[gmp] Switch sources on windows away from SMP fork. #23466
[gmp] Switch sources on windows away from SMP fork. #23466dan-shaw merged 18 commits intomicrosoft:masterfrom
Conversation
There was a problem hiding this comment.
You have modified or added at least one vcpkg.json where a "license" field is missing.
Details
If you feel able to do so, please consider adding a "license" field to the following files:
ports/gmp/vcpkg.json
Valid values for the license field can be found in the documentation
There was a problem hiding this comment.
You have modified or added at least one vcpkg.json where a "license" field is missing.
Details
If you feel able to do so, please consider adding a "license" field to the following files:
ports/gmp/vcpkg.json
Valid values for the license field can be found in the documentation
ras0219-msft
left a comment
There was a problem hiding this comment.
This is an awesome improvement.
|
@Neumann-A There are regression error on CI pipeline: |
|
There was a problem hiding this comment.
This is a new experimental fast check for PR issues. Please let us know if this bot is helpful!
After committing all other changes, the version database must be updated
git add -u && git commit
git checkout 5a8a1dfff8e2dffd4c8cc2d1fc0da379ffd18b25 -- versions
./vcpkg x-add-version --allDiff
diff --git a/versions/g-/gmp.json b/versions/g-/gmp.json
index f6ad9ad..6e7b3ca 100644
--- a/versions/g-/gmp.json
+++ b/versions/g-/gmp.json
@@ -1,7 +1,7 @@
{
"versions": [
{
- "git-tree": "66917e3a6421c994cf676763cd48c30f8117c8ec",
+ "git-tree": "d252d4fbdebdfeff6c7d17087e9f66cdc3a0c07e",
"version": "6.2.1",
"port-version": 9
},You have modified or added at least one vcpkg.json where a "license" field is missing.
Details
If you feel able to do so, please consider adding a "license" field to the following files:
ports/gmp/vcpkg.json
Valid values for the license field can be found in the documentation
There was a problem hiding this comment.
You have modified or added at least one vcpkg.json where a "license" field is missing.
Details
If you feel able to do so, please consider adding a "license" field to the following files:
ports/gmp/vcpkg.jsonports/mpc/vcpkg.jsonports/nettle/vcpkg.json
Valid values for the license field can be found in the documentation
There was a problem hiding this comment.
You have modified or added at least one vcpkg.json where a "license" field is missing.
Details
If you feel able to do so, please consider adding a "license" field to the following files:
ports/gmp/vcpkg.jsonports/mpc/vcpkg.jsonports/nettle/vcpkg.json
Valid values for the license field can be found in the documentation
ras0219-msft
left a comment
There was a problem hiding this comment.
LGTM once the manual-tools directory is renamed to tools. If the concern is that users might be confused, I'm fine with adding more subdirectories underneath tools/${PORT}/... to clarify
| --enable-cxx | ||
| ) | ||
| if(VCPKG_CROSSCOMPILING) | ||
| list(TRANSFORM tool_names PREPEND "${CURRENT_HOST_INSTALLED_DIR}/manual-tools/${PORT}/") |
There was a problem hiding this comment.
| list(TRANSFORM tool_names PREPEND "${CURRENT_HOST_INSTALLED_DIR}/manual-tools/${PORT}/") | |
| list(TRANSFORM tool_names PREPEND "${CURRENT_HOST_INSTALLED_DIR}/tools/${PORT}/") |
I don't think this is special enough to need a new top-level directory "manual-tools"
There was a problem hiding this comment.
Manual-tools already exists in e.g. the vcpkg-tool-python2 port and it is intended that those tools dont get automatically added to CMAKE_PROGRAM_PATH.
