This method will allow preloading the VGA_BIOS_FILE. By preloading the
file, into cbfs_cache we reduce boot time. In the future we can also add
support for loading the second VGA_BIOS_FILE and the DGPU VGA_BIOS_FILE.
BUG=b:179699789
TEST=Boot guybrush to OS and verify 12 ms reduction in boot time
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: Icb54fe3a942e9507ff6f1173ba5620a8f4ce6549
Reviewed-on: https://review.coreboot.org/c/coreboot/+/56581
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Rob Barnes <robbarnes@google.com>
This change makes it so the timers run after each boot state callback,
and after each boot state. This gives coop threads the opportunity to
run more frequently and predictably.
BUG=b:179699789
TEST=Boot guybrush to OS, see SPI transactions progress faster.
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: I9508e7777d52fe934cc09d486abc0dab5cf7dad8
Reviewed-on: https://review.coreboot.org/c/coreboot/+/58955
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Rob Barnes <robbarnes@google.com>
This reverts commit adb393bdd6.
This relands commit 6260bf712a.
Reason for revert:
The original CL did not handle some devices correctly.
With the fixes:
* commit 36721a4 (mb/google/brya: Add GPIO_IN_RW to all variants'
early GPIO tables)
* commit 3bfe46c (mb/google/guybrush: Add GPIO EC in RW to early
GPIO tables)
* commit 3a30cf9 (mb/google/guybrush: Build chromeos.c in verstage
This CL also fix the following platforms:
* Change to always trusted: cyan.
* Add to early GPIO table: dedede, eve, fizz, glados, hatch, octopus,
poppy, reef, volteer.
* Add to both Makefile and early GPIO table: zork.
For mb/intel:
* adlrvp: Add support for get_ec_is_trusted().
* glkrvp: Add support for get_ec_is_trusted() with always trusted.
* kblrvp: Add support for get_ec_is_trusted() with always trusted.
* kunimitsu: Add support for get_ec_is_trusted() and initialize it as
early GPIO.
* shadowmountain: Add support for get_ec_is_trusted() and initialize
it as early GPIO.
* tglrvp: Add support for get_ec_is_trusted() with always trusted.
For qemu-q35: Add support for get_ec_is_trusted() with always trusted.
We could attempt another land.
Change-Id: I66b8b99d6e6bf259b18573f9f6010f9254357bf9
Signed-off-by: Hsuan Ting Chen <roccochen@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/58253
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Yu-Ping Wu <yupingso@google.com>
None of the *_DEVID defines was used in the code, so drop those. The SoC
code uses the PCI ID defines from include/device/pci_ids.h instead.
Since it might still be useful to have the PCI device IDs as a reference
in the SoC's pci_devs.h, add those as comments instead.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I7c77d648dac57b15b56f631bd8b2494676c00a8b
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59268
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Create the vell variant of the brya0 reference board by copying
the template files to a new directory named for the variant.
(Auto-Generated by create_coreboot_variant.sh version 4.5.0).
BUG=b:205908918
BRANCH=None
TEST=util/abuild/abuild -p none -t google/brya -x -a
make sure the build includes GOOGLE_VELL
Signed-off-by: Shon Wang <shon.wang@quanta.corp-partner.google.com>
Change-Id: Ide8ba1c0dd9b5d9ad90556053abf2a597136a10c
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59242
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Software Connection Manager doesn't work with Linux 5.13 or later and
results in TBT ports timing out. Not advertising this results in
Firmware Connection Manager being used and TBT works correctly.
Linux patch:
c6da62a219
Tested on:
* StarBook Mk V
* System76 Oryx Pro 8
Signed-off-by: Sean Rhodes <sean@starlabs.systems>
Change-Id: Ib947c3c9cd843e54d4664509c15336178c0bc99e
Reviewed-on: https://review.coreboot.org/c/coreboot/+/58798
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Tim Crawford <tcrawford@system76.com>
1. Turn off L2C SRAM and reconfigure as L2 cache
Mediatek SoC uses part of the L2 cache as SRAM before DRAM is ready.
