Merge branch 'master' into bug13658-branch

4 files changed, 418 insertions, 77 deletions

diff --git a/sysdeps/x86_64/fpu/e_expf.S b/sysdeps/x86_64/fpu/e_expf.S
index f1ce2851b0..340a1c2f07 100644
--- a/sysdeps/x86_64/fpu/e_expf.S
+++ b/sysdeps/x86_64/fpu/e_expf.S
@@ -14,9 +14,8 @@
    Lesser General Public License for more details.
 
    You should have received a copy of the GNU Lesser General Public
-   License along with the GNU C Library; if not, write to the Free
-   Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
-   02111-1307 USA.  */
+   License along with the GNU C Library; if not, see
+   <http://www.gnu.org/licenses/>.  */
 
 #include <sysdep.h>
 
diff --git a/sysdeps/x86_64/fpu/libm-test-ulps b/sysdeps/x86_64/fpu/libm-test-ulps
index 269dca6d03..33efe9df51 100644
--- a/sysdeps/x86_64/fpu/libm-test-ulps
+++ b/sysdeps/x86_64/fpu/libm-test-ulps
@@ -41,13 +41,151 @@ ildouble: 1
 ldouble: 1
 
 # cacos
+Test "Imaginary part of: cacos (+0 + 0.5 i) == pi/2 - 0.4812118250596034474977589134243684231352 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+Test "Imaginary part of: cacos (+0 + 1.0 i) == pi/2 - 0.8813735870195430252326093249797923090282 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+ildouble: 2
+ldouble: 2
+Test "Imaginary part of: cacos (+0 + 1.5 i) == pi/2 - 1.194763217287109304111930828519090523536 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+Test "Imaginary part of: cacos (+0 - 0.5 i) == pi/2 + 0.4812118250596034474977589134243684231352 i":
+float: 1
+ifloat: 1
+Test "Imaginary part of: cacos (+0 - 1.0 i) == pi/2 + 0.8813735870195430252326093249797923090282 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Imaginary part of: cacos (+0 - 1.5 i) == pi/2 + 1.194763217287109304111930828519090523536 i":
+double: 1
+idouble: 1
+Test "Imaginary part of: cacos (-0 + 0.5 i) == pi/2 - 0.4812118250596034474977589134243684231352 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+Test "Imaginary part of: cacos (-0 + 1.0 i) == pi/2 - 0.8813735870195430252326093249797923090282 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+ildouble: 2
+ldouble: 2
+Test "Imaginary part of: cacos (-0 + 1.5 i) == pi/2 - 1.194763217287109304111930828519090523536 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+Test "Imaginary part of: cacos (-0 - 0.5 i) == pi/2 + 0.4812118250596034474977589134243684231352 i":
+float: 1
+ifloat: 1
+Test "Imaginary part of: cacos (-0 - 1.0 i) == pi/2 + 0.8813735870195430252326093249797923090282 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Imaginary part of: cacos (-0 - 1.5 i) == pi/2 + 1.194763217287109304111930828519090523536 i":
+double: 1
+idouble: 1
+Test "Imaginary part of: cacos (-1.5 + +0 i) == pi - 0.9624236501192068949955178268487368462704 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Imaginary part of: cacos (-1.5 - 0 i) == pi + 0.9624236501192068949955178268487368462704 i":
+ildouble: 1
+ldouble: 1
+Test "Real part of: cacos (0.5 + +0 i) == 1.047197551196597746154214461093167628066 - 0 i":
+double: 1
+idouble: 1
+ildouble: 1
+ldouble: 1
+Test "Real part of: cacos (0.5 - 0 i) == 1.047197551196597746154214461093167628066 + +0 i":
+double: 1
+idouble: 1
+ildouble: 1
+ldouble: 1
 Test "Imaginary part of: cacos (0.75 + 1.25 i) == 1.11752014915610270578240049553777969 - 1.13239363160530819522266333696834467 i":
 float: 1
 ifloat: 1
 ildouble: 2
 ldouble: 2
+Test "Imaginary part of: cacos (1.5 + +0 i) == +0 - 0.9624236501192068949955178268487368462704 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Imaginary part of: cacos (1.5 - 0 i) == +0 + 0.9624236501192068949955178268487368462704 i":
+ildouble: 1
+ldouble: 1
 
 # cacosh
+Test "Real part of: cacosh (+0 + 0.5 i) == 0.4812118250596034474977589134243684231352 + pi/2 i":
+float: 1
+ifloat: 1
+Test "Real part of: cacosh (+0 + 1.0 i) == 0.8813735870195430252326093249797923090282 + pi/2 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Real part of: cacosh (+0 + 1.5 i) == 1.194763217287109304111930828519090523536 + pi/2 i":
+double: 1
+idouble: 1
+Test "Real part of: cacosh (+0 - 0.5 i) == 0.4812118250596034474977589134243684231352 - pi/2 i":
+float: 1
+ifloat: 1
+Test "Real part of: cacosh (+0 - 1.0 i) == 0.8813735870195430252326093249797923090282 - pi/2 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Real part of: cacosh (+0 - 1.5 i) == 1.194763217287109304111930828519090523536 - pi/2 i":
+double: 1
+idouble: 1
+Test "Real part of: cacosh (-0 + 0.5 i) == 0.4812118250596034474977589134243684231352 + pi/2 i":
+float: 1
+ifloat: 1
+Test "Real part of: cacosh (-0 + 1.0 i) == 0.8813735870195430252326093249797923090282 + pi/2 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Real part of: cacosh (-0 + 1.5 i) == 1.194763217287109304111930828519090523536 + pi/2 i":
+double: 1
+idouble: 1
+Test "Real part of: cacosh (-0 - 0.5 i) == 0.4812118250596034474977589134243684231352 - pi/2 i":
+float: 1
+ifloat: 1
+Test "Real part of: cacosh (-0 - 1.0 i) == 0.8813735870195430252326093249797923090282 - pi/2 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Real part of: cacosh (-0 - 1.5 i) == 1.194763217287109304111930828519090523536 - pi/2 i":
+double: 1
+idouble: 1
+Test "Imaginary part of: cacosh (-0.5 + +0 i) == +0 + 2.094395102393195492308428922186335256131 i":
+double: 1
+idouble: 1
+Test "Imaginary part of: cacosh (-0.5 - 0 i) == +0 - 2.094395102393195492308428922186335256131 i":
+double: 1
+idouble: 1
+Test "Real part of: cacosh (-1.5 + +0 i) == 0.9624236501192068949955178268487368462704 + pi i":
+float: 1
+ifloat: 1
+Test "Real part of: cacosh (-1.5 - 0 i) == 0.9624236501192068949955178268487368462704 - pi i":
+float: 1
+ifloat: 1
 Test "Real part of: cacosh (-2 - 3 i) == 1.9833870299165354323470769028940395 - 2.1414491111159960199416055713254211 i":
 double: 1
 float: 7
@@ -65,8 +203,78 @@ ldouble: 1
 Test "Real part of: cacosh (0.75 + 1.25 i) == 1.13239363160530819522266333696834467 + 1.11752014915610270578240049553777969 i":
 ildouble: 1
 ldouble: 1
+Test "Real part of: cacosh (1.5 + +0 i) == 0.9624236501192068949955178268487368462704 + +0 i":
+float: 1
+ifloat: 1
+Test "Real part of: cacosh (1.5 - 0 i) == 0.9624236501192068949955178268487368462704 - 0 i":
+float: 1
+ifloat: 1
 
 # casin
+Test "Imaginary part of: casin (+0 + 0.5 i) == +0 + 0.4812118250596034474977589134243684231352 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+Test "Imaginary part of: casin (+0 + 1.0 i) == +0 + 0.8813735870195430252326093249797923090282 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+ildouble: 2
+ldouble: 2
+Test "Imaginary part of: casin (+0 + 1.5 i) == +0 + 1.194763217287109304111930828519090523536 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+Test "Imaginary part of: casin (+0 - 0.5 i) == +0 - 0.4812118250596034474977589134243684231352 i":
+float: 1
+ifloat: 1
+Test "Imaginary part of: casin (+0 - 1.0 i) == +0 - 0.8813735870195430252326093249797923090282 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Imaginary part of: casin (+0 - 1.5 i) == +0 - 1.194763217287109304111930828519090523536 i":
+double: 1
+idouble: 1
+Test "Imaginary part of: casin (-0 + 0.5 i) == -0 + 0.4812118250596034474977589134243684231352 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+Test "Imaginary part of: casin (-0 + 1.0 i) == -0 + 0.8813735870195430252326093249797923090282 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+ildouble: 2
+ldouble: 2
+Test "Imaginary part of: casin (-0 + 1.5 i) == -0 + 1.194763217287109304111930828519090523536 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+Test "Imaginary part of: casin (-0 - 0.5 i) == -0 - 0.4812118250596034474977589134243684231352 i":
+float: 1
+ifloat: 1
+Test "Imaginary part of: casin (-0 - 1.0 i) == -0 - 0.8813735870195430252326093249797923090282 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Imaginary part of: casin (-0 - 1.5 i) == -0 - 1.194763217287109304111930828519090523536 i":
+double: 1
+idouble: 1
+Test "Imaginary part of: casin (-1.5 + +0 i) == -pi/2 + 0.9624236501192068949955178268487368462704 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Imaginary part of: casin (-1.5 - 0 i) == -pi/2 - 0.9624236501192068949955178268487368462704 i":
+ildouble: 1
+ldouble: 1
 Test "Real part of: casin (0.75 + 1.25 i) == 0.453276177638793913448921196101971749 + 1.13239363160530819522266333696834467 i":
 double: 1
 float: 1
@@ -79,8 +287,66 @@ float: 1
 ifloat: 1
 ildouble: 2
 ldouble: 2
+Test "Imaginary part of: casin (1.5 + +0 i) == pi/2 + 0.9624236501192068949955178268487368462704 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Imaginary part of: casin (1.5 - 0 i) == pi/2 - 0.9624236501192068949955178268487368462704 i":
+ildouble: 1
+ldouble: 1
 
 # casinh
+Test "Real part of: casinh (+0 + 1.5 i) == 0.9624236501192068949955178268487368462704 + pi/2 i":
+ildouble: 1
+ldouble: 1
+Test "Real part of: casinh (+0 - 1.5 i) == 0.9624236501192068949955178268487368462704 - pi/2 i":
+ildouble: 1
+ldouble: 1
+Test "Real part of: casinh (-0 + 1.5 i) == -0.9624236501192068949955178268487368462704 + pi/2 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Real part of: casinh (-0 - 1.5 i) == -0.9624236501192068949955178268487368462704 - pi/2 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Real part of: casinh (-0.5 + +0 i) == -0.4812118250596034474977589134243684231352 + +0 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+Test "Real part of: casinh (-0.5 - 0 i) == -0.4812118250596034474977589134243684231352 - 0 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+Test "Real part of: casinh (-1.0 + +0 i) == -0.8813735870195430252326093249797923090282 + +0 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+ildouble: 2
+ldouble: 2
+Test "Real part of: casinh (-1.0 - 0 i) == -0.8813735870195430252326093249797923090282 - 0 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+ildouble: 2
+ldouble: 2
+Test "Real part of: casinh (-1.5 + +0 i) == -1.194763217287109304111930828519090523536 + +0 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
+Test "Real part of: casinh (-1.5 - 0 i) == -1.194763217287109304111930828519090523536 - 0 i":
+double: 2
+float: 1
+idouble: 2
+ifloat: 1
 Test "Real part of: casinh (-2 - 3 i) == -1.9686379257930962917886650952454982 - 0.96465850440760279204541105949953237 i":
 double: 5
 float: 1
@@ -95,6 +361,12 @@ idouble: 3
 ifloat: 6
 ildouble: 5
 ldouble: 5
+Test "Real part of: casinh (0.5 + +0 i) == 0.4812118250596034474977589134243684231352 + +0 i":
+float: 1
+ifloat: 1
+Test "Real part of: casinh (0.5 - 0 i) == 0.4812118250596034474977589134243684231352 - 0 i":
+float: 1
+ifloat: 1
 Test "Real part of: casinh (0.75 + 1.25 i) == 1.03171853444778027336364058631006594 + 0.911738290968487636358489564316731207 i":
 float: 1
 ifloat: 1
@@ -105,6 +377,22 @@ idouble: 1
 ifloat: 1
 ildouble: 1
 ldouble: 1
+Test "Real part of: casinh (1.0 + +0 i) == 0.8813735870195430252326093249797923090282 + +0 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Real part of: casinh (1.0 - 0 i) == 0.8813735870195430252326093249797923090282 - 0 i":
+double: 1
+float: 1
+idouble: 1
+ifloat: 1
+Test "Real part of: casinh (1.5 + +0 i) == 1.194763217287109304111930828519090523536 + +0 i":
+double: 1
+idouble: 1
+Test "Real part of: casinh (1.5 - 0 i) == 1.194763217287109304111930828519090523536 - 0 i":
+double: 1
+idouble: 1
 
 # catan
 Test "Real part of: catan (-2 - 3 i) == -1.4099210495965755225306193844604208 - 0.22907268296853876629588180294200276 i":
@@ -560,6 +848,35 @@ ifloat: 1
 Test "Real part of: csqrt (-2 - 3 i) == 0.89597747612983812471573375529004348 - 1.6741492280355400404480393008490519 i":
 float: 1
 ifloat: 1
+Test "Imaginary part of: csqrt (0x1.fffffep+127 + 1.0 i) == 1.844674352395372953599975585936590505260e+19 + 2.710505511993121390769065968615872097053e-20 i":
+float: 1
+ifloat: 1
+Test "Real part of: csqrt (0x1.fffffffffffffp+1023 + 0x1.fffffffffffffp+1023 i) == 1.473094556905565378990473658199034571917e+154 + 6.101757441282702188537080005372547713595e+153 i":
+double: 1
+idouble: 1
+Test "Imaginary part of: csqrt (0x1.fffffffffffffp+1023 + 0x1.fffffffffffffp+1023 i) == 1.473094556905565378990473658199034571917e+154 + 6.101757441282702188537080005372547713595e+153 i":
+double: 1
+idouble: 1
+Test "Imaginary part of: csqrt (0x1.fffffffffffffp+1023 + 0x1p+1023 i) == 1.379778091031440685006200821918878702861e+154 + 3.257214233483129514781233066898042490248e+153 i":
+double: 1
+idouble: 1
+ildouble: 1
+ldouble: 1
+Test "Imaginary part of: csqrt (0x1.fp+16383 + 0x1.fp+16383 i) == 1.179514222452201722651836720466795901016e+2466 + 4.885707879516577666702435054303191575148e+2465 i":
+ildouble: 1
+ldouble: 1
+Test "Imaginary part of: csqrt (0x1p-1073 + 0x1p-1073 i) == 3.453664695497464982856905711457966660085e-162 + 1.430554756764195530630723976279903095110e-162 i":
+ildouble: 1
+ldouble: 1
+Test "Imaginary part of: csqrt (0x1p-1074 + 0x1p-1074 i) == 2.442109726130830256743814843868934877597e-162 + 1.011554969366634726113090867589031782487e-162 i":
+ildouble: 1
+ldouble: 1
+Test "Imaginary part of: csqrt (0x1p-147 + 0x1p-147 i) == 8.225610928685557596194006925540350401606e-23 + 3.407159605465907500737319471202779419102e-23 i":
+ildouble: 1
+ldouble: 1
+Test "Imaginary part of: csqrt (0x1p-149 + 0x1p-149 i) == 4.112805464342778798097003462770175200803e-23 + 1.703579802732953750368659735601389709551e-23 i":
+ildouble: 1
+ldouble: 1
 
 # ctan
 Test "Real part of: ctan (-2 - 3 i) == 0.376402564150424829275122113032269084e-2 - 1.00323862735360980144635859782192726 i":
@@ -1498,8 +1815,16 @@ ifloat: 1
 ildouble: 1
 ldouble: 1
 
+Function: Real part of "cacos":
+double: 1
+idouble: 1
+ildouble: 1
+ldouble: 1
+
 Function: Imaginary part of "cacos":
+double: 2
 float: 1
+idouble: 2
 ifloat: 1
 ildouble: 2
 ldouble: 2
@@ -1529,7 +1854,9 @@ ildouble: 2
 ldouble: 2
 
 Function: Imaginary part of "casin":
+double: 2
 float: 1
+idouble: 2
 ifloat: 1
 ildouble: 2
 ldouble: 2
@@ -1735,8 +2062,18 @@ ildouble: 2
 ldouble: 2
 
