aco/optimizer_postRA: don't combine DPP across exec on GFX8/9

GFX8/9 seem to use FI=0 behaviour.

fossil-db (vega10):
Totals from 1 (0.00% of 63053) affected shaders:
Instrs: 542 -> 570 (+5.17%)
CodeSize: 2928 -> 3040 (+3.83%)
Latency: 2087 -> 2118 (+1.49%)
InvThroughput: 1103 -> 1143 (+3.63%)

Affected shader is from Cyberpunk 2077 fossil.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Cc: 23.2 <mesa-stable>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9784
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25471>

src/amd/compiler/aco_optimizer_postRA.cpp

@@ -507,6 +507,11 @@ try_combine_dpp(pr_opt_ctx& ctx, aco_ptr<Instruction>& instr)
       if (is_overwritten_since(ctx, mov->operands[0], op_instr_idx))
          continue;
 
+      /* GFX8/9 don't have fetch-inactive. */
+      if (ctx.program->gfx_level < GFX10 &&
+          is_overwritten_since(ctx, Operand(exec, ctx.program->lane_mask), op_instr_idx))
+         continue;
+
       /* We won't eliminate the DPP mov if the operand is used twice */
       bool op_used_twice = false;
       for (unsigned j = 0; j < instr->operands.size(); j++)