On Mon, 9 Jun 2003, Bart Oldeman wrote:
> On Mon, 9 Jun 2003, Daniel Greenberg wrote: > > > It turns out the divide-by-zero problem isn't "really new," just new. > > That is, 1.1.4.15 has the problem, but 1.1.4.0 is fine. I picked > > something in the middle - 1.1.4.8 - and found that it too had the "new" > > divide-by-zero errors. A couple more downloads and compiles and I > > was able to isolate the introduction of the problem to 1.1.4.7. That is, > > 1.1.4.6 runs without the "new" divide-by-zero errors, while > > 1.1.4.7 runs with the errors. So apparently 1.1.4.7 was the first to > > introduce this problem. > > Thanks, that narrows it down a bit. I suspect that the initialization > (which was shuffled around a little in 1.1.4.7) is the culprit, but I > fail to see where. > > Can you make hexdumps of the area from 0:0 to 0:500 for both DOSEMU's > (using DEBUG in DOS or using dosdebug with the "d" command) and see if > there are any differences, beyond the timer (4 bytes at 0:46c) and the > keyboard buffer (0:41a- - 0:43d)? > > Bart > > - > To unsubscribe from this list: send the line "unsubscribe linux-msdos" in > the body of a message to [EMAIL PROTECTED] > More majordomo info at http://vger.kernel.org/majordomo-info.html > Comparing 1.1.4 & 1.1.5 the only difference I found was at 0:01C. This is in the interrupt vector table, no? Not sure the significance of this though. Otherwise - with the exception of the timer and keyboad buffer - they are identical. Dan - To unsubscribe from this list: send the line "unsubscribe linux-msdos" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
