Skip to content

Disambiguate apple and darwin triples #6478

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Apr 26, 2023
Merged

Disambiguate apple and darwin triples #6478

merged 2 commits into from
Apr 26, 2023

Conversation

rauhul
Copy link
Member

@rauhul rauhul commented Apr 25, 2023

Fixes a bug where all triples with vendor apple are considered to be os variants of darwin. This change special cases "apple-none" to not match Triple.isDarwin(), but all other "apple-*" triples will still be considered to be darwin variant.

Fixes some fallout of making the above change to check for the newly added Triple.isApple() instead of Triple.isDarwin() to allow the triple "apple-none" to opt into behaviors like dead stripping, using libtool, etc.

As a result of these changes spm no longer appends a macOS version number to the "armv7em-apple-none-macho" triple (e.g. "armv7em-apple-none-macho10.13") removing the work around of passing "-target armv7em-apple-none-macho" via cli args or via a toolset.json.

Additionally spm no longer passes unexpected -rpath args to the linker when building a static binary for "armv7em-apple-none-macho".

@rauhul
Copy link
Member Author

rauhul commented Apr 25, 2023

this needs to be rebased on main once #6438 lands due to the use of .noneOS.

Copy link
Contributor

@tomerd tomerd left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

very nice. could we add a small unit test to prevent regressions?

@kubamracek kubamracek self-requested a review April 25, 2023 19:00
rauhul added 2 commits April 25, 2023 17:12
Fixes a bug where all triples with vendor apple are considered to be os
variants of darwin. This change special cases "apple-none" to not match
`Triple.isDarwin()`, but all other "apple-*" triples will still be
considered to be darwin variant.

Fixes some fallout of making the above change to check for the newly
added `Triple.isApple()` instead of `Triple.isDarwin()` to allow the
triple "apple-none" to opt into behaviors like dead stripping, using
libtool, etc.

As a result of these changes spm no longer appends a macOS version
number to the "armv7em-apple-none-macho" triple (e.g.
"armv7em-apple-none-macho10.13") removing the work around of passing
"-target armv7em-apple-none-macho" via cli args or via a toolset.json.

Additionally spm no longer passes unexpected -rpath args to the linker
when building a static binary for "armv7em-apple-none-macho".
@rauhul rauhul force-pushed the apple-not-equal-darwin branch from 0cf84e2 to 2e6d5bf Compare April 26, 2023 02:11
@rauhul
Copy link
Member Author

rauhul commented Apr 26, 2023

@swift-ci please smoke test

@rauhul rauhul enabled auto-merge (squash) April 26, 2023 02:25
@rauhul
Copy link
Member Author

rauhul commented Apr 26, 2023

@swift-ci please smoke test windows

@rauhul rauhul merged commit ec77f08 into main Apr 26, 2023
@rauhul rauhul deleted the apple-not-equal-darwin branch April 26, 2023 04:51
MaxDesiatov pushed a commit that referenced this pull request Jul 18, 2023
Cherry-pick of #6478.

Fixes a bug where all triples with vendor `apple` are considered to be OS variants of `darwin`. This change special cases `apple-none` to not match `Triple.isDarwin()`, but all other "apple-*" triples will still be
considered to be a variant `darwin`.

Fixes some fallout of making the above change to check for the newly added `Triple.isApple()` instead of `Triple.isDarwin()` to allow the triple `apple-none` to opt into behaviors like dead stripping, using
libtool, etc.

As a result of these changes SwiftPM no longer appends a macOS version number to the `armv7em-apple-none-macho` triple (e.g. `armv7em-apple-none-macho10.13`) removing the work around of passing `-target armv7em-apple-none-macho` via cli args or via a `toolset.json`.

Additionally SwiftPM no longer passes unexpected `-rpath` args to the linker when building a static binary for `armv7em-apple-none-macho`.

# Conflicts:
#	Tests/BasicsTests/TripleTests.swift
MaxDesiatov added a commit that referenced this pull request Jul 19, 2023
Cherry-pick of #6478.

Fixes a bug where all triples with vendor `apple` are considered to be OS variants of `darwin`. This change special cases `apple-none` to not match `Triple.isDarwin()`, but all other "apple-*" triples will still be
considered to be a variant `darwin`.

Fixes some fallout of making the above change to check for the newly added `Triple.isApple()` instead of `Triple.isDarwin()` to allow the triple `apple-none` to opt into behaviors like dead stripping, using
libtool, etc.

As a result of these changes SwiftPM no longer appends a macOS version number to the `armv7em-apple-none-macho` triple (e.g. `armv7em-apple-none-macho10.13`) removing the work around of passing `-target armv7em-apple-none-macho` via cli args or via a `toolset.json`.

Additionally SwiftPM no longer passes unexpected `-rpath` args to the linker when building a static binary for `armv7em-apple-none-macho`.

# Conflicts:
#	Tests/BasicsTests/TripleTests.swift

Co-authored-by: Rauhul Varma <rauhul@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants