Re: First Annual EMC Users meeting?
- Subject: Re: First Annual EMC Users meeting?
- From: Matt Shaver <mshaver-at-erols.com>
- Date: Sat, 12 May 2001 01:38:44 -0400
- Content-Transfer-Encoding: 7bit
- Content-Type: text/plain; charset=us-ascii; format=flowed
- References: <01051110373504.07674-at-ray1domain.com>
- User-Agent: Mozilla/5.0 (Windows; U; Win95; en-US; 0.8.1) Gecko/20010323
Ray wrote:
> Ray nominated Matt to serve as Chair of a steering committee. The rather
> interrupted/fragmented vote confirmed Matt with three favorable and one
> opposed.
I've thought about demanding a recount ;)
> No firm bylaws were discussed and no purposes for the group or steering
> committee were formulated but all within range of this message are
> encouraged to support Matt and this initiative.
>
> Matt?
Well, lately things have gone pretty well on the EMC front. The biggest
advance lately has been Paul Corner's BDI CD which may finally get us
past the "it's too hard to install" problem. Ray did a test install on a
computer I brought to NAMES, and after answering a few questions the
install was completed in less than 8 minutes. A few folks have reported
problems with certain video cards, but overall it's a great success.
The next goal has to be rock-solid reliability. I know Jon Elson and
perhaps a few other users are using 2.0.36/0.9J/dec-1999 systems because
of random lockups experienced with the later kernel, RT, and EMC
versions. We've got to get to the bottom of this! Ray Henry has been
doing a lot of experiments trying to reproduce the problem and has had
some luck in causing the lockups (see his e-mails for details). I've got
a PC that I brought to NAMES which is now scratch and I'm going to load
the BDI and try to confirm Ray's observations. Since the problem results
in a hard lockup of the PC, Dave Engvall opined that we might need a
hardware debugger or (gulp) an emulator to track down the problem. I do
recall a simpler method of regaining control of a hung PC. What you do
is hook the NMI vector and point to your own monitor code. When you lose
control, it's possible to assert NMI by yanking on the I/O Channel Check
pin on the ISA bus. What you do from there to find the problem, I don't
know...
I'll report back when the problem is reproducable for me and/or when I
think up some more ideas.
Thanks,
Matt
Date Index |
Thread Index |
Back to archive index |
Back to Mailing List Page
Problems or questions? Contact