The name `..._index` is confusing since the maximum index of an array is
not `ARRAY_SIZE(array)` but `ARRAY_SIZE(array) - 1`.
Rename `uart_max_index` to `uart_ctrlr_config_size` to make the name
match the variable´s value.
Change-Id: I7409c9dc040c3c6ad718abc96f268c187d50d79c
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49305
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Two USB2 ports 4 and 9 are assigned to type C connectors on Voxel board.
This update configures these USB2 ports for Type C which will allow USB2
port reset message upstream from PCH to CPU to recover a USB3 device
that downgraded to USB2 to upgrade back to USB3.
BUG=b:176575892
TEST=Booted to kernel on Voxel board and verified usb2 port reset
message enable bits through pch xhci_mmio_base + R_XHCI_MEM_U2PRM_U2PRDE
where the offset register R_XHCI_MEM_U2PRM_U2PRDE has value 0x92f4.
Validated various USB3 devices enumeration.
Signed-off-by: John Zhao <john.zhao@intel.com>
Change-Id: Ia370a449a41701e690c1c507d70bedfce2076a65
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49053
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nick Vaccaro <nvaccaro@google.com>
Reviewed-by: Chiranjeevi Rapolu <chiranjeevi.rapolu@intel.corp-partner.google.com>
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
When building as part of the coreboot build system, use the same
mechanism as other tools (cbfstool, amdfwtool, ...) so that abuild
builds ifdtool once into sharedutils instead of once per board (while
avoiding other race conditions, too).
Change-Id: I42c7b43cc0859916174d59cba6b62630e70287fd
Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49312
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
The bugs happen on real hardware or in qemu with KVM enabled.
The very same code runs on some real devices and it runs in qemu
with KVM disabled.
The bugs are so strange that no root cause could be found yet.
Change-Id: I01050f2e38f92c6b96e3258a5b619aa9ee685acc
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/44733
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
This change adds a helper function `pcie_rp_enable_mask()` that
returns a 32-bit mask indicating the status (enabled/disabled) of PCIe
root ports (in the groups table) as configured by the mainboard in the
device tree.
With this helper function, SoC chip config does not need to add
another `PcieRpEnable[]` config to identify what root ports are
enabled.
Change-Id: I7ce5fca1c662064fd21f0961dac13cda1fa2ca44
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48968
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: EricR Lai <ericr_lai@compal.corp-partner.google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change updates the definition of config_of_soc() to a macro that
expands to __pci_0_00_0_config instead of accessing the config
structure by referencing the struct device. This allows linker to
optimize out unused portions of the device tree from early stages.
With this change, bootblock .text section size drops as follows:
Platform | Size without change | Size with change | Reduction |
---------------|---------------------|------------------|-------------|
GLK (ampton) | 27112 bytes | 9832 bytes | 17280 bytes |
APL (reef) | 26488 bytes | 17528 bytes | 8960 bytes |
TGL (volteer2) | 47760 bytes | 21648 bytes | 26112 bytes |
CML (hatch) | 40616 bytes | 22792 bytes | 17824 bytes |
JSL (waddledee)| 37872 bytes | 19408 bytes | 18464 bytes |
KBL (soraka) | 31840 bytes | 21568 bytes | 10272 bytes |
As static.h is now included in device.h which gets pulled in during
the unit tests, a dummy static.h is added under tests/include.
Change-Id: I1fbf5b9817065e967e46188739978a1cc96c2c7e
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49215
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Nico Huber <nico.h@gmx.de>
By default, the HS400 mode of GL9763E is slow mode (150MHz).
Therefore, the slow mode is disabled for HS400 running at 200MHz.
For eMMCs such as Hynix (H26M74002HMR) on HS400, adjust the internal
Rx latch dealy of HS400 to have better compatibility.
Signed-off-by: Ben Chuang <benchuanggli@gmail.com>
Change-Id: I84844c2432d4223d9929182c5c430915e52875b7
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49079
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Commit 542307b815 (broadwell: Add small delay before Flex Ratio reboot)
introduced a workaround for Broadwell. Implement it on Haswell as well.
Since this is only necessary when a TPM is present on a system, only do
the delay (which is not that small, to be honest) on TPM-enabled builds.
Change-Id: Id8b58e9fa2a1c81989305f5b4b765b82c01e1596
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/46941
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The MSR only needs to be set when IO MWAIT redirection is to be enabled.
This was copied from Sandy Bridge, which already had this inconsistency.
Change-Id: I424333afd654db9a7e180e9a2c31d369e3d92fd6
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/46917
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
We only need `$_OBJ` in the include path for in-tree builds. Also,
curses only need special handling for those and PDCurses turned out
to need many more include paths.
