Ok, so I'm testing the 1.4.x release branch under VMWARE on Windows
2000.  I have a 32MB VM with a 2GB EIDE virtual disk, and the very
latest -release.  Doing a 'make cleandir' in /usr/src will eventually
cause make to SEGV.  If I relink make as a static executable, it appears
to work fine.  (at least 'make build' completes, minus make)

My -current VM is exactly the same (32MB/2GB), except for the MAC
address, and it is the very latest -current.  The problem does *NOT*
happen under -current.  I am able to run 'make build' repeatedly with
no problems.

What is the best way to debug this?  Using gdb on the core, I see
the following: (from 'make cleandir >& make.out' in /usr/src)

Core was generated by `make'.
Program terminated with signal 11, Segmentation fault.
Reading symbols from /usr/libexec/ld.so...done.
Reading symbols from /usr/lib/libc.so.12.40...done.
#0  0x48aa in Cond_Eval (line=3D0x284a00 ".if !defined(UPDATE)") at cond.c:=
1274
1274		condStack[condTop] =3D value;
(gdb) where=0D
#0  0x48aa in Cond_Eval (line=3D0x284a00 ".if !defined(UPDATE)") at cond.c:=
1274
#1  0xd958 in ParseReadLine () at parse.c:2373
#2  0xdd70 in Parse_File (name=3D0x9e1f "Makefile", stream=3D0x400a7534)
    at parse.c:2624
#3  0xa85a in ReadMakefile (p=3D0x9e1f, q=3D0x0) at main.c:897
#4  0xa463 in main (argc=3D3, argv=3D0xefbfd9d4) at main.c:719
[ a second core dump ]
Core was generated by `make'.
Program terminated with signal 11, Segmentation fault.
Reading symbols from /usr/libexec/ld.so...done.
Reading symbols from /usr/lib/libc.so.12.40...done.
#0  0x48aa in Cond_Eval (line=3D0x34200 ".else") at cond.c:1274
1274		condStack[condTop] =3D value;
(gdb) where=0D
#0  0x48aa in Cond_Eval (line=3D0x34200 ".else") at cond.c:1274
#1  0xd97e in ParseReadLine () at parse.c:2378
#2  0xdd70 in Parse_File (name=3D0x9e1f "Makefile", stream=3D0x400a7534)
    at parse.c:2624
#3  0xa85a in ReadMakefile (p=3D0x9e1f, q=3D0x0) at main.c:897
#4  0xa463 in main (argc=3D3, argv=3D0xefbfd9d0) at main.c:719

One of the reasons I'm asking about this, is because NetBSD 1.4.2=20
really needs to be stable under VMWARE, at least for PR reasons.
Currently FreeBSD/Linux/Windows are "supported" guest operating
systems, while NetBSD/OpenBSD/BSDi aren't even mentioned.

I can get a ktrace, or anything else that is necessary.=20

-Andrew
--=20
-----------------------------------------------------------------
Andrew Gillham                            | This space left blank
gillham@whirlpool.com                     | inadvertently.
I speak for myself, not for my employer.  | Contact the publisher.