On Fri, Sep 24 2004 - 06:01, Sean Davis wrote:
> On Thu, 23 Sep 2004 22:00:18 +0100, Steve Woodford <scw@netbsd.org> wrote:
> > 
> > 
> > On Thursday 23 September 2004 08:43, matthew green wrote:
> > > my ultra10/440 experienced the sleep forever bug twice a couple of
> > > weeks ago.  it happens on any sparc64 box.  it seems to be more
> > > likely to occur if the disk and network are busy (eg, both the
> > > above happened when i started writing and read heavily over NFS
> > > at the same time.)
> > 
> > Just on a whim, can someone try out the attached patch to
> > sys/arch/sparc64/include/psl.h to see if it cures the sleep forever
> > bug?
> > 
> > There's a chance that gcc is reordering instructions around some spl*
> > calls. The patch should address this.
> 
> Well, the problem is, I haven't found a way to trigger the bug. Other
> people have stated that heavy network/disk IO seem to cause it, but
> yesterday I threw a gig and a half over my network to my Ultra 5, over
> FTP then over NFS, and it handled it just fine (although NFS was about
> 25% slower than FTP, obviously...)
> 

my U10 "died" (at about 9pm, according the the last cron logs) while
doing a whole build.sh in /usr/src. I don't know is rebuilding the whole
stuff can generate that heavey IO...

> Once I get my main workstation back up (its AGP slot (or video card)
> died yesterday when putting in a faster cpu) I'll be happy to do more
> testing, but for now the only machines I have running are my Ultra 1
> which is doing NAT, and my mother's dell windows XP box. (I've got a
> sparcbook 3gx going in my room, but 10baseT/HDX is hardly enough for
> testing network speeds on a 100mbit lan...)
> 

-- 
,-- This mail runs ---------.
`------------ NetBSD/i386 --'