All of lore.kernel.org
 help / color / mirror / Atom feed
From: Chris Wilson <chris@chris-wilson.co.uk>
To: Dave Airlie <airlied@gmail.com>, dri-devel@lists.freedesktop.org
Subject: Re: [PATCH 3/3] drm/gem: start adding support for per-file object mmaps
Date: Wed, 19 Dec 2012 09:47:59 +0000	[thread overview]
Message-ID: <6c3329$7nh8o3@orsmga002.jf.intel.com> (raw)
In-Reply-To: <1355892119-13926-4-git-send-email-airlied@gmail.com>

From: Chris Wilson <chris@chris-wilson.co.uk>
Subject: Re: [proof-of-concept/rfc] per object file mmaps
To: Dave Airlie <airlied@gmail.com>, dri-devel@lists.freedesktop.org
In-Reply-To: <1355892119-13926-1-git-send-email-airlied@gmail.com>
References: <1355892119-13926-1-git-send-email-airlied@gmail.com>

On Wed, 19 Dec 2012 14:41:56 +1000, Dave Airlie <airlied@gmail.com> wrote:
> The 2+3 patches are actually the code, the first was just a cleanup.
> 
> Anyways the second patch describes it best, but I've taken the approach
> that we just need to keep track of what filp this object has had a handle
> created on, and block mmaps on it. We could also probably block a bit
> more explicitly in the driver respective mmap offset retrieval ioctls,
> however we need to block in mmap itself to stop people just picking misc
> address and trying them.
> 
> It doesn't really add much now, but with render nodes where we have flink_to
> or fd passing it would be more useful.
> 
> I've no idea if I'll get to do much more with this, so consider it an
> indication of how I'd like it done for someone else to run with :-)

Locking looks entirely non-existent for per-file mmap add/remove in the
generic gem code. Also we have a pair of hooks there that look like they
would be a convenient point to place that code.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre

      reply	other threads:[~2012-12-19  9:48 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-12-19  4:41 [proof-of-concept/rfc] per object file mmaps Dave Airlie
2012-12-19  4:41 ` [PATCH 1/3] drm/vma/gem/ttm: consolide unmapping range Dave Airlie
2012-12-19  4:41 ` [PATCH 2/3] drm/ttm: add object per-file mmap validation Dave Airlie
2012-12-19  4:41 ` [PATCH 3/3] drm/gem: start adding support for per-file object mmaps Dave Airlie
2012-12-19  9:47   ` Chris Wilson [this message]

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='6c3329$7nh8o3@orsmga002.jf.intel.com' \
    --to=chris@chris-wilson.co.uk \
    --cc=airlied@gmail.com \
    --cc=dri-devel@lists.freedesktop.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.