Re: [PATCH 15/27] target/s390x: Use tcg_gen_qemu_{ld,st}_i128 for LPQ, STPQ

[Richard Henderson]

On 5/22/23 01:35, David Hildenbrand wrote:

@@ -4533,14 +4522,11 @@ static DisasJumpType op_stmh(DisasContext *s, DisasOps 
*o)
  static DisasJumpType op_stpq(DisasContext *s, DisasOps *o)
  {
-    if (!(tb_cflags(s->base.tb) & CF_PARALLEL)) {
-        gen_helper_stpq(cpu_env, o->in2, o->out2, o->out);
-    } else if (HAVE_ATOMIC128) {
-        gen_helper_stpq_parallel(cpu_env, o->in2, o->out2, o->out);
-    } else {
-        gen_helper_exit_atomic(cpu_env);
-        return DISAS_NORETURN;
-    }
+    TCGv_i128 t16 = tcg_temp_new_i128();
+
+    tcg_gen_concat_i64_i128(t16, o->out2, o->out);
+    tcg_gen_qemu_st_i128(t16, o->in2, get_mem_index(s),
+                         MO_TE | MO_128 | MO_ALIGN);

I briefly glimpsed at tcg_gen_qemu_ld_i128_int (and use_two_i64_for_i128()), does this really provide the atomic guarantees we need in all cases?

Yes.  The CF_PARALLEL check is the same as the one removed above.


r~