Re: [avrdude-dev] [patch #4372] Better synchronization for stk500

[Joerg Wunsch]

As Bernard Fouché wrote:

> So this is a cable/driver problem, not related to the upper
> application.

But that's no excuse for not retrying after a communications failure.
The STK500 is suitable to offer retries, as it can for sure recover
from that.  I'm suffering from a PCI multiport board myself that
occasionally loses an interrupt (and thus a response from the STK).
I've already made the STK500v2 and JTAG code retrying failed
communication (albeit the STK500v2 code takes abysmally long to
actually start a retry).

Based on the description of the patch, I'm not sure though the
proposed solution (blindly sending a few Sync commands, and only
then start communication) is the best one though, as it would only
catch a single case of communication failure.

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

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