> The ire entry is not maintained correctly during the plumb/unplumb
 > process. The packet destined to your driver was sent out through bge
 > interface and bge does not check the packet length and it tries to copy
 > the large LSO packet to it's own buffer. 6586787 has been filed for
 > this.

I'm not sure if I follow the claimed cause of this bug.  Specifically, the
claim seems to be that the IRE changed between the time tcp_send() did the
tcp_send_find_ire_ill() and the time tcp_lsosend_data() actually used the
IRE.  But we have a refhold on that IRE the entire time so the IRE can't
go away, and furthermore the ire_stq for a given IRE never changes over
its lifetime.

--
meem
_______________________________________________
networking-discuss mailing list
[email protected]

Reply via email to