PASD Fails to Start

jazzed

Hello.

I've been using PASD and have become somewhat addicted to it.
Now the program tries to start, but crashes before opening fully.
Anyone else have this problem? Is there a·fix?

TIA

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
jazzed·... about·living in·http://en.wikipedia.org/wiki/Silicon_Valley

Traffic is slow at times, but Parallax orders·always get here fast 8)

Ariba

Sorry jazzed, and all the other users of PASD

I have accidentally left an Expire Date ( 1. Mars 2008) in the source code. This was only intended for the early Beta-versions.
Now I have an updated version (0.3) on my website, that shoud work forever

Only the .exe has changed, so you only have to replace this. I also fixed the bug with F2.

Link:
www.insonix.ch/propeller/prop_pasd.html

Cheers
Andy

jazzed

Thank you Andy [noparse]:)[/noparse]

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
jazzed·... about·living in·http://en.wikipedia.org/wiki/Silicon_Valley

Traffic is slow at times, but Parallax orders·always get here fast 8)

jazzed

Andy this product is very well done.
If you ever revisit it for feature additions for which I would happily donate ...

Please consider in order of importance:

1) Provide a "continue" function that allows running after a break;
2) Make the "on top" feature of RAM viewers optional;
3) Make the virtual LEDs resizable;

Another interesting idea is a .spin debugger ... I've considered writing one.
This is the single biggest missing feature in propeller today.

Cheers 8)

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
jazzed·... about·living in·http://en.wikipedia.org/wiki/Silicon_Valley

Traffic is slow at times, but Parallax orders·always get here fast 8)

Ariba

1) You can continue after a Break with F5 (Run). The first press executes the actual Instruction as a single step (necessary because this instruction was replaced by a Break jump), a second F5 press continues the code until the next Breakpoint is reached.

2) This is doable, but If they are not on top, you have to bring it on top after every single step, if needed. I found it easier to minimize and restore the viewer windows.

3) This can be done

Yes a Spin debugger would be great, but the problem is the assignement of the bytecode to the Spin source lines. For PASM this is relativly easy because every line produces a long in the Cog RAM (with some exeptions). Today we know much more about the bytecode (thanks hippy) then last year when I have made the PASD, so it may be possible... (I don't plan to make it).

Andy

jazzed

Ariba said...
1) You can continue after a Break with F5 (Run). The first press executes the actual Instruction as a single step (necessary because this instruction was replaced by a Break jump), a second F5 press continues the code until the next Breakpoint is reached.

Ok, the second F5·does work, but·the run takes a while.
My somewhat impatient nature got the best of me [noparse]:)[/noparse]
Thanks.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
jazzed·... about·living in·http://en.wikipedia.org/wiki/Silicon_Valley

Traffic is slow at times, but Parallax orders·always get here fast 8)

hippy

jazzed said...
Another interesting idea is a .spin debugger ... I've considered writing one.
This is the single biggest missing feature in propeller today.

Agreed. The main obstacle as Andy notes is being able to relate the bytecode to source line and knowing where variables are in memory. The PropTool may one day produce a listing or cross-reference which would help there, and hopefully any third-party compilers will do that.

The other solution is to decompile the bytecode to Spin source code but that would by necessity be much more artificial and would require a fair bit of work.

jazzed

Perhaps the community .spin compiler project will produce something useful in that area.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
jazzed·... about·living in·http://en.wikipedia.org/wiki/Silicon_Valley

Traffic is slow at times, but Parallax orders·always get here fast 8)