kern/165863

[Ryan Stone]

2012/4/10 Gleb Smirnoff <glebius at freebsd.org>:
>  Thanks, Ryan!
(snip)
> Looks okay from my viewpoint. Have you stress tested it? My snap patch
> definitely had problems, AFAIR.
>
> If this patch fixes panics observed by kern/165863 and passes stress
> testing, then it should be committed ASAP, and shouldn't depend on
> IPv6 part.
>
> --
> Totus tuus, Glebius.

Hm, I was all ready to commit this, but I've realized that there is a
problem.  According to callout_reset(9), any caller to callout_reset
must hold any mutex associated with the callout, but the two places
that call callout_reset on the la_timer are not done with the
if_afdata_lock held, and changing that looks to be non-trivial without
making the if_afdata_lock a huge contention point.

At this point I'm not sure what the best way to proceed is.