From: Yu Dai <yu.dai@intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/i915/guc: Defer LRC context unpin or release
Date: Fri, 6 Nov 2015 15:42:36 -0800 [thread overview]
Message-ID: <563D3AEC.1000203@intel.com> (raw)
In-Reply-To: <20151106220929.GD18324@nuc-i3427.alporthouse.com>
On 11/06/2015 02:09 PM, Chris Wilson wrote:
> On Fri, Nov 06, 2015 at 01:56:59PM -0800, yu.dai@intel.com wrote:
> > From: Alex Dai <yu.dai@intel.com>
> >
> > Can't immediately free LRC context (neither unpin it) even all
> > its referenced requests are completed, because HW still need a
> > short period of time to save data to LRC status page. It is safe
> > to free LRC when HW completes a request from a different LRC.
>
> See the legacy context switch mechanism for code to reuse - at least
> reuse the pointers rather than add yet another almost identically named
> one to intel_engine_cs.
> -Chris
>
Sorry that I accidentally submitted wrong version of this patch. The
correct one is here: https://patchwork.freedesktop.org/patch/64094/.
I use 'retired_ctx' to avoid confusing with legacy last_context in
intel_engine_cs. The 'retired' means all gem_request ref on it have
retired. However, the unpin or free of its backing BO is deferred until
HW completes another batch from a different LRC. The ref/unref concept
is similar to the legacy context switch. But I don't believe there is
any code we can reuse here.
Thanks,
Alex
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2015-11-06 23:44 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-11-06 21:56 [PATCH] drm/i915/guc: Defer LRC context unpin or release yu.dai
2015-11-06 22:03 ` Yu Dai
2015-11-06 22:09 ` Chris Wilson
2015-11-06 23:42 ` Yu Dai [this message]
2015-11-06 22:12 ` kbuild test robot
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=563D3AEC.1000203@intel.com \
--to=yu.dai@intel.com \
--cc=chris@chris-wilson.co.uk \
--cc=intel-gfx@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox