On 11 Jun 2009, at 5:41 PM, Paul B. Mahol wrote:
Looks like boot(8) is problematic.
Anything in /etc/src.conf or /etc/make.conf?
I have never touched or created a src.conf. If there was one there,
it has been unmodified by me.
I HAVE modified make.conf. Here is its contents:
--- /etc/make.conf ---
BATCH=yes
NO_PROFILE=yes
KERNCONF=DKA
USE_FORTRAN=yes
WITH_JADETEX=no
PERL_VERSION=5.8.9
---
My custom kernel named DKA has only three modifications from GENERIC:
I commented out the following three lines:
#cpu I486_CPU
#cpu I586_CPU
#makeoptions DEBUG="-g"
I have run with such a kernel on many machines for many years now.
However, my experiments have shown that the kernel build is not to
blame.
Isn't boot part of the kernel build? Why would installing the kernel
not cause this problem?
It is by installing the world via make installworld that my drive gets
munged.
I obviously am missing something, but boot(8) sounds like it is in the
neighborhood of where the problem is.
There were changes to /usr/src/sys/boot/i386/libi386/biosdisk.c and
biospnp.c that really look suspect to me.
Dan
_______________________________________________
[email protected] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[email protected]"
