[ofa-general] Re: [PATCH] ipoib/cm: compliance fix

[Roland Dreier]

thanks, applied

 > Important enough for -stable?

I don't think so -- there's no way to hit this in practice right now,
so I don't think it meets the -stable criteria.

BTW, looking at the code that happens to be in the patch context:

 > 	if (!likely(wr_id & IPOIB_CM_RX_UPDATE_MASK)) {

I think this annotation is unclear and I'm not sure gcc will do what
is intended here (and I'm not sure what is intended).  Should this be

	if (likely(!(wr_id & IPOIB_CM_RX_UPDATE_MASK))) {

or

	if (unlikely(!(wr_id & IPOIB_CM_RX_UPDATE_MASK))) {

...seems as if "unlikely" is appropriate.

 - R.