On Wed, Jul 30, 2008 at 7:41 AM, Michael Fischer <[EMAIL PROTECTED]> wrote:
> Hello Spen,
>
> please add your patch if it is working. The time will
> show if it is OK or not. Like with other feature before.

Lots of changes have been rejected to this code. Just
becuse this is the latest change, doesn't mean that
it shouldn't compete on the same terms as other
performance suggestions to this code.

The problem with this change is that we already know it will
cause some performance regressions.

It is a performance optimisation and performance optimisations
should always be backed by numbers/measurements.
Optimization should be 90% profiling 10% programming,
generally.

This is code that has already been optimized, so this change
should compete with other changes to this code on the same
terms: documented performance improvement.

I'm not asking for reems of documentation, but *some*
evidence that it is in fact faster.

One idea that came to me is that packets smaller than 512
bytes or so will probably be slower. There are many small
packets. The ethernet hardware has a lower limit to the
packet length for ethernet, so it makes no sense to try
to compress those packets.

Also the operating system will have a lot of TCP/IP optimisations.

-- 
Øyvind Harboe
http://www.zylin.com/zy1000.html
ARM7 ARM9 XScale Cortex
JTAG debugger and flash programmer
_______________________________________________
Openocd-development mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/openocd-development

Reply via email to