Date: Tue, 05 Mar 2019 13:19:31 +0000 From: bugzilla-noreply@freebsd.org To: bugs@FreeBSD.org Subject: [Bug 236233] Python multiprocessing is 20x slower with 11.2 compared to 11.0 kernel Message-ID: <bug-236233-227-AIAXWDjglx@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-236233-227@https.bugs.freebsd.org/bugzilla/>
index | next in thread | previous in thread | raw e-mail
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=236233 --- Comment #1 from Alex Richardson <arichardson@FreeBSD.org> --- I can confirm that this is a kernel regression between 11.0 and 11.2. I updated the kernel on one of the build machines from 11.0 to 11.2 (same userspace) and now the testsuite is much slower. Running `./bin/llvm-lit -j32 test/CodeGen/Mips` from the LLVM build directory now takes 121 seconds instead of 29, i.e. 4x slower. On a Linux build slave with the same hardware running the test/CodeGen/Mips tests only takes 6.8 seconds. Once a I have some more time I can try bisecting but that will probably take a while. -- You are receiving this mail because: You are the assignee for the bug.home | help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-236233-227-AIAXWDjglx>