koobs added the comment: I recently updated the buildbot host to the latest 9-STABLE sources (changeset delta is approximately 4-6 weeks)
On a hunch I asked our src committers if there had been any sendfile related changes and was pointed to 3 changesets that were merged from current (MFC) to stable/9 4 weeks ago: http://svnweb.freebsd.org/base?view=revision&revision=250907 I reverted the kernel to r250906 (hunch commit -1) that resulted in a clean run: http://buildbot.python.org/all/builders/AMD64%20FreeBSD%209.0%20dtrace%203.x/builds/1563 Manual Rebuild: Trying to isolate test_trailing with older kernel I have it on preliminary evidence that our regression tests are passing on this functionality, but analysis at the Python end for a root cause and details on how the test implementation plays a role would be helpful. ---------- _______________________________________ Python tracker <rep...@bugs.python.org> <http://bugs.python.org/issue18296> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
