radv: avoid context rolls when binding graphics pipelines

It's common in some applications to bind a new graphics pipeline without
ending up changing any context registers.

This has a pipline have two command buffers: one for setting context
registers and one for everything else. The context register command buffer
is only emitted if it differs from the previous pipeline's.

v2: ensure late scissor emission is done when radv_emit_rbplus_state() is
    called
v2: make use of cmd_buffer->state.workaround_scissor_bug
v3: rename "workaround_scissor_bug" to
    "context_roll_without_scissor_emitted"

Signed-off-by: Rhys Perry <[email protected]>
Reviewed-by: Bas Nieuwenhuizen <[email protected]>

3 files changed, 141 insertions, 108 deletions

diff --git a/src/amd/vulkan/radv_cmd_buffer.c b/src/amd/vulkan/radv_cmd_buffer.c
index f6c93284db3..5f1fefee869 100644
--- a/src/amd/vulkan/radv_cmd_buffer.c
+++ b/src/amd/vulkan/radv_cmd_buffer.c
@@ -661,6 +661,8 @@ radv_update_multisample_state(struct radv_cmd_buffer *cmd_buffer,
 		radeon_emit(cmd_buffer->cs, PKT3(PKT3_EVENT_WRITE, 0, 0));
 		radeon_emit(cmd_buffer->cs, EVENT_TYPE(V_028A90_FLUSH_DFSM) | EVENT_INDEX(0));
 	}
+
+	cmd_buffer->state.context_roll_without_scissor_emitted = true;
 }
 
 static void
@@ -857,10 +859,13 @@ radv_emit_rbplus_state(struct radv_cmd_buffer *cmd_buffer)
 		sx_blend_opt_control |= S_02875C_MRT0_COLOR_OPT_DISABLE(1) << (i * 4);
 		sx_blend_opt_control |= S_02875C_MRT0_ALPHA_OPT_DISABLE(1) << (i * 4);
 	}
+	/* TODO: avoid redundantly setting context registers */
 	radeon_set_context_reg_seq(cmd_buffer->cs, R_028754_SX_PS_DOWNCONVERT, 3);
 	radeon_emit(cmd_buffer->cs, sx_ps_downconvert);
 	radeon_emit(cmd_buffer->cs, sx_blend_opt_epsilon);
 	radeon_emit(cmd_buffer->cs, sx_blend_opt_control);
+
+	cmd_buffer->state.context_roll_without_scissor_emitted = true;
 }
 
 static void
@@ -884,6 +889,15 @@ radv_emit_graphics_pipeline(struct radv_cmd_buffer *cmd_buffer)
 
 	radeon_emit_array(cmd_buffer->cs, pipeline->cs.buf, pipeline->cs.cdw);
 
+	if (!cmd_buffer->state.emitted_pipeline ||
+	    cmd_buffer->state.emitted_pipeline->ctx_cs.cdw != pipeline->ctx_cs.cdw ||
+	    cmd_buffer->state.emitted_pipeline->ctx_cs_hash != pipeline->ctx_cs_hash ||
+	    memcmp(cmd_buffer->state.emitted_pipeline->ctx_cs.buf,
+	           pipeline->ctx_cs.buf, pipeline->ctx_cs.cdw * 4)) {
+		radeon_emit_array(cmd_buffer->cs, pipeline->ctx_cs.buf, pipeline->ctx_cs.cdw);
+		cmd_buffer->state.context_roll_without_scissor_emitted = true;
+	}
+
 	for (unsigned i = 0; i < MESA_SHADER_COMPUTE; i++) {
 		if (!pipeline->shaders[i])
 			continue;
@@ -2939,6 +2953,8 @@ radv_emit_compute_pipeline(struct radv_cmd_buffer *cmd_buffer)
 	if (!pipeline || pipeline == cmd_buffer->state.emitted_compute_pipeline)
 		return;
 
+	assert(!pipeline->ctx_cs.cdw);
+
 	cmd_buffer->state.emitted_compute_pipeline = pipeline;
 
 	radeon_check_space(cmd_buffer->device->ws, cmd_buffer->cs, pipeline->cs.cdw);
@@ -3630,20 +3646,16 @@ static bool radv_need_late_scissor_emission(struct radv_cmd_buffer *cmd_buffer,
 	uint32_t used_states = cmd_buffer->state.pipeline->graphics.needed_dynamic_state | ~RADV_CMD_DIRTY_DYNAMIC_ALL;
 
 	/* Index, vertex and streamout buffers don't change context regs, and
-	 * pipeline is handled later.
+	 * pipeline is already handled.
 	 */
 	used_states &= ~(RADV_CMD_DIRTY_INDEX_BUFFER |
 			 RADV_CMD_DIRTY_VERTEX_BUFFER |
 			 RADV_CMD_DIRTY_STREAMOUT_BUFFER |
 			 RADV_CMD_DIRTY_PIPELINE);
 
-	/* Assume all state changes except  these two can imply context rolls. */
 	if (cmd_buffer->state.dirty & used_states)
 		return true;
 
-	if (cmd_buffer->state.emitted_pipeline != cmd_buffer->state.pipeline)
-		return true;
-
 	if (info->indexed && state->pipeline->graphics.prim_restart_enable &&
 	    (state->index_type ? 0xffffffffu : 0xffffu) != state->last_primitive_reset_index)
 		return true;
@@ -3655,7 +3667,7 @@ static void
 radv_emit_all_graphics_states(struct radv_cmd_buffer *cmd_buffer,
 			      const struct radv_draw_info *info)
 {
-	bool late_scissor_emission = radv_need_late_scissor_emission(cmd_buffer, info);
+	bool late_scissor_emission;
 
 	if ((cmd_buffer->state.dirty & RADV_CMD_DIRTY_FRAMEBUFFER) ||
 	    cmd_buffer->state.emitted_pipeline != cmd_buffer->state.pipeline)
@@ -3664,6 +3676,12 @@ radv_emit_all_graphics_states(struct radv_cmd_buffer *cmd_buffer,
 	if (cmd_buffer->state.dirty & RADV_CMD_DIRTY_PIPELINE)
 		radv_emit_graphics_pipeline(cmd_buffer);
 
+	/* This should be before the cmd_buffer->state.dirty is cleared
+	 * (excluding RADV_CMD_DIRTY_PIPELINE) and after
+	 * cmd_buffer->state.context_roll_without_scissor_emitted is set. */
+	late_scissor_emission =
+		radv_need_late_scissor_emission(cmd_buffer, info);
+
 	if (cmd_buffer->state.dirty & RADV_CMD_DIRTY_FRAMEBUFFER)
 		radv_emit_framebuffer_state(cmd_buffer);
 
diff --git a/src/amd/vulkan/radv_pipeline.c b/src/amd/vulkan/radv_pipeline.c
index 3bf3c7ffef7..138e153f9a4 100644
--- a/src/amd/vulkan/radv_pipeline.c
+++ b/src/amd/vulkan/radv_pipeline.c
@@ -2526,7 +2526,7 @@ radv_compute_bin_size(struct radv_pipeline *pipeline, const VkGraphicsPipelineCr
 }
 
 static void
-radv_pipeline_generate_binning_state(struct radeon_cmdbuf *cs,
+radv_pipeline_generate_binning_state(struct radeon_cmdbuf *ctx_cs,
 				     struct radv_pipeline *pipeline,
 				     const VkGraphicsPipelineCreateInfo *pCreateInfo)
 {
@@ -2576,15 +2576,15 @@ radv_pipeline_generate_binning_state(struct radeon_cmdbuf *cs,
 	                S_028C44_OPTIMAL_BIN_SELECTION(1);
 	}
 
-	radeon_set_context_reg(cs, R_028C44_PA_SC_BINNER_CNTL_0,
+	radeon_set_context_reg(ctx_cs, R_028C44_PA_SC_BINNER_CNTL_0,
 			       pa_sc_binner_cntl_0);
-	radeon_set_context_reg(cs, R_028060_DB_DFSM_CONTROL,
+	radeon_set_context_reg(ctx_cs, R_028060_DB_DFSM_CONTROL,
 			       db_dfsm_control);
 }
 
 
 static void
-radv_pipeline_generate_depth_stencil_state(struct radeon_cmdbuf *cs,
+radv_pipeline_generate_depth_stencil_state(struct radeon_cmdbuf *ctx_cs,
                                            struct radv_pipeline *pipeline,
                                            const VkGraphicsPipelineCreateInfo *pCreateInfo,
                                            const struct radv_graphics_pipeline_create_info *extra)
@@ -2657,35 +2657,35 @@ radv_pipeline_generate_depth_stencil_state(struct radeon_cmdbuf *cs,
 		db_render_override |= S_02800C_DISABLE_VIEWPORT_CLAMP(1);
 	}
 
-	radeon_set_context_reg(cs, R_028800_DB_DEPTH_CONTROL, db_depth_control);
-	radeon_set_context_reg(cs, R_02842C_DB_STENCIL_CONTROL, db_stencil_control);
+	radeon_set_context_reg(ctx_cs, R_028800_DB_DEPTH_CONTROL, db_depth_control);
+	radeon_set_context_reg(ctx_cs, R_02842C_DB_STENCIL_CONTROL, db_stencil_control);
 
-	radeon_set_context_reg(cs, R_028000_DB_RENDER_CONTROL, db_render_control);
-	radeon_set_context_reg(cs, R_02800C_DB_RENDER_OVERRIDE, db_render_override);
-	radeon_set_context_reg(cs, R_028010_DB_RENDER_OVERRIDE2, db_render_override2);
+	radeon_set_context_reg(ctx_cs, R_028000_DB_RENDER_CONTROL, db_render_control);
+	radeon_set_context_reg(ctx_cs, R_02800C_DB_RENDER_OVERRIDE, db_render_override);
+	radeon_set_context_reg(ctx_cs, R_028010_DB_RENDER_OVERRIDE2, db_render_override2);
 }
 
 static void
-radv_pipeline_generate_blend_state(struct radeon_cmdbuf *cs,
+radv_pipeline_generate_blend_state(struct radeon_cmdbuf *ctx_cs,
                                    struct radv_pipeline *pipeline,
                                    const struct radv_blend_state *blend)
 {
-	radeon_set_context_reg_seq(cs, R_028780_CB_BLEND0_CONTROL, 8);
-	radeon_emit_array(cs, blend->cb_blend_control,
+	radeon_set_context_reg_seq(ctx_cs, R_028780_CB_BLEND0_CONTROL, 8);
+	radeon_emit_array(ctx_cs, blend->cb_blend_control,
 			  8);
-	radeon_set_context_reg(cs, R_028808_CB_COLOR_CONTROL, blend->cb_color_control);
-	radeon_set_context_reg(cs, R_028B70_DB_ALPHA_TO_MASK, blend->db_alpha_to_mask);
+	radeon_set_context_reg(ctx_cs, R_028808_CB_COLOR_CONTROL, blend->cb_color_control);
+	radeon_set_context_reg(ctx_cs, R_028B70_DB_ALPHA_TO_MASK, blend->db_alpha_to_mask);
 
 	if (pipeline->device->physical_device->has_rbplus) {
 
-		radeon_set_context_reg_seq(cs, R_028760_SX_MRT0_BLEND_OPT, 8);
-		radeon_emit_array(cs, blend->sx_mrt_blend_opt, 8);
+		radeon_set_context_reg_seq(ctx_cs, R_028760_SX_MRT0_BLEND_OPT, 8);
+		radeon_emit_array(ctx_cs, blend->sx_mrt_blend_opt, 8);
 	}
 
-	radeon_set_context_reg(cs, R_028714_SPI_SHADER_COL_FORMAT, blend->spi_shader_col_format);
+	radeon_set_context_reg(ctx_cs, R_028714_SPI_SHADER_COL_FORMAT, blend->spi_shader_col_format);
 
-	radeon_set_context_reg(cs, R_028238_CB_TARGET_MASK, blend->cb_target_mask);
-	radeon_set_context_reg(cs, R_02823C_CB_SHADER_MASK, blend->cb_shader_mask);
+	radeon_set_context_reg(ctx_cs, R_028238_CB_TARGET_MASK, blend->cb_target_mask);
+	radeon_set_context_reg(ctx_cs, R_02823C_CB_SHADER_MASK, blend->cb_shader_mask);
 
 	pipeline->graphics.col_format = blend->spi_shader_col_format;
 	pipeline->graphics.cb_target_mask = blend->cb_target_mask;
@@ -2703,7 +2703,7 @@ radv_get_conservative_raster_mode(const VkPipelineRasterizationStateCreateInfo *
 }
 
 static void
-radv_pipeline_generate_raster_state(struct radeon_cmdbuf *cs,
+radv_pipeline_generate_raster_state(struct radeon_cmdbuf *ctx_cs,
 				    struct radv_pipeline *pipeline,
                                     const VkGraphicsPipelineCreateInfo *pCreateInfo)
 {
@@ -2712,14 +2712,14 @@ radv_pipeline_generate_raster_state(struct radeon_cmdbuf *cs,
 		radv_get_conservative_raster_mode(vkraster);
 	uint32_t pa_sc_conservative_rast = S_028C4C_NULL_SQUAD_AA_MASK_ENABLE(1);
 
-	radeon_set_context_reg(cs, R_028810_PA_CL_CLIP_CNTL,
+	radeon_set_context_reg(ctx_cs, R_028810_PA_CL_CLIP_CNTL,
 	                       S_028810_DX_CLIP_SPACE_DEF(1) | // vulkan uses DX conventions.
 	                       S_028810_ZCLIP_NEAR_DISABLE(vkraster->depthClampEnable ? 1 : 0) |
 	                       S_028810_ZCLIP_FAR_DISABLE(vkraster->depthClampEnable ? 1 : 0) |
 	                       S_028810_DX_RASTERIZATION_KILL(vkraster->rasterizerDiscardEnable ? 1 : 0) |
 	                       S_028810_DX_LINEAR_ATTR_CLIP_ENA(1));
 
-	radeon_set_context_reg(cs, R_0286D4_SPI_INTERP_CONTROL_0,
+	radeon_set_context_reg(ctx_cs, R_0286D4_SPI_INTERP_CONTROL_0,
 	                       S_0286D4_FLAT_SHADE_ENA(1) |
 	                       S_0286D4_PNT_SPRITE_ENA(1) |
 	                       S_0286D4_PNT_SPRITE_OVRD_X(V_0286D4_SPI_PNT_SPRITE_SEL_S) |
@@ -2728,12 +2728,12 @@ radv_pipeline_generate_raster_state(struct radeon_cmdbuf *cs,
 	                       S_0286D4_PNT_SPRITE_OVRD_W(V_0286D4_SPI_PNT_SPRITE_SEL_1) |
 	                       S_0286D4_PNT_SPRITE_TOP_1(0)); /* vulkan is top to bottom - 1.0 at bottom */
 
-	radeon_set_context_reg(cs, R_028BE4_PA_SU_VTX_CNTL,
+	radeon_set_context_reg(ctx_cs, R_028BE4_PA_SU_VTX_CNTL,
 	                       S_028BE4_PIX_CENTER(1) | // TODO verify
 	                       S_028BE4_ROUND_MODE(V_028BE4_X_ROUND_TO_EVEN) |
 	                       S_028BE4_QUANT_MODE(V_028BE4_X_16_8_FIXED_POINT_1_256TH));
 
-	radeon_set_context_reg(cs, R_028814_PA_SU_SC_MODE_CNTL,
+	radeon_set_context_reg(ctx_cs, R_028814_PA_SU_SC_MODE_CNTL,
 	                       S_028814_FACE(vkraster->frontFace) |
 	                       S_028814_CULL_FRONT(!!(vkraster->cullMode & VK_CULL_MODE_FRONT_BIT)) |
 	                       S_028814_CULL_BACK(!!(vkraster->cullMode & VK_CULL_MODE_BACK_BIT)) |
@@ -2774,37 +2774,37 @@ radv_pipeline_generate_raster_state(struct radeon_cmdbuf *cs,
 		}
 	}
 
-	radeon_set_context_reg(cs, R_028C4C_PA_SC_CONSERVATIVE_RASTERIZATION_CNTL,
+	radeon_set_context_reg(ctx_cs, R_028C4C_PA_SC_CONSERVATIVE_RASTERIZATION_CNTL,
 				   pa_sc_conservative_rast);
 }
 
 
 static void
-radv_pipeline_generate_multisample_state(struct radeon_cmdbuf *cs,
+radv_pipeline_generate_multisample_state(struct radeon_cmdbuf *ctx_cs,
                                          struct radv_pipeline *pipeline)
 {
 	struct radv_multisample_state *ms = &pipeline->graphics.ms;
 
-	radeon_set_context_reg_seq(cs, R_028C38_PA_SC_AA_MASK_X0Y0_X1Y0, 2);
-	radeon_emit(cs, ms->pa_sc_aa_mask[0]);
-	radeon_emit(cs, ms->pa_sc_aa_mask[1]);
+	radeon_set_context_reg_seq(ctx_cs, R_028C38_PA_SC_AA_MASK_X0Y0_X1Y0, 2);
+	radeon_emit(ctx_cs, ms->pa_sc_aa_mask[0]);
+	radeon_emit(ctx_cs, ms->pa_sc_aa_mask[1]);
 
-	radeon_set_context_reg(cs, R_028804_DB_EQAA, ms->db_eqaa);
-	radeon_set_context_reg(cs, R_028A4C_PA_SC_MODE_CNTL_1, ms->pa_sc_mode_cntl_1);
+	radeon_set_context_reg(ctx_cs, R_028804_DB_EQAA, ms->db_eqaa);
+	radeon_set_context_reg(ctx_cs, R_028A4C_PA_SC_MODE_CNTL_1, ms->pa_sc_mode_cntl_1);
 
 	/* The exclusion bits can be set to improve rasterization efficiency
 	 * if no sample lies on the pixel boundary (-8 sample offset). It's
 	 * currently always TRUE because the driver doesn't support 16 samples.
 	 */
 	bool exclusion = pipeline->device->physical_device->rad_info.chip_class >= CIK;
-	radeon_set_context_reg(cs, R_02882C_PA_SU_PRIM_FILTER_CNTL,
+	radeon_set_context_reg(ctx_cs, R_02882C_PA_SU_PRIM_FILTER_CNTL,
 			       S_02882C_XMAX_RIGHT_EXCLUSION(exclusion) |
 			       S_02882C_YMAX_BOTTOM_EXCLUSION(exclusion));
 }
 
 static void
-radv_pipeline_generate_vgt_gs_mode(struct radeon_cmdbuf *cs,
-                                   const struct radv_pipeline *pipeline)
+radv_pipeline_generate_vgt_gs_mode(struct radeon_cmdbuf *ctx_cs,
+                                   struct radv_pipeline *pipeline)
 {
 	const struct radv_vs_output_info *outinfo = get_vs_output_info(pipeline);
 
@@ -2822,12 +2822,13 @@ radv_pipeline_generate_vgt_gs_mode(struct radeon_cmdbuf *cs,
 		vgt_primitiveid_en = true;
 	}
 
-	radeon_set_context_reg(cs, R_028A84_VGT_PRIMITIVEID_EN, vgt_primitiveid_en);
-	radeon_set_context_reg(cs, R_028A40_VGT_GS_MODE, vgt_gs_mode);
+	radeon_set_context_reg(ctx_cs, R_028A84_VGT_PRIMITIVEID_EN, vgt_primitiveid_en);
+	radeon_set_context_reg(ctx_cs, R_028A40_VGT_GS_MODE, vgt_gs_mode);
 }
 
 static void
-radv_pipeline_generate_hw_vs(struct radeon_cmdbuf *cs,
+radv_pipeline_generate_hw_vs(struct radeon_cmdbuf *ctx_cs,
+			     struct radeon_cmdbuf *cs,
 			     struct radv_pipeline *pipeline,
 			     struct radv_shader_variant *shader)
 {
@@ -2848,10 +2849,10 @@ radv_pipeline_generate_hw_vs(struct radeon_cmdbuf *cs,
 		outinfo->writes_layer ||
 		outinfo->writes_viewport_index;
 
-	radeon_set_context_reg(cs, R_0286C4_SPI_VS_OUT_CONFIG,
+	radeon_set_context_reg(ctx_cs, R_0286C4_SPI_VS_OUT_CONFIG,
 	                       S_0286C4_VS_EXPORT_COUNT(MAX2(1, outinfo->param_exports) - 1));
 
-	radeon_set_context_reg(cs, R_02870C_SPI_SHADER_POS_FORMAT,
+	radeon_set_context_reg(ctx_cs, R_02870C_SPI_SHADER_POS_FORMAT,
 	                       S_02870C_POS0_EXPORT_FORMAT(V_02870C_SPI_SHADER_4COMP) |
 	                       S_02870C_POS1_EXPORT_FORMAT(outinfo->pos_exports > 1 ?
 	                                                   V_02870C_SPI_SHADER_4COMP :
@@ -2863,13 +2864,13 @@ radv_pipeline_generate_hw_vs(struct radeon_cmdbuf *cs,
 	                                                   V_02870C_SPI_SHADER_4COMP :
 	                                                   V_02870C_SPI_SHADER_NONE));
 
-	radeon_set_context_reg(cs, R_028818_PA_CL_VTE_CNTL,
+	radeon_set_context_reg(ctx_cs, R_028818_PA_CL_VTE_CNTL,
 			       S_028818_VTX_W0_FMT(1) |
 			       S_028818_VPORT_X_SCALE_ENA(1) | S_028818_VPORT_X_OFFSET_ENA(1) |
 			       S_028818_VPORT_Y_SCALE_ENA(1) | S_028818_VPORT_Y_OFFSET_ENA(1) |
 			       S_028818_VPORT_Z_SCALE_ENA(1) | S_028818_VPORT_Z_OFFSET_ENA(1));
 
-	radeon_set_context_reg(cs, R_02881C_PA_CL_VS_OUT_CNTL,
+	radeon_set_context_reg(ctx_cs, R_02881C_PA_CL_VS_OUT_CNTL,
 	                       S_02881C_USE_VTX_POINT_SIZE(outinfo->writes_pointsize) |
 	                       S_02881C_USE_VTX_RENDER_TARGET_INDX(outinfo->writes_layer) |
 	                       S_02881C_USE_VTX_VIEWPORT_INDX(outinfo->writes_viewport_index) |
@@ -2881,7 +2882,7 @@ radv_pipeline_generate_hw_vs(struct radeon_cmdbuf *cs,
 	                       clip_dist_mask);
 
 	if (pipeline->device->physical_device->rad_info.chip_class <= VI)
-		radeon_set_context_reg(cs, R_028AB4_VGT_REUSE_OFF,
+		radeon_set_context_reg(ctx_cs, R_028AB4_VGT_REUSE_OFF,
 		                       outinfo->writes_viewport_index);
 }
 
@@ -2949,7 +2950,8 @@ radv_pipeline_generate_hw_hs(struct radeon_cmdbuf *cs,
 }
 
 static void
-radv_pipeline_generate_vertex_shader(struct radeon_cmdbuf *cs,
+radv_pipeline_generate_vertex_shader(struct radeon_cmdbuf *ctx_cs,
+				     struct radeon_cmdbuf *cs,
 				     struct radv_pipeline *pipeline,
 				     const struct radv_tessellation_state *tess)
 {
@@ -2965,11 +2967,12 @@ radv_pipeline_generate_vertex_shader(struct radeon_cmdbuf *cs,
 	else if (vs->info.vs.as_es)
 		radv_pipeline_generate_hw_es(cs, pipeline, vs);
 	else
-		radv_pipeline_generate_hw_vs(cs, pipeline, vs);
+		radv_pipeline_generate_hw_vs(ctx_cs, cs, pipeline, vs);
 }
 
 static void
-radv_pipeline_generate_tess_shaders(struct radeon_cmdbuf *cs,
+radv_pipeline_generate_tess_shaders(struct radeon_cmdbuf *ctx_cs,
+				    struct radeon_cmdbuf *cs,
 				    struct radv_pipeline *pipeline,
 				    const struct radv_tessellation_state *tess)
 {
@@ -2985,24 +2988,25 @@ radv_pipeline_generate_tess_shaders(struct radeon_cmdbuf *cs,
 		if (tes->info.tes.as_es)
 			radv_pipeline_generate_hw_es(cs, pipeline, tes);
 		else
-			radv_pipeline_generate_hw_vs(cs, pipeline, tes);
+			radv_pipeline_generate_hw_vs(ctx_cs, cs, pipeline, tes);
 	}
 
 	radv_pipeline_generate_hw_hs(cs, pipeline, tcs, tess);
 
-	radeon_set_context_reg(cs, R_028B6C_VGT_TF_PARAM,
+	radeon_set_context_reg(ctx_cs, R_028B6C_VGT_TF_PARAM,
 			       tess->tf_param);
 
 	if (pipeline->device->physical_device->rad_info.chip_class >= CIK)
-		radeon_set_context_reg_idx(cs, R_028B58_VGT_LS_HS_CONFIG, 2,
+		radeon_set_context_reg_idx(ctx_cs, R_028B58_VGT_LS_HS_CONFIG, 2,
 					   tess->ls_hs_config);
 	else
-		radeon_set_context_reg(cs, R_028B58_VGT_LS_HS_CONFIG,
+		radeon_set_context_reg(ctx_cs, R_028B58_VGT_LS_HS_CONFIG,
 				       tess->ls_hs_config);
 }
 
 static void
-radv_pipeline_generate_geometry_shader(struct radeon_cmdbuf *cs,
+radv_pipeline_generate_geometry_shader(struct radeon_cmdbuf *ctx_cs,
+				       struct radeon_cmdbuf *cs,
 				       struct radv_pipeline *pipeline,
 				       const struct radv_gs_state *gs_state)
 {
@@ -3023,32 +3027,32 @@ radv_pipeline_generate_geometry_shader(struct radeon_cmdbuf *cs,
 
 	offset = num_components[0] * gs_max_out_vertices;
 
