diff --git a/src/amd/compiler/aco_optimizer.cpp b/src/amd/compiler/aco_optimizer.cpp
index 418f61bd61f..25914429f48 100644
--- a/src/amd/compiler/aco_optimizer.cpp
+++ b/src/amd/compiler/aco_optimizer.cpp
@@ -5089,8 +5089,10 @@ combine_instruction(opt_ctx& ctx, aco_ptr<Instruction>& instr)
          add_opt(v_mul_f32, v_mad_f32, 0x3, "120");
          add_opt(v_mul_legacy_f32, v_mad_legacy_f32, 0x3, "120");
       }
-      if (ctx.program->dev.has_fast_fma32)
+      if (ctx.program->dev.has_fast_fma32) {
          add_opt(v_mul_f32, v_fma_f32, 0x3, "120", create_fma_cb);
+         add_opt(s_mul_f32, v_fma_f32, 0x3, "120", create_fma_cb);
+      }
       if (ctx.program->gfx_level >= GFX10_3)
          add_opt(v_mul_legacy_f32, v_fma_legacy_f32, 0x3, "120", create_fma_cb);
    } else if (info.opcode == aco_opcode::v_add_f16) {
@@ -5099,14 +5101,20 @@ combine_instruction(opt_ctx& ctx, aco_ptr<Instruction>& instr)
       else if (ctx.program->gfx_level < GFX10 && ctx.fp_mode.denorm16_64 == 0)
          add_opt(v_mul_f16, v_mad_f16, 0x3, "120");
 
-      if (ctx.program->gfx_level < GFX9)
+      if (ctx.program->gfx_level < GFX9) {
          add_opt(v_mul_f16, v_fma_legacy_f16, 0x3, "120", create_fma_cb);
-      else
+      } else {
          add_opt(v_mul_f16, v_fma_f16, 0x3, "120", create_fma_cb);
+         add_opt(s_mul_f16, v_fma_f16, 0x3, "120", create_fma_cb);
+      }
    } else if (info.opcode == aco_opcode::v_add_f64) {
       add_opt(v_mul_f64, v_fma_f64, 0x3, "120", create_fma_cb);
    } else if (info.opcode == aco_opcode::v_add_f64_e64) {
       add_opt(v_mul_f64_e64, v_fma_f64, 0x3, "120", create_fma_cb);
+   } else if (info.opcode == aco_opcode::s_add_f32) {
+      add_opt(s_mul_f32, s_fmac_f32, 0x3, "120", create_fma_cb);
+   } else if (info.opcode == aco_opcode::s_add_f16) {
+      add_opt(s_mul_f16, s_fmac_f16, 0x3, "120", create_fma_cb);
    }
 
    if (match_and_apply_patterns(ctx, info, patterns)) {