Message-Id:	<199603071117.GAA13791@delorie.com>
Date:	Thu, 07 Mar 96 12:34:32 LIT
From:	Martynas Kunigelis <martynas DOT kunigelis AT VM DOT KTU DOT LT>
To:	J DOT Rouwe AT cpedu DOT RUG DOT NL, djgpp AT delorie DOT com
In-Reply-To:	Your message of Wed, 06 Mar 1996 02:11:54 +0100

On Wed, 06 Mar 1996 02:11:54 +0100 J. Rouwe wrote:
>
>...<skipped>...             I've read somewhere that you should also
>acknoledge the keyboard controller (but I don't remember where because the
>old code is already 2 years old), so perhaps some keyboards need this
>acknoledge (mine doesn't seem to care). ...<skipped>...
>Maybe this is your problem....

 BINGO! I'm sure it is| Actualy I put the 'acknowledgement' code into the
initial version, but my keyboard didn't seem to care also, so I thought only
very old PC's would need it. So I removed it, because I thought it might
confuse people. But my friend just told me, there are still a lot of keyboards
that need the acknowledgement, especially ones that come with older models
( the bug was reported by person with 386 ). J. Rouwe also discovered
another bug, something like 'movw variable, %edx', but I think it should only
hurt when closing the handler, not when it is active. I'll correct that,
anyway. Now, I think people are sick of two-line patches, so I think I'll
modify the code, test it on more PC's, and post it to DJ again, so he can
upload it once more. That would bring less confusion than posting you those
nasty little patches. I also received a crash report with QDPMI. I'm not sure
why is that, but I think maybe it'll work when existing bugs are corrected. If
not, I'll see what could be wrong. Looks like I'll have to install QEMM...


 that's it for now.. bye (thanks for help, everybody!)

Martynas Kunigelis

- Raw text -