[vcpkg] Remove the tombstones and 'ignore' baseline concepts. #12197
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
I forgot about our 'save_failure_logs' function which will be broken by this, switching to draft. |
ras0219
reviewed
Jul 1, 2020
strega-nil
approved these changes
Jul 2, 2020
This changes our PR builds to treat 'fail' in the ci.baseline.txt as 'skip' instead of using tombstones. We currently have large numbers of spurious failures that get enshrined in PRs through no fault of a PR author, removing the tombstones concept will fix those by allowing the user to retry. This does mean we accept some risk of not detecting when a port is 'fixed', but that failure is reasonable for us to handle after we see it in CI, but that seems worth it given that it lets us get rid of the tombstone concept. This also helps out the binary caching feature, because we don't have to figure out how to productize tombstones.
c39d715 to
78d82f2
Compare
|
Example build with failures: https://dev.azure.com/vcpkg/public/_build/results?buildId=39679&view=results |
317f68c to
8cd3f8f
Compare
6c95633 to
129f482
Compare
ras0219-msft
approved these changes
Jul 2, 2020
toolsrc/include/vcpkg/vcpkgpaths.h
Outdated
| @@ -1,4 +1,5 @@ | |||
| #pragma once | |||
| #pragma once | |||
toolsrc/src/vcpkg/build.cpp
Outdated
| fs.create_directories(paths.buildtrees / name, ec); | ||
| const auto abi_file_path = paths.buildtrees / name / (triplet.canonical_name() + ".vcpkg_abi_info.txt"); | ||
| auto current_build_tree = paths.build_dir(action.spec); | ||
| fs.create_directory(current_build_tree, ec); |
There was a problem hiding this comment.
Could use ignore_errors here.
There was a problem hiding this comment.
Is paths.buildtrees guaranteed to exist at this point? Should this be create_directories()?
There was a problem hiding this comment.
Hmmm or maybe it should be VCPKG_LINE_INFO. No reason to continue if we can't create a build tree.
I was trying to avoid create_directories because in profiling that showed up as being expensive. Let me see if I can find where we create buildtrees (because we do, it worked in testing so something must have done it already...)
There was a problem hiding this comment.
It looks like we create buildtrees when we make the "CMakeVarProvider"
toolsrc/src/vcpkg/commands.ci.cpp
Outdated
| { | ||
| using namespace vcpkg::Build; | ||
|
|
||
| const fs::path log_path = fs::u8path(".log"); |
There was a problem hiding this comment.
log_extension or log_ext? I confused log_path with "prefix path under which logs are found"
There was a problem hiding this comment.
Renamed to 'dot_log' and 'readme_dot_log' respectively.
strega-nil
approved these changes
Jul 2, 2020
E452003
added a commit
to E452003/vcpkg
that referenced
this pull request
Jul 6, 2020
* 'master' of https://github.com/microsoft/vcpkg: (1418 commits) [vcpkg integrate] Clean up vcpkg.target file (microsoft#4608) [vcpkg_from_sourceforge] Add retry mirror function (2/2) (microsoft#12018) [pcre2] Restore the https://ftp.pcre.org/ mirror in addition to the SourceForge mirrors. (microsoft#12233) [xercesc] rename feature from xmlch_wchar to xmlch-wchar (microsoft#12205) [safeint] Update to 3.24 (microsoft#12217) [vcpkg] Remove the tombstones and 'ignore' baseline concepts. (microsoft#12197) [msbuild] Revert the importance to Normal (microsoft#12212) [vtk] Added opengl feature. (microsoft#11399) [span-lite] Update to 0.7.0 (microsoft#12206) [cmocka libarchive libiconv libpq libxml2 plibsys] fix drive-by error in vcpkg-cmake-wrappers (microsoft#12196) [azure-iot-sdk-c] Fix feature name and enable to build (microsoft#12209) [vcpkg] Improve performance of compiler tracking by suppressing aspects of CMake's compiler detection. (microsoft#12203) [vcpkg] Remove all uses of Foo::Foo() noexcept = default; to fix microsoft#9955 (microsoft#12201) [sqlite3] update to 3.32.3 to deal with security issues (microsoft#12185) [infoware] Bump version to 0.5.4 (microsoft#12167) [imgui] Update to 1.77 (microsoft#12155) [vcpkg] Update message in bootstrap.ps1 (microsoft#12145) [vcpkg] Enable NuGet-based binary caching via mono (microsoft#12170) Don't change manifest root when manifest isn't enabled. (microsoft#12191) Fix sourceparagraph:BooleanField (microsoft#12192) ...
strega-nil
pushed a commit
to strega-nil/vcpkg
that referenced
this pull request
May 5, 2021
…oft#12197) This changes our PR builds to treat 'fail' in the ci.baseline.txt as 'skip' instead of using tombstones. We currently have large numbers of spurious failures that get enshrined in PRs through no fault of a PR author, removing the tombstones concept will fix those by allowing the user to retry. This does mean we accept some risk of not detecting when a port is 'fixed', but that failure is reasonable for us to handle after we see it in CI, but that seems worth it given that it lets us get rid of the tombstone concept. This also helps out the binary caching feature, because we don't have to figure out how to productize tombstones.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This changes our PR builds to treat 'fail' in the ci.baseline.txt as 'skip' instead of using tombstones.
We currently have large numbers of spurious failures that get enshrined in PRs through no fault of a PR author, removing the tombstones concept will fix those by allowing the user to retry. This does mean we accept some risk of not detecting when a port is 'fixed', but that failure is reasonable for us to handle after we see it in CI, but that seems worth it given that it lets us get rid of the tombstone concept.
This also helps out the binary caching feature, because we don't have to figure out how to productize tombstones.