Commit ba8b7ccb authored by Oscar Mateo's avatar Oscar Mateo Committed by Daniel Vetter

drm/i915/bdw: Workload submission mechanism for Execlists

This is what i915_gem_do_execbuffer calls when it wants to execute some
worload in an Execlists world.

v2: Check arguments before doing stuff in intel_execlists_submission. Also,
get rel_constants parsing right.
Signed-off-by: default avatarOscar Mateo <oscar.mateo@intel.com>
Reviewed-by: default avatarDamien Lespiau <damien.lespiau@intel.com>
[danvet: Drop the chipset flush, that's pre-gen6. And appease
checkpatch a bit .... again!]
Signed-off-by: default avatarDaniel Vetter <daniel.vetter@ffwll.ch>
parent 15648585
...@@ -2266,6 +2266,12 @@ int i915_gem_set_domain_ioctl(struct drm_device *dev, void *data, ...@@ -2266,6 +2266,12 @@ int i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
struct drm_file *file_priv); struct drm_file *file_priv);
int i915_gem_sw_finish_ioctl(struct drm_device *dev, void *data, int i915_gem_sw_finish_ioctl(struct drm_device *dev, void *data,
struct drm_file *file_priv); struct drm_file *file_priv);
void i915_gem_execbuffer_move_to_active(struct list_head *vmas,
struct intel_engine_cs *ring);
void i915_gem_execbuffer_retire_commands(struct drm_device *dev,
struct drm_file *file,
struct intel_engine_cs *ring,
struct drm_i915_gem_object *obj);
int i915_gem_ringbuffer_submission(struct drm_device *dev, int i915_gem_ringbuffer_submission(struct drm_device *dev,
struct drm_file *file, struct drm_file *file,
struct intel_engine_cs *ring, struct intel_engine_cs *ring,
......
...@@ -942,7 +942,7 @@ i915_gem_validate_context(struct drm_device *dev, struct drm_file *file, ...@@ -942,7 +942,7 @@ i915_gem_validate_context(struct drm_device *dev, struct drm_file *file,
return ctx; return ctx;
} }
static void void
i915_gem_execbuffer_move_to_active(struct list_head *vmas, i915_gem_execbuffer_move_to_active(struct list_head *vmas,
struct intel_engine_cs *ring) struct intel_engine_cs *ring)
{ {
...@@ -983,7 +983,7 @@ i915_gem_execbuffer_move_to_active(struct list_head *vmas, ...@@ -983,7 +983,7 @@ i915_gem_execbuffer_move_to_active(struct list_head *vmas,
} }
} }
static void void
i915_gem_execbuffer_retire_commands(struct drm_device *dev, i915_gem_execbuffer_retire_commands(struct drm_device *dev,
struct drm_file *file, struct drm_file *file,
struct intel_engine_cs *ring, struct intel_engine_cs *ring,
......
...@@ -91,6 +91,55 @@ int intel_sanitize_enable_execlists(struct drm_device *dev, int enable_execlists ...@@ -91,6 +91,55 @@ int intel_sanitize_enable_execlists(struct drm_device *dev, int enable_execlists
return 0; return 0;
} }
static int logical_ring_invalidate_all_caches(struct intel_ringbuffer *ringbuf)
{
struct intel_engine_cs *ring = ringbuf->ring;
uint32_t flush_domains;
int ret;
flush_domains = 0;
if (ring->gpu_caches_dirty)
flush_domains = I915_GEM_GPU_DOMAINS;
ret = ring->emit_flush(ringbuf, I915_GEM_GPU_DOMAINS, flush_domains);
if (ret)
return ret;
ring->gpu_caches_dirty = false;
return 0;
}
static int execlists_move_to_gpu(struct intel_ringbuffer *ringbuf,
struct list_head *vmas)
{
struct intel_engine_cs *ring = ringbuf->ring;
struct i915_vma *vma;
uint32_t flush_domains = 0;
bool flush_chipset = false;
int ret;
list_for_each_entry(vma, vmas, exec_list) {
struct drm_i915_gem_object *obj = vma->obj;
ret = i915_gem_object_sync(obj, ring);
if (ret)
return ret;
if (obj->base.write_domain & I915_GEM_DOMAIN_CPU)
flush_chipset |= i915_gem_clflush_object(obj, false);
flush_domains |= obj->base.write_domain;
}
