avrdude-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [avrdude-dev] Good news: STK500v2 and JTAG ICE mkII code committed t


From: Joerg Wunsch
Subject: Re: [avrdude-dev] Good news: STK500v2 and JTAG ICE mkII code committed to CVS
Date: Thu, 26 May 2005 22:51:34 +0200
User-agent: Mutt/1.4.2.1i

As Colin O'Flynn wrote:

> > Is the dW interface also capable of flashing an entire ROM?

> Hmm.. while it's similar to JTAG. One of the differences is though
> that you need to enable DW mode first, which requires SPI if the
> fuse isn't set. So you use the limited SPI interface of the MK-II to
> enable the DW fuse then use DW to download code, then disable the DW
> fuse.

Btw., I've meanwhile got some explanation from one of those Atmel
Noway guys about how dW works.  That ``limited SPI interface'' is
basically limited in that it can only transfer a single reply byte
from the ISP protocol back to the host.  Good enough for programming a
fuse byte, and that's all it is intented for.

debugWire is somehow still on my agenda, but probably first as a
debugging protocol, not as a method to download something.

Back to the 5.0-BETA: using Colin's USB hack (for the JTAG ICE mkII)
on Linux, I recently ran into reproducible troubles when trying to
read back the flash of something I just had downloaded.  It always
hung at the same percentange, accompanied by a message that it
received a preposterous packet length indication.  The reason for this
is not yet quite clear to me, perhaps it's even an ICE firmware bug
where the ICE is intermixing two packets (a regular reply and an event
packet for a break event).  I've got a suspicion that that could
occasionally happen (and if that's reproducible, I'll report it to
Atmel), however, upon reviewing that code, the receiver tries to
ignore the preposterous number of bytes it's just being told -- which
of course makes no sense, as it will wait forever then.  Instead, it
should simply re-enter the state machine, looking for yet another
start-of-frame indication.  I'll have to fix that, it's a showstopper
as it is right now.

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

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




reply via email to

[Prev in Thread] Current Thread [Next in Thread]