Skip to contentSkip to conversation

drm/i915: Use atomic for dev_priv->mm.bsd_engine_dispatch_index

PATCH In review Joonas Lahtinen 3 participants 6 messages Sep 1, 2016

Subsystems (2)
lore
Share & tools
lore

Joonas Lahtinen

author maintainer
#0 lore raw

PATCH 1/1 drm/i915: Use atomic for dev_priv->mm.bsd_engine_dispatch_index

Use atomic type and operands for dev_priv->mm.bsd_engine_dispatch_index to avoid one struct_mutex locking scenario.

Cc Chris Wilson <[email protected]>
Cc Imre Deak <[email protected]>
Cc Zhao Yakui <[email protected]>
Cc Daniel Vetter <[email protected]>
---
 drivers/gpu/drm/i915/i915_drv.h            |  2 +-
 drivers/gpu/drm/i915/i915_gem.c            |  2 ++
 drivers/gpu/drm/i915/i915_gem_execbuffer.c | 10 +++-------
 3 files changed, 6 insertions(+), 8 deletions(-)
drivers/gpu/drm/i915/i915_drv.h +1 -1
index c413587..43d5298 100644
@@ -1338,7 +1338,7 @@ struct i915_gem_mm {
 	bool interruptible;
 
 	/* the indicator for dispatch video commands on two BSD rings */
-	unsigned int bsd_engine_dispatch_index;
+	atomic_t bsd_engine_dispatch_index;
 
 	/** Bit 6 swizzling required for X tiling */
 	uint32_t bit_6_swizzle_x;
drivers/gpu/drm/i915/i915_gem.c +2 -0
index d37b441..2401818 100644
@@ -4622,6 +4622,8 @@ i915_gem_load_init(struct drm_device *dev)
 
 	dev_priv->mm.interruptible = true;
 
+	atomic_set(&dev_priv->mm.bsd_engine_dispatch_index, 0);
+
 	spin_lock_init(&dev_priv->fb_tracking.lock);
 }
 
drivers/gpu/drm/i915/i915_gem_execbuffer.c +3 -7
index 57d096a..9432d4c 100644
@@ -1535,13 +1535,9 @@ gen8_dispatch_bsd_engine(struct drm_i915_private *dev_priv,
 	struct drm_i915_file_private *file_priv = file->driver_priv;
 
 	/* Check whether the file_priv has already selected one ring. */
-	if ((int)file_priv->bsd_engine < 0) {
-		/* If not, use the ping-pong mechanism to select one. */
-		mutex_lock(&dev_priv->drm.struct_mutex);
-		file_priv->bsd_engine = dev_priv->mm.bsd_engine_dispatch_index;
-		dev_priv->mm.bsd_engine_dispatch_index ^= 1;
-		mutex_unlock(&dev_priv->drm.struct_mutex);
-	}
+	if ((int)file_priv->bsd_engine < 0)
+		file_priv->bsd_engine = atomic_fetch_xor(1,
+			 &dev_priv->mm.bsd_engine_dispatch_index);
 
 	return file_priv->bsd_engine;
 }
Signature · 6 lines
2.5.5

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Reviewed-by on v1 1/1
#1 lore raw
Quoted 8 lines · Joonas Lahtinen

Use atomic type and operands for dev_priv->mm.bsd_engine_dispatch_index to avoid one struct_mutex locking scenario.

Cc Chris Wilson <[email protected]>
Cc Imre Deak <[email protected]>
Cc Zhao Yakui <[email protected]>
Cc Daniel Vetter <[email protected]>

Looks fine, it's one that I've simply been too lazy to worry about.

-Chris

Signature · 5 lines
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#2 lore raw

✗ Fi.CI.BAT: failure for drm/i915: Use atomic for dev_priv->mm.bsd_engine_dispatch_index

In reply to Joonas Lahtinen
== Series Details ==

== Series Details ==

Series: drm/i915: Use atomic for dev_priv->mm.bsd_engine_dispatch_index
URL : https://patchwork.freedesktop.org/series/11890/
State : failure

== Summary ==

Series 11890v1 drm/i915: Use atomic for dev_priv->mm.bsd_engine_dispatch_index
http://patchwork.freedesktop.org/api/1.0/series/11890/revisions/1/mbox/

Test prime_vgem:

        Subgroup basic-fence-flip:
                pass       -> FAIL       (fi-bdw-5557u)
fi-bdw-5557u     total:252  pass:235  dwarn:0   dfail:0   fail:2   skip:15 
fi-bsw-n3050     total:252  pass:205  dwarn:0   dfail:0   fail:1   skip:46 
fi-byt-n2820     total:252  pass:209  dwarn:0   dfail:0   fail:2   skip:41 
fi-hsw-4770r     total:252  pass:225  dwarn:0   dfail:0   fail:1   skip:26 
fi-ivb-3520m     total:252  pass:220  dwarn:0   dfail:0   fail:1   skip:31 
fi-skl-6260u     total:252  pass:237  dwarn:0   dfail:0   fail:1   skip:14 
fi-skl-6700k     total:252  pass:223  dwarn:0   dfail:0   fail:1   skip:28 
fi-snb-2520m     total:252  pass:207  dwarn:0   dfail:0   fail:2   skip:43 
fi-snb-2600      total:252  pass:207  dwarn:0   dfail:0   fail:2   skip:43 

Results at /archive/results/CI_IGT_test/Patchwork_2459/

cb8b5ca97c2d7ae7514c61e214c69fd23b0d5039 drm-intel-nightly: 2016y-08m-31d-11h-30m-30s UTC integration manifest 096c16e drm/i915: Use atomic for dev_priv->mm.bsd_engine_dispatch_index

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#3 lore raw

Re: ✗ Fi.CI.BAT: failure for drm/i915: Use atomic for dev_priv->mm.bsd_engine_dispatch_index

Quoted 6 lines · Patchwork

== Series Details ==

Series: drm/i915: Use atomic for dev_priv->mm.bsd_engine_dispatch_index
URL : https://patchwork.freedesktop.org/series/11890/
State : failure

Quoted 8 lines

== Summary ==

Series 11890v1 drm/i915: Use atomic for dev_priv->mm.bsd_engine_dispatch_index
http://patchwork.freedesktop.org/api/1.0/series/11890/revisions/1/mbox/

Test prime_vgem:

        Subgroup basic-fence-flip:
                pass       -> FAIL       (fi-bdw-5557u)

Hmm, this is probably some condition I forgot to drain before starting the test. What's the actual failure?
-Chris

Signature · 5 lines
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
author maintainer
#4 lore raw

Re: ✗ Fi.CI.BAT: failure for drm/i915: Use atomic for dev_priv->mm.bsd_engine_dispatch_index

Quoted 6 lines · Chris Wilson
Quoted 3 lines · Patchwork

Test prime_vgem:
        Subgroup basic-fence-flip:
                pass       -> FAIL       (fi-bdw-5557u)

Hmm, this is probably some condition I forgot to drain before starting the test. What's the actual failure?

(prime_vgem:29095) CRITICAL: Test assertion failure function flip_to_vgem, file prime_vgem.c:651:
(prime_vgem:29095) CRITICAL: Failed assertion: poll(&pfd, 1, 20000) == 1 (prime_vgem:29095) CRITICAL: error: 0 != 1
Subtest basic-fence-flip failed.
**** DEBUG ****
(prime_vgem:29089) ioctl-wrappers-DEBUG: Test requirement passed: has_modifiers (prime_vgem:29089) ioctl-wrappers-DEBUG: Test requirement passed: has_modifiers (prime_vgem:29089) DEBUG: Test requirement passed: (crtc_id = set_fb_on_crtc(i915, 0, &bo[0], fb_id[0])) ****  END  ****

Regards, Joonas

Signature · 7 lines
Joonas Lahtinen
Open Source Technology Center
Intel Corporation
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
author maintainer
#5 lore raw

Re: ✗ Fi.CI.BAT: failure for drm/i915: Use atomic for dev_priv->mm.bsd_engine_dispatch_index

Quoted 9 lines · Joonas Lahtinen
Quoted 8 lines · Chris Wilson
Quoted 4 lines · Patchwork

Test prime_vgem:
        Subgroup basic-fence-flip:
                pass       -> FAIL       (fi-bdw-5557u)

Hmm, this is probably some condition I forgot to drain before starting the test. What's the actual failure?

Quoted 10 lines

(prime_vgem:29095) CRITICAL: Test assertion failure function flip_to_vgem, file prime_vgem.c:651:
(prime_vgem:29095) CRITICAL: Failed assertion: poll(&pfd, 1, 20000) == 1 (prime_vgem:29095) CRITICAL: error: 0 != 1
Subtest basic-fence-flip failed.
**** DEBUG ****
(prime_vgem:29089) ioctl-wrappers-DEBUG: Test requirement passed: has_modifiers (prime_vgem:29089) ioctl-wrappers-DEBUG: Test requirement passed: has_modifiers (prime_vgem:29089) DEBUG: Test requirement passed: (crtc_id = set_fb_on_crtc(i915, 0, &bo[0], fb_id[0])) ****  END  ****

Anyway, pushed the patch, thanks for review.

Regards, Joonas

Quoted 1 line

Regards, Joonas

Signature · 7 lines
Joonas Lahtinen
Open Source Technology Center
Intel Corporation
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx