i965/urb: fixes division by zero

[mesa.git] / src / mesa / drivers / dri / i965 / brw_vec4_visitor.cpp

diff --git a/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp b/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
index 4cfbc143d5a605e670d9fa355049df952b4440fc..f73d678252894e21856c209247d51c19466fb57d 100644 (file)
--- a/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
+++ b/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
@@ -587,15 +587,18 @@ type_size_vec4(const struct glsl_type *type)
    case GLSL_TYPE_INT:
    case GLSL_TYPE_FLOAT:
    case GLSL_TYPE_BOOL:
+   case GLSL_TYPE_DOUBLE:
       if (type->is_matrix()) {
-        return type->matrix_columns;
+         const glsl_type *col_type = type->column_type();
+         unsigned col_slots = col_type->is_dual_slot_double() ? 2 : 1;
+         return type->matrix_columns * col_slots;
       } else {
-        /* Regardless of size of vector, it gets a vec4. This is bad
-         * packing for things like floats, but otherwise arrays become a
-         * mess.  Hopefully a later pass over the code can pack scalars
-         * down if appropriate.
-         */
-        return 1;
+         /* Regardless of size of vector, it gets a vec4. This is bad
+          * packing for things like floats, but otherwise arrays become a
+          * mess.  Hopefully a later pass over the code can pack scalars
+          * down if appropriate.
+          */
+         return type->is_dual_slot_double() ? 2 : 1;
       }
    case GLSL_TYPE_ARRAY:
       assert(type->length > 0);
@@ -619,7 +622,6 @@ type_size_vec4(const struct glsl_type *type)
    case GLSL_TYPE_IMAGE:
       return DIV_ROUND_UP(BRW_IMAGE_PARAM_SIZE, 4);
    case GLSL_TYPE_VOID:
-   case GLSL_TYPE_DOUBLE:
    case GLSL_TYPE_ERROR:
    case GLSL_TYPE_INTERFACE:
    case GLSL_TYPE_FUNCTION:
@@ -758,7 +760,7 @@ vec4_visitor::emit_pull_constant_load_reg(dst_reg dst,
       pull->mlen = 2;
       pull->header_size = 1;
    } else if (devinfo->gen >= 7) {
-      dst_reg grf_offset = dst_reg(this, glsl_type::int_type);
+      dst_reg grf_offset = dst_reg(this, glsl_type::uint_type);
 
       grf_offset.type = offset_reg.type;
 
@@ -1056,10 +1058,16 @@ vec4_visitor::emit_texture(ir_texture_opcode op,
    /* fixup num layers (z) for cube arrays: hardware returns faces * layers;
     * spec requires layers.
     */
-   if (op == ir_txs && is_cube_array) {
-      emit_math(SHADER_OPCODE_INT_QUOTIENT,
-                writemask(inst->dst, WRITEMASK_Z),
-                src_reg(inst->dst), brw_imm_d(6));
+   if (op == ir_txs) {
+      if (is_cube_array) {
+         emit_math(SHADER_OPCODE_INT_QUOTIENT,
+                   writemask(inst->dst, WRITEMASK_Z),
+                   src_reg(inst->dst), brw_imm_d(6));
+      } else if (devinfo->gen < 7) {
+         /* Gen4-6 return 0 instead of 1 for single layer surfaces. */
+         emit_minmax(BRW_CONDITIONAL_GE, writemask(inst->dst, WRITEMASK_Z),
+                     src_reg(inst->dst), brw_imm_d(1));
+      }
    }
 
    if (devinfo->gen == 6 && op == ir_tg4) {
@@ -1581,21 +1589,21 @@ vec4_visitor::move_grf_array_access_to_scratch()
 void
 vec4_visitor::emit_pull_constant_load(bblock_t *block, vec4_instruction *inst,
                                      dst_reg temp, src_reg orig_src,
-                                     int base_offset, src_reg indirect)
+                                      int base_offset, src_reg indirect)
 {
    int reg_offset = base_offset + orig_src.reg_offset;
    const unsigned index = prog_data->base.binding_table.pull_constants_start;
 
    src_reg offset;
    if (indirect.file != BAD_FILE) {
-      offset = src_reg(this, glsl_type::int_type);
+      offset = src_reg(this, glsl_type::uint_type);
 
       emit_before(block, inst, ADD(dst_reg(offset), indirect,
-                                   brw_imm_d(reg_offset * 16)));
+                                   brw_imm_ud(reg_offset * 16)));
    } else if (devinfo->gen >= 8) {
       /* Store the offset in a GRF so we can send-from-GRF. */
-      offset = src_reg(this, glsl_type::int_type);
-      emit_before(block, inst, MOV(dst_reg(offset), brw_imm_d(reg_offset * 16)));
+      offset = src_reg(this, glsl_type::uint_type);
+      emit_before(block, inst, MOV(dst_reg(offset), brw_imm_ud(reg_offset * 16)));
    } else {
       offset = brw_imm_d(reg_offset * 16);
    }
@@ -1623,6 +1631,14 @@ vec4_visitor::emit_pull_constant_load(bblock_t *block, vec4_instruction *inst,
 void
 vec4_visitor::move_uniform_array_access_to_pull_constants()
 {
+   /* The vulkan dirver doesn't support pull constants other than UBOs so
+    * everything has to be pushed regardless.
+    */
+   if (stage_prog_data->pull_param == NULL) {
+      split_uniform_registers();
+      return;
+   }
+
    int pull_constant_loc[this->uniforms];
    memset(pull_constant_loc, -1, sizeof(pull_constant_loc));