On Fri, 3 Feb 2006 14:02:46 +0000, George Gwilt wrote:
>
> On 1 Feb 2006, at 17:46, Thierry Godefroy wrote:
>
> > IIRC, this is an issue with QMON using the VBR... I got a patch,
> > somewhere...
> > Will check this week-end.
> >
> > Thierry.
>
> For use on Q40/60 and my QXL which has an FPU QMON's instructions to
> zero the VBR have been omitted.
>
> George
> _______________________________________________
> QL-Users Mailing List
> http://www.q-v-d.demon.co.uk/smsqe.htm
I found the archives... It's a problem with George's FPU extension using
the VBR, and QMON resetting it to 0... Here is the message I sent three
years ago to Simon Goodwin:
Date: Sun Jan 12 21:44:43 1997
From: Thierry Godefroy <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Cc:
Subject: Re: FPUFns
.../...
I made a more serious check about this on my own version and it does reset
the VBR to zero too ;-(
The offending instructions are:
MOVEQ #0,D0
MOVEC D0,VBR (opcode $4E7B0801)
RTS
So I just replaced the MOVEC with two NOPs, et voila ! I do not think that
QMON will be armed by this patch but as I did not make many testing with the
patched version, I can't guarantee this. Here is a small patching program:
100 QM$="win5_lm_Qmon_bin"
110 COPY QM$ TO QM$&"_orig":REMark Saves an original QMON copy !
120 L=FLEN(\QM$):AD=ALCHP(L):LBYTES QM$,AD
140 FOR I=0 TO L STEP 2
150 IF HEX$(PEEK_W(AD+I),16)="4E7B" AND HEX$(PEEK_W(AD+I+2),12)="801"
160 POKE_L AD+I,HEX("4E714E71"):PRINT "Patched !":EXIT I
170 END IF
180 END FOR I
190 SBYTES_O QM$,AD,L:RECHP AD
QDOS / SMS forever !
Thierry.
_______________________________________________
QL-Users Mailing List
http://www.q-v-d.demon.co.uk/smsqe.htm