From: Daniel Vetter <daniel@ffwll.ch>
To: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ben Widawsky <ben@bwidawsk.net>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 3/3] drm/i915: Use PIPE_CONTROL for flushing on gen6+.
Date: Thu, 6 Oct 2011 01:54:47 +0200 [thread overview]
Message-ID: <20111005235447.GB2858@phenom.ffwll.local> (raw)
In-Reply-To: <e39f63$27lah8@fmsmga002.fm.intel.com>
On Thu, Oct 06, 2011 at 12:36:03AM +0100, Chris Wilson wrote:
> On Wed, 5 Oct 2011 15:57:13 -0700, Ben Widawsky <ben@bwidawsk.net> wrote:
> > I think we also want a TLB invalidate here, bit 18. This requires another
> > workaround before issuing this flush: We need 2 Store Data Commands (such as
> > MI_STORE_DATA_IMM or MI_STORE_DATA_INDEX) before sending PIPE_CONTROL w/ stall
> > (20) and TLB inv bit (18) set
>
> Isn't that workaround itself rather hand-wavy? As in it gives the
> hardware sufficient time to complete outstanding writes, but not
> necessarily. Or am I thinking of yet another workaround...
MI_STORE seems to be the hw teams favourite for adding these delays to
paper over hw issues. I've seen it all over bspec ...
-Daniel
--
Daniel Vetter
Mail: daniel@ffwll.ch
Mobile: +41 (0)79 365 57 48
next prev parent reply other threads:[~2011-10-05 23:54 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-10-04 6:02 [PATCH 1/3] i915: Remove implied length of 2 from GFX_OP_PIPE_CONTROL #define Kenneth Graunke
2011-10-04 6:02 ` [PATCH 2/3] i915: Rename PIPE_CONTROL bit defines to be less terse Kenneth Graunke
2011-10-04 8:30 ` Daniel Vetter
2011-10-05 5:13 ` Kenneth Graunke
2011-10-05 10:05 ` Daniel Vetter
2011-10-05 22:29 ` Ben Widawsky
2011-10-04 6:02 ` [PATCH 3/3] drm/i915: Use PIPE_CONTROL for flushing on gen6+ Kenneth Graunke
2011-10-05 22:57 ` Ben Widawsky
2011-10-05 23:36 ` Chris Wilson
2011-10-05 23:54 ` Daniel Vetter [this message]
2011-10-06 0:59 ` Eric Anholt
2011-10-06 5:15 ` Ben Widawsky
2011-10-06 18:00 ` Eric Anholt
2011-10-06 19:01 ` Ben Widawsky
2011-10-05 23:39 ` Ben Widawsky
2011-10-05 20:35 ` [PATCH 1/3] i915: Remove implied length of 2 from GFX_OP_PIPE_CONTROL #define Ben Widawsky
2011-10-11 11:09 ` Daniel Vetter
2011-10-11 17:20 ` Jesse Barnes
2011-10-11 18:39 ` Ben Widawsky
2011-10-11 18:53 ` Jesse Barnes
2011-10-11 22:17 ` Keith Packard
-- strict thread matches above, loose matches on Subject: below --
2011-10-11 21:41 [PATCH 1/3] drm/i915: " Daniel Vetter
2011-10-11 21:41 ` [PATCH 3/3] drm/i915: Use PIPE_CONTROL for flushing on gen6+ Daniel Vetter
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=20111005235447.GB2858@phenom.ffwll.local \
--to=daniel@ffwll.ch \
--cc=ben@bwidawsk.net \
--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