[ofiwg] completion flags as actually defined by OFI

[Jason Gunthorpe]

On Tue, Apr 14, 2015 at 06:37:50PM +0000, Hefty, Sean wrote:
> > But why should an app author care? What is the use case?
> 
> This was an MPI requirement.

Okay, but reading the docs you provided, I can't understand what the
requirement is.

> The point of my post wasn't even about whether these 3 flags were
> desirable.  It was whether a 4th flag
> (FI_DATA_VISIBLE_AT_REMOTE_SIDE_BUT_NOT_YET_IN_FINAL_DESTINATION_BUFFER)
> was needed.

You mean, the WR is guarenteed to complete on the peer, but the peer
may not have seen it yet?

I think that is very useful, and should replace FI_TRANSMIT_COMPLETE.

I would suggest a weak semantic: true under normal operation, clean
socket close, no program crash, no peer error.

FI_COMMIT_COMPLETE is the strong semantic of the same idea.

-- 
Jason Gunthorpe <jgunthorpe at obsidianresearch.com>        (780)4406067x832
Chief Technology Officer, Obsidian Research Corp         Edmonton, Canada