lkml.org 
[lkml]   [1998]   [Aug]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Winmodem support, some performance tradeoff estimates
On Sun, 16 Aug 1998, Alan Cox wrote:

> > means the performance of using a soft modem is about 8 Pentium MHz,
> > on the grounds that 1 32-bit Pentium MHz =~ 1 64-bit R4000 MHz in the
>
> Way off. Firstly Im dubious about the 8Mhz, secondly the R4K's tend to
> have DSP instructions (ones that are useful). Pentiums dont have an
> add/multiply with no stall instruction once per clock.

An unrolled multiply accumulate _can_ be done in 2 clocks per argument on
a Pentium, however (hint: the fxchg instruction can be made to take 0(!!)
clocks if ordered properly). I put together a signal processing app that
did dot products at 45 mflops on a P90 last year. But this was only if its
working set fit within the L1 cache.

With MMX, a 16-bit fixed point multiply accumulate could probably be done
with about about a cycle per operand (multiply, shift, add, exchange)
provided your vectors could be into four chunks large enough to unroll
nicely.

--
"Love the dolphins," she advised him. "Write by W.A.S.T.E.."


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html

\
 
 \ /
  Last update: 2005-03-22 13:44    [W:1.159 / U:0.316 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site