target/i386: optimize computation of JL and JLE from flags

Take advantage of the fact that there can be no 1 bits between SF and OF.
If they were adjacent, you could sum SF and get a carry only if SF was
already set.  Then the value of OF in the sum is the XOR of OF itself,
the carry (which is SF) and 0 (the value of the OF bit in the addend):
this is OF^SF exactly.

Because OF and SF are not adjacent, just place more 1 bits to the
left so that the carry propagates, which means summing CC_O - CC_S.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c
index 037bc47e7c27ca01e955aa92e9c3e660cd9a9a77..8fb80011a22aee509d93896e9a76445eba53786f 100644 (file)
--- a/target/i386/tcg/translate.c
+++ b/target/i386/tcg/translate.c
@@ -1126,10 +1126,9 @@ static CCPrepare gen_prepare_cc(DisasContext *s, int b, TCGv reg)
             if (reg == cpu_cc_src) {
                 reg = s->tmp0;
             }
-            tcg_gen_shri_tl(reg, cpu_cc_src, 4); /* CC_O -> CC_S */
-            tcg_gen_xor_tl(reg, reg, cpu_cc_src);
+            tcg_gen_addi_tl(reg, cpu_cc_src, CC_O - CC_S);
             cc = (CCPrepare) { .cond = TCG_COND_NE, .reg = reg,
-                               .mask = CC_S };
+                               .mask = CC_O };
             break;
         default:
         case JCC_LE:
@@ -1137,10 +1136,9 @@ static CCPrepare gen_prepare_cc(DisasContext *s, int b, TCGv reg)
             if (reg == cpu_cc_src) {
                 reg = s->tmp0;
             }
-            tcg_gen_shri_tl(reg, cpu_cc_src, 4); /* CC_O -> CC_S */
-            tcg_gen_xor_tl(reg, reg, cpu_cc_src);
+            tcg_gen_addi_tl(reg, cpu_cc_src, CC_O - CC_S);
             cc = (CCPrepare) { .cond = TCG_COND_NE, .reg = reg,
-                               .mask = CC_S | CC_Z };
+                               .mask = CC_O | CC_Z };
             break;
         }
         break;