Luigi,
I've been seeing the same thing for about 6 weeks now :)
I, too can reproduce the panic _at will_. If anyone's interested I've got
a large selection of kernel cores to choose from :) ...
Andrew.
Some comments below ...
On Fri, 20 Aug 1999, Luigi Rizzo wrote:
> Hi,
>
> trying to build picobsd on 19990816snap of -current, i am consistently
> hitting a problem, which is totally reproducible from the command line.
> The following sequence
>
> vnconfig -s labels -c /dev/rvn0 fs.PICOBSD
> disklabel -rw /dev/rvn0 fd1440
>
> causes a panic with a page fault accessing a low address (presumably
> through a null pointer) in vnstrategy.
Yeah, I instrumented my kernel - and it is NULL. I think it's dev_t
related breakage but I have no proof :) . Also slices don't work.
I try doing:
vnconfig -e -s labels /dev/vn0
and then
disklabel -rw vn0 auto
and disklabel says '/dev/vn0c' not configured, or some such - it seems the
mapping between
/dev/vn0 /dev/vn0s1 (and friends)
and
/dev/vn0a /dev/vn0b (and friends)
does not work _at all_.
This has been the case since early July.
>
> fs.PICOBSD is a file filled with 0's with dd if=/dev/zero...
> and the problem appears to occur independently from the size of the
> file.
Correct. Same here.
>
> can someone investigate this ? the code seems to work fine on 3.x
>
As stated above it worked fine until sometime in July.
A point that may be of interest -
I've found panics are more likely to occur if the command pair is
executed in a script vs. from the command line. This may suggest some
sort of race is occurring.
Hope some of this is helpful... :)
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
