[Intel-gfx] [PATCH 3/3] drm/i915/icl: Introduce gen11 flush/invalidate

[Mika Kuoppala]

There are more pipe control levers to pull with icl.
Take them into use both for flushing and invalidating to
ensure completeness.

Doing so, avoid overloading the gen8 flush/invalidate
further and make a gen11 specific callback.

Cc: Chris Wilson <chris at chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
Signed-off-by: Mika Kuoppala <mika.kuoppala at linux.intel.com>
---
 drivers/gpu/drm/i915/intel_gpu_commands.h |  2 +
 drivers/gpu/drm/i915/intel_lrc.c          | 61 ++++++++++++++++++++++-
 2 files changed, 62 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/intel_gpu_commands.h b/drivers/gpu/drm/i915/intel_gpu_commands.h
index 105e2a9e874a..62112e9da610 100644
--- a/drivers/gpu/drm/i915/intel_gpu_commands.h
+++ b/drivers/gpu/drm/i915/intel_gpu_commands.h
@@ -195,6 +195,8 @@
 #define   DISPLAY_PLANE_A           (0<<20)
 #define   DISPLAY_PLANE_B           (1<<20)
 #define GFX_OP_PIPE_CONTROL(len)	((0x3<<29)|(0x3<<27)|(0x2<<24)|((len)-2))
+#define   PIPE_CONTROL_COMMAND_CACHE_INVALIDATE		(1<<29) /* gen11+ */
+#define   PIPE_CONTROL_TILE_CACHE_FLUSH			(1<<28) /* gen11+ */
 #define   PIPE_CONTROL_FLUSH_L3				(1<<27)
 #define   PIPE_CONTROL_GLOBAL_GTT_IVB			(1<<24) /* gen7+ */
 #define   PIPE_CONTROL_MMIO_WRITE			(1<<23)
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 039e2a54b1fc..d4c7dab9cdb4 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -2021,6 +2021,60 @@ static int gen8_emit_flush_render(struct i915_request *request,
 	return 0;
 }
 
+static int gen11_emit_flush_render(struct i915_request *request,
+				   u32 mode)
+{
+	struct intel_engine_cs *engine = request->engine;
+	const u32 scratch_addr =
+		i915_scratch_offset(engine->i915) + 2 * CACHELINE_BYTES;
+	u32 *cs;
+
+	if (mode & EMIT_FLUSH) {
+		u32 flags = PIPE_CONTROL_CS_STALL;
+
+		flags |= PIPE_CONTROL_TILE_CACHE_FLUSH;
+		flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH;
+		flags |= PIPE_CONTROL_STALL_AT_SCOREBOARD;
+		flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH;
+		flags |= PIPE_CONTROL_DC_FLUSH_ENABLE;
+		flags |= PIPE_CONTROL_FLUSH_ENABLE;
+		flags |= PIPE_CONTROL_QW_WRITE;
+		flags |= PIPE_CONTROL_GLOBAL_GTT_IVB;
+
+		cs = intel_ring_begin(request, 6);
+		if (IS_ERR(cs))
+			return PTR_ERR(cs);
+
+		cs = gen8_emit_pipe_control(cs, flags, scratch_addr);
+
+		intel_ring_advance(request, cs);
+	}
+
+	if (mode & EMIT_INVALIDATE) {
+		u32 flags = PIPE_CONTROL_CS_STALL;
+
+		flags |= PIPE_CONTROL_COMMAND_CACHE_INVALIDATE;
+		flags |= PIPE_CONTROL_TLB_INVALIDATE;
+		flags |= PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE;
+		flags |= PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE;
+		flags |= PIPE_CONTROL_VF_CACHE_INVALIDATE;
+		flags |= PIPE_CONTROL_CONST_CACHE_INVALIDATE;
+		flags |= PIPE_CONTROL_STATE_CACHE_INVALIDATE;
+		flags |= PIPE_CONTROL_QW_WRITE;
+		flags |= PIPE_CONTROL_GLOBAL_GTT_IVB;
+
+		cs = intel_ring_begin(request, 6);
+		if (IS_ERR(cs))
+			return PTR_ERR(cs);
+
+		cs = gen8_emit_pipe_control(cs, flags, scratch_addr);
+
+		intel_ring_advance(request, cs);
+	}
+
+	return 0;
+}
+
 /*
  * Reserve space for 2 NOOPs at the end of each request to be
  * used as a workaround for not being allowed to do lite
@@ -2268,7 +2322,12 @@ int logical_render_ring_init(struct intel_engine_cs *engine)
 
 	/* Override some for render ring. */
 	engine->init_context = gen8_init_rcs_context;
-	engine->emit_flush = gen8_emit_flush_render;
+
+	if (INTEL_GEN(dev_priv) >= 11)
+		engine->emit_flush = gen11_emit_flush_render;
+	else
+		engine->emit_flush = gen8_emit_flush_render;
+
 	engine->emit_breadcrumb = gen8_emit_breadcrumb_rcs;
 	engine->emit_breadcrumb_sz = gen8_emit_breadcrumb_rcs_sz;
 
-- 
2.17.1