Nope.  Again panic:

ex0: too many segments, panic: m_copydata: m == 0, len 8
Stopped at      netbsd:cpu_Debugger+0x4:        popl    %ebp
db{1}> bt
cpu_Debugger(c04bec6e,cbe37aec,7,c052b4e0,7) at netbsd:cpu_Debugger+0x4
panic(c04ba8da,8,3c,c034ca5d,7) at netbsd:panic+0x12a
m_copydata(c1b23400,0,52,c1b22e34,c1b23400) at netbsd:m_copydata+0x9b
ex_start(c1d2d03c,cbe37ba0,c1d2d03c,2,cbe37bb8) at netbsd:ex_start+0x2f6
ether_output(c1d2d03c,c1b22e00,c054f5d8,c1b38c60,c061d61c) at netbsd:ether_output+0x363
ip_output(c1b22e00,0,c054f5d4,1,0) at netbsd:ip_output+0x6b8
ip_forward(c1b22e00,1,c1f2dc00,1,c0526e14) at netbsd:ip_forward+0x189
ip_input(c1b22e00,0,cbe37f50,0,c0100e31) at netbsd:ip_input+0x2de
ipintr(c04b0010,30,c04b0010,10,cbe34000) at netbsd:ipintr+0x88
DDB lost frame for netbsd:Xsoftnet+0x49, trying 0xcbe37f58
Xsoftnet() at netbsd:Xsoftnet+0x49
--- interrupt ---
Bad frame pointer: 0xc052b4e0
0x246:

On Wed, Feb 16, 2005 at 06:12:32PM -0500, Christos Zoulas scribed:
> On Feb 16,  8:25pm, dlagno@mail.ru (Denis Lagno) wrote:
> -- Subject: kern/29407: panic: m_copyback0: read-only
> 
> | >How-To-Repeat:
> | >From scarce observations this panic arises when my first ftp access
> | is from some other host in the local network.  The host that panics
> | is a gateway to Internet.  I use NAT and ipsec.
> | I first observed this panic with week-old -current.  I updated system
> | and the panic persist.
> | >Fix:
> | N/A
> 
> Try this:
> 
> Index: uipc_mbuf.c
> ===================================================================
> RCS file: /cvsroot/src/sys/kern/uipc_mbuf.c,v
> retrieving revision 1.92
> diff -u -u -r1.92 uipc_mbuf.c
> --- uipc_mbuf.c	24 Jan 2005 21:25:09 -0000	1.92
> +++ uipc_mbuf.c	16 Feb 2005 23:11:52 -0000
> @@ -1077,7 +1077,7 @@
>  #if defined(DEBUG)
>  	error =
>  #endif /* defined(DEBUG) */
> -	m_copyback0(&m0, off, len, cp,
> +	m_copyback0(&m0, off, len, cp, M_COPYBACK0_COW|
>  	    M_COPYBACK0_COPYBACK|M_COPYBACK0_EXTEND, M_DONTWAIT);
>  
>  #if defined(DEBUG)