Re: Timings for optimised poll(2)
Keith Owens (kaos@ocs.com.au)
Mon, 25 Aug 1997 12:08:26 +1000
On Sat, 23 Aug 1997 21:07:44 -0400,
Peter Leif Rasmussen <peter@udgaard.isgtec.com> wrote:
>And just for fun I compiled it and ran it on a 486DX2/66MHz, a 486SLC2/50MHz
>and a 386SX/20MHz, and my numbers said:
>
>486DX2/66MHz: 4200 us (linux 2.0.30)
>486SLC2/50MHz: 10000 us (linux 2.1.51)
>386SX/20MHz: 24400 us (linux 2.0.29)
>
>Without having access to any Pentium computer, I find it difficult to believe
>that you got 1.5 us with your test program?! If you did, could you give a point
>of view as to why you think the numbers are so different?
386DX/40 ISA 27,700 us (2.0.30)
486DX2/66 ISA 12,000 us (2.0.30)
P100 PCI 1,600 us (2.0.30)
P166 (dual) PCI 1,300 us (2.1.51)
Binary compiled with -O2 on P166, copied to other systems. Suspect
cache effect.