On Wed, Jan 04, 2017 at 07:37:32PM +0000, Christos Zoulas wrote: > In article <20170104185302.GB5015@quark>, > Patrick Welche <pr...@cam.ac.uk> wrote: > >On Wed, Jan 04, 2017 at 12:52:54PM -0500, Christos Zoulas wrote: > >> On Jan 4, 3:45pm, pr...@cam.ac.uk (Patrick Welche) wrote: > >> -- Subject: Re: npf related panic > >> > >> | npf seems blameless: see John Baker's message - we have the same stack > >> trace > >> | below pfil_run_addrhooks. (amd64 stack trace has pfil.c 1.30, the > >RPI probably > >> > >> Can you add a check printing: > >> > >> printf("npf=%p ifa=%p\n", npf, ifa); > >> npf_ifaddr_sync(npf, ifa->ifa_ifp); > >> > >> in src/sys/netinet/npf/npf_os.c in the address hook function? > > > >npf=0xfffffe81162fe7d0 ifa=0x0 > > > >sure enough... null ifa... > > Should be fixed...
Thanks! (RPI images good too...) P