* [Most recent nightly build failed](https://dev.azure.com/vcpkg/public/_build/results?buildId=69427) * [Validation of this tool update failed](https://dev.azure.com/vcpkg/public/_build/results?buildId=69417) ## Common to both: PASSING, REMOVE FROM FAIL LIST: chartdir:x64-windows (.\scripts\ci.baseline.txt) PASSING, REMOVE FROM FAIL LIST: chartdir:x64-windows-static-md (.\scripts\ci.baseline.txt) PASSING, REMOVE FROM FAIL LIST: chartdir:x86-windows (.\scripts\ci.baseline.txt) Probably fixed by microsoft#23701 PASSING, REMOVE FROM FAIL LIST: gmp:x64-uwp (.\scripts\ci.baseline.txt) PASSING, REMOVE FROM FAIL LIST: gmp:x64-windows-static-md (.\scripts\ci.baseline.txt) Probably fixed by microsoft#23466 ? REGRESSION: colmap:x64-windows-static-md failed with BUILD_FAILED. If expected, add colmap:x64-windows-static-md=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. I don't know exactly what changed. I observe that * this thing depends on a *lot* of stuff * on March 14 we didn't even attempt to build this * the x64-windows ones are already in the baseline so I skipped it. REGRESSION: qtdeclarative:x64-windows. If expected, add qtdeclarative:x64-windows=fail to .\scripts\ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\1\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. This is a reporting change: The new world order also includes host build failures which is why it's duplicated. See also microsoft#23714 See also microsoft#23490 I'm nervous about baslining this because it seems most of the qt world is built on top of this port I filed microsoft#23824 about this and @Neumann-A indicated this should be fixed by microsoft#23755 REGRESSION: nettle:x64-uwp. If expected, add nettle:x64-uwp=fail to .\scripts\ci.baseline.txt. REGRESSION: nettle:x64-windows-static-md. If expected, add nettle:x64-windows-static-md=fail to .\scripts\ci.baseline.txt. REGRESSION: nettle:x64-uwp failed with BUILD_FAILED. If expected, add nettle:x64-uwp=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: nettle:x64-windows-static-md failed with POST_BUILD_CHECKS_FAILED. If expected, add nettle:x64-windows-static-md=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. Didn't analyze, probably fixed by microsoft#23519 ? REGRESSION: libgpg-error:x64-uwp. If expected, add libgpg-error:x64-uwp=fail to .\scripts\ci.baseline.txt. REGRESSION: libgpg-error:x64-uwp failed with BUILD_FAILED. If expected, add libgpg-error:x64-uwp=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. This was broken by VS2022 update: ``` C:\Program Files\Microsoft Visual Studio\2022\Enterprise\MSBuild\Microsoft\VisualStudio\v17.0\AppxPackage\Microsoft.AppXPackage.Targets(892,25): error MSB4086: A numeric comparison was attempted on "$(TargetPlatformMinVersion)" that evaluates to "" instead of a number, in condition "'$(TargetPlatformMinVersion)' >= '10.0.17200.0'". [C:\Dev\vcpkg\buildtrees\libgpg-error\x64-uwp-rel\error-1.42-2324ddbc71.clean\SMP\libgpg-error_winrt.vcxproj] ``` REGRESSION: libmikmod:x64-osx. If expected, add libmikmod:x64-osx=fail to .\scripts\ci.baseline.txt. REGRESSION: libmikmod:x64-osx failed with BUILD_FAILED. If expected, add libmikmod:x64-osx=fail to /Users/vagrant/Data/work/2/s/scripts/azure-pipelines/../ci.baseline.txt. Broken between [2022-03-16](https://dev.azure.com/vcpkg/public/_build/results?buildId=68947) and [2022-03-18](https://dev.azure.com/vcpkg/public/_build/results?buildId=69051). Unfortunately I don't see obvious reasons why. Nothing else depends on this and nobody has noticed in 2 weeks, so I'm baslining it for now. (Will investigate shortly...) ## Only broken in tool update: REGRESSION: mesa:x64-windows failed with BUILD_FAILED. If expected, add mesa:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. ``` -- Downloading https://gitlab.freedesktop.org/mesa/mesa/-/archive/mesa-21.2.5/mesa-mesa-21.2.5.tar.gz -> mesa-mesa-mesa-21.2.5-1.tar.gz... -- Extracting source /Users/vagrant/Data/downloads/mesa-mesa-mesa-21.2.5-1.tar.gz -- Applying patch swravx512-post-static-link.patch -- Applying patch swr-msvc-2.patch -- Applying patch swr-llvm13.patch -- Applying patch radv-msvc-llvm13-2.patch -- Applying patch d3d10sw.patch -- Using source at /Users/vagrant/Data/buildtrees/mesa/src/esa-21.2.5-2df234d2b1.clean Traceback (most recent call last): File "<string>", line 1, in <module> ModuleNotFoundError: No module named 'mako' CMake Error at ports/mesa/portfile.cmake:85 (message): Python package 'mako' needs to be installed for port 'mesa'. Complete list of required python packages: setuptools;mako Call Stack (most recent call first): ports/mesa/portfile.cmake:91 (vcpkg_get_python_package) scripts/ports.cmake:145 (include) ``` Looks like this is being tracked by microsoft#23089 ; perhaps that we don't have as aggressive a recycling strategy for macos boxes as we do for the others has let different machines give different results? ## Only broken without tool update: REGRESSION: chromium-base:x64-osx. If expected, add chromium-base:x64-osx=fail to .\scripts\ci.baseline.txt. This one has been constantly flaky; I baselined it. REGRESSION: libxml2:x64-osx. If expected, add libxml2:x64-osx=fail to .\scripts\ci.baseline.txt. This port uses vcpkg_from_git and the upstream server was down during the build.
* Update vcpkg-tool to 2022-03-24 * Hook up microsoft/vcpkg-tool#345 * Hook up microsoft/vcpkg-tool#442 * Update vcpkg-tool to 2022-03-25 * Analysis of failures. * [Most recent nightly build failed](https://dev.azure.com/vcpkg/public/_build/results?buildId=69427) * [Validation of this tool update failed](https://dev.azure.com/vcpkg/public/_build/results?buildId=69417) ## Common to both: PASSING, REMOVE FROM FAIL LIST: chartdir:x64-windows (.\scripts\ci.baseline.txt) PASSING, REMOVE FROM FAIL LIST: chartdir:x64-windows-static-md (.\scripts\ci.baseline.txt) PASSING, REMOVE FROM FAIL LIST: chartdir:x86-windows (.\scripts\ci.baseline.txt) Probably fixed by #23701 PASSING, REMOVE FROM FAIL LIST: gmp:x64-uwp (.\scripts\ci.baseline.txt) PASSING, REMOVE FROM FAIL LIST: gmp:x64-windows-static-md (.\scripts\ci.baseline.txt) Probably fixed by #23466 ? REGRESSION: colmap:x64-windows-static-md failed with BUILD_FAILED. If expected, add colmap:x64-windows-static-md=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. I don't know exactly what changed. I observe that * this thing depends on a *lot* of stuff * on March 14 we didn't even attempt to build this * the x64-windows ones are already in the baseline so I skipped it. REGRESSION: qtdeclarative:x64-windows. If expected, add qtdeclarative:x64-windows=fail to .\scripts\ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\1\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: qtdeclarative:x64-windows failed with BUILD_FAILED. If expected, add qtdeclarative:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. This is a reporting change: The new world order also includes host build failures which is why it's duplicated. See also #23714 See also #23490 I'm nervous about baslining this because it seems most of the qt world is built on top of this port I filed #23824 about this and @Neumann-A indicated this should be fixed by #23755 REGRESSION: nettle:x64-uwp. If expected, add nettle:x64-uwp=fail to .