On Fri, Dec 28, 2012 at 8:57 AM, frantisek holop <min...@obiit.org> wrote: ... > i started a 'boot crash', but i am not sure > if it was finished correctly, it was sitting there > forever with the disk led on, so in the end > i just power cycled it.
It depends on how much memory you have and how fast your disk is. With 4GB of mem on my laptop, it feels like it takes 5 minutes, though I haven't actually timed it. If I'm *trying* get a crash dump (as part of debugging something particularly complicated, say), then I'll usually tell boot to lie and only use 128MB, via "machine memory =128M", before "boot bsd.test". (Note: if you do that, you must *also* do that when booting after the crash dump, so that savecore can find the dump in the swap partition.) > savecore came on and i have in the logs: > > Dec 28 00:25:25 amaaq savecore: reboot after panic: kernel diagnostic > assertion "wp->wp_new_futex == f" failed: file > "../../../../compat/linux/linux_futex.c", line 568 Excellent. The next question is whether that's the only bug that you're hitting, or if there's something else going on that should also be debugged. As for that particular failed assertion, it would be interesting to know what the actual values of wp->wp_mew_futex was (if it was NULL, then I have a guess as to the bug; if it wasn't NULL, then uh, good luck!) Philip Guenther
