well as other mainboards supported in coreboot already.
Working:
- Serial port I/O
- S3 suspend/resume. Untested with SeaBIOS since it failed
to resume on a similar board. It is likely to be due to
low memory corruption, but I have not worked on it.
- USB ports and headers
- USB3 ports attached to the ASM1042 controller. SeaBIOS can
boot from them, and it is likely GRUB can detect devices on
those ports as well. The chip has a small SPI flash nearby,
which seems to hold an Option ROM.
- Gigabit Ethernet
- Integrated graphics (libgfxinit)
- VGA BIOS for integrated graphics init
- PCIe x16 graphics
- PCIe x1
- SATA controller
- Hardware Monitor
- Fan Control (fancontrol on linux works well)
- Native raminit
- flashrom, using the internal programmer. Tested with coreboot,
as well as with the vendor firmware.
- NVRAM settings. Only debug_level has been tested.
Untested:
- DVI port. It can detect a "fake" display, that is, an
EEPROM connected to the DVI port. Thus, gma-mainboard.ads
has been setup accordingly.
- PS/2 port.
- Audio: Only rear output (green) has been tested.
- EHCI debug.
- Parallel port header.
- Non-Linux OSes
- ACPI thermal zone and fan control (probably not working)
Not working:
- Booting from devices attached to the ASM1061 controller.
Devices on ports work fine once Linux has loaded.
- Any SATA devices with Tianocore (payload issue)
Change-Id: I7e89ebe43a2e1ff0308f4876e98bbf2f5a0d85f2
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/26419
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Shuffle words and drop the _DATA_FILE suffix.
Change-Id: I0b0d50ea729e5580c0bc7b43f250ff387ce59cfc
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/26898
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Provide empty stub implementations for set_pcie_reset() and
set_pcie_dereset(), many boards do not provide a proper one.
Change-Id: Ia6811442905ef1776fa5a8e3f5d4433e86e42f88
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/26776
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Callsite declared returning int, which makes more sense
than u8 the motherboard side code defined the functions
with.
Change-Id: I8ee83aa2833408ad163c9011a076e08578f3ca6f
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/26774
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Along with other patches submitted for review to get the chipset
parts working, this allows Linux or other OS to use a TPM module
plugged into the 20-pin LPC header on the board, by exposing its
presence through the ACPI and PNP tables.
This patch adds to the Kconfig and devicetree.cb files.
Tested with the TPM/FW 3.19 and the trousers tools.
Change-Id: I8c1aea245f81fa44a6bdd5301bbee958cbcdfaaa
Signed-off-by: Kevin Cody-Little <kcodyjr@gmail.com>
Reviewed-on: https://review.coreboot.org/26193
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Platforms with LATE_CBMEM_INIT were agreed to be
removed with 4.7 release late 2017.
Change-Id: I0ecbb40f8c7ebdf68217f50af5624905d9005c64
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/26671
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Since the SIOW method doesn't use any arguments, don't pass it any,
and initialize it as not using any.
Change-Id: I3fa2ab8afb7d09c176a94bbd1db27587c36030cd
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/26126
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Incorrect interrupt routing configuration prevented handling
interrupts for devices behind PCIe bridges 00:02.1 and 00:02.5.
With the new configuration, devices work as expected.
Tested with Linux 4.10 booted with the "pci=nomsi" parameter.
Change-Id: I3c95be7ba6207697afc7983d4b5f9d9a28584723
Signed-off-by: Gergely Kiss <mail.gery@gmail.com>
Reviewed-on: https://review.coreboot.org/23771
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Use of device_t has been abandoned in ramstage.
Change-Id: I8fe817bb514c69a647c2208a0573a2c5fe98722d
Signed-off-by: Elyes HAOUAS <ehaouas@noos.fr>
Reviewed-on: https://review.coreboot.org/26076
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
