This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: Performance of Integer Multiplication on PIII (gcc-2.95.2 onP6)



And now something completely different:

gcc-2.95.2 (haifa) on PII (P6 Modell II)-400
(SuSE-6.1, glibc-2.0.7, GNU ld version 2.9.1 (with BFD 2.9.1.0.22),
 unmodified)

gcc -O2 -march=i686 -s -o imul imul.c read_empty.c read.c
 Loop: 1.93, Code: 3.13
 Clocks: 23.32

-ffplfj=2:= -fomit-frame-pointer -malign-loops=2 -malign-functions=2
            -malign-jumps=2

gcc -O2 -march=i686 -ffplfj=2 -s -o imul imul.c read_empty.c read.c
 Loop: 1.18, Code: 3.48
 Clocks: 25.93

-flfj=2:= -fomit-frame-pointer -malign-loops=2 -malign-functions=2
            -malign-jumps=2

gcc -O2 -march=i686 -flfj=2 -s -o imul imul.c read_empty.c read.c
 Loop: 1.92, Code: 3.12
 Clocks: 23.24

... read_[i]asm.s ... will not work with intel_syntax ... (Tex motto
applies: user can work it out ...)

Thus, it seems, we are at status quo ante?

Peter


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]