-	radeon_set_context_reg_seq(cs, R_028A60_VGT_GSVS_RING_OFFSET_1, 3);
-	radeon_emit(cs, offset);
+	radeon_set_context_reg_seq(ctx_cs, R_028A60_VGT_GSVS_RING_OFFSET_1, 3);
+	radeon_emit(ctx_cs, offset);
 	if (max_stream >= 1)
 		offset += num_components[1] * gs_max_out_vertices;
-	radeon_emit(cs, offset);
+	radeon_emit(ctx_cs, offset);
 	if (max_stream >= 2)
 		offset += num_components[2] * gs_max_out_vertices;
-	radeon_emit(cs, offset);
+	radeon_emit(ctx_cs, offset);
 	if (max_stream >= 3)
 		offset += num_components[3] * gs_max_out_vertices;
-	radeon_set_context_reg(cs, R_028AB0_VGT_GSVS_RING_ITEMSIZE, offset);
+	radeon_set_context_reg(ctx_cs, R_028AB0_VGT_GSVS_RING_ITEMSIZE, offset);
 
-	radeon_set_context_reg(cs, R_028B38_VGT_GS_MAX_VERT_OUT, gs->info.gs.vertices_out);
+	radeon_set_context_reg(ctx_cs, R_028B38_VGT_GS_MAX_VERT_OUT, gs->info.gs.vertices_out);
 
-	radeon_set_context_reg_seq(cs, R_028B5C_VGT_GS_VERT_ITEMSIZE, 4);
-	radeon_emit(cs, num_components[0]);
-	radeon_emit(cs, (max_stream >= 1) ? num_components[1] : 0);
-	radeon_emit(cs, (max_stream >= 2) ? num_components[2] : 0);
-	radeon_emit(cs, (max_stream >= 3) ? num_components[3] : 0);
+	radeon_set_context_reg_seq(ctx_cs, R_028B5C_VGT_GS_VERT_ITEMSIZE, 4);
+	radeon_emit(ctx_cs, num_components[0]);
+	radeon_emit(ctx_cs, (max_stream >= 1) ? num_components[1] : 0);
+	radeon_emit(ctx_cs, (max_stream >= 2) ? num_components[2] : 0);
+	radeon_emit(ctx_cs, (max_stream >= 3) ? num_components[3] : 0);
 
 	uint32_t gs_num_invocations = gs->info.gs.invocations;
-	radeon_set_context_reg(cs, R_028B90_VGT_GS_INSTANCE_CNT,
+	radeon_set_context_reg(ctx_cs, R_028B90_VGT_GS_INSTANCE_CNT,
 			       S_028B90_CNT(MIN2(gs_num_invocations, 127)) |
 			       S_028B90_ENABLE(gs_num_invocations > 0));
 
-	radeon_set_context_reg(cs, R_028AAC_VGT_ESGS_RING_ITEMSIZE,
+	radeon_set_context_reg(ctx_cs, R_028AAC_VGT_ESGS_RING_ITEMSIZE,
 			       gs_state->vgt_esgs_ring_itemsize);
 
 	va = radv_buffer_get_va(gs->bo) + gs->bo_offset;
@@ -3062,8 +3066,8 @@ radv_pipeline_generate_geometry_shader(struct radeon_cmdbuf *cs,
 		radeon_emit(cs, gs->rsrc1);
 		radeon_emit(cs, gs->rsrc2 | S_00B22C_LDS_SIZE(gs_state->lds_size));
 
-		radeon_set_context_reg(cs, R_028A44_VGT_GS_ONCHIP_CNTL, gs_state->vgt_gs_onchip_cntl);
-		radeon_set_context_reg(cs, R_028A94_VGT_GS_MAX_PRIMS_PER_SUBGROUP, gs_state->vgt_gs_max_prims_per_subgroup);
+		radeon_set_context_reg(ctx_cs, R_028A44_VGT_GS_ONCHIP_CNTL, gs_state->vgt_gs_onchip_cntl);
+		radeon_set_context_reg(ctx_cs, R_028A94_VGT_GS_MAX_PRIMS_PER_SUBGROUP, gs_state->vgt_gs_max_prims_per_subgroup);
 	} else {
 		radeon_set_sh_reg_seq(cs, R_00B220_SPI_SHADER_PGM_LO_GS, 4);
 		radeon_emit(cs, va >> 8);
@@ -3072,7 +3076,7 @@ radv_pipeline_generate_geometry_shader(struct radeon_cmdbuf *cs,
 		radeon_emit(cs, gs->rsrc2);
 	}
 
-	radv_pipeline_generate_hw_vs(cs, pipeline, pipeline->gs_copy_shader);
+	radv_pipeline_generate_hw_vs(ctx_cs, cs, pipeline, pipeline->gs_copy_shader);
 }
 
 static uint32_t offset_to_ps_input(uint32_t offset, bool flat_shade)
@@ -3094,8 +3098,8 @@ static uint32_t offset_to_ps_input(uint32_t offset, bool flat_shade)
 }
 
 static void
-radv_pipeline_generate_ps_inputs(struct radeon_cmdbuf *cs,
-                                 struct radv_pipeline *pipeline)
+radv_pipeline_generate_ps_inputs(struct radeon_cmdbuf *ctx_cs,
+				 struct radv_pipeline *pipeline)
 {
 	struct radv_shader_variant *ps = pipeline->shaders[MESA_SHADER_FRAGMENT];
 	const struct radv_vs_output_info *outinfo = get_vs_output_info(pipeline);
@@ -3166,9 +3170,9 @@ radv_pipeline_generate_ps_inputs(struct radeon_cmdbuf *cs,
 	}
 
