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

RE: Software reboot problem

After a flashit we jump to the start of flash which is different from a
RESET. The SDRAM problem occurs when we are up and running in SDRAM and
ETRAX is reset. The SDRAM may be in a non idle state when this happens
and thus not be ready for new commands. The same applies to flashes
(but the flash memories actually have a RESET input). One solution to
the SDRAM problem may be do disable the SDRAM clock long enough for
the SDRAM to forget every state it has (not sure how long that time
is). I hope that someone else can comment on this because I haven't
investigated this thoroughly.


-----Original Message-----
From: owner-dev-etrax@xxxxxxx.com">mailto:owner-dev-etrax@xxxxxxx.com] On
Behalf Of Pieter Grimmerink
Sent: Thursday, March 11, 2004 5:12 PM
To: Mikael Starvik
Cc: dev-etrax
Subject: Re: Software reboot problem

On Thursday 11 March 2004 11:18, Mikael Starvik wrote:
> >And regarding the SDRAM clock issue, can the watchdog suffer from
> >problem?
> Yes. One idea I have had is to to shutdown the SDRAM interface
> when we receive the watchdog NMI (which happens 3.4ms before the
> real RESET).. I have not tested this yet.

What's strange is that after a ./flashit has completed, the Etrax boots
new image just fine.
Shouldn't the same problem occur, when the SDRAM causes it?