Dwayne Reid wrote:

> At 01:27 PM 10/20/2011, Peter Johansson wrote:
>=20
>> My suggestion is that you learn C and assembly side-by-side.  Write
>> all of your small sample programs in both C and assembly, and then
>> compare the code the C compiler generates with your own code.
>> Knowledge of just how your compiler does things can really help when
>> stepping through your code in the debugger.=20
>>=20
>> Chances are you will probably wind up coding entirely in C,
>> particularly if your compiler has a very good optimizer for your
>> MCU, but the time spent learning assembly will be time well spent.
>=20
> I've been deliberately staying out this discussion (I don't grok 'C',
> so I code exclusively in Assembler), but I think that the above
> advice is the best suggestion.=20
>=20
> If you already code in 'C' for other processors, it should be a short
> learning curve.=20
>=20
> Great advice, Peter.=20

Agreed. The first thing on a new architecture that I do is to read the
architecture and assembly manuals, then look at compiler-generated code
and try to understand it.=20

What most arguments of the type "you need to understand assembly anyway"
seem to miss is that there is a huge difference between being able to
(slowly) read and understand assembly (which I agree is necessary,
together with an understanding of the architecture), and being able to
efficiently write assembly. Programming in C doesn't make the first part
unnecessary, but the second part. You don't have to memorize all the
assembly commands, which flags they affect etc... that's the compiler's
job; I just have to know the general outlines of the architecture -- and
where I can look up the details in case I need them.

Gerhard
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.