Implement C23 powr

C23 adds various <math.h> function families originally defined in TS 18661-4. Add the powr functions, which are like pow, but with simpler handling of special cases (based on exp(y*log(x)), so negative x and 0^0 are domain errors, powers of -0 are always +0 or +Inf never -0 or -Inf, and 1^+-Inf and Inf^0 are also domain errors, while NaN^0 and 1^NaN are NaN). The test inputs are taken from those for pow, with appropriate adjustments (including removing all tests that would be domain errors from those in auto-libm-test-in and adding some more such tests in libm-test-powr.inc). The underlying implementation uses __ieee754_pow functions after dealing with all special cases that need to be handled differently. It might be a little faster (avoiding a wrapper and redundant checks for special cases) to have an underlying implementation built separately for both pow and powr with compile-time conditionals for special-case handling, but I expect the benefit of that would be limited given that both functions will end up needing to use the same logic for computing pow outside of special cases. My understanding is that powr(negative, qNaN) should raise "invalid": that the rule on "invalid" for an argument outside the domain of the function takes precedence over a quiet NaN argument producing a quiet NaN result with no exceptions raised (for rootn it's explicit that the 0th root of qNaN raises "invalid"). I've raised this on the WG14 reflector to confirm the intent. Tested for x86_64 and x86, and with build-many-glibcs.py.
2025-03-14 15:58:11 +00:00 · 2025-03-14 15:58:11 +00:00 · 409668f6e8
parent c7c4a5906f
commit 409668f6e8
48 changed files with 18866 additions and 2 deletions
--- a/2
+++ b/2
@ -14,7 +14,7 @@ Major new features:
  functions for float, double, long double, _FloatN and _FloatNx, and a
  type-generic macro in <tgmath.h>.

-  - Power and absolute-value functions: rsqrt.
+  - Power and absolute-value functions: powr, rsqrt.

 * On Linux, the pthread_gettid_np function has been added.

--- a/manual/math.texi
+++ b/manual/math.texi
@ -774,6 +774,21 @@ do that, so instead it signals a domain error. @code{pow} may also
 underflow or overflow the destination type.
@end deftypefun

+@deftypefun double powr (double @var{base}, double @var{power})
+@deftypefunx float powrf (float @var{base}, float @var{power})
+@deftypefunx {long double} powrl (long double @var{base}, long double @var{power})
+@deftypefunx _FloatN powrfN (_Float@var{N} @var{base}, _Float@var{N} @var{power})
+@deftypefunx _FloatNx powrfNx (_Float@var{N}x @var{base}, _Float@var{N}x @var{power})
+@standards{TS 18661-4:2015, math.h}
+@safety{@prelim{}@mtsafe{}@assafe{}@acsafe{}}
+These return @var{base} raised to @var{power}, but with special cases
+based on @code{exp (power * log (base))}.  For example, if @var{base}
+is negative, the @code{powr} functions always signal a domain error,
+but this is valid for @code{pow} if @var{power} is an integer.
+
+The @code{powr} functions are from TS 18661-4:2015.
+@end deftypefun
+
@cindex square root function
@deftypefun double sqrt (double @var{x})
@deftypefunx float sqrtf (float @var{x})
--- a/math/Makefile
+++ b/math/Makefile
@ -139,6 +139,7 @@ gen-libm-calls = \
  s_log2p1F \
  s_nanF \
  s_nextdownF \
+  s_powrF \
  s_rsqrtF \
  s_significandF \
  s_sinpiF \
@ -685,6 +686,7 @@ libm-test-funcs-auto = \
  log1p \
  log2p1 \
  pow \
+  powr \
  rsqrt \
  sin \
  sincos \
@ -1016,6 +1018,7 @@ tgmath3-macros = \
  nexttoward \
  nextup \
  pow \
+  powr \
  remainder \
  remquo \
  rint \
@ -1439,6 +1442,7 @@ CFLAGS-s_nexttoward.c += -fno-builtin-nexttoward -fno-builtin-nexttowardl
 CFLAGS-s_nexttowardf.c += -fno-builtin-nexttowardf
 CFLAGS-s_nextup.c += -fno-builtin-nextupl
 CFLAGS-e_pow.c += -fno-builtin-powl
+CFLAGS-s_powr.c += -fno-builtin-powrl
 CFLAGS-w_remainder.c += -fno-builtin-remainderl -fno-builtin-dreml
 CFLAGS-s_remquo.c += -fno-builtin-remquol
 CFLAGS-s_rint.c += -fno-builtin-rintl
@ -1574,6 +1578,7 @@ CFLAGS-s_nextafter.c += -fno-builtin-nextafterf32x -fno-builtin-nextafterf64
 CFLAGS-s_nextdown.c += -fno-builtin-nextdownf32x -fno-builtin-nextdownf64
 CFLAGS-s_nextup.c += -fno-builtin-nextupf32x -fno-builtin-nextupf64
 CFLAGS-e_pow.c += -fno-builtin-powf32x -fno-builtin-powf64
+CFLAGS-s_powr.c += -fno-builtin-powrf32x -fno-builtin-powrf64
 CFLAGS-w_remainder.c += -fno-builtin-remainderf32x -fno-builtin-remainderf64
 CFLAGS-s_remquo.c += -fno-builtin-remquof32x -fno-builtin-remquof64
 CFLAGS-s_rint.c += -fno-builtin-rintf32x -fno-builtin-rintf64
@ -1700,6 +1705,7 @@ CFLAGS-s_nextafterf.c += -fno-builtin-nextafterf32
 CFLAGS-s_nextdownf.c += -fno-builtin-nextdownf32
 CFLAGS-s_nextupf.c += -fno-builtin-nextupf32
 CFLAGS-e_powf.c += -fno-builtin-powf32
+CFLAGS-s_powrf.c += -fno-builtin-powrf32
 CFLAGS-w_remainderf.c += -fno-builtin-remainderf32
 CFLAGS-s_remquof.c += -fno-builtin-remquof32
 CFLAGS-s_rintf.c += -fno-builtin-rintf32
--- a/math/Versions
+++ b/math/Versions
@ -673,8 +673,10 @@ libm {
  }
  GLIBC_2.42 {
    # Functions not involving _Float64x or _Float128, for all configurations.
+    powr; powrf; powrl; powrf32; powrf64; powrf32x;
    rsqrt; rsqrtf; rsqrtl; rsqrtf32; rsqrtf64; rsqrtf32x;
    # Functions involving _Float64x or _Float128, for some configurations.
+    powrf64x; powrf128;
    rsqrtf64x; rsqrtf128;
  }
 }
--- a/math/auto-libm-test-in
+++ b/math/auto-libm-test-in
@ -8151,6 +8151,313 @@ pow 0x1.059c76p+0 0x1.ff80bep+11
 pow 0x1.7ac7cp+5 23
 pow -0x1.7ac7cp+5 23

+powr 10 0
+powr 10 -0
+
+powr 1 1
+powr 1 -1
+powr 1 1.25
+powr 1 -1.25
+powr 1 0x1p62
+powr 1 0x1p63
+powr 1 0x1p64
+powr 1 0x1p72
+powr 1 min_subnorm
+powr 1 -min_subnorm
+
+# powr (x, +-0) == 1.
+powr 32.75 0
+powr 32.75 -0
+powr 0x1p72 0
+powr 0x1p72 -0
+powr 0x1p-72 0
+powr 0x1p-72 -0
+
+powr 0x1p72 0x1p72
+powr 10 -0x1p72
+powr max max
+powr 10 -max
+
+powr 0 1
+powr 0 11
+
+powr -0 1
+powr -0 11
+
+powr 0 2
+powr 0 11.1
+
+powr -0 2
+powr -0 11.1
+
+# powr (+0, y) == +0 for y an odd integer > 0.
+powr 0.0 27
+powr 0.0 0xffffff
+powr 0.0 0x1.fffffffffffffp+52
+powr 0.0 0x1.fffffffffffffffep+63
+powr 0.0 0x1.ffffffffffffffffffffffffff8p+105
+powr 0.0 0x1.ffffffffffffffffffffffffffffp+112
+
+# powr (-0, y) == +0 (unlike pow) for y an odd integer > 0.
+powr -0 27
+powr -0 0xffffff
+powr -0 0x1fffffe
+powr -0 0x1.fffffffffffffp+52
+powr -0 0x1.fffffffffffffp+53
+powr -0 0x1.fffffffffffffffep+63
+powr -0 0x1.fffffffffffffffep+64 xfail-rounding:ibm128-libgcc
+powr -0 0x1.ffffffffffffffffffffffffff8p+105 xfail-rounding:ibm128-libgcc
+powr -0 0x1.ffffffffffffffffffffffffff8p+106 xfail-rounding:ibm128-libgcc
+powr -0 0x1.ffffffffffffffffffffffffffffp+112 xfail-rounding:ibm128-libgcc
+powr -0 0x1.ffffffffffffffffffffffffffffp+113 xfail-rounding:ibm128-libgcc
+
+# powr (+0, y) == +0 for y > 0 and not an odd integer.
+powr 0.0 4
+powr 0.0 0x1p24
+powr 0.0 0x1p127
+powr 0.0 max
+powr 0.0 min_subnorm
+
+# powr (-0, y) == +0 for y > 0 and not an odd integer.
+powr -0 0.5
+powr -0 4
+powr -0 0x1p24
+powr -0 0x1p127
+powr -0 max
+powr -0 min_subnorm
+
+powr 16 0.25
+powr 0x1p64 0.125
+powr 2 4
+powr 256 8
+
+powr 0.75 1.25
+
+powr 0x0.ffffffp0 10
+powr 0x0.ffffffp0 100
+powr 0x0.ffffffp0 1000
+powr 0x0.ffffffp0 0x1p24
+powr 0x0.ffffffp0 0x1p30
+powr 0x0.ffffffp0 0x1.234566p30
+powr 0x0.ffffffp0 -10
+powr 0x0.ffffffp0 -100
+powr 0x0.ffffffp0 -1000
+powr 0x0.ffffffp0 -0x1p24
+powr 0x0.ffffffp0 -0x1p30
+powr 0x0.ffffffp0 -0x1.234566p30
+powr 0x1.000002p0 0x1p24
+powr 0x1.000002p0 0x1.234566p29
+powr 0x1.000002p0 -0x1.234566p29
+
+powr 0x0.fffffffffffff8p0 0x1.23456789abcdfp62
+powr 0x0.fffffffffffff8p0 -0x1.23456789abcdfp62
+powr 0x1.0000000000001p0 0x1.23456789abcdfp61
+powr 0x1.0000000000001p0 -0x1.23456789abcdfp61
+
+powr 0x0.ffffffffffffffffp0 0x1.23456789abcdef0ep77
+powr 0x0.ffffffffffffffffp0 -0x1.23456789abcdef0ep77
+powr 0x1.0000000000000002p0 0x1.23456789abcdef0ep76
+powr 0x1.0000000000000002p0 -0x1.23456789abcdef0ep76
+
+powr 0x0.ffffffffffffffffffffffffffff8p0 0x1.23456789abcdef0123456789abcdp126
+powr 0x0.ffffffffffffffffffffffffffff8p0 -0x1.23456789abcdef0123456789abcdp126
+powr 0x1.0000000000000000000000000001p0 0x1.23456789abcdef0123456789abcdp125
+powr 0x1.0000000000000000000000000001p0 -0x1.23456789abcdef0123456789abcdp125
+
+powr 0x1.000002p0 0x1p30
+powr 0x1.000002p0 max
+powr 0x1.00000ep0 0x1p30
+powr 0x1.00000ep0 max
+
+powr 1e4932 0.75
+powr 1e4928 0.75
+powr 1e4924 0.75
+powr 1e4920 0.75
+powr 10.0 4932.0
+powr 10.0 4931.0
+powr 10.0 4930.0
+powr 10.0 4929.0
+powr 10.0 -4931.0
+powr 10.0 -4930.0
+powr 10.0 -4929.0
+powr 1e27 182.0
+powr 1e27 -182.0
+
+powr min_subnorm min_subnorm
+powr min_subnorm -min_subnorm
+powr max min_subnorm
+powr max -min_subnorm
+powr 0.99 min_subnorm
+powr 0.99 -min_subnorm
+powr 1.01 min_subnorm
+powr 1.01 -min_subnorm
+
+powr 2.0 -100000.0
+
+powr 2 -126.125
+powr 2 -126.25
+powr 2 -126.375
+powr 2 -126.5
+powr 2 -126.625
+powr 2 -126.75
+powr 2 -126.875
+powr 2 -969.125
+powr 2 -969.25
+powr 2 -969.375
+powr 2 -969.5
+powr 2 -969.625
+powr 2 -969.75
+powr 2 -969.875
+powr 2 -1022.125
+powr 2 -1022.25
+powr 2 -1022.375
+powr 2 -1022.5
+powr 2 -1022.625
+powr 2 -1022.75
+powr 2 -1022.875
+powr 2 -16382.125
+powr 2 -16382.25
+powr 2 -16382.375
+powr 2 -16382.5
+powr 2 -16382.625
+powr 2 -16382.75
+powr 2 -16382.875
+powr 2 -16383.125
+powr 2 -16383.25
+powr 2 -16383.375
+powr 2 -16383.5
+powr 2 -16383.625
+powr 2 -16383.75
+powr 2 -16383.875
+powr 0.5 126.125
+powr 0.5 126.25
+powr 0.5 126.375
+powr 0.5 126.5
+powr 0.5 126.625
+powr 0.5 126.75
+powr 0.5 126.875
+powr 0.5 969.125
+powr 0.5 969.25
+powr 0.5 969.375
+powr 0.5 969.5
+powr 0.5 969.625
+powr 0.5 969.75
+powr 0.5 969.875
+powr 0.5 1022.125
+powr 0.5 1022.25
+powr 0.5 1022.375
+powr 0.5 1022.5
+powr 0.5 1022.625
+powr 0.5 1022.75
+powr 0.5 1022.875
+powr 0.5 16382.125
+powr 0.5 16382.25
+powr 0.5 16382.375
+powr 0.5 16382.5
+powr 0.5 16382.625
+powr 0.5 16382.75
+powr 0.5 16382.875
+powr 0.5 16383.125
+powr 0.5 16383.25
+powr 0.5 16383.375
+powr 0.5 16383.5
+powr 0.5 16383.625
+powr 0.5 16383.75
+powr 0.5 16383.875
+powr 0x1.00001p1 -126
+powr 0x1.00003p1 -126
+powr 0x1.00005p1 -126
+powr 0x1.00007p1 -126
+powr 0x1.00001p1 -969
+powr 0x1.00003p1 -969
+powr 0x1.00005p1 -969
+powr 0x1.00007p1 -969
+powr 0x1.00001p1 -1022
+powr 0x1.00003p1 -1022
+powr 0x1.00005p1 -1022
+powr 0x1.00007p1 -1022
+powr 0x1.00001p1 -16382
+powr 0x1.00003p1 -16382
+powr 0x1.00005p1 -16382
+powr 0x1.00007p1 -16382
+powr 0x1.00001p1 -16383
+powr 0x1.00003p1 -16383
+powr 0x1.00005p1 -16383
+powr 0x1.00007p1 -16383
+powr 0x0.ffff1p-1 126
+powr 0x0.ffff3p-1 126
+powr 0x0.ffff5p-1 126
+powr 0x0.ffff7p-1 126
+powr 0x0.ffff1p-1 969
+powr 0x0.ffff3p-1 969
+powr 0x0.ffff5p-1 969
+powr 0x0.ffff7p-1 969
+powr 0x0.ffff1p-1 1022
+powr 0x0.ffff3p-1 1022
+powr 0x0.ffff5p-1 1022
+powr 0x0.ffff7p-1 1022
+powr 0x0.ffff1p-1 16382
+powr 0x0.ffff3p-1 16382
+powr 0x0.ffff5p-1 16382
+powr 0x0.ffff7p-1 16382
+powr 0x0.ffff1p-1 16383
+powr 0x0.ffff3p-1 16383
+powr 0x0.ffff5p-1 16383
+powr 0x0.ffff7p-1 16383
+powr 0x2.000b3p0 -1022
+powr 0x2.000582p0 -1022
+powr 2 -0x3.fe513p+8
+powr 2 -0x3.fe4e8p+8
+
+powr 10 -1
+powr 10 -2
+powr 10 -3
+powr 10 -4
+powr 10 -5
+powr 10 -6
+powr 10 -7
+
+powr 0x0.ffffffffffffffffp0 1
+powr 0x0.ffffffffffffffffp0 2
+powr 0x0.ffffffffffffffffp0 3
+powr 0x0.ffffffffffffffffp0 4
+powr 0x0.ffffffffffffffffp0 5
+powr 0x0.ffffffffffffffffp0 6
+powr 0x0.ffffffffffffffffp0 7
+powr 0x0.ffffffffffffffffp0 -1
+powr 0x0.ffffffffffffffffp0 -2
+powr 0x0.ffffffffffffffffp0 -3
+powr 0x0.ffffffffffffffffp0 -4
+powr 0x0.ffffffffffffffffp0 -5
+powr 0x0.ffffffffffffffffp0 -6
+powr 0x0.ffffffffffffffffp0 -7
+
+powr 0x1.0000000000000002p0 1
+powr 0x1.0000000000000002p0 2
+powr 0x1.0000000000000002p0 3
+powr 0x1.0000000000000002p0 4
+powr 0x1.0000000000000002p0 5
+powr 0x1.0000000000000002p0 6
+powr 0x1.0000000000000002p0 7
+powr 0x1.0000000000000002p0 -1
+powr 0x1.0000000000000002p0 -2
+powr 0x1.0000000000000002p0 -3
+powr 0x1.0000000000000002p0 -4
+powr 0x1.0000000000000002p0 -5
+powr 0x1.0000000000000002p0 -6
+powr 0x1.0000000000000002p0 -7
+
+powr 1.0625 1.125
+powr 1.5 1.03125
+powr 0x1.7d1a0a6f2p+681 1.5
+powr 0x1.ce78f2p+0 -0x2.7f1f78p+4
+powr 0xf.fffffp+124 -0x5.b5b648p+0
+powr 0x1.430d4cp+0 0x5.0e462p+4
+powr 0x9.8b82ap-4 -0x1.99907ap+12
+powr 0xd.73035p-4 -0x1.47bb8p+8
+powr 0x1.059c76p+0 0x1.ff80bep+11
+powr 0x1.7ac7cp+5 23
+
 rsqrt 2209
 rsqrt 4
 rsqrt 2
--- a/math/auto-libm-test-out-powr
+++ b/math/auto-libm-test-out-powr
--- a/math/bits/mathcalls.h
+++ b/math/bits/mathcalls.h
@ -186,6 +186,9 @@ __MATHCALL_VEC (cbrt,, (_Mdouble_ __x));
 #endif

 #if __GLIBC_USE (IEC_60559_FUNCS_EXT_C23)
+/* Return X to the Y power.  */
+__MATHCALL (powr,, (_Mdouble_ __x, _Mdouble_ __y));
+
 /* Return the reciprocal of the square root of X.  */
 __MATHCALL (rsqrt,, (_Mdouble_ __x));
 #endif
--- a/math/gen-auto-libm-tests.c
+++ b/math/gen-auto-libm-tests.c
@ -589,6 +589,7 @@ static test_function test_functions[] =
    FUNC_mpfr_f_f ("log2p1", mpfr_log2p1, false),
    FUNC_mpfr_ff_f ("mul", mpfr_mul, true),
    FUNC_mpfr_ff_f ("pow", mpfr_pow, false),
+    FUNC_mpfr_ff_f ("powr", mpfr_powr, false),
    /* mpfr_rec_sqrt differs from rsqrt on -0, but gen-auto-libm-tests
       does not handle results that are exact infinities anyway.  */
    FUNC_mpfr_f_f ("rsqrt", mpfr_rec_sqrt, true),
--- a/math/gen-tgmath-tests.py
+++ b/math/gen-tgmath-tests.py
@ -753,6 +753,7 @@ class Tests(object):
        self.add_tests('log2p1', 'r', ['r'])
        self.add_tests('log10p1', 'r', ['r'])
        self.add_tests('logp1', 'r', ['r'])
+        self.add_tests('powr', 'r', ['r', 'r'])
        self.add_tests('rsqrt', 'r', ['r'])
        self.add_tests('sinpi', 'r', ['r'])
        self.add_tests('tanpi', 'r', ['r'])
--- a/math/libm-test-powr.inc
+++ b/math/libm-test-powr.inc
@ -0,0 +1,454 @@
+/* Test powr.
+   Copyright (C) 1997-2025 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
+   modify it under the terms of the GNU Lesser General Public
+   License as published by the Free Software Foundation; either
+   version 2.1 of the License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   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, see
+   <https://www.gnu.org/licenses/>.  */
+
+#include "libm-test-driver.c"
+
+static const struct test_ff_f_data powr_test_data[] =
+  {
+    TEST_ff_f (powr, qnan_value, 0, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, 0, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, qnan_value, minus_zero, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, minus_zero, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, snan_value, 0, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, 0, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, snan_value, minus_zero, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, minus_zero, qnan_value, INVALID_EXCEPTION),
+
+    TEST_ff_f (powr, plus_zero, plus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, plus_zero, minus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_zero, plus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_zero, minus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, -min_subnorm_value, plus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_subnorm_value, minus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_subnorm_value, -min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_subnorm_value, min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_subnorm_value, -min_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_subnorm_value, min_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_subnorm_value, -1.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_subnorm_value, 1.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_subnorm_value, -2.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_subnorm_value, 2.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_subnorm_value, -max_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_subnorm_value, max_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, -min_value, plus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_value, minus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_value, -min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_value, min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_value, -min_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_value, min_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_value, -1.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_value, 1.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_value, -2.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_value, 2.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_value, -max_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_value, max_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, -1.0, plus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.0, minus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.0, -min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.0, min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.0, -min_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.0, min_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.0, -1.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.0, 1.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.0, -2.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.0, 2.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.0, -max_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.0, max_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, -2.0, plus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -2.0, minus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -2.0, -min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -2.0, min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -2.0, -min_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -2.0, min_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -2.0, -1.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -2.0, 1.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -2.0, -2.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -2.0, 2.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -2.0, -max_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -2.0, max_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, -max_value, plus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -max_value, minus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -max_value, -min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -max_value, min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -max_value, -min_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -max_value, min_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -max_value, -1.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -max_value, 1.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -max_value, -2.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -max_value, 2.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -max_value, -max_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -max_value, max_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, 1.1L, plus_infty, plus_infty, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, plus_infty, plus_infty, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -1.1L, plus_infty, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, plus_infty, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, 0.9L, plus_infty, 0, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, 1e-7L, plus_infty, 0, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -0.9L, plus_infty, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1e-7L, plus_infty, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, 1.1L, minus_infty, 0, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, minus_infty, 0, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -1.1L, minus_infty, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, minus_infty, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, 0.9L, minus_infty, plus_infty, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, 1e-7L, minus_infty, plus_infty, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -0.9L, minus_infty, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1e-7L, minus_infty, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, plus_infty, 1e-7L, plus_infty, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, 1, plus_infty, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, 1e7L, plus_infty, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, min_subnorm_value, plus_infty, ERRNO_UNCHANGED),
+
+    TEST_ff_f (powr, plus_infty, -1e-7L, 0, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, -1, 0, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, -1e7L, 0, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, -min_subnorm_value, 0, ERRNO_UNCHANGED),
+
+    TEST_ff_f (powr, minus_infty, 1, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 11, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 1001, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, minus_infty, 2, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 12, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 1002, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 0.1L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 1.1L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 11.1L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 1001.1L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, minus_infty, -1, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -11, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -1001, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, minus_infty, -2, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -12, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -1002, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -0.1L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -1.1L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -11.1L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -1001.1L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, qnan_value, qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, qnan_value, -qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, -qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, qnan_value, snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, qnan_value, -snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -qnan_value, snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -qnan_value, -snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, snan_value, qnan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, snan_value, -qnan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, qnan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, -qnan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, snan_value, snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, snan_value, -snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, -snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, 0, qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, 0, -qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, 0, snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, 0, -snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, 1, qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, 1, -qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, 1, snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, 1, -snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -1, qnan_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1, -qnan_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1, snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -1, -snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, qnan_value, 1, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, 1, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, snan_value, 1, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, 1, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, qnan_value, -1, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, -1, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, snan_value, -1, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, -1, qnan_value, INVALID_EXCEPTION),
+
+    /* powr (x, qNaN or sNaN) == qNaN.  */
