[Mplayer-cvslog] CVS: main/mp3lib decode_k7.s,1.6,1.7
Nick Kurshev
nickols_k at users.sourceforge.net
Fri Jun 8 12:56:06 CEST 2001
Update of /cvsroot/mplayer/main/mp3lib
In directory usw-pr-cvs1:/tmp/cvs-serv18096/main/mp3lib
Modified Files:
decode_k7.s
Log Message:
Last minute improvements before release. Added banchmark
Index: decode_k7.s
===================================================================
RCS file: /cvsroot/mplayer/main/mp3lib/decode_k7.s,v
retrieving revision 1.6
retrieving revision 1.7
diff -C2 -r1.6 -r1.7
*** decode_k7.s 2001/06/08 09:41:22 1.6
--- decode_k7.s 2001/06/08 10:56:04 1.7
***************
*** 8,19 ****
/// - decreased number of opcodes (as it was suggested by k7 manual)
/// (using memory reference as operand of instructions)
! /// - added PREFETCHW opcode. It has different semantic than k6-2
/// and saves 15-25 cpu clocks for athlon.
/// - partial unrolling loops for removing slower MOVW insns.
! /// (Note probably same operation should be done for decode_3dnow)
/// - change function name for support 3DNowEx! automatic detect
///
/// note: because K7 processors are an aggresive out-of-order three-way
/// superscalar ones instruction order is not significand for them.
///
/// Modified by Nick Kurshev <nickols_k at mail.ru>
--- 8,24 ----
/// - decreased number of opcodes (as it was suggested by k7 manual)
/// (using memory reference as operand of instructions)
! /// - added PREFETCHW opcode. It has different semantic on k7 than on k6-2
/// and saves 15-25 cpu clocks for athlon.
/// - partial unrolling loops for removing slower MOVW insns.
! /// (Note: probably same operation should be done for decode_3dnow.s)
/// - change function name for support 3DNowEx! automatic detect
///
/// note: because K7 processors are an aggresive out-of-order three-way
/// superscalar ones instruction order is not significand for them.
+ ///
+ /// Benchmark: measured by mplayer on Duron-700:
+ /// 3dNow! optimized code - 1.5% of cpu usage
+ /// k7 optimized code - 1.1% of cpu usage
+ /// Note: K6-2 users have an chance with partial loops unrolling
///
/// Modified by Nick Kurshev <nickols_k at mail.ru>
_______________________________________________
Mplayer-cvslog mailing list
Mplayer-cvslog at lists.sourceforge.net
http://lists.sourceforge.net/lists/listinfo/mplayer-cvslog
More information about the MPlayer-cvslog
mailing list