Re: [Ql-Users] atari smsqe revisited
https://dl.dropboxusercontent.com/u/15146505/IMG_20150728_230655.jpg. SMSQ/E 3.22 on my STe. Michael On 30 July 2015 at 06:30, Wolf w...@wlenerz.com wrote: Oops, this ans(t for the list, sorry Wolfgang On 07/30/2015 05:41 AM, Wolf wrote: Hi, you can read / write bytes with the bget and bput commands. They shoud be documented in the TK2 manual. Have fun Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Nice! That means you were able to fix it? Great. Can’t wait to try that myself. Tobias Am 30.07.2015 um 10:00 schrieb Michael Grunditz michael.grund...@gmail.com: https://dl.dropboxusercontent.com/u/15146505/IMG_20150728_230655.jpg. SMSQ/E 3.22 on my STe. Michael On 30 July 2015 at 06:30, Wolf w...@wlenerz.com wrote: Oops, this ans(t for the list, sorry Wolfgang On 07/30/2015 05:41 AM, Wolf wrote: Hi, you can read / write bytes with the bget and bput commands. They shoud be documented in the TK2 manual. Have fun Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Hi, When using Hatari, beware that you don't get access to the hard disk. I dunno why, I **presume** that it's a timing question, perhaps the emulation does not really get to emulate a 68000 running at 8 Mhz. . One can see that smsqe is trying to access it, but you'll get a bad medium error. First I thought that the entire win code was somehow bad, but then Michael told me that this works OK on a real machine. This is probably also the reason why it takes so long for the cursor to appear after the windows are drawn - smsqe must be trying to find a boot file on the disk. Be patient. If the cursor doesn't appear at all, try to type something in. If that works, configure smsqe NOT to use a sprite as a cursor. Emulated floppy disks work OK, though. Have fun Wolfgang On 07/30/2015 05:11 PM, Ralf Reköndt wrote: Great, Congratulations. Again a reason to try Hatari. Cheers...Ralf - Original Message - From: Michael Grunditz Sent: Thursday, July 30, 2015 10:00 AM Subject: Re: [Ql-Users] atari smsqe revisited https://dl.dropboxusercontent.com/u/15146505/IMG_20150728_230655.jpg. SMSQ/E 3.22 on my STe. Michael ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Wolfgang fixed it, see the release mail! On 30 July 2015 at 10:26, Tobias Fröschle tobias.froesc...@t-online.de wrote: Nice! That means you were able to fix it? Great. Can’t wait to try that myself. Tobias Am 30.07.2015 um 10:00 schrieb Michael Grunditz michael.grund...@gmail.com: https://dl.dropboxusercontent.com/u/15146505/IMG_20150728_230655.jpg. SMSQ/E 3.22 on my STe. Michael On 30 July 2015 at 06:30, Wolf w...@wlenerz.com wrote: Oops, this ans(t for the list, sorry Wolfgang On 07/30/2015 05:41 AM, Wolf wrote: Hi, you can read / write bytes with the bget and bput commands. They shoud be documented in the TK2 manual. Have fun Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Hi, yeah I was lucky. There's still a little screen corruption when moving the pointer. Wolfgang Nice! That means you were able to fix it? Great. Can’t wait to try that myself. Tobias Am 30.07.2015 um 10:00 schrieb Michael Grunditz michael.grund...@gmail.com: https://dl.dropboxusercontent.com/u/15146505/IMG_20150728_230655.jpg. SMSQ/E 3.22 on my STe. Michael On 30 July 2015 at 06:30, Wolf w...@wlenerz.com wrote: Oops, this ans(t for the list, sorry Wolfgang On 07/30/2015 05:41 AM, Wolf wrote: Hi, you can read / write bytes with the bget and bput commands. They shoud be documented in the TK2 manual. Have fun Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Hi, you can read / write bytes with the bget and bput commands. They shoud be documented in the TK2 manual. Have fun Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
On 22 July 2015 at 10:19, Wolfgang Lenerz w...@wlenerz.com wrote: On 07/21/2015 10:03 PM, Michael Grunditz wrote: It enters the code without corrupting it 9 times. Fails after that. So which is the last(?) module? This should be smsq_atari_driver_dv3_asm. Could you check by setting a breakpoint at $12b40 and then follow the code after the last good break, to see whether it is indeed this module? You should find the name of the module a few bytes before the address at which it is called. I am home again and can do something serious with this. I tried to set a breakpoint at the rts after the copy and it seems like the last entry doesn't finish , i.e. rts is never reached. Can I (just for test) disable the 9th module? I know I've asked this before, but isn't there a special code one can use to cause a breakpoint in your debugger? (i.e. same A line instruction or something like that). I don't know. It should be something that causes a exception. I get exceptions from the hw probing of VME bus. So perhaps a read from impossible address can do the trick here. Marcel, I will du a attempt with serial ports if above fails. ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Hi, perhaps what you could do is modify dev8_smsq_atari_driver_dv3_asm by putting an rts at label d3r_init. That way that module will do nothing. Recompile everything and try to see what happens... Wolfgang On 07/23/2015 01:34 PM, Michael Grunditz wrote: On 22 July 2015 at 10:19, Wolfgang Lenerz w...@wlenerz.com wrote: On 07/21/2015 10:03 PM, Michael Grunditz wrote: It enters the code without corrupting it 9 times. Fails after that. So which is the last(?) module? This should be smsq_atari_driver_dv3_asm. Could you check by setting a breakpoint at $12b40 and then follow the code after the last good break, to see whether it is indeed this module? You should find the name of the module a few bytes before the address at which it is called. I am home again and can do something serious with this. I tried to set a breakpoint at the rts after the copy and it seems like the last entry doesn't finish , i.e. rts is never reached. Can I (just for test) disable the 9th module? I know I've asked this before, but isn't there a special code one can use to cause a breakpoint in your debugger? (i.e. same A line instruction or something like that). I don't know. It should be something that causes a exception. I get exceptions from the hw probing of VME bus. So perhaps a read from impossible address can do the trick here. Marcel, I will du a attempt with serial ports if above fails. ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
