Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 19 Sep 2018 16:26:22 -0600
From:   Keith Busch <keith.busch@intel.com>
To:     Dave Hansen <dave.hansen@intel.com>
Cc:     linux-mm@kvack.org, linux-kernel@vger.kernel.org,
        Kirill Shutemov <kirill.shutemov@linux.intel.com>,
        Dan Williams <dan.j.williams@intel.com>
Subject: Re: [PATCH 0/7] mm: faster get user pages
Message-ID: <20180919222621.GA29003@localhost.localdomain>
References: <20180919210250.28858-1-keith.busch@intel.com>
 <40b392d0-0642-2d9b-5325-664a328ff677@intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <40b392d0-0642-2d9b-5325-664a328ff677@intel.com>
User-Agent: Mutt/1.9.1 (2017-09-22)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, Sep 19, 2018 at 02:15:28PM -0700, Dave Hansen wrote:
> On 09/19/2018 02:02 PM, Keith Busch wrote:
> > Pinning user pages out of nvdimm dax memory is significantly slower
> > compared to system ram. Analysis points to software overhead incurred
> > from a radix tree lookup. This patch series fixes that by removing the
> > relatively costly dev_pagemap lookup that was repeated for each page,
> > significantly increasing gup time.
> 
> Could you also remind us why DAX pages are such special snowflakes and
> *require* radix tree lookups in the first place?

Yeah, ZONE_DEVICE memory is special. It has struct page mappings, but
not for online general use. The dev_pagemap is the metadata to the zone
device memory, and that metadata is stashed in a radix tree.

We're looking up the dev_pagemap in this path to take a reference so
the zone device can't be unmapped.