Date: Sun, 5 Jun 2005 11:41:27 +0200 From: Oliver Lehmann <lehmann@ans-netz.de> To: Martin <nakal@nurfuerspam.de> Cc: freebsd-current@freebsd.org, matt@gsicomp.on.ca Subject: Re: Optimizing libc/string for amd64(?) Message-ID: <20050605114127.018ad7ec.lehmann@ans-netz.de> In-Reply-To: <42A2BC87.5040103@nurfuerspam.de> References: <42A242D1.4000002@nurfuerspam.de> <000701c5696f$0dee3b10$1200a8c0@gsicomp.on.ca> <42A2BC87.5040103@nurfuerspam.de>
next in thread | previous in thread | raw e-mail | index | archive | help
Martin wrote:
> But before thinking about a change, it would be better if someone
> confirms the degraded performance (just run nbench). I've only checked
> it on 1 machine and cannot access it very often (it does not belong
> to me).
AMD64 3500+ running 6.0-CURRENT/amd64.
I've no i386 with aprox. the same speed I could compare with.
TEST : Iterations/sec. : Old Index : New Index
: : Pentium 90* : AMD K6/233*
--------------------:------------------:-------------:------------
NUMERIC SORT : 2151.5 : 55.18 : 18.12
STRING SORT : 185.65 : 82.95 : 12.84
BITFIELD : 4.8454e+08 : 83.12 : 17.36
FP EMULATION : 140.8 : 67.56 : 15.59
FOURIER : 17614 : 20.03 : 11.25
ASSIGNMENT : 25.529 : 97.14 : 25.20
IDEA : 2779.5 : 42.51 : 12.62
HUFFMAN : 1715 : 47.56 : 15.19
NEURAL NET : 24.339 : 39.10 : 16.45
LU DECOMPOSITION : 1145 : 59.32 : 42.83
--
Oliver Lehmann
http://www.pofo.de/
http://wishlist.ans-netz.de/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050605114127.018ad7ec.lehmann>