Re: "panic: fxp_start_body: attempted use of a free mbuf!" in RELENG_5

On Thu, Oct 07, 2004 at 04:48:44PM -0400, Brian Fundakowski Feldman wrote:
> On Thu, Oct 07, 2004 at 09:57:19AM -0700, Kris Kennaway wrote:
> > Top of RELENG_5 on an smp machine:
> > 
> > panic: fxp_start_body: attempted use of a free mbuf!
> > cpuid = 1
> > KDB: enter: panic
> > [thread 100042]
> > Stopped at      kdb_enter+0x30: leave
> > db> trace
> > kdb_enter(c06de69a,1,c06d47c5,ebbd2b14,c34cd320) at kdb_enter+0x30
> > panic(c06d47c5,c06c1030,4,c0499300,c3539000) at panic+0x14e
> > fxp_start_body(c3539000,0,c06d5639,4ca,c3539000) at fxp_start_body+0x462
> > fxp_start(c3539000,122,0,c3539000) at fxp_start+0x3f
> > if_start(c3539000,0,c06e7822,179,ffff8002) at if_start+0x99
> > ether_output_frame(c3539000,c38b5300,6,c4106b00,ebbd2bc8) at ether_output_frame+0x218
> > ether_output(c3539000,c38b5300,c4106b00,c95ec210,c04fe48d) at ether_output+0x43e
> > in_arpinput(c38b8700,122,c38b8700,c07656b8,c38b8700) at in_arpinput+0x686
> > arpintr(c38b8700,0,c06e7c88,e5,c07656b8) at arpintr+0x104
> > netisr_processqueue(c07656b8,c0738ac0,1,c06dda24,c3462940) at netisr_processqueue+0x8e
> > swi_net(0,0,c06dc037,269,c0738ac0) at swi_net+0xe9
> > ithread_loop(c345d880,ebbd2d48,c06dbe2a,323,d552d9c9) at ithread_loop+0x172
> > fork_exit(c04f1210,c345d880,ebbd2d48) at fork_exit+0xc6
> > fork_trampoline() at fork_trampoline+0x8
> > --- trap 0x1, eip = 0, esp = 0xebbd2d7c, ebp = 0 ---
> 
> Does this fix it?

I don't yet know whether this problem is repeatable, so there's no
point in testing the patch yet.  If it keeps happening, I'll try your
patch and see what it does so I can distinguish "fixed" from "null
result".

Kris