IRQ latency (was Re: RTEMS-MVME5500 BSP v1.3 available)
feng1 at bnl.gov
Tue Aug 9 18:49:58 UTC 2005
"Joel Sherrill " wrote:
> Do you know the difference between your 1.3 and what is on the
> 4.6 branch?
The difference is in the ChangeLog.
> Kate Feng wrote:
> > I will try to sync with the current CVS and new tools as soon as I
> > managed to get more disk space on my PC.
> > As I was passing by the New York city last week, I started to think that
> > one might get better performace on IRQ latency and context switch
> > using the new powerPC shared files and new compiler.
> You mean it might be possible to squeeze even more performance out of
> this BSP? :) It is already doing well in the benchmarks.
If you think it's already doing well, then I probably do'nt have to
try the RTEMS4.7.
I thought there is some discrepancy between my result and Peter's.
It seems to me the following description uses interrupts.
Peter Dufault wrote :
> Data acquisition interrupt (copies data off PCI board, posts
> semaphore): 22KHz 15KHz
The former one is for RTEMS and the later one is for vxWorks.
RTEMS interrupt outperformed vxWorks by 30%, which I interprete
it as the "average case" instead of the "worst case". However,
looking at the average performance I got even for the loaded
system (attached at the end), it seems that RTEMS is 30%
slower on the "average" case. This puzzles me. The difference
between my RTEMS system and his is :
1. RTEMS4.6 with gcc-3.2 v.s. RTEMS4.7 with gcc-4.0
2. optimized v.s. non-optimized.
3. system bus counter's IRQ v.s. PMC module's IRQ, which should
not make a difference for the optimized version.
4. native O.S. thread v.s. pthread for the benchmark software.
Friday, July 29:
MVME5500 Interrupt Latency | Context Switching
max (average) | max (average)
RTEMS 5.04 (3.45) | 6.80 (0.96)
vxWorks 6.10 (1.58) | 9.65 (0.91)
RTEMS 8.17 (3.74) | 17.48 (1.69)
vxWorks 13.90 (1.68) | 20.80 (1.90)
** All units are in usec.
More information about the users