Subject: RE: speedy FP
To: None <port-arm32@NetBSD.ORG>
From: Neil A. Carson <CARSON@rmcs.cranfield.ac.uk>
List: port-arm32
Date: 04/13/1997 07:15:04
> A better solution is to write it as softfloat library for the gcc. (can not a
> existing softfloat library from a other machine be compiled/ported to our
> system ?)
Hi,
Yep, the FP library idea is a nice one. We're currently evaluating a couple
of these to use in conjunction with the gcc soft float option. Early results
have shown a 3-6 times speed increase over using the FPE under RiscBSD.
Maybe you will all be able to see something like this in action at Wakefield.
It seems that, using a lkibrary, the AtrongARM runs FP at somewhere around the same
speed as an ARM700 with FPA11.
> Next idea: If a DSP would be present in the system, would this improve the FP
> performance if the floatingpoint requests from a (special) softfloat library
> would be calculated by the DSP ?
> I'm not shure whether the encoding/transmitting/decoding overhead for this
> mechanism still keeps the speed improvement of the DSP. (Native DSP soft is
> regions faster).
> (I ask this not without a reason ;)
> Ciao, Peter
> /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
> | |
> | >>> FutureGroove <<< |
> | |
> | Peter Berg Music production & |
> | softwaredevelopment |
> | w3: http://www.ww.tu-freiberg.de/~pberg |
> | email: Peter.Berg@ww.tu-freiberg.de |
> | part time system administrator |
> | at *.ww.tu-freiberg.de |
> \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/
Yours aye,
Neil
--
* Neil A Carson
* The Royal Military College of Science, Shrivenham
* e-mail carson@rmcs.cranfield.ac.uk
* Address: Kitchener Mess, RMCS, Shrivenham SN6 8LA. Tel: (0370) 593183