Anyone else having problems with lots of segfaults? NetBSD 1.2D (March 21)

I booted a 1.2D kernel yesterday and started a make build,
The make build ran fine up until it was building gnu/libexec/uucp 
where make died with a segmentation violation.

I cd'd to gnu/lib/uucp and ran make
which soon segfaulted again.  I installed the newly built make,
which also segfaulted.  gdb showed that the segfault happend in
malloc().  I'd have to re-build libc with -g to know more.
BTW, I used to always build libc.a with -g - very handy for debugging. 

Now, looking at the mail from overnight cron jobs shows lots 
of segmentation faults.  For example:

root:82# ls -l /var/log/*.core
-rw-------  1 root  wheel   12564 Mar 23 06:02 /var/log/[.core
-rw-------  1 root  wheel   49452 Mar 23 03:05 /var/log/atrun.core
-rw-------  1 root  wheel  156008 Mar 23 02:34 /var/log/awk.core
-rw-------  1 root  wheel   12564 Mar 23 02:04 /var/log/cat.core
-rw-------  1 root  wheel  143708 Mar 23 03:02 /var/log/grep.core
-rw-------  1 root  wheel   65812 Mar 23 02:00 /var/log/kill.core
-rw-------  1 root  wheel  106820 Mar 23 02:02 /var/log/sh.core
-rw-------  1 root  wheel   12564 Mar 23 10:02 /var/log/test.core

gdb shows:
gdb /usr/obj/bin/test/test /var/log/[.core
...
(no debugging symbols found)...
Core was generated by `['.
Program terminated with signal 11, Segmentation fault.
#0  0x23f8 in __main ()
(gdb) q
gdb /usr/obj/libexec/atrun/atrun /var/log/atrun.core
(no debugging symbols found)...
Core was generated by `atrun'.
Program terminated with signal 11, Segmentation fault.
#0  0x10003ae4 in end ()

Looks like something is seriously hosed.

BTW, it would be great if /usr/src could be built without having to 
install headrs or libraries... using make -I ${BSDSRCDIR}/share/mk
would even avoid having to install those... until the build has worked 
and the results can be tested - to some extent at least, installing
is rather risky to say the least.


--sjg


--sjg