Created attachment 898116 [details] (slightly truncated) catalyst.log At first glance this seems to be a return of bug 904338, the patch from that fix doesn't seem to have survived the bump to 1.47.1. This particular failure for fallocate.o was among the patched make targets in that patch, and the fix does not exist in the new upstream code. (All of this is "as far as I can tell", there could be more to the issue, or something more subtle). Saw this failure on x32 autobuilds which happen on a 32-thread amd64 host.
On x32 e2fsprogs needs ext2_err.h inside the /usr/include/ext2fs/ directory to build but amd64 (glibc and musl) and x86 does not hit this issue. We notice this in releng as we don't build e2fsprogs in stage1 so this file isn't there. Simply copying over the file allows it to build. I can't see why this is needed on x32 only but hopefully is enough of a clue for someone else.
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=fe128bcd6e3f2b834e804a13360ed297ad60f5ab commit fe128bcd6e3f2b834e804a13360ed297ad60f5ab Author: Mike Gilbert <floppym@gentoo.org> AuthorDate: 2024-09-02 17:51:42 +0000 Commit: Mike Gilbert <floppym@gentoo.org> CommitDate: 2024-09-02 17:53:18 +0000 sys-fs/e2fsprogs: work around broken Makefile Bug: https://bugs.gentoo.org/936493 Signed-off-by: Mike Gilbert <floppym@gentoo.org> sys-fs/e2fsprogs/e2fsprogs-1.47.1.ebuild | 4 ++++ 1 file changed, 4 insertions(+)
Please let me know if this resolves the issue.
Works a treat in both my catalyst build and chroot, thanks.