Preserve existing UI and add the note about pressing Space to go to
the systemd-boot menu for Pop!_OS Recovery.
Signed-off-by: Tim Crawford <tcrawford@system76.com>
In coreboot, we mark GPU prefmem above 4GB, because NVIDIA wants a lot
(16GB region on the 30 series), otherwise coreboot will obviously fail
to allocate the resources.
In EDK2, we then end up hitting this assert:
InitRootBridge: populated root bus 0, with room for 36 subordinate bus(es)
RootBridge: PciRoot(0x0)
Support/Attr: 7001F / 7001F
DmaAbove4G: No
NoExtConfSpace: No
AllocAttr: 0 ()
Bus: 0 - 24 Translation=0
Io: 1000 - EFFF Translation=0
Mem: 80400000 - 1001FFFFFF Translation=0
MemAbove4G: FFFFFFFFFFFFFFFF - 0 Translation=0
PMem: FFFFFFFFFFFFFFFF - 0 Translation=0
PMemAbove4G: FFFFFFFFFFFFFFFF - 0 Translation=0
ASSERT [PciHostBridgeDxe] .../edk2/MdeModulePkg/Bus/Pci/PciHostBridgeDxe/PciRootBridgeIo.c(120): Bridge->Mem.Limit < 0x0000000100000000ULL
So, bring back Pci*NoEnumerationDxe from the deleted DuetPkg, which
doesn't check anything and let's us boot.
Signed-off-by: Tim Crawford <tcrawford@system76.com>
Register an event to trigger when a block device is added or removed
that will update the list of boot options. Use a refreshguid to force
the form to display the changes if it already open.
Make use of the SMFI command interface to forward logs from edk2 to
System76 EC.
Signed-off-by: Jeremy Soller <jeremy@system76.com>
Signed-off-by: Tim Crawford <tcrawford@system76.com>
The keyword is not defined and will end as public variable beeing declared
in every source that includes the header.
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Make sure that:
* FaultTolerantDxe is started before VariableRuntimeDxe
This ensures that FailedTolerantPei is not required and faults when writing
the variable store are discovered
* SMBUSConfigLoader is started right after VariableRuntimeDxe
This ensured that the board specific variables are updated as soon as possible
* Start SmbiosDxe and BlSupportDxeSmbios early to install board specific NULL protocols
* Start SecureBootConfigDxe right after VariableRuntimeDxe to update SecureBootKeys before
Tcg2Dxe starts
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Read the coreboot table containing the TPM PPI handoff buffer and
place it in gEfiTcgPhysicalPresenceInfoHob.
coreboot uses the same PPI interface as QEMU does and installs the
corresponding ACPI code to provide a full PPI interface to the OS.
The OS must reboot in order to execute the requests.
The corresponding coreboot patch can be found here:
https://review.coreboot.org/c/coreboot/+/45568
In a follow up commit the OvmfPkg PhysicalPresence library will be used
to confirm TPM PPI request. This is necessary as coreboot doesn't have
input drivers or a graphical UI that could be used.
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Instead of waiting and infinite time, abort the TPM PPI request after
3 minutes. This allow to boot headless platforms where no keyboard is
attached or the user didn't made it in time.
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Put the PPI configuration retriveal into an own library.
That will allow to reuse the code in the UefipayloadPkg, where the
firmware provides the ACPI tables, like QEMU does on OvmfPkg.
However one major difference is that the PPI interface in UefiPayloadPkg
is not backed by a MMIO device, but resides in DRAM and is shared with ACPI code.
Add an additional parameter to provide the location of the PPI and
test if the memory region has the correct attributes.
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
* Add support for TPM1.2 and TPM2.
This adds measured boot support and will be extended with Secureboot.
Signed-off-by: Christian Walter <christian.walter@9elements.com>
In BlSupportDxe read the AcpiBoardInfo and set PcdTpmInstanceGuid
to indicate that a TPM is likely present as TPM tables had been installed
by the bootloader.
The Tcg*Dxes will probe for the TPM, so no need to do it here as well.
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Probe for ACPI tables
* TPM2
* TCPA
and store the result in AcpiBoardInfo.
Will be used to determine if a TPM1.2 or TPM2.0 is present.
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Install the gPciPlatformProtocol to scan for Option ROMs.
For every device we probe the Option ROM and provide a pointer
to the activated BAR if found.
It's safe to assume that all ROM bars have been enumerated,
reserved in the bridge resources and are disabled by default.
This is made a mandatory bootloader requirement in the next commit.
Enabling them and leaving them enabled will do no harm.
This can easily be tested on qemu, where it will start finding Option ROMs
for VGA and network cards.
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Signed-off-by: Marcello Sylvester Bauer <marcello.bauer@9elements.com>
Cc: Patrick Rudolph <patrick.rudolph@9elements.com>
Cc: Christian Walter <christian.walter@9elements.com>
Cc: Maurice Ma <maurice.ma@intel.com>
Cc: Nate DeSimone <nathaniel.l.desimone@intel.com>
Cc: Star Zeng <star.zeng@intel.com>
Recent model Chromebooks only return ACK, but not
BAT_SUCCESS, which causes hanging and failed ps2k init.
To mitigate this, make the absence of BAT_SUCCESS reply
non-fatal, and reduce the no-reply timeout from 4s to 1s.
Tested on google/dracia and purism/librem_14
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Don't set PcdDebugPropertyMask for release builds, and properly set it
for debug builds based on SOURCE_DEBUG_ENABLE.
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Previous 1s timeout causeed stalls on boot splash with no benefit.
Reduced to 100ms and no interruptions to boot when no SD card inserted.
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
array.fromstring and array.tostring deprecated, and alias for
array.frombytes and array.tobytes. Deprecated since version 3.2,
have been removed in version python 3.9.
Cc: Bob Feng <bob.c.feng@intel.com>
Cc: Liming Gao <gaoliming@byosoft.com.cn>
Signed-off-by: Yunhua Feng <fengyunhua@byosoft.com.cn>
Startup script does nothing other than confuse users.
Show a welcome banner and tell users how to exit.
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Make sure that:
* FaultTolerantDxe is started before VariableRuntimeDxe
This ensures that FailedTolerantPei is not required and faults when writing
the variable store are discovered
* Start BlSupportDxeSmbios early to install board specific NULL protocols
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
This adds support for FVB in order to support a platform independent
and non-volatile variable store on UefiPayloadPkg. It is required for
non-volatile variable support, TPM support, Secureboot support and more.
Since commit bc744f5893fc4d53275ed26dd8d968011c6a09c1 coreboot supports
the SMMSTORE v2 feature. It implements a SMI handler that is able to
write, read and erase pages in the boot media (SPI flash).
The communication is done using a fixed communication buffer that is
allocated in CBMEM. The existence of this optional feature is advertised
by a coreboot table.
When the SMMSTORE feature is not available the variable emulation is used
by setting PcdEmuVariableNvModeEnable to TRUE.
Add a library for SMMStore to be used in DXE.
The DXE component provides runtime services and takes care of virtual to
physical mapping the communication buffers between SMM and OS.
Make use of the APRIORI DXE to initialize an empty store on the first boot
and set the PCDs to sane values before the variable driver is loaded.
Tests on Intel(R) Xeon(R) E-2288G CPU @ 3.70G showed that the SMI isn't
triggered with a probability of 1:40 of all cases when called in a tight
loop. The CPU continues running and the SMI is triggeres asynchronously
a few clock cycles later. coreboot only handels synchronous APM request
and does nothing on asynchronous APM triggers.
As there's no livesign from SMM it's impossible to tell if the handler
has run. Just wait a bit and try again to trigger a synchronous SMI.
Tests confirmed that out of 5 million tries the SMI is now always handled.
Tested on Linux and Windows 10 on real hardware.
Currently this cannot be tested on coreboot and qemu as it doesn't support
the SMMSTORE on qemu.
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
This fixes an issue where the framebuffer provided by coreboot or
slimbootloader will only work on the primary VGA device. If the
framebuffer corresponds to a different device the screen will stay black.
In addition, the code doesn't work for multiple graphic cards, has
reference to non existing functions, and is a duplication of common code.
Call EfiBootManagerConnectVideoController on every display device found,
not only the legacy VGA device. This is the same as OvmfPkg does.
Allows to display output on the framebuffer set up by firmware, which might
not be the VGA device.
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Uses the RDRAND instruction if available and install EfiRngProtocol.
The protocol may be used by iPXE or the Linux kernel to gather entropy.
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
