> I can't take credit for peekpoke.h; that was done by another
developer
> (Jo if I recall); i know it's not mandatory but looks good for
speed
> optimization.
"Jo" , (J)esse (O)ff maybe? ;-)
Those inline assembly routines are more than just for speed
optimization. Actually, gcc will emit the equivalent on regular
pointer dereferences, so its probably not really any faster most of
the time.
The real reason for those is for the byte and halfword (16bit)
reads/writes. GCC actually won't use the byte and halfword
loads/stores unless compiled explicity with the option "-mcpu=arm9".
Instead it will emit regular 32-bit load/stores with appropriate
shifts/masks. This isn't a big deal on ep93xx register space since
mostly every register is 32 bits anyway, but on the memory mapped
8/16bit PC104 bus it could cause havoc and make things 2-4x slower
than necessary. Rather than make proper functioning behavior of a
program depend on the -mcpu option, I've gotten into a habit of just
using those inline asm routines.
//Jesse Off
------------------------ Yahoo! Groups Sponsor --------------------~-->
<font face=arial size=-1><a
href="http://us.ard.yahoo.com/SIG=12hs02gri/M=362329.6886308.7839368.1510227/D=groups/S=1705005378:TM/Y=YAHOO/EXP=1124901924/A=2894321/R=0/SIG=11dvsfulr/*http://youthnoise.com/page.php?page_id=1992
">Fair play? Video games influencing politics. Click and talk back!</a>.</font>
--------------------------------------------------------------------~->
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/ts-7000/
<*> To unsubscribe from this group, send an email to:
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|