Re: faster strcpy()

Re: faster strcpy()

[Philip Blundell]

>It's something like subtracting 0x01010101 from the dword and oring with 
>0x80808080 to detect the carry, but that's not quite it. Perhaps you do 

/* The following magic check was designed by A. Mycroft. It yields a     */
/* nonzero value if the argument w has a zero byte in it somewhere. The  */
/* messy constants have been unfolded a bit in this code so that they    */
/* get preloaded into registers before relevant loops.                   */

#ifdef _copywords
#  define ONES_WORD   0x01010101
#  define EIGHTS_WORD (ones_word << 7)
#  define nullbyte_prologue_() \
      int ones_word = ONES_WORD;
#  define word_has_nullbyte(w) (((w) - ones_word) & ~(w) & EIGHTS_WORD)
#endif

p.



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu

Re: faster strcpy()

[Alexander Kjeldaas]

On Fri, Apr 24, 1998 at 11:26:09AM +0100, Philip Blundell wrote:
> >It's something like subtracting 0x01010101 from the dword and oring with 
> >0x80808080 to detect the carry, but that's not quite it. Perhaps you do 
> 
> /* The following magic check was designed by A. Mycroft. It yields a     */
> /* nonzero value if the argument w has a zero byte in it somewhere. The  */
> /* messy constants have been unfolded a bit in this code so that they    */
> /* get preloaded into registers before relevant loops.                   */
> 
> #ifdef _copywords
> #  define ONES_WORD   0x01010101
> #  define EIGHTS_WORD (ones_word << 7)
> #  define nullbyte_prologue_() \
>       int ones_word = ONES_WORD;
> #  define word_has_nullbyte(w) (((w) - ones_word) & ~(w) & EIGHTS_WORD)
> #endif
> 

I'm not sure why Mycroft used '& ~(w)' in the above. Unless something
obvious escapes me, you can optimize the above down to 3
instructions. Both codepaths have a dependency chain of 3
instructions.

#define word_has_nullbyte(w) ((((w) - 0x01010101) ^ (w)) & 0x80808080)

astor

-- 
 Alexander Kjeldaas, Guardian Networks AS, Trondheim, Norway
 http://www.guardian.no/

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu

Re: faster strcpy()

[Pavel Machek]

Hi!

> > I'm not sure why Mycroft used '& ~(w)' in the above. Unless something
> > obvious escapes me, you can optimize the above down to 3
> > instructions. Both codepaths have a dependency chain of 3
> > instructions.
> > 
> > #define word_has_nullbyte(w) ((((w) - 0x01010101) ^ (w)) & 0x80808080)
> > 
> 
> Hmm.. two things obviously escaped me:
> 
> 1) The above fails horribly and detects 0x80 as a null.

It does not matter too much: you usually want 

word_has_probably_nullbyte(), since you usually want to know where the
nullbyte is and copy rest byte-by-byte.

								Pavel
-- 
I'm really pavel@atrey.karlin.mff.cuni.cz. 	   Pavel
Look at http://atrey.karlin.mff.cuni.cz/~pavel/ ;-).

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu