drm/amdgpu:add switch buffer to end of CS (v2)

sync switch buffer scheme with windows kmd for gfx v8,
step1: append a switch_buffer to the end of CS

v2:rebase on latest staging
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drivers/gpu/drm/amd/amdgpu/amdgpu.h

@@ -319,6 +319,7 @@ struct amdgpu_ring_funcs {
 	/* note usage for clock and power gating */
 	void (*begin_use)(struct amdgpu_ring *ring);
 	void (*end_use)(struct amdgpu_ring *ring);
+	void (*emit_switch_buffer) (struct amdgpu_ring *ring);
 };
 
 /*
@@ -2254,6 +2255,7 @@ amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
 #define amdgpu_ring_emit_gds_switch(r, v, db, ds, wb, ws, ab, as) (r)->funcs->emit_gds_switch((r), (v), (db), (ds), (wb), (ws), (ab), (as))
 #define amdgpu_ring_emit_hdp_flush(r) (r)->funcs->emit_hdp_flush((r))
 #define amdgpu_ring_emit_hdp_invalidate(r) (r)->funcs->emit_hdp_invalidate((r))
+#define amdgpu_ring_emit_switch_buffer(r) (r)->funcs->emit_switch_buffer((r))
 #define amdgpu_ring_pad_ib(r, ib) ((r)->funcs->pad_ib((r), (ib)))
 #define amdgpu_ring_init_cond_exec(r) (r)->funcs->init_cond_exec((r))
 #define amdgpu_ring_patch_cond_exec(r,o) (r)->funcs->patch_cond_exec((r),(o))

drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c

@@ -210,6 +210,8 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 		amdgpu_ring_patch_cond_exec(ring, patch_offset);
 
 	ring->current_ctx = ctx;
+	if (ring->funcs->emit_switch_buffer)
+		amdgpu_ring_emit_switch_buffer(ring);
 	amdgpu_ring_commit(ring);
 	return 0;
 }

drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c

@@ -6087,6 +6087,12 @@ static void gfx_v8_0_ring_emit_fence_compute(struct amdgpu_ring *ring,
 	amdgpu_ring_write(ring, upper_32_bits(seq));
 }
 
+static void gfx_v8_ring_emit_sb(struct amdgpu_ring *ring)
+{
+	amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0));
+	amdgpu_ring_write(ring, 0);
+}
+
 static void gfx_v8_0_set_gfx_eop_interrupt_state(struct amdgpu_device *adev,
 						 enum amdgpu_interrupt_state state)
 {
@@ -6268,6 +6274,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = {
 	.test_ib = gfx_v8_0_ring_test_ib,
 	.insert_nop = amdgpu_ring_insert_nop,
 	.pad_ib = amdgpu_ring_generic_pad_ib,
+	.emit_switch_buffer = gfx_v8_ring_emit_sb,
 };
 
 static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = {