This commit oves StackCheckLib from a NULL lib to an instance of
StackCheckLib. This requires every entry point to add a library
dependency on StackCheckLib. It also requires every SEC module
to have a dependency on StackCheckLib because there is no
standard SEC entry point.
It allows for greater flexibility for a platform to apply stack
cookies and simplifies DSC logic.
Continuous-integration-options: PatchCheck.ignore-multi-package
Signed-off-by: Oliver Smith-Denny <osde@microsoft.com>
The magic page is allocated from a fixed address specified
by PcdPeiServicesTablePage. This allocation has been observed
to sometimes fail. Move the allocation of this buffer to the
very beginning of main() to minimize the change that another
allocation is allocated from the PcdPeiServicesTablePage
address.
Signed-off-by: Michael D Kinney <michael.d.kinney@intel.com>
IA32 EmulatorPkg had many linker warnings because with the current
set of linker flags, the MSVC linker was expecting the __stdcall
calling convention on all entry points. This was an effect of
having /SUBSYSTEM:CONSOLE on all binaries built in EmulatorPkg; this
is only needed on WinHost, as that is what Windows launches. The linker
options are adjusted to only set /SUBSYSTEM:CONSOLE on WinHost.
Signed-off-by: Oliver Smith-Denny <osde@microsoft.com>
The Visual Studio Windows debugger will only load symbols for PE/COFF images
that Windows is aware of. Therefore, to enable source level debugging, all
PEI/DXE modules must be loaded via LoadLibrary() or LoadLibraryEx() and the
the instance in memory created by LoadLibrary() must be the one that is
actually executed.
The current source level debug implementation in EmulatorPkg for Windows is
inherited from the old Nt32Pkg. This implementation makes the assumption that
all PEI/DXE modules have a DLL export tables with a symbol named
InitializeDriver. Therefore, this source level debug implementation requires
all modules to be linked in a non-PI spec defined manner. Support for adding
the InitializeDriver symbol was removed in EmulatorPkg, which broke source
level debugging.
To fix this, the source level debugging implementation has been modified to
use the PE/COFF entry point directly. This brings the implementation into
compliance with the PI spec and should work with any PEIM/DXE driver.
Implementing this requires parsing the in-memory instance of the PE/COFF image
created by Windows to find the entrypoint and since PEIMs/DXE drivers are not
garunteed to have 4KB aligned sections, it also requires explicit configuration
of the page table using VirtualProtect().
With this fix, the debugging experience is now so good it is unprecedented!
In Visual Studio Code, add the following to launch.json:
{
"version": "0.2.0",
"configurations": [
{
"name": "EmulatorPkg Launch",
"type": "cppvsdbg",
"request": "launch",
"program": "${workspaceFolder}/<path_to_build>/Build/EmulatorX64/DEBUG_<tool_chain>/X64/WinHost",
"args": [],
"stopAtEntry": false,
"cwd": "${workspaceFolder}/<path_to_build>/Build/EmulatorX64/DEBUG_<tool_chain>/X64/",
"environment": [],
"console": false,
}
]
}
Make modifications to the above template as nessesary and build EmulatorPkg.
Now, just add breakpoints directly in Visual Studio Code the way you would with
any other software project. When you start the debugger, it will halt at the
breakpoint automatically without any extra configuration required.
Reviewed-by: Michael D Kinney <michael.d.kinney@intel.com>
Cc: Andrew Fish <afish@apple.com>
Cc: Ray Ni <ray.ni@intel.com>
Cc: Chasel Chiu <chasel.chiu@intel.com>
Signed-off-by: Nate DeSimone <nathaniel.l.desimone@intel.com>
Update SecPrint() to use _vsnprintf_s() instead of
_vsnprintf() that is a safe function and allows the
defines _CRT_SECURE_NO_WARNINGS and _CRT_SECURE_NO_DEPRECATE
to be removed from WinHost builds.
Signed-off-by: Michael D Kinney <michael.d.kinney@intel.com>
WinNtBlockIoWriteBlocks can spuriously fail on X64. This occurs because
&BytesWritten is a `UINTN*` (i.e. `UINT64*`) but is cast to `LPDWORD`
(i.e. `UINT32*`). Only the low 32 bits are initialized by WriteFile, so
the high 32 bits are uninitialized. This means we will spuriously fail
the `BytesWritten != BufferSize` test.
This doesn't occur on X86-32 since UINTN is the same as DWORD in that
case.
Fix is to declare BytesWritten as DWORD to match the type expected by
WriteFile. This also makes the cast unnecessary.
Signed-off-by: Doug Cook <idigdoug@gmail.com>
The Python version used for build and CI should always be at least
the minimum version supported by edk2-pytool-extensions. A badge
is added that keeps this information dynamically up-to-date based
on the minimum version specified in edk2-pytool-extensions
pyproject.toml file.
