Created attachment 896814 [details] build.log and emerge --info Presumably this is the same as bug 933494 https://sourceware.org/bugzilla/show_bug.cgi?id=30584 This IS A REGRESSION from 9.4-r1 FAIL: test-random_r =================== FAIL test-random_r (exit status: 138)
I wonder what change in 9.4 vs 9.5 exposed it (a gnulib change, perhaps) or whether it's a glibc one (doesn't line up with the timing)?
(In reply to Sam James from comment #1) > I wonder what change in 9.4 vs 9.5 exposed it (a gnulib change, perhaps) or > whether it's a glibc one (doesn't line up with the timing)? 9.4-r1 passes with same setup. The test is under gnulib, and is specifically called out: https://github.com/coreutils/gnulib/commit/ea3dd857725e176232ae61a1ad7aff498c2939ce