\scripts\ci.baseline.txt. REGRESSION: nettle:x64-windows-static-md. If expected, add nettle:x64-windows-static-md=fail to .\scripts\ci.baseline.txt. REGRESSION: nettle:x64-uwp failed with BUILD_FAILED. If expected, add nettle:x64-uwp=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. REGRESSION: nettle:x64-windows-static-md failed with POST_BUILD_CHECKS_FAILED. If expected, add nettle:x64-windows-static-md=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. Didn't analyze, probably fixed by #23519 ? REGRESSION: libgpg-error:x64-uwp. If expected, add libgpg-error:x64-uwp=fail to .\scripts\ci.baseline.txt. REGRESSION: libgpg-error:x64-uwp failed with BUILD_FAILED. If expected, add libgpg-error:x64-uwp=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. This was broken by VS2022 update: ``` C:\Program Files\Microsoft Visual Studio\2022\Enterprise\MSBuild\Microsoft\VisualStudio\v17.0\AppxPackage\Microsoft.AppXPackage.Targets(892,25): error MSB4086: A numeric comparison was attempted on "$(TargetPlatformMinVersion)" that evaluates to "" instead of a number, in condition "'$(TargetPlatformMinVersion)' >= '10.0.17200.0'". [C:\Dev\vcpkg\buildtrees\libgpg-error\x64-uwp-rel\error-1.42-2324ddbc71.clean\SMP\libgpg-error_winrt.vcxproj] ``` REGRESSION: libmikmod:x64-osx. If expected, add libmikmod:x64-osx=fail to .\scripts\ci.baseline.txt. REGRESSION: libmikmod:x64-osx failed with BUILD_FAILED. If expected, add libmikmod:x64-osx=fail to /Users/vagrant/Data/work/2/s/scripts/azure-pipelines/../ci.baseline.txt. Broken between [2022-03-16](https://dev.azure.com/vcpkg/public/_build/results?buildId=68947) and [2022-03-18](https://dev.azure.com/vcpkg/public/_build/results?buildId=69051). Unfortunately I don't see obvious reasons why. Nothing else depends on this and nobody has noticed in 2 weeks, so I'm baslining it for now. (Will investigate shortly...) ## Only broken in tool update: REGRESSION: mesa:x64-windows failed with BUILD_FAILED. If expected, add mesa:x64-windows=fail to C:\a\2\s\scripts\azure-pipelines/../ci.baseline.txt. ``` -- Downloading https://gitlab.freedesktop.org/mesa/mesa/-/archive/mesa-21.2.5/mesa-mesa-21.2.5.tar.gz -> mesa-mesa-mesa-21.2.5-1.tar.gz... -- Extracting source /Users/vagrant/Data/downloads/mesa-mesa-mesa-21.2.5-1.tar.gz -- Applying patch swravx512-post-static-link.patch -- Applying patch swr-msvc-2.patch -- Applying patch swr-llvm13.patch -- Applying patch radv-msvc-llvm13-2.patch -- Applying patch d3d10sw.patch -- Using source at /Users/vagrant/Data/buildtrees/mesa/src/esa-21.2.5-2df234d2b1.clean Traceback (most recent call last): File "<string>", line 1, in <module> ModuleNotFoundError: No module named 'mako' CMake Error at ports/mesa/portfile.cmake:85 (message): Python package 'mako' needs to be installed for port 'mesa'. Complete list of required python packages: setuptools;mako Call Stack (most recent call first): ports/mesa/portfile.cmake:91 (vcpkg_get_python_package) scripts/ports.cmake:145 (include) ``` Looks like this is being tracked by #23089 ; perhaps that we don't have as aggressive a recycling strategy for macos boxes as we do for the others has let different machines give different results? ## Only broken without tool update: REGRESSION: chromium-base:x64-osx. If expected, add chromium-base:x64-osx=fail to .\scripts\ci.baseline.txt. This one has been constantly flaky; I baselined it. REGRESSION: libxml2:x64-osx. If expected, add libxml2:x64-osx=fail to .\scripts\ci.baseline.txt. This port uses vcpkg_from_git and the upstream server was down during the build. * Restore chartdir to the baseline, I thought #23732 had been merged.
|
FYI, you seem to have left |
No description provided.