Subject: Re: details on KASSERT in uvm_map.c
To: YAMAMOTO Takashi <email@example.com>
From: Bruce ONeel <firstname.lastname@example.org>
Date: 01/24/2005 14:49:55
Thanks, I finally got back to this.
I filed pr port-mac68k/29052 which should describe things
somewhat well. I added your stack traceback call and put
that in there as well.
YAMAMOTO Takashi <email@example.com> wrote:
> Date: Tue, 11 Jan 2005 07:03:24 +0900
> From: YAMAMOTO Takashi <firstname.lastname@example.org>
> Subject: Re: details on KASSERT in uvm_map.c
> To: email@example.com
> Cc: current-users@NetBSD.org, port-mac68k@NetBSD.org
> > Hi,
> > A followup to my KASSERT in uvm_map.c. I rebuilt the install kernel
> > with a print statement so I could see what the different values
> > that caused the assert at line 478 in uvm_map.c on a mac68k
> > system.
> > The values are
> > VM_MAP_USE_KMAPENT(map) is 0
> > old_entry->flags is 5
> > UVM_MAP_QUANTUM is 8
> > UMR_EMPTY(umr) is 0
> > So it fails in that both VM_MAP_USE_KMAPENT(map) is 0 and
> > UMR_EMPTY(umr) is 0. Surely someone can point me to
> > where these might be an issue :-)
> > cheers
> > bruce
> can you check backtrace?
> TRUE, 65535, "", printf);
> will print it.
> YAMAMOTO Takashi