Date: Mon, 9 Oct 2000 16:34:34 -0500 (CDT) From: Dave Uhring <duhring@charter.net> To: Gary Kline <kline@thought.org> Cc: Gerhard Sittig <Gerhard.Sittig@gmx.net>, freebsd-stable@FreeBSD.ORG Subject: Re: the ``stray irq 7'' is back Message-ID: <Pine.NEB.4.21.0010091622100.259-100000@dave.uhring.com> In-Reply-To: <20001009140940.A17746@tao.thought.org>
next in thread | previous in thread | raw e-mail | index | archive | help
The usual reason for getting these spurious IRQ's is that the equipment is not properly connected, leaving the Centronics ACK pin floating. A floating input can take just about any voltage level, depending on ambient EMI noise. When the pin voltage floats (damn if I remember the exact logic level anymore) the IRQ is triggered. Same thing happened with a firewall box I had on IRQ 15 with no secondary IDE drives attached. The real fun with these spurious IRQ's happens when you are doing a humongous download and one packet out of 5000 gets corrupted by an overflow on your NIC's buffer because the CPU was busy dealing with the IRQ. The NIC drivers apparently do not check for underflow/overflow conditions. Then when you attempt to unzip your tarball, you get the message 'header not found...' I had that happen when I was trying to get the StarOffice-5.1a package from Sun - 69MB shot to hell. Dave On Mon, 9 Oct 2000, Gary Kline wrote: > Date: Mon, 9 Oct 2000 14:09:40 -0700 > From: Gary Kline <kline@thought.org> > To: Gerhard Sittig <Gerhard.Sittig@gmx.net> > Cc: freebsd-stable@FreeBSD.ORG > Subject: Re: the ``stray irq 7'' is back > > On Mon, Oct 09, 2000 at 07:53:42PM +0200, Gerhard Sittig wrote: > > On Sun, Oct 08, 2000 at 21:31 -0700, Doug White wrote: > > > > > > Gary, your motherboard is cursed. I have one too; live with it. > > > > > > Stray IRQ 7s are _normal_. > > > > Jumping in late, but ... > > > > I know "stray irqs" from when hardware is "disabled" or you have > > hardware in the machine the OS doesn't support and thus doesn't > > register drivers or resources for. That's when the events end up > > in the "stray" (i.e. "unexpected since not caused by me") > > handler. Sound cards were known to cause this some five to eight > > years ago. Some irq7 got triggered "by chance" without the lpt > > driver thinking it has done it or some such. And I've seem > > machines with crackling sound in the speaker boxes whenever print > > jobs via lpt were executed. It must have been some kind of cross > > over noise, maybe not enough or defective separation of > > electrical paths or something. > > > > After cleaning both ends of the cable, it's possible that > one end or the other (or both) weren't reconnected > correctly; thanks for this information. When something as > mundane as this suddenly quits after several years, > I'm ready to believe anything! > > gary > > > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.NEB.4.21.0010091622100.259-100000>