diff --git a/newlib/libm/common/Makefile.am b/newlib/libm/common/Makefile.am
index 97b24315b..3e2620c7e 100644
--- a/newlib/libm/common/Makefile.am
+++ b/newlib/libm/common/Makefile.am
@@ -25,7 +25,7 @@ fsrc =	sf_finite.c sf_copysign.c sf_modf.c sf_scalbn.c \
 	sf_scalbln.c sf_trunc.c \
 	sf_exp.c sf_exp2.c sf_exp2_data.c sf_log.c sf_log_data.c \
 	sf_log2.c sf_log2_data.c sf_pow_log2_data.c sf_pow.c \
-	sinf.c cosf.c sincosf.c sincosf_data.c math_errf.c
+	math_errf.c
 
 lsrc =	atanl.c cosl.c sinl.c tanl.c tanhl.c frexpl.c modfl.c ceill.c fabsl.c \
 	floorl.c log1pl.c expm1l.c acosl.c asinl.c atan2l.c coshl.c sinhl.c \
@@ -38,7 +38,6 @@ lsrc =	atanl.c cosl.c sinl.c tanl.c tanhl.c frexpl.c modfl.c ceill.c fabsl.c \
 	sl_finite.c
 
 libcommon_la_LDFLAGS = -Xcompiler -nostdlib
-lib_a_CFLAGS = -fbuiltin -fno-math-errno
 
 if USE_LIBTOOL
 noinst_LTLIBRARIES = libcommon.la
@@ -53,7 +52,7 @@ lib_a_SOURCES = $(src) $(fsrc)
 if HAVE_LONG_DOUBLE
 lib_a_SOURCES += $(lsrc)
 endif # HAVE_LONG_DOUBLE
-lib_a_CFLAGS += $(AM_CFLAGS)
+lib_a_CFLAGS = $(AM_CFLAGS)
 noinst_DATA =
 endif # USE_LIBTOOL
 
diff --git a/newlib/libm/common/Makefile.in b/newlib/libm/common/Makefile.in
index c944ac790..6a0ba73d2 100644
--- a/newlib/libm/common/Makefile.in
+++ b/newlib/libm/common/Makefile.in
@@ -55,7 +55,6 @@ build_triplet = @build@
 host_triplet = @host@
 @HAVE_LONG_DOUBLE_TRUE@@USE_LIBTOOL_TRUE@am__append_1 = $(lsrc)
 @HAVE_LONG_DOUBLE_TRUE@@USE_LIBTOOL_FALSE@am__append_2 = $(lsrc)
-@USE_LIBTOOL_FALSE@am__append_3 = $(AM_CFLAGS)
 DIST_COMMON = $(srcdir)/../../Makefile.shared $(srcdir)/Makefile.in \
 	$(srcdir)/Makefile.am
 subdir = common
@@ -115,8 +114,6 @@ am__objects_2 = lib_a-sf_finite.$(OBJEXT) lib_a-sf_copysign.$(OBJEXT) \
 	lib_a-sf_log.$(OBJEXT) lib_a-sf_log_data.$(OBJEXT) \
 	lib_a-sf_log2.$(OBJEXT) lib_a-sf_log2_data.$(OBJEXT) \
 	lib_a-sf_pow_log2_data.$(OBJEXT) lib_a-sf_pow.$(OBJEXT) \
-	lib_a-sinf.$(OBJEXT) lib_a-cosf.$(OBJEXT) \
-	lib_a-sincosf.$(OBJEXT) lib_a-sincosf_data.$(OBJEXT) \
 	lib_a-math_errf.$(OBJEXT)
 am__objects_3 = lib_a-atanl.$(OBJEXT) lib_a-cosl.$(OBJEXT) \
 	lib_a-sinl.$(OBJEXT) lib_a-tanl.$(OBJEXT) \
@@ -172,8 +169,7 @@ am__objects_6 = sf_finite.lo sf_copysign.lo sf_modf.lo sf_scalbn.lo \
 	sf_nearbyint.lo sf_remquo.lo sf_round.lo sf_scalbln.lo \
 	sf_trunc.lo sf_exp.lo sf_exp2.lo sf_exp2_data.lo sf_log.lo \
 	sf_log_data.lo sf_log2.lo sf_log2_data.lo sf_pow_log2_data.lo \
-	sf_pow.lo sinf.lo cosf.lo sincosf.lo sincosf_data.lo \
-	math_errf.lo
+	sf_pow.lo math_errf.lo
 am__objects_7 = atanl.lo cosl.lo sinl.lo tanl.lo tanhl.lo frexpl.lo \
 	modfl.lo ceill.lo fabsl.lo floorl.lo log1pl.lo expm1l.lo \
 	acosl.lo asinl.lo atan2l.lo coshl.lo sinhl.lo expl.lo \
@@ -365,7 +361,7 @@ fsrc = sf_finite.c sf_copysign.c sf_modf.c sf_scalbn.c \
 	sf_scalbln.c sf_trunc.c \
 	sf_exp.c sf_exp2.c sf_exp2_data.c sf_log.c sf_log_data.c \
 	sf_log2.c sf_log2_data.c sf_pow_log2_data.c sf_pow.c \
-	sinf.c cosf.c sincosf.c sincosf_data.c math_errf.c
+	math_errf.c
 
 lsrc = atanl.c cosl.c sinl.c tanl.c tanhl.c frexpl.c modfl.c ceill.c fabsl.c \
 	floorl.c log1pl.c expm1l.c acosl.c asinl.c atan2l.c coshl.c sinhl.c \
@@ -378,7 +374,6 @@ lsrc = atanl.c cosl.c sinl.c tanl.c tanhl.c frexpl.c modfl.c ceill.c fabsl.c \
 	sl_finite.c
 
 libcommon_la_LDFLAGS = -Xcompiler -nostdlib
-lib_a_CFLAGS = -fbuiltin -fno-math-errno $(am__append_3)
 @USE_LIBTOOL_TRUE@noinst_LTLIBRARIES = libcommon.la
 @USE_LIBTOOL_TRUE@libcommon_la_SOURCES = $(src) $(fsrc) \
 @USE_LIBTOOL_TRUE@	$(am__append_1)
@@ -386,6 +381,7 @@ lib_a_CFLAGS = -fbuiltin -fno-math-errno $(am__append_3)
 @USE_LIBTOOL_TRUE@noinst_DATA = objectlist.awk.in
 @USE_LIBTOOL_FALSE@noinst_LIBRARIES = lib.a
 @USE_LIBTOOL_FALSE@lib_a_SOURCES = $(src) $(fsrc) $(am__append_2)
+@USE_LIBTOOL_FALSE@lib_a_CFLAGS = $(AM_CFLAGS)
 
 #
 # documentation rules
@@ -948,30 +944,6 @@ lib_a-sf_pow.o: sf_pow.c
 lib_a-sf_pow.obj: sf_pow.c
 	$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(lib_a_CFLAGS) $(CFLAGS) -c -o lib_a-sf_pow.obj `if test -f 'sf_pow.c'; then $(CYGPATH_W) 'sf_pow.c'; else $(CYGPATH_W) '$(srcdir)/sf_pow.c'; fi`
 
-lib_a-sinf.o: sinf.c
-	$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(lib_a_CFLAGS) $(CFLAGS) -c -o lib_a-sinf.o `test -f 'sinf.c' || echo '$(srcdir)/'`sinf.c
-
-lib_a-sinf.obj: sinf.c
-	$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(lib_a_CFLAGS) $(CFLAGS) -c -o lib_a-sinf.obj `if test -f 'sinf.c'; then $(CYGPATH_W) 'sinf.c'; else $(CYGPATH_W) '$(srcdir)/sinf.c'; fi`
-
-lib_a-cosf.o: cosf.c
-	$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(lib_a_CFLAGS) $(CFLAGS) -c -o lib_a-cosf.o `test -f 'cosf.c' || echo '$(srcdir)/'`cosf.c
-
-lib_a-cosf.obj: cosf.c
-	$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(lib_a_CFLAGS) $(CFLAGS) -c -o lib_a-cosf.obj `if test -f 'cosf.c'; then $(CYGPATH_W) 'cosf.c'; else $(CYGPATH_W) '$(srcdir)/cosf.c'; fi`
-
-lib_a-sincosf.o: sincosf.c
-	$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(lib_a_CFLAGS) $(CFLAGS) -c -o lib_a-sincosf.o `test -f 'sincosf.c' || echo '$(srcdir)/'`sincosf.c
-
-lib_a-sincosf.obj: sincosf.c
-	$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(lib_a_CFLAGS) $(CFLAGS) -c -o lib_a-sincosf.obj `if test -f 'sincosf.c'; then $(CYGPATH_W) 'sincosf.c'; else $(CYGPATH_W) '$(srcdir)/sincosf.c'; fi`
-
-lib_a-sincosf_data.o: sincosf_data.c
-	$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(lib_a_CFLAGS) $(CFLAGS) -c -o lib_a-sincosf_data.o `test -f 'sincosf_data.c' || echo '$(srcdir)/'`sincosf_data.c
-
-lib_a-sincosf_data.obj: sincosf_data.c
-	$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(lib_a_CFLAGS) $(CFLAGS) -c -o lib_a-sincosf_data.obj `if test -f 'sincosf_data.c'; then $(CYGPATH_W) 'sincosf_data.c'; else $(CYGPATH_W) '$(srcdir)/sincosf_data.c'; fi`
-
 lib_a-math_errf.o: math_errf.c
 	$(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(lib_a_CFLAGS) $(CFLAGS) -c -o lib_a-math_errf.o `test -f 'math_errf.c' || echo '$(srcdir)/'`math_errf.c
 
diff --git a/newlib/libm/common/cosf.c b/newlib/libm/common/cosf.c
deleted file mode 100644
index aac0a9aee..000000000
--- a/newlib/libm/common/cosf.c
+++ /dev/null
@@ -1,88 +0,0 @@
-/* Single-precision cos function.
-   Copyright (c) 2018 Arm Ltd.  All rights reserved.
-
-   Redistribution and use in source and binary forms, with or without
-   modification, are permitted provided that the following conditions
-   are met:
-   1. Redistributions of source code must retain the above copyright
-      notice, this list of conditions and the following disclaimer.
-   2. Redistributions in binary form must reproduce the above copyright
-      notice, this list of conditions and the following disclaimer in the
-      documentation and/or other materials provided with the distribution.
-   3. The name of the company may not be used to endorse or promote
-      products derived from this software without specific prior written
-      permission.
-
-   THIS SOFTWARE IS PROVIDED BY ARM LTD ``AS IS AND ANY EXPRESS OR IMPLIED
-   WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
-   MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
-   IN NO EVENT SHALL ARM LTD BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
-   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
-   TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
-   PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
-   LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
-   NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
-   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */
-
-#include "fdlibm.h"
-#if !__OBSOLETE_MATH
-
-#include <stdint.h>
-#include <math.h>
-#include "math_config.h"
-#include "sincosf.h"
-
-/* Fast cosf implementation.  Worst-case ULP is 0.56072, maximum relative
-   error is 0.5303p-23.  A single-step signed range reduction is used for
-   small values.  Large inputs have their range reduced using fast integer
-   arithmetic.
-*/
-float
-cosf (float y)
-{
-  double x = y;
-  double s;
-  int n;
-  sincos_t *p = &sincosf_table[0];
-
-  if (abstop12 (y) < abstop12 (pio4))
-    {
-      double x2 = x * x;
-
-      if (unlikely (abstop12 (y) < abstop12 (0x1p-12f)))
-	return 1.0f;
-
-      return sinf_poly (x, x2, p, 1);
-    }
-  else if (likely (abstop12 (y) < abstop12 (120.0f)))
-    {
-      x = reduce_fast (x, p, &n);
-
-      /* Setup the signs for sin and cos.  */
-      s = p->sign[n & 3];
-
-      if (n & 2)
-	p = &sincosf_table[1];
-
-      return sinf_poly (x * s, x * x, p, n ^ 1);
-    }
-  else if (abstop12 (y) < abstop12 (INFINITY))
-    {
-      uint32_t xi = asuint (y);
-      int sign = xi >> 31;
-
-      x = reduce_large (xi, &n);
-
-      /* Setup signs for sin and cos - include original sign.  */
-      s = p->sign[(n + sign) & 3];
-
-      if ((n + sign) & 2)
-	p = &sincosf_table[1];
-
-      return sinf_poly (x * s, x * x, p, n ^ 1);
-    }
-  else
-    return __math_invalidf (y);
-}
-
-#endif
diff --git a/newlib/libm/common/math_config.h b/newlib/libm/common/math_config.h
index 6d87376ab..c5667844a 100644
--- a/newlib/libm/common/math_config.h
+++ b/newlib/libm/common/math_config.h
@@ -1,5 +1,5 @@
 /* Configuration for math routines.
-   Copyright (c) 2017-2018 Arm Ltd.  All rights reserved.
+   Copyright (c) 2017 ARM Ltd.  All rights reserved.
 
    Redistribution and use in source and binary forms, with or without
    modification, are permitted provided that the following conditions
@@ -43,44 +43,6 @@
 # define WANT_ERRNO_UFLOW (WANT_ROUNDING && WANT_ERRNO)
 #endif
 
-/* Compiler can inline round as a single instruction.  */
-#ifndef HAVE_FAST_ROUND
-# if __aarch64__
-#   define HAVE_FAST_ROUND 1
-# else
-#   define HAVE_FAST_ROUND 0
-# endif
-#endif
-
-/* Compiler can inline lround, but not (long)round(x).  */
-#ifndef HAVE_FAST_LROUND
-# if __aarch64__ && (100*__GNUC__ + __GNUC_MINOR__) >= 408 && __NO_MATH_ERRNO__
-#   define HAVE_FAST_LROUND 1
-# else
-#   define HAVE_FAST_LROUND 0
-# endif
-#endif
-
-#if HAVE_FAST_ROUND
-# define TOINT_INTRINSICS 1
-
-static inline double_t
-roundtoint (double_t x)
-{
-  return round (x);
-}
-
-static inline uint64_t
-converttoint (double_t x)
-{
-# if HAVE_FAST_LROUND
-  return lround (x);
-# else
-  return (long) round (x);
-# endif
-}
-#endif
-
 #ifndef TOINT_INTRINSICS
 # define TOINT_INTRINSICS 0
 #endif
@@ -147,56 +109,12 @@ issignalingf_inline (float x)
   return 2 * (ix ^ 0x00400000) > 2u * 0x7fc00000;
 }
 
-/* Force the evaluation of a floating-point expression for its side-effect.  */
-#if __aarch64__ && __GNUC__
-static inline void
-force_eval_float (float x)
-{
-  __asm__ __volatile__ ("" : "+w" (x));
-}
-static inline void
-force_eval_double (double x)
-{
-  __asm__ __volatile__ ("" : "+w" (x));
-}
-#else
-static inline void
-force_eval_float (float x)
-{
-  volatile float y = x;
-}
-static inline void
-force_eval_double (double x)
-{
-  volatile double y = x;
-}
-#endif
-
-/* Evaluate an expression as the specified type, normally a type
-   cast should be enough, but compilers implement non-standard
-   excess-precision handling, so when FLT_EVAL_METHOD != 0 then
-   these functions may need to be customized.  */
-static inline float
-eval_as_float (float x)
-{
-  return x;
-}
-static inline double
-eval_as_double (double x)
-{
-  return x;
-}
-
 #ifdef __GNUC__
 # define HIDDEN __attribute__ ((__visibility__ ("hidden")))
 # define NOINLINE __attribute__ ((noinline))
-# define likely(x) __builtin_expect (!!(x), 1)
-# define unlikely(x) __builtin_expect (x, 0)
 #else
 # define HIDDEN
 # define NOINLINE
-# define likely(x) (x)
-# define unlikely(x) (x)
 #endif
 
 HIDDEN float __math_oflowf (unsigned long);
diff --git a/newlib/libm/common/sincosf.c b/newlib/libm/common/sincosf.c
deleted file mode 100644
index 85f12642f..000000000
--- a/newlib/libm/common/sincosf.c
+++ /dev/null
@@ -1,104 +0,0 @@
-/* Single-precision sincos function.
-   Copyright (c) 2018 Arm Ltd.  All rights reserved.
-
-   Redistribution and use in source and binary forms, with or without
-   modification, are permitted provided that the following conditions
-   are met:
-   1. Redistributions of source code must retain the above copyright
-      notice, this list of conditions and the following disclaimer.
-   2. Redistributions in binary form must reproduce the above copyright
-      notice, this list of conditions and the following disclaimer in the
-      documentation and/or other materials provided with the distribution.
-   3. The name of the company may not be used to endorse or promote
-      products derived from this software without specific prior written
-      permission.
-
-   THIS SOFTWARE IS PROVIDED BY ARM LTD ``AS IS AND ANY EXPRESS OR IMPLIED
-   WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
-   MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
-   IN NO EVENT SHALL ARM LTD BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
-   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
-   TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
-   PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
-   LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
-   NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
-   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */
-
-#include "fdlibm.h"
-#if !__OBSOLETE_MATH
-
-#include <stdint.h>
-#include <math.h>
-#include "math_config.h"
-#include "sincosf.h"
-
-/* Fast sincosf implementation.  Worst-case ULP is 0.56072, maximum relative
-   error is 0.5303p-23.  A single-step signed range reduction is used for
-   small values.  Large inputs have their range reduced using fast integer
-   arithmetic.
-*/
-void
-sincosf (float y, float *sinp, float *cosp)
-{
-  double x = y;
-  double s;
-  int n;
-  sincos_t *p = &sincosf_table[0];
-
-  if (abstop12 (y) < abstop12 (pio4))
-    {
-      double x2 = x * x;
-
-      if (unlikely (abstop12 (y) < abstop12 (0x1p-12f)))
-      {
-	if (unlikely (abstop12 (y) < abstop12 (0x1p-126f)))
-	  /* Force underflow for tiny y.  */
-	  force_eval_float (x2);
-	*sinp = y;
-	*cosp = 1.0f;
-	return;
-      }
-
-      sincosf_poly (x, x2, p, 0, sinp, cosp);
-    }
-  else if (abstop12 (y) < abstop12 (120.0f))
-    {
-      x = reduce_fast (x, p, &n);
-
-      /* Setup the signs for sin and cos.  */
-      s = p->sign[n & 3];
-
-      if (n & 2)
-	p = &sincosf_table[1];
-
-      sincosf_poly (x * s, x * x, p, n, sinp, cosp);
-    }
-  else if (likely (abstop12 (y) < abstop12 (INFINITY)))
-    {
-      uint32_t xi = asuint (y);
-      int sign = xi >> 31;
-
-      x = reduce_large (xi, &n);
-
-      /* Setup signs for sin and cos - include original sign.  */
-      s = p->sign[(n + sign) & 3];
-
-      if ((n + sign) & 2)
-	p = &sincosf_table[1];
-
-      sincosf_poly (x * s, x * x, p, n, sinp, cosp);
-    }
-  else
-    {
-      /* Return NaN if Inf or NaN for both sin and cos.  */
-      *sinp = *cosp = y - y;
-#if WANT_ERRNO
-      /* Needed to set errno for +-Inf, the add is a hack to work
-	 around a gcc register allocation issue: just passing y
-	 affects code generation in the fast path.  */
-      __math_invalidf (y + y);
-#endif
-    }
-}
-
-#endif
diff --git a/newlib/libm/common/sincosf.h b/newlib/libm/common/sincosf.h
deleted file mode 100644
index 2f3b2f83b..000000000
--- a/newlib/libm/common/sincosf.h
+++ /dev/null
@@ -1,172 +0,0 @@
-/* Header for single-precision sin/cos/sincos functions.
-   Copyright (c) 2018 Arm Ltd.  All rights reserved.
-
-   Redistribution and use in source and binary forms, with or without
-   modification, are permitted provided that the following conditions
-   are met:
-   1. Redistributions of source code must retain the above copyright
-      notice, this list of conditions and the following disclaimer.
-   2. Redistributions in binary form must reproduce the above copyright
-      notice, this list of conditions and the following disclaimer in the
-      documentation and/or other materials provided with the distribution.
-   3. The name of the company may not be used to endorse or promote
-      products derived from this software without specific prior written
-      permission.
-
-   THIS SOFTWARE IS PROVIDED BY ARM LTD ``AS IS AND ANY EXPRESS OR IMPLIED
-   WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
-   MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
-   IN NO EVENT SHALL ARM LTD BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
-   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
-   TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
-   PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
-   LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
-   NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
-   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */
-
-#include <stdint.h>
-#include <math.h>
-#include "math_config.h"
-
-/* PI * 2^-64.  */
-static const double pi64 = 0x1.921FB54442D18p-62;
-/* PI / 4.  */
-static const double pio4 = 0x1.921FB54442D18p-1;
-
-typedef const struct
-{
-  double sign[4];
-  double hpi_inv, hpi, c0, c1, c2, c3, c4, s1, s2, s3;
-} sincos_t;
-
-extern sincos_t sincosf_table[2] HIDDEN;
-
-extern const uint32_t inv_pio4[] HIDDEN;
-
-/* abstop12 assumes floating point reinterpret is fast by default.
-   If floating point comparisons are faster, define PREFER_FLOAT_COMPARISON.  */
-#if PREFER_FLOAT_COMPARISON
-static inline float
-abstop12 (float x)
-{
-  return fabsf (x);
-}
-#else
-static inline uint32_t
-abstop12 (float x)
-{
-  return (asuint (x) >> 20) & 0x7ff;
-}
-#endif
-
-/* Compute the sine and cosine of inputs X and X2 (X squared), using the
-   polynomial P and store the results in SINP and COSP.  N is the quadrant,
-   if odd the cosine and sine polynomials are swapped.  */
-static inline void
-sincosf_poly (double x, double x2, sincos_t *p, int n, float *sinp, float *cosp)
-{
-  double x3, x4, x5, x6, s, c, c1, c2, s1;
-
-  x4 = x2 * x2;
-  x3 = x2 * x;
-  c2 = p->c3 + x2 * p->c4;
-  s1 = p->s2 + x2 * p->s3;
-
-  /* Swap sin/cos result based on quadrant.  */
-  float *tmp = (n & 1 ? cosp : sinp);
-  cosp = (n & 1 ? sinp : cosp);
-  sinp = tmp;
-
-  c1 = p->c0 + x2 * p->c1;
-  x5 = x3 * x2;
-  x6 = x4 * x2;
-
-  s = x + x3 * p->s1;
-  c = c1 + x4 * p->c2;
-
-  *sinp = s + x5 * s1;
-  *cosp = c + x6 * c2;
-}
-
-/* Return the sine of inputs X and X2 (X squared) using the polynomial P.
-   N is the quadrant, and if odd the cosine polynomial is used.  */
-static inline float
-sinf_poly (double x, double x2, sincos_t *p, int n)
-{
-  double x3, x4, x6, x7, s, c, c1, c2, s1;
-
-  if ((n & 1) == 0)
-    {
-      x3 = x * x2;
-      s1 = p->s2 + x2 * p->s3;
-
-      x7 = x3 * x2;
-      s = x + x3 * p->s1;
-
-      return s + x7 * s1;
-    }
-  else
-    {
-      x4 = x2 * x2;
-      c2 = p->c3 + x2 * p->c4;
-      c1 = p->c0 + x2 * p->c1;
-
-      x6 = x4 * x2;
-      c = c1 + x4 * p->c2;
-
-      return c + x6 * c2;
-    }
-}
-
-/* Fast range reduction using single multiply-subtract.  Return the modulo of
-   X as a value between -PI/4 and PI/4 and store the quadrant in NP.
-   The values for PI/2 and 2/PI are accessed via P.  Since PI/2 as a double
-   is accurate to 55 bits and the worst-case cancellation happens at 6 * PI/4,
-   only 2 multiplies are required and the result is accurate for |X| <= 120.0.
-   Use round/lround if inlined, otherwise convert to int.  To avoid inaccuracies
-   introduced by truncating negative values, compute the quadrant * 2^24.  */
-static inline double
-reduce_fast (double x, sincos_t *p, int *np)
-{
-  double r;
-#if TOINT_INTRINSICS
-  r = x * p->hpi_inv;
-  *np = converttoint (r);
-  return x - roundtoint (r) * p->hpi;
-#else
-  r = x * p->hpi_inv;
-  int n = ((int)r + 0x800000) >> 24;
-  *np = n;
-  return x - n * p->hpi;
-#endif
-}
-
-/* Reduce the range of XI to a multiple of PI/4 using fast integer arithmetic.
-   XI is a reinterpreted float and must be >= 2.0f (the sign bit is ignored).
-   Return the modulo between -PI/4 and PI/4 and store the quadrant in NP.
-   Reduction uses a table of 4/PI with 192 bits of precision.  A 32x96->128 bit
-   multiply computes the exact 2.62-bit fixed-point modulo.  Since the result
-   can have at most 29 leading zeros after the binary point, the double
-   precision result is accurate to 33 bits.  */
-static inline double
-reduce_large (uint32_t xi, int *np)
-{
-  const uint32_t *arr = &inv_pio4[(xi >> 26) & 15];
-  int shift = (xi >> 23) & 7;
-  uint64_t n, res0, res1, res2;
-
-  xi = (xi & 0xffffff) | 0x800000;
-  xi <<= shift;
-
-  res0 = xi * arr[0];
-  res1 = (uint64_t)xi * arr[4];
-  res2 = (uint64_t)xi * arr[8];
-  res0 = (res2 >> 32) | (res0 << 32);
-  res0 += res1;
-
-  n = (res0 + (1ULL << 61)) >> 62;
-  res0 -= n << 62;
-  double x = (int64_t)res0;
-  *np = n;
-  return x * pi64;
-}
diff --git a/newlib/libm/common/sincosf_data.c b/newlib/libm/common/sincosf_data.c
deleted file mode 100644
index 069ee7bf6..000000000
--- a/newlib/libm/common/sincosf_data.c
+++ /dev/null
@@ -1,87 +0,0 @@
-/* Data definitions for sinf, cosf and sincosf.
-   Copyright (c) 2018 Arm Ltd.  All rights reserved.
-
-   Redistribution and use in source and binary forms, with or without
-   modification, are permitted provided that the following conditions
-   are met:
-   1. Redistributions of source code must retain the above copyright
-      notice, this list of conditions and the following disclaimer.
-   2. Redistributions in binary form must reproduce the above copyright
-      notice, this list of conditions and the following disclaimer in the
-      documentation and/or other materials provided with the distribution.
-   3. The name of the company may not be used to endorse or promote
-      products derived from this software without specific prior written
-      permission.
-
-   THIS SOFTWARE IS PROVIDED BY ARM LTD ``AS IS AND ANY EXPRESS OR IMPLIED
-   WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
-   MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
-   IN NO EVENT SHALL ARM LTD BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
-   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
-   TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
-   PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
-   LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
-   NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
-   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */
-
-#include "fdlibm.h"
-#if !__OBSOLETE_MATH
-
-#include <stdint.h>
-#include <math.h>
-#include "math_config.h"
-#include "sincosf.h"
-
-/* The constants and polynomials for sine and cosine.  The 2nd entry
-   computes -cos (x) rather than cos (x) to get negation for free.  */
-sincos_t sincosf_table[2] =
-{
-  {
-    { 1.0, -1.0, -1.0, 1.0 },
-#if TOINT_INTRINSICS
-    0x1.45F306DC9C883p-1,
-#else
-    0x1.45F306DC9C883p+23,
-#endif
-    0x1.921FB54442D18p0,
-    0x1p0,
-    -0x1.ffffffd0c621cp-2,
-    0x1.55553e1068f19p-5,
-    -0x1.6c087e89a359dp-10,
-    0x1.99343027bf8c3p-16,
-    -0x1.555545995a603p-3,
-    0x1.1107605230bc4p-7,
-    -0x1.994eb3774cf24p-13
-  },
-  {
-    { 1.0, -1.0, -1.0, 1.0 },
-#if TOINT_INTRINSICS
-    0x1.45F306DC9C883p-1,
-#else
-    0x1.45F306DC9C883p+23,
-#endif
-    0x1.921FB54442D18p0,
-    -0x1p0,
-    0x1.ffffffd0c621cp-2,
-    -0x1.55553e1068f19p-5,
-    0x1.6c087e89a359dp-10,
-    -0x1.99343027bf8c3p-16,
-    -0x1.555545995a603p-3,
-    0x1.1107605230bc4p-7,
-    -0x1.994eb3774cf24p-13
-  }
-};
-
-/* Table with 4/PI to 192 bit precision.  To avoid unaligned accesses
-   only 8 new bits are added per entry, making the table 4 times larger.  */
-const uint32_t inv_pio4[24] =
-{
-  0xa2,       0xa2f9,	  0xa2f983,   0xa2f9836e,
-  0xf9836e4e, 0x836e4e44, 0x6e4e4415, 0x4e441529,
-  0x441529fc, 0x1529fc27, 0x29fc2757, 0xfc2757d1,
-  0x2757d1f5, 0x57d1f534, 0xd1f534dd, 0xf534ddc0,
-  0x34ddc0db, 0xddc0db62, 0xc0db6295, 0xdb629599,
-  0x6295993c, 0x95993c43, 0x993c4390, 0x3c439041
-};
-
-#endif
diff --git a/newlib/libm/common/sinf.c b/newlib/libm/common/sinf.c
deleted file mode 100644
index c1baf4237..000000000
--- a/newlib/libm/common/sinf.c
+++ /dev/null
@@ -1,92 +0,0 @@
-/* Single-precision sin function.
-   Copyright (c) 2018 Arm Ltd.  All rights reserved.
-
-   Redistribution and use in source and binary forms, with or without
-   modification, are permitted provided that the following conditions
-   are met:
-   1. Redistributions of source code must retain the above copyright
-      notice, this list of conditions and the following disclaimer.
-   2. Redistributions in binary form must reproduce the above copyright
-      notice, this list of conditions and the following disclaimer in the
-      documentation and/or other materials provided with the distribution.
-   3. The name of the company may not be used to endorse or promote
-      products derived from this software without specific prior written
-      permission.
-
-   THIS SOFTWARE IS PROVIDED BY ARM LTD ``AS IS AND ANY EXPRESS OR IMPLIED
-   WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
-   MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
-   IN NO EVENT SHALL ARM LTD BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
-   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
-   TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
-   PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
-   LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
-   NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
-   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */
-
-#include "fdlibm.h"
-#if !__OBSOLETE_MATH
-
-#include <math.h>
-#include "math_config.h"
-#include "sincosf.h"
-
-/* Fast sinf implementation.  Worst-case ULP is 0.56072, maximum relative
-   error is 0.5303p-23.  A single-step signed range reduction is used for
-   small values.  Large inputs have their range reduced using fast integer
-   arithmetic.
-*/
-float
-sinf (float y)
-{
-  double x = y;
-  double s;
-  int n;
-  sincos_t *p = &sincosf_table[0];
-
-  if (abstop12 (y) < abstop12 (pio4))
-    {
-      s = x * x;
-
-      if (unlikely (abstop12 (y) < abstop12 (0x1p-12f)))
-      {
-	if (unlikely (abstop12 (y) < abstop12 (0x1p-126f)))
-	  /* Force underflow for tiny y.  */
-	  force_eval_float (s);
-	return y;
-      }
-
-      return sinf_poly (x, s, p, 0);
-    }
-  else if (likely (abstop12 (y) < abstop12 (120.0f)))
-    {
-      x = reduce_fast (x, p, &n);
-
-      /* Setup the signs for sin and cos.  */
-      s = p->sign[n & 3];
-
-      if (n & 2)
-	p = &sincosf_table[1];
-
-      return sinf_poly (x * s, x * x, p, n);
-    }
-  else if (abstop12 (y) < abstop12 (INFINITY))
-    {
-      uint32_t xi = asuint (y);
-      int sign = xi >> 31;
-
-      x = reduce_large (xi, &n);
-
-      /* Setup signs for sin and cos - include original sign.  */
-      s = p->sign[(n + sign) & 3];
-
-      if ((n + sign) & 2)
-	p = &sincosf_table[1];
-
-      return sinf_poly (x * s, x * x, p, n);
-    }
-  else
-    return __math_invalidf (y);
-}
-
-#endif
diff --git a/newlib/libm/math/sf_cos.c b/newlib/libm/math/sf_cos.c
index 8cb0eb008..4c0a9a535 100644
--- a/newlib/libm/math/sf_cos.c
+++ b/newlib/libm/math/sf_cos.c
@@ -14,7 +14,6 @@
  */
 
 #include "fdlibm.h"
-#if __OBSOLETE_MATH
 
 #ifdef __STDC__
 static const float one=1.0;
@@ -67,4 +66,3 @@ static float one=1.0;
 }
 
 #endif /* defined(_DOUBLE_IS_32BITS) */
-#endif /* __OBSOLETE_MATH */
diff --git a/newlib/libm/math/sf_sin.c b/newlib/libm/math/sf_sin.c
index c3ec6e2e4..da81845d9 100644
--- a/newlib/libm/math/sf_sin.c
+++ b/newlib/libm/math/sf_sin.c
@@ -14,7 +14,6 @@
  */
 
 #include "fdlibm.h"
-#if __OBSOLETE_MATH
 
 #ifdef __STDC__
 	float sinf(float x)
@@ -61,4 +60,3 @@
 }
 
 #endif /* defined(_DOUBLE_IS_32BITS) */
-#endif /* __OBSOLETE_MATH */
diff --git a/newlib/libm/math/wf_sincos.c b/newlib/libm/math/wf_sincos.c
index 69eb922c9..477c60401 100644
--- a/newlib/libm/math/wf_sincos.c
+++ b/newlib/libm/math/wf_sincos.c
@@ -1,8 +1,6 @@
 /* sincos -- currently no more efficient than two separate calls to
    sin and cos. */
 #include "fdlibm.h"
-#if __OBSOLETE_MATH
-
 #include <errno.h>
 
 #ifdef __STDC__
@@ -33,4 +31,3 @@
   *cosx = cosf((float) x);
 }
 #endif /* defined(_DOUBLE_IS_32BITS) */
-#endif /* __OBSOLETE_MATH */