[ofa-general] rdma_cm issues in 2.6.21-rc1

[Sean Hefty]

Sean Hefty wrote:
> As just a note, I'm investigating two issues with the rdma_cm and 2.6.21-rc1.
> 
> Running ucmatose twice results in a failure binding to an address the second
> time that it's run.
> 
> I'm also seeing a kernel crash if ucmatose is killed while waiting for a
> connection.

Just an update - I think I've found the issue.  The merger of the multicast 
support added with kzalloc -> kmalloc cleanups left a field uninitialized.  I'll 
have a patch shortly.

- Sean