libevent: call autoreconf directly instead of via autogen.sh#41057
libevent: call autoreconf directly instead of via autogen.sh#41057alalazo merged 1 commit intospack:developfrom
Conversation
|
Hi @blue42u! I noticed that the following package(s) don't yet have maintainers:
Are you interested in adopting any of these package(s)? If so, simply add the following to the package class: maintainers("blue42u")If not, could you contact the developers of this package and see if they are interested? You can quickly see who has worked on a package with $ spack blame libeventThank you for your help! Please don't add maintainers without their consent. You don't have to be a Spack expert or package developer in order to be a "maintainer," it just gives us a list of users willing to review PRs or debug issues relating to this package. A package can have multiple maintainers; just add a list of GitHub handles of anyone who wants to volunteer. |
3946efb to
e0c7691
Compare
Head branch was pushed to by a user without write access
e0c7691 to
e834226
Compare
|
Looking into the CI failure, apparently Spack's # # On ppc64le
# PATH=$(spack location -i which)/bin:$PATH which which
which: no which in (/home/software/spack/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeh/linux-ubuntu20.04-ppc64le/gcc-9.4.0/which-2.21-57arjijrsllc6tozz3mnhirvq47xpei6/bin:/opt/bootstrap/view/bin:/builds/spack/spack/bin:/opt/bootstrap/view/bin:/opt/bootstrap:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin)
# # On x86-64
# PATH=$(spack location -i which)/bin:$PATH which which
/home/software/spack/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeholder__/__spack_path_placeh/linux-ubuntu20.04-x86_64_v3/gcc-11.4.0/which-2.21-e7g4ojtne4qs2qsjnd4fyt2agwncljdm/bin/whichAside: Upon further investigation
Given the wide range of So, take three: I've adjusted the recipe to just run |
The deps were added in spack#40945 to make it work on macOS 11, because the old configure scripts only detect macOS 10. Apparently people the autoreconf script caused issues, later fixed in spack#41057. However, also with that fix, things are incorrect, cause people now report: ``` libtool: You should recreate aclocal.m4 with macros from libtool 2.4.7 libtool: and run autoconf again. ``` I did not check what that error is about exactly, but it looks like Spack's `def autoreconf` step does way less than the `autgen.sh` script provided in the source tarball. HOWEVER, all this is unnecessary, because the underlying issue was already fixed long ago, it's just that it regressed at some point, but it's back in place since spack#41205, therefore, get rid of all this madness.
The deps were added in #40945 to make it work on macOS 11, because the old configure scripts only detect macOS 10. Apparently people reported the autoreconf script caused issues, later fixed in #41057. However, also with that fix, things are incorrect, cause people now report: ``` libtool: You should recreate aclocal.m4 with macros from libtool 2.4.7 libtool: and run autoconf again. ``` HOWEVER, all this is unnecessary, because the underlying issue was already fixed long ago, it's just that it regressed at some point, but it's back in place since #41205.
The deps were added in spack#40945 to make it work on macOS 11, because the old configure scripts only detect macOS 10. Apparently people reported the autoreconf script caused issues, later fixed in spack#41057. However, also with that fix, things are incorrect, cause people now report: ``` libtool: You should recreate aclocal.m4 with macros from libtool 2.4.7 libtool: and run autoconf again. ``` HOWEVER, all this is unnecessary, because the underlying issue was already fixed long ago, it's just that it regressed at some point, but it's back in place since spack#41205.
The deps were added in spack#40945 to make it work on macOS 11, because the old configure scripts only detect macOS 10. Apparently people reported the autoreconf script caused issues, later fixed in spack#41057. However, also with that fix, things are incorrect, cause people now report: ``` libtool: You should recreate aclocal.m4 with macros from libtool 2.4.7 libtool: and run autoconf again. ``` HOWEVER, all this is unnecessary, because the underlying issue was already fixed long ago, it's just that it regressed at some point, but it's back in place since spack#41205.
Fixes #41056.
The
autogen.shscript shipped by libevent useswhichto determine ifautoreconfis available, butwhichisn't installed by default on all systems (e.g. Fedora). If Spack is building withlibtool @2.4.7this causes build failures since the libtool M4 fragments in libevent are libtool 2.4.6.This PR adds a patch to replace the use ofwhichwithcommand -v, which is a POSIX shell builtin and thus more portable.EDIT: This PR fixes this by adding a build dependency onwhich.EDIT: This PR fixes this by calling
autoreconfdirectly, instead of trying to getautogen.shto do so.