[ewg] rping/cxgb3 regression

[Steve Wise]

Hey Sean,

Can you peruse:

http://bugs.openfabrics.org/bugzilla/show_bug.cgi?id=2230

The changes I added to rping to fix problems I was seeing when running rping over the upstream iw_cxgb3 have been 
included in OFED-1.5.3.  That change, however breaks rping over iw_cxgb3 in 1.5.3.  It causes a hang at the end of the 
rping run.  The problem is not with rping, but with the down-level iw_cxgb3 code.  The upstream change to fix iw_cxgb3, 
however, isn't trivial and it bumps the cxgb3 uverbs ABI.  That's why I didn't pull it into 1.5.3.  I'm hesitant to pull 
in the iw_cxgb3 (and required libcxgb3-1.3.0) since we're at RC4 and going to try and GA next week.

I'm wondering if pulling the rping changes for ofed-1.5.3 would be ok?  I guess to do this you would have to push a 
1-off librdmacm without those changes?  Or maybe back up what is in OFED-1.5.3 to the previous release without this 
rping change?

Thoughts?

Sorry about this regression.

Steve.