Kernel panic on boot at dm_crypt

[Stefan Unterweger]

Commit f41543e16b49e0294028d1f4bb5a5667694536f1 looks interesting.  At
least from a cursory glance it seems as if it might be close to the
parts of the code that are giving me problems.  I am compiling right
now, as soon as I manage to push it to the server, I‘ll report back.

| Author: Sepherosa Ziehau <sephe at dragonflybsd.org>  2016-07-06 10:17:21
| Committer: Sepherosa Ziehau <sephe at dragonflybsd.org>  2016-07-06 10:19:45
| Parent: 695396069af271146f6e4b3d2a4d3012466e11e1 (libc/confstr: Fix comment indentation.)
| Child:  bc6e3a6df1cd227d9b372bb33d9c17c3c0c23d1a (Merge branch 'master' of git://git.dragonflybsd.org/dragonfly into Rechner/Nur-Ab-Sal)
| Branch: Rechner/Nur-Ab-Sal
| Follows: Stackframe/344/nachher
| Precedes: 
| 
|     virtio: Fix MSI support; thus unbreak booting on bhyve
|     
|     I'd like to thank Peter Grehan <grehan freebsd org> very much for
|     providing various information on bhyve side and helping testing
|     this patch.
| 
| ------------------- sys/dev/virtual/virtio/pci/virtio_pci.c -------------------
| index f131109..8c67213 100644
| @@ -851,7 +851,7 @@ vtpci_alloc_msi(struct vtpci_softc *sc)
|              &sc->vtpci_irq_rid, &irq_flags);
|  
|  
| -	return (1);
| +	return (0);
|  }
 

* Stefan Unterweger on Wed, Jul 06, 2016 at 02:59:20AM +0200:
> I forgot to add the output of ‚trace‘, here it is, beginning from the
> most recent panic.
> 
> | db> trace
> | mpipe_alloc_callback() at mpipe_alloc_callback+0x13d 0xffffffff80606056
> | dmtc_crypto_read_start() at dmtc_crypto_read_start+0x54 0xffffffff826bd572
> | dmtc_bio_read_done() at dmtc_bio_read_done+0x2c 0xffffffff826bd6ff
> | biodone() at biodone+0x110 0xffffffff8066bc77
> | vtblk_vq_intr() at vtblk_vq_intr+0x129 0xffffffff808cd087
> | virtqueue_intr() at virtqueue_intr+0x26 0xffffffff808d2dc7
> | vtpci_legacy_intr() at vtpci_legacy_intr+0x58 0xffffffff808d17b5
> | lwkt_serialize_handler_call() at lwkt_serialize_handler_call+0x79 0xffffffff80620542
> | ithread_handler() at ithread_handler+0x242 0xffffffff805d898c
> 
> In case of typos, I’ve again attached a screenshot.  I have been
> observing the traces of the last few crashes.  The traces are
> practically identical, down to most of the exact memory locations of
> every single line.

-- 
The reasonable man adapts himself to the world; the unreasonable man
persists in trying to adapt the world to himself. Therefore, all
progress depends on the unreasonable man.
		-- George Bernard Shaw