Introduce .editorconfig

[mesa.git] / src / gallium / drivers / r600 / r600_asm.c

diff --git a/src/gallium/drivers/r600/r600_asm.c b/src/gallium/drivers/r600/r600_asm.c
index ba17909bf7c73df33f5079cb798876da2ac6485d..f85993d45118146be388b217df23dbd43952fe41 100644 (file)
--- a/src/gallium/drivers/r600/r600_asm.c
+++ b/src/gallium/drivers/r600/r600_asm.c
@@ -27,6 +27,7 @@
 #include "r600d.h"
 
 #include <errno.h>
+#include "util/u_bitcast.h"
 #include "util/u_dump.h"
 #include "util/u_memory.h"
 #include "util/u_math.h"
@@ -237,7 +238,7 @@ int r600_bytecode_add_output(struct r600_bytecode *bc,
 /* alu instructions that can ony exits once per group */
 static int is_alu_once_inst(struct r600_bytecode *bc, struct r600_bytecode_alu *alu)
 {
-       return r600_isa_alu(alu->op)->flags & (AF_KILL | AF_PRED);
+       return r600_isa_alu(alu->op)->flags & (AF_KILL | AF_PRED) || alu->is_lds_idx_op || alu->op == ALU_OP0_GROUP_BARRIER;
 }
 
 static int is_alu_reduction_inst(struct r600_bytecode *bc, struct r600_bytecode_alu *alu)
@@ -268,6 +269,24 @@ static int alu_uses_rel(struct r600_bytecode *bc, struct r600_bytecode_alu *alu)
        return 0;
 }
 
+static int is_lds_read(int sel)
+{
+  return sel == EG_V_SQ_ALU_SRC_LDS_OQ_A_POP || sel == EG_V_SQ_ALU_SRC_LDS_OQ_B_POP;
+}
+
+static int alu_uses_lds(struct r600_bytecode *bc, struct r600_bytecode_alu *alu)
+{
+       unsigned num_src = r600_bytecode_get_num_operands(bc, alu);
+       unsigned src;
+
+       for (src = 0; src < num_src; ++src) {
+               if (is_lds_read(alu->src[src].sel)) {
+                       return 1;
+               }
+       }
+       return 0;
+}
+
 static int is_alu_64bit_inst(struct r600_bytecode *bc, struct r600_bytecode_alu *alu)
 {
        const struct alu_op_info *op = r600_isa_alu(alu->op);
@@ -787,6 +806,8 @@ static int merge_inst_groups(struct r600_bytecode *bc, struct r600_bytecode_alu
                                }
                                have_rel = 1;
                        }
+                       if (alu_uses_lds(bc, prev[i]))
+                               return 0;
 
                        num_once_inst += is_alu_once_inst(bc, prev[i]);
                }
@@ -800,7 +821,7 @@ static int merge_inst_groups(struct r600_bytecode *bc, struct r600_bytecode_alu
                } else if (prev[i] && slots[i]) {
                        if (max_slots == 5 && result[4] == NULL && prev[4] == NULL && slots[4] == NULL) {
                                /* Trans unit is still free try to use it. */
-                               if (is_alu_any_unit_inst(bc, slots[i])) {
+                               if (is_alu_any_unit_inst(bc, slots[i]) && !alu_uses_lds(bc, slots[i])) {
                                        result[i] = prev[i];
                                        result[4] = slots[i];
                                } else if (is_alu_any_unit_inst(bc, prev[i])) {
@@ -1670,7 +1691,7 @@ int r600_bytecode_build(struct r600_bytecode *bc)
        if (!bc->nstack) // If not 0, Stack_size already provided by llvm
                bc->nstack = bc->stack.max_entries;
 
-       if (bc->type == TGSI_PROCESSOR_VERTEX && !bc->nstack) {
+       if ((bc->type == PIPE_SHADER_VERTEX || bc->type == PIPE_SHADER_TESS_EVAL || bc->type == PIPE_SHADER_TESS_CTRL) && !bc->nstack) {
                bc->nstack = 1;
        }
 
@@ -1936,7 +1957,7 @@ static int print_src(struct r600_bytecode_alu *alu, unsigned idx)
                        need_chan = 1;
                        break;
                case V_SQ_ALU_SRC_LITERAL:
-                       o += fprintf(stderr, "[0x%08X %f]", src->value, *(float*)&src->value);
+                       o += fprintf(stderr, "[0x%08X %f]", src->value, u_bitcast_u2f(src->value));
                        break;
                case V_SQ_ALU_SRC_0_5:
                        o += fprintf(stderr, "0.5");
@@ -2323,6 +2344,12 @@ void r600_vertex_data_type(enum pipe_format pformat,
                return;
        }
 
+       if (pformat == PIPE_FORMAT_B5G6R5_UNORM) {
+               *format = FMT_5_6_5;
+               *endian = r600_endian_swap(16);
+               return;
+       }
+
        desc = util_format_description(pformat);
        if (desc->layout != UTIL_FORMAT_LAYOUT_PLAIN) {
                goto out_unknown;
@@ -2595,7 +2622,8 @@ void *r600_create_vertex_fetch_shader(struct pipe_context *ctx,
                return NULL;
        }
 
-       u_suballocator_alloc(rctx->allocator_fetch_shader, fs_size, &shader->offset,
+       u_suballocator_alloc(rctx->allocator_fetch_shader, fs_size, 256,
+                            &shader->offset,
                             (struct pipe_resource**)&shader->buffer);
        if (!shader->buffer) {
                r600_bytecode_clear(&bc);
@@ -2613,7 +2641,7 @@ void *r600_create_vertex_fetch_shader(struct pipe_context *ctx,
        } else {
                memcpy(bytecode, bc.bytecode, fs_size);
        }
-       rctx->b.ws->buffer_unmap(shader->buffer->cs_buf);
+       rctx->b.ws->buffer_unmap(shader->buffer->buf);
 
        r600_bytecode_clear(&bc);
        return shader;