[4561] in linux-net channel archive
Re: seeing crashes? read this... YUP... on intel also
daemon@ATHENA.MIT.EDU (babydr)
Sat Sep 28 09:51:07 1996
Date: Fri, 27 Sep 1996 22:25:41 -0700 (PDT)
From: babydr <babydr@nwrain.net>
Reply-To: babydr <babydr@nwrain.net>
To: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: "Linux Network Info. List" <linux-net@vger.rutgers.edu>
In-Reply-To: <m0v5ecG-0005aNC@lightning.swansea.linux.org.uk>
> > has made no difference the 'divide error: 0000' still persists
> > at the location (from /System.map)
> > 0010ec3c t do_fast_gettimeoffset
>
> Go to the bottom of arch/i386/kernel/time.c and you will see a piece
> of code checking x86_capability&16 - just change that so it always fails
> ie to if(0)
>
> Or if you dont want to hack code, enable APM support
Oh Heck... why not. I'll just get out the scalpel.. ;-)
Ok, This message will be postponed until 26-Sep-1996 or later
Just to see how bad the damages...
Well since the evening of the 24th there hasn't been a problem
of the type I had mentioned.
Although do_fast_timeoffset is still in the /System.map ?
I have been doing a little digging around in the src tree
for away to basically ask at 'make (x/menu)config' time
if the individual would like to disable 'do_fast_timeoffset'
for his AMD5x86 CPU .
But, then Ken said (below, And what I sent him)
>Ken Jordan <kenjordan@massmedia.com>
>> Hello again,
>>
>> I just saw Alan's suggestion about modifing the capibilities check but
>> in my examinations it seems that it should already fail (on the
>> AMD5x86-P75).
>
>> I would appreciate it if you could send me your "/proc/cpuinfo". This
>> should show what capibilities Linux thinks your CPU has. Apparently
>> Linux thinks you have the Pentuim cycle clock instructions and I'm a
>> bit curious why.
>
> Ken, Here you go..
>
>processor : 0
>cpu : 586
>model : 0
>vendor_id : AuthenticAMD
>stepping : unknown
>fdiv_bug : no
>hlt_bug : no
>fpu : yes
>fpu_exception : yes
>cpuid : yes
>wp : yes
>flags : fpu vme de pse tsc msr mce cx8 apic
>bogomips : 179.81