Change-Id: Idd29ef33065033e26ba61b09d412d8ca3566d643
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/47631
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Booting on Kingston (EMMC64G-TA29/TX29-HP) and Hynix (H26M74002HMR) eMMC
currently fails due to R/W error. This is a workaround to finetune the
data latch timing by verdor-specific setting of GL9763E. For improving
the compatibility of GL9763E with these two eMMC.
Signed-off-by: Renius Chen <reniuschengl@gmail.com>
Change-Id: Iddb145ed6a9edb2d7a50248e64659cda78b88ae6
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48941
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: YH Lin <yueherngl@google.com>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Test: Linux adds the cpuidle sysfs interface; Windows with s0ix_enable=1
boots without crashing with an INTERNAL_POWER_ERROR.
Change-Id: Icccd9d15a9e9a22c9bfe7a9843e95d77013c9c8f
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49047
Reviewed-by: Lance Zhao
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Test: Linux adds the cpuidle sysfs interface; Windows with s0ix_enable=1
boots without crashing with an INTERNAL_POWER_ERROR.
- Windows and Linux tested on google/akemi
- Linux tested on clevo/cml-u
Change-Id: I51fdf52419aa7f059b70a906fd8bdac88d5b6046
Tested-By: Matt DeVillier <matt.devillier@gmail.com>
Tested-by: Michael Niewöhner <foss@mniewoehner.de>
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49046
Reviewed-by: Lance Zhao
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Add support for the Intel LPIT table to support reading Low Power Idle
Residency counters by the OS. On platforms supporting S0ix sleep states
there can be two types of residencies:
* CPU package PC10 residency counter (read from MSR via FFH interface)
* PCH SLP_S0 assertion residency counter (read via memory mapped
interface)
With presence of one or both of these counters in the LPIT table, Linux
dynamically adds the corresponding attributes to the cpuidle sysfs
interface, that can be used to read the residency timers:
* /sys/devices/system/cpu/cpuidle/low_power_idle_cpu_residency_us
* /sys/devices/system/cpu/cpuidle/low_power_idle_system_residency_us
The code in src/acpi implements generic LPIT support. Each SoC or
platform has to implement `acpi_fill_lpit` to fill the table with
platform-specific LPI state entries. This is done in this change for
soc/intel/common, while being added as its own compilation unit, so SoCs
not yet using common acpi code (like Skylake) can use it, too.
Reference:
https://uefi.org/sites/default/files/resources/Intel_ACPI_Low_Power_S0_Idle.pdf
Test: Linux adds the cpuidle sysfs interface; Windows with s0ix_enable=1
boots without crashing with an INTERNAL_POWER_ERROR.
- Windows and Linux tested on google/akemi together with CB:49046
- Linux tested on clevo/cml-u, supermicro/x11ssmf together with CB:49046
Change-Id: I816888e8788e2f04c89f20d6ea1654d2f35cf18e
Tested-by: Matt DeVillier <matt.devillier@gmail.com>
Tested-by: Michael Niewöhner <foss@mniewoehner.de>
Signed-off-by: Shaunak Saha <shaunak.saha@intel.com>
Signed-off-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49045
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Drop the support for the Intel Cannon Lake SoC for various reasons:
* Most people can't use coreboot on Cannon Lake, since the required FSP
binaries aren't publicly available. Given that FSP binaries for several
newer platforms have been released, it's very unlikely that Cannon Lake
FSP will ever be released.
* It seems there is no interest in this, since the reference mainboard
is the only available mainboard in tree.
Also, remove the related reference mainboard intel/cannonlake_rvp and
its FSP headers in intel/fsp2_0/cannonlake.
Change-Id: I8f698e16099acb45444b2bc675642d161ff8c237
Signed-off-by: Felix Singer <felixsinger@posteo.net>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48775
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change emits chip config pointers for PCI devices on root bus in
static_devices.h so that the config structure can be accessed directly
without having to reference the device structure. This allows the
linker to optimize out unused parts of the device tree from early
stages like bootblock.
Change-Id: I1d42e926dbfae14b889ade6dda363d8607974cae
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49214
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change updates various uart_* functions to use simple(_s_)
variants of PCI functions. This is done for a few reasons:
* __SIMPLE_DEVICE__ check can be dropped since the same data type can
be used in early stages and ramstage.
* Removes the requirement on early stage to walk the device tree to
get access to the device structure. This allows linker-based device
tree optimizations for early stages.
As part of this change, uart_get_device() is refactored and a new
function uart_console_get_devfn() is added which returns pci_devfn_t
in MMCONF format. It is then used directly by the _s_ variants of PCI
functions.
