Topic : Still about identifiers...

Forum : 8051

Original Post
Post Information Post
October 25, 2008 - 11:12am
Guest

Dear sirs,
I am still porting old firmware, and I have such a situation:
- a global variable named--> xdata unsigned char passw[6];
- a routine -->cmppassw(xdata unsigned char *passw)
- the code is generated ok, but the debugger won't show me the correct passw value/address, it tells to me that passw is at xdata[0] instaed of eg xdata[151f]; it seems that the debugger points to the formal routine parameter instead to the variable;
- I have modified the routine parameter name -->cmppassw(xdata unsigned char *pointer_to_passw)
and now this runs fine
- is there any switch to avoid such a behavior? This is a little boring 'cause only during debug U can note the misalignement

Thanks

Replies
Post Information Post
+1
0
-1
November 10, 2008 - 2:31pm
Guest

Thanks for the report. We will fix this issue.
Francis

+1
0
-1
November 10, 2008 - 5:26pm
Raisonance Support Team

Dear sir,

As you can see in the image above, we aren't able to reproduce your trouble.

What are your Ride version (BN) ?

Stéphane

+1
0
-1
November 17, 2008 - 3:22pm
Guest

Ride Environment
Version 06.10.22
Build Number: BN746-51

Please take note that I am working on banked code...

+1
0
-1
November 25, 2008 - 1:55pm
Guest

The 8051 tools will be soon released with various fixes for banked code on Silabs. However, the problem above should not be linked to banked mode (and it is not reproducible in the new release).