Messages in this thread | | | Date | Sat, 26 Oct 2002 14:11:09 +0200 | From | Jurjen Oskam <> | Subject | Re: [CFT] faster athlon/duron memory copy implementation |
| |
On Thu, Oct 24, 2002 at 07:15:43PM +0200, Manfred Spraul wrote:
> Attached is a test app that compares several memory copy implementations. > Could you run it and report the results to me, together with cpu, > chipset and memory type?
joskam@hobbes:~> gcc -v Reading specs from /usr/lib/gcc-lib/i486-suse-linux/2.95.3/specs gcc version 2.95.3 20010315 (SuSE) joskam@hobbes:~> gcc athlon.c joskam@hobbes:~> ./a.out Athlon test program $Id: fast.c,v 1.6 2000/09/23 09:05:45 arjan Exp $
copy_page() tests copy_page function 'warm up run' took 19425 cycles per page copy_page function '2.4 non MMX' took 22285 cycles per page copy_page function '2.4 MMX fallback' took 21698 cycles per page copy_page function '2.4 MMX version' took 19587 cycles per page copy_page function 'faster_copy' took 11082 cycles per page copy_page function 'even_faster' took 11203 cycles per page copy_page function 'no_prefetch' took 7140 cycles per page joskam@hobbes:~> ./a.out Athlon test program $Id: fast.c,v 1.6 2000/09/23 09:05:45 arjan Exp $
copy_page() tests copy_page function 'warm up run' took 19463 cycles per page copy_page function '2.4 non MMX' took 22378 cycles per page copy_page function '2.4 MMX fallback' took 21863 cycles per page copy_page function '2.4 MMX version' took 19558 cycles per page copy_page function 'faster_copy' took 11036 cycles per page copy_page function 'even_faster' took 11292 cycles per page copy_page function 'no_prefetch' took 7152 cycles per page joskam@hobbes:~> ./a.out Athlon test program $Id: fast.c,v 1.6 2000/09/23 09:05:45 arjan Exp $
copy_page() tests copy_page function 'warm up run' took 21519 cycles per page copy_page function '2.4 non MMX' took 22224 cycles per page copy_page function '2.4 MMX fallback' took 21707 cycles per page copy_page function '2.4 MMX version' took 19399 cycles per page copy_page function 'faster_copy' took 11002 cycles per page copy_page function 'even_faster' took 11211 cycles per page copy_page function 'no_prefetch' took 7147 cycles per page joskam@hobbes:~> cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 6 model : 6 model name : AMD Athlon(TM) XP1700+ stepping : 2 cpu MHz : 1477.400 cache size : 256 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov pat pse36 mmx fxsr sse syscall mmxext 3dnowext 3dnow bogomips : 2949.12
This was run on an Asus A7V266-E motherboard with a KT266A chipset, with 512 MB of DDR SDRAM.
-- Jurjen Oskam
PGP Key available at http://www.stupendous.org/ - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |