Quoting Jack Morgenstein <[EMAIL PROTECTED]>: > Prevent issuing multiple MAD transactions with the same TID. > Could happen if duplicate requests are posted. > > Signed-off-by: Jack Morgenstein <[EMAIL PROTECTED]> > Signed-off-by: Michael S. Tsirkin <[EMAIL PROTECTED]>
This is actually affecting duplicate responses as well, isn't it? As far as I remember, duplicate RMPP responses corrupting each other is something that we saw in the lab with opensm, so it would be nice to have that fixed for 2.6.17 (or even for 2.6.16 if possible?). I think the problem with this specific patch was that ib_find_send_mad ignores the remote address (and method?) when doing lookup, so its wrong to use it to check that the transaction is in progress. This shouldnt be hard to fix, though, should it? We should just check the packet method, and compare source/destination GIDs as appropriate. The remote GID can be got from ib_query_ah. Right? -- Michael S. Tsirkin Staff Engineer, Mellanox Technologies _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
