On 6/30/21 3:49 AM, Ulrich Weigand wrote:
I'm thinking that should be able to remove TCGv_i32 cc_op in the
translator and manually write back to the slot instead. We already
do a good job of caching the value within DisasContext -- I imagine
that the final code wouldn't even change too much.
I
On Mon, Jun 28, 2021 at 11:45:27AM -0700, Richard Henderson wrote:
> On 6/28/21 9:35 AM, Ulrich Weigand wrote:
> >@@ -506,6 +534,7 @@ uint64_t HELPER(cgeb)(CPUS390XState *env, uint64_t v2,
> >uint32_t m34)
> > {
> > int old_mode = s390_swap_bfp_rounding_mode(env, round_from_m34(m34));
> >
On 6/28/21 9:35 AM, Ulrich Weigand wrote:
@@ -506,6 +534,7 @@ uint64_t HELPER(cgeb)(CPUS390XState *env, uint64_t v2,
uint32_t m34)
{
int old_mode = s390_swap_bfp_rounding_mode(env, round_from_m34(m34));
int64_t ret = float32_to_int64(v2, &env->fpu_status);
+env->cc_op = set_cc_
The FP-to-integer conversion instructions need to set CC 3 whenever
a "special case" occurs; this is the case whenever the instruction
also signals the IEEE invalid exception. (See e.g. figure 19-18
in the Principles of Operation.)
However, qemu currently will set CC 3 only in the case where the
