This brings the ACPI code more in line with both what the new code for
the AMD SoCs will do and also what the current Intel code does. This was
mainly done to have a reduced delta to the new AMD domain resource
handling functions to debug it, but it might still be useful to upstream
this change.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I8cca05976b1c9d4e994e407b8c0197da7dd35eb2
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75487
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Matt DeVillier <matt.devillier@amd.corp-partner.google.com>
Reviewed-by: Martin Roth <martin.roth@amd.corp-partner.google.com>
With the coreboot build process, `UniversalPayloadBuild.sh` calls
`UniversalPayloadBuild.py`. That Python script will unconditionally
build DXE as 64-bit, but accepts an argument for the entry point:
parser.add_argument('-a', '--Arch', choices=['IA32', 'X64'],
help='Specify the ARCH for payload entry module. Default build X64
image.', default ='X64')
Currently, ` -a IA32 -a X64` is passed, and the Python script will
use the `X64` argument, resulting in a payload that won't work with
coreboot.
Remove the `-a X64`, so the resulting build is a 32-bit entry point,
and 64-bit DXE, which works with coreboot.
Signed-off-by: Sean Rhodes <sean@starlabs.systems>
Change-Id: I8a557d6e155a2938b44036d98f9274cc8b38f156
Reviewed-on: https://review.coreboot.org/c/coreboot/+/73668
Reviewed-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-by: Paul Menzel <paulepanter@mailbox.org>
Reviewed-by: Martin L Roth <gaumless@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Benjamin Doron <benjamin.doron00@gmail.com>
Reviewed-by: Felix Singer <service+coreboot-gerrit@felixsinger.de>
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
This function can be called to more easily add a file to CBFS.
Additional file attributes can be added later:
cbfs-files-y += pagetables
pagetables-file := $(objcbfs)/pt
pagetables-type := raw
pagetables-compression := none
pagetables-COREBOOT-position := $(CONFIG_ARCH_X86_64_PGTBL_LOC)
becomes
$(call add-cbfs-file-simple, pagetables, $(objcbfs)/pt, raw, none )
pagetables-COREBOOT-position := $(CONFIG_ARCH_X86_64_PGTBL_LOC)
This is especially useful inside macros where you may want to add
an unknown number of entries.
Signed-off-by: Martin Roth <gaumless@gmail.com>
Change-Id: I72bb2f21fb22f650b7970c7a37a48c10a4af0ed5
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75108
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Singer <service+coreboot-gerrit@felixsinger.de>
Incorrect memory part was used in CB:74745 to generate the DRAM Strap
ID. Amend the memory_parts_used.txt and regenerate the DRAM Strap ID.
BUG=b:272746814
TEST=Generate the DRAM Strap ID.
Change-Id: I0668d7e02345610a11f9113d8bbe99a474f33f1a
Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75456
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Eric Lai <eric_lai@quanta.corp-partner.google.com>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jon Murphy <jpmurphy@google.com>
The acpigen_resource_[bus_number,io,mmio*] functions didn't make it very
clear that they are generating resource producer ranges and not resource
consumer ranges. To clarify this, change the function names to
acpigen_resource_producer_[bus_number,io,mmio*] and explicitly add the
ADDR_SPACE_GENERAL_FLAG_PRODUCER flag which evaluates to 0, so this
doesn't change the functionality.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I334f38aa8ab418d5577f92b980ff750504e2bb4e
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75486
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <inforichland@gmail.com>
cbfs_boot_map_optionrom will generate lower case hex digits for the
filename to look for in CBFS, so make sure that the file name will use
lower case hex digits and no upper case hex digits.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I1d4daa04120de0f2c853a44691b7e2c52eb2af20
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75483
Reviewed-by: Fred Reitberger <reitbergerfred@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Timeless build for QEMU (i440fx/piix4) does not modify the binary.
New patch is add to fix the build in a separate directory from the source.
Change-Id: Ib69437be8ee69ad62fb1dfbbafabc2c4c885b7b2
Signed-off-by: Elyes Haouas <ehaouas@noos.fr>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/73740
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Singer <service+coreboot-gerrit@felixsinger.de>
The IBM SBP1 is an evaluation platform.
It's utilising:
- 4 SPR sockets, having 16 DIMMs each
- 240C/480T at maximum
- 32x CPU PCIe slots
- 2x M.2 PCH PCIe slots
- Dual 200Gbit/s NIC
- SPI TPM
It has an AST2600 BMC for remote management.
It doesn't have:
- External facing USB ports
- Video outputs
- Audio codec
Test:
The board boots to Linux 5.15 with all 480 cores available.
All PCIe devices are working and no errors in ACPI.
All 64 memory DIMMS are working and M.2 devices can be used.
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Change-Id: Ie21c744224e8d9e5232d63b8366d2981c9575d70
Reviewed-on: https://review.coreboot.org/c/coreboot/+/73392
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Lean Sheng Tan <sheng.tan@9elements.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This commit enables the build for IO Margining, ensuring that ASPM is
disabled and certain FSP knobs are adjusted in coreboot as below
1. Enable DFXEnable
2. Disable PcieGlobalAspm
3. Disable KtiLinkL1En & KtiLinkL0pEn
Since the FSP UPD does not provide all the necessary knobs for IO
Margining, the following settings need to be applied during the FSP
build process:
1. Enable PcdBiosDfxKnobEnabled
2. Disable PchDmiAspm
3. Enable SataTestMode
4. Enable WmphyMargining
5. Disable IioErrorEn
TEST=Build for IBM sbp1 board.
Signed-off-by: Naresh Solanki <Naresh.Solanki@9elements.com>
Change-Id: Ie306d12943adb76411d55358548b5cb2eb3a95be
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75415
Reviewed-by: Lean Sheng Tan <sheng.tan@9elements.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The list of outstanding issues on ticket.coreboot.org is formatted as
a Markdown table, which is not supported by Recommonmark. Reformat it
as an embedded reStructuredText table.
Change-Id: Id885e268d55348a365e38c536aed17f974f47840
Signed-off-by: Nicholas Chin <nic.c3.14@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75463
Reviewed-by: Felix Singer <service+coreboot-gerrit@felixsinger.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
This commit adds support for Intel 800 series chipset. The new chipset
can be uniquely identified by its SPI speed, eSPI speed, and
chipset name.
This commit message is clear and concise, and it accurately describes
the changes that were made to the code. It also includes the following
information:
- Specify the correct chipset name.
"PCH Revision: 800 series Meteor Lake"
- Show the valid eSPI/EC frequency.
"Read eSPI/EC Bus Frequency: 20MHz"
Signed-off-by: Subrata Banik <subratabanik@google.com>
Change-Id: I70619d9e3ed2bcad86f84a0527e3a0ad13acd706
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75433
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Lean Sheng Tan <sheng.tan@9elements.com>
Reviewed-by: Kapil Porwal <kapilporwal@google.com>
This patch adds two new chromeos_*.fmd files for release and debug FSP
builds targeting rex_ec_ish.
`rex_ec_ish` variant would pack ISH firmware into the CSE boot partition
hence, the blob size is expected to increase. Creates separate flash map
layout to ensure ISH work is not impacting on the regular `rex0` project
SPI flash usage.
BUG=b:284254353
TEST=Able to build google/rex_ish_ec board and boot on target hardware.
Change-Id: Ife4663d3ccf80a928646eadaac4c9ab49ad29055
Signed-off-by: Subrata Banik <subratabanik@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75471
Reviewed-by: Kapil Porwal <kapilporwal@google.com>
Reviewed-by: Tarun Tuli <taruntuli@google.com>
Reviewed-by: YH Lin <yueherngl@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This patch creates a new variant to support the ISH enablement using
Rex platform.The idea here is to leverage the `rex0` code as much as
possible and add specific support for ISH enablement as per the hardware
schematic differences.
BUG=b:284254353
TEST=Able to build google/rex_ish_ec board and boot on target hardware.
Change-Id: I625fd0b31aed998f4e8f2d139827bc212ee8a90b
Signed-off-by: Subrata Banik <subratabanik@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75470
Reviewed-by: Tarun Tuli <taruntuli@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: YH Lin <yueherngl@google.com>
Reviewed-by: Kapil Porwal <kapilporwal@google.com>
EFS header is mapped during PSP verstage and bootblock to read some SPI
configuration. After use it is left unmapped. Unmap the EFS region after
use.
BUG=b:240664755
TEST=Build and boot to OS in Skyrim with unsigned PSP verstage.
Change-Id: I865f45a3d25bc639eb8435b54aa80895ec4afd27
Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75455
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
cbfs_unmap does not unmap the mapped region from the boot device. This
leads to some resource leaks eg. TLB slots in PSP. Explicitly call
rdev_munmap on the address mapped by cbfs_map.
BUG=b:240664755
TEST=Build and boot to OS in Skyrim with unsigned PSP verstage.
Change-Id: I51b9d066a40103f2ebdf2ef2fc3da13beb467921
Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75454
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
cbfs_unmap does not unmap the mapped region from the boot device. This
leads to some resource leaks eg. TLB slots in PSP. Explicitly call
rdev_munmap on the address mapped by cbfs_map.
BUG=b:240664755
TEST=Build and boot to OS in Skyrim with unsigned PSP verstage.
Change-Id: If1d355972cc743b8d8c451e1b3f827abd15e98fe
Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75453
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
On FMAP without RW slots, PSP verstage fails to build because of
reference to FMAP_SECTION_FW_MAIN_A_*. Instead extract the offset and
size of relevant sections using fmap_locate_area().
BUG=b:240664755
TEST=Build and boot to OS in Skyrim with unsigned PSP verstage.
Change-Id: I29997534c6843b47a36655431f79e5c70bd17f9b
Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75452
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Only the Intel Quark SoC selected this option and that SoC was dropped
in commit 531023285e ("soc/intel/quark: Drop support"), so drop this
Kconfig option too.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Ic4f1c7530cd8ac7a1945b1493a2d53a7904daa06
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75473
Reviewed-by: Felix Singer <service+coreboot-gerrit@felixsinger.de>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The Galileo board was dropped in commit 037c25d4dd ("mb/intel/
galileo: Drop support"), so also drop this config file.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: I8c85dc03c3a4a016d6e13f1bee170d1bc6439470
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75472
Reviewed-by: Felix Singer <service+coreboot-gerrit@felixsinger.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nicholas Chin <nic.c3.14@gmail.com>
The previous regular expression only matches the line starting with
"Signed-off-by:". If the name and mail address are missing, it can not
find out. The following words should be "name <mail@xxx.com>".
Change-Id: I42cc399e79b65928a6aef87c51e5476c7158d166
Signed-off-by: Zheng Bao <fishbaozi@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/73340
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Lean Sheng Tan <sheng.tan@9elements.com>
The SuperIO ACPI name was hard-coded to "SIO0". Allow setting the name
in the device tree so more than one SuperIO can be named.
An upcoming board (purism/librem_l1um_v2) has two SuperIOs - one in the
AST2500 BMC, and a Nuvoton NCT6791D used for hardware monitor, POST
display, etc.
Many boards have references to SIO0 already, so the default name is
still the same for a single SuperIO.
Change-Id: Ibfa6ab7622749e6310ee91530bc3722e8e28d9bb
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75089
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Paul Menzel <paulepanter@mailbox.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Add support of variant_devtree_update() function to override
devtree settings for variant boards. Also, add CPU power limit
values for rex baseboard.
BRANCH=None
BUG=b:270664854
TEST=Built and verified power limit values as below log message
for 15W SKU on Rex board.
Overriding power limits PL1 (mW) (10000, 15000) PL2 (mW)
(57000, 57000) PL4 (W) (114)
Change-Id: If46445157358e3e0f227e26a35b4303fc9189a4b
Signed-off-by: Sumeet Pawnikar <sumeet.r.pawnikar@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/74754
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Eric Lai <eric_lai@quanta.corp-partner.google.com>
Reviewed-by: Harsha B R <harsha.b.r@intel.com>
Reviewed-by: Subrata Banik <subratabanik@google.com>
Add support to update power limit values for variants.
Until now, each SoC implements this themselves.
To avoid code duplication, add this to common code.
BRANCH=None
BUG=b:270664854
TEST=Built and verified power limit values as below log message
for 15W SKU on Rex board.
Overriding power limits PL1 (mW) (10000, 15000) PL2 (mW)
(57000, 57000) PL4 (W) (114)
Change-Id: I414715f211d816bbfad03a673ca96dd5df94caeb
Signed-off-by: Sumeet Pawnikar <sumeet.r.pawnikar@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/74620
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Subrata Banik <subratabanik@google.com>
Reviewed-by: Eric Lai <eric_lai@quanta.corp-partner.google.com>
Hook the SaGvFreq, SaGvGear upds so that mainboard can change the
settings via devicetree. Meteor Lake supports 4 SaGv work points and it
can dynamically scale the work point based on memory bandwidth
utilization. Dynamic gearing technology allows the Memory Controller to
run at 1:1, 1:2 or 1:4 ratio of DRAM speed. The gear ratio is the ratio
of DRAM speed to Memory Controller Clock.
BUG=b:282164577
TEST=Verified the settings on google/rex using debug FSP logs.
Signed-off-by: Bora Guvendik <bora.guvendik@intel.com>
Change-Id: I37169880af4019675374594e90735b5d7d0873b8
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75290
Reviewed-by: Subrata Banik <subratabanik@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Use verbs enabling jack detect if the EC firmware has been updated
with fixed jack detection.
Test: Build Librem 14 and boot with latest EC, test headset jack
detection.
Change-Id: I57a27b1d51e4f6c7c712bcb2823d21692b9c5ce6
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/74364
Reviewed-by: Paul Menzel <paulepanter@mailbox.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Provide system76_ec_cmd() to send arbitrary commands to the EC.
Provide librem_ec_has_jack_detect() to probe for the jack detect fix.
Change-Id: Ic7bda0ce230a3ad68dfeb7b01a0e04f70dab9e5d
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/74390
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Crawford <tcrawford@system76.com>
Add the acpigen_resource_mmio helper function to generate an MMIO range
resource.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Change-Id: I38d55dfcc2892bcb5d253a3aef6ed993cfdba0a5
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75045
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Based on the schematic and vendor ASL code, PCI interrupt lines ABC of
the Ricoh R5C847 PC Card/Media Card/FireWire controller are routed DBC.
From lspci and the schematic this chip is PCI device 1. The original
config copied from the T400 was routed ABCD->BCDA, causing Linux to
issue an "irq 18: nobody cared" message when inserting an SD card.
This is fixed by this patch and the SD card now works properly.
Change-Id: Iede1de72d5369f1aebbac170792733739add3431
Signed-off-by: Nicholas Chin <nic.c3.14@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75411
Reviewed-by: Paul Menzel <paulepanter@mailbox.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
The MAX98363 smart speaker amp can be connected over SoundWire and be
configured for mainboards to use:
- Data Port 0 and Bulk Register Access is not supported
- Data Port 1 is the 32bit data input for the speaker path
The data port and audio mode properties are filled out as best as
possible with the datasheet as a reference.
The ACPI address for the codec is calculated with the information in
the codec driver combined with the devicetree.cb hierarchy where the
link and unique IDs are extracted from the device path.
For example this device is connected to master link ID 2 and has strap
settings configuring it for unique ID 0.
chip drivers/soundwire/max98363
register "desc" = ""Left Speaker Amp""
device generic 2.0 on end
end
This driver was tested with the rex0 reference design by booting
and disassembling the runtime SSDT to ensure that the devices have the
expected address and properties.
Device (SW20)
{
Name (_ADR, 0x000230019F836300) // _ADR: Address
Name (_DDN, "Left Speaker Amp") // _DDN: DOS Device Name
Method (_STA, 0, NotSerialized) // _STA: Status
{
Return (0x0F)
}
Name (_DSD, Package ()
{
ToUUID ("daffd814-6eba-4d8c-8a91-bc9bbf4aa301"),
Package ()
{
Package () { "mipi-sdw-sw-interface-revision", 0x00010000 },
[...]
Package () { "mipi-sdw-source-port-list", Zero },
Package () { "mipi-sdw-sink-port-list", 0x02 }
},
ToUUID ("dbb8e3e6-5886-4ba6-8795-1319f52a966b"),
Package ()
{
Package () { "mipi-sdw-port-audio-mode-0", "MOD0" },
Package () { "mipi-sdw-dp-1-sink-subproperties", "SNK1" }
}
})
Name (MOD0, Package ()
{
ToUUID ("daffd814-6eba-4d8c-8a91-bc9bbf4aa301"),
Package ()
{
Package () { "mipi-sdw-audio-mode-bus-frequency-configs",
Package () { 0x00927C00, ... }
},
Package () { "mipi-sdw-audio-mode-sampling-frequency-configs",
Package () { 0x3E80, ... }
},
[...]
}
})
Name (SNK1, Package ()
{
ToUUID ("daffd814-6eba-4d8c-8a91-bc9bbf4aa301"),
Package ()
{
Package () { "mipi-sdw-data-port-type", Zero },
[...]
},
ToUUID ("dbb8e3e6-5886-4ba6-8795-1319f52a966b"),
Package ()
{
Package () { "mipi-sdw-port-audio-mode-0", "MOD0" }
}
})
}
BUG=b:269497731
TEST=Verified SSDT for SNDW in the OS
Signed-off-by: Kapil Porwal <kapilporwal@google.com>
Change-Id: Ie56109d615759e3e5e32782c8782cb2f47014ec4
Reviewed-on: https://review.coreboot.org/c/coreboot/+/73278
Reviewed-by: Sridhar Siricilla <sridhar.siricilla@intel.com>
Reviewed-by: Subrata Banik <subratabanik@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The PCI config space access via IO ports uses two 32 bit IO ports.
Signed-off-by: Felix Held <felix-coreboot@felixheld.de>
Change-Id: Ie99b4f5fc01fb0405243ff108d813ee1a3d35e5d
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75408
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Now -mno-mmx is statically set in arch/x86 so remove this option.
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Change-Id: I0da7f9f1afb0c8ecae728c45591897ca1d4dfb11
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75318
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Lean Sheng Tan <sheng.tan@9elements.com>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Even though coreboot does not allow floating point operations some
compilers like clang generate code using hw floating point registers,
e.g. SSE %XMMx registers on 64bit code by default. Floating point
operations need to be enabled in hardware for this to work (CR4). Also
in SMM we explicitly need to save and restore floating point registers
for this reason. If we instruct the compiler to not generate code with
FPU ops, this simplifies our code as we can skip that step.
With clang this reduces the binary size a bit. For instance ramstage for
qemu/Q35 drops from 216600 bytes decompressed to 212768.
TEST: See that with x86_64 bit and clang coreboot reaches the payload
without setting the CR4_OSFXSR bit in CR4. Without this change it would
bootloop very early in the bootblock on Qemu Q35.
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Change-Id: Ib8590c55e7aed1ece2aa23b8ea99463396435e11
Reviewed-on: https://review.coreboot.org/c/coreboot/+/75316
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Felix Held