Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread tom ehlert via Freedos-devel
Hallo Herr Bernd Böckmann via Freedos-devel, am Freitag, 30. Mai 2025 um 19:17 schrieben Sie: > Hi Tom, >> after all this work, it's time for the EXTKEYB.EXE utility. >> >> It's quick and efficient, patent free and listed below >> >> >> #include >> >> main() >> { >> *(unsigned char fa

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread Bob Smith via Freedos-devel
On 5/30/2025 3:14 PM, tom ehlert via Freedos-devel wrote: Bob, It can be loaded from CONFIG.SYS. Perhaps the events you want to trap appear before CONFIG.SYS is processed, in which case this tool won't help. Otherwise, it can be very helpful. The events should happen before CONFIG.SYS. Luc

[Freedos-devel] 386SWAT

2025-05-30 Thread Bernd Böckmann via Freedos-devel
Hello Bob, > Sorry to intrude on this technical discussion, but an open source PM debugger > such as 386SWAT can trivially be set to detect reads and/or writes to any > location such as 40:96. Set it to detect writes, and it pops up exactly when > they happen. nothing to be sorry about :) Tha

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread tom ehlert via Freedos-devel
Bob, >> It can be loaded from CONFIG.SYS. Perhaps the events you want to trap appear >> before CONFIG.SYS is processed, in which case this tool won't help. >> Otherwise, it can be very helpful. The events should happen before CONFIG.SYS. Luckily S-ICE can be loaded before CONFIG.SYS. Unfortuna

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread tom ehlert via Freedos-devel
>> after all this work, it's time for the EXTKEYB.EXE utility. >> >> It's quick and efficient, patent free and listed below >> >> >> #include >> >> main() >> { >> *(unsigned char far *)MK_FP(0x40,0x96) |= 0x10; >> } >> >> to be run on buggy machines, before mKEYB runs. >> > thanks!

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread tom ehlert via Freedos-devel
Bob, > On 5/30/2025 4:58 AM, Bernd Böckmann via Freedos-devel wrote: >> I finally got my hands on a machine that shows the symptom of BDA 40:96 >> getting changed, and I was able to identify the problem. >> > When the PC starts, BDA 40:96 is initially zero. As soon as the users >> > enters some

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread Bob Smith via Freedos-devel
On 5/30/2025 4:58 AM, Bernd Böckmann via Freedos-devel wrote: I finally got my hands on a machine that shows the symptom of BDA 40:96 getting changed, and I was able to identify the problem. When the PC starts, BDA 40:96 is initially zero. As soon as the users enters some specific keys (like t

Re: [Freedos-devel] Kernel package question

2025-05-30 Thread Bernd Böckmann via Freedos-devel
So as there were no further responses to this thread I renamed the unstable branch on Gitlab to preview. https://gitlab.com/FreeDOS/base/kernel/-/tree/preview As a result, this will not be in T2505, but everyone interested may download the binaries from the above link. Maybe we manage to includ

Re: [Freedos-devel] Will FreeDOS work to reflash LSI SCSI SAS cards like 9300 series?

2025-05-30 Thread Louis Santillan via Freedos-devel
I've used the DOS versions of sas2flash and sas3flash with FreeDOS. I think what I did was: * take the FreeDOS 1.2 or 1.3 USB image * strip out autoexec.bat (to keep it from attempting to install FreeDOS) * delete a bunch of packages until I had about 64MB or so free for the ROM/FW images, sas/lsi

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread Bernd Böckmann via Freedos-devel
Hi Tom, > after all this work, it's time for the EXTKEYB.EXE utility. > > It's quick and efficient, patent free and listed below > > > #include > > main() > { > *(unsigned char far *)MK_FP(0x40,0x96) |= 0x10; > } > > to be run on buggy machines, before mKEYB runs. > thanks! May I add

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread tom ehlert via Freedos-devel
Hi Bernd, after all this work, it's time for the EXTKEYB.EXE utility. It's quick and efficient, patent free and listed below #include main() { *(unsigned char far *)MK_FP(0x40,0x96) |= 0x10; } to be run on buggy machines, before mKEYB runs. Tom ___

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread Bernd Böckmann via Freedos-devel
Hi Eric > Am 30.05.2025 um 14:34 schrieb Eric Auer via Freedos-devel > : > > > /* BIOS keyboard routines with 84 and 102 key keyboard support */ > int Xbioskey(int cmd) > { >static int keybase = -1; >union REGS kregs; >if (keybase < 0) { >volatile char far *kbtype

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread Eric Auer via Freedos-devel
Hi! As this is in context of EDIT, here are some EDIT 0.7d snippets :-) In message.c function collect_events() we have: - refresh the clock display, but only once per second, and in a COUNTRY setting sensitive way :-) - compatibility-friendly BIOS keyboard handling: if ((sk = getsh

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread Joe Forster/STA via Freedos-devel
Hi Bernd, Update: I found the code in your Star Commander source. It seems to be an altered version of the TurboVision shipped with Borland Pascal 7. Did this alterations come from Borland directly, or from some other source? I'm sorry for the misinformation; you're right, it's NOT original

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread Bernd Böckmann via Freedos-devel
> Am 30.05.2025 um 12:56 schrieb Bernd Böckmann via Freedos-devel > : > > does speak against it (AH=0 on return) Have to correct myself: 11H-2 should be AH=0FH :) ___ Freedos-devel mailing list Freedos-devel@lists.sourceforge.net https://lists.sour

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread Bernd Böckmann via Freedos-devel
@Jim: Update: I found the code in your Star Commander source. It seems to be an altered version of the TurboVision shipped with Borland Pascal 7. Did this alterations come from Borland directly, or from some other source? I consider the implemented test to be flawed. I think this will fail on a

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread Bernd Böckmann via Freedos-devel
Hi Joe, > In Turbo Pascal's Turbo Vision, extended keyboards are detected once, upon > program startup, like this: > >mov ax, 11ffh >int 16h >xor bl, bl >cmp ax, 11ffh >je @@1 >mov bl, 10h > @@1:mov ReadKeyFunc, bl > > and then keyboard functions are called like this: >

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread Joe Forster/STA via Freedos-devel
Hi Bernd, When the PC starts, BDA 40:96 is initially zero. As soon as the users enters some specific keys (like the arrow keys), the value at 40:96 changes to 0x10. So the BIOS seems to do some keyboard auto detection. In Turbo Pascal's Turbo Vision, extended keyboards are detected once, upo

Re: [Freedos-devel] Extended keyboard flag at 0040:0096 gets unexpectedly cleared

2025-05-30 Thread Bernd Böckmann via Freedos-devel
I finally got my hands on a machine that shows the symptom of BDA 40:96 getting changed, and I was able to identify the problem. When the PC starts, BDA 40:96 is initially zero. As soon as the users enters some specific keys (like the arrow keys), the value at 40:96 changes to 0x10. So the BIOS