Re: ser_posix.c: is select()ing for the write descriptoruseful at all?

As Nick Lott wrote:

> I have tested the patch on the following with no problems.

Thanks for the test!

> I did notice that the new version is 1~2 seconds slower but this
> could be to do with other factors as I'm comparing to version 5.1.

Hmm, it would be interesting to analyze anyway.  Maybe you could
ktrace both, and try nailing it down a bit where that happens?

The only thing where it is possible that it might be slower is using
plain "-c stk500", as that now first tries to use version 1 protocol,
and failing that (i. e. timing out on it), proceeds to version 2.

-- 
cheers, J"org               .-.-.   --... ...--   -.. .  DL8DTL

http://www.sax.de/~joerg/                        NIC: JW11-RIPE
Never trust an operating system you don't have sources for. ;-)