freedreno/a3xx: only emit dirty consts

[mesa.git] / src / gallium / drivers / radeonsi / si_compute.c

diff --git a/src/gallium/drivers/radeonsi/si_compute.c b/src/gallium/drivers/radeonsi/si_compute.c
index fc842d437194f392acf750e6f617862f46ce439b..be6441817963b1b6f99391194fee59a911f0cd2c 100644 (file)
--- a/src/gallium/drivers/radeonsi/si_compute.c
+++ b/src/gallium/drivers/radeonsi/si_compute.c
@@ -38,14 +38,14 @@
 #define NUM_USER_SGPRS 4
 #endif
 
-struct si_pipe_compute {
+struct si_compute {
        struct si_context *ctx;
 
        unsigned local_size;
        unsigned private_size;
        unsigned input_size;
        unsigned num_kernels;
-       struct si_pipe_shader *kernels;
+       struct si_shader *kernels;
        unsigned num_user_sgprs;
 
        struct r600_resource *input_buffer;
@@ -59,8 +59,7 @@ static void *si_create_compute_state(
        const struct pipe_compute_state *cso)
 {
        struct si_context *sctx = (struct si_context *)ctx;
-       struct si_pipe_compute *program =
-                                       CALLOC_STRUCT(si_pipe_compute);
+       struct si_compute *program = CALLOC_STRUCT(si_compute);
        const struct pipe_llvm_program_header *header;
        const unsigned char *code;
        unsigned i;
@@ -77,12 +76,12 @@ static void *si_create_compute_state(
 
        program->num_kernels = radeon_llvm_get_num_kernels(program->llvm_ctx, code,
                                                        header->num_bytes);
-       program->kernels = CALLOC(sizeof(struct si_pipe_shader),
+       program->kernels = CALLOC(sizeof(struct si_shader),
                                                        program->num_kernels);
        for (i = 0; i < program->num_kernels; i++) {
                LLVMModuleRef mod = radeon_llvm_get_kernel_module(program->llvm_ctx, i,
                                                        code, header->num_bytes);
-               si_compile_llvm(sctx, &program->kernels[i], mod);
+               si_compile_llvm(sctx->screen, &program->kernels[i], mod);
                LLVMDisposeModule(mod);
        }
 
@@ -95,7 +94,7 @@ static void *si_create_compute_state(
 static void si_bind_compute_state(struct pipe_context *ctx, void *state)
 {
        struct si_context *sctx = (struct si_context*)ctx;
-       sctx->cs_shader_state.program = (struct si_pipe_compute*)state;
+       sctx->cs_shader_state.program = (struct si_compute*)state;
 }
 
 static void si_set_global_binding(
@@ -105,7 +104,7 @@ static void si_set_global_binding(
 {
        unsigned i;
        struct si_context *sctx = (struct si_context*)ctx;
-       struct si_pipe_compute *program = sctx->cs_shader_state.program;
+       struct si_compute *program = sctx->cs_shader_state.program;
 
        if (!resources) {
                for (i = first; i < first + n; i++) {
@@ -169,7 +168,8 @@ static void si_launch_grid(
                uint32_t pc, const void *input)
 {
        struct si_context *sctx = (struct si_context*)ctx;
-       struct si_pipe_compute *program = sctx->cs_shader_state.program;
+       struct radeon_winsys_cs *cs = sctx->b.rings.gfx.cs;
+       struct si_compute *program = sctx->cs_shader_state.program;
        struct si_pm4_state *pm4 = CALLOC_STRUCT(si_pm4_state);
        struct r600_resource *input_buffer = program->input_buffer;
        unsigned kernel_args_size;
@@ -181,22 +181,22 @@ static void si_launch_grid(
        uint64_t shader_va;
        unsigned arg_user_sgpr_count = NUM_USER_SGPRS;
        unsigned i;
-       struct si_pipe_shader *shader = &program->kernels[pc];
+       struct si_shader *shader = &program->kernels[pc];
        unsigned lds_blocks;
        unsigned num_waves_for_scratch;
 
-       pm4->compute_pkt = true;
-       si_cmd_context_control(pm4);
+       radeon_emit(cs, PKT3(PKT3_CONTEXT_CONTROL, 1, 0) | PKT3_SHADER_TYPE_S(1));
+       radeon_emit(cs, 0x80000000);
+       radeon_emit(cs, 0x80000000);
 
-       si_pm4_cmd_begin(pm4, PKT3_EVENT_WRITE);
-       si_pm4_cmd_add(pm4, EVENT_TYPE(EVENT_TYPE_CACHE_FLUSH) |
-                           EVENT_INDEX(0x7) |
-                           EVENT_WRITE_INV_L2);
-       si_pm4_cmd_end(pm4, false);
+       sctx->b.flags |= R600_CONTEXT_INV_TEX_CACHE |
+                        R600_CONTEXT_INV_SHADER_CACHE |
+                        R600_CONTEXT_INV_CONST_CACHE |
+                        R600_CONTEXT_FLUSH_WITH_INV_L2 |
+                        R600_CONTEXT_FLAG_COMPUTE;
+       si_emit_cache_flush(&sctx->b, NULL);
 
-       si_pm4_inval_texture_cache(pm4);
-       si_pm4_inval_shader_cache(pm4);
-       si_cmd_surface_sync(pm4, pm4->cp_coher_cntl);
+       pm4->compute_pkt = true;
 
        /* Upload the kernel arguments */
 
@@ -361,14 +361,6 @@ static void si_launch_grid(
        si_pm4_cmd_add(pm4, 1); /* DISPATCH_INITIATOR */
         si_pm4_cmd_end(pm4, false);
 
-       si_pm4_cmd_begin(pm4, PKT3_EVENT_WRITE);
-       si_pm4_cmd_add(pm4, EVENT_TYPE(V_028A90_CS_PARTIAL_FLUSH | EVENT_INDEX(0x4)));
-       si_pm4_cmd_end(pm4, false);
-
-       si_pm4_inval_texture_cache(pm4);
-       si_pm4_inval_shader_cache(pm4);
-       si_cmd_surface_sync(pm4, pm4->cp_coher_cntl);
-
        si_pm4_emit(sctx, pm4);
 
 #if 0
@@ -379,11 +371,18 @@ static void si_launch_grid(
 #endif
 
        si_pm4_free_state(sctx, pm4, ~0);
+
+       sctx->b.flags |= R600_CONTEXT_CS_PARTIAL_FLUSH |
+                        R600_CONTEXT_INV_TEX_CACHE |
+                        R600_CONTEXT_INV_SHADER_CACHE |
+                        R600_CONTEXT_INV_CONST_CACHE |
+                        R600_CONTEXT_FLAG_COMPUTE;
+       si_emit_cache_flush(&sctx->b, NULL);
 }
 
 
 static void si_delete_compute_state(struct pipe_context *ctx, void* state){
-       struct si_pipe_compute *program = (struct si_pipe_compute *)state;
+       struct si_compute *program = (struct si_compute *)state;
 
        if (!state) {
                return;
@@ -392,7 +391,7 @@ static void si_delete_compute_state(struct pipe_context *ctx, void* state){
        if (program->kernels) {
                for (int i = 0; i < program->num_kernels; i++){
                        if (program->kernels[i].bo){
-                               si_pipe_shader_destroy(ctx, &program->kernels[i]);
+                               si_shader_destroy(ctx, &program->kernels[i]);
                        }
                }
                FREE(program->kernels);