In case PcdFirmwareReleaseDateString is not set use a valid date
as fallback. Using "unknown" makes Windows unhappy.
Fixes: 4cb94f20b0 ("OvmfPkg/SmbiosPlatformDxe: use PcdFirmware*")
Reported-by: ruifeng.gao@intel.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
The hard-coded attributes for the re-added memory space should instead
forward the replaced descriptor's capabilities.
Tested on Linux with efi=debug. Prior to this change, an 8GiB VM running
a kernel without unaccepted memory support shows this entry
efi: mem94: [Conventional| | |CC| | | | | | | | | | | ]
range=[0x0000000100000000-0x000000023fffffff] (5120MB)
This does not have the cache capabilities one would expect for system
memory, UC|WC|WT|WB.
After this change, the same entry becomes
efi: mem94: [Conventional| | |CC| | | | | | | |WB|WT|WC|UC]
range=[0x0000000100000000-0x000000023fffffff] (5120MB)
This has all the expected attributes.
Cc: Erdem Aktas <erdemaktas@google.com>
Cc: James Bottomley <jejb@linux.ibm.com>
Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Min Xu <min.m.xu@intel.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Michael Roth <michael.roth@amd.com>
Signed-off-by: Dionna Glaze <dionnaglaze@google.com>
[ardb: drop the EFI_MEMORY_CPU_CRYPTO flag - it isn't used anywhere else
in EDK2 or Linux so it doesn't actually do anything, and it is
unclear whether it is intended for use by the guest in the first
place]
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Allow users to build OVMF then run QEMU by moving the build block above
the run block and removing the exit line.
Signed-off-by: Rebecca Cran <rebecca@quicinc.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
BZ: https://bugzilla.tianocore.org/show_bug.cgi?id=4243
This patch enables Tdx measurement in OvmfPkgX64 with below changes:
1) CC_MEASUREMENT_ENABLE is introduced in OvmfPkgX64.dsc. This flag
indicates if Intel TDX measurement is enabled in OvmfPkgX64. Its
default value is FALSE.
2) Include TdTcg2Dxe in OvmfPkgX64 so that CC_MEASUREMENT_PROTOCOL
is installed in a Td-guest. TdTcg2Dxe is controlled by
TDX_MEASUREMENT_ENABLE because it is only valid when Intel TDX
measurement is enabled.
3) OvmfTpmLibs.dsc.inc and OvmfTpmSecurityStub.dsc.inc are updated
because DxeTpm2MeasureBootLib.inf and DxeTpmMeasurementLib.inf
should be included to support CC_MEASUREMENT_PROTOCOL.
Cc: Erdem Aktas <erdemaktas@google.com>
Cc: James Bottomley <jejb@linux.ibm.com>
Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Michael Roth <michael.roth@amd.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
Signed-off-by: Min Xu <min.m.xu@intel.com>
BZ: https://bugzilla.tianocore.org/show_bug.cgi?id=4243
MeasureFvImage once was implemented in PeilessStartupLib and it does
measurement and logging for Configuration FV (Cfv) image in one go,
using TpmMeasureAndLogData(). But it doesn't work in SEC.
This patch splits MeasureFvImage into 2 functions and implement them in
SecTdxHelperLib.
- TdxHelperMeasureCfvImage
- TdxHelperBuildGuidHobForTdxMeasurement
TdxHelperMeasureCfvImage measures the Cfv image and stores the hash value
in WorkArea. TdxHelperBuildGuidHobForTdxMeasurement builds GuidHob for the
measurement based on the hash value in WorkArea.
After these 2 functions are introduced, PeilessStartupLib should also be
updated:
- Call these 2 functions instead of the MeasureFvImage
- Delete the duplicated codes in PeilessStartupLib
Cc: Erdem Aktas <erdemaktas@google.com>
Cc: James Bottomley <jejb@linux.ibm.com>
Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Michael Roth <michael.roth@amd.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
Signed-off-by: Min Xu <min.m.xu@intel.com>
BZ: https://bugzilla.tianocore.org/show_bug.cgi?id=4243
MeasureHobList once was implemented in PeilessStartupLib and it does
measurement and logging for TdHob in one go, using TpmMeasureAndLogData().
But it doesn't work in SEC.
This patch splits MeasureHobList into 2 functions and implement them in
SecTdxHelperLib.
