[ofa-general] Re: [PATCH] ib/cm: fix stale connection detection

[Sean Hefty]

> The patch looks obvious enough for 2.6.22, safe enough in that it replaces a
> timeout with a reject, and it addresses a real problem.  Sean? Roland? What do
> you think?

To make it easier, I've added the patch to:

	git://git.openfabrics.org/~shefty/rdma-dev.git for-roland

commit 2fbe169db0c6bddcc7b28d03eb51d057277ffd6a

I'm comfortable with this merging into 2.6.22 myself.

> Something that I think would be very useful for 2.6.22 already: could you please
> document which portions of chapter 12 are not currently implemented in cm.c, and
> put this in some file in kernel tree?  This way people will be able to figure
> out whether something that they need is missing, and contribute.

This isn't something that I know without comparing the code against the 
spec.

>> Also, I left the duplicate request handling
>> as it was, since that should go in as a separate patch.
> 
> Could you please describe what is missing currently?
> Is the missing handling likely to cause timeouts?

If two REQs are received with matching local IDs, but the REQs 
themselves differ on one or more fields, such as the QPN, the second REQ 
is dropped as a duplicate.  This causes timeouts, so I need to figure 
out what the correct behavior should be here.

- Sean