+    TEST_ff_f (powr, 3.0, qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, 3.0, -qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, minus_zero, qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, minus_zero, -qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, -qnan_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -3.0, qnan_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -3.0, -qnan_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, qnan_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -qnan_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, 3.0, snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, 3.0, -snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, minus_zero, snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, minus_zero, -snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, plus_infty, snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, plus_infty, -snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -3.0, snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -3.0, -snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, minus_infty, snan_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, minus_infty, -snan_value, qnan_value, INVALID_EXCEPTION),
+
+    TEST_ff_f (powr, qnan_value, 3.0, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, 3.0, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, qnan_value, -3.0, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, -3.0, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, qnan_value, plus_infty, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, plus_infty, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, qnan_value, minus_infty, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, minus_infty, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, qnan_value, 2.5, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, 2.5, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, qnan_value, -2.5, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, -2.5, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, qnan_value, min_subnorm_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, min_subnorm_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, qnan_value, -min_subnorm_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, -qnan_value, -min_subnorm_value, qnan_value, NO_INEXACT_EXCEPTION|ERRNO_UNCHANGED),
+    TEST_ff_f (powr, snan_value, 3.0, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, 3.0, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, snan_value, -3.0, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, -3.0, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, snan_value, plus_infty, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, plus_infty, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, snan_value, minus_infty, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, minus_infty, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, snan_value, 2.5, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, 2.5, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, snan_value, -2.5, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, -2.5, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, snan_value, min_subnorm_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, min_subnorm_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, snan_value, -min_subnorm_value, qnan_value, INVALID_EXCEPTION),
+    TEST_ff_f (powr, -snan_value, -min_subnorm_value, qnan_value, INVALID_EXCEPTION),
+
+    TEST_ff_f (powr, 1, plus_infty, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1, plus_infty, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, 1, minus_infty, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1, minus_infty, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    /* powr (inf, +-0) is a NaN.  */
+    TEST_ff_f (powr, plus_infty, 0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, plus_infty, minus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, minus_zero, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, -0.1L, 1.1L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -0.1L, -1.1L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -10.1L, 1.1L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -10.1L, -1.1L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.01L, min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.01L, -min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.0L, min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -1.0L, -min_subnorm_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, 0, -1, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, 0, -11, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, 0, -0xffffff, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+#if !TEST_COND_binary32
+    TEST_ff_f (powr, 0, -0x1.fffffffffffffp+52L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+#endif
+#if MANT_DIG >= 64
+    TEST_ff_f (powr, 0, -0x1.fffffffffffffffep+63L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+# if MANT_DIG >= 106
+    TEST_ff_f (powr, 0, -0x1.ffffffffffffffffffffffffff8p+105L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+# endif
+# if MANT_DIG >= 113
+    TEST_ff_f (powr, 0, -0x1.ffffffffffffffffffffffffffffp+112L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+# endif
+#endif
+    TEST_ff_f (powr, minus_zero, -1, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, minus_zero, -11L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, minus_zero, -0xffffff, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, minus_zero, -0x1fffffe, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+#if !TEST_COND_binary32
+    TEST_ff_f (powr, minus_zero, -0x1.fffffffffffffp+52L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, minus_zero, -0x1.fffffffffffffp+53L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+#endif
+#if MANT_DIG >= 64
+    TEST_ff_f (powr, minus_zero, -0x1.fffffffffffffffep+63L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, minus_zero, -0x1.fffffffffffffffep+64L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE|XFAIL_ROUNDING_IBM128_LIBGCC),
+# if MANT_DIG >= 106
+    TEST_ff_f (powr, minus_zero, -0x1.ffffffffffffffffffffffffff8p+105L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, minus_zero, -0x1.ffffffffffffffffffffffffff8p+106L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE|XFAIL_ROUNDING_IBM128_LIBGCC),
+# endif
+# if MANT_DIG >= 113
+    TEST_ff_f (powr, minus_zero, -0x1.ffffffffffffffffffffffffffffp+112L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, minus_zero, -0x1.ffffffffffffffffffffffffffffp+113L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+# endif
+#endif
+
+    TEST_ff_f (powr, 0, -2, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, 0, -11.1L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, 0, -min_subnorm_value, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, 0, -0x1p24, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, 0, -0x1p127, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, 0, -max_value, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, minus_zero, -2, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, minus_zero, -11.1L, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, minus_zero, -min_subnorm_value, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, minus_zero, -0x1p24, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, minus_zero, -0x1p127, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+    TEST_ff_f (powr, minus_zero, -max_value, plus_infty, DIVIDE_BY_ZERO_EXCEPTION|ERRNO_ERANGE),
+
+    TEST_ff_f (powr, 0, plus_infty, 0, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, minus_zero, plus_infty, 0, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, 0, minus_infty, plus_infty, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, minus_zero, minus_infty, plus_infty, ERRNO_UNCHANGED),
+
+    /* powr (x, +inf) == +inf for |x| > 1.  */
+    TEST_ff_f (powr, 1.5, plus_infty, plus_infty, ERRNO_UNCHANGED),
+
+    /* powr (x, +inf) == +0 for |x| < 1.  */
+    TEST_ff_f (powr, 0.5, plus_infty, 0.0, ERRNO_UNCHANGED),
+
+    /* powr (x, -inf) == +0 for |x| > 1.  */
+    TEST_ff_f (powr, 1.5, minus_infty, 0.0, ERRNO_UNCHANGED),
+
+    /* powr (x, -inf) == +inf for |x| < 1.  */
+    TEST_ff_f (powr, 0.5, minus_infty, plus_infty, ERRNO_UNCHANGED),
+
+    /* powr (+inf, y) == +inf for y > 0.  */
+    TEST_ff_f (powr, plus_infty, 2, plus_infty, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, 0xffffff, plus_infty, ERRNO_UNCHANGED),
+#if !TEST_COND_binary32
+    TEST_ff_f (powr, plus_infty, 0x1.fffffffffffffp+52L, plus_infty, ERRNO_UNCHANGED),
+#endif
+#if MANT_DIG >= 64
+    TEST_ff_f (powr, plus_infty, 0x1.fffffffffffffffep+63L, plus_infty, ERRNO_UNCHANGED),
+# if MANT_DIG >= 106
+    TEST_ff_f (powr, plus_infty, 0x1.ffffffffffffffffffffffffff8p+105L, plus_infty, ERRNO_UNCHANGED),
+# endif
+# if MANT_DIG >= 113
+    TEST_ff_f (powr, plus_infty, 0x1.ffffffffffffffffffffffffffffp+112L, plus_infty, ERRNO_UNCHANGED),
+# endif
+#endif
+    TEST_ff_f (powr, plus_infty, 0x1p24, plus_infty, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, 0x1p127, plus_infty, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, max_value, plus_infty, ERRNO_UNCHANGED),
+
+    /* powr (+inf, y) == +0 for y < 0.  */
+    TEST_ff_f (powr, plus_infty, -1, 0.0, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, -0xffffff, 0.0, ERRNO_UNCHANGED),
+#if !TEST_COND_binary32
+    TEST_ff_f (powr, plus_infty, -0x1.fffffffffffffp+52L, 0.0, ERRNO_UNCHANGED),
+#endif
+#if MANT_DIG >= 64
+    TEST_ff_f (powr, plus_infty, -0x1.fffffffffffffffep+63L, 0.0, ERRNO_UNCHANGED),
+# if MANT_DIG >= 106
+    TEST_ff_f (powr, plus_infty, -0x1.ffffffffffffffffffffffffff8p+105L, 0.0, ERRNO_UNCHANGED),
+# endif
+# if MANT_DIG >= 113
+    TEST_ff_f (powr, plus_infty, -0x1.ffffffffffffffffffffffffffffp+112L, 0.0, ERRNO_UNCHANGED),
+# endif
+#endif
+    TEST_ff_f (powr, plus_infty, -0x1p24, 0.0, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, -0x1p127, 0.0, ERRNO_UNCHANGED),
+    TEST_ff_f (powr, plus_infty, -max_value, 0.0, ERRNO_UNCHANGED),
+
+    /* powr (-inf, y) is a NaN (unlike pow) for y an odd integer > 0.  */
+    TEST_ff_f (powr, minus_infty, 27, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 0xffffff, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 0x1fffffe, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+#if !TEST_COND_binary32
+    TEST_ff_f (powr, minus_infty, 0x1.fffffffffffffp+52L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 0x1.fffffffffffffp+53L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+#endif
+#if MANT_DIG >= 64
+    TEST_ff_f (powr, minus_infty, 0x1.fffffffffffffffep+63L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 0x1.fffffffffffffffep+64L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM|XFAIL_ROUNDING_IBM128_LIBGCC),
+# if MANT_DIG >= 106
+    TEST_ff_f (powr, minus_infty, 0x1.ffffffffffffffffffffffffff8p+105L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 0x1.ffffffffffffffffffffffffff8p+106L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM|XFAIL_ROUNDING_IBM128_LIBGCC),
+# endif
+# if MANT_DIG >= 113
+    TEST_ff_f (powr, minus_infty, 0x1.ffffffffffffffffffffffffffffp+112L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 0x1.ffffffffffffffffffffffffffffp+113L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+# endif
+#endif
+
+    /* powr (-inf, y) is a NaN (unlike pow) for y > 0 and not an odd integer.  */
+    TEST_ff_f (powr, minus_infty, 0.5, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 28, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 0x1p24, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, 0x1p127, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, max_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    /* powr (-inf, y) is a NaN (unlike pow) for y an odd integer < 0. */
+    TEST_ff_f (powr, minus_infty, -3, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -0xffffff, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -0x1fffffe, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+#if !TEST_COND_binary32
+    TEST_ff_f (powr, minus_infty, -0x1.fffffffffffffp+52L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -0x1.fffffffffffffp+53L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+#endif
+#if MANT_DIG >= 64
+    TEST_ff_f (powr, minus_infty, -0x1.fffffffffffffffep+63L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -0x1.fffffffffffffffep+64L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM|XFAIL_ROUNDING_IBM128_LIBGCC),
+# if MANT_DIG >= 106
+    TEST_ff_f (powr, minus_infty, -0x1.ffffffffffffffffffffffffff8p+105L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -0x1.ffffffffffffffffffffffffff8p+106L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM|XFAIL_ROUNDING_IBM128_LIBGCC),
+# endif
+# if MANT_DIG >= 113
+    TEST_ff_f (powr, minus_infty, -0x1.ffffffffffffffffffffffffffffp+112L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -0x1.ffffffffffffffffffffffffffffp+113L, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+# endif
+#endif
+    /* powr (-inf, y) is a NaN (unlike pow) for y < 0 and not an odd integer.  */
+    TEST_ff_f (powr, minus_infty, -2.0, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -0x1p24, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -0x1p127, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, minus_infty, -max_value, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, -max_value, 0.5, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -max_value, 1.5, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -max_value, 1000.5, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+
+    TEST_ff_f (powr, -min_value, 0.5, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_value, 1.5, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    TEST_ff_f (powr, -min_value, 1000.5, qnan_value, INVALID_EXCEPTION|ERRNO_EDOM),
+    AUTO_TESTS_ff_f (powr),
+  };
+
+static void
+powr_test (void)
+{
+  ALL_RM_TEST (powr, 0, powr_test_data, RUN_TEST_LOOP_ff_f, END);
+}
+
+static void
+do_test (void)
+{
+  powr_test ();
+}
+
+/*
+ * Local Variables:
+ * mode:c
+ * End:
+ */
--- a/math/s_powr_template.c
+++ b/math/s_powr_template.c
@ -0,0 +1,50 @@
+/* Return X^Y, with special cases based on exp(Y*log(X)).
+   Copyright (C) 2025 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
+   modify it under the terms of the GNU Lesser General Public
+   License as published by the Free Software Foundation; either
+   version 2.1 of the License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   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, see
+   <https://www.gnu.org/licenses/>.  */
+
+#include <errno.h>
+#include <math.h>
+#include <math_private.h>
+
+FLOAT
+M_DECL_FUNC (__powr) (FLOAT x, FLOAT y)
+{
+  if (isless (x, M_LIT (0.0))
+      || (x == M_LIT (0.0) && y == M_LIT (0.0))
+      || (x == M_LIT (1.0) && isinf (y))
+      || (isinf (x) && y == M_LIT (0.0)))
+    {
+      __set_errno (EDOM);
+      return (x - x) / (x - x);
+    }
+  if (isnan (x) || isnan (y))
+    return x + y;
+  x = M_FABS (x);
+  FLOAT ret = M_SUF (__ieee754_pow) (x, y);
+  if (!isfinite (ret))
+    {
+      if (isfinite (x) && isfinite (y))
+	__set_errno (ERANGE);
+    }
+  else if (ret == 0
+	   && isfinite (x)
+	   && x != 0
+	   && isfinite (y))
+    __set_errno (ERANGE);
+  return ret;
+}
+declare_mgen_alias (__powr, powr);
--- a/math/test-tgmath.c
+++ b/math/test-tgmath.c
@ -48,7 +48,7 @@ volatile int count_cdouble;
 volatile int count_cfloat;
 volatile int count_cldouble;

-#define NCALLS     185
+#define NCALLS     188
 #define NCALLS_INT 4
 #define NCCALLS    47

@ -268,6 +268,7 @@ F(compile_test) (void)
  a = log10p1 (log10p1 (x));
  a = logp1 (logp1 (x));
  a = pow (pow (x, a), pow (c, b));
+  a = powr (powr (x, a), powr (c, b));
  b = sqrt (sqrt (a));
  a = rsqrt (rsqrt (b));
  a = hypot (hypot (x, b), hypot (c, a));
@ -394,6 +395,7 @@ F(compile_test) (void)
      a = log10p1 (y);
      a = logp1 (y);
      a = pow (y, y);
+      a = powr (y, y);
      a = sqrt (y);
      a = rsqrt (y);
      a = hypot (y, y);
@ -771,6 +773,14 @@ TYPE
  return x + y;
 }

+TYPE
+(F(powr)) (TYPE x, TYPE y)
+{
+  ++count;
+  P ();
+  return x + y;
+}
+
 TYPE
 (F(sqrt)) (TYPE x)
 {
--- a/math/tgmath.h
+++ b/math/tgmath.h
@ -924,6 +924,10 @@
 #define cbrt(Val) __TGMATH_UNARY_REAL_ONLY (Val, cbrt)

 #if __GLIBC_USE (IEC_60559_FUNCS_EXT_C23)
+/* Return X to the Y power.  */
+# define powr(Val1, Val2) __TGMATH_BINARY_REAL_ONLY (Val1, Val2, powr)
+
+/* Return 1/sqrt(X).  */
 # define rsqrt(Val) __TGMATH_UNARY_REAL_ONLY (Val, rsqrt)
 #endif

--- a/sysdeps/ieee754/ldbl-128ibm-compat/Versions
+++ b/sysdeps/ieee754/ldbl-128ibm-compat/Versions
@ -154,6 +154,7 @@ libm {
    __tanpiieee128;
  }
  GLIBC_2.42 {
+    __powrieee128;
    __rsqrtieee128;
  }
 }
--- a/sysdeps/ieee754/ldbl-opt/Makefile
+++ b/sysdeps/ieee754/ldbl-opt/Makefile
@ -166,6 +166,7 @@ libnldbl-calls = \
  obstack_vprintf \
  obstack_vprintf_chk \
  pow \
+  powr \
  printf \
  printf_chk \
  printf_fp \
@ -375,6 +376,7 @@ CFLAGS-nldbl-nexttoward.c = -fno-builtin-nexttoward -fno-builtin-nexttowardl
 CFLAGS-nldbl-nexttowardf.c = -fno-builtin-nexttowardf
 CFLAGS-nldbl-nextup.c = -fno-builtin-nextupl
 CFLAGS-nldbl-pow.c = -fno-builtin-powl
+CFLAGS-nldbl-powr.c = -fno-builtin-powrl
 CFLAGS-nldbl-remainder.c = -fno-builtin-remainderl -fno-builtin-dreml
 CFLAGS-nldbl-remquo.c = -fno-builtin-remquol
 CFLAGS-nldbl-rint.c = -fno-builtin-rintl
--- a/sysdeps/ieee754/ldbl-opt/nldbl-powr.c
+++ b/sysdeps/ieee754/ldbl-opt/nldbl-powr.c
@ -0,0 +1,8 @@
+#include "nldbl-compat.h"
+
+double
+attribute_hidden
+powrl (double x, double y)
+{
+  return powr (x, y);
+}
--- a/sysdeps/mach/hurd/i386/libm.abilist
+++ b/sysdeps/mach/hurd/i386/libm.abilist
@ -1277,6 +1277,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/mach/hurd/x86_64/libm.abilist
+++ b/sysdeps/mach/hurd/x86_64/libm.abilist
@ -1134,6 +1134,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/aarch64/libm.abilist
+++ b/sysdeps/unix/sysv/linux/aarch64/libm.abilist
@ -1245,6 +1245,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/alpha/libm.abilist
+++ b/sysdeps/unix/sysv/linux/alpha/libm.abilist
@ -1404,6 +1404,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/arc/libm.abilist
+++ b/sysdeps/unix/sysv/linux/arc/libm.abilist
@ -829,6 +829,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/arm/be/libm.abilist
+++ b/sysdeps/unix/sysv/linux/arm/be/libm.abilist
@ -920,6 +920,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/arm/le/libm.abilist
+++ b/sysdeps/unix/sysv/linux/arm/le/libm.abilist
@ -920,6 +920,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/csky/libm.abilist
+++ b/sysdeps/unix/sysv/linux/csky/libm.abilist
@ -895,6 +895,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/hppa/libm.abilist
+++ b/sysdeps/unix/sysv/linux/hppa/libm.abilist
@ -920,6 +920,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/i386/libm.abilist
+++ b/sysdeps/unix/sysv/linux/i386/libm.abilist
@ -1284,6 +1284,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/loongarch/lp64/libm.abilist
+++ b/sysdeps/unix/sysv/linux/loongarch/lp64/libm.abilist
@ -1124,6 +1124,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/m68k/coldfire/libm.abilist
+++ b/sysdeps/unix/sysv/linux/m68k/coldfire/libm.abilist
@ -920,6 +920,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/m68k/m680x0/libm.abilist
+++ b/sysdeps/unix/sysv/linux/m68k/m680x0/libm.abilist
@ -956,6 +956,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/microblaze/be/libm.abilist
+++ b/sysdeps/unix/sysv/linux/microblaze/be/libm.abilist
@ -920,6 +920,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/microblaze/le/libm.abilist
+++ b/sysdeps/unix/sysv/linux/microblaze/le/libm.abilist
@ -920,6 +920,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/mips/mips32/libm.abilist
+++ b/sysdeps/unix/sysv/linux/mips/mips32/libm.abilist
@ -920,6 +920,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/mips/mips64/libm.abilist
+++ b/sysdeps/unix/sysv/linux/mips/mips64/libm.abilist
@ -1245,6 +1245,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/or1k/libm.abilist
+++ b/sysdeps/unix/sysv/linux/or1k/libm.abilist
@ -829,6 +829,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/powerpc/powerpc32/fpu/libm.abilist
+++ b/sysdeps/unix/sysv/linux/powerpc/powerpc32/fpu/libm.abilist
@ -1067,6 +1067,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/powerpc/powerpc32/nofpu/libm.abilist
+++ b/sysdeps/unix/sysv/linux/powerpc/powerpc32/nofpu/libm.abilist
@ -1066,6 +1066,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/powerpc/powerpc64/be/libm.abilist
+++ b/sysdeps/unix/sysv/linux/powerpc/powerpc64/be/libm.abilist
@ -1060,6 +1060,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/powerpc/powerpc64/le/libm.abilist
+++ b/sysdeps/unix/sysv/linux/powerpc/powerpc64/le/libm.abilist
@ -1429,7 +1429,16 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 __powrieee128 F
 GLIBC_2.42 __rsqrtieee128 F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/riscv/rv32/libm.abilist
+++ b/sysdeps/unix/sysv/linux/riscv/rv32/libm.abilist
@ -1124,6 +1124,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/riscv/rv64/libm.abilist
+++ b/sysdeps/unix/sysv/linux/riscv/rv64/libm.abilist
@ -1221,6 +1221,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/s390/s390-32/libm.abilist
+++ b/sysdeps/unix/sysv/linux/s390/s390-32/libm.abilist
@ -1348,6 +1348,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/s390/s390-64/libm.abilist
+++ b/sysdeps/unix/sysv/linux/s390/s390-64/libm.abilist
@ -1348,6 +1348,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/sh/be/libm.abilist
+++ b/sysdeps/unix/sysv/linux/sh/be/libm.abilist
@ -920,6 +920,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/sh/le/libm.abilist
+++ b/sysdeps/unix/sysv/linux/sh/le/libm.abilist
@ -920,6 +920,12 @@ GLIBC_2.41 tanpif32 F
 GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf32 F
--- a/sysdeps/unix/sysv/linux/sparc/sparc32/libm.abilist
+++ b/sysdeps/unix/sysv/linux/sparc/sparc32/libm.abilist
@ -1355,6 +1355,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/sparc/sparc64/libm.abilist
+++ b/sysdeps/unix/sysv/linux/sparc/sparc64/libm.abilist
@ -1245,6 +1245,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/x86_64/64/libm.abilist
+++ b/sysdeps/unix/sysv/linux/x86_64/64/libm.abilist
@ -1278,6 +1278,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F
--- a/sysdeps/unix/sysv/linux/x86_64/x32/libm.abilist
+++ b/sysdeps/unix/sysv/linux/x86_64/x32/libm.abilist
@ -1278,6 +1278,14 @@ GLIBC_2.41 tanpif32x F
 GLIBC_2.41 tanpif64 F
 GLIBC_2.41 tanpif64x F
 GLIBC_2.41 tanpil F
+GLIBC_2.42 powr F
+GLIBC_2.42 powrf F
+GLIBC_2.42 powrf128 F
+GLIBC_2.42 powrf32 F
+GLIBC_2.42 powrf32x F
+GLIBC_2.42 powrf64 F
+GLIBC_2.42 powrf64x F
+GLIBC_2.42 powrl F
 GLIBC_2.42 rsqrt F
 GLIBC_2.42 rsqrtf F
 GLIBC_2.42 rsqrtf128 F