From: "Terje Bergström" <tbergstrom@nvidia.com>
To: Dave Airlie <airlied@gmail.com>
Cc: "thierry.reding@avionic-design.de"
<thierry.reding@avionic-design.de>,
"linux-tegra@vger.kernel.org" <linux-tegra@vger.kernel.org>,
"dri-devel@lists.freedesktop.org"
<dri-devel@lists.freedesktop.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
Arto Merilainen <amerilainen@nvidia.com>
Subject: Re: [RFC v2 8/8] drm: tegra: Add gr2d device
Date: Tue, 27 Nov 2012 10:16:33 +0200 [thread overview]
Message-ID: <50B476E1.4070403@nvidia.com> (raw)
In-Reply-To: <CAPM=9txCuPJcFAfD7Hu5o2BVFK=pVah7B8HhG0ctLCyFPwNEnA@mail.gmail.com>
On 27.11.2012 09:33, Dave Airlie wrote:
>> Third would be having a firewall in 2D driver checking the stream and
>> ensuring all registers that accept addresses are written by values
>> derived from dmabufs. I haven't tried implementing this, but it'd
>> involve a lookup table in kernel and CPU reading through the command
>> stream. Offsets and sizes would also need to be validated. There would
>> be a performance hit.
>
> This is the standard mechanism, and what exynos does as well.
>
> The per process VM method is also used as an extension to this on some hw.
Hi,
Thanks for the pointer, I looked at exynos code. It indeed checks the
registers written to, but it doesn't prevent overrun by checking sizes
of buffers and compare against requests.
Based on my experience with Tegra graphics stack, going through command
streams in kernel is bad for performance. For 2D operations this is
probably ok as the command streams are pretty simple. Anything more
complex is going to cause severe degradation of performance, but it's
also outside the scope of this patch set.
If this is the way to go, I'll put the firewall behind a Kconfig flag so
that system integrator can decide if his system needs it.
Terje
next prev parent reply other threads:[~2012-11-27 8:12 UTC|newest]
Thread overview: 145+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-26 13:19 [RFC v2 0/8] Support for Tegra 2D hardware Terje Bergstrom
2012-11-26 13:19 ` [RFC v2 1/8] video: tegra: Add nvhost driver Terje Bergstrom
2012-11-27 10:52 ` Sivaram Nair
2012-11-28 21:23 ` Thierry Reding
2012-11-29 10:21 ` Terje Bergström
2012-11-29 11:47 ` Thierry Reding
2012-11-29 18:38 ` Stephen Warren
2012-11-30 6:52 ` Thierry Reding
2012-11-30 8:50 ` Lucas Stach
2012-12-01 11:44 ` Terje Bergström
2012-12-01 15:10 ` Thierry Reding
2012-12-01 16:55 ` Terje Bergström
2012-12-01 17:34 ` Lucas Stach
2012-12-01 19:29 ` Terje Bergström
2012-12-01 21:42 ` Dave Airlie
2012-12-01 22:39 ` Thierry Reding
2012-12-02 11:24 ` Terje Bergström
2012-12-02 20:55 ` Thierry Reding
2012-12-03 6:26 ` Terje Bergström
2012-11-30 8:56 ` Terje Bergström
2012-11-30 10:38 ` Thierry Reding
2012-12-01 11:31 ` Terje Bergström
2012-12-01 13:42 ` Daniel Vetter
2012-12-01 16:22 ` Terje Bergström
2012-12-01 14:58 ` Thierry Reding
2012-12-01 17:13 ` Terje Bergström
2012-12-03 19:23 ` Stephen Warren
2012-12-04 21:31 ` Thierry Reding
2012-12-03 19:20 ` Stephen Warren
2012-12-03 21:03 ` Thierry Reding
2012-12-04 2:08 ` Mark Zhang
2012-12-04 2:11 ` Mark Zhang
2012-12-04 6:17 ` Terje Bergström
2012-11-29 18:34 ` Stephen Warren
2012-11-30 6:54 ` Terje Bergström
2012-11-30 6:53 ` Thierry Reding
2012-11-29 9:10 ` [RFC,v2,1/8] " Mark Zhang
2012-12-10 10:28 ` Terje Bergström
2012-11-26 13:19 ` [RFC v2 2/8] video: tegra: Add syncpoint wait and interrupts Terje Bergstrom
2012-11-27 11:02 ` Sivaram Nair
2012-11-29 8:44 ` Thierry Reding
2012-11-29 10:39 ` Terje Bergström
2012-11-30 7:22 ` Thierry Reding
2012-11-30 7:41 ` Terje Bergström
2012-11-29 18:41 ` Stephen Warren
2012-11-30 7:23 ` Thierry Reding
2012-11-26 13:19 ` [RFC v2 3/8] video: tegra: host: Add channel and client support Terje Bergstrom
2012-11-29 10:01 ` [RFC,v2,3/8] " Mark Zhang
2012-11-29 10:46 ` Terje Bergström
2012-11-30 6:13 ` Mark Zhang
2012-11-29 10:04 ` [RFC v2 3/8] " Thierry Reding
2012-11-29 11:00 ` Terje Bergström
2012-11-30 7:46 ` Thierry Reding
2012-11-26 13:19 ` [RFC v2 4/8] video: tegra: Add debug support Terje Bergstrom
2012-11-26 13:19 ` [RFC v2 5/8] ARM: tegra: Add auxiliary data for nvhost Terje Bergstrom
2012-11-26 23:39 ` Stephen Warren
2012-11-27 6:33 ` Terje Bergström
2012-11-27 17:17 ` Stephen Warren
2012-11-26 13:19 ` [RFC v2 6/8] gpu: drm: tegra: Remove redundant host1x Terje Bergstrom
2012-12-05 8:33 ` Thierry Reding
2012-12-05 10:10 ` Terje Bergström
2012-12-05 11:13 ` Thierry Reding
2012-12-05 11:47 ` Terje Bergström
2012-12-05 12:02 ` Lucas Stach
2012-12-05 12:03 ` Daniel Vetter
2012-12-05 12:08 ` Daniel Vetter
2012-12-05 12:22 ` Thierry Reding
2012-12-05 12:31 ` Daniel Vetter
2012-12-05 13:28 ` Thierry Reding
2012-12-05 16:34 ` Daniel Vetter
2012-12-05 20:44 ` Thierry Reding
2012-12-05 12:04 ` Thierry Reding
2012-12-05 15:43 ` Terje Bergström
2012-12-10 11:42 ` Terje Bergström
2012-12-12 16:08 ` Thierry Reding
2012-12-12 16:56 ` Terje Bergström
2012-12-13 8:48 ` Terje Bergström
2012-12-13 8:57 ` Thierry Reding
2012-12-13 17:58 ` Stephen Warren
2012-12-13 20:32 ` Thierry Reding
2012-12-14 6:09 ` Terje Bergström
2012-12-14 16:21 ` Stephen Warren
2012-12-14 19:59 ` Terje Bergström
2012-12-16 12:16 ` Thierry Reding
2012-12-16 16:37 ` Terje Bergström
2012-12-17 20:55 ` Stephen Warren
2012-12-18 6:37 ` Terje Bergström
2012-12-20 9:17 ` Terje Bergström
2012-12-20 17:14 ` Stephen Warren
2012-12-20 17:46 ` Terje Bergström
2012-12-20 17:55 ` Stephen Warren
2012-12-20 18:01 ` Terje Bergström
2012-12-20 20:30 ` Thierry Reding
2012-12-20 21:34 ` Terje Bergström
2012-12-20 21:50 ` Thierry Reding
2012-12-20 22:29 ` Stephen Warren
2012-12-20 22:28 ` Stephen Warren
2012-12-21 6:31 ` Terje Bergström
2012-12-21 8:57 ` Arto Merilainen
2012-12-21 21:19 ` Stephen Warren
2013-01-02 5:41 ` Terje Bergström
2013-01-04 10:09 ` Terje Bergström
2013-01-04 20:25 ` Stephen Warren
2013-01-07 8:20 ` Terje Bergström
2013-01-07 17:07 ` Stephen Warren
2013-01-15 11:30 ` Thierry Reding
2013-01-15 11:41 ` Terje Bergström
2012-11-26 13:19 ` [RFC v2 7/8] gpu: drm: tegra: Prime support Terje Bergstrom
2012-11-26 13:19 ` [RFC v2 8/8] drm: tegra: Add gr2d device Terje Bergstrom
2012-11-26 21:59 ` Rob Clark
2012-11-26 22:15 ` Dave Airlie
2012-11-27 6:52 ` Terje Bergström
2012-11-27 7:33 ` Dave Airlie
2012-11-27 8:16 ` Terje Bergström [this message]
2012-11-27 8:32 ` Dave Airlie
2012-11-27 8:45 ` Terje Bergström
2012-11-27 10:22 ` Lucas Stach
2012-11-27 10:37 ` Thierry Reding
2012-11-27 11:31 ` Terje Bergström
2012-11-27 11:47 ` Lucas Stach
2012-11-27 12:59 ` Terje Bergström
2012-11-27 23:00 ` Dave Airlie
2012-11-28 13:17 ` Terje Bergström
2012-11-28 13:33 ` Lucas Stach
2012-11-28 13:57 ` Terje Bergström
2012-11-28 14:06 ` Lucas Stach
2012-11-28 14:45 ` Terje Bergström
2012-11-28 15:13 ` Lucas Stach
2012-11-28 16:23 ` Terje Bergström
2012-11-28 18:46 ` Lucas Stach
2012-11-29 8:17 ` Terje Bergström
2012-11-29 9:09 ` Lucas Stach
2012-11-29 12:14 ` Thierry Reding
2012-11-30 7:44 ` Terje Bergström
2012-11-30 7:53 ` Lucas Stach
2012-11-29 13:36 ` Terje Bergström
2012-11-28 16:24 ` Stephen Warren
2012-11-28 20:53 ` Thomas Hellstrom
2012-12-03 9:30 ` Mark Zhang
2012-12-03 9:40 ` Daniel Vetter
2012-12-04 1:49 ` Mark Zhang
2012-11-29 7:37 ` [RFC,v2,8/8] " Mark Zhang
2012-12-01 14:45 ` [RFC v2 0/8] Support for Tegra 2D hardware Thierry Reding
2012-12-01 17:08 ` Terje Bergström
2012-12-01 19:29 ` Thierry Reding
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=50B476E1.4070403@nvidia.com \
--to=tbergstrom@nvidia.com \
--cc=airlied@gmail.com \
--cc=amerilainen@nvidia.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tegra@vger.kernel.org \
--cc=thierry.reding@avionic-design.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).