On Fri, Jan 31, 2020 at 11:51:43AM -0700, Andrzej Jakowski wrote:

On 1/30/20 11:34 PM, Christoph Hellwig wrote:

quoted

Can you explain this check?  This looks weird to me  I think we need
a generalized check if a make_request based driver supports REQ_NOWAIT
instead (and as a separate patch / patchset).

Original check used to reject polled IO for stackable block devices as "not
supported". To solve that situation I introduced additional check to reject
all non REQ_HIPRI requests. That check is not intended to generalize, like
you indicated, but to conservatively select which requests to accept.
Perhaps there is better way to do that. Any suggestions?

REQ_NOWAIT and REQ_HIPRI are completley unrelated concepts and they need
to be dealt with entirely separately.