aco: coalesce parallelcopies during register allocation

These are the result of lowering to CSSA, and should be removed if possible

Totals from affected shaders: (VEGA)
SGPRS: 544544 -> 544544 (0.00 %)
VGPRS: 418224 -> 418224 (0.00 %)
Spilled SGPRs: 141826 -> 141826 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 65853740 -> 64703380 (-1.75 %) bytes
LDS: 0 -> 0 (0.00 %) blocks
Max Waves: 13669 -> 13669 (0.00 %)

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4952>

diff --git a/src/amd/compiler/aco_register_allocation.cpp b/src/amd/compiler/aco_register_allocation.cpp
index 8d6971e1c4dc545c0b1eda5dafe539fcbd01e216..318e29b9cf33ae979bbae962c3b28f3fdc3fd4d6 100644 (file)
--- a/src/amd/compiler/aco_register_allocation.cpp
+++ b/src/amd/compiler/aco_register_allocation.cpp
@@ -1837,13 +1837,10 @@ void register_allocation(Program *program, std::vector<TempSet>& live_out_per_bl
                   reg.reg_b += instr->definitions[j].bytes();
                if (get_reg_specified(ctx, register_file, definition.regClass(), parallelcopy, instr, reg))
                   definition.setFixed(reg);
-            } else if (instr->opcode == aco_opcode::p_wqm) {
-               PhysReg reg;
-               if (instr->operands[0].isKillBeforeDef() && instr->operands[0].getTemp().type() == definition.getTemp().type()) {
-                  reg = instr->operands[0].physReg();
+            } else if (instr->opcode == aco_opcode::p_wqm || instr->opcode == aco_opcode::p_parallelcopy) {
+               PhysReg reg = instr->operands[i].physReg();
+               if (get_reg_specified(ctx, register_file, definition.regClass(), parallelcopy, instr, reg))
                   definition.setFixed(reg);
-                  assert(register_file[reg.reg()] == 0);
-               }
             } else if (instr->opcode == aco_opcode::p_extract_vector) {
                PhysReg reg;
                if (instr->operands[0].isKillBeforeDef() &&