Date: Thu, 26 Aug 2004 09:53:24 -0500 From: Craig Boston <craig@xfoil.gank.org> To: Doug White <dwhite@gumbysoft.com> Cc: freebsd-current@freebsd.org Subject: Re: PLEASE TEST: IPI deadlock avoidance patch Message-ID: <20040826145324.GA40029@nowhere> In-Reply-To: <20040822115345.Y94593@carver.gumbysoft.com> References: <20040822115345.Y94593@carver.gumbysoft.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Aug 22, 2004 at 12:05:39PM -0700, Doug White wrote: > If you have a reasonably fast i386 or amd64 multiprocessor and/or > hyperthreading machine and are experiencing reproducible hangs during -j > buildwords and other highly parallel operations, please try this patch: Just a follow-up to my off-list message and another data point, with this patch I no longer get deadlocks, however I now get random data corruption. The first time, it managed to complete the installworld and I didn't notice anything until trying to portupgrade Xorg failed. It turns out that the installed ranlib coredumped when run, but the one in /usr/obj worked fine. After that I was not able to successfully buildworld again. Every time one of the source files would inevitably end up corrupted with blocks of garbage, even after removing /usr/src entirely and re-cvsupping. Disabling the second processor or falling back to an older kernel (one from before the IPI hangs started) both fix the problem. Craig
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040826145324.GA40029>