On Wed, 2005-05-04 at 16:00 -0400, Paul Mather wrote:
> I upgraded my -CURRENT system (kernel and userland) not too long ago
> (from 2.99.* to 3.99.3) and noticed a day or so afterwards the panic
> "panic: m_copydata: m == 0, len 1" had occurred.  I decided to ignore it
> unless it happened again.  Well, it happened again, today, ~1 week since
> the last occurrence.
> 
> This is what I have in my dmesg:
> 
> panic: m_copydata: m == 0, len 1
> syncing disks... 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 giving up
[snip]
> Has anyone else noticed this panic?  I guess I could try updating to a
> newer -CURRENT and hope this was just a transient problem in the code
> base.

I'm pretty sure this is related to the new TCP SACK code. I've worked
around the panic on my two affected machines (one alpha, one sparc) by
setting:

  sysctl -w net.inet.tcp.sack=0
  sysctl -w net.inet6.tcp6.sack=0

Cheers,
Jared