OpenHarmony/third_party_mesa3d
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

i965/gen7+: Use NIR for lowering of pack/unpack opcodes.

Browse Source 
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
This commit is contained in:
Matt Turner
2016-01-25 11:07:28 -08:00
parent f4952421cd
commit 75c9def8ee
3 changed files with 29 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
src/mesa/drivers/dri/i965/brw_compiler.c
View File

@@ -87,7 +87,15 @@ shader_perf_log_mesa(void *data, const char *fmt, ...)
static const struct nir_shader_compiler_options scalar_nir_options = { static const struct nir_shader_compiler_options scalar_nir_options = {
COMMON_OPTIONS, COMMON_OPTIONS,
.lower_pack_half_2x16 = true, .lower_pack_half_2x16 = true,
.lower_pack_snorm_2x16 = true,
.lower_pack_snorm_4x8 = true,
.lower_pack_unorm_2x16 = true,
.lower_pack_unorm_4x8 = true,
.lower_unpack_half_2x16 = true, .lower_unpack_half_2x16 = true,
.lower_unpack_snorm_2x16 = true,
.lower_unpack_snorm_4x8 = true,
.lower_unpack_unorm_2x16 = true,
.lower_unpack_unorm_4x8 = true,
}; };
static const struct nir_shader_compiler_options vector_nir_options = { static const struct nir_shader_compiler_options vector_nir_options = {
@@ -98,6 +106,13 @@ static const struct nir_shader_compiler_options vector_nir_options = {
* instructions because it can optimize better for us. * instructions because it can optimize better for us.
*/ */
.fdot_replicates = true, .fdot_replicates = true,
.lower_pack_snorm_2x16 = true,
.lower_pack_unorm_2x16 = true,
.lower_unpack_snorm_2x16 = true,
.lower_unpack_unorm_2x16 = true,
.lower_extract_byte = true,
.lower_extract_word = true,
}; };
struct brw_compiler * struct brw_compiler *

8
src/mesa/drivers/dri/i965/brw_fs_channel_expressions.cpp
View File

@@ -73,6 +73,10 @@ channel_expressions_predicate(ir_instruction *ir)
switch (expr->operation) { switch (expr->operation) {
case ir_unop_pack_half_2x16: case ir_unop_pack_half_2x16:
case ir_unop_pack_snorm_2x16:
case ir_unop_pack_snorm_4x8:
case ir_unop_pack_unorm_2x16:
case ir_unop_pack_unorm_4x8:
return false; return false;
/* these opcodes need to act on the whole vector, /* these opcodes need to act on the whole vector,
@@ -166,6 +170,10 @@ ir_channel_expressions_visitor::visit_leave(ir_assignment *ir)
switch (expr->operation) { switch (expr->operation) {
case ir_unop_pack_half_2x16: case ir_unop_pack_half_2x16:
case ir_unop_pack_snorm_2x16:
case ir_unop_pack_snorm_4x8:
case ir_unop_pack_unorm_2x16:
case ir_unop_pack_unorm_4x8:
case ir_unop_interpolate_at_centroid: case ir_unop_interpolate_at_centroid:
case ir_binop_interpolate_at_offset: case ir_binop_interpolate_at_offset:
case ir_binop_interpolate_at_sample: case ir_binop_interpolate_at_sample:

25
src/mesa/drivers/dri/i965/brw_link.cpp
View File

@@ -73,26 +73,13 @@ brw_lower_packing_builtins(struct brw_context *brw,
gl_shader_stage shader_type, gl_shader_stage shader_type,
exec_list *ir) exec_list *ir)
{ {
const struct brw_compiler *compiler = brw->intelScreen->compiler; /* Gens < 7 don't have instructions to convert to or from half-precision,
* and Gens < 6 don't expose that functionality.
*/
if (brw->gen != 6)
return;
int ops = LOWER_PACK_SNORM_2x16 lower_packing_builtins(ir, LOWER_PACK_HALF_2x16 | LOWER_UNPACK_HALF_2x16);
| LOWER_UNPACK_SNORM_2x16
| LOWER_PACK_UNORM_2x16
| LOWER_UNPACK_UNORM_2x16;
if (compiler->scalar_stage[shader_type]) {
ops |= LOWER_UNPACK_UNORM_4x8
| LOWER_UNPACK_SNORM_4x8
| LOWER_PACK_UNORM_4x8
| LOWER_PACK_SNORM_4x8;
}
if (brw->gen < 7) {
ops |= LOWER_PACK_HALF_2x16
| LOWER_UNPACK_HALF_2x16;
}
lower_packing_builtins(ir, ops);
} }
static void static void