Looks like either the header got corrupted or your memory size goes into
the territory of unsigned numbers.

...after all, you don't *really* have 8GB of RAM in there, do you?

I do have a related question to ask, though:

Are there any plans for making a "smart" crash dump, i.e. one which dumps
only the active/required pages/mappings?  It seems absurd to me to require
a 640MB chunk of disk just for swap space.  I mean, that's about seven
hours of music, or even room to compile something.

On 30 Jan 2002, Wolfgang Rupprecht wrote:

#     # savecore -v -z /var/crash/
#     dumplo = -804327936 (-1570953 * 512)
#     savecore: /netbsd: kvm_dump_mkheader: Pread: Invalid argument
#
#     # uname -a
#     NetBSD capsicum.wsrcc.com 1.5ZA NetBSD 1.5ZA (WSRCC_ATHLON) #56:
#     Wed Jan 30 16:49:42 PST 2002
#     wolfgang@capsicum.wsrcc.com:/v/src/netbsd/src/sys/arch/i386/compile/WSRCC_ATHLON
#     i386
#
# Is savecore trying to tell me that I shouldn't be running w. an 8 gig
# swap or is there some other problem?  The (more or less) same kernel
# runs on other i386 machines without such lossage.

				--*greywolf;
--
NetBSD, Net Profit.