From mboxrd@z Thu Jan  1 00:00:00 1970
From: keith.busch@intel.com (Keith Busch)
Date: Wed, 20 Feb 2019 09:39:22 -0700
Subject: Read-only Mapping of Program Text using Large THP Pages
In-Reply-To: <20190220144345.GG12668@bombadil.infradead.org>
References: <379F21DD-006F-4E33-9BD5-F81F9BA75C10@oracle.com>
 <20190220134454.GF12668@bombadil.infradead.org>
 <07B3B085-C844-4A13-96B1-3DB0F1AF26F5@oracle.com>
 <20190220144345.GG12668@bombadil.infradead.org>
Message-ID: <20190220163921.GA4451@localhost.localdomain>

On Wed, Feb 20, 2019@06:43:46AM -0800, Matthew Wilcox wrote:
> What NVMe doesn't have is a way for the host to tell the controller
> "Here's a 2MB sized I/O; bytes 40960 to 45056 are most important to
> me; please give me a completion event once those bytes are valid and
> then another completion event once the entire I/O is finished".
> 
> I have no idea if hardware designers would be interested in adding that
> kind of complexity, but this is why we also have I/O people at the same
> meeting, so we can get these kinds of whole-stack discussions going.

We have two unused PRP bits, so I guess there's room to define something
like a "me first" flag. I am skeptical we'd get committee approval for
that or partial completion events, though.

I think the host should just split the more important part of the transfer
into a separate command. The only hardware support we have to prioritize
that command ahead of others is with weighted priority queues, but we're
missing driver support for that at the moment.