Discussion:
How do I debug this?
(too old to reply)
Jonathan de Boyne Pollard
2011-03-04 04:13:49 UTC
Permalink
Raw Message
I suspect that Dani's IDE driver cannot write to the dump partition.
No, the protect mode drivers are not involved. System dumps to the
SADUMP partition are written using BIOS I/O and the OS2DUMP code is
running in real mode.
It's more likely, given what's said elsewhere in this thread, that the
failure of the system dump is because the ATA channel is in a state that
the firmware code doesn't expect as an initial state, and is really a
red herring.
Steven Levine
2011-03-05 01:25:57 UTC
Permalink
Raw Message
On Fri, 4 Mar 2011 04:13:49 UTC, Jonathan de Boyne Pollard
<J.deBoynePollard-***@NTLWorld.COM> wrote:

Hi,
Post by Jonathan de Boyne Pollard
It's more likely, given what's said elsewhere in this thread, that the
failure of the system dump is because the ATA channel is in a state that
the firmware code doesn't expect as an initial state, and is really a
red herring.
I would agree for the case where the system is already hung. However,
at the moment, Al is testing with Ctrl-Alt-Numlock-Numlock while the
system is in a stable state. I've not seen problem with this, unless
the system has more than 2GB of RAM in which case there's the DUMPFS
version of OS2DUMP which works on some systems.

Steven
--
---------------------------------------------------------------------
Steven Levine <***@earthlink.bogus.net>
eCS/Warp/DIY etc. www.scoug.com www.ecomstation.com
---------------------------------------------------------------------
Loading...