[backport proposed 4.8 improvement to __cxa_guard_acquire <mikpe@it.uu.se>]

From: Thiago Macieira <thiago dot macieira at intel dot com>
Subject: [PATCH, libstdc++] Improve slightly __cxa_guard_acquire
Date: Thu, 30 Aug 2012 12:48:34 +0200
List-Id: <gcc-patches.gcc.gnu.org>

Hello

The attached patch is a simple improvement to make a thread that failed to set
the waiting bit to exit the function earlier, if it detects that another
thread has successfully finished initialising. It matches the CAS code from a
few lines above.

The change from RELAXED to ACQUIRE is noted in the previous patch I've just
sent.

2012-08-30  Thiago Macieira <thiago.macieira@intel.com>

	* libsupc++/guard.cc (__cxa_guard_acquire): exit the loop earlier if
	we detect that another thread has had success.

--- gcc-4.7.1/libstdc++-v3/libsupc++/guard.cc.~1~	2012-09-08 22:16:41.000000000 +0200
+++ gcc-4.7.1/libstdc++-v3/libsupc++/guard.cc	2012-09-08 22:19:01.000000000 +0200
@@ -269,8 +269,16 @@ namespace __cxxabiv1 
 		 int newv = expected | waiting_bit;
 		 if (!__atomic_compare_exchange_n(gi, &expected, newv, false,
 						  __ATOMIC_ACQ_REL, 
-						  __ATOMIC_RELAXED))
-		   continue;
+						  __ATOMIC_ACQUIRE))
+		   {
+		     if (expected == guard_bit)
+		       {
+			 // Already initialized.
+			 return 0;
+		       }
+		     if (expected == 0)
+		       continue;
+		   }
 		 
 		 expected = newv;
 	       }