* add a single config community triplet based on x64-windows
* add a failed port, otherwise mechanism does not support empty list
* add failures to baseline
* remove duplicate entry
* second round of failures added to baseline
* third round of regressions added to CI
* [ade] fix single-config builds
* [ade] update refs
* remove duplicate entries
* [unix2dos] restore eol in script file
* add even more ports now failing in singleConfig...
* [libjpeg-turbo] remove unnecessary check
* [libjpeg-turbo] fix references
* [baseline] restore check for ports that were added recently
* [libmodplug] enable single config triplets
* [tensorflow-common] enable single config triplets
* update baseline
* fix references
* remove wrong check for single config triplets
* fix references
* [ogre] fix single config builds
* [ogre] fix references
* [untested] force x64-windows-release as host triplet when testing x64-windows-release in CI
* fix
* [x264] fix for single config
* [x264] update refs
* [libffi] enable single config builds
* [libffi] update refs
* remove cli broken arguments
* [calceph] enable single-config builds
* [calceph] fix references
* [kf5config] enable single-config builds
* [kf5config] fix references
* [detours] enable single-config builds
* [detours] fix references
* [kf5*] enable single-config builds
* [kf5*] fix references
* fix
* kf5config fix reference
* trying again to force host triplet in CI
* [boost] fix for single-config builds
* [boost] update references
* fix
* [boost] update refs
* disable host triplet in ci because it's not necessary
* retrigger a build for ports previously failing
* remove host triplet in ci, not necessary anymore?
* make PR dry
* [tensorflow-common] bump version
* [tensorflow-common] fix references
* Fix version database for boost-modular-build-helper.
* fix references
* restore CRLF on some ports
* bump versions
* fix references
* bump port versions
* fix references
* boost-modular-build-helper: Reverted due to no changes.
detours, kf5config, kf5holidays, libbacktrace: Reverted due to only whitespace changes.
libffi: bumped port-version
libmodplug: reverted due to no functional changes
ogre, ogre-next: reverted due to incorrect case of "Debug" vs. "Release" already fixed in master.
And rebuilt version database.
Co-authored-by: Billy Robert O'Neal III <bion@microsoft.com>
* Use -llibffi on Windows
Replace -lffi with -llibffi for Windows development, not just mingw
* Update portfile.cmake
* Update baseline.json
* Update libffi.json
* Add port-version and make path quoted
* Update version files
Co-authored-by: NancyLi1013 <lirui09@beyondsoft.com>
* [libffi] Add pkgconfig file
- Configure and add the libffi.pc file.
- Furthermore, enable the check for same 32/64bit-ness in
libffiConfigVersion.cmake again, which had caused issues in the past
with newer versions of glib using meson build, e.g. when building for
x86-windows target on an x64-windows host.
Now, libffi is found by current glib (e.g. 2.64.3) during
meson build using the libffi pkgconfig file.
* Check VCPKG_BUILD_TYPE to handle release/debug
* [libffi] Support arm/arm64
* [libffi] Use VCPKG_TARGET_ARCHITECTURE to determine triplet
* [libffi] Use VCPKG_TARGET_ARCHITECTURE in CMakeLists.txt
* update baseline
* re-trigger ci test
* libffi 3.3-rc0 custom port, with arm/arm64 windows support
* fix linux build
* enable EFI64 in linux x64, should fix build errors in some cases
* libffi: move to the arm64 support pr branch
* libffi: move to latest upstream master
* libffi: update to upstream master
* libffi: update to v3.3-rc2
* libffi: update to v3.3 release
* libffi: update version in fficonfig.h
* libffi: Remove CMAKE_WINDOWS_EXPORT_ALL_SYMBOLS
* [libffi] update portfile.cmake
* [libnice] fix Windows ARM64 build
* CI: set atk:arm64-windows to fail
atk compilation requires running glib(arm64) tools, which is not possible on a x86 system.
* CI: libffi:arm64-windows passing
* [gdk-pixbuf] fix Windows ARM64 build
* [pangomm] show unsupported architecture error on ARM
* [pangomm] update portfile.cmake
* [libffi] Avoid depending on VCPKG_TARGET_TRIPLET; use native cmake variables instead. Avoid double expansion inside if().
* [libffi] TARGET is a keyword for if() -- use quoted expansion instead.
Co-authored-by: Robert Schumacher <roschuma@microsoft.com>
