diff options
author | Chris Wilson <chris@chris-wilson.co.uk> | 2010-11-10 16:40:20 +0000 |
---|---|---|
committer | Chris Wilson <chris@chris-wilson.co.uk> | 2010-12-02 10:07:05 +0000 |
commit | d9e86c0ee60f323e890484628f351bf50fa9a15d (patch) | |
tree | 4755a8ec484340bc12e2d1f88f35fbd266df6001 /drivers/gpu/drm/i915/i915_gem_tiling.c | |
parent | 87ca9c8a7ea9c8c7ce1561edaad1aa8570f1a01e (diff) | |
download | linux-d9e86c0ee60f323e890484628f351bf50fa9a15d.tar.gz linux-d9e86c0ee60f323e890484628f351bf50fa9a15d.tar.bz2 linux-d9e86c0ee60f323e890484628f351bf50fa9a15d.zip |
drm/i915: Pipelined fencing [infrastructure]
With this change, every batchbuffer can use all available fences (save
pinned and scanout, of course) without ever stalling the gpu!
In theory. Currently the actual pipelined update of the register is
disabled due to some stability issues. However, just the deferred update
is a significant win.
Based on a series of patches by Daniel Vetter.
The premise is that before every access to a buffer through the GTT we
have to declare whether we need a register or not. If the access is by
the GPU, a pipelined update to the register is made via the ringbuffer,
and we track the last seqno of the batches that access it. If by the
CPU we wait for the last GPU access and update the register (either
to clear or to set it for the current buffer).
One advantage of being able to pipeline changes is that we can defer the
actual updating of the fence register until we first need to access the
object through the GTT, i.e. we can eliminate the stall on set_tiling.
This is important as the userspace bo cache does not track the tiling
status of active buffers which generate frequent stalls on gen3 when
enabling tiling for an already bound buffer.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Diffstat (limited to 'drivers/gpu/drm/i915/i915_gem_tiling.c')
-rw-r--r-- | drivers/gpu/drm/i915/i915_gem_tiling.c | 23 |
1 files changed, 7 insertions, 16 deletions
diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c index 1c5fdb30f272..22a32b9932c5 100644 --- a/drivers/gpu/drm/i915/i915_gem_tiling.c +++ b/drivers/gpu/drm/i915/i915_gem_tiling.c @@ -244,9 +244,6 @@ i915_gem_object_fence_ok(struct drm_i915_gem_object *obj, int tiling_mode) if (INTEL_INFO(obj->base.dev)->gen >= 4) return true; - if (!obj->gtt_space) - return true; - if (INTEL_INFO(obj->base.dev)->gen == 3) { if (obj->gtt_offset & ~I915_FENCE_START_MASK) return false; @@ -345,27 +342,21 @@ i915_gem_set_tiling(struct drm_device *dev, void *data, * tiling mode. Otherwise we can just leave it alone, but * need to ensure that any fence register is cleared. */ - if (!i915_gem_object_fence_ok(obj, args->tiling_mode)) - ret = i915_gem_object_unbind(obj); - else if (obj->fence_reg != I915_FENCE_REG_NONE) - ret = i915_gem_object_put_fence_reg(obj, true); - else - i915_gem_release_mmap(obj); + i915_gem_release_mmap(obj); - if (ret != 0) { - args->tiling_mode = obj->tiling_mode; - args->stride = obj->stride; - goto err; - } + obj->map_and_fenceable = + obj->gtt_space == NULL || + (obj->gtt_offset + obj->base.size <= dev_priv->mm.gtt_mappable_end && + i915_gem_object_fence_ok(obj, args->tiling_mode)); + obj->tiling_changed = true; obj->tiling_mode = args->tiling_mode; obj->stride = args->stride; } -err: drm_gem_object_unreference(&obj->base); mutex_unlock(&dev->struct_mutex); - return ret; + return 0; } /** |