gpasm and MPlab 8.92 - anyone made it work?

[augustinetez]

I've got some PIC's (16F18855) that aren't available in MPlab 8.92 but can be compiled with the latest Gpasm.

I've installed Gpasm and can select it as the toolchain in MPlab but it won't compile a known program that has compiled with MPasm (on a 16F1827).

Just keeps giving a "Build failed message".

Trying to invoke the error file of Gpasm with the -k tag also doesn't work.

Any idea's?

 

[augustinetez]

Also tried Gpasm direct from the command prompt as per the manual and that does a lot of nothing as well.

 

[tumbleweed]

If you can't get it to assemble from a command line prompt there's not much point in trying to run it from an IDE.

MPLAB 8.92 isn't going to let you do much with an unsupported processor anyway.

 

[upand_at_them]

I don't bother with the IDE. I use a text editor and compile with MPASMX from the command prompt.

 

[augustinetez]

There's another program that lets you link gpasm in to the old MPlab - Mplab then just becomes the editor.

I don't bother with the IDE. I use a text editor and compile with MPASMX from the command prompt.

Quick pointer to some info on how to do that? I've been using Notepad++ for the editor.

 

[tumbleweed]

Instead of farting around with getting gpasm to work, mpasmx 5.87 supports the 16f18855.

You can get it from MPLABX v5.35 (the last one to support mpasm).
After you install it you can always copy/move the mpasmx folder somewhere else and remove mplabx if you don't want it.

mpasmx is the same executable as mpasm, just renamed. You can run it from a commandline or the little standalone app just the same as mpasm.

 

[upand_at_them]

What tumbleweed said. Sorry, I didn't mention the version.

Calling mpasmx is easy in my text editor (old version of UltraEdit), as it allows creating buttons that run commands. So I have it call the EXE with the name of the open file plus some other parameters. Compiling is as simple as clicking the button (or hot-keying).

 

[Ian Rogers]

There's another program that lets you link gpasm in to the old MPlab - Mplab then just becomes the editor.


Quick pointer to some info on how to do that? I've been using Notepad++ for the editor.

There is a cracking editor that I use for several builds.... It's called Textpad... I have used it for years.
It has the ability to run assemblers compilers and linkers...

Its £16.50 and that's for life... I've been using it since V4... they are on V8.11 now..
Or eclipse ( but it has a learning curve to get command line tools to work )

 

[augustinetez]

Thanks for all the comments, I'll follow them up and see where it leads.

 

[augustinetez]

There is a cracking editor that I use for several builds.... It's called Textpad...

Downloaded Textpad but can't make it work with Mpasmx.

Can you share your integration details Ian?

 

[augustinetez]

Update, got it working for the most part.

Need to work out how to make it generate the .err file with the same name as the file being compiled instead of %1.ERR and also to stop it outputting an .0 (object) file.

Tool preferences window below and this is what is in the parameters window:-

/e%1.ERR /aINHX8M /W0 /Q /L+ /X- /C- $File

 

[augustinetez]

Ok, final update got it all working bar stopping generating the .o file - I can live with that seeing as it's only for chips MPlab 8.92 won't handle.

 

[tumbleweed]

...and also to stop it outputting an .0 (object) file.

I don't think you can, even though there's supposedly a commandline switch to stop it.

I've never been able to get it to work and that's going all the way back to mpasmwin 5.51 from mplab 8.92.
Seems to be hard-wired to generate a .O file.

 

[Nigel Goodwin]

I don't think you can, even though there's supposedly a commandline switch to stop it.

I've never been able to get it to work and that's going all the way back to mpasmwin 5.51 from mplab 8.92.
Seems to be hard-wired to generate a .O file.

Does it matter anyway?.

 

[augustinetez]

Does it matter anyway?.

No, as I said above, I can live with it.

 

[upand_at_them]

stopping generating the .o file

Can you write a batch file to delete it and call that after compiling?

 

[augustinetez]

It's not really a big deal, just my OCD saying why generate it unless it's needed - I generally delete all the extraneous files after a compile anyway.

MPlab 8.92 is the same - I have a vague recollection of a discussion (possibly on the Microchip forum) quite some time ago about stopping it being generated but not sure if a solution was found then either.

 

[Nigel Goodwin]

Long ago I wrote a version of WinPicProg that had a simple in-built BASIC compiler, this generated assembler code, and then called MPASM to assemble it, the resultant HEX file was then automatically loaded back in to WinPicProg, and (optionally) automatically uploaded to the target PIC.

I can't really remember now?, but I suspect as part of that process any extraneous files were deleted - no command line flags or whatever, just done in the Windows source code.