diff mbox

[RFC] drm/i915: Remove FIXME_lrc_ctx backpointer

Message ID 1413366514-12115-1-git-send-email-nicholas.hoath@intel.com (mailing list archive)
State New, archived
Headers show

Commit Message

Nick Hoath Oct. 15, 2014, 9:48 a.m. UTC 
The first pass implementation of execlists required a backpointer to the context to be held
in the intel_ringbuffer. However the context pointer is available higher in the call stack.
Remove the backpointer from the ring buffer structure and instead pass it down through the
call stack.
Signed-off-by: Nick Hoath <nicholas.hoath@intel.com>
CC: Chris Harris <chris.harris@intel.com>
---
 drivers/gpu/drm/i915/i915_gem.c         |  7 ++--
 drivers/gpu/drm/i915/intel_lrc.c        | 65 ++++++++++++++++++++-------------
 drivers/gpu/drm/i915/intel_lrc.h        | 10 +++--
 drivers/gpu/drm/i915/intel_ringbuffer.h | 14 +++----
 4 files changed, 56 insertions(+), 40 deletions(-)

Comments

Chris Wilson Oct. 15, 2014, 10:02 a.m. UTC | #1 
On Wed, Oct 15, 2014 at 10:48:34AM +0100, Nick Hoath wrote:
> The first pass implementation of execlists required a backpointer to the context to be held
> in the intel_ringbuffer. However the context pointer is available higher in the call stack.
> Remove the backpointer from the ring buffer structure and instead pass it down through the
> call stack.

If you look at my fixes for execlist, you don't even need this. The only
usecase I have for ring->ctx is in debugfs, where you don't have the
information in the stack anyway.
-Chris
diff mbox

Patch

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index e9c783d..724bbdf 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -2322,6 +2322,7 @@  int __i915_add_request(struct intel_engine_cs *ring,
 	struct drm_i915_gem_request *request;
 	struct intel_ringbuffer *ringbuf;
 	u32 request_ring_position, request_start;
+	struct intel_context *ctx = NULL;
 	int ret;
 
 	request = ring->preallocated_lazy_request;
@@ -2329,7 +2330,7 @@  int __i915_add_request(struct intel_engine_cs *ring,
 		return -ENOMEM;
 
 	if (i915.enable_execlists) {
-		struct intel_context *ctx = request->ctx;
+		ctx = request->ctx;
 		ringbuf = ctx->engine[ring->id].ringbuf;
 	} else
 		ringbuf = ring->buffer;
@@ -2343,7 +2344,7 @@  int __i915_add_request(struct intel_engine_cs *ring,
 	 * what.
 	 */
 	if (i915.enable_execlists) {
-		ret = logical_ring_flush_all_caches(ringbuf);
+		ret = logical_ring_flush_all_caches(ringbuf, ctx);
 		if (ret)
 			return ret;
 	} else {
@@ -2360,7 +2361,7 @@  int __i915_add_request(struct intel_engine_cs *ring,
 	request_ring_position = intel_ring_get_tail(ringbuf);
 
 	if (i915.enable_execlists) {
-		ret = ring->emit_request(ringbuf);
+		ret = ring->emit_request(ringbuf, ctx);
 		if (ret)
 			return ret;
 	} else {
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 803fc38..1be836a 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -578,7 +578,8 @@  static int execlists_context_queue(struct intel_engine_cs *ring,
 	return 0;
 }
 
-static int logical_ring_invalidate_all_caches(struct intel_ringbuffer *ringbuf)
+static int logical_ring_invalidate_all_caches(struct intel_ringbuffer *ringbuf,
+					      struct intel_context *ctx)
 {
 	struct intel_engine_cs *ring = ringbuf->ring;
 	uint32_t flush_domains;
@@ -588,7 +589,8 @@  static int logical_ring_invalidate_all_caches(struct intel_ringbuffer *ringbuf)
 	if (ring->gpu_caches_dirty)
 		flush_domains = I915_GEM_GPU_DOMAINS;
 
-	ret = ring->emit_flush(ringbuf, I915_GEM_GPU_DOMAINS, flush_domains);
+	ret = ring->emit_flush(ringbuf, ctx,
+			       I915_GEM_GPU_DOMAINS, flush_domains);
 	if (ret)
 		return ret;
 
@@ -597,6 +599,7 @@  static int logical_ring_invalidate_all_caches(struct intel_ringbuffer *ringbuf)
 }
 
 static int execlists_move_to_gpu(struct intel_ringbuffer *ringbuf,
+				 struct intel_context *ctx,
 				 struct list_head *vmas)
 {
 	struct intel_engine_cs *ring = ringbuf->ring;
@@ -624,7 +627,7 @@  static int execlists_move_to_gpu(struct intel_ringbuffer *ringbuf,
 	/* Unconditionally invalidate gpu caches and ensure that we do flush
 	 * any residual writes from the previous batch.
 	 */
-	return logical_ring_invalidate_all_caches(ringbuf);
+	return logical_ring_invalidate_all_caches(ringbuf, ctx);
 }
 
