这是indexloc提供的服务，不要输入任何密码
Skip to content

explicitly build against system-libcxxabi #177

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.

Sign up for GitHub

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

Jump to bottom
Open
wants to merge 9 commits into
base: main
Choose a base branch
from
Open

explicitly build against system-libcxxabi #177

wants to merge 9 commits into from

Conversation

@h-vetinari
Copy link
Member

@h-vetinari h-vetinari commented Jul 31, 2024 •
edited
Loading

This option landed in libcxx 15 (llvm/llvm-project@ba4f1e4) but I didn't notice it until now (but then, I also understood the abi part even less well than today... 😅)

See also docs.

@conda-forge-webservices
Copy link

conda-forge-webservices bot commented Jul 31, 2024 •
edited by conda-forge-admin
Loading

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe/meta.yaml) and found it was in an excellent condition.

@h-vetinari h-vetinari changed the base branch from main to rc August 17, 2024 10:14
@h-vetinari h-vetinari mentioned this pull request Aug 17, 2024
Closed
@h-vetinari h-vetinari changed the base branch from rc to main October 4, 2024 13:34
@h-vetinari h-vetinari mentioned this pull request Oct 6, 2024
Merged
@conda-forge-admin
Copy link
Contributor

conda-forge-admin commented Dec 18, 2024 •
edited
Loading

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe/meta.yaml) and found it was in an excellent condition.

I do have some suggestions for making it better though...

For recipe/meta.yaml:

  • ℹ️ The recipe is not parsable by parser conda-souschef (grayskull). This parser is not currently used by conda-forge, but may be in the future. We are collecting information to see which recipes are compatible with grayskull.
  • ℹ️ The recipe is not parsable by parser conda-recipe-manager. The recipe can only be automatically migrated to the new v1 format if it is parseable by conda-recipe-manager.

This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/15802021717. Examine the logs at this URL for more detail.

@h-vetinari
get rid of CMAKE_EXTRA_ARGS
d243ea6
@h-vetinari
explicitly build against system-libcxxabi
70d88f3
@h-vetinari
set LIBCXX_CXX_ABI_INCLUDE_PATHS
db4a3e2
@h-vetinari
unbreak circularity between clangxx and libcxx-devel
86396f5 
the only difference between clang & clangxx on osx is a dependence on libcxx-devel,
with a version matching that of clangxx; we cannot do that here because, by necessity,
we're building a newer libcxx-devel before we have a matching clangxx.
@h-vetinari
move libcxx-testing downstream tests to libcxx; add v20
dc66a8b
@h-vetinari
bump build number
9dac5a9
@h-vetinari
Copy link
Member Author

h-vetinari commented Jun 22, 2025

@mgorny, since you were involved on the upstream issue for this - any ideas what's going on here? I'm getting

dyld[29511]: Symbol not found: __ZTVN10__cxxabiv117__class_type_infoE
  Referenced from: <071C190C-8BFC-3957-84B7-1688F23DE737> /Users/runner/miniforge3/conda-bld/libcxx_and_libcxxabi_1750559514100/_test_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_pla/bin/clang-20
  Expected in:     <74BCAF44-3A02-301E-8F16-B7DA6ED48C6F> /Users/runner/miniforge3/conda-bld/libcxx_and_libcxxabi_1750559514100/_test_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_pla/lib/libc++.1.0.dylib
/bin/sh: line 4: 29511 Abort trap: 6           x86_64-apple-darwin13.4.0-clang++ -march=core2 -mtune=haswell -mssse3 -ftree-vectorize -fPIC -fstack-protector-strong -O2 -pipe -stdlib=libc++ -fvisibility-inlines-hidden -fmessage-length=0 -isystem /Users/runner/miniforge3/conda-bld/libcxx_and_libcxxabi_1750559514100/_test_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_pla/include -fdebug-prefix-map=/Users/runner/miniforge3/conda-bld/libcxx_and_libcxxabi_1750559514100/test_tmp=/usr/local/src/conda/libcxx-testing-0.20.99 -fdebug-prefix-map=/Users/runner/miniforge3/conda-bld/libcxx_and_libcxxabi_1750559514100/_test_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_pla=/usr/local/src/conda-prefix -Wl,-headerpad_max_install_names -Wl,-dead_strip_dylibs -Wl,-rpath,/Users/runner/miniforge3/conda-bld/libcxx_and_libcxxabi_1750559514100/_test_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_pla/lib -L/Users/runner/miniforge3/conda-bld/libcxx_and_libcxxabi_1750559514100/_test_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_pla/lib -undefined dynamic_lookup -shared -fPIC Exception.cc -o libException.dylib

in our https://github.com/conda-forge/libcxx-testing-feedstock downstream tests. These tests are a bit weird

libcxx-feedstock/recipe/meta.yaml

Lines 120 to 123 in 5e730f0

# test current libcxx against old clang builds;
# version correspondence is 0.{{ CLANG_MAJOR }}
# these tests are unusual in that they use -Wl,-rpath, but not -L.
- libcxx-testing 0.19 # [osx]

but a priori we should avoid breaking this.

@mgorny
Copy link

mgorny commented Jun 22, 2025

I'm afraid I don't know the issue offhand (and to be honest, I don't really want to try to figure it out, my brain is melting in this heat). Plus, I have no Darwin experience here, and everything related to libc++ is driver-specific, so Darwin driver has probably a different logic and different expectations than Linux or NetBSD drivers I've worked with.

If that's any help, here are the Gentoo ebuilds for respective libc++abi and libc++ versions:
https://gitweb.gentoo.org/repo/gentoo.git/tree/llvm-runtimes/libcxxabi/libcxxabi-20.1.7.ebuild
https://gitweb.gentoo.org/repo/gentoo.git/tree/llvm-runtimes/libcxx/libcxx-20.1.7.ebuild

In particular, our custom ldscripts may be relevant:
https://gitweb.gentoo.org/repo/gentoo.git/tree/llvm-runtimes/libcxx/libcxx-20.1.7.ebuild#n183

We need them since the Linux driver only passes -lcxx, and then the ldscript takes care of including libc++abi.so (or .a).

However, I'm afraid that's how far as my knowledge goes. Most of this hackery really predates me, and I've only been trying to keep it working the best I could.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@SylvainCorlay SylvainCorlay Awaiting requested review from SylvainCorlay SylvainCorlay is a code owner

@beckermr beckermr Awaiting requested review from beckermr beckermr is a code owner

@inducer inducer Awaiting requested review from inducer inducer is a code owner

@isuruf isuruf Awaiting requested review from isuruf isuruf is a code owner

@jakirkham jakirkham Awaiting requested review from jakirkham jakirkham is a code owner

@timsnyder timsnyder Awaiting requested review from timsnyder timsnyder is a code owner

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@h-vetinari @conda-forge-admin @mgorny