Tom Vijlbrief wrote:
> Performance is based on getting 4 things right:
>
> 1) Designing a performance capable architecture
> 2) Pinpointing performance problem areas
> 3) Choosing the right algoritms
> 4) Efficient low level coding
>
> I have been busy working on computer chess programs the last 15 years,
> a class of programs where every cycle wasted is a death crime.
> This included all 4 subjects mentioned above, including writing
> Pentium assembly code to keep both instruction pipelines occupied.
> So I guess this should qualify me to have an opinion on performance.
>
> I would estimate the impact of the write-cache performance
> to be less than 0,001% on the total plex86 performance based on
> the following reasoning:
Good 4 points above. I like when people do some thinkings on
performance impact before they chose tools. Obviously you've
done much of this. Based on your estimates of total performance
impact, you have a good argument for using the C++ class to
yield better coding productivity.
I'll commit your write-cache code mods soon.
If library execution performance is not an issue, I'm still
interested in memory consumption by the map algorithms. Memory
use has a way of pushing other critical pieces of data out of
the processor caches. At some point, it would be cool to have some
stats on memory consumption for the write cache on some real guest
OS runs.
-Kevin
--
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Kevin Lawton [EMAIL PROTECTED]
MandrakeSoft, Inc. Plex86 developer
http://www.linux-mandrake.com/ http://www.plex86.org/
