[PATCH 0/3] Fix request completion holes

[Jason Gunthorpe]

On Wed, Nov 01, 2017 at 06:02:38PM +0200, idanb wrote:

> (3) No send signal with invalidation signal w/o moderation 5.2M / 8.3M

Actually, isn't this valid and reasonable? You don't need to see the
SEND completion, the only thing that matters is the invalidate
completion, and the invalidate will naturally order after the send
is finished, yes?

Jason