Use compiler intrinsic instead of assembly for popcnt

lucasart · lucasart · commit 85b08ce3adaf · 2014-07-03T18:22:53.000+08:00
This time, do not break compatibility with some AMD machines that have SSE3 and popcnt, but do
not have SSE4.2.

No functional change.
diff --git a/src/Makefile b/src/Makefile
@@ -285,7 +285,7 @@ endif
 
 ### 3.9 popcnt
 ifeq ($(popcnt),yes)
-	CXXFLAGS += -msse3 -DUSE_POPCNT
+	CXXFLAGS += -msse3 -mpopcnt -DUSE_POPCNT
 endif
 
 ### 3.10 pext
diff --git a/src/bitcount.h b/src/bitcount.h
@@ -96,8 +96,7 @@ inline int popcount<CNT_HW_POPCNT>(Bitboard b) {
 
 #else
 
-  __asm__("popcnt %1, %0" : "=r" (b) : "r" (b));
-  return b;
+  return __builtin_popcountll(b);
 
 #endif
 }

Original file line number	Diff line number	Diff line change
`@@ -96,8 +96,7 @@ inline int popcount<CNT_HW_POPCNT>(Bitboard b) {`
`96`	`96`
`97`	`97`	`#else`
`98`	`98`
`99`		`- __asm__("popcnt %1, %0" : "=r" (b) : "r" (b));`
`100`		`- return b;`
	`99`	`+ return __builtin_popcountll(b);`
`101`	`100`
`102`	`101`	`#endif`
`103`	`102`	`}`