From owner-freebsd-alpha Tue Apr 13 10:59:26 1999 Delivered-To: freebsd-alpha@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [209.157.86.2]) by hub.freebsd.org (Postfix) with ESMTP id D263D15890 for ; Tue, 13 Apr 1999 10:59:24 -0700 (PDT) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id KAA18655; Tue, 13 Apr 1999 10:57:03 -0700 (PDT) (envelope-from dillon) Date: Tue, 13 Apr 1999 10:57:03 -0700 (PDT) From: Matthew Dillon Message-Id: <199904131757.KAA18655@apollo.backplane.com> To: Matthew Jacob Cc: alpha@FreeBSD.ORG Subject: Re: Hmm... still NFS problems for Alpha? References: Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I missed one. Also try #if 0'ing out the pmap_zero_page_area() call in vm_page_zero_invalid(), also in vm/vm_page.c. I'm working on a more permanent fix but it is not trivial enough to post at the moment. Basically, the problem is that NFS's piecemeal writes confuse the VM system. The new code that zeros invalid portions of buffers, that the temporary patch #if 0's out, is stomping on pieces of the NFS system. The bug is in NFS, though, not the new code. -Matt Matthew Dillon :*after* the VM patch, I still get: : : make depend && make :cc -c -nostdinc -I- -I. -I../.. -I/usr/include -DKERNEL -include :opt_global.h -UKERNEL ../../alpha/alpha/genassym.c :cc -static -nostdinc -I- -I. -I../.. -I/usr/include -DKERNEL -include :opt_global.h genassym.o -o genassym :./genassym >assym.s :Illegal instruction - core dumped :*** Error code 132 : :Stop. :farrago.feral.com > uname -a :FreeBSD farrago.feral.com 4.0-CURRENT FreeBSD 4.0-CURRENT #0: Mon Apr 12 :23:02:43 PDT 1999 mjacob@farrago.feral.com:/tmp/sys/compile/FARRAGO :alpha : :(same source builds in local or mfs based filesystems fine) : : :Anybody else still having NFS problems with alpha? : : : : To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message