RE: [ECOS] c++ ecos performance

[Schmidt Henning Larsen <HenningLS at danfoss dot com>]

> 
> Grant Edwards wrote:
> > I've compared timing info for eCos, ThreadX, Nucleus, and one
> > other I don't remember.  All but eCos were written in C.  eCos
> > times for context switching and whatnot were all pretty much
> > the same as the others.

Hi.
Is it possible to see the testresults, and maby you could tell how you made
them.
Because the only times I have, is a sheet from ThreadX saying:
(ARM7, 20MHz)
context switch: 4,8 uS
thread suspend: 7,5 uS
semaphore put:  1,8 uS
queue send:     5,9 uS
...

If I compare these to ECos:
(ARM7TDMI, 20MHz)
thread switch:  49,14 uS
thread suspend: 15,65 uS
semaphore wait: 17,35 uS
mbox put:       20,70 uS
...

The difference is eye-catching.
Unfortunately I hasnt got the time to do any test myself, but I find it hard
to believe that the difference is that high.
Are there others who have compared eCos and TreadX?

Henning S Larsen



-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss