[Intel-gfx] [PATCH 158/190] drm/i915: Skip holding an object reference for execbuf preparation

Chris Wilson chris at chris-wilson.co.uk
Mon Jan 11 03:00:59 PST 2016


This is a golden oldie! We can shave a couple of locked instructions for
about 10% of the per-object overhead by not taking an extra kref whilst
reserving objects for an execbuf. Due to lock management this is safe,
as we cannot lose the original object reference without the lock.
Equally, because this relies on the heavy BKL^W struct_mutex, it is also
likely to be only a temporary optimisation until we have fine grained
locking. (That's what we said 5 years ago, so there's probably another
10 years before we get around to finer grained locking!)

Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
---
 drivers/gpu/drm/i915/i915_gem_execbuffer.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 6ccce848f3e2..b7424f1b1293 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -158,7 +158,6 @@ eb_lookup_vmas(struct eb_vmas *eb,
 			goto err;
 		}
 
-		drm_gem_object_reference(&obj->base);
 		list_add_tail(&obj->obj_exec_link, &objects);
 	}
 	spin_unlock(&file->table_lock);
@@ -272,7 +271,6 @@ static void eb_destroy(struct eb_vmas *eb)
 				       exec_list);
 		list_del_init(&vma->exec_list);
 		i915_gem_execbuffer_unreserve_vma(vma);
-		drm_gem_object_unreference(&vma->obj->base);
 	}
 	kfree(eb);
 }
@@ -947,7 +945,6 @@ i915_gem_execbuffer_relocate_slow(struct drm_device *dev,
 		vma = list_first_entry(&eb->vmas, struct i915_vma, exec_list);
 		list_del_init(&vma->exec_list);
 		i915_gem_execbuffer_unreserve_vma(vma);
-		drm_gem_object_unreference(&vma->obj->base);
 	}
 
 	mutex_unlock(&dev->struct_mutex);
@@ -1325,7 +1322,6 @@ i915_gem_execbuffer_parse(struct intel_engine_cs *ring,
 
 	vma->exec_entry = shadow_exec_entry;
 	vma->exec_entry->flags = __EXEC_OBJECT_HAS_PIN;
-	drm_gem_object_reference(&shadow_batch_obj->base);
 	list_add_tail(&vma->exec_list, &eb->vmas);
 
 err:
-- 
2.7.0.rc3



More information about the Intel-gfx mailing list