public inbox for intel-gfx@lists.freedesktop.org
 help / color / mirror / Atom feed
From: Daniel Vetter <daniel@ffwll.ch>
To: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 5/7] drm/i915: avoid emiting semaphore wait on GEN8 when seqno wrap happened.
Date: Tue, 5 Aug 2014 10:36:44 +0200	[thread overview]
Message-ID: <20140805083644.GE8727@phenom.ffwll.local> (raw)
In-Reply-To: <1407176119-5294-5-git-send-email-rodrigo.vivi@intel.com>

On Mon, Aug 04, 2014 at 11:15:17AM -0700, Rodrigo Vivi wrote:

This commit message is too thin.

> C: Ben Widawsky <ben@bwidawsk.net>
> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

Does this blow up with Mika's seqno wrap testcase? If so please add the
right Testcase: line, if not, how can we repro this?
-Daniel
> ---
>  drivers/gpu/drm/i915/intel_ringbuffer.c | 30 +++++++++++++++++++++---------
>  1 file changed, 21 insertions(+), 9 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
> index 2e566e0..cd30b39 100644
> --- a/drivers/gpu/drm/i915/intel_ringbuffer.c
> +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
> @@ -867,15 +867,26 @@ gen8_ring_sync(struct intel_engine_cs *waiter,
>  	if (ret)
>  		return ret;
>  
> -	intel_ring_emit(waiter, MI_SEMAPHORE_WAIT |
> -				MI_SEMAPHORE_GLOBAL_GTT |
> -				MI_SEMAPHORE_POLL |
> -				MI_SEMAPHORE_SAD_GTE_SDD);
> -	intel_ring_emit(waiter, seqno);
> -	intel_ring_emit(waiter,
> -			lower_32_bits(GEN8_WAIT_OFFSET(waiter, signaller->id)));
> -	intel_ring_emit(waiter,
> -			upper_32_bits(GEN8_WAIT_OFFSET(waiter, signaller->id)));
> +	/* If seqno wrap happened, omit the wait with no-ops */
> +	if (likely(!i915_gem_has_seqno_wrapped(waiter->dev, seqno))) {
> +		intel_ring_emit(waiter, MI_SEMAPHORE_WAIT |
> +					MI_SEMAPHORE_GLOBAL_GTT |
> +					MI_SEMAPHORE_POLL |
> +					MI_SEMAPHORE_SAD_GTE_SDD);
> +		intel_ring_emit(waiter, seqno);
> +		intel_ring_emit(waiter,
> +				lower_32_bits(GEN8_WAIT_OFFSET(waiter,
> +							       signaller->id)));
> +		intel_ring_emit(waiter,
> +				upper_32_bits(GEN8_WAIT_OFFSET(waiter,
> +							       signaller->id)));
> +	} else {
> +		intel_ring_emit(waiter, MI_NOOP);
> +		intel_ring_emit(waiter, MI_NOOP);
> +		intel_ring_emit(waiter, MI_NOOP);
> +		intel_ring_emit(waiter, MI_NOOP);
> +	}
> +
>  	intel_ring_advance(waiter);
>  	return 0;
>  }
> @@ -2572,6 +2583,7 @@ int intel_init_vebox_ring_buffer(struct drm_device *dev)
>  			ring->semaphore.mbox.signal[BCS] = GEN6_BVESYNC;
>  			ring->semaphore.mbox.signal[VECS] = GEN6_NOSYNC;
>  			ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
> +
>  		}
>  	}
>  	ring->init = init_ring_common;
> -- 
> 1.9.3
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch

  reply	other threads:[~2014-08-05  8:36 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-08-04 18:15 [PATCH 1/7] drm/i915/bdw: Always issue a force restore Rodrigo Vivi
2014-08-04 18:15 ` [PATCH 2/7] drm/i915: Add thread stall DOP clock gating workaround on Broadwell Rodrigo Vivi
2014-08-05 17:10   ` Ville Syrjälä
2014-08-04 18:15 ` [PATCH 3/7] drm/i915/bdw: MI_FLUSH_DW a qword instead of dword Rodrigo Vivi
2014-08-05  1:21   ` Ben Widawsky
2014-08-04 18:15 ` [PATCH 4/7] drm/i915/bdw: cs-stall before state cache invld w/a Rodrigo Vivi
2014-08-05 12:39   ` Ville Syrjälä
2014-08-04 18:15 ` [PATCH 5/7] drm/i915: avoid emiting semaphore wait on GEN8 when seqno wrap happened Rodrigo Vivi
2014-08-05  8:36   ` Daniel Vetter [this message]
2014-08-04 18:15 ` [PATCH 6/7] drm/i915: BDW Semaphore signal with Post Sync Rodrigo Vivi
2014-08-04 18:15 ` [PATCH 7/7] Revert "drm/i915: Enable semaphores on BDW" Rodrigo Vivi
2014-08-07 20:05   ` Rodrigo Vivi
2014-08-08  7:10     ` Daniel Vetter
2014-08-08  7:13   ` Daniel Vetter
2014-08-08  7:31     ` Chris Wilson
2014-08-05  1:20 ` [PATCH 1/7] drm/i915/bdw: Always issue a force restore Ben Widawsky
2014-08-05  8:41   ` 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=20140805083644.GE8727@phenom.ffwll.local \
    --to=daniel@ffwll.ch \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=rodrigo.vivi@intel.com \
    /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