[ofa-general] [PATCH/RFC] Add support for "send with invalidate" to libibverbs

[Roland Dreier]

 > Why do you need the flag IBV_DEVICE_MEM_WINDOW?
 > If the value of device_attributes.num_mw is more than zero => the
 > device supports memory windows, so i think this flag
 > can be safely removed.

OK, I'll delete it from the libibverbs changes.  I guess we can kill it
on the kernel side too.

 > I think that the send & invalidate should be a new opcode instead of a
 > send flag.

That makes sense.  All existing hardware seems to use a separate opcode
in the HW WQE format, so it makes things cleaner to use a new opcode at
the verbs API level too.

I'll update my patches.

Thanks,
  Roland