Change-Id: I344037828118572ae5eb27c82c496d5e7a508a53
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49213
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
This change renames `struct uart_gpio_pad_config` to `struct
uart_controller_config` and adds a new parameter devfn (which expects
devfn for the UART controller corresponding to the index in
PCI_DEVFN() format). This gets rid of the SoC callback to get `struct
device` pointer to the UART controller device.
Change-Id: Id0712a0038f2cc1a61b8b5a58fa155f14e7949a5
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49212
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Prevent the FSP from writing its default SVID SDID values of 8086:7270
for internal devices as this locks most of the registers. Allows the
subsystemid values set in devicetree to be used.
A description of this SSID table override behavior, along with example
code, is provided in the TigerLake FSP Integration Guide, section
15.178 ("SI_CONFIG Struct Reference").
The xHCI and HDA devices have RW/L registers rather than RW/O registers.
They can be written to multiple times but cannot be modified after
being locked, which happens during FspSiliconInit. Because coreboot
populates subsystem IDs after SiliconInit, these devices specifically
must be written beforehand or will otherwise be locked with their
default values of 0:0.
Tested by checking lspci output on System76 galp3-c (WHL), oryp5 (CFL),
and oryp6 (CML).
References:
- TigerLake FSP Integration Guide
- Intel Document Number 337868-002
Change-Id: Ieaa45ef7fa8e0da4a25b9174ded1ea0c5d9c4b4e
Signed-off-by: Jeremy Soller <jeremy@system76.com>
Signed-off-by: Tim Crawford <tcrawford@system76.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49104
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
While FORCE_PWR is set high, it prevents retimer from entering low power
state. S0ix failure occurs while USB4 Gatkex is connected on Port-0.
This change sets FORCE_PWR(GPP_H10) low. This FORCE_PWR GPIO will be
toggled by kernel through DSM method while updating retimer firmware.
BUG=b:174166586
Cq-Depend: chromium:2594438
TEST=Verifed s0ix cycles with USB4 Gatkex connected on Port-0.
Signed-off-by: John Zhao <john.zhao@intel.com>
Change-Id: Ie4b442e1078379c522a94bfdc00cd99e6f9b8170
Reviewed-on: https://review.coreboot.org/c/coreboot/+/48649
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
With Intel CPX-SP FSP ww01 release, CidBitMap field is added to
DimmDevice struct in hob_memmap.h.
The copyright statements were updated to accomodate year 2021.
gpio_fsp.h is not needed any more as coreboot takes over GPIO
configuration.
Signed-off-by: Jonathan Zhang <jonzhang@fb.com>
Change-Id: I3242c8b50401757a28de8a9e9c71fb95bc0515dc
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49246
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
To work around various bugs running KVM enabled, copy page tables to
DRAM in assembly before jumping to x86_64 mode.
Tested on QEMU using KVM, no more stange bugs happen:
Tested on host
- CPU Intel(R) Core(TM) i7-7700HQ
- Linux 5.9
- qemu 4.2.1
Used to crash on emulating MMX instructions and failed to translate
some addresses using the virtual MMU when running in long mode.
Tested on host
- CPU AMD EPYC 7401P 24-Core Processor
- Linux 5.4
- qemu 4.2.1
Used to crash on jumping to long mode.
Change-Id: Ic0bdd2bef7197edd2e7488a8efdeba7eb4ab0dd4
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49228
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Wrap `r` in parentheses to avoid unexpected behavior with compound
expressions. This prevents `CxDRBy_BOUND_MB(r+1, base)` from triggering
undefined behavior when `r = 2`, as the shift would be greater than 32.
Change-Id: I14235b2708ab502d842da677451c14203a469b45
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49261
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Platform can now select VGA_ROM_RUN_DEFAULT Kconfig to perform graphics
initialization for PCI-E based discrete card through VGA OpRom
(SoC or Mainboard user can't select VGA_ROM_RUN directly because
it's part of choice option).
(Note: Some payloads, like SeaBIOS, are also able to run Option ROMs,
so coreboot does not need to enable VGA_ROM_RUN Kconfig)
For payload like depthcharge, create VGA_ROM_RUN_DEFAULT Kconfig
for mainboard to select design with DGPU where OpROM is embedded
inside the DGPU card.
Allow auto selection of VGA_ROM_RUN_DEFAULT from VGA_BIOS Kconfig.
Also NO_GFX_INIT Kconfig to avoid running VGA_ROM_RUN
by default in case SeaBIOS is used.
TEST=Able to get Pre-OS splash screen with AMD Radeon RX 5700 PCI-E
DGPU when mainboard user selects VGA_ROM_RUN_DEFAULT.
Change-Id: Iecb2fcdb105af449bc20ad727759cdef17d5e376
Signed-off-by: Subrata Banik <subrata.banik@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49016
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
