It looks like there's a race condition with FEDORA-EPEL-2026-81afbdedb7
From what I can tell, the updates should get combined, as either one landing without the other looks like it would break things.
It looks like there's a race condition with FEDORA-EPEL-2026-e989a60156
From what I can tell, the updates should get combined, as either one landing without the other looks like it would break things.
Looks like this was missed in this update. I submitted the missing build: FEDORA-EPEL-2026-920b6955e3
This one ended up in /dev/null, filed https://forge.fedoraproject.org/releng/tickets/issues/13331
This appears to be a false positive. I just downloaded the builds (bodhi updates download --updateid <this update>) and they all install correctly in a clean Fedora 43 mock chroot (mock -r fedora-43-x86_64 install ./*.rpm).
This update seems to have accidentally bumped the soname of libcpuinfo.so. Dependent packages (fbgemm, nnpack, onnxruntime) were not rebuilt for this so they now fail to install.
22.1.2 appears to fix some weird codegen bugs on ppc64le :thumbsup:
The builds from this update all install successfully in a alma+epel-9 environment. The uu_* and uutils-coreutils commands appear to work as expected. This update also fixes preexisting FailsToInstall bugs for rust-uu_{dir,ls,vdir} and a FailsToBuild issue for rust-coreutils.
Yup, this "security fix" breaks applications, including Steam: https://github.com/flatpak/flatpak/issues/6568
No issues encountered.
Works fine for running tests in the tox py312 env.
No longer crashes!
It appears that all packages that use setuptools_rust build correctly with 1.12.1.
I tested this with
And all packages installed without dependency issues.