Signed-off-by: Michael Kubacki <michael.kubacki@microsoft.com>
This PR updates the CI pipelines to use VS2022 instead of VS2019
as that is the latest supported VS toolchain on edk2.
Continuous-integration-options: PatchCheck.ignore-multi-package
Signed-off-by: Oliver Smith-Denny <osde@microsoft.com>
BlockIo2 Read/Write/Flush APIs should signal the token's event when the
I/O operation completes, but the Emulator APIs do not. As a result, any
code that tries to implement async I/O will hang on emulator.
Both Windows and Unix emulator hosts work the same way:
- All I/O is completed synchronously.
- All I/O implementations contain the comment:
`// Caller is responsible for signaling EFI Event`
However, the protocol implementations do not signal the event, so the
event is never signalled.
Fix is to signal the event in the appropriate protocol implementations.
- If the host API returns success then the I/O is complete since it's
always synchronous.
- If there is a Token and Token->Event is not null and the I/O is
successful then the event should be signalled.
Signed-off-by: Doug Cook <idigdoug@gmail.com>
Then we can use correct TimerLib in another code,
such as dpDynamicCommand(PerformanceLib).
These API are from profileapi.h header and can refer to the link:
https://learn.microsoft.com/en-us/windows/win32/api/profileapi/
Signed-off-by: Yang Gang <yanggang@byosoft.com.cn>
Now that the ResetVectors are USER_DEFINED modules, they will not
be linked against StackCheckLibNull, which were the only modules
causing issues. So, we can now remove the kludge we had before
and the requirement for every DSC to include StackCheckLibNull
for SEC modules and just apply StackCheckLibNull globally.
This also changes every DSC to drop the SEC definition of
StackCheckLibNull.
Continuous-integration-options: PatchCheck.ignore-multi-package
Signed-off-by: Oliver Smith-Denny <osde@linux.microsoft.com>
When updating MdePkg's CR macro to enforce signature checking in
all usages, it was discovered that EmulatorPkg was initializing
a structure without setting the signature for it, causing an error
to be returned when CR now checked the signature.
This commit updates the graphics stack in EmulatorPkg to set the
signature of the data structure and check the return value of
the wrapper for the CR macro.
Signed-off-by: Oliver Smith-Denny <osde@linux.microsoft.com>
Add the necessary toolchain override flags for ms_abi and LTO on X64 for
the unversioned GCC toolchain profile.
This resolves a runtime segmentation fault.
Signed-off-by: Leif Lindholm <quic_llindhol@quicinc.com>
Adds an entry to the package's CI configuration file that enable policy
5 for stuart_pr_eval. With this Policy, all INFs used by the package are
extracted from the provided DSC file and compared against the list of
changed *.inf (INF) files in the PR. If there is a match, stuart_pr_eval
will specify that this package is affected by the PR and needs to be
tested.
Signed-off-by: Joey Vagedes <joey.vagedes@gmail.com>
Currently EDK2 is supporting VS2022, with VS2022 EmulatorPkg build is
failing, this patch is to add the VS2022 support for WinHost
Signed-off-by: Ashraf Ali <ashraf.ali.s@intel.com>
This patch adds Hash2DxeCrypto to EmulatorPkg. The Hash2DxeCrypto is
used to provide the hashing protocol services.
Cc: Andrew Fish <afish@apple.com>
Cc: Ray Ni <ray.ni@intel.com>
Signed-off-by: Doug Flick [MSFT] <doug.edk2@gmail.com>
Reviewed-by: Ray Ni <ray.ni@Intel.com>
This patch adds RngDxe to EmulatorPkg. The RngDxe is used to provide
random number generation services to the UEFI firmware.
Cc: Andrew Fish <afish@apple.com>
Cc: Ray Ni <ray.ni@intel.com>
Signed-off-by: Doug Flick [MSFT] <doug.edk2@gmail.com>
Reviewed-by: Ray Ni <ray.ni@intel.com>
Refer to Uefi spec 2.10 section 12.3.3, Add a new retval
EFI_UNSUPPORTED to EFI_SIMPLE_TEXT_INPUT_EX_PROTOCOL.ReadKeyStrokeEx
and EFI_SIMPLE_TEXT_INPUT_PROTOCOL.ReadKeyStroke().
Cc: Andrew Fish <afish@apple.com>
Cc: Ray Ni <ray.ni@intel.com>
Cc: Michael D Kinney <michael.d.kinney@intel.com>
Cc: Liming Gao <gaoliming@byosoft.com.cn>
Signed-off-by: Qingyu <qingyu.shang@intel.com>
Reviewed-by: Liming Gao <gaoliming@byosoft.com.cn>
Rely on AutoGen for declaring ProcessLibraryConstructorList().
Build-tested with:
build -a X64 -b DEBUG -m EmulatorPkg/Sec/Sec.inf \
-p EmulatorPkg/EmulatorPkg.dsc -t GCC5
Cc: Andrew Fish <afish@apple.com>
Cc: Ray Ni <ray.ni@intel.com>
Ref: https://bugzilla.tianocore.org/show_bug.cgi?id=990
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Message-Id: <20240305113843.68812-8-lersek@redhat.com>
Reviewed-by: Michael D Kinney <michael.d.kinney@intel.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
There is an inconsistency between the UNIX and Windows
implementations of EMU_THUNK_PROTOCOL.SetTime(). The Windows
version returns an EFI_STATUS value whereas the the UNIX
implementation is VOID. However, the UNIX implementation is an
unimplemented stub whereas the Windows version is implementated.
The current EMU_THUNK_PROTOCOL function pointer definition
specifies a VOID return type. However, EMU_THUNK_PROTOCOL.SetTime()
is close to the spec defined gRT->SetTime() except for missing the
EFI_STATUS return type.
Therefore, I conclude that the most sensible reconciliation is to
add the EFI_STATUS return type to the protocol definition.
Cc: Andrew Fish <afish@apple.com>
Reviewed-by: Ray Ni <ray.ni@intel.com>
Cc: Michael D Kinney <michael.d.kinney@intel.com>
Cc: Chasel Chiu <chasel.chiu@intel.com>
Signed-off-by: Nate DeSimone <nathaniel.l.desimone@intel.com>
In the early 2000s as part of the x64 transition the definition
for LPTIMECALLBACK changed from (UINT, UINT, DWORD, DWORD, DWORD)
to (UINT, UINT, DWORD_PTR, DWORD_PTR, DWORD_PTR). However,
the MMTimerThread() function was never updated to the new
signature. Since the implementation does not use the last three
parameters, this issue has not been caught until now.
Cc: Andrew Fish <afish@apple.com>
Reviewed-by: Ray Ni <ray.ni@intel.com>
Cc: Michael D Kinney <michael.d.kinney@intel.com>
Signed-off-by: Nate DeSimone <nathaniel.l.desimone@intel.com>
File description has not been updated since Nt32Pkg was merged
with EmulatorPkg, and several details were no longer technically accurate.
Cc: Andrew Fish <afish@apple.com>
Reviewed-by: Ray Ni <ray.ni@intel.com>
Cc: Michael D Kinney <michael.d.kinney@intel.com>
Cc: Chasel Chiu <chasel.chiu@intel.com>
Signed-off-by: Nate DeSimone <nathaniel.l.desimone@intel.com>
Cc: Andrew Fish <afish@apple.com>
Cc: Ray Ni <ray.ni@intel.com>
Signed-off-by: Michael Kubacki <michael.kubacki@microsoft.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Ray Ni <ray.ni@intel.com>
Add an instance of ImagePropertiesRecordLib which will be used by the
DXE Core.
Cc: Andrew Fish <afish@apple.com>
Cc: Ray Ni <ray.ni@intel.com>
Signed-off-by: Taylor Beebe <taylor.d.beebe@gmail.com>
Reviewed-by: Michael D Kinney <michael.d.kinney@intel.com>
After running EmulatorPkg, one will notice that their terminal acts
strangely. This is caused by the EmulatorPkg Host changing the terminal
mode and not restoring the original mode, which is now fixed.
Reviewed-by: Michael D Kinney <michael.d.kinney@intel.com>
Cc: Andrew Fish <afish@apple.com>
Cc: Ray Ni <ray.ni@intel.com>
Cc: Chasel Chiu <chasel.chiu@intel.com>
Signed-off-by: Nate DeSimone <nathaniel.l.desimone@intel.com>
BZ: https://bugzilla.tianocore.org/show_bug.cgi?id=4504
The BaseRngLibTimerLib allows to generate number based on a timer.
This mechanism allows to have a basic non-secure implementation
for non-production platforms.
To bind and identify Random Number Generators implementations with
a GUID, an unsafe GUID should be added. This GUID cannot be added
to the MdePkg unless it is also added to a specification.
To keep the MdePkg self-contained, copy the BaseRngLibTimerLib to
the MdeModulePkg. This will allow to define an unsafe Rng GUID
in a later patch in the MdeModulePkg.
The MdePkg implementation will be removed later. This allows to give
some time to platform owners to switch to the MdeModulePkg
implementation.
Signed-off-by: Pierre Gondois <pierre.gondois@arm.com>
Reviewed-by: Sami Mujawar <sami.mujawar@arm.com>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Tested-by: Kun Qin <kun.qin@microsoft.com>
