On Tue, Mar 30, 2010 at 8:34 PM, Kustaa Nyholm
<[email protected]>wrote:
>
> > This approach requires two byte more RAM.
> That is true
>
> >Then bitwise operation or ((unsigned char *)&add)[0] method.
>
> I did not quite follow what you wrote ... but extrapolating
> that you were offering ((unsigned char *)&add)[0] as a better
> alternative, shouldn't that be *((unsigned char *)&add)[0]?
>
>
I just need to divide the pointer into high/low byte but not need to get the
content of high/low address pointed to ;)
> Also SDCC/PIC generates poor code for pointers ...especially
> for generic pointers, so *((__data unsigned char *)&add)[0]
> could potentially be better. Don't know how well the
> optimizer would optimize any of above but I do know that
> access to global variables/arrays with constant indexes
> and struct members with are handled reasonably optimized.
>
>
I have only used sdcc to develop mcs51 programs. And the generate pointer
there is 3 bytes long. The format is a bit complicated to handle.(But I
think it's complicated for compiler, users need not to take care of that in
most cases)
--
Cheers,
Grissiom
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Sdcc-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/sdcc-user