After DRAM is ready, we should invoke disable_l2c_sram to reconfigure
the L2C SRAM as L2 cache.
2. Configure DMA buffer in DRAM
Set DRAM DMA to be non-cacheable to load blob correctly.
TEST=build pass
BUG=b:202871018
Signed-off-by: Rex-BC Chen <rex-bc.chen@mediatek.com>
Change-Id: If56d29cdd7d9dfaed05e129754aa1f887a581482
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59247
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Yu-Ping Wu <yupingso@google.com>
CONFIG(SMP) was an invalid condition to use in cases where one
stage requires spinlocks and another one does not. The
stage not requiring spinlock still required <smp/spinlock.h>
to be implemented with no-op stubs.
This reverts commit 037ee4b556
soc/amd/picasso: Add dummy spinlock for psp_verstage
Change-Id: Iba52febdeee78294f916775ee9ce8a82d6203570
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59094
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Add separate gpio table for early eSPI bus init. Remove espi GPIO from
early_gpio_table. This allows for initializing eSPI separately from
other GPIOs. Simplify verstage_mainboard_early_init.
BUG=b:200578885
BRANCH=None
TEST=Build and boot guybrush
Change-Id: I0cd439f207df7c27575ae363b207293d40485bf8
Signed-off-by: Rob Barnes <robbarnes@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59082
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Reviewed-by: Kangheui Won <khwon@chromium.org>
This change adds about 30 KiB to FSP-M. When not using the SPI DMA
controller, this change actually has a ~7 ms boot time penalty. When
we use the DMA engine, we end up with about a 5 ms decrease. Once we
switch to 100 MHz SPI this will help even more since we have effectively
eliminated the decompression time.
BUG=b:179699789
TEST=Boot nipperkin to OS and take boot time measurements
fspm.bin 0x2efc0 fsp 90953 LZMA (233472 decompressed)
fspm.bin 0x2cfc0 fsp 121156 LZ4 (233472 decompressed)
- FSP-M / no async -
| 508 - finished loading body | 177.019 | 179.384 Δ( 2.36, 0.16%) |
...
| 970 - loading FSP-M | 0.346 | 0.346 Δ( 0.00, 0.00%) |
| 17 - starting LZ4 decompress (ignore for x86) | 0.009 | 0.01 Δ( 0.00, 0.00%) |
| 18 - finished LZ4 decompress (ignore for x86) | 53.916 | 59.475 Δ( 5.56, 0.37%) |
- FSP-M / async -
| 508 - finished loading body | 177.185 | 179.689 Δ( 2.50, 0.18%) |
...
| 970 - loading FSP-M | 0.989 | 0.99 Δ( 0.00, 0.00%) |
| 17 - starting LZ4 decompress (ignore for x86) | 9.483 | 12.877 Δ( 3.39, 0.24%) |
| 18 - finished LZ4 decompress (ignore for x86) | 10.833 | 0.312 Δ(-10.52, -0.75%) |
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: I7d0363d27d98d4ed3afc6f802a13ff7986391921
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59029
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Reviewed-by: Rob Barnes <robbarnes@google.com>
Calling reboot_into_recovery with NULL context fails. Initializing ctx
early also fails because the cmos is not ready until after
verstage_soc_early_init. So just reboot and hope for the best.
BUG=None
TEST=Boot guybrush, suspend/resume guybrush
BRANCH=None
Change-Id: I7267a14ab048781b8998d3a6f4220de10e7df250
Signed-off-by: Rob Barnes <robbarnes@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59017
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Reviewed-by: Kangheui Won <khwon@chromium.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
The 64-bit compiler x86_64-linux-gnu-gcc-10 aborts the build with the format
warning below:
CC romstage/soc/intel/xeon_sp/memmap.o
src/soc/intel/xeon_sp/memmap.c: In function 'fill_postcar_frame':
src/soc/intel/xeon_sp/memmap.c:39:62: error: format '%lx' expects argument of type 'long unsigned int', but argument 4 has type 'size_t' {aka 'unsigned int'} [-Werror=format=]
39 | printk(BIOS_DEBUG, "cbmem base_ptr: 0x%lx, size: 0x%lx\n", cbmem_base, cbmem_size);
| ~~^ ~~~~~~~~~~
| | |
| long unsigned int size_t {aka unsigned int}
| %x
As `cbmem_size` is of type `size_t` use the appropriate length modifier
`z`.
