 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : CMM2: Bug reports
  Page 6 of 17   |
|||||
| Author | Message | ||||
KeepIS Guru  Joined: 13/10/2014 Location: AustraliaPosts: 1835 |
I've been editing since my last post, no errors. I just confirmed your test above as I had already found the answer to that, but tried it again now and - the dam editor would not paste, and when I looked at the last line in the edit window it had an ESC sequence.  When you mark a line of text by moving down past the last text line, you are also copying the CRLF at the end of the last line of text. If you put a space at the end of the last word on the line and only highlight to the last character, you don't get an extra CR inserted in the paste. Mike. EDIT: I've been using Ctrl-W a lot in the last edit session, no problems here. Could be a keyboard sticking or that auto repeat hick-up has returned? .. Edited 2020-06-12 12:20 by KeepIS NANO Inverter: Full download - Only Hex Ver 8.1Ks |
||||
Chopperp Guru Joined: 03/01/2018 Location: AustraliaPosts: 1090 |
Thanks Jim. I'll keep that in mind when I go reading the manual again. Brian ChopperP |
||||
vegipete Guru Joined: 29/01/2013 Location: CanadaPosts: 1122 |
Auto-repeat thing is still happening, but _only_ the [ENTER] key after the CTRL-W save file in the editor. Pressing any key, including [ENTER] makes it stop. Interesting, the effect persists after pressing the reset button on the waveshare board. And the effect remains after power cycling the whole system. Both [ENTER] keys do the same. But CTRL-M does not! Microsoft Wired Keyboard 400, for what it's worth. Visit Vegipete's *Mite Library for cool programs. |
||||
| robert.rozee Guru Joined: 31/12/2012 Location: New ZealandPosts: 2401 |
well, this is embarrassing... using gw-basic 3.23, running under DosBox: 110 For i=0 To 1 120 Print "example " i+1 ":" 130 If i Then Print ,, "green eggs ";: Print "on ham" 140 Next i 150 ' 160 For i=0 To 1 170 Print "example " i+3 ":" 180 If i Then Print ,, "oxfords" Else Print ,, "brouges" 190 Next i 200 ' 210 For i=0 To 1 220 Print "example " i+5 ":" 230 If i Then Print ,, "jam on "; Else Print ,, "french ";: Print "toast" 240 Next i 250 rem ' 255 ? 260 For i=0 To 1 270 Print "example " i+7 ":" 280 If i Then Print ,, "1st ";: ? "2nd" Else Print ,, "3rd ";: Print "4th" 290 Next i we get the following output: example 1 : example 2 : green eggs on ham example 3 : brouges example 4 : oxfords example 5 : french toast example 6 : jam on example 7 : 3rd 4th example 8 : 1st 2nd Ok whereas micromite basic running on an MX170 gives: > RUN example 1: example 2: green eggs on ham example 3: brouges example 4: oxfords example 5: french toast example 6: jam on toast example 7: example 8: 1st [19] 280 If i Then Print ,, "1st ";: Print "2nd" Else Print ,, "3rd ";: Print "4th" Error: Expected closing bracket > note the deviation of example 6, that is NOT syntactically incorrect. one might speculate that the origins of the discrepancy lay in code written by Mrs Verda Spell, whose premature departure from this mortal plane prevented her completing the implementation of single line if-then-else. but then, i couldn't possibly comment! cheers, rob :-) Edited 2020-06-12 14:15 by robert.rozee |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1835 |
Strange one that, could still be a local issue as I've been hammering Ctrl-W and enter all day. Using a small LINDEN wireless keyboard, I grabbed the last two a few years back at a stock runout for around $12.00 each. Haven't missed a beat and get used every day. Hope you find the problem, I'll keep checking on it to be sure. NANO Inverter: Full download - Only Hex Ver 8.1Ks |
||||
panky Guru Joined: 02/10/2012 Location: AustraliaPosts: 1111 |
@KeepIS/matherp I have gone through the sequences by KeepIS above quite a number of times with no errors, EXCEPT for a 1 time Part 2 error with the ?[?255l?[H error appearing in white over black. Have not been able to repeat this and have never got the @@@@@ sequence anywhere. Tried at least thirty times. version 5.05.03b1c CMM2 from Circuit Gizmos, Microsoft USB keyboard, VGA output panky ... almost all of the Maximites, the MicromMites, the MM Extremes, the ArmMites, the PicoMite and loving it! |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1835 |
@panky, thanks for trying, I'm in the same boat, it can take an hour or so of coding before it happens, other times after a few seconds. That's what happened today, no rhyme nor reason, and appears to have nothing to do with Ctrl-C or errors, just seems totally random. Edited 2020-06-12 16:13 by KeepIS NANO Inverter: Full download - Only Hex Ver 8.1Ks |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 10070 |
All many thanks for keep trying to get to the bottom of the cut/paste issue. The editor shows control characters in reverse video and the @ symbol is a 0. What is happening is that for some reason the number of characters in the buffer is getting incorrectly set so it is pasting beyond the end of the valid data and into the zero initialised buffer - the question is why? Please try the attached version V5.05.03b1e http://geoffg.net/Downloads/Maximite/CMM2_Beta.zip In it I've included on the status line the number of characters in the paste buffer  This should be equal to the number of characters that were highlighted + the number of CRLF end of line pairs. It gets set when you press F4 or F5 to cut or copy the buffer. In addition I've included a check of the difference between my calculated buffer size and the length of the C string in the buffer - strlen(buffer). If these don't match then the number of chars is displayed as 999999 None of these changes should fix the problem but they may help you to diagnose when an issue has occurred. If you ever see the number of chars as 999999 then please let me know. In addition, if you see it, what happens if rather than pasting you do another selection? Does the error stay or does it self correct? |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1835 |
If I try another selection, it does the same thing. Just now I tried to paste a line and got garbage, I tried to copy the same line and paste, different garbage. Pressed SAVE this time and went back into edit, cleared the garbage and tried again to copy & paste, same problem. Pressed save again and went back into Edit, this time the copy & paste worked without error. I'll download the Zip file now and install. NANO Inverter: Full download - Only Hex Ver 8.1Ks |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1835 |
New firmware and spent some time before I got an error, however the behaviour has changed. The Error is nothing was pasted, copied again and pasted, still nothing, did it a 3rd time and the line pasted, editor then worked correctly every time. Difference this time, no garbage inserted. No 999999. EDIT, forgot to add, the Paste Buffer said 23 chars which was correct but nothing pasted. Mike EDIT 2: Same thing again, worked the second time, correct buffer size and once again, no garbage anywhere and the editor was stable again. I have never NOT had garbage pasted or other effects, so something has indeed changed for the better. . Edited 2020-06-12 19:17 by KeepIS NANO Inverter: Full download - Only Hex Ver 8.1Ks |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 10070 |
Please download again - sorry I b.....d cut/paste even though copy/paste was OK |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1835 |
Running new/new FW and was about to give up as all ok, however this time after a paste with nothing being pasted, I pressed F4 on a blank spot to clear the buffer, no chars in buffer display. When I pressed enter key, the text that didn't paste was inserted and another small line of text near the copy point was also inserted, trying to edit was making a mess as every enter key press inserted more text, so aborted edit, went back in and all ok and could not get another error so far. Still no Zero or Esc chars showing up. I've had this happen before but always followed by ESC and zero garbage. I will have to go shortly, but I will try again tomorrow. Mike. NANO Inverter: Full download - Only Hex Ver 8.1Ks |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 10070 |
Please download and try V5.05.03b1f http://geoffg.net/Downloads/Maximite/CMM2_Beta.zip Based on the above feedback it appears the cut/copy algorithm is robust but the issue is probably somewhere in paste. Accordingly I have re-engineered this slightly so please test this version and see if the editor issue has been solved. I've left the paste buffer count in the code as this is probably a useful feature. |
||||
| robert.rozee Guru Joined: 31/12/2012 Location: New ZealandPosts: 2401 |
just had a lightbulb moment. it looks very much like what some folks are seeing may be escape sequences sent out the console TxD line being echoed back on the RxD line. are you powering your CMM2 from a PSU, or from a USB port attached to a PC? if powered from a PSU, then the 1455/2221 will not be enumerating, and in that state may not be configuring the UART I/O pins running back to the H7. if this is the case, the console RxD input (PA10) to the H7 may float, picking up signals from all sorts of different places, including the adjacent console TxD output (PA9). to check this hypotheses, power your CMM2 from a USB port on a running PC and see if the garbage disappears. if it does, the solution is to either: 1. have the CMM2 firmware configure a weak pullup on PA10 (if this is possible), or, 2. add a 10k pullup from PA10 to 3v3 at the H7 module, or, 3. add a 10k pullup between pins 6 and 1 of the 1455/2221. note that a floating RxD pin picking up 'random' noise may create framing errors and/or massive interrupt storms that can cause bad issues. cheers, rob :-) |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 10070 |
Interesting idea that is certainly worth considering. I've added an internal pullup to the console RX port on the STM32H743 as a precaution. V5.05.03b1f was updated at 14:04 GMT 12/06/2020 with this change Questions for those seeing the issue. Are you running with the USB console port not enunciated? Which USB adapter are you using MCP2221A or PIC16F1455? What happens if you set OPTION CONSOLE SCREEN, OPTION CONSOLE SAVE so that the serial port is disabled? UPDATE Please hold off downloading for the moment Edited 2020-06-13 00:27 by matherp |
||||
| vegipete Guru Joined: 29/01/2013 Location: CanadaPosts: 1122 |
Still have my [ENTER] key repeat problem after using CTRL-W in the editor. This persists after power cycling and even removing the back-up battery. (latest copy of version 5.0503, US USB keyboard, power from a PSU.) Another keyboard issue, upon fresh power-up, ie, battery removed to clear saved config, I cannot select any keyboard type other than "1" UK. "2", "3" and other numbers do nothing. Other numbers do work when I'm typing in the time and date. I can change the keyboard type afterwards from the command prompt with option usbkeyboard. Visit Vegipete's *Mite Library for cool programs. |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 10070 |
This is deliberate to limit choices to those that are valid thwill: I'm struggling with zmim. I think it is corrupting memory with a rogue POKE or similar. If you write a trivial program to output scrolling and line-wrap work perfectly but somehow zmim is breaking it. Depending on trivial changes to the firmware I see very different effects - all bad, suggesting corruption which impacts differently depending on where the internal variables are placed by the linker |
||||
thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 4251 |
Thanks Peter, I hadn't dare hope you would find time to look at my issue. Currently what I've almost finished doing is changing the code to pass all "printing" through a single user subroutine. I need to do this anyway to implement word-wrapping and handle the case where more than a page of output is generated between input prompts. Perhaps this will *magically* fix it. Either way I'll then look at the POKEs. Ignoring trivial ones in "debug.inc" and "main.bas" they are all in "mem_cmm2.inc" so it should be easy to find if this is the issue. On a first glance it looks like the only possibility is if 'a' is < 0 or 'BASE_STATIC' contains nonsense on calls to wb() or ww(). I can't test this immediately as I'm away from the hardware. What does "or similar" mean? Isn't POKE the only way I can make a mess unless there is a bug in the firmware ? Thanks again, Tom Edited 2020-06-13 05:22 by thwill MMBasic for Linux, Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
| vegipete Guru Joined: 29/01/2013 Location: CanadaPosts: 1122 |
On my setup, the only valid response is "1". No other choice works. (Cicruit Gizmo CMM2 with Waveshare, 5.0503.) Visit Vegipete's *Mite Library for cool programs. |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 10070 |
I think this or something similar may be causing the problem ElseIf oc = &h7 Then ' RESTART Run Can you easily recode? This is an issue in the early versions of 5.05.03. For the moment just run main.bas. Easy fix but I need to get to the bottom of thwill's issue first |
||||
| Page 6 of 17 |
|||||
 Print this page |
 |
The Back Shed's forum code is written, and hosted, in Australia. | © JAQ Software 2025 |