if (flush_domains & I915_GEM_DOMAIN_GTT)
wmb();
/* Unconditionally invalidate gpu caches and ensure that we do flush
* any residual writes from the previous batch.
*/
return logical_ring_invalidate_all_caches(ringbuf);
}
int intel_execlists_submission(struct drm_device *dev, struct drm_file *file, int intel_execlists_submission(struct drm_device *dev, struct drm_file *file,
struct intel_engine_cs *ring, struct intel_engine_cs *ring,
struct intel_context *ctx, struct intel_context *ctx,
...@@ -99,7 +148,84 @@ int intel_execlists_submission(struct drm_device *dev, struct drm_file *file, ...@@ -99,7 +148,84 @@ int intel_execlists_submission(struct drm_device *dev, struct drm_file *file,
struct drm_i915_gem_object *batch_obj, struct drm_i915_gem_object *batch_obj,
u64 exec_start, u32 flags) u64 exec_start, u32 flags)
{ {
/* TODO */ struct drm_i915_private *dev_priv = dev->dev_private;
struct intel_ringbuffer *ringbuf = ctx->engine[ring->id].ringbuf;
int instp_mode;
u32 instp_mask;
int ret;
instp_mode = args->flags & I915_EXEC_CONSTANTS_MASK;
instp_mask = I915_EXEC_CONSTANTS_MASK;
switch (instp_mode) {
case I915_EXEC_CONSTANTS_REL_GENERAL:
case I915_EXEC_CONSTANTS_ABSOLUTE:
case I915_EXEC_CONSTANTS_REL_SURFACE:
if (instp_mode != 0 && ring != &dev_priv->ring[RCS]) {
DRM_DEBUG("non-0 rel constants mode on non-RCS\n");
return -EINVAL;
}
if (instp_mode != dev_priv->relative_constants_mode) {
if (instp_mode == I915_EXEC_CONSTANTS_REL_SURFACE) {
DRM_DEBUG("rel surface constants mode invalid on gen5+\n");
return -EINVAL;
}
/* The HW changed the meaning on this bit on gen6 */
instp_mask &= ~I915_EXEC_CONSTANTS_REL_SURFACE;
}
break;
default:
DRM_DEBUG("execbuf with unknown constants: %d\n", instp_mode);
return -EINVAL;
}
if (args->num_cliprects != 0) {
DRM_DEBUG("clip rectangles are only valid on pre-gen5\n");
return -EINVAL;
} else {
if (args->DR4 == 0xffffffff) {
DRM_DEBUG("UXA submitting garbage DR4, fixing up\n");
args->DR4 = 0;
}
if (args->DR1 || args->DR4 || args->cliprects_ptr) {
DRM_DEBUG("0 cliprects but dirt in cliprects fields\n");
return -EINVAL;
}
}
if (args->flags & I915_EXEC_GEN7_SOL_RESET) {
DRM_DEBUG("sol reset is gen7 only\n");
return -EINVAL;
}
ret = execlists_move_to_gpu(ringbuf, vmas);
if (ret)
return ret;
if (ring == &dev_priv->ring[RCS] &&
instp_mode != dev_priv->relative_constants_mode) {
ret = intel_logical_ring_begin(ringbuf, 4);
if (ret)
return ret;
intel_logical_ring_emit(ringbuf, MI_NOOP);
intel_logical_ring_emit(ringbuf, MI_LOAD_REGISTER_IMM(1));
intel_logical_ring_emit(ringbuf, INSTPM);
intel_logical_ring_emit(ringbuf, instp_mask << 16 | instp_mode);
intel_logical_ring_advance(ringbuf);
dev_priv->relative_constants_mode = instp_mode;
}
ret = ring->emit_bb_start(ringbuf, exec_start, flags);
if (ret)
return ret;
i915_gem_execbuffer_move_to_active(vmas, ring);
i915_gem_execbuffer_retire_commands(dev, file, ring, batch_obj);
return 0; return 0;
} }
...@@ -363,8 +489,6 @@ static int gen8_init_render_ring(struct intel_engine_cs *ring) ...@@ -363,8 +489,6 @@ static int gen8_init_render_ring(struct intel_engine_cs *ring)
static int gen8_emit_bb_start(struct intel_ringbuffer *ringbuf, static int gen8_emit_bb_start(struct intel_ringbuffer *ringbuf,
u64 offset, unsigned flags) u64 offset, unsigned flags)
{ {
struct intel_engine_cs *ring = ringbuf->ring;
struct drm_i915_private *dev_priv = ring->dev->dev_private;
bool ppgtt = !(flags & I915_DISPATCH_SECURE); bool ppgtt = !(flags & I915_DISPATCH_SECURE);
int ret; int ret;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment