summaryrefslogtreecommitdiffstats
path: root/src/compiler/glsl/glsl_to_nir.cpp
diff options
context:
space:
mode:
authorJuan A. Suarez Romero <[email protected]>2016-12-16 10:24:43 +0100
committerJuan A. Suarez Romero <[email protected]>2017-01-09 10:42:22 +0100
commitc2acf97fcc9b32eaa9778771282758e5652a8ad4 (patch)
treea7c8890b9a6a1532b55170d531310382742cb9de /src/compiler/glsl/glsl_to_nir.cpp
parent3551a2d3ad2661477ba3b0a36a13eeb68e28fe85 (diff)
nir/i965: use two slots from inputs_read for dvec3/dvec4 vertex input attributes
So far, input_reads was a bitmap tracking which vertex input locations were being used. In OpenGL, an attribute bigger than a vec4 (like a dvec3 or dvec4) consumes just one location, any other small attribute. So we mark the proper bit in inputs_read, and also the same bit in double_inputs_read if the attribute is a dvec3/dvec4. But in Vulkan, this is slightly different: a dvec3/dvec4 attribute consumes two locations, not just one. And hence two bits would be marked in inputs_read for the same vertex input attribute. To avoid handling two different situations in NIR, we just choose the latest one: in OpenGL, when creating NIR from GLSL/IR, any dvec3/dvec4 vertex input attribute is marked with two bits in the inputs_read bitmap (and also in the double_inputs_read), and following attributes are adjusted accordingly. As example, if in our GLSL/IR shader we have three attributes: layout(location = 0) vec3 attr0; layout(location = 1) dvec4 attr1; layout(location = 2) dvec3 attr2; then in our NIR shader we put attr0 in location 0, attr1 in locations 1 and 2, and attr2 in location 3 and 4. Checking carefully, basically we are using slots rather than locations in NIR. When emitting the vertices, we do a inverse map to know the corresponding location for each slot. v2 (Jason): - use two slots from inputs_read for dvec3/dvec4 NIR from GLSL/IR. v3 (Jason): - Fix commit log error. - Use ladder ifs and fix braces. - elements_double is divisible by 2, don't need DIV_ROUND_UP(). - Use if ladder instead of a switch. - Add comment about hardware restriction in 64bit vertex attributes. Reviewed-by: Jason Ekstrand <[email protected]>
Diffstat (limited to 'src/compiler/glsl/glsl_to_nir.cpp')
-rw-r--r--src/compiler/glsl/glsl_to_nir.cpp28
1 files changed, 28 insertions, 0 deletions
diff --git a/src/compiler/glsl/glsl_to_nir.cpp b/src/compiler/glsl/glsl_to_nir.cpp
index 69d4c2b20c6..33f71bf416a 100644
--- a/src/compiler/glsl/glsl_to_nir.cpp
+++ b/src/compiler/glsl/glsl_to_nir.cpp
@@ -129,6 +129,19 @@ private:
} /* end of anonymous namespace */
+static void
+nir_remap_attributes(nir_shader *shader)
+{
+ nir_foreach_variable(var, &shader->inputs) {
+ var->data.location += _mesa_bitcount_64(shader->info->double_inputs_read &
+ BITFIELD64_MASK(var->data.location));
+ }
+
+ /* Once the remap is done, reset double_inputs_read, so later it will have
+ * which location/slots are doubles */
+ shader->info->double_inputs_read = 0;
+}
+
nir_shader *
glsl_to_nir(const struct gl_shader_program *shader_prog,
gl_shader_stage stage,
@@ -146,6 +159,13 @@ glsl_to_nir(const struct gl_shader_program *shader_prog,
nir_lower_constant_initializers(shader, (nir_variable_mode)~0);
+ /* Remap the locations to slots so those requiring two slots will occupy
+ * two locations. For instance, if we have in the IR code a dvec3 attr0 in
+ * location 0 and vec4 attr1 in location 1, in NIR attr0 will use
+ * locations/slots 0 and 1, and attr1 will use location/slot 2 */
+ if (shader->stage == MESA_SHADER_VERTEX)
+ nir_remap_attributes(shader);
+
shader->info->name = ralloc_asprintf(shader, "GLSL%d", shader_prog->Name);
if (shader_prog->Label)
shader->info->label = ralloc_strdup(shader, shader_prog->Label);
@@ -322,6 +342,14 @@ nir_visitor::visit(ir_variable *ir)
var->data.compact = ir->type->without_array()->is_scalar();
}
}
+
+ /* Mark all the locations that require two slots */
+ if (glsl_type_is_dual_slot(glsl_without_array(var->type))) {
+ for (uint i = 0; i < glsl_count_attribute_slots(var->type, true); i++) {
+ uint64_t bitfield = BITFIELD64_BIT(var->data.location + i);
+ shader->info->double_inputs_read |= bitfield;
+ }
+ }
break;
case ir_var_shader_out: