From: Paolo Bonzini <pbonzini@redhat.com>
Date: Wed, 25 Oct 2023 11:13:13 +0000 (+0200)
Subject: target/i386: speedup JO/SETO after MUL or IMUL
X-Git-Tag: qemu-xen-4.20.0~357^2~43
X-Git-Url: http://xenbits.xensource.com/gitweb?a=commitdiff_plain;h=1e7dde8008e4973faed3c6a0d7a909cc35cbdf74;p=qemu-xen.git

target/i386: speedup JO/SETO after MUL or IMUL

OF is equal to the carry flag, so use the same CCPrepare.

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 8fb80011a2..a16eb8d400 100644
--- a/target/i386/tcg/translate.c
+++ b/target/i386/tcg/translate.c
@@ -1020,6 +1020,9 @@ static CCPrepare gen_prepare_eflags_o(DisasContext *s, TCGv reg)
     case CC_OP_CLR:
     case CC_OP_POPCNT:
         return (CCPrepare) { .cond = TCG_COND_NEVER, .mask = -1 };
+    case CC_OP_MULB ... CC_OP_MULQ:
+        return (CCPrepare) { .cond = TCG_COND_NE,
+                             .reg = cpu_cc_src, .mask = -1 };
     default:
         gen_compute_eflags(s);
         return (CCPrepare) { .cond = TCG_COND_NE, .reg = cpu_cc_src,