about summary refs log tree commit diff
path: root/sysdeps/powerpc/fpu
diff options
context:
space:
mode:
authorJoseph Myers <joseph@codesourcery.com>2015-02-13 16:20:36 +0000
committerJoseph Myers <joseph@codesourcery.com>2015-02-13 16:20:36 +0000
commit95c26233a1c7dd73a852a4f3bffd2d2eeba5e2fd (patch)
treef6e64cd9b63a30a3c22cb4200515f99b376acbf1 /sysdeps/powerpc/fpu
parent3f293d614c9e641a0d96d347df5c1c5ee687762f (diff)
downloadglibc-95c26233a1c7dd73a852a4f3bffd2d2eeba5e2fd.tar.gz
glibc-95c26233a1c7dd73a852a4f3bffd2d2eeba5e2fd.tar.xz
glibc-95c26233a1c7dd73a852a4f3bffd2d2eeba5e2fd.zip
Fix powerpc software sqrtf (bug 17967).
Similarly to sqrt in
<https://sourceware.org/ml/libc-alpha/2015-02/msg00353.html>, the
powerpc sqrtf implementation for when _ARCH_PPCSQ is not defined also
relies on a * b + c being contracted into a fused multiply-add.
Although this contraction is not explicitly disabled for e_sqrtf.c, it
still seems appropriate to make the file explicit about its
requirements by using __builtin_fmaf; this patch does so.
Furthermore, it turns out that doing so fixes the observed inaccuracy
and missing exceptions (that is, that without explicit __builtin_fmaf
usage, it was not being compiled as intended).

Tested for powerpc32 (hard float).

	[BZ #17967]
	* sysdeps/powerpc/fpu/e_sqrtf.c (__slow_ieee754_sqrtf): Use
	__builtin_fmaf instead of relying on contraction of a * b + c.
Diffstat (limited to 'sysdeps/powerpc/fpu')
-rw-r--r--sysdeps/powerpc/fpu/e_sqrtf.c24
1 files changed, 13 insertions, 11 deletions
diff --git a/sysdeps/powerpc/fpu/e_sqrtf.c b/sysdeps/powerpc/fpu/e_sqrtf.c
index 034b6f58ca..a684cf977a 100644
--- a/sysdeps/powerpc/fpu/e_sqrtf.c
+++ b/sysdeps/powerpc/fpu/e_sqrtf.c
@@ -87,26 +87,28 @@ __slow_ieee754_sqrtf (float x)
 	  /* Here we have three Newton-Raphson iterations each of a
 	     division and a square root and the remainder of the
 	     argument reduction, all interleaved.   */
-	  sd = -(sg * sg - sx);
+	  sd = -__builtin_fmaf (sg, sg, -sx);
 	  fsgi = (xi + 0x40000000) >> 1 & 0x7f800000;
 	  sy2 = sy + sy;
-	  sg = sy * sd + sg;	/* 16-bit approximation to sqrt(sx). */
-	  e = -(sy * sg - almost_half);
+	  sg = __builtin_fmaf (sy, sd, sg);	/* 16-bit approximation to
+						   sqrt(sx). */
+	  e = -__builtin_fmaf (sy, sg, -almost_half);
 	  SET_FLOAT_WORD (fsg, fsgi);
-	  sd = -(sg * sg - sx);
-	  sy = sy + e * sy2;
+	  sd = -__builtin_fmaf (sg, sg, -sx);
+	  sy = __builtin_fmaf (e, sy2, sy);
 	  if ((xi & 0x7f800000) == 0)
 	    goto denorm;
 	  shx = sx * fsg;
-	  sg = sg + sy * sd;	/* 32-bit approximation to sqrt(sx),
-				   but perhaps rounded incorrectly.  */
+	  sg = __builtin_fmaf (sy, sd, sg);	/* 32-bit approximation to
+						   sqrt(sx), but perhaps
+						   rounded incorrectly.  */
 	  sy2 = sy + sy;
 	  g = sg * fsg;
-	  e = -(sy * sg - almost_half);
-	  d = -(g * sg - shx);
-	  sy = sy + e * sy2;
+	  e = -__builtin_fmaf (sy, sg, -almost_half);
+	  d = -__builtin_fmaf (g, sg, -shx);
+	  sy = __builtin_fmaf (e, sy2, sy);
 	  fesetenv_register (fe);
-	  return g + sy * d;
+	  return __builtin_fmaf (sy, d, g);
 	denorm:
 	  /* For denormalised numbers, we normalise, calculate the
 	     square root, and return an adjusted result.  */