[nas] HPUX

[Jon Trulson]

On 30 Sep 2002 morano at ele.uri.edu wrote:

> Date: 30 Sep 2002 13:22:09 -0400 (EDT)
> From: morano at ele.uri.edu
> To: nas at radscan.com
> Subject: [nas] HPUX
> 
> > 	Oh, that's disgusting ;-)  I forgot all about that crap.  
> > Apparently, in earlier versions of HPUX (don't know about now), there was
> > a version of bcopy in some syslib that conflicted with generally accepted
> > 'standards' of bcopy.  So this mess above, would extract a known working
> > copy of bcopy() from some library and then embed it within NAS, thereby
> > removing the need for the linker to bring in the 'broken' version.
> 
> Amazing !!  I didn't know that HPUX was so messed up !!
> 

	;-) Actually, I think the original developer could have been a
little less hacky... Oh well.

> > 	I think that in this day and age of flying machines and
> > light bulbs, most systems these days support memmove().  Perhaps it's time
> > to remove the bcopy dependancies and replace them with memmove (at least
> > for the server).  That would also allow me to gleefully remove the mess
> > above.
> 
> Yes, please migrate to 'memcpy()', 'memmove()' et cetera !
> 
> What many people do for maintenance of old codes is to make #defines
> for 'bcopy()' and friends to actually call 'memcpy()' et cetera.
> But it is still better to just change the code to call the new standard 
> functions.
> 

	Yes, these defines already exist and are used on some platforms
(like linux).  I could try to do this 'in the blind' as it were and send
you the changed file(s) for you to try.

> The U.S. went to the moon back in 1969 !  I do not know how long SVID,
> X/OPEN, POSIX, et cetera have had 'memcpy()' and friends standardized
> but it would be nice to finally take advantage of that standard !! :-)
> 

	I hear you.


-- 
Jon Trulson    mailto:jon at radscan.com
ID: 1A9A2B09, FP: C23F328A721264E7 B6188192EC733962
PGP keys at http://radscan.com/~jon/PGPKeys.txt
#include <std/disclaimer.h>
Bad Color Temperature, Too much Peach.