Change-Id: I1ca77de1ce33ce1e97d7c8895c6e75424f0769f5
Found-by: gcc (Debian 11.2.0-10) 11.2.0
Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59054
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Lance Zhao
Do not place console output in low-level GPIO functions.
The caller of get_recovery_mode_switch() is in vboot_logic.c
that is linked in romstage. So presumably recovery mode
is broken and is not fixed with this commit either.
Change-Id: I2a0fdbb370d54898c72adb29a0e9b990a5fc0ce1
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59003
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Add APU device apc driver and set up permissions.
APU has its own device apc for control access by domains.
For Domain 0, the access to the following slaves are restricted to
security read and write:
apusys_ao-2, apusys_ao-4, apusys_ao-5, apu_sctrl_reviser,
apu_iommu0_r1 apu_iommu0_r2, apu_iommu0_r3, apu_iommu0_r4
apu_iommu1_r1, apu_iommu1_r2, apu_iommu1_r3,apu_iommu1_r4
For VPU, D0/D5 are set as no protection, other domains are forbidden.
For other slaves, the D0 is no protection, other domains are forbidden.
BUG=b:203145462
BRANCH=cherry
TEST=boot cherry, check dump log and test permissions
Signed-off-by: Flora Fu <flora.fu@mediatek.com>
Change-Id: If92d3b02ac4966332315b85d68e0f48c6a9fce85
Reviewed-on: https://review.coreboot.org/c/coreboot/+/58969
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Yu-Ping Wu <yupingso@google.com>
FSP-S is normally memmapped and then decompressed. There are about 7 ms
between starting ramstage, and loading FSP-S. By preloading we can
ensure the fsps.bin is already in RAM by the time we need it. This
reduces boot time by about 7 ms.
BUG=b:
TEST=Boot nipperkin and see ~7ms reduction in boot time
| 10 - start of ramstage | 0.044 | 0.044 Δ( 0.00, 0.00%) |
| 30 - device enumeration | 1.899 | 2.073 Δ( 0.17, 0.01%) |
| 971 - loading FSP-S | 6.645 | 6.628 Δ( -0.02, -0.00%) |
| 15 - starting LZMA decompress (ignore for x86) | 0.016 | 0.01 Δ( -0.01, -0.00%) |
| 16 - finished LZMA decompress (ignore for x86) | 15.266 | 8.316 Δ( -6.95, -0.47%) |
| 954 - calling FspSiliconInit | 0.08 | 0.09 Δ( 0.01, 0.00%) |
CBFS DEBUG: _cbfs_alloc(name='fsps.bin', alloc=0xc9761e5c(0xc97a3f0c), force_ro=false, type=-1)
CBFS: Found 'fsps.bin' @0x1a1fc0 size 0x102cd in mcache @0xc97dd208
waiting for thread
took 1 us <-- fsps.bin was preloaded
CBFS DEBUG: get_preload_rdev(name='fsps.bin') preload successful
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: I5a728047b8ad92d70bba8485017579aa3df48d95
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59025
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Reviewed-by: Rob Barnes <robbarnes@google.com>
This will enable reading FSP-S/M using the SPI DMA controller.
BUG=B:179699789
TEST=Build guybrush with SPI DMA enabled and verify alignment is set
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: I282b9989d8e95c93603c6f69616a8f236a4e2e35
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59130
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
Used H1 firmware where the last version number is 0.0.22, 0.3.22 or
less to production that will need to disable autonomous GPIO power
management and then can get H1 version by gsctool -a -f -M
BUG=b:201054849
TEST=USE="project_primus emerge-brya coreboot" and verify it builds
without error.
Change-Id: If5a99a96e5d4b84be3f2c1165283ce249ca75d58
Signed-off-by: Casper Chang <casper_chang@wistron.corp-partner.google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59079
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
