https://bugzilla.redhat.com/show_bug.cgi?id=1176595 --- Comment #17 from Dave Love <d.love@xxxxxxxxxxxxxxx> --- (In reply to Zbigniew Jędrzejewski-Szmek from comment #16) > Use %autosetup please with so many patches. Than I wouldn't be able to use it on the platform I care about (el6). > A single Requires/Provides per line. Where's that documented? > %doc is used for some license files. Thanks. > .so files would usually be named like libHYPRE.so.2.10.1 not > libHYPRE-2.10.1.so. With the current naming, the number is part of the file > name. See the comment in the spec file. That's kept in Debian and, I think, OpenSuSE, and I didn't see anything prohibiting it when I checked. > I don't really know what the effect of this is. It will probably break > linking against the library... Do you have any programs using the headers > and libraries in -devel? Yes. This was done as a dependency of petsc and trilinos. $ rpm -q --requires trilinos-openmpi|grep -i hypre libHYPRE-2.10.1.so()(64bit) I haven't fixed the spec for the latest petsc properly yet, but it builds. > It seem that at least it breaks automatic MPI provides/requires generation: > $ rpm -qp -R results/hypre-openmpi-2.10.1-1.fc24.x86_64.rpm > openmpi(x86-64) > rpmlib(CompressedFileNames) <= 3.0.4-1 > rpmlib(FileDigests) <= 4.6.0-1 > rpmlib(PayloadFilesHavePrefix) <= 4.0-1 > rpmlib(PayloadIsXz) <= 5.2-1 > > as you can see, there are no requirements on libraries. That sounds like a bug in the automatic MPI stuff. An MPI LD_PRELOAD library for profiling or debugging will typically not have a version suffix, for instance, and as far as I can tell, the form that hypre uses is legitimate. -- You are receiving this mail because: You are on the CC list for the bug. You are always notified about changes to this product and component _______________________________________________ package-review mailing list package-review@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/package-review