Date: Wed, 6 Nov 2002 20:48:14 -0500 (EST) From: Jeff Roberson <jroberson@chesapeake.net> To: Andrew Gallatin <gallatin@cs.duke.edu> Cc: mjacob@feral.com, <alpha@FreeBSD.ORG>, <alc@FreeBSD.ORG>, John Baldwin <jhb@FreeBSD.ORG> Subject: Re: alpha: top of tree kernel blooie Message-ID: <20021106204331.F17759-100000@mail.chesapeake.net> In-Reply-To: <15817.44172.809351.382956@grasshopper.cs.duke.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 6 Nov 2002, Andrew Gallatin wrote: > > alpha/include/atomic.h rev 1.17 (try 1.16) > alpha/alpha/atomic.s rev 1.6 (try 1.5) > > John thinks the expensive extra memory barriers in the atomic ops may > have masked existing pmap problems. I've been running the patch for > 2 months on a UP config and not seen any problems. I did once see a > strange Heisenbug when using an SMP kernel on a UP machine, but, > again, not sure if that's related to the atomic changes. Once I put > in a ktr to trace the trap, the bug disappeared. > > Since you're using the K0SEG, I"m not sure how pmap comes into play. > > Drew > Well, I have actually seen this panic one other time on a SMP system. Although this was a two way and not four. I have single proc alphas that get almost as much usage and have never had this panic. Unfortunately, I too have only seen this once. This was way before the K0SEG changes as well. I bring up pmap because it was already somewhat suspect and we're panicing through some vm weirdness. There is definitely some bug here that is alpha specific. Hopefully when I get my 4way system up again I can reproduce it and get some more insight into the problem. Cheers, Jeff To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20021106204331.F17759-100000>