[Integration PR] Noncopyable generics in stdlib (phase 2) + Use new noncopyable types infrastructure #72346
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
|
swiftlang/llvm-project#8384 |
|
swiftlang/llvm-project#8384 |
|
swiftlang/llvm-project#8384 |
|
Hm, I see some local failures and unexpected passes: |
|
Well this isn't annoying at all |
lorentey
force-pushed
the
noncopyable-primitives-and-more!
branch
2 times, most recently
from
8fa25b7 to
9b8ddc2
Compare
|
swiftlang/llvm-project#8384 |
|
@swift-ci benchmark |
|
swiftlang/llvm-project#8384 |
|
Oh hey, the benchmarks succeeded! |
lorentey
force-pushed
the
noncopyable-primitives-and-more!
branch
from
6b168dc to
3bf9ef0
Compare
|
swiftlang/llvm-project#8384 |
|
@swift-ci benchmark |
lorentey
force-pushed
the
noncopyable-primitives-and-more!
branch
from
8cfcd5e to
c821c2f
Compare
|
swiftlang/llvm-project#8384 |
|
@swift-ci benchmark |
|
@swift-ci build toolchain macOS |
12 tasks
|
D'oh, I have a duplicate entry in the ABI expectations list: |
|
#72348 has probably obsoleted some symbol mismatch expectations; I'll rebase and try to verify locally |
[stdlib] MemoryLayout: Update Swift version numbers [stdlib] MemoryLayout: Actually hide legacy ABI
… types [stdlib] Pull back @_aeic on pointer → integer conversions [stdlib] UnsafeMutablePointer.allocate: Fix thinko [stdlib] Disable support for noncopyable pointees on some pointer operations We have to temporarily pull back support for noncopyable pointees for UnsafeMutablePointer.initialize(to:), .moveInitialize, .moveUpdate, as the builtins they’re calling are no longer accepting such types. These will return following a builtin audit. [stdlib] Remove workarounds for certain builtins not supporting noncopyable use swiftlang#71733 fixed this! [stdlib] Update FIXME [stdlib] UnsafePointer: Update Swift version numbers [stdlib] UnsafePointer: Actually hide legacy ABI [stdlib] Remove workaround for U[M]BP.withMemoryRebound
- Enable BorrowingSwitch feature within the stdlib - ExpressibleByNilLiteral: Add retroactive support for noncopyable conforming types - Optional: draft an API surface for noncopyable payloads [stdlib] Oops, the ExpressibleByNilLiteral conformance kept its implicit copyability
The test was relying on UnsafeMutablePointer (UMP) not having a Copyable requirement. Those Copyable requirements are more prominent when using the new infrastructure for noncopyable types. Since UMP is planned to have `~Copyable` on its generic parameter quite soon, this shouldn't remain broken for too long.
It's not currently clear whether this is a regression or just a test that needs updating. Investigation is needed.
It's not currently clear whether this is a regression or just a test that needs updating. For now it seems harmless to just update the test rather than disable it.
This test is failing to verify the generic signature for this bogus program, because the generic signature isn't minimal. We should gracefully handle this situation and still have a minimal signature, despite rejecting this program.
The presence of Copyable/Escapable conformances doesn't affect ABI. Only their absence in terms of suppressed requirements like `~Copyable` need to be output for diffing by the APIDigester.
This appears to be a code path that wasn't previously stressed when deserializing a bogus module, but now it is with NoncopyableGenerics, as Copyable is often emitted as a builtin conformance.
After removing the implicit conditional conformance synthesis for inverses, this test wasn't updated.
[gardening] update copyright notice [gardening] update copyright notice [gardening] update copyright notice [gardening] update copyright notice
…rk when using Optional and other stdlib features
lorentey
force-pushed
the
noncopyable-primitives-and-more!
branch
from
c821c2f to
8ea0a2d
Compare
|
Closing as obsolete, as #72276 has now landed on main 🎉 |
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 integrates #71688 and #72276 into a single PR, to let me look at test failures in the former before waiting on the latter.