From mboxrd@z Thu Jan  1 00:00:00 1970
From: Chris Wilson <chris@chris-wilson.co.uk>
Subject: Re: [PATCH 1/2] split render engine batch buffer and
	BLT engine
Date: Tue, 26 Oct 2010 10:13:24 +0100
Message-ID: <b9dded$gt4at7@orsmga002.jf.intel.com>
References: <1288078396-6002-1-git-send-email-nanhai.zou@intel.com>
	<b9dded$gt3srv@orsmga002.jf.intel.com>
	<41EFD7A46E18724CAB128DAD0073348018E24C713F@shsmsx502.ccr.corp.intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Return-path: <intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org>
Received: from mga09.intel.com (mga09.intel.com [134.134.136.24])
	by gabe.freedesktop.org (Postfix) with ESMTP id D5DD49E748
	for <intel-gfx@lists.freedesktop.org>;
	Tue, 26 Oct 2010 02:13:27 -0700 (PDT)
In-Reply-To: <41EFD7A46E18724CAB128DAD0073348018E24C713F@shsmsx502.ccr.corp.intel.com>
List-Unsubscribe: <http://lists.freedesktop.org/mailman/options/intel-gfx>,
	<mailto:intel-gfx-request@lists.freedesktop.org?subject=unsubscribe>
List-Archive: <http://lists.freedesktop.org/archives/intel-gfx>
List-Post: <mailto:intel-gfx@lists.freedesktop.org>
List-Help: <mailto:intel-gfx-request@lists.freedesktop.org?subject=help>
List-Subscribe: <http://lists.freedesktop.org/mailman/listinfo/intel-gfx>,
	<mailto:intel-gfx-request@lists.freedesktop.org?subject=subscribe>
Sender: intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org
Errors-To: intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org
To: Zou Nanhai <nanhai.zou@intel.com>, "intel-gfx@lists.freedesktop.org" <intel-gfx@lists.freedesktop.org>
List-Id: intel-gfx@lists.freedesktop.org

On Tue, 26 Oct 2010 16:23:24 +0800, "Zou, Nanhai" <nanhai.zou@intel.com> wrote:
> 	I do not quite understand your point.
> How can we mix blitter command with render command in a batch buffer?

We can't. We also can't mix render targets/sources between concurrent
batch buffers either, at the moment. (I'd much rather finish the kernel
synchronisation before teaching X how to synchronise a pair of
interdependent concurrent batch buffers).  So in order to get any
synchronisation between the rings you have to flush the BLT batch when
starting a RENDER operation (if that operation accesses a BLT buffer and
vice versa).  Hence why I think that a single, modal batch buffer will be
no less efficient than maintaining a pair (or more) concurrent batch
buffers. X's request model means that we will process many sequential ops
from a single client in a single pass which will intermix BLT/RENDER
commands (or just use RENDER predominantly). I do not foresee X attaining
any great level of parallelism between the rings.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre