Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 31 May 1998 00:05:13 -0700 (PDT)
From:      Julian Elischer <julian@whistle.com>
To:        Brian Feldman <brianfeldman@hotmail.com>
Cc:        freebsd-current@FreeBSD.ORG
Subject:   Re: fd crash (in isa_dmastart)
Message-ID:  <Pine.BSF.3.95.980530234338.10668D-100000@current1.whistle.com>
In-Reply-To: <19980531042432.16101.qmail@m2.findmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
I'm not sure what causes this but it should be easy enough to reproduce..
I'll add it to my list of things to clean up...

Interestingly enoungh the stack trace shows no softupdates related
functions.
I will guess however that the softupdate code may not fully appreciate the
subtlties of bounce buffers
as they are a FreeBSD oddity.
(I can actually think of a few things to check out.)
this may be a problem with aha1542 scsi systems too.



julian


On 31 May 1998, Brian Feldman wrote:

> Well, out of curiousity, I wanted to try SoftUpdates on a floppy. Let's just say that SoftUpdates on a floppy is a Bad Thing, causing Too Much Disk Access and A Bad Panic. (out out evil capitals!) Here's the bt, for those interested, and I'll delete this huge vmcore tomorrow if noone tells me not to and we don't want to try to get any more info out of it:
> (kgdb) bt
> #0  boot (howto=256) at ../../kern/kern_shutdown.c:281
> #1  0xf0118fc7 in panic (fmt=0xf01f1f56 "isa_dmastart: bad bounce buffer")
>     at ../../kern/kern_shutdown.c:421
> #2  0xf01f1fcf in isa_dmastart (flags=587333684, addr=0xf2d4d014 "ð\004", 
>     nbytes=4096, chan=2) at ../../i386/isa/isa.c:767
> #3  0xf01ec1fd in fdstate (fdcu=0, fdc=0xf02622d4) at ../../i386/isa/fd.c:1640
> #4  0xf01ebcb3 in fdintr (fdcu=0) at ../../i386/isa/fd.c:1445
> #5  0xf01ebc75 in fd_pseudointr (arg1=0x0) at ../../i386/isa/fd.c:1425
> #6  0xf011d203 in softclock () at ../../kern/kern_timeout.c:124
> #7  0xf01d6cc7 in doreti_swi ()
> Cannot access memory at address 0x274e35f8.
> 
> Now I'm relatively certain that this was specifically caused by way to much disk access on SoftUpdates' part; any idea how we should fix this?
> 
> Cheers,
> Brian Feldman
> 
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-current" in the body of the message
> 


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



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.95.980530234338.10668D-100000>