From mboxrd@z Thu Jan  1 00:00:00 1970
From: =?ISO-8859-1?Q?Christian_K=F6nig?= <deathsimple@vodafone.de>
Subject: Re: [drm-next] drm/radeon: use IBs for VM page table updates
Date: Fri, 01 Feb 2013 17:35:08 +0100
Message-ID: <510BEEBC.9070901@vodafone.de>
References: <1359668271-3697-1-git-send-email-alexdeucher@gmail.com>
	<510B980F.8080002@vodafone.de>
	<CADnq5_PjUXB-iLSx8Q9XTZ+uddViXg4oQX0bc17BJvvERR1NDw@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"; Format="flowed"
Content-Transfer-Encoding: quoted-printable
Return-path: <dri-devel-bounces+sf-dri-devel=m.gmane.org@lists.freedesktop.org>
Received: from outgoing.email.vodafone.de (outgoing.email.vodafone.de
	[139.7.28.128])
	by gabe.freedesktop.org (Postfix) with ESMTP id 9C437E6C03
	for <dri-devel@lists.freedesktop.org>;
	Fri,  1 Feb 2013 08:35:14 -0800 (PST)
In-Reply-To: <CADnq5_PjUXB-iLSx8Q9XTZ+uddViXg4oQX0bc17BJvvERR1NDw@mail.gmail.com>
List-Unsubscribe: <http://lists.freedesktop.org/mailman/options/dri-devel>,
	<mailto:dri-devel-request@lists.freedesktop.org?subject=unsubscribe>
List-Archive: <http://lists.freedesktop.org/archives/dri-devel>
List-Post: <mailto:dri-devel@lists.freedesktop.org>
List-Help: <mailto:dri-devel-request@lists.freedesktop.org?subject=help>
List-Subscribe: <http://lists.freedesktop.org/mailman/listinfo/dri-devel>,
	<mailto:dri-devel-request@lists.freedesktop.org?subject=subscribe>
Sender: dri-devel-bounces+sf-dri-devel=m.gmane.org@lists.freedesktop.org
Errors-To: dri-devel-bounces+sf-dri-devel=m.gmane.org@lists.freedesktop.org
To: Alex Deucher <alexdeucher@gmail.com>
Cc: Alex Deucher <alexander.deucher@amd.com>, dri-devel@lists.freedesktop.org
List-Id: dri-devel@lists.freedesktop.org

Am 01.02.2013 15:04, schrieb Alex Deucher:
> On Fri, Feb 1, 2013 at 5:25 AM, Christian K=F6nig <deathsimple@vodafone.d=
e> wrote:
>> [SNAP]
>> Wouldn't it make sense to use the IB functions (ib_get, ib_schedule,
>> ib_free) here instead of filling the IB structure manually?
> I tried that initially, but ran into several problems and it ended up
> being easier to just allocate and schedule the IB directly.  The IB
> functions have self contained code to deal with semaphores and fences
> that I couldn't figure out how to make work cleanly with the vm
> semaphore and fence.  radeon_ib_schedule() takes the ring lock so if I
> needed to do any direct ring manipuation prior, I'd need to drop the
> lock before calling ib_schedule().  Let me know if you have any
> suggestions.

Just send out an updated version of the patch, which seems to look quite =

a bit cleaner and still works fine (at least on my SI board).

Christian.

>
> Alex
>
>
>> Apart from that it looks quite good.
>>
>> Christian.
>>