This fixes the bug when VidInitialize asks for different translated addresses
and VgaIsPresent always returns FALSE either because the machine does not have
VGA controller at all (e.g. 86Duino Zero), or it does have video hardware
that is not compatible with VGA standard (e.g. Original Xbox).
The problem investigated by Stanislav Motylkov. Suggested fix by Thomas Faber.
CORE-14625 CORE-16216 CORE-16222
Co-authored-by: Thomas Faber <thomas.faber@reactos.org>
- Use a separate function for detecting blacklisted PCI slots
- Blacklist PCI-to-PCI bridge to avoid stack overflow on real hardware Microsoft Xbox 1.3
CORE-16319 CORE-16216
- Create xbox.cmake based on legacy.cmake and pic.cmake
- Revive Xbox HAL files and add them to the build
- Fix boot with Xbox HAL, add missing parts from halinit_up.c
- Code style and debug logging improvements in pcibus.c
- Allow to choose Xbox HAL in usetup
CORE-16216
There is no need to compile our DLLs as shared libraries since we are
managing symbols exports and imports through spec files.
On my system, this reduces the configure-time by a factor of two.
* Add an NDK header to define INIT_FUNCTION/INIT_SECTION globally
* Use _declspec(allocate(x)) and _declspec(code_seg(x)) on MSVC versions that support it
* Use INIT_FUNCTION on functions only and INIT_SECTION on data only (required by MSVC)
* Place INIT_FUNCTION before the return type (required by MSVC)
* Make sure declarations and implementations share the same modifiers (required by MSVC)
* Add a global linker option to suppress warnings about defined but unused INIT section
* Merge INIT section into .text in freeldr
Extracted from the pci.ids database at https://pci-ids.ucw.cz/ from 2018-11-21 03:15:02 .
Maintained by Albert Pool, Martin Mares, Michal Vaner, and other volunteers from the PCI ID Project.
"warning: cast between incompatible calling conventions 'cdecl' and 'fastcall'; calls through this pointer may abort at runtime [-Wcast-calling-convention]"
CORE-14306
Because we can encounter pending interrupts repeatedly,
HalpEndSoftwareInterrupt and HalEndSystemInterrupt already clean up the stack
space for their arguments (done for CORE-11123/CORE-14076).
However these functions are called from C functions such as KiInterruptDispatch
and HalpDispatchInterrupt2[ndEntry]. These callers also use up stack space,
and it is unknown how much.
To fix this, we simply reset the stack pointer to the location of the trap
frame, which is where it points during a first-level dispatch. This cleans
up the stack usage of any callers higher up, and is okay because a return
will happen through the trap frame anyway.
Dedicated to Pierre.