It is the debugger in the Hatari emulator. On 23 July 2015 at 15:10, Wolf w...@wlenerz.com wrote: Michael, what's the name of your debugger? Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
I think, I am completely lost in this thread. As far as I know, SMSQ.PRG (with SMSQ/E) only works either with the Extended Mode 4 emulator board in an Atari ST or with the QVME card in an STe or TT or just with the latter. I assumed, that only SMS2 uses the native screen on Atari. Am I wrong? Cheers...Ralf - Original Message - From: Michael Grunditz Sent: Thursday, July 23, 2015 3:20 PM Subject: Re: [Ql-Users] atari smsqe revisited It is the debugger in the Hatari emulator. On 23 July 2015 at 15:10, Wolf wrote: Michael, what's the name of your debugger? Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Aha, what I have thought. Though it was new for me, that there is a way to let it run on native screen resolution. Is that 640 x400? Cheers...Ralf - Original Message - From: Wolf Sent: Thursday, July 23, 2015 5:13 PM Subject: Re: [Ql-Users] atari smsqe revisited Hi, Ah, OK, thanks. I downloaded this hatari emulator. I see the following behaviour: If I use this with the tos that came with it (emultos), in STE mode (32 Mhz, 4 MB mem) and use mono as Atari monitor, then smsq/e works fine. If I use any other screen mode, it doesn't work, which is not surprising as Smsqe then seems to need a screen adapter (e.g. qvme). If now I use the mono monitor mode but with a real tos image (2.06) still in STE mode (32 Mhz, 4 MB mem),then this crashes somewhere. Is that the behaviour you also find, i.e. it works with emultos but not real tos? Wolfgang NB: How to set a breakpoint in the smsqe sources that will call the hatari debugger: insert the following sequence in the smsqe sources where you want the breakpoint to happen : move.l D0,-(A7) move.l #$12345678,d0 nop move.l (a7)+,d0 compile etc. Start the emulation and call up the debugger (alt gr + pause) then insert the follwing in the debugger: b d0 = $12345678 followed by c to continue the emulation. Start smsqe and the debugger will pop up as soon as d0 = $12345678 I can already tell you that the crash seem to happen before the smsq_atari_driver_dv3_asm module is called. Wolfgang On 07/23/2015 03:20 PM, Michael Grunditz wrote: It is the debugger in the Hatari emulator. On 23 July 2015 at 15:10, Wolf wrote: Michael, what's the name of your debugger? Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Hi, I can not see SMSQ/E working on the Hatari emulator. Does SMsQ/E intereact with the physical hardware, which a Hatari emulator does try to emulate, but not taht good. I think you need Atari Hardware to make this work correctly. I have my STFM with MOde 8 emulator, but I have to make a video cable for the CM8833-II monitor. Regards, Derek On 23/07/15 14:20, Michael Grunditz wrote:, It is the debugger in the Hatari emulator. On 23 July 2015 at 15:10, Wolf w...@wlenerz.com wrote: Michael, what's the name of your debugger? Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Derek, SMSQ.PRG 2.98 works well on my real STe and on Hatari STe config. Current SMSQE works on Hatari with EmuTOS. On 23 July 2015 at 17:38, Ralf Reköndt ralf.rekoe...@t-online.de wrote: Aha, what I have thought. Though it was new for me, that there is a way to let it run on native screen resolution. Is that 640 x400? Cheers...Ralf - Original Message - From: Wolf Sent: Thursday, July 23, 2015 5:13 PM Subject: Re: [Ql-Users] atari smsqe revisited Hi, Ah, OK, thanks. I downloaded this hatari emulator. I see the following behaviour: If I use this with the tos that came with it (emultos), in STE mode (32 Mhz, 4 MB mem) and use mono as Atari monitor, then smsq/e works fine. If I use any other screen mode, it doesn't work, which is not surprising as Smsqe then seems to need a screen adapter (e.g. qvme). If now I use the mono monitor mode but with a real tos image (2.06) still in STE mode (32 Mhz, 4 MB mem),then this crashes somewhere. Is that the behaviour you also find, i.e. it works with emultos but not real tos? Wolfgang NB: How to set a breakpoint in the smsqe sources that will call the hatari debugger: insert the following sequence in the smsqe sources where you want the breakpoint to happen : move.l D0,-(A7) move.l #$12345678,d0 nop move.l (a7)+,d0 compile etc. Start the emulation and call up the debugger (alt gr + pause) then insert the follwing in the debugger: b d0 = $12345678 followed by c to continue the emulation. Start smsqe and the debugger will pop up as soon as d0 = $12345678 I can already tell you that the crash seem to happen before the smsq_atari_driver_dv3_asm module is called. Wolfgang On 07/23/2015 03:20 PM, Michael Grunditz wrote: It is the debugger in the Hatari emulator. On 23 July 2015 at 15:10, Wolf wrote: Michael, what's the name of your debugger? Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Hi, Ah, OK, thanks. I downloaded this hatari emulator. I see the following behaviour: If I use this with the tos that came with it (emultos), in STE mode (32 Mhz, 4 MB mem) and use mono as Atari monitor, then smsq/e works fine. If I use any other screen mode, it doesn't work, which is not surprising as Smsqe then seems to need a screen adapter (e.g. qvme). If now I use the mono monitor mode but with a real tos image (2.06) still in STE mode (32 Mhz, 4 MB mem),then this crashes somewhere. Is that the behaviour you also find, i.e. it works with emultos but not real tos? Wolfgang NB: How to set a breakpoint in the smsqe sources that will call the hatari debugger: insert the following sequence in the smsqe sources where you want the breakpoint to happen : move.l D0,-(A7) move.l #$12345678,d0 nop move.l (a7)+,d0 compile etc. Start the emulation and call up the debugger (alt gr + pause) then insert the follwing in the debugger: b d0 = $12345678 followed by c to continue the emulation. Start smsqe and the debugger will pop up as soon as d0 = $12345678 I can already tell you that the crash seem to happen before the smsq_atari_driver_dv3_asm module is called. Wolfgang On 07/23/2015 03:20 PM, Michael Grunditz wrote: It is the debugger in the Hatari emulator. On 23 July 2015 at 15:10, Wolf w...@wlenerz.com wrote: Michael, what's the name of your debugger? Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Hi, I can not see SMSQ/E working on the Hatari emulator. Yes it does, in mono mode, under Linux. I've now sent a version to Michael that works, for me, under the emultos rom and also under tos 2.06. Hopefully that'll work also for Michael, and also on a real machine... I think, I am completely lost in this thread. As far as I know, SMSQ.PRG (with SMSQ/E) only works either with the Extended Mode 4 emulator board in an Atari ST or with the QVME card in an STe or TT or just with the latter. I assumed, that only SMS2 uses the native screen on Atari. Errr - I don't know. I can see that when I configure the atari smsqe file, I can choose several screen modes: mono, plus others which all seem to need an adapter, so I **presume** that colour modes all need an adapter... I have no idea what SMS2 supported. I'd be astonished, though, if it supported a screen mode which SMSQ/E does not... Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Yes, SMS2 runs on native ST in native 640x400 mono with the SM124 monochrome monitor. Cheers...Ralf - Original Message - From: Wolf Sent: Thursday, July 23, 2015 5:43 PM Subject: Re: [Ql-Users] atari smsqe revisited I have no idea what SMS2 supported. I'd be astonished, though, if it supported a screen mode which SMSQ/E does not... Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
On 07/21/2015 10:03 PM, Michael Grunditz wrote: It enters the code without corrupting it 9 times. Fails after that. So which is the last(?) module? This should be smsq_atari_driver_dv3_asm. Could you check by setting a breakpoint at $12b40 and then follow the code after the last good break, to see whether it is indeed this module? You should find the name of the module a few bytes before the address at which it is called. I know I've asked this before, but isn't there a special code one can use to cause a breakpoint in your debugger? (i.e. same A line instruction or something like that). Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
On 21 July 2015 at 21:56, Michael Grunditz michael.grund...@gmail.com wrote: On 21 July 2015 at 20:53, Wolf w...@wlenerz.com wrote: Hi, disasm $12b40 $012b40 : 6052 bra.s $12b94 $012b42 : 7a4b moveq #$4b,d5 $012b44 : 6002 bra.s $12b48 $012b46 : 7a41 moveq #$41,d5 $012b48 : b3ca cmpa.la2,a1 $012b4a : 6dfa blt.s $12b46 $012b4c : 4e75 rts $012b4e : 6012 bra.s $12b62 This is right after the string. Yes, and it's very wrong. (...) No. I can set a breakpoint at a address. Good, set it at $12b40 HEXdump: 05e0: 0003 5965 7300 2689 2649 4299 32d8 ..YesIB.2. 05f0: b3ca 6dfa 4e75 6012 46fc 2700 4e70 ..m.Nu`...F.'.Np 0600: 4cf8 0300 2e48 4ed1 2f07 6622 41fa L..HN./.fA. 0610: 053a 2010 671a 0cb0 514d 4f4e 0804 6708 .: .g...QMON..g. 0620: d0a8 0004 d1c0 60ea 41f0 0808 70ff 4e90 ..`.A...p.N. 0630: 2e1f 4eba 0306 1a3a fbda 0885 6604 ..N:..f. 0640: 4eba 03fa 8c05 1a3c 001e ca06 0c05 0008 N.. 32d8 b3ca 6dfa , the debugger reports 7a41 instead of 32d8 Now, this is the correct content, it corresponds to the SMSQ.PRG file. That is the SMSQ.PRG file :) I just pasted it for reference. What kind of machine or rather processor is this runnung on (could this be some kind of cache problem?). 68000 STe (tested with 1 and 4MB ram) If you set the breakpoint at $12b40, you will notice that this routine gets called quite a few times - it copies the individual modules.. You could check whether the code is still good each time it is called, so we could perhaps find out when the corruption occurs. Ok will do. It enters the code without corrupting it 9 times. Fails after that. So which is the last(?) module? ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Hi, disasm $12b40 $012b40 : 6052 bra.s $12b94 $012b42 : 7a4b moveq #$4b,d5 $012b44 : 6002 bra.s $12b48 $012b46 : 7a41 moveq #$41,d5 $012b48 : b3ca cmpa.la2,a1 $012b4a : 6dfa blt.s $12b46 $012b4c : 4e75 rts $012b4e : 6012 bra.s $12b62 This is right after the string. Yes, and it's very wrong. (...) No. I can set a breakpoint at a address. Good, set it at $12b40 HEXdump: 05e0: 0003 5965 7300 2689 2649 4299 32d8 ..YesIB.2. 05f0: b3ca 6dfa 4e75 6012 46fc 2700 4e70 ..m.Nu`...F.'.Np 0600: 4cf8 0300 2e48 4ed1 2f07 6622 41fa L..HN./.fA. 0610: 053a 2010 671a 0cb0 514d 4f4e 0804 6708 .: .g...QMON..g. 0620: d0a8 0004 d1c0 60ea 41f0 0808 70ff 4e90 ..`.A...p.N. 0630: 2e1f 4eba 0306 1a3a fbda 0885 6604 ..N:..f. 0640: 4eba 03fa 8c05 1a3c 001e ca06 0c05 0008 N.. 32d8 b3ca 6dfa , the debugger reports 7a41 instead of 32d8 Now, this is the correct content, it corresponds to the SMSQ.PRG file. What kind of machine or rather processor is this runnung on (could this be some kind of cache problem?). If you set the breakpoint at $12b40, you will notice that this routine gets called quite a few times - it copies the individual modules.. You could check whether the code is still good each time it is called, so we could perhaps find out when the corruption occurs. Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
On 21 July 2015 at 20:53, Wolf w...@wlenerz.com wrote: Hi, disasm $12b40 $012b40 : 6052 bra.s $12b94 $012b42 : 7a4b moveq #$4b,d5 $012b44 : 6002 bra.s $12b48 $012b46 : 7a41 moveq #$41,d5 $012b48 : b3ca cmpa.la2,a1 $012b4a : 6dfa blt.s $12b46 $012b4c : 4e75 rts $012b4e : 6012 bra.s $12b62 This is right after the string. Yes, and it's very wrong. (...) No. I can set a breakpoint at a address. Good, set it at $12b40 HEXdump: 05e0: 0003 5965 7300 2689 2649 4299 32d8 ..YesIB.2. 05f0: b3ca 6dfa 4e75 6012 46fc 2700 4e70 ..m.Nu`...F.'.Np 0600: 4cf8 0300 2e48 4ed1 2f07 6622 41fa L..HN./.fA. 0610: 053a 2010 671a 0cb0 514d 4f4e 0804 6708 .: .g...QMON..g. 0620: d0a8 0004 d1c0 60ea 41f0 0808 70ff 4e90 ..`.A...p.N. 0630: 2e1f 4eba 0306 1a3a fbda 0885 6604 ..N:..f. 0640: 4eba 03fa 8c05 1a3c 001e ca06 0c05 0008 N.. 32d8 b3ca 6dfa , the debugger reports 7a41 instead of 32d8 Now, this is the correct content, it corresponds to the SMSQ.PRG file. That is the SMSQ.PRG file :) I just pasted it for reference. What kind of machine or rather processor is this runnung on (could this be some kind of cache problem?). 68000 STe (tested with 1 and 4MB ram) If you set the breakpoint at $12b40, you will notice that this routine gets called quite a few times - it copies the individual modules.. You could check whether the code is still good each time it is called, so we could perhaps find out when the corruption occurs. Ok will do. ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
On 21/07/15 08:28, Marcel Kilgus wrote: Michael Grunditz wrote: This might be a good idea. Does it need input to work or can it be configured to perhaps break when something fails? It's been a decade since I've used it, but IIRC it breaks before the boot process where you can give it a g command to get it running again. Then it'll only break again after a fault. If the serial init works, is it possible to send information to the serial port directly (printouts ) and in that case is there some functions to do this? It expects a VT100 terminal emulator on the other end. Then it handles like a normal QMON debugger. Marcel ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm Hi, I have a Atari STFM Mode 8 Emulator, but not used it in a while. I will see if I can get it to work and try an upto date version of SMSQ/E. I also have a Mega ST and a extended 4 Emulator board, not fitted, maybe worth a look... Regards, Derek ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
On 21 July 2015 at 09:28, Marcel Kilgus ql-us...@mail.kilgus.net wrote: Michael Grunditz wrote: This might be a good idea. Does it need input to work or can it be configured to perhaps break when something fails? It's been a decade since I've used it, but IIRC it breaks before the boot process where you can give it a g command to get it running again. Then it'll only break again after a fault. If the serial init works, is it possible to send information to the serial port directly (printouts ) and in that case is there some functions to do this? It expects a VT100 terminal emulator on the other end. Then it handles like a normal QMON debugger. Ok. I don't have any problem with serial ports and terminal emulation. But before Thursday I can't do proper serial. However I can spool the serial port to a text file. So I wonder, can I send data to serialport just by moving data to the serial port address? I don't know which to use. I tried scc_data , but that did just bomb. ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Michael Grunditz wrote: But before Thursday I can't do proper serial. However I can spool the serial port to a text file. What for? So I wonder, can I send data to serialport just by moving data to the serial port address? I don't know which to use. I tried scc_data , but that did just bomb. Frankly, I have no idea about Atari hardware. But I don't understand what you're trying to do anyway, sorry. Marcel ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Hi, This is the loop: $012b46 : 7a41 moveq #$41,d5 $012b48 : b3ca cmpa.la2,a1 $012b4a : 6dfa blt.s $12b46 $012b4c : 4e75 rts registerdump: D0: 00016a4c D1: 0002d9de D2: 6a40 D3: 0032 D4: D5: 0041 D6: D7: 000f6b8c A0: 00029ae6 A1: 00012b48 A2: 00016a40 A3: fffc A4: 03ec A5: 03f0 A6: 00012b40 A7: 03cc USP=000f7ff2 ISP=03cc MSP= VBR= Ok, this is definitely wrong. I very strongly suspect that this is part of smsq_atari_hwinit_asm, which should however go : at_mdloop move.w (a0)+,(a1)+ (NOT moveq #$41,d5) cmpa.l a2,a1 blt.s at_mdlopp rts You can check this, hopefully, by checking whether a few bytes before these instructions you find the string Use sprite for cursor. If this is really there, then somehow the instructions get corrupted, making the move.w (a0)+,(a1)+ into a moveq. Could you disassemble the code as of the address contained in A6, which is a few bytes before the loop? Is there really no way you can insert a breakpoint into the code which would call up your debugger? Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
On 21 July 2015 at 12:27, Marcel Kilgus ql-us...@mail.kilgus.net wrote: Michael Grunditz wrote: But before Thursday I can't do proper serial. However I can spool the serial port to a text file. What for? So I wonder, can I send data to serialport just by moving data to the serial port address? I don't know which to use. I tried scc_data , but that did just bomb. Frankly, I have no idea about Atari hardware. But I don't understand what you're trying to do anyway, sorry. In lack of a proper source level debugger I would like to do printf like debugging. If it is possible to print to the serial port I can add loads of printouts to the modules / moduleloader. ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Michael Grunditz wrote: Frankly, I have no idea about Atari hardware. But I don't understand what you're trying to do anyway, sorry. In lack of a proper source level debugger I would like to do printf like debugging. If it is possible to print to the serial port I can add loads of printouts to the modules / moduleloader. Ah, I see. Well, this is how QMON accesses the hardware: io_setup lea io_mfptab,a0 ; set up mfp registers lea at_mfp,a1 moveq #io_mfpend-io_mfptab-1,d0 io_mfp move.b (a0)+,(a1) addq.l #2,a1 dbrad0,io_mfp rts io_mfptab dc.b0; no output data dc.bmfp.acte ; active edge dc.bmfp.ddir ; data direction dc.b0,0,0,0,0,0,0,0 ; no interrupts dc.b$68 ; software end of interrupt (should be automatic) dc.b0,0,1; no timers ABC and serial port D 9600 dc.b0,0,0,2 ; no timers ABC and serial port D 9600 io_mfpend ds.w0 io_setio moveq #$fff0,d0 and.b mfp_cdct+at_mfp,d0 addq.b #1,d0 move.b d0,mfp_cdct+at_mfp move.b #2,mfp_ddat+at_mfp move.b #mfp.npty+mfp.lstp+mfp.8bit+mfp.adiv,mfp_uctl+at_mfp ; set usart ; no parity, 1.5 stop bits, 8 bit async move.b #mfp.rxen,mfp_rstt+at_mfp ; receiver enabled move.b #mfp.txen,mfp_tstt+at_mfp ; transmitter enabled bclr#mfp..rxi,mfp_rxedisable interrupts bclr#mfp..txi,mfp_txe bclr#mfp..rxi,mfp_rxmand mask bclr#mfp..txi,mfp_txm rts * * send string * io_ssprt moveq #0,d1; set count ioss_loop tst.w d2 ; end of string? ble.s io_ok ioss_wait btst#mfp..txr,mfp_tstt+at_mfp; check status beq.s ioss_wait; ... wait btst#mfp..cts,mfp_cts; clear to send bne.s ioss_wait; ... no move.b (a1)+,mfp_sdat+at_mfp; write output subq.w #1,d2 addq.w #1,d1 bra.s ioss_loop; ... next io_ok moveq #0,d0 rts Marcel ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Michael Grunditz wrote: This might be a good idea. Does it need input to work or can it be configured to perhaps break when something fails? It's been a decade since I've used it, but IIRC it breaks before the boot process where you can give it a g command to get it running again. Then it'll only break again after a fault. If the serial init works, is it possible to send information to the serial port directly (printouts ) and in that case is there some functions to do this? It expects a VT100 terminal emulator on the other end. Then it handles like a normal QMON debugger. Marcel ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Ok, this is definitely wrong. I very strongly suspect that this is part of smsq_atari_hwinit_asm, which should however go : at_mdloop move.w (a0)+,(a1)+ (NOT moveq #$41,d5) cmpa.l a2,a1 blt.s at_mdlopp rts That's exactly what I thought about. You can check this, hopefully, by checking whether a few bytes before these instructions you find the string Use sprite for cursor. Yep I found that : disasm $12b40 $012b40 : 6052 bra.s $12b94 $012b42 : 7a4b moveq #$4b,d5 $012b44 : 6002 bra.s $12b48 $012b46 : 7a41 moveq #$41,d5 $012b48 : b3ca cmpa.la2,a1 $012b4a : 6dfa blt.s $12b46 $012b4c : 4e75 rts $012b4e : 6012 bra.s $12b62 This is right after the string. If this is really there, then somehow the instructions get corrupted, making the move.w (a0)+,(a1)+ into a moveq. Could you disassemble the code as of the address contained in A6, which is a few bytes before the loop? See above. Is there really no way you can insert a breakpoint into the code which would call up your debugger? No. I can set a breakpoint at a address. As I explained in the beginning of this thread it somehow relates to the TOS. The same SMSQ.PRG work with emutos but not with 1.62. I guess I can patch the binary. But if this (the bad moveq ) fails there are probably more places that gets wrong. ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
On 21 July 2015 at 13:13, Michael Grunditz michael.grund...@gmail.com wrote: Ok, this is definitely wrong. I very strongly suspect that this is part of smsq_atari_hwinit_asm, which should however go : at_mdloop move.w (a0)+,(a1)+ (NOT moveq #$41,d5) cmpa.l a2,a1 blt.s at_mdlopp rts That's exactly what I thought about. You can check this, hopefully, by checking whether a few bytes before these instructions you find the string Use sprite for cursor. Yep I found that : disasm $12b40 $012b40 : 6052 bra.s $12b94 $012b42 : 7a4b moveq #$4b,d5 $012b44 : 6002 bra.s $12b48 $012b46 : 7a41 moveq #$41,d5 $012b48 : b3ca cmpa.la2,a1 $012b4a : 6dfa blt.s $12b46 $012b4c : 4e75 rts $012b4e : 6012 bra.s $12b62 This is right after the string. If this is really there, then somehow the instructions get corrupted, making the move.w (a0)+,(a1)+ into a moveq. Could you disassemble the code as of the address contained in A6, which is a few bytes before the loop? See above. Is there really no way you can insert a breakpoint into the code which would call up your debugger? No. I can set a breakpoint at a address. As I explained in the beginning of this thread it somehow relates to the TOS. The same SMSQ.PRG work with emutos but not with 1.62. I guess I can patch the binary. But if this (the bad moveq ) fails there are probably more places that gets wrong. HEXdump: 05e0: 0003 5965 7300 2689 2649 4299 32d8 ..YesIB.2. 05f0: b3ca 6dfa 4e75 6012 46fc 2700 4e70 ..m.Nu`...F.'.Np 0600: 4cf8 0300 2e48 4ed1 2f07 6622 41fa L..HN./.fA. 0610: 053a 2010 671a 0cb0 514d 4f4e 0804 6708 .: .g...QMON..g. 0620: d0a8 0004 d1c0 60ea 41f0 0808 70ff 4e90 ..`.A...p.N. 0630: 2e1f 4eba 0306 1a3a fbda 0885 6604 ..N:..f. 0640: 4eba 03fa 8c05 1a3c 001e ca06 0c05 0008 N.. 32d8 b3ca 6dfa , the debugger reports 7a41 instead of 32d8 ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
I think I can answer the last question myself. It seems to build , but it fails on the last atari component. sys_boot_st_host_link. Is that the PRG creation? Is sys_boot_st_host supposed to exist , in that case it is missing. ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
-Original-Nachricht- Betreff: [Ql-Users] atari smsqe revisited Datum: Mon, 20 Jul 2015 16:09:40 +0200 Von: Michael Grunditz michael.grund...@gmail.com An: ql-us...@q-v-d.com Hi As you may remember I have been trying to get recent SMSQE to run on my 1024 STe. I have tried to run in it the hatari emulator and it does work when booted with EMUTOS, but not with the standard STe TOS. With the standard TOS I get exactly the same behaviour as on real STe, a garbled screen. It might be possible to load EMUTOS on the real STe , but I wonders what might cause this error. As I reported before a very old SMSQ.PRG runs just fine. I have been looking on the initialisation code in SMSQE. One thing that might go wrong is the detection of wether it is ST or STe. Another idea is that the old SMSQ.PRG doesn't have support for STe and that it is that support that is broken somehow. I will try to rebuild SMSQE without the auto detection and also trying to force ST mode even if it is a STe. Can I build the system with SMSQmulator or do I need QPC2? In that case I need to wait a few days :) Michael ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm Michael, Very much appreciate someone volunteers to look into supporting (kindof) neglected SMSQ/E hardware. QPC2 would probably be a tad faster. Other than that, I don't see why SMSQmulator would not be able to build SMSQ/E. And I would also bet that Wolfgang has already tried it ;) Tobias ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
On 20 July 2015 at 16:50, tobias.froesc...@t-online.de tobias.froesc...@t-online.de wrote: -Original-Nachricht- Betreff: [Ql-Users] atari smsqe revisited Datum: Mon, 20 Jul 2015 16:09:40 +0200 Von: Michael Grunditz michael.grund...@gmail.com An: ql-us...@q-v-d.com Hi As you may remember I have been trying to get recent SMSQE to run on my 1024 STe. I have tried to run in it the hatari emulator and it does work when booted with EMUTOS, but not with the standard STe TOS. With the standard TOS I get exactly the same behaviour as on real STe, a garbled screen. It might be possible to load EMUTOS on the real STe , but I wonders what might cause this error. As I reported before a very old SMSQ.PRG runs just fine. I have been looking on the initialisation code in SMSQE. One thing that might go wrong is the detection of wether it is ST or STe. Another idea is that the old SMSQ.PRG doesn't have support for STe and that it is that support that is broken somehow. I will try to rebuild SMSQE without the auto detection and also trying to force ST mode even if it is a STe. Can I build the system with SMSQmulator or do I need QPC2? In that case I need to wait a few days :) Michael ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm Michael, Very much appreciate someone volunteers to look into supporting (kindof) neglected SMSQ/E hardware. QPC2 would probably be a tad faster. Other than that, I don't see why SMSQmulator would not be able to build SMSQ/E. And I would also bet that Wolfgang has already tried it ;) Tobias My mistake , builds just fine , I just forgot to build the generic files :) ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Ralf, it used to work. STe has some special initialisation parameters. Anyway, The detection works from what I can see. If I force it to go in ST mode,SMSQ just bombs. I also tried to force mono monitor ,but that did not help at all. It seems like it fails in the final step when it tries to create the screen. Perhaps it uses a TOS call that is unsupported in this version of TOS. Investigation continues... On 20 July 2015 at 17:22, Ralf Reköndt ralf.rekoe...@t-online.de wrote: As far as I know, there is no different program for the three variations of Atari (ST, STe, TT). I have used the Level D and from TT a pre-assembled Level E (alpha) on both MegaST and TT. Works without problems. Ok, it was the time, when TT still was programming SMSQ. Cheers...Ralf - Original Message - From: Michael Grunditz michael.grund...@gmail.com To: ql-us...@q-v-d.com Sent: Monday, July 20, 2015 4:09 PM Subject: [Ql-Users] atari smsqe revisited Hi As you may remember I have been trying to get recent SMSQE to run on my 1024 STe. I have tried to run in it the hatari emulator and it does work when booted with EMUTOS, but not with the standard STe TOS. With the standard TOS I get exactly the same behaviour as on real STe, a garbled screen. It might be possible to load EMUTOS on the real STe , but I wonders what might cause this error. As I reported before a very old SMSQ.PRG runs just fine. I have been looking on the initialisation code in SMSQE. One thing that might go wrong is the detection of wether it is ST or STe. Another idea is that the old SMSQ.PRG doesn't have support for STe and that it is that support that is broken somehow. I will try to rebuild SMSQE without the auto detection and also trying to force ST mode even if it is a STe. Can I build the system with SMSQmulator or do I need QPC2? In that case I need to wait a few days :) Michael ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
By the way I would like to know if anybody here use the recent version of SMSQ/E on his Atari? I always thought, that the SuperGoldCard way is the better one. Load a JS and all the drivers, then SMSQ/E. SMSQE_GOLD is always loaded after JS or Minerva is run on the SuperGoldCard. Can't understand Tony's way on the Atari. He'd done it with the first QL-Emulator (just JS). Cheers...Ralf - Original Message - From: Michael Grunditz To: ql-us...@q-v-d.com Sent: Monday, July 20, 2015 5:47 PM Subject: Re: [Ql-Users] atari smsqe revisited Ralf, it used to work. STe has some special initialisation parameters. Anyway, The detection works from what I can see. If I force it to go in ST mode,SMSQ just bombs. I also tried to force mono monitor ,but that did not help at all. It seems like it fails in the final step when it tries to create the screen. Perhaps it uses a TOS call that is unsupported in this version of TOS. Investigation continues... On 20 July 2015 at 17:22, Ralf Reköndt ralf.rekoe...@t-online.de wrote: As far as I know, there is no different program for the three variations of Atari (ST, STe, TT). I have used the Level D and from TT a pre-assembled Level E (alpha) on both MegaST and TT. Works without problems. Ok, it was the time, when TT still was programming SMSQ. Cheers...Ralf - Original Message - From: Michael Grunditz michael.grund...@gmail.com To: ql-us...@q-v-d.com Sent: Monday, July 20, 2015 4:09 PM Subject: [Ql-Users] atari smsqe revisited Hi As you may remember I have been trying to get recent SMSQE to run on my 1024 STe. I have tried to run in it the hatari emulator and it does work when booted with EMUTOS, but not with the standard STe TOS. With the standard TOS I get exactly the same behaviour as on real STe, a garbled screen. It might be possible to load EMUTOS on the real STe , but I wonders what might cause this error. As I reported before a very old SMSQ.PRG runs just fine. I have been looking on the initialisation code in SMSQE. One thing that might go wrong is the detection of wether it is ST or STe. Another idea is that the old SMSQ.PRG doesn't have support for STe and that it is that support that is broken somehow. I will try to rebuild SMSQE without the auto detection and also trying to force ST mode even if it is a STe. Can I build the system with SMSQmulator or do I need QPC2? In that case I need to wait a few days :) Michael ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Is there documentation on how the kernel starts up? Ie where are the drivers? Does SMSQ/E always depend on another kernel? The _driver_ files seems most empty. For example where does it get the framebuffer (the screen memory)? Michael On 20 July 2015 at 17:54, Ralf Reköndt ralf.rekoe...@t-online.de wrote: Ok, I can't understand this but if you say so, it should be ok ;-). What is the latest version (D or E), which works on your STe? I have a lot of disks from TT, ok, not the recent SMSQ/E... but they all should work. Cheers...Ralf - Original Message - From: Michael Grunditz To: ql-us...@q-v-d.com Sent: Monday, July 20, 2015 5:47 PM Subject: Re: [Ql-Users] atari smsqe revisited Ralf, it used to work. STe has some special initialisation parameters. Anyway, The detection works from what I can see. If I force it to go in ST mode,SMSQ just bombs. I also tried to force mono monitor ,but that did not help at all. It seems like it fails in the final step when it tries to create the screen. Perhaps it uses a TOS call that is unsupported in this version of TOS. Investigation continues... On 20 July 2015 at 17:22, Ralf Reköndt wrote: As far as I know, there is no different program for the three variations of Atari (ST, STe, TT). I have used the Level D and from TT a pre-assembled Level E (alpha) on both MegaST and TT. Works without problems. Ok, it was the time, when TT still was programming SMSQ. Cheers...Ralf - Original Message - From: Michael Grunditz To: ql-us...@q-v-d.com Sent: Monday, July 20, 2015 4:09 PM Subject: [Ql-Users] atari smsqe revisited Hi As you may remember I have been trying to get recent SMSQE to run on my 1024 STe. I have tried to run in it the hatari emulator and it does work when booted with EMUTOS, but not with the standard STe TOS. With the standard TOS I get exactly the same behaviour as on real STe, a garbled screen. It might be possible to load EMUTOS on the real STe , but I wonders what might cause this error. As I reported before a very old SMSQ.PRG runs just fine. I have been looking on the initialisation code in SMSQE. One thing that might go wrong is the detection of wether it is ST or STe. Another idea is that the old SMSQ.PRG doesn't have support for STe and that it is that support that is broken somehow. I will try to rebuild SMSQE without the auto detection and also trying to force ST mode even if it is a STe. Can I build the system with SMSQmulator or do I need QPC2? In that case I need to wait a few days :) Michael ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
As far as I know, there is no different program for the three variations of Atari (ST, STe, TT). I have used the Level D and from TT a pre-assembled Level E (alpha) on both MegaST and TT. Works without problems. Ok, it was the time, when TT still was programming SMSQ. Cheers...Ralf - Original Message - From: Michael Grunditz michael.grund...@gmail.com To: ql-us...@q-v-d.com Sent: Monday, July 20, 2015 4:09 PM Subject: [Ql-Users] atari smsqe revisited Hi As you may remember I have been trying to get recent SMSQE to run on my 1024 STe. I have tried to run in it the hatari emulator and it does work when booted with EMUTOS, but not with the standard STe TOS. With the standard TOS I get exactly the same behaviour as on real STe, a garbled screen. It might be possible to load EMUTOS on the real STe , but I wonders what might cause this error. As I reported before a very old SMSQ.PRG runs just fine. I have been looking on the initialisation code in SMSQE. One thing that might go wrong is the detection of wether it is ST or STe. Another idea is that the old SMSQ.PRG doesn't have support for STe and that it is that support that is broken somehow. I will try to rebuild SMSQE without the auto detection and also trying to force ST mode even if it is a STe. Can I build the system with SMSQmulator or do I need QPC2? In that case I need to wait a few days :) Michael ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Michael Grunditz wrote: So somewhere in the loading process it crashes. If I break with the hatari debugger and disassemble I end up in a loop that doesn't end. I have no idea where the loop is in the source tree. There are multiple ways to find this, pretty much all involve noting a few bytes of code and looking them up in your binary. There you can work out in which module they are and then have a look in the sources for the module. But from your description I'm pretty sure that the loop you're looking at is not your problem but just a symptom. In many cases the error is fairly trivial but debugging this stuff is _hard_. I can probably provide the Atari QMON kernel debugger module, but getting it to work might of course be a quest in itself (communication is done through a serial port). Marcel ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
On 20 July 2015 at 22:44, Marcel Kilgus ql-us...@mail.kilgus.net wrote: Michael Grunditz wrote: So somewhere in the loading process it crashes. If I break with the hatari debugger and disassemble I end up in a loop that doesn't end. I have no idea where the loop is in the source tree. There are multiple ways to find this, pretty much all involve noting a few bytes of code and looking them up in your binary. There you can work out in which module they are and then have a look in the sources for the module. But from your description I'm pretty sure that the loop you're looking at is not your problem but just a symptom. In many cases the error is fairly trivial but debugging this stuff is _hard_. Figured out that it is rather a symptom.. I can probably provide the Atari QMON kernel debugger module, but getting it to work might of course be a quest in itself (communication is done through a serial port). This might be a good idea. Does it need input to work or can it be configured to perhaps break when something fails? If the serial init works, is it possible to send information to the serial port directly (printouts ) and in that case is there some functions to do this? ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Am 20.07.2015 um 19:28 schrieb Michael Grunditz michael.grund...@gmail.com: Is there documentation on how the kernel starts up? Ie where are the drivers? Does SMSQ/E always depend on another kernel? The _driver_ files seems most empty. For example where does it get the framebuffer (the screen memory)? Michael Very limited, http://www.dilwyn.me.uk/docs/smsqegd2/index.html http://www.dilwyn.me.uk/docs/smsqegd2/index.html the SMSQ Modules document describes a bit of the boot process. Apart from that, there’s only the sources afaik Tobias ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Is there a way to let the user generate a SMSQ.PRG for Atari? SMSQ/E is not only SGC or QPC2 or SMSQmulator. Atari was in the first place, so still has to be maintained. Cheers...Ralf - Original Message - From: Tobias Fröschle To: ql-us...@q-v-d.com Sent: Monday, July 20, 2015 9:18 PM Subject: Re: [Ql-Users] atari smsqe revisited Am 20.07.2015 um 19:28 schrieb Michael Grunditz Is there documentation on how the kernel starts up? Ie where are the drivers? Does SMSQ/E always depend on another kernel? The _driver_ files seems most empty. For example where does it get the framebuffer (the screen memory)? Michael Very limited, http://www.dilwyn.me.uk/docs/smsqegd2/index.html http://www.dilwyn.me.uk/docs/smsqegd2/index.html the SMSQ Modules document describes a bit of the boot process. Apart from that, there’s only the sources afaik Tobias ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Am 20.07.2015 um 21:23 schrieb Ralf Reköndt ralf.rekoe...@t-online.de: Is there a way to let the user generate a SMSQ.PRG for Atari? SMSQ/E is not only SGC or QPC2 or SMSQmulator. Atari was in the first place, so still has to be maintained. Cheers...Ralf That’s what Michael is trying to do - But he doesn’t end up with a working binary for the ST/E Tobias ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Very limited, http://www.dilwyn.me.uk/docs/smsqegd2/index.html http://www.dilwyn.me.uk/docs/smsqegd2/index.html the SMSQ Modules document describes a bit of the boot process. Apart from that, there’s only the sources afaik Ok thanks, The error is in the kernel (I guess in one of its modules). It fails after the jump to the modules (?).. When SMSQ starts as expected the garbled screen appears but becomes replaced with the QL screen after a while. So somewhere in the loading process it crashes. If I break with the hatari debugger and disassemble I end up in a loop that doesn't end. I have no idea where the loop is in the source tree. ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Ok, I can't understand this but if you say so, it should be ok ;-). What is the latest version (D or E), which works on your STe? I have a lot of disks from TT, ok, not the recent SMSQ/E... but they all should work. Cheers...Ralf - Original Message - From: Michael Grunditz To: ql-us...@q-v-d.com Sent: Monday, July 20, 2015 5:47 PM Subject: Re: [Ql-Users] atari smsqe revisited Ralf, it used to work. STe has some special initialisation parameters. Anyway, The detection works from what I can see. If I force it to go in ST mode,SMSQ just bombs. I also tried to force mono monitor ,but that did not help at all. It seems like it fails in the final step when it tries to create the screen. Perhaps it uses a TOS call that is unsupported in this version of TOS. Investigation continues... On 20 July 2015 at 17:22, Ralf Reköndt wrote: As far as I know, there is no different program for the three variations of Atari (ST, STe, TT). I have used the Level D and from TT a pre-assembled Level E (alpha) on both MegaST and TT. Works without problems. Ok, it was the time, when TT still was programming SMSQ. Cheers...Ralf - Original Message - From: Michael Grunditz To: ql-us...@q-v-d.com Sent: Monday, July 20, 2015 4:09 PM Subject: [Ql-Users] atari smsqe revisited Hi As you may remember I have been trying to get recent SMSQE to run on my 1024 STe. I have tried to run in it the hatari emulator and it does work when booted with EMUTOS, but not with the standard STe TOS. With the standard TOS I get exactly the same behaviour as on real STe, a garbled screen. It might be possible to load EMUTOS on the real STe , but I wonders what might cause this error. As I reported before a very old SMSQ.PRG runs just fine. I have been looking on the initialisation code in SMSQE. One thing that might go wrong is the detection of wether it is ST or STe. Another idea is that the old SMSQ.PRG doesn't have support for STe and that it is that support that is broken somehow. I will try to rebuild SMSQE without the auto detection and also trying to force ST mode even if it is a STe. Can I build the system with SMSQmulator or do I need QPC2? In that case I need to wait a few days :) Michael ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] atari smsqe revisited
Hi, So somewhere in the loading process it crashes. If I break with the hatari debugger and disassemble I end up in a loop that doesn't end. I have no idea where the loop is in the source tree. What kind of loop is it? Is it a branch to the same address? If yes, you might want to check the A7 stack for a return address, that could show you where this is called from. Does your debugger have a function where you can call it from the source code (something like My problem is that I don't have a working St any more... I can probably provide the Atari QMON kernel debugger module, but getting it to work might of course be a quest in itself (communication is done through a serial port). This might be a good idea. Does it need input to work or can it be configured to perhaps break when something fails? IIRC you connect it to a machine that runs qmon and that then lets you pilot the atari from there. Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
