Author: das
Date: Sat Oct 15 04:24:54 2011
New Revision: 226374
URL: http://svn.freebsd.org/changeset/base/226374
Log:
Don't define FP_FAST_FMA on sparc64; with the recent fixes, fma() is
no longer "fast" on sparc64. (It really wasn't to begin with, since
the old implementation was using long doubles, and long doubles are
emulated in software on sparc64.)
Modified:
head/lib/msun/src/math.h
Modified: head/lib/msun/src/math.h
==============================================================================
--- head/lib/msun/src/math.h Sat Oct 15 04:22:55 2011 (r226373)
+++ head/lib/msun/src/math.h Sat Oct 15 04:24:54 2011 (r226374)
@@ -68,14 +68,11 @@ extern const union __nan_un {
#define MATH_ERREXCEPT 2
#define math_errhandling MATH_ERREXCEPT
-/* XXX We need a <machine/math.h>. */
-#if defined(__ia64__) || defined(__sparc64__)
-#define FP_FAST_FMA 1
-#endif
+#define FP_FAST_FMAF 1
#ifdef __ia64__
+#define FP_FAST_FMA 1
#define FP_FAST_FMAL 1
#endif
-#define FP_FAST_FMAF 1
/* Symbolic constants to classify floating point numbers. */
#define FP_INFINITE 0x01
_______________________________________________
[email protected] mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "[email protected]"