 /**
@@ -704,13 +707,13 @@  int intel_execlists_submission(struct drm_device *dev, struct drm_file *file,
 		return -EINVAL;
 	}
 
-	ret = execlists_move_to_gpu(ringbuf, vmas);
+	ret = execlists_move_to_gpu(ringbuf, ctx, 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);
+		ret = intel_logical_ring_begin(ringbuf, ctx, 4);
 		if (ret)
 			return ret;
 
@@ -723,7 +726,7 @@  int intel_execlists_submission(struct drm_device *dev, struct drm_file *file,
 		dev_priv->relative_constants_mode = instp_mode;
 	}
 
-	ret = ring->emit_bb_start(ringbuf, exec_start, flags);
+	ret = ring->emit_bb_start(ringbuf, ctx, exec_start, flags);
 	if (ret)
 		return ret;
 
@@ -755,7 +758,8 @@  void intel_logical_ring_stop(struct intel_engine_cs *ring)
 	I915_WRITE_MODE(ring, _MASKED_BIT_DISABLE(STOP_RING));
 }
 
-int logical_ring_flush_all_caches(struct intel_ringbuffer *ringbuf)
+int logical_ring_flush_all_caches(struct intel_ringbuffer *ringbuf,
+				  struct intel_context *ctx)
 {
 	struct intel_engine_cs *ring = ringbuf->ring;
 	int ret;
@@ -763,7 +767,7 @@  int logical_ring_flush_all_caches(struct intel_ringbuffer *ringbuf)
 	if (!ring->gpu_caches_dirty)
 		return 0;
 
-	ret = ring->emit_flush(ringbuf, 0, I915_GEM_GPU_DOMAINS);
+	ret = ring->emit_flush(ringbuf, ctx, 0, I915_GEM_GPU_DOMAINS);
 	if (ret)
 		return ret;
 
@@ -780,10 +784,10 @@  int logical_ring_flush_all_caches(struct intel_ringbuffer *ringbuf)
  * on a queue waiting for the ELSP to be ready to accept a new context submission. At that
  * point, the tail *inside* the context is updated and the ELSP written to.
  */
-void intel_logical_ring_advance_and_submit(struct intel_ringbuffer *ringbuf)
+void intel_logical_ring_advance_and_submit(struct intel_ringbuffer *ringbuf,
+					   struct intel_context *ctx)
 {
 	struct intel_engine_cs *ring = ringbuf->ring;
-	struct intel_context *ctx = ringbuf->FIXME_lrc_ctx;
 
 	intel_logical_ring_advance(ringbuf);
 
@@ -860,6 +864,7 @@  static int logical_ring_wait_request(struct intel_ringbuffer *ringbuf,
 }
 
 static int logical_ring_wait_for_space(struct intel_ringbuffer *ringbuf,
+				       struct intel_context *ctx,
 				       int bytes)
 {
 	struct intel_engine_cs *ring = ringbuf->ring;
@@ -873,7 +878,7 @@  static int logical_ring_wait_for_space(struct intel_ringbuffer *ringbuf,
 		return ret;
 
 	/* Force the context submission in case we have been skipping it */
-	intel_logical_ring_advance_and_submit(ringbuf);
+	intel_logical_ring_advance_and_submit(ringbuf, ctx);
 
 	/* With GEM the hangcheck timer should kick us out of the loop,
 	 * leaving it early runs the risk of corrupting GEM state (due
@@ -911,13 +916,14 @@  static int logical_ring_wait_for_space(struct intel_ringbuffer *ringbuf,
 	return ret;
 }
 
-static int logical_ring_wrap_buffer(struct intel_ringbuffer *ringbuf)
+static int logical_ring_wrap_buffer(struct intel_ringbuffer *ringbuf,
+				    struct intel_context *ctx)
 {
 	uint32_t __iomem *virt;
 	int rem = ringbuf->size - ringbuf->tail;
 
 	if (ringbuf->space < rem) {
-		int ret = logical_ring_wait_for_space(ringbuf, rem);
+		int ret = logical_ring_wait_for_space(ringbuf, ctx, rem);
 
 		if (ret)
 			return ret;
@@ -934,18 +940,19 @@  static int logical_ring_wrap_buffer(struct intel_ringbuffer *ringbuf)
 	return 0;
 }
 
-static int logical_ring_prepare(struct intel_ringbuffer *ringbuf, int bytes)
+static int logical_ring_prepare(struct intel_ringbuffer *ringbuf,
+				struct intel_context *ctx, int bytes)
 {
 	int ret;
 
 	if (unlikely(ringbuf->tail + bytes > ringbuf->effective_size)) {
-		ret = logical_ring_wrap_buffer(ringbuf);
+		ret = logical_ring_wrap_buffer(ringbuf, ctx);
 		if (unlikely(ret))
 			return ret;
 	}
 
 	if (unlikely(ringbuf->space < bytes)) {
-		ret = logical_ring_wait_for_space(ringbuf, bytes);
+		ret = logical_ring_wait_for_space(ringbuf, ctx, bytes);
 		if (unlikely(ret))
 			return ret;
 	}
@@ -966,7 +973,9 @@  static int logical_ring_prepare(struct intel_ringbuffer *ringbuf, int bytes)
  *
  * Return: non-zero if the ringbuffer is not ready to be written to.
  */
-int intel_logical_ring_begin(struct intel_ringbuffer *ringbuf, int num_dwords)
+int intel_logical_ring_begin(struct intel_ringbuffer *ringbuf,
+			     struct intel_context *ctx,
+			     int num_dwords)
 {
 	struct intel_engine_cs *ring = ringbuf->ring;
 	struct drm_device *dev = ring->dev;
@@ -978,12 +987,12 @@  int intel_logical_ring_begin(struct intel_ringbuffer *ringbuf, int num_dwords)
 	if (ret)
 		return ret;
 
-	ret = logical_ring_prepare(ringbuf, num_dwords * sizeof(uint32_t));
+	ret = logical_ring_prepare(ringbuf, ctx, num_dwords * sizeof(uint32_t));
 	if (ret)
 		return ret;
 
 	/* Preallocate the olr before touching the ring */
-	ret = logical_ring_alloc_seqno(ring, ringbuf->FIXME_lrc_ctx);
+	ret = logical_ring_alloc_seqno(ring, ctx);
 	if (ret)
 		return ret;
 
@@ -1038,12 +1047,13 @@  static int gen8_init_render_ring(struct intel_engine_cs *ring)
 }
 
 static int gen8_emit_bb_start(struct intel_ringbuffer *ringbuf,
+			      struct intel_context *ctx,
 			      u64 offset, unsigned flags)
 {
 	bool ppgtt = !(flags & I915_DISPATCH_SECURE);
 	int ret;
 
-	ret = intel_logical_ring_begin(ringbuf, 4);
+	ret = intel_logical_ring_begin(ringbuf, ctx, 4);
 	if (ret)
 		return ret;
 
@@ -1091,6 +1101,7 @@  static void gen8_logical_ring_put_irq(struct intel_engine_cs *ring)
 }
 
 static int gen8_emit_flush(struct intel_ringbuffer *ringbuf,
+			   struct intel_context *ctx,
 			   u32 invalidate_domains,
 			   u32 unused)
 {
@@ -1100,7 +1111,7 @@  static int gen8_emit_flush(struct intel_ringbuffer *ringbuf,
 	uint32_t cmd;
 	int ret;
 
-	ret = intel_logical_ring_begin(ringbuf, 4);
+	ret = intel_logical_ring_begin(ringbuf, ctx, 4);
 	if (ret)
 		return ret;
 
@@ -1129,6 +1140,7 @@  static int gen8_emit_flush(struct intel_ringbuffer *ringbuf,
 }
 
 static int gen8_emit_flush_render(struct intel_ringbuffer *ringbuf,
+				  struct intel_context *ctx,
 				  u32 invalidate_domains,
 				  u32 flush_domains)
 {
@@ -1155,7 +1167,7 @@  static int gen8_emit_flush_render(struct intel_ringbuffer *ringbuf,
 		flags |= PIPE_CONTROL_GLOBAL_GTT_IVB;
 	}
 
-	ret = intel_logical_ring_begin(ringbuf, 6);
+	ret = intel_logical_ring_begin(ringbuf, ctx, 6);
 	if (ret)
 		return ret;
 
@@ -1180,13 +1192,14 @@  static void gen8_set_seqno(struct intel_engine_cs *ring, u32 seqno)
 	intel_write_status_page(ring, I915_GEM_HWS_INDEX, seqno);
 }
 
-static int gen8_emit_request(struct intel_ringbuffer *ringbuf)
+static int gen8_emit_request(struct intel_ringbuffer *ringbuf,
+			     struct intel_context *ctx)
 {
 	struct intel_engine_cs *ring = ringbuf->ring;
 	u32 cmd;
 	int ret;
 
-	ret = intel_logical_ring_begin(ringbuf, 6);
+	ret = intel_logical_ring_begin(ringbuf, ctx, 6);
 	if (ret)
 		return ret;
 
@@ -1201,7 +1214,7 @@  static int gen8_emit_request(struct intel_ringbuffer *ringbuf)
 	intel_logical_ring_emit(ringbuf, ring->outstanding_lazy_seqno);
 	intel_logical_ring_emit(ringbuf, MI_USER_INTERRUPT);
 	intel_logical_ring_emit(ringbuf, MI_NOOP);
-	intel_logical_ring_advance_and_submit(ringbuf);
+	intel_logical_ring_advance_and_submit(ringbuf, ctx);
 
 	return 0;
 }
@@ -1474,6 +1487,7 @@  int intel_lr_context_render_state_init(struct intel_engine_cs *ring,
 		return 0;
 
 	ret = ring->emit_bb_start(ringbuf,
+			ctx,
 			so.ggtt_offset,
 			I915_DISPATCH_SECURE);
 	if (ret)
@@ -1702,7 +1716,6 @@  int intel_lr_context_deferred_create(struct intel_context *ctx,
 	}
 
 	ringbuf->ring = ring;
-	ringbuf->FIXME_lrc_ctx = ctx;
 
 	ringbuf->size = 32 * PAGE_SIZE;
 	ringbuf->effective_size = ringbuf->size;
diff --git a/drivers/gpu/drm/i915/intel_lrc.h b/drivers/gpu/drm/i915/intel_lrc.h
index 33c3b4b..ee647db 100644
--- a/drivers/gpu/drm/i915/intel_lrc.h
+++ b/drivers/gpu/drm/i915/intel_lrc.h
@@ -36,8 +36,10 @@  void intel_logical_ring_stop(struct intel_engine_cs *ring);
 void intel_logical_ring_cleanup(struct intel_engine_cs *ring);
 int intel_logical_rings_init(struct drm_device *dev);
 
-int logical_ring_flush_all_caches(struct intel_ringbuffer *ringbuf);
-void intel_logical_ring_advance_and_submit(struct intel_ringbuffer *ringbuf);
+int logical_ring_flush_all_caches(struct intel_ringbuffer *ringbuf,
+					   struct intel_context *ctx);
+void intel_logical_ring_advance_and_submit(struct intel_ringbuffer *ringbuf,
+					   struct intel_context *ctx);
 /**
  * intel_logical_ring_advance() - advance the ringbuffer tail
  * @ringbuf: Ringbuffer to advance.
@@ -59,7 +61,9 @@  static inline void intel_logical_ring_emit(struct intel_ringbuffer *ringbuf,
 	iowrite32(data, ringbuf->virtual_start + ringbuf->tail);
 	ringbuf->tail += 4;
 }
-int intel_logical_ring_begin(struct intel_ringbuffer *ringbuf, int num_dwords);
+int intel_logical_ring_begin(struct intel_ringbuffer *ringbuf,
+			     struct intel_context *ctx,
+			     int num_dwords);
 
 /* Logical Ring Contexts */
 int intel_lr_context_render_state_init(struct intel_engine_cs *ring,
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 96479c8..5cd09d3 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -99,13 +99,6 @@  struct intel_ringbuffer {
 
 	struct intel_engine_cs *ring;
 
-	/*
-	 * FIXME: This backpointer is an artifact of the history of how the
-	 * execlist patches came into being. It will get removed once the basic
-	 * code has landed.
-	 */
-	struct intel_context *FIXME_lrc_ctx;
-
 	u32 head;
 	u32 tail;
 	int space;
@@ -123,6 +116,8 @@  struct intel_ringbuffer {
 	u32 last_retired_head;
 };
 
+struct intel_context;
+
 struct  intel_engine_cs {
 	const char	*name;
 	enum intel_ring_id {
@@ -237,11 +232,14 @@  struct  intel_engine_cs {
 	struct list_head execlist_queue;
 	u8 next_context_status_buffer;
 	u32             irq_keep_mask; /* bitmask for interrupts that should not be masked */
-	int		(*emit_request)(struct intel_ringbuffer *ringbuf);
+	int		(*emit_request)(struct intel_ringbuffer *ringbuf,
+					struct intel_context *ctx);
 	int		(*emit_flush)(struct intel_ringbuffer *ringbuf,
+				      struct intel_context *ctx,
 				      u32 invalidate_domains,
 				      u32 flush_domains);
 	int		(*emit_bb_start)(struct intel_ringbuffer *ringbuf,
+					 struct intel_context *ctx,
 					 u64 offset, unsigned flags);
 
 	/**