From owner-freebsd-current Tue Nov 21 01:20:06 1995 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id BAA28297 for current-outgoing; Tue, 21 Nov 1995 01:20:06 -0800 Received: from bunyip.cc.uq.oz.au (pp@bunyip.cc.uq.oz.au [130.102.2.1]) by freefall.freebsd.org (8.6.12/8.6.6) with SMTP id BAA28253 for ; Tue, 21 Nov 1995 01:19:57 -0800 Received: from cc.uq.oz.au by bunyip.cc.uq.oz.au id <21868-0@bunyip.cc.uq.oz.au>; Tue, 21 Nov 1995 19:19:33 +1000 Received: from netfl15a.devetir.qld.gov.au by pandora.devetir.qld.gov.au (8.6.10/DEVETIR-E0.3a) with ESMTP id TAA10822 for ; Tue, 21 Nov 1995 19:25:04 +1000 Received: from localhost by netfl15a.devetir.qld.gov.au (8.6.8.1/DEVETIR-0.1) id JAA01770; Tue, 21 Nov 1995 09:24:35 GMT Message-Id: <199511210924.JAA01770@netfl15a.devetir.qld.gov.au> X-Mailer: exmh version 1.6.4 10/10/95 To: current@freebsd.org cc: sysseh@devetir.qld.gov.au Subject: Re: QIC-02 tape drive problemette In-reply-to: Your message of "Tue, 21 Nov 1995 16:21:55 +1000." <199511210621.GAA00811@netfl15a.devetir.qld.gov.au> X-Face: 3}heU+2?b->-GSF-G4T4>jEB9~FR(V9lo&o>kAy=Pj&;oVOc<|pr%I/VSG"ZD32J>5gGC0N 7gj]^GI@M:LlqNd]|(2OxOxy@$6@/!,";-!OlucF^=jq8s57$%qXd/ieC8DhWmIy@J1AcnvSGV\|*! >Bvu7+0h4zCY^]{AxXKsDTlgA2m]fX$W@'8ev-Qi+-;%L'CcZ'NBL!@n?}q!M&Em3*eW7,093nOeV8 M)(u+6D;%B7j\XA/9j4!Gj~&jYzflG[#)E9sI&Xe9~y~Gn%fA7>F:YKr"Wx4cZU*6{^2ocZ!YyR Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 21 Nov 1995 19:24:35 +1000 From: Stephen Hocking Sender: owner-current@freebsd.org Precedence: bulk > > [Patches deleted] > > I've got one of these beasts and have noticed that removing a tape at critical > points in restore will reliably cause the machine to crash (While it's still > open, just prior to closing & rewinding it). It also calls tsleep from within > wtintr (actually nested withing a call to wtsense->wtpoll), a definite no-no, > but wtintr is sometimes called (aaggh!) from the non-interrupt portions of the > code to do some stuff, so one is never sure if it's legal to do the tsleep. > I managed to put in the patches, even though they needed a fair bit of massaging (although mt.c did not need patching) and also persuaded the code to do a DELAY instead of an tsleep when wtpoll was called from wtintr. This stopped the crashes but also caused the machine to lock up until the tape rewound. I'll look at reducing the sizes of the DELAY parameter and see if that makes any difference. Stephen -- I do not speak for the Worker's Compensation Board of Queensland - They don't pay me enough for that!