Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 13 Aug 2000 06:00:02 -0700 (PDT)
From:      Bruce Evans <bde@zeta.org.au>
To:        freebsd-doc@freebsd.org
Subject:   Re: kern/17774: stray irq7 
Message-ID:  <200008131300.GAA30420@freefall.freebsd.org>

next in thread | raw e-mail | index | archive | help

The following reply was made to PR kern/17774; it has been noted by GNATS.

From: Bruce Evans <bde@zeta.org.au>
To: Johan Karlsson <k@numeri.campus.luth.se>
Cc: freebsd-gnats-submit@FreeBSD.ORG
Subject: Re: kern/17774: stray irq7 
Date: Sun, 13 Aug 2000 22:52:53 +1000 (EST)

 On Sun, 13 Aug 2000, Johan Karlsson wrote:
 
 > At Sun, 13 Aug 2000 08:55:04 +1000, Bruce Evans wrote:
 > > On Fri, 11 Aug 2000, Johan Karlsson wrote:
 > > 
 > > >  I suggest the following is added to the faq
 > > 
 > > The fwa? :-).
 > 
 > ???
 
 Frequently Wrong Answers.
 
 > > problem by breaking detection of it.  The correct fix is to detect
 > > the stray interrupts caused by interrupt timing glitches and decide
 > > what to do about them (whatever is done, it shouldn't involve
 > > printing a faq magnet).
 > 
 > Ok
 > 
 > Should the print of this text be removed from kernel then?
 
 Either that or turned into a KASSERT panic or something that annoys
 developers enough to fix the problem.  (The irq7 and irq15 handlers
 need extra code which is hard to fit into the macros used to generate
 the code.  This may be easier when the code is rewritten for SMPng.)
 
 > Is the following suggestion for the FAQ better or should we not
 > have an FAQ entry at all for this (which I think we should since 
 > the question pops up)?
 > 
 > ============
 > Q: What does 'stray irq' mean?
 > A: Stray IRQs are indications of hardware IRQ glitches, mostly 
 > from hardware that removes its interrupt request in the middle 
 > of the interrupt request acknowledge cycle.
 > J Wunsch writes in a response to a PR 
 > "Stray IRQs are a known phenomenon.  Obviously (if you think about it :), 
 > there's nothing the kernel could do about it. Unless you have
 > misconfigured your kernel so there's no driver assigned to a device
 > that actually issues IRQs, they are a sign of flakey hardware, often
 > caused by glitches on an IRQ line."
 >        
 > You have three choices:
 >  1) Live with the warnings 
 >  2) Get a driver for the hardware into your kernel
 >  3) Remove the hardware that generates the interrupts
 > =============
 
 This is very confusing, since it gives conflicting answers.
 
 The only reasonable choices are:
 
 1) Live with the warnings.  All except the first 5 per irq are suppressed
    anyway.
 2) Break the warnings by changing 5 to 0 in isa_strayintr() so that all the
    warnings are suppressed.
 3) Break the warnings by installing parallel port hardware that uses irq 7
    and the ppp driver for it (this happens on most systems), and install
    an ide drive or other hardware that uses irq 15 and a suitable driver for
    it.
 
 Stray irqs on irqs other than irq7 and irq15 cause very frequently asked
 questions since they "can't happen".
 
 Bruce
 
 


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-doc" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200008131300.GAA30420>