 	if (ps_offset) {
-		radeon_set_context_reg_seq(cs, R_028644_SPI_PS_INPUT_CNTL_0, ps_offset);
+		radeon_set_context_reg_seq(ctx_cs, R_028644_SPI_PS_INPUT_CNTL_0, ps_offset);
 		for (unsigned i = 0; i < ps_offset; i++) {
-			radeon_emit(cs, ps_input_cntl[i]);
+			radeon_emit(ctx_cs, ps_input_cntl[i]);
 		}
 	}
 }
@@ -3206,7 +3210,8 @@ radv_compute_db_shader_control(const struct radv_device *device,
 }
 
 static void
-radv_pipeline_generate_fragment_shader(struct radeon_cmdbuf *cs,
+radv_pipeline_generate_fragment_shader(struct radeon_cmdbuf *ctx_cs,
+				       struct radeon_cmdbuf *cs,
 				       struct radv_pipeline *pipeline)
 {
 	struct radv_shader_variant *ps;
@@ -3222,22 +3227,22 @@ radv_pipeline_generate_fragment_shader(struct radeon_cmdbuf *cs,
 	radeon_emit(cs, ps->rsrc1);
 	radeon_emit(cs, ps->rsrc2);
 
-	radeon_set_context_reg(cs, R_02880C_DB_SHADER_CONTROL,
+	radeon_set_context_reg(ctx_cs, R_02880C_DB_SHADER_CONTROL,
 	                       radv_compute_db_shader_control(pipeline->device,
 							      pipeline, ps));
 
-	radeon_set_context_reg(cs, R_0286CC_SPI_PS_INPUT_ENA,
+	radeon_set_context_reg(ctx_cs, R_0286CC_SPI_PS_INPUT_ENA,
 			       ps->config.spi_ps_input_ena);
 
-	radeon_set_context_reg(cs, R_0286D0_SPI_PS_INPUT_ADDR,
+	radeon_set_context_reg(ctx_cs, R_0286D0_SPI_PS_INPUT_ADDR,
 			       ps->config.spi_ps_input_addr);
 
-	radeon_set_context_reg(cs, R_0286D8_SPI_PS_IN_CONTROL,
+	radeon_set_context_reg(ctx_cs, R_0286D8_SPI_PS_IN_CONTROL,
 			       S_0286D8_NUM_INTERP(ps->info.fs.num_interp));
 
-	radeon_set_context_reg(cs, R_0286E0_SPI_BARYC_CNTL, pipeline->graphics.spi_baryc_cntl);
+	radeon_set_context_reg(ctx_cs, R_0286E0_SPI_BARYC_CNTL, pipeline->graphics.spi_baryc_cntl);
 
-	radeon_set_context_reg(cs, R_028710_SPI_SHADER_Z_FORMAT,
+	radeon_set_context_reg(ctx_cs, R_028710_SPI_SHADER_Z_FORMAT,
 	                       ac_get_spi_shader_z_format(ps->info.info.ps.writes_z,
 	                                                  ps->info.info.ps.writes_stencil,
 	                                                  ps->info.info.ps.writes_sample_mask));
@@ -3250,7 +3255,7 @@ radv_pipeline_generate_fragment_shader(struct radeon_cmdbuf *cs,
 }
 
 static void
-radv_pipeline_generate_vgt_vertex_reuse(struct radeon_cmdbuf *cs,
+radv_pipeline_generate_vgt_vertex_reuse(struct radeon_cmdbuf *ctx_cs,
 					struct radv_pipeline *pipeline)
 {
 	if (pipeline->device->physical_device->rad_info.family < CHIP_POLARIS10)
@@ -3261,7 +3266,7 @@ radv_pipeline_generate_vgt_vertex_reuse(struct radeon_cmdbuf *cs,
 	    radv_get_shader(pipeline, MESA_SHADER_TESS_EVAL)->info.tes.spacing == TESS_SPACING_FRACTIONAL_ODD) {
 		vtx_reuse_depth = 14;
 	}
-	radeon_set_context_reg(cs, R_028C58_VGT_VERTEX_REUSE_BLOCK_CNTL,
+	radeon_set_context_reg(ctx_cs, R_028C58_VGT_VERTEX_REUSE_BLOCK_CNTL,
 	                       S_028C58_VTX_REUSE_DEPTH(vtx_reuse_depth));
 }
 
@@ -3331,38 +3336,46 @@ radv_pipeline_generate_pm4(struct radv_pipeline *pipeline,
                            const struct radv_gs_state *gs,
                            unsigned prim, unsigned gs_out)
 {
-	pipeline->cs.buf = malloc(4 * 256);
-	pipeline->cs.max_dw = 256;
-
-	radv_pipeline_generate_depth_stencil_state(&pipeline->cs, pipeline, pCreateInfo, extra);
-	radv_pipeline_generate_blend_state(&pipeline->cs, pipeline, blend);
-	radv_pipeline_generate_raster_state(&pipeline->cs, pipeline, pCreateInfo);
-	radv_pipeline_generate_multisample_state(&pipeline->cs, pipeline);
-	radv_pipeline_generate_vgt_gs_mode(&pipeline->cs, pipeline);
-	radv_pipeline_generate_vertex_shader(&pipeline->cs, pipeline, tess);
-	radv_pipeline_generate_tess_shaders(&pipeline->cs, pipeline, tess);
-	radv_pipeline_generate_geometry_shader(&pipeline->cs, pipeline, gs);
-	radv_pipeline_generate_fragment_shader(&pipeline->cs, pipeline);
-	radv_pipeline_generate_ps_inputs(&pipeline->cs, pipeline);
-	radv_pipeline_generate_vgt_vertex_reuse(&pipeline->cs, pipeline);
-	radv_pipeline_generate_binning_state(&pipeline->cs, pipeline, pCreateInfo);
-
-	radeon_set_context_reg(&pipeline->cs, R_0286E8_SPI_TMPRING_SIZE,
+	struct radeon_cmdbuf *ctx_cs = &pipeline->ctx_cs;
+	struct radeon_cmdbuf *cs = &pipeline->cs;
+
+	cs->max_dw = 64;
+	ctx_cs->max_dw = 256;
+	cs->buf = malloc(4 * (cs->max_dw + ctx_cs->max_dw));
+	ctx_cs->buf = cs->buf + cs->max_dw;
+
+	radv_pipeline_generate_depth_stencil_state(ctx_cs, pipeline, pCreateInfo, extra);
+	radv_pipeline_generate_blend_state(ctx_cs, pipeline, blend);
+	radv_pipeline_generate_raster_state(ctx_cs, pipeline, pCreateInfo);
+	radv_pipeline_generate_multisample_state(ctx_cs, pipeline);
+	radv_pipeline_generate_vgt_gs_mode(ctx_cs, pipeline);
+	radv_pipeline_generate_vertex_shader(ctx_cs, cs, pipeline, tess);
+	radv_pipeline_generate_tess_shaders(ctx_cs, cs, pipeline, tess);
+	radv_pipeline_generate_geometry_shader(ctx_cs, cs, pipeline, gs);
+	radv_pipeline_generate_fragment_shader(ctx_cs, cs, pipeline);
+	radv_pipeline_generate_ps_inputs(ctx_cs, pipeline);
+	radv_pipeline_generate_vgt_vertex_reuse(ctx_cs, pipeline);
+	radv_pipeline_generate_binning_state(ctx_cs, pipeline, pCreateInfo);
+
+	radeon_set_context_reg(ctx_cs, R_0286E8_SPI_TMPRING_SIZE,
 			       S_0286E8_WAVES(pipeline->max_waves) |
 			       S_0286E8_WAVESIZE(pipeline->scratch_bytes_per_wave >> 10));
 
-	radeon_set_context_reg(&pipeline->cs, R_028B54_VGT_SHADER_STAGES_EN, radv_compute_vgt_shader_stages_en(pipeline));
+	radeon_set_context_reg(ctx_cs, R_028B54_VGT_SHADER_STAGES_EN, radv_compute_vgt_shader_stages_en(pipeline));
 
 	if (pipeline->device->physical_device->rad_info.chip_class >= CIK) {
-		radeon_set_uconfig_reg_idx(&pipeline->cs, R_030908_VGT_PRIMITIVE_TYPE, 1, prim);
+		radeon_set_uconfig_reg_idx(cs, R_030908_VGT_PRIMITIVE_TYPE, 1, prim);
 	} else {
-		radeon_set_config_reg(&pipeline->cs, R_008958_VGT_PRIMITIVE_TYPE, prim);
+		radeon_set_config_reg(cs, R_008958_VGT_PRIMITIVE_TYPE, prim);
 	}
-	radeon_set_context_reg(&pipeline->cs, R_028A6C_VGT_GS_OUT_PRIM_TYPE, gs_out);
+	radeon_set_context_reg(ctx_cs, R_028A6C_VGT_GS_OUT_PRIM_TYPE, gs_out);
 
-	radeon_set_context_reg(&pipeline->cs, R_02820C_PA_SC_CLIPRECT_RULE, radv_compute_cliprect_rule(pCreateInfo));
+	radeon_set_context_reg(ctx_cs, R_02820C_PA_SC_CLIPRECT_RULE, radv_compute_cliprect_rule(pCreateInfo));
 
-	assert(pipeline->cs.cdw <= pipeline->cs.max_dw);
+	pipeline->ctx_cs_hash = _mesa_hash_data(ctx_cs->buf, ctx_cs->cdw * 4);
+
+	assert(ctx_cs->cdw <= ctx_cs->max_dw);
+	assert(cs->cdw <= cs->max_dw);
 }
 
 static struct radv_ia_multi_vgt_param_helpers
diff --git a/src/amd/vulkan/radv_private.h b/src/amd/vulkan/radv_private.h
index 064c221549c..280504ea03f 100644
--- a/src/amd/vulkan/radv_private.h
+++ b/src/amd/vulkan/radv_private.h
@@ -1367,6 +1367,8 @@ struct radv_pipeline {
 	VkShaderStageFlags                           active_stages;
 
 	struct radeon_cmdbuf                      cs;
+	uint32_t                                  ctx_cs_hash;
+	struct radeon_cmdbuf                      ctx_cs;
 
 	struct radv_vertex_elements_info             vertex_elements;