- TdxHelperMeasureTdHob
- TdxHelperBuildGuidHobForTdxMeasurement
TdxHelperMeasureTdHob measures the TdHob and stores the hash value in
WorkArea. TdxHelperBuildGuidHobForTdxMeasurement builds GuidHob for the
measurement based on the hash value in WorkArea.
After these 2 functions are introduced, PeilessStartupLib should also be
updated:
- Call these 2 functions instead of the MeasureHobList
- Delete the duplicated codes in PeilessStartupLib
Cc: Erdem Aktas <erdemaktas@google.com>
Cc: James Bottomley <jejb@linux.ibm.com>
Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Michael Roth <michael.roth@amd.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
Signed-off-by: Min Xu <min.m.xu@intel.com>
BZ: https://bugzilla.tianocore.org/show_bug.cgi?id=4243
TdxHelperLib provides below helper functions for a td-guest.
- TdxHelperProcessTdHob
- TdxHelperMeasureTdHob
- TdxHelperMeasureCfvImage
- TdxHelperBuildGuidHobForTdxMeasurement
SecTdxHelperLib is the SEC instance of TdxHelperLib. It implements 4
functions for tdx in SEC phase:
- TdxHelperProcessTdHob consumes TdHob to accept un-accepted memories.
Before the TdHob is consumed, it is first validated.
- TdxHelperMeasureTdHob measure/extend TdHob and store the measurement
value in workarea.
- TdxHelperMeasureCfvImage measure/extend the Configuration FV image and
store the measurement value in workarea.
- TdxHelperBuildGuidHobForTdxMeasurement builds GuidHob for tdx
measurement.
This patch implements the stubs of the functions. The actual
implementations are in the following patches. Because they are moved from
other files.
Cc: Erdem Aktas <erdemaktas@google.com>
Cc: James Bottomley <jejb@linux.ibm.com>
Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Michael Roth <michael.roth@amd.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
Signed-off-by: Min Xu <min.m.xu@intel.com>
BZ: https://bugzilla.tianocore.org/show_bug.cgi?id=4243
From the perspective of security any external input should be measured
and extended to some registers (TPM PCRs or TDX RTMR registers).
There are below 2 external input in a Td guest:
- TdHob
- Configuration FV (CFV)
TdHob contains the resource information passed from VMM, such as
unaccepted memory region. CFV contains the configurations, such as
secure boot variables.
TdHob and CFV should be measured and extended to RTMRs before they're
consumed. TdHob is consumed in the very early stage of boot process.
At that moment the memory service is not ready. Cfv is consumed in
PlatformPei to initialize the EmuVariableNvStore. To make the
implementation simple and clean, these 2 external input are measured
and extended to RTMRs in SEC phase. That is to say the tdx measurement
is only supported in SEC phase.
After the measurement the hash values are stored in WorkArea. Then after
the Hob service is available, these 2 measurement values are retrieved
and GuidHobs for these 2 tdx measurements are generated.
This patch defines the structure of TDX_MEASUREMENTS_DATA in
SEC_TDX_WORK_AREA to store above 2 tdx measurements. It can be extended
to store more tdx measurements if needed in the future.
Cc: Erdem Aktas <erdemaktas@google.com>
Cc: James Bottomley <jejb@linux.ibm.com>
Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Michael Roth <michael.roth@amd.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
Signed-off-by: Min Xu <min.m.xu@intel.com>
https://bugzilla.tianocore.org/show_bug.cgi?id=4245
QEMU provides the following three files for guest to install the ACPI
tables:
- etc/acpi/rsdp
- etc/acpi/tables
- etc/table-loader
"etc/acpi/rsdp" and "etc/acpi/tables" are similar, they are only kept
separate because they have different allocation requirements in SeaBIOS.
Both of these fw_cfg files contain preformatted ACPI payload.
"etc/acpi/rsdp" contains only the RSDP table, while "etc/acpi/tables"
contains all other tables, concatenated. To be noted, the tables in these
two files have been filled in by qemu, but two kinds of fields are
incomplete: pointers to other tables and checksums (which depend on the
pointers).
"/etc/table-loader" is a linker/loader which provides the commands to
"patch" the tables in "etc/acpi/tables" and then install them. "Patch"
means to fill the pointers and compute the checksum.
From the security perspective these 3 files are the raw data downloaded
from qemu. They should be measured and extended before they're consumed.
Cc: Erdem Aktas <erdemaktas@google.com>
Cc: James Bottomley <jejb@linux.ibm.com>
Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Michael Roth <michael.roth@amd.com>
Signed-off-by: Min Xu <min.m.xu@intel.com>
Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Instead of eagerly accepting all memory in PEI, only accept memory under
the 4GB address. This allows a loaded image to use the
MEMORY_ACCEPTANCE_PROTOCOL to disable the accept behavior and indicate
that it can interpret the memory type accordingly.
This classification is safe since ExitBootServices will accept and
reclassify the memory as conventional if the disable protocol is not
used.
Cc: Ard Biescheuvel <ardb@kernel.org>
Cc: "Min M. Xu" <min.m.xu@intel.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: James Bottomley <jejb@linux.ibm.com>
Cc: Tom Lendacky <Thomas.Lendacky@amd.com>
Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Erdem Aktas <erdemaktas@google.com>
Signed-off-by: Dionna Glaze <dionnaglaze@google.com>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
In QEMU v5.1.0, the CPU hotplug register block misbehaves: the negotiation
protocol is (effectively) broken such that it suggests that switching from
the legacy interface to the modern interface works, but in reality the
switch never happens. The symptom has been witnessed when using TCG
acceleration; KVM seems to mask the issue. The issue persists with the
following (latest) stable QEMU releases: v5.2.0, v6.2.0, v7.2.0. Currently
there is no stable release that addresses the problem.
The QEMU bug confuses the Present and Possible counting in function
PlatformMaxCpuCountInitialization(), in
"OvmfPkg/Library/PlatformInitLib/Platform.c". OVMF ends up with Present=0
Possible=1. This in turn further confuses MpInitLib in UefiCpuPkg (hence
firmware-time multiprocessing will be broken). Worse, CPU hot(un)plug with
SMI will be summarily broken in OvmfPkg/CpuHotplugSmm, which (considering
the privilege level of SMM) is not that great.
Detect the issue in PlatformCpuCountBugCheck(), and print an error message
and *hang* if the issue is present.
Users willing to take risks can override the hang with the experimental
QEMU command line option
-fw_cfg name=opt/org.tianocore/X-Cpuhp-Bugcheck-Override,string=yes
(The "-fw_cfg" QEMU option itself is not experimental; its above argument,
as far it concerns the firmware, is experimental.)
The problem was originally reported by Ard [0]. We analyzed it at [1] and
[2]. A QEMU patch was sent at [3]; now merged as commit dab30fbef389
("acpi: cpuhp: fix guest-visible maximum access size to the legacy reg
block", 2023-01-08), to be included in QEMU v8.0.0.
[0] https://bugzilla.tianocore.org/show_bug.cgi?id=4234#c2
[1] https://bugzilla.tianocore.org/show_bug.cgi?id=4234#c3
[2] IO port write width clamping differs between TCG and KVM
http://mid.mail-archive.com/aaedee84-d3ed-a4f9-21e7-d221a28d1683@redhat.comhttps://lists.gnu.org/archive/html/qemu-devel/2023-01/msg00199.html
[3] acpi: cpuhp: fix guest-visible maximum access size to the legacy reg block
http://mid.mail-archive.com/20230104090138.214862-1-lersek@redhat.comhttps://lists.gnu.org/archive/html/qemu-devel/2023-01/msg00278.html
NOTE: PlatformInitLib is used in the following platform DSCs:
OvmfPkg/AmdSev/AmdSevX64.dsc
OvmfPkg/CloudHv/CloudHvX64.dsc
OvmfPkg/IntelTdx/IntelTdxX64.dsc
OvmfPkg/Microvm/MicrovmX64.dsc
OvmfPkg/OvmfPkgIa32.dsc
OvmfPkg/OvmfPkgIa32X64.dsc
OvmfPkg/OvmfPkgX64.dsc
but I can only test this change with the last three platforms, running on
QEMU.
Test results:
TCG QEMU OVMF override result
patched patched
--- ------- ------- -------- --------------------------------------
0 0 0 0 CPU counts OK (KVM masks the QEMU bug)
0 0 1 0 CPU counts OK (KVM masks the QEMU bug)
0 1 0 0 CPU counts OK (QEMU fix, but KVM masks
the QEMU bug anyway)
0 1 1 0 CPU counts OK (QEMU fix, but KVM masks
the QEMU bug anyway)
1 0 0 0 boot with broken CPU counts (original
QEMU bug)
1 0 1 0 broken CPU count caught (boot hangs)
1 0 1 1 broken CPU count caught, bug check
overridden, boot continues
1 1 0 0 CPU counts OK (QEMU fix)
1 1 1 0 CPU counts OK (QEMU fix)
Cc: Ard Biesheuvel <ardb+tianocore@kernel.org>
Cc: Brijesh Singh <brijesh.singh@amd.com>
Cc: Erdem Aktas <erdemaktas@google.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: James Bottomley <jejb@linux.ibm.com>
Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Michael Brown <mcb30@ipxe.org>
Cc: Min Xu <min.m.xu@intel.com>
Cc: Oliver Steffen <osteffen@redhat.com>
Cc: Sebastien Boeuf <sebastien.boeuf@intel.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Bugzilla: https://bugzilla.tianocore.org/show_bug.cgi?id=4250
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Message-Id: <20230119110131.91923-3-lersek@redhat.com>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Hugely-appreciated-by: Michael Brown <mcb30@ipxe.org>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
QEMU for x86 has a nasty CPU hotplug bug of which the ramifications are
difficult to oversee, even though KVM acceleration seems to be
unaffected. This has been addressed in QEMU mainline, and will percolate
through the ecosystem at its usual pace. In the mean time, due to the
potential impact on production workloads, we will be updating OVMF to
abort the boot when it detects a QEMU build that is affected.
Tiancore's platform CI uses QEMU in TCG mode, and is therefore impacted
by this mitigation, unless its QEMU builds are updated. This has been
done for Ubuntu-GCC5, but Windows-VS2019 still uses a QEMU build that is
affected.
Aborting the boot upon detecting the QEMU issue will render all boot
tests carried out on Windows-VS2019 broken unless we implement the
'escape hatch' that enables proceed-at-your-own-risk mode, and permits
the boot to proceed even if the QEMU issue is detected.
So let's enable this for Windows-VS2019, and remove it again once it is
no longer needed.
Cc: Laszlo Ersek <lersek@redhat.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Michael Brown <mcb30@ipxe.org>
Cc: Oliver Steffen <osteffen@redhat.com>
Cc: Michael Kubacki <michael.kubacki@microsoft.com>
Bugzilla: https://bugzilla.tianocore.org/show_bug.cgi?id=4250
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Message-Id: <20230119134302.1524569-1-ardb@kernel.org>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Michael Kubacki <michael.kubacki@microsoft.com>
Acked-by: Jiewen Yao <Jiewen.yao@Intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Commit c673216f53 introduces FFS_CHECK_SECTION_HOOK and add it as the
second input parameter in FfsFindSectionData. This change breaks the build
of ArmVirtPkg. To fix this issue, the new version of FfsFindSectionData
is renamed as FfsFindSectionDataWithHook in this patch. In the following
patch the original FfsFindSectionData will be added back.
FfsFindSectionData is renamed as FfsFindSectionDataWithHook. Accordingly
PeilessStartupLib in OvmfPkg should be updated as well. To prevent the
build from being broken, the changes in OvmfPkg are in this patch.
Signed-off-by: Min Xu <min.m.xu@intel.com>
Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
BZ: https://bugzilla.tianocore.org/show_bug.cgi?id=4152
In current DXE FV there are 100+ drivers. Some of the drivers are not
used in Td guest. (Such as USB support drivers, network related drivers,
etc).
From the security perspective if a driver is not used, we'd should prevent
it from being loaded / started. There are 2 benefits:
1. Reduce the attack surface
2. Improve the boot performance
So we separate DXEFV into 2 FVs: DXEFV and NCCFV. All the drivers which
are not needed by a Confidential Computing guest are moved from DXEFV
to NCCFV.
The following patch will find NCCFV for non-cc guest and build FVHob
so that NCCFV drivers can be loaded / started in DXE phase.
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Erdem Aktas <erdemaktas@google.com>
Cc: James Bottomley <jejb@linux.ibm.com>
Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Min Xu <min.m.xu@intel.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
Run the Linux jobs of the OvmfPkg platform CI inside a container,
in the same way the general CI does now. Make use of the default image
specified in the defaults.yml template.
Do not run apt-get in CI jobs to install qemu and gcc dependencies.
Assume the container image provides these.
Use Python from the container image, do not download at runtime.
Signed-off-by: Oliver Steffen <osteffen@redhat.com>
Acked-by: Jiewen Yao <jiewen.yao@intel.com>
Reviewed-by: Michael Kubacki <michael.kubacki@microsoft.com>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Chris Fernald <chfernal@microsoft.com>
