DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com 7731361D05
Date: Fri, 23 Jun 2017 10:40:43 -0600
From: Alex Williamson <alex.williamson@redhat.com>
To: Gerd Hoffmann <kraxel@redhat.com>
Cc: Zhi Wang <zhi.a.wang@intel.com>,
        "Wang, Zhenyu Z" <zhenyu.z.wang@intel.com>,
        "intel-gfx@lists.freedesktop.org" <intel-gfx@lists.freedesktop.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "Chen, Xiaoguang" <xiaoguang.chen@intel.com>,
        "Zhang, Tina" <tina.zhang@intel.com>,
        Kirti Wankhede <kwankhede@nvidia.com>,
        "Lv, Zhiyuan" <zhiyuan.lv@intel.com>,
        "intel-gvt-dev@lists.freedesktop.org" 
        <intel-gvt-dev@lists.freedesktop.org>
Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf
 operations
Message-ID: <20170623104043.7f3505af@w520.home>
In-Reply-To: <1498206688.24807.5.camel@redhat.com>
References: <1497513611-2814-1-git-send-email-xiaoguang.chen@intel.com>
        <1497513611-2814-6-git-send-email-xiaoguang.chen@intel.com>
        <f19da207-a7f5-d3d3-4577-b071e7d75cf7@nvidia.com>
        <1497542438.29252.1.camel@redhat.com>
        <20170615143833.7526351b@w520.home>
        <24c4880b-24f5-ea07-834c-c77d3e895c78@nvidia.com>
        <1497854312.4207.4.camel@redhat.com>
        <20170619085530.1f5e46dc@w520.home>
        <237F54289DF84E4997F34151298ABEBC7C56EBE0@SHSMSX101.ccr.corp.intel.com>
        <1497956256.16795.7.camel@redhat.com>
        <20170620090004.44ac7fbc@w520.home>
        <237F54289DF84E4997F34151298ABEBC7C56F3DC@SHSMSX101.ccr.corp.intel.com>
        <20170620172204.09405cf4@w520.home>
        <237F54289DF84E4997F34151298ABEBC7C56F843@SHSMSX101.ccr.corp.intel.com>
        <1498043011.5802.5.camel@redhat.com>
        <20170621125938.1a92abda@w520.home>
        <1498120215.25651.5.camel@redhat.com>
        <20170622125458.01c953ab@w520.home>
        <1498202819.24807.3.camel@redhat.com>
        <594CC7F6.7050507@intel.com>
        <1498206688.24807.5.camel@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8BIT
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1340
Lines: 31

On Fri, 23 Jun 2017 10:31:28 +0200
Gerd Hoffmann <kraxel@redhat.com> wrote:

> On Fri, 2017-06-23 at 15:49 +0800, Zhi Wang wrote:
> > Hi:
> >      Thanks for the discussions! If the userspace application has 
> > already maintained a LRU list, it looks like we don't need
> > generation 
> > anymore,  
> 
> generation isn't required, things are working just fine without that. 
> It is just a small optimization, userspace can skip the LRU lookup
> altogether if the generation didn't change.
> 
> But of couse that only pays off if the kernel doesn't has to put much
> effort into maintaining the generation id.  Something simple like
> increasing it each time the guest writes a register which affects
> plane_info.

But it seems like that simple management algorithm pretty much
guarantees that the kernel will never revisit a generation and
therefore caching dmabuf fds is pointless.  AIUI the optimization is to
allow userspace to 'at a glance' test one plane_info vs another.  The
user could also do this with a memcmp of the plane_info structs if
that's its only purpose.  A randomly incremented field within that
struct could actually be a hindrance to the user for such a
comparison.  Are there cases where the plane_info struct is otherwise
identical where the user would need to get a new dmabuf fd anyway?
Thanks,

Alex