Re: [avr-gcc-list] RE: avr-gcc-list Digest, Vol 6, Issue 10

[Joerg Wunsch]

"Dafni & Robert Berger" <address@hidden> wrote:

> If took me a couple of weeks to figure out, that the latest WinAVR
> (also with beta patches) corrupts your code in the elf to coff
> conversion.

Well, you know that the COFF conversion is currently marked `beta'.
So if you feel that something corrupts your code (which IMHO cannot
even happen since the conversion only touches the debugging symbols
but not the code itself), i'd like to hear the details from you, so
it can eventually be fixed.

> Now the biggest drawback not using AvrStudio is, that you loose a
> nice (and in the latest version of AVR Studio also mainly working)
> simulator. In my experiments it exhibits exactly the same behavior
> as running with the JTAG-ICE on the actual hardware, which is pretty
> neat.

Well, except for debugging algorithms, i have yet to see a simulation
that comes close enough to simulate reality. ;-)  Even with VMLAB which
i found offers some more options for simulating the electrical
environment around the controller, i'd have a hard time simulating
simple things like the LC-meter i'm currently building.

But you're right, simulavr is still far away from both, AVR Studio and
VMLAB, in terms of IO simulation.  OTOH, for debugging an algorithm,
i've found it much more valuable to have the full power of GDB instead
of a GUI that seems to be always getting into my way instead of
helping me debugging...  That's just my personal opinion, of course.

-- 
J"org Wunsch                                           Unix support engineer
address@hidden        http://www.interface-systems.de/~j/