From: Julian Seward <jseward@acm.org>
Date: Sat, 5 May 2018 07:49:02 +0000 (+0200)
Subject: Comment out test "(INSNA(6,6)  == 0 || INSNA(6,6) == 1)" that gcc 8 complains (correc... 
X-Git-Tag: VALGRIND_3_14_0~119
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=98deefdfea5f5dc3ec2b6b54f7e5b27ca0a60bb2;p=thirdparty%2Fvalgrind.git

Comment out test "(INSNA(6,6)  == 0 || INSNA(6,6) == 1)" that gcc 8 complains (correctly) is always true.  No functional change.
---

diff --git a/VEX/priv/guest_arm_toIR.c b/VEX/priv/guest_arm_toIR.c
index 26c5174647..d858c85e0b 100644
--- a/VEX/priv/guest_arm_toIR.c
+++ b/VEX/priv/guest_arm_toIR.c
@@ -9145,8 +9145,11 @@ static Bool decode_V6MEDIA_instruction (
         }
      } else {
         if (INSNA(27,20) == BITS8(0,1,1,0,1,0,0,0) &&
-            INSNA(5,4)   == BITS2(0,1)             &&
-            (INSNA(6,6)  == 0 || INSNA(6,6) == 1) ) {
+            INSNA(5,4)   == BITS2(0,1) /*          &&
+            (INSNA(6,6)  == 0 || INSNA(6,6) == 1)
+            This last bit with INSNA(6,6) is correct, but gcc 8 complains
+            (correctly) that it is always true.  So I commented it out
+            to keep gcc quiet. */ ) {
            regD = INSNA(15,12);
            regN = INSNA(19,16);
            regM = INSNA(3,0);