Errors when using Hitech

[Lieuwe]

Hello,

I Have some problems when I use the Hitech compiler.

Used Software:
Flowcode 4.5.18.74
Hitech 9.82
Source boost 7.04

Display_16F887.fcf

(7.5 KiB) Downloaded 231 times

When I use the default settings with the Source boost compiler, everything works as expected.

When I chance the compiler settings in flowcode to:

Compiler Location = C:\Program Files\HI-TECH Software\PICC\9.82\bin\picc.exe
Params = --CHIP=%p -N48 %f.c

Linker Location = C:\Program Files\Matrix Multimedia\Flowcode V4\Tools\DoNothing\DoNothing.exe
Linker Parameters =

I get the following message:

Launching the compiler...
C:\Program Files\HI-TECH Software\PICC\9.82\bin\picc.exe --CHIP=16F887 -N48 Display_16F887.c
Display_16F887.c: main()
663: OPTION = 0xC0;
^ (192) undefined identifier "OPTION"
HI-TECH C Compiler for PIC10/12/16 MCUs (Standard Mode) V9.82
Copyright (C) 2011 Microchip Technology Inc.

(908) exit status = 1

Return code = 1

Flowcode kan de C-code niet vertalen vanwege de volgende fout:


Als de flowchart C-code bevat, controleer die dan zorgvuldig. Als de flowchart geen C-code bevat of als die gegarandeerd vrij is van fouten, neem dan contact op met de technische ondersteuning.

FINISHED



Also importing the C file in MPlab gives the same messages.
When I use Sourceboost as compiler it works in MPlab, Hitech won't.

Lieuwe

[medelec35]

Hi Lieuwe
http://www.matrixmultimedia.com/mmforum ... 043#p25010
May or may not help you.

But I am a bit surprised about this as I thought the new definitions file accounted for this issue?
I compiled with hitec and Flowchart compiled OK without altering definition file.
If you do decide to alter your definitions file, it's best to copy the original fist, so if does not work, there's no harm done.
Just replace the modified file with original.

If still an issue then you may need to wait for matrix to response on Monday
Martin

[Lieuwe]

Thank you Medelec35,
I thought the definition files would be OK now, with the new release.
there are also some problems here with the new PIC16F1xxx devices.
I hope my comments help a bit to make Flowcode a better product.
(it is not a bad product at all, but improvements are welcome )

Waiting for a new update from matrix works the best for me now, be course I have both, Sourceboost and Hitech compilers standalone.
I switch compilers in this case (too bad Source boost seems to have some problems with the PIC16F1xxx ones).
But thanks for the suggestion.

Lieuwe

[Benj]

Hello,

Have you tried the new v4.5 update for Flowcode PIC? This allows BoostC to work with those newer 16f18xx amd 16f19xx devices.

[Lieuwe]

Benj,

When you have a look at my posting, I mentioned, that I use Flowcode 4.5
So I would like to know, are the settings for the compiler, as I mentioned above the right ones?
Also I used an older Target PIC16F887, to avoid problems with (possible, bugged) target definition files.

But you are right that I would actually like to use a PIC16F1939, to compile a file to drive a LCD I struggle with.
To look if changing compiler from sourceboost to hitech helps.

The good news is that compiling C files generated with Flowcode, outside flowcode in MPlab works again with Sourceboost V7.04 (older flowcode did not, be course of the new _ in the boostc_pic16.exe name.

Somehow switching compiler to Hitech in MPlab, AND / OR straight in PPP in Flowcode won't generate code.

Lieuwe

[Benj]

Hi Leuwe,

Sorry about that must have missed that bit of info.

The sourceboost compiler should handle the chip and the LCD ok is it having problems?

Regarding your HiTech problem i'm surprised you are getting that error.

Adding the define to your supplementary code window should allow you to compile correctly using HiTech.

http://www.matrixmultimedia.com/mmforum ... 043#p25010

[Lieuwe]

Some results using Hitech in Flowcode 4.5

Step 1: Using original settings (using sourceboost)
Result: successful.

Step 2: Using HI-TECH Software 9.70
Result: For half a seccond it shows only “Hi”, after that, strange symbols on Display.
Instead of “Hi look i am working” it shows strange characters.

Step 3: Using Hitech Software 9.70 AND replace definitions file:
Result: the same as step 2.

Step 4: updating Hitech compiler from 9.70 to 9.82
Result: no Code be course of: ^ (192) undefined identifier "OPTION"

Chanching definition file has no effect.

[Benj]

Hello,

What if you place the following line of code into your supplementary code window. View -> Project options -> View Supplementary code.

#define OPTION OPTION_REG

[Lieuwe]

If you place the following line of code into your supplementary code window. View -> Project options -> View Supplementary code.

#define OPTION OPTION_REG

Then: Hitech 9.7 does not compile and returns ^ (192) undefined identifier "OPTION"

But Hitech 9.82 does compile.

So for compiling with Hitech 9.7 do NOT include Suplementary code.
And for compiling with Hitech 9.82 DO include Supplementary code.

That was about compiling, now the results on the hardware:

Used: Eblock EB006 multiprogrammer, PIC16F887, LED Eblock, and LCD Eblock on port D.
D7 is 1 sec ON, on startup, after that second it stays OFF.
D6 Alternates ON/OFF and at the same time, text “Hi I am Working” switches between row 1 and 2 of the LCD.

HitechDisplayTest.fcf

(7.5 KiB) Downloaded 231 times

This wil work when you use the Sourceboost compiler.
When you use the Hitech compiler (9.70 or 9.82)
The Led D7 and D6 work as planed, but....
The display wil give strange symbols.

So now I can compile with Hitech, but the code does not work,
while the generated code, using Source boost does work.

Lieuwe