Always check if found libraries are linkable #15134
Open
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
thesamesam
force-pushed
the
sam-libatomic
branch
from
c0b3e68 to
e52cf63
Compare
thesamesam
force-pushed
the
sam-libatomic
branch
2 times, most recently
from
7fc99ba to
c31a10f
Compare
|
This seems like the right approach to me. |
In an Ubuntu Bionic container with Python 3.7.5: ``` $ python3 -c 'import tempfile; print(tempfile.mkdtemp(dir=""))' tmpycdjfo6m $ python3 -c 'import tempfile; print(tempfile.mkdtemp(dir=None))' /tmp/tmpy6dlpv0r ``` Pass dir as None to get the behaviour we need - which is how newer Pythons act, as otherwise we aren't running in the cwd for the compile test that we expect. This shows up as a failure in _test_all_naming when _find_library_real is modified to use links().
We don't always check if a library is actually linkable because of concerns that a library may not be standalone, so linking against it may still have unresolved references. We can workaround that by building a shared library as a test rather than an executable, and in fact we already do that in a bunch of cases since bb5f2ca. This comes up in particular with Fedora and dependency('atomic') because on x86_64, they provide a libatomic.so linker script (so our file existence check passes), but trying to use it later on will fail if the backing package isn't installed. _get_file_from_list has not been deleted because the Ninja backend's guess_library_absolute_path uses it. We might be able to change it to also use a link test but I've left that alone. Test notes: * The _test_all_naming unittest has been tweaked because we were using a blank file rather than an actual shared library, which (as expected), now fails. * I've also added a test for dependency('atomic') providing a result that can actually be linked against. I've not made it check generally whether dependency('atomic') finds something when we expect to (at least for now) as it'll involve some CI whack-a-mole. Bug: https://bugzilla.redhat.com/show_bug.cgi?id=2352531 Bug: mesonbuild#14946 Closes: mesonbuild#10936
thesamesam
force-pushed
the
sam-libatomic
branch
from
c31a10f to
8f3f0e7
Compare
dcbaker
approved these changes
Oct 20, 2025
|
@eli-schwartz, this looks right to me, and like it would fix a lot issues. I think you're more qualified to review this than I am though, so, do you want to have a look or just trust Sam and I? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Always check if found libraries are linkable
We don't always check if a library is actually linkable because of concerns
that a library may not be standalone, so linking against it may still have
unresolved references. We can workaround that by building a shared library
as a test rather than an executable, and in fact we already do that in a
bunch of cases since bb5f2ca.
This comes up in particular with Fedora and dependency('atomic') because
on x86_64, they provide a libatomic.so linker script (so our file existence
check passes), but trying to use it later on will fail if the backing package
isn't installed.
The _test_all_naming unittest has been tweaked because we were using a
blank file rather than an actual shared library, which (as expected),
now fails.
I've also added a test for dependency('atomic') providing a result that
can actually be linked against. I've not made it check generally whether
dependency('atomic') finds something when we expect to (at least for now)
as it'll involve some CI whack-a-mole.
_get_file_from_list has not been deleted because the Ninja backend's
guess_library_absolute_path uses it. We might be able to change it to
also use a link test but I've left that alone.
Bug: https://bugzilla.redhat.com/show_bug.cgi?id=2352531
Bug: #14946
Closes: #10936