lmm_pc is the PC in the LMM kernel. Manipulating it directly allows for jumps to +64/-63 instructions.
lmm_jump
rdlong lmm_pc, lmm_pc ' access long just after call to lmm_jumpadd lmm_pc, #16' adjust for real addressjmp #lmm_fetch
lmm_fetch
rdlong lmm_i, lmm_pc
add lmm_pc, #4' next inst
lmm_i
nopjmp #lmm_fetch
...
somewhere your LMM code:
mov somevar, #500jmp #lmm_jump
long @some_label
...
somelabel
mov somevar, #200mov var_1, #500
...
sub lmm_pc, #($-somelabel-1)*4' This is a jump backwards to somelabel
someloop:
add lmm_pc, #(someotherlabel - 1 - $)*4' jump forwards
That kind of LMM does not need anything extra from the compiler and works very well!
The rest of the code that access variables and subroutines in COG do not need any change, because it will be run from COG RAM as any other instruction. Only jumps or calls to LMM code need to work around different address range. Two small routines can take care of that (lmm_jump, lmm_call).
I have two threads which I believe should be stickies.·Anyone can·add (post) descriptions and·links·to their threads. I reference·these threads·in my signature (as suggested by OBC). I have maintained the second one much better than the first (as I gained more experience - I just need to update the first thread to match).
The third possibility is currently·based on my pcb, but I hope that others will follow suit with various implementations on various hardware (all propeller based). So it needs to be an index as the two above, not·my thread. I am willing to create it and maintain it if you would like me to.
I have a couple of questions.. I just installed 0.17 on my intel mac laptop and I have 2 issues, the first is it doesn't see my propeller, and second.. the code is all black on a white background, no colour in the editor..
KPR said...
I have a couple of questions.. I just installed 0.17 on my intel mac laptop and I have 2 issues, the first is it doesn't see my propeller, and second.. the code is all black on a white background, no colour in the editor..
BUG: The compiler crashes if fed with code containing #define and/or #ifdef (Not sure which).
When the compiler fails the IDE becomes unusable, menu items no longer work, can't close it etc.
I have a little bit of weird set up here. I'm using BST under Linux (Debian) as the editor and Prop loader.
The code under development contains #define and #ifdef (See attachment)and is compiled with HomeSpun.
Of course I had to hit F8 to see what happens!
Any chance of directives in BST at some point ? Or dare I say a way for the BSC IDE to use the homespun compiler ?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
heater said...
BUG: The compiler crashes if fed with code containing #define and/or #ifdef (Not sure which).
When the compiler fails the IDE becomes unusable, menu items no longer work, can't close it etc.
I have a little bit of weird set up here. I'm using BST under Linux (Debian) as the editor and Prop loader.
The code under development contains #define and #ifdef (See attachment)and is compiled with HomeSpun.
That is a rather obscure configuration. Interesting that it locks up. If you send me a file that it locks up on I'll fix it and get a quick pre-release out for you.
heater said...
Of course I had to hit F8 to see what happens!
Any chance of directives in BST at some point ? Or dare I say a way for the BSC IDE to use the homespun compiler ?
Directives are certainly on my todo list. They are bubbling their way to the top.
Not a hope in hell of having bst use homespun. I don't just "call" the compiler, I have full interactive access to the list file, symbol list, binary output and each individual object. If I was to enable external compilers I'd lose lots of the little integrations that make the compile/download experience so smooth.
You might also have noticed that you can make changes to source files and compile/download them without saving anything to disk. You can't do that with external compilers without saving temporary files.. and that is so 1980's.
I'm still struggling with the widgetset bug on OSX that Ale pointed out. As soon as I've got that nailed I'm ready for a new release (have a few new bits in there now)
My OSX machine is being completely recalcitrant and needs a total re-load (I borked the X-Code development environment and it has screwed a great deal of stuff up), so I've still got a little way to go. (Like finding my original OSX 10.5 disk.. )
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cardinal Fang! Fetch the comfy chair.
Not so obscure, just bst on Debian. Only we wan to use directives to support different hardware configurations on the Prop.
Attached is a file containing the messages that pop out. The first two messages about signals "enter" and "leave" always happen on start up.
The rest of the messages are the result of selecting Compile->Compile Information followed by File->Quit (did that twice).
The menues appear but they do not respond. I can see the compiler continues to run as it sits in the KDE panel.
Now, I also have Windows XP running in a virtual machine under Debian. The same experiment there results in the same compiler lock up but the menues do seem to continue working.
By the way, I'm using Sun's VirtualBox for virtualization which is excellent and free. If you have a enough RAM this is a great way to test cross platform code like bst, just run a Linux and an XP on a Mac for example.
Point taken about the alternative compilers, how about away to run external shell commands on the current file, say from F12 ?. Opens up a lot of possibilties for running other tools. One thing I'd like to see is a way to get rid of all the wide characters in a file and save as ASCII text so that normal Unix tools can be used with spin files.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
BradC: heaters code (ZiCog) is what I've been discussing with you offline. I just had a thought - I have just added special prop characters, so the file is no longer ascii, but unicode. I know you don't handle #define/#ifdef/#else/#endif. But could it be the unicode?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔ Links to other interesting threads:
Cluso99 said...
BradC: heaters code (ZiCog) is what I've been discussing with you offline. I just had a thought - I have just added special prop characters, so the file is no longer ascii, but unicode. I know you don't handle #define/#ifdef/#else/#endif. But could it be the unicode?
I doubt it. If Propeller Tool can compile it, bst can compile it. If it can't then its a bug. There is a lot of unicode in the OBEX and I compile over 1,500 files from that every time I run a regression test.
@heater. Thanks, it's a compiler crash and I've only just added some code to cope with that in the main application thread. Otherwise when it crashes it just locks up the bst UI.
The actual bug is the compiler is seeing #define in the CON block and trying to resolve the number to begin enumeration. It then ties itself tightly in a knot. I did say the error handling might be a bit flaky [noparse];)[/noparse]
I'll get that fixed asap and get another pre-release cooked up for you to play with.
As for the option to strip out the unicode to convert it to plaintext, I can add that very easily. I already have a program "ucat" I use that translates propeller tool UTF-16 so I can view it with less or vi. It'd be easy to add an edit menu option to gut the unicode out. I'll pop it on the todo list.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cardinal Fang! Fetch the comfy chair.
Ok, I found the bug and have donned requisite brown paper bag. It's one of those " I can't believe I could have been so stupid " cut and paste bugs. I'll get a regression test done and new build out some time tomorrow. Sorry lads.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cardinal Fang! Fetch the comfy chair.
BUG: Wants to change to UNICODE when non UNICODE character added to a file. Then saves as ASCII anyway.
1. Open a file that contains no unicode characters
2. Enter a ' somewhere for a comment
3. Message box pops up "File contains new UNICODE characters and will be saved as UNICODE"
4. Hit OK
5. Use unix "file" command to see what the file now is, it reports ISO-8859 English text, with CRLF terminators"
6. Inspect the file with the vim editor, looks fine.
Compile and run with homespun no problem.
Again v17 and v18pre3 on Debian Lenny.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
heater said...
BUG: This is something that hangs BST consistently on my Debian Lenny machine.
1. Open BST
2. View -> Serial Terminal
3. Port -> /dev/ttyUSB0
4. Communication -> Connect
5. minimize the serial terminal
6. File -> Open
We are now hung up, every time. No messages came out of BST except the usual thing about "enter" and "leave".
Curses.. I can't reproduce this here at all. I'll keep looking.
I did have the file selector dialog box hide behind the main window once, but a alt-tab brought it forwards.
heater said...
BUG: Wants to change to UNICODE when non UNICODE character added to a file. Then saves as ASCII anyway.
1. Open a file that contains no unicode characters
2. Enter a ' somewhere for a comment
3. Message box pops up "File contains new UNICODE characters and will be saved as UNICODE"
4. Hit OK
5. Use unix "file" command to see what the file now is, it reports ISO-8859 English text, with CRLF terminators"
6. Inspect the file with the vim editor, looks fine.
I've seen this one several times in the last couple of days (I've been away on the boat doing some development while I was disconnected).
I am investigating it further. I also have some other serial terminal bugs to squash which I discovered when trying to view some corrupt NMEA data.
I've just uploaded -pre4 which has the fix for the compiler crash when encountering a #define. It may have a number of broken bits though as it's just a snapshot of where I am right at the moment, and I'm in the middle of some pretty serious re-architecture of a couple of components (which is one of the reasons I've been slow getting a release out). I've used it for the last couple of days as-is, but its not been tested on Win or Mac.
@Ale, I've nailed the slow startup on OSX. It's related to the way I populate the directory tree on startup. I'm working on a fix for it.
I have managed to fix the crash on OSX Intel when double clicking the top of the directory tree in this -pre.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!
OK it does not hang but the file selector dialog box does hide behind the main window when following those steps. Never though to check that it was underneath because it normally comes out on top and it does not show up in my GNOME panel as do the main window and the terminal window.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
heater said...
OK it does not hang but the file selector dialog box does hide behind the main window when following those steps. Never though to check that it was underneath because it normally comes out on top and it does not show up in my GNOME panel as do the main window and the terminal window.
Excellent.. so I see what you see.. now I can reproduce it I can fix it [noparse]:)[/noparse]
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!
Today I've been trying to use FullDuplexSerial and i cannot get the Serial Terminal in BST to display anything. Also PST will not display anything. If I use PropTool to load the prop, with BST closed i can see the serial data. I cannot say if any previous versions have worked, though i am happy to test if you want.
Also, i know this is non-working code, but the follow code hangs BST when compiling for ~10 seconds, then says there is a Compile Error, but reports no errors.
I will hopefully get a bit of prop time this week, so i should be able to do some more testing with the 64-bit version of vista. Also the internet will be connected to the new house in the next few days, should i should be back on the prop IRC channel soon.
Many thanks as always.
sevs
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Advertisement sponsored by dfletch:
Come and join us on the Propeller IRC channel for fast and easy help!
Channel: #propeller
Server: irc.freenode.net or freenode.net
If you don't want to bother installing an IRC client, use Mibbit. www.mibbit.com
@sevs. I have intermittent access to a Vista machine, so I'll check the serial stuff when I get a chance.
@McGRAW, There is currently no way to re-arrange the tab order short of closing the tabs and re-opening them in the order you want. It's certainly on my todo list though.
You may have noticed I've not been as responsive lately. I apologise for that, but have some other issues that have taken priority.. I hope to get some more time in the near future to at least get these last bugs taken care of and a release out the door.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!
I'd like to request if possible to add cmd + arrows movement in the Mac version. I know Mac software is very inconsistent (at best) in that regard (I mean the Home/End/PgUp and PgDown do not work equal in all apps) But that way it could help because I do not get custom to Fn+Arrows to do exactly that. With time.
I'd like to request if possible to add cmd + arrows movement in the Mac version. I know Mac software is very inconsistent (at best) in that regard (I mean the Home/End/PgUp and PgDown do not work equal in all apps) But that way it could help because I do not get custom to Fn+Arrows to do exactly that. With time.
Thanks !
No worries Ale! Changing the keymaps for the Mac Version to make it more "Macified" is on the immediate priority list, so as soon as I get some time to work on it (hopefully in the next 10 days or so) I'll be sorting that out. My Intel Mac is still playing silly buggers, so it needs to be re-installed, but I can't find my OSX 10.5 install disk... I'll get there [noparse]:)[/noparse]
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!
If you are there can you use the same color (syntax highlight) for and and or as for the rest of the instructions ?. Do they have black for some reason (I did not read about) ?
Ale said...
If you are there can you use the same color (syntax highlight) for and and or as for the rest of the instructions ?. Do they have black for some reason (I did not read about) ?
Thanks !
Yes, they are black as I had a bug in the syntax highlighter and they were being treated in that context as a SPIN instruction rather than assembly or an operator as required. I have already fixed that one [noparse]:)[/noparse]
There are quite a few oddities in the syntax highlighter if you look too closely [noparse];)[/noparse] The Parallax one has a very clever parsing table and defined rules to prevent that sort of thing from happening. Mine kinda evolved as a spin-off of the compilers parser as I learned about writing a highlighter for the editor I'm using. It could do with a bit of cleaning up!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!
BradC, thanks for making this program. It has been working flawlessly for me on my Mac. Saves me from having to run a virtual machine.
Once in a while (three times now) I get this screen that vanishes with resizing the window or clicking on another tab. Hasn't caused me any problems though.
W9GFO said...
BradC, thanks for making this program. It has been working flawlessly for me on my Mac. Saves me from having to run a virtual machine.
Once in a while (three times now) I get this screen that vanishes with resizing the window or clicking on another tab. Hasn't caused me any problems though.
Rich H
Haha! You found the easter egg!
No, but seriously, I put that there to expose precisely this kind of bug. Are you on PPC or Intel, 10.4 or 10.5?
Have you any idea what you might be doing that tickles it? I know what it is, I just don't know why or how to solve it. The message you are seeing is written on the tab component canvas. That we can see it at all means the editor component associated with that tab has become invisible. Resizing the window causes a re-paint and it must become visible again.
I might have to look at that bit of code a bit harder.
Thanks for the report [noparse]:)[/noparse]
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!
I can't say what causes it. After the second time I saw it I decided to try to make it happen again so that I could get a screenshot. Probably spent half an hour clicking tabs, closing tabs, editing then closing without saving etc...
Then one time upon closing a tab, it was there.
The first two times it happened soon after coming back to the program when it had been in the background for quite a while.
Just a small question: I saw in the status ba that it says "INS" meaning insert mode. But Where is the insert key ? My Mac keyboard does not have an Insert key (I could recognize) neither has my MacBook. Do you happen to know which key configuration will trigger it ?, thanks !
Ale said...
Just a small question: I saw in the status ba that it says "INS" meaning insert mode. But Where is the insert key ? My Mac keyboard does not have an Insert key (I could recognize) neither has my MacBook. Do you happen to know which key configuration will trigger it ?, thanks !
No, I have no idea. On my Mac (running Linux) it's the help key. What toggles between insert and overwrite on a Mac? I'll make sure it's correct if you can tell me the right key combination to use.
@W9GFO, Ok, so you saw it closing a tab. That makes sense actually. Let me look at it a bit and see if I can get it to happen here.
It's a bit of a bugger, as the machine I use to do my cross-compiling for Mac is also my main desktop Linux machine (which happens to be my Intel Mac), so there is a _lot_ of rebooting involved in testing this stuff as I have to boot OSX, test, boot linux, change/compile, wash, rinse, repeat.. Most of this weird stuff just refuses to happen on my 10.4 PPC G4 400Mhz.
I don't have any time to look at any of this in the next week, but hopefully on the weekend or early next week I can get to it.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!
Comments
I do the following for LMM code:
lmm_pc is the PC in the LMM kernel. Manipulating it directly allows for jumps to +64/-63 instructions.
lmm_jump rdlong lmm_pc, lmm_pc ' access long just after call to lmm_jump add lmm_pc, #16 ' adjust for real address jmp #lmm_fetch lmm_fetch rdlong lmm_i, lmm_pc add lmm_pc, #4 ' next inst lmm_i nop jmp #lmm_fetch ... somewhere your LMM code: mov somevar, #500 jmp #lmm_jump long @some_label ... somelabel mov somevar, #200 mov var_1, #500 ... sub lmm_pc, #($-somelabel-1)*4 ' This is a jump backwards to somelabel someloop: add lmm_pc, #(someotherlabel - 1 - $)*4 ' jump forwards
That kind of LMM does not need anything extra from the compiler and works very well!
The rest of the code that access variables and subroutines in COG do not need any change, because it will be run from COG RAM as any other instruction. Only jumps or calls to LMM code need to work around different address range. Two small routines can take care of that (lmm_jump, lmm_call).
I have two threads which I believe should be stickies.·Anyone can·add (post) descriptions and·links·to their threads. I reference·these threads·in my signature (as suggested by OBC). I have maintained the second one much better than the first (as I gained more experience - I just need to update the first thread to match).
Prop Tools under Development or Completed (Index) http://forums.parallax.com/showthread.php?p=753439
Emulators: Micros eg Altair, and Terminals eg VT100 (Index)··http://forums.parallax.com/showthread.php?p=778427
The third possibility is currently·based on my pcb, but I hope that others will follow suit with various implementations on various hardware (all propeller based). So it needs to be an index as the two above, not·my thread. I am willing to create it and maintain it if you would like me to.
Single Board Computer - 3 Propeller ICs and TriBladeProp PCB··http://forums.parallax.com/showthread.php?p=790917
The topic needs to·be·more generic, perhaps "Single Board Computers: Propeller based (Index)"
All of these type of threads are attracting a lot of interest, as can be seen by the number of replies and views.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps: TriBladeProp, SixBladeProp, website (Multiple propeller pcbs)
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
· Prop Tools under Development or Completed (Index)
· Emulators: Micros eg Altair, and Terminals eg VT100 (Index)
· Search the Propeller forums (via Google)
My cruising website is: ·www.bluemagic.biz·· MultiBladeProp is: www.bluemagic.biz/cluso.htm
Any Ideas
Kevin
Several things to try:
Do you have the FTDI drivers installed? http://www.ftdichip.com/FTDrivers.htm
Is the board powered up?
Do you have the use colors checkbox in Tools->IDE Preferences->Colors and Fonts checked?
I'll have to check tonight when I get home from work..
Thanks for the the help..
When the compiler fails the IDE becomes unusable, menu items no longer work, can't close it etc.
I have a little bit of weird set up here. I'm using BST under Linux (Debian) as the editor and Prop loader.
The code under development contains #define and #ifdef (See attachment)and is compiled with HomeSpun.
Of course I had to hit F8 to see what happens!
Any chance of directives in BST at some point ? Or dare I say a way for the BSC IDE to use the homespun compiler ?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
That is a rather obscure configuration. Interesting that it locks up. If you send me a file that it locks up on I'll fix it and get a quick pre-release out for you.
Directives are certainly on my todo list. They are bubbling their way to the top.
Not a hope in hell of having bst use homespun. I don't just "call" the compiler, I have full interactive access to the list file, symbol list, binary output and each individual object. If I was to enable external compilers I'd lose lots of the little integrations that make the compile/download experience so smooth.
You might also have noticed that you can make changes to source files and compile/download them without saving anything to disk. You can't do that with external compilers without saving temporary files.. and that is so 1980's.
I'm still struggling with the widgetset bug on OSX that Ale pointed out. As soon as I've got that nailed I'm ready for a new release (have a few new bits in there now)
My OSX machine is being completely recalcitrant and needs a total re-load (I borked the X-Code development environment and it has screwed a great deal of stuff up), so I've still got a little way to go. (Like finding my original OSX 10.5 disk.. )
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cardinal Fang! Fetch the comfy chair.
Attached is a file containing the messages that pop out. The first two messages about signals "enter" and "leave" always happen on start up.
The rest of the messages are the result of selecting Compile->Compile Information followed by File->Quit (did that twice).
The menues appear but they do not respond. I can see the compiler continues to run as it sits in the KDE panel.
Now, I also have Windows XP running in a virtual machine under Debian. The same experiment there results in the same compiler lock up but the menues do seem to continue working.
By the way, I'm using Sun's VirtualBox for virtualization which is excellent and free. If you have a enough RAM this is a great way to test cross platform code like bst, just run a Linux and an XP on a Mac for example.
Point taken about the alternative compilers, how about away to run external shell commands on the current file, say from F12 ?. Opens up a lot of possibilties for running other tools. One thing I'd like to see is a way to get rid of all the wide characters in a file and save as ASCII text so that normal Unix tools can be used with spin files.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps: TriBladeProp, SixBladeProp, website (Multiple propeller pcbs)
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
· Prop Tools under Development or Completed (Index)
· Emulators: Micros eg Altair, and Terminals eg VT100 (Index)
· Search the Propeller forums (via Google)
My cruising website is: ·www.bluemagic.biz·· MultiBladeProp is: www.bluemagic.biz/cluso.htm
I doubt it. If Propeller Tool can compile it, bst can compile it. If it can't then its a bug. There is a lot of unicode in the OBEX and I compile over 1,500 files from that every time I run a regression test.
@heater. Thanks, it's a compiler crash and I've only just added some code to cope with that in the main application thread. Otherwise when it crashes it just locks up the bst UI.
The actual bug is the compiler is seeing #define in the CON block and trying to resolve the number to begin enumeration. It then ties itself tightly in a knot. I did say the error handling might be a bit flaky [noparse];)[/noparse]
I'll get that fixed asap and get another pre-release cooked up for you to play with.
As for the option to strip out the unicode to convert it to plaintext, I can add that very easily. I already have a program "ucat" I use that translates propeller tool UTF-16 so I can view it with less or vi. It'd be easy to add an edit menu option to gut the unicode out. I'll pop it on the todo list.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cardinal Fang! Fetch the comfy chair.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cardinal Fang! Fetch the comfy chair.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps: TriBladeProp, SixBladeProp, website (Multiple propeller pcbs)
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
· Prop Tools under Development or Completed (Index)
· Emulators: Micros eg Altair, and Terminals eg VT100 (Index)
· Search the Propeller forums (via Google)
My cruising website is: ·www.bluemagic.biz·· MultiBladeProp is: www.bluemagic.biz/cluso.htm
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
1. Open BST
2. View -> Serial Terminal
3. Port -> /dev/ttyUSB0
4. Communication -> Connect
5. minimize the serial terminal
6. File -> Open
We are now hung up, every time. No messages came out of BST except the usual thing about "enter" and "leave".
This is annoying because what I want to do is have the terminal all up and running prior to opening an eeprom file for download.
Edit: Forgot mention this is true for v17 and v18pre3
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
Post Edited (heater) : 4/9/2009 5:39:40 PM GMT
1. Open a file that contains no unicode characters
2. Enter a ' somewhere for a comment
3. Message box pops up "File contains new UNICODE characters and will be saved as UNICODE"
4. Hit OK
5. Use unix "file" command to see what the file now is, it reports ISO-8859 English text, with CRLF terminators"
6. Inspect the file with the vim editor, looks fine.
Compile and run with homespun no problem.
Again v17 and v18pre3 on Debian Lenny.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
Curses.. I can't reproduce this here at all. I'll keep looking.
I did have the file selector dialog box hide behind the main window once, but a alt-tab brought it forwards.
I've seen this one several times in the last couple of days (I've been away on the boat doing some development while I was disconnected).
I am investigating it further. I also have some other serial terminal bugs to squash which I discovered when trying to view some corrupt NMEA data.
I've just uploaded -pre4 which has the fix for the compiler crash when encountering a #define. It may have a number of broken bits though as it's just a snapshot of where I am right at the moment, and I'm in the middle of some pretty serious re-architecture of a couple of components (which is one of the reasons I've been slow getting a release out). I've used it for the last couple of days as-is, but its not been tested on Win or Mac.
@Ale, I've nailed the slow startup on OSX. It's related to the way I populate the directory tree on startup. I'm working on a fix for it.
I have managed to fix the crash on OSX Intel when double clicking the top of the directory tree in this -pre.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
Excellent.. so I see what you see.. now I can reproduce it I can fix it [noparse]:)[/noparse]
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!
Today I've been trying to use FullDuplexSerial and i cannot get the Serial Terminal in BST to display anything. Also PST will not display anything. If I use PropTool to load the prop, with BST closed i can see the serial data. I cannot say if any previous versions have worked, though i am happy to test if you want.
Also, i know this is non-working code, but the follow code hangs BST when compiling for ~10 seconds, then says there is a Compile Error, but reports no errors.
CON OBJ VAR byte Count_UP byte Count_Down PUB Start Count_Up := 1 Count_Down := 2 dira[noparse][[/noparse]Count_Up] := 1 dira[noparse][[/noparse]Count_Down] := 1 repeat if
Im using BST Pre-4 and Vista 64 Bit.
I will hopefully get a bit of prop time this week, so i should be able to do some more testing with the 64-bit version of vista. Also the internet will be connected to the new house in the next few days, should i should be back on the prop IRC channel soon.
Many thanks as always.
sevs
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Advertisement sponsored by dfletch:
Come and join us on the Propeller IRC channel for fast and easy help!
Channel: #propeller
Server: irc.freenode.net or freenode.net
If you don't want to bother installing an IRC client, use Mibbit. www.mibbit.com
First I want to thank you for this application.
I haven't experienced any bugs yet but I was curious about the ability to arrange tabs in different orders.
I understand this is a minimal priority but for someone like me who is still learning spin and constantly has several tabs open it proves useful
-- AJ
@McGRAW, There is currently no way to re-arrange the tab order short of closing the tabs and re-opening them in the order you want. It's certainly on my todo list though.
You may have noticed I've not been as responsive lately. I apologise for that, but have some other issues that have taken priority.. I hope to get some more time in the near future to at least get these last bugs taken care of and a release out the door.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!
I'd like to request if possible to add cmd + arrows movement in the Mac version. I know Mac software is very inconsistent (at best) in that regard (I mean the Home/End/PgUp and PgDown do not work equal in all apps) But that way it could help because I do not get custom to Fn+Arrows to do exactly that. With time.
Thanks !
No worries Ale! Changing the keymaps for the Mac Version to make it more "Macified" is on the immediate priority list, so as soon as I get some time to work on it (hopefully in the next 10 days or so) I'll be sorting that out. My Intel Mac is still playing silly buggers, so it needs to be re-installed, but I can't find my OSX 10.5 install disk... I'll get there [noparse]:)[/noparse]
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!
Thanks !
Yes, they are black as I had a bug in the syntax highlighter and they were being treated in that context as a SPIN instruction rather than assembly or an operator as required. I have already fixed that one [noparse]:)[/noparse]
There are quite a few oddities in the syntax highlighter if you look too closely [noparse];)[/noparse] The Parallax one has a very clever parsing table and defined rules to prevent that sort of thing from happening. Mine kinda evolved as a spin-off of the compilers parser as I learned about writing a highlighter for the editor I'm using. It could do with a bit of cleaning up!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!
Once in a while (three times now) I get this screen that vanishes with resizing the window or clicking on another tab. Hasn't caused me any problems though.
Rich H
Haha! You found the easter egg!
No, but seriously, I put that there to expose precisely this kind of bug. Are you on PPC or Intel, 10.4 or 10.5?
Have you any idea what you might be doing that tickles it? I know what it is, I just don't know why or how to solve it. The message you are seeing is written on the tab component canvas. That we can see it at all means the editor component associated with that tab has become invisible. Resizing the window causes a re-paint and it must become visible again.
I might have to look at that bit of code a bit harder.
Thanks for the report [noparse]:)[/noparse]
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!
Yay for me!
10.5.6 Intel
I can't say what causes it. After the second time I saw it I decided to try to make it happen again so that I could get a screenshot. Probably spent half an hour clicking tabs, closing tabs, editing then closing without saving etc...
Then one time upon closing a tab, it was there.
The first two times it happened soon after coming back to the program when it had been in the background for quite a while.
Rich H
No, I have no idea. On my Mac (running Linux) it's the help key. What toggles between insert and overwrite on a Mac? I'll make sure it's correct if you can tell me the right key combination to use.
@W9GFO, Ok, so you saw it closing a tab. That makes sense actually. Let me look at it a bit and see if I can get it to happen here.
It's a bit of a bugger, as the machine I use to do my cross-compiling for Mac is also my main desktop Linux machine (which happens to be my Intel Mac), so there is a _lot_ of rebooting involved in testing this stuff as I have to boot OSX, test, boot linux, change/compile, wash, rinse, repeat.. Most of this weird stuff just refuses to happen on my 10.4 PPC G4 400Mhz.
I don't have any time to look at any of this in the next week, but hopefully on the weekend or early next week I can get to it.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"VOOM"?!? Mate, this bird wouldn't "voom" if you put four million volts through it! 'E's bleedin' demised!