bugfix: 修复浮点数打印错误的bug

kernel/common/math/fabs.c

@@ -0,0 +1,30 @@
+#include <common/math.h>
+#include <common/sys/types.h>
+#include "libm.h"
+
+double fabs(double x)
+{
+    union
+    {
+        double f;
+        uint64_t i;
+    } u = {x};
+    u.i &= -1ULL / 2;
+    return u.f;
+}
+
+
+#if __LDBL_MANT_DIG__ == 53 &&  __LDBL_MAX_EXP__ == 1024
+long double fabsl(long double x)
+{
+	return fabs(x);
+}
+#elif (__LDBL_MANT_DIG__ == 64 || __LDBL_MANT_DIG__ == 113) && __LDBL_MAX_EXP__ == 16384
+long double fabsl(long double x)
+{
+	union ldshape u = {x};
+
+	u.i.se &= 0x7fff;
+	return u.f;
+}
+#endif