From: Richard Henderson <rth@twiddle.net>
Date: Thu, 3 Jul 2014 19:37:59 +0000 (-0700)
Subject: target-alpha: Fix cvttq vs inf
X-Git-Tag: TizenStudio_2.0_p2.3.2~208^2~189^2~8
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=7b4dde839e86ca6c254d4e3cd28260e9d668afb5;p=sdk%2Femulator%2Fqemu.git

target-alpha: Fix cvttq vs inf

We should raise INV for infinities as well, not OVR+INE.

Reported-by: Al Viro <viro@ZenIV.linux.org.uk>
Signed-off-by: Richard Henderson <rth@twiddle.net>
---

diff --git a/target-alpha/fpu_helper.c b/target-alpha/fpu_helper.c
index 9449c57..db523fb 100644
--- a/target-alpha/fpu_helper.c
+++ b/target-alpha/fpu_helper.c
@@ -444,7 +444,7 @@ static uint64_t do_cvttq(CPUAlphaState *env, uint64_t a, int roundmode)
             goto do_underflow;
         }
     } else if (exp == 0x7ff) {
-        exc = (frac ? FPCR_INV : FPCR_IOV | FPCR_INE);
+        exc = FPCR_INV;
     } else {
         /* Restore implicit bit.  */
         frac |= 0x10000000000000ull;