 Function: Real part of "csqrt":
+double: 1
 float: 1
+idouble: 1
+ifloat: 1
+
+Function: Imaginary part of "csqrt":
+double: 1
+float: 1
+idouble: 1
 ifloat: 1
+ildouble: 1
+ldouble: 1
 
 Function: Real part of "ctan":
 double: 1
diff --git a/sysdeps/x86_64/fpu/math_private.h b/sysdeps/x86_64/fpu/math_private.h
index 8e79718262..c8616f654a 100644
--- a/sysdeps/x86_64/fpu/math_private.h
+++ b/sysdeps/x86_64/fpu/math_private.h
@@ -1,4 +1,5 @@
-#ifndef _MATH_PRIVATE_H
+#ifndef X86_64_MATH_PRIVATE_H
+#define X86_64_MATH_PRIVATE_H 1
 
 #define math_opt_barrier(x) \
   ({ __typeof(x) __x;							      \
@@ -15,7 +16,7 @@
       __asm __volatile ("" : : "f" (x));				      \
   } while (0)
 
-#include <math/math_private.h>
+#include_next <math_private.h>
 
 /* We can do a few things better on x86-64.  */
 
@@ -67,7 +68,6 @@
     f = f__;								      \
   } while (0)
 
-#endif
 
 #define __isnan(d) \
   ({ long int __di; EXTRACT_WORDS64 (__di, (double) (d));		      \
@@ -90,85 +90,87 @@
   ({ int __di; GET_FLOAT_WORD (__di, (float) d);			      \
      (__di & 0x7fffffff) < 0x7f800000; })
 
+extern __always_inline double
+__ieee754_sqrt (double d)
+{
+  double res;
 #if defined __AVX__ || defined SSE2AVX
-# define __ieee754_sqrt(d) \
-  ({ double __res;							      \
-    asm ("vsqrtsd %1, %0, %0" : "=x" (__res) : "xm" ((double) (d)));	      \
-     __res; })
-# define __ieee754_sqrtf(d) \
-  ({ float __res;							      \
-    asm ("vsqrtss %1, %0, %0" : "=x" (__res) : "xm" ((float) (d)));	      \
-     __res; })
+  asm ("vsqrtsd %1, %0, %0" : "=x" (res) : "xm" (d));
 #else
-# define __ieee754_sqrt(d) \
-  ({ double __res;							      \
-    asm ("sqrtsd %1, %0" : "=x" (__res) : "xm" ((double) (d)));		      \
-     __res; })
-# define __ieee754_sqrtf(d) \
-  ({ float __res;							      \
-    asm ("sqrtss %1, %0" : "=x" (__res) : "xm" ((float) (d)));		      \
-     __res; })
+  asm ("sqrtsd %1, %0" : "=x" (res) : "xm" (d));
 #endif
-#define __ieee754_sqrtl(d) \
-  ({ long double __res;							      \
-    asm ("fsqrt" : "=t" (__res) : "0" ((long double) (d)));		      \
-     __res; })
+  return res;
+}
+
+extern __always_inline float
+__ieee754_sqrtf (float d)
+{
+  float res;
+#if defined __AVX__ || defined SSE2AVX
+  asm ("vsqrtss %1, %0, %0" : "=x" (res) : "xm" (d));
+#else
+  asm ("sqrtss %1, %0" : "=x" (res) : "xm" (d));
+#endif
+  return res;
+}
+
+extern __always_inline long double
+__ieee754_sqrtl (long double d)
+{
+  long double res;
+  asm ("fsqrt" : "=t" (res) : "0" (d));
+  return res;
+}
 
 #ifdef __SSE4_1__
-# ifndef __rint
-#  if defined __AVX__ || defined SSE2AVX
-#   define __rint(d) \
-  ({ double __res; \
-    asm ("vroundsd $4, %1, %0, %0" : "=x" (__res) : "xm" ((double) (d)));      \
-     __res; })
-#  else
-#   define __rint(d) \
-  ({ double __res; \
-    asm ("roundsd $4, %1, %0" : "=x" (__res) : "xm" ((double) (d)));	      \
-     __res; })
-#  endif
+extern __always_inline double
+__rint (double d)
+{
+  double res;
+# if defined __AVX__ || defined SSE2AVX
+  asm ("vroundsd $4, %1, %0, %0" : "=x" (res) : "xm" (d));
+# else
+  asm ("roundsd $4, %1, %0" : "=x" (res) : "xm" (d));
 # endif
-# ifndef __rintf
-#  if defined __AVX__ || defined SSE2AVX
-#   define __rintf(d) \
-  ({ float __res; \
-    asm ("vroundss $4, %1, %0, %0" : "=x" (__res) : "xm" ((float) (d)));      \
-     __res; })
-#  else
-#   define __rintf(d) \
-  ({ float __res; \
-    asm ("roundss $4, %1, %0" : "=x" (__res) : "xm" ((float) (d)));	      \
-     __res; })
-#  endif
+  return res;
+}
+
+extern __always_inline float
+__rintf (float d)
+{
+  float res;
+# if defined __AVX__ || defined SSE2AVX
+  asm ("vroundss $4, %1, %0, %0" : "=x" (res) : "xm" (d));
+# else
+  asm ("roundss $4, %1, %0" : "=x" (res) : "xm" (d));
 # endif
-
-# ifndef __floor
-#  if defined __AVX__ || defined SSE2AVX
-#   define __floor(d) \
-  ({ double __res; \
-    asm ("vroundsd $1, %1, %0, %0" : "=x" (__res) : "xm" ((double) (d)));      \
-     __res; })
-#  else
-#   define __floor(d) \
-  ({ double __res; \
-    asm ("roundsd $1, %1, %0" : "=x" (__res) : "xm" ((double) (d)));	      \
-     __res; })
-#  endif
+  return res;
+}
+
+extern __always_inline double
+__floor (double d)
+{
+  double res;
+# if defined __AVX__ || defined SSE2AVX
+  asm ("vroundsd $1, %1, %0, %0" : "=x" (res) : "xm" (d));
+# else
+  asm ("roundsd $1, %1, %0" : "=x" (res) : "xm" (d));
 # endif
-# ifndef __floorf
-#  if defined __AVX__ || defined SSE2AVX
-#   define __floorf(d) \
-  ({ float __res; \
-    asm ("vroundss $1, %1, %0, %0" : "=x" (__res) : "xm" ((float) (d)));      \
-     __res; })
-#  else
-#   define __floorf(d) \
-  ({ float __res; \
-    asm ("roundss $1, %1, %0" : "=x" (__res) : "xm" ((float) (d)));	      \
-     __res; })
+  return res;
+}
+
+extern __always_inline float
+__floorf (float d)
+{
+  float res;
+# if defined __AVX__ || defined SSE2AVX
+  asm ("vroundss $1, %1, %0, %0" : "=x" (res) : "xm" (d));
+# else
+  asm ("roundss $1, %1, %0" : "=x" (res) : "xm" (d));
 #  endif
-# endif
-#endif
+  return res;
+}
+#endif /* __SSE4_1__ */
 
 
 /* Specialized variants of the <fenv.h> interfaces which only handle
@@ -226,3 +228,5 @@
 #undef libc_feupdateenvf
 #define libc_feupdateenvf(e) libc_feupdateenv (e)
 // #define libc_feupdateenvl(e) (void) feupdateenv (e)
+
+#endif /* X86_64_MATH_PRIVATE_H */
diff --git a/sysdeps/x86_64/fpu/printf_fphex.c b/sysdeps/x86_64/fpu/printf_fphex.c
index 0b6bfc6bdc..7de7283c4c 100644
--- a/sysdeps/x86_64/fpu/printf_fphex.c
+++ b/sysdeps/x86_64/fpu/printf_fphex.c
@@ -1,5 +1,5 @@
 /* Print floating point number in hexadecimal notation according to ISO C99.
-   Copyright (C) 1997,1998,1999,2000,2001,2005 Free Software Foundation, Inc.
+   Copyright (C) 1997-2012 Free Software Foundation, Inc.
    This file is part of the GNU C Library.
 
    The GNU C Library is free software; you can redistribute it and/or
@@ -57,6 +57,7 @@ do {									      \
 									      \
       /* We use a full nibble for the leading digit.  */		      \
       leading = *numstr++;						      \
+      wnumstr++;							      \
 									      \
       /* We have 3 bits from the mantissa in the leading nibble.	      \
 	 Therefore we are here using `IEEE854_LONG_DOUBLE_BIAS + 3'.  */      \