about summary refs log tree commit diff
path: root/math
Commit message (Collapse)AuthorAgeFilesLines
* Stop applying a GCC-specific workaround on clang [BZ #30550]Tulio Magno Quites Machado Filho2023-06-301-1/+2
| | | | | | | | | | GCC was the only compiler affected by the issue with __builtin_isinf_sign and float128. Fix BZ #30550. Reported-by: Qiu Chaofan <qiucofan@cn.ibm.com> Reviewed-by: Florian Weimer <fweimer@redhat.com>
* Fix all the remaining misspellings -- BZ 25337Paul Pluzhnikov2023-06-026-6/+6
|
* math: Remove the error handling wrapper from fmod and fmodfAdhemerval Zanella Netto2023-04-033-6/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The error handling is moved to sysdeps/ieee754 version with no SVID support. The compatibility symbol versions still use the wrapper with SVID error handling around the new code. There is no new symbol version nor compatibility code on !LIBM_SVID_COMPAT targets (e.g. riscv). The ia64 is unchanged, since it still uses the arch specific __libm_error_region on its implementation. For both i686 and m68k, which provive arch specific implementation, wrappers are added so no new symbol are added (which would require to change the implementations). It shows an small improvement, the results for fmod: Architecture | Input | master | patch -----------------|-----------------|----------|-------- x86_64 (Ryzen 9) | subnormals | 12.5049 | 9.40992 x86_64 (Ryzen 9) | normal | 296.939 | 296.738 x86_64 (Ryzen 9) | close-exponents | 16.0244 | 13.119 aarch64 (N1) | subnormal | 6.81778 | 4.33313 aarch64 (N1) | normal | 155.620 | 152.915 aarch64 (N1) | close-exponents | 8.21306 | 5.76138 armhf (N1) | subnormal | 15.1083 | 14.5746 armhf (N1) | normal | 244.833 | 241.738 armhf (N1) | close-exponents | 21.8182 | 22.457 Checked on x86_64-linux-gnu, i686-linux-gnu, and aarch64-linux-gnu. Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
* math: Improve fmodfAdhemerval Zanella Netto2023-04-031-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This uses a new algorithm similar to already proposed earlier [1]. With x = mx * 2^ex and y = my * 2^ey (mx, my, ex, ey being integers), the simplest implementation is: mx * 2^ex == 2 * mx * 2^(ex - 1) while (ex > ey) { mx *= 2; --ex; mx %= my; } With mx/my being mantissa of double floating pointer, on each step the argument reduction can be improved 8 (which is sizeof of uint32_t minus MANTISSA_WIDTH plus the signal bit): while (ex > ey) { mx << 8; ex -= 8; mx %= my; } */ The implementation uses builtin clz and ctz, along with shifts to convert hx/hy back to doubles. Different than the original patch, this path assume modulo/divide operation is slow, so use multiplication with invert values. I see the following performance improvements using fmod benchtests (result only show the 'mean' result): Architecture | Input | master | patch -----------------|-----------------|----------|-------- x86_64 (Ryzen 9) | subnormals | 17.2549 | 12.0318 x86_64 (Ryzen 9) | normal | 85.4096 | 49.9641 x86_64 (Ryzen 9) | close-exponents | 19.1072 | 15.8224 aarch64 (N1) | subnormal | 10.2182 | 6.81778 aarch64 (N1) | normal | 60.0616 | 20.3667 aarch64 (N1) | close-exponents | 11.5256 | 8.39685 I also see similar improvements on arm-linux-gnueabihf when running on the N1 aarch64 chips, where it a lot of soft-fp implementation (for modulo, and multiplication): Architecture | Input | master | patch -----------------|-----------------|----------|-------- armhf (N1) | subnormal | 11.6662 | 10.8955 armhf (N1) | normal | 69.2759 | 34.1524 armhf (N1) | close-exponents | 13.6472 | 18.2131 Instead of using the math_private.h definitions, I used the math_config.h instead which is used on newer math implementations. Co-authored-by: kirill <kirill.okhotnikov@gmail.com> [1] https://sourceware.org/pipermail/libc-alpha/2020-November/119794.html Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
* math: Improve fmodAdhemerval Zanella Netto2023-04-031-0/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This uses a new algorithm similar to already proposed earlier [1]. With x = mx * 2^ex and y = my * 2^ey (mx, my, ex, ey being integers), the simplest implementation is: mx * 2^ex == 2 * mx * 2^(ex - 1) while (ex > ey) { mx *= 2; --ex; mx %= my; } With mx/my being mantissa of double floating pointer, on each step the argument reduction can be improved 11 (which is sizeo of uint64_t minus MANTISSA_WIDTH plus the signal bit): while (ex > ey) { mx << 11; ex -= 11; mx %= my; } */ The implementation uses builtin clz and ctz, along with shifts to convert hx/hy back to doubles. Different than the original patch, this path assume modulo/divide operation is slow, so use multiplication with invert values. I see the following performance improvements using fmod benchtests (result only show the 'mean' result): Architecture | Input | master | patch -----------------|-----------------|----------|-------- x86_64 (Ryzen 9) | subnormals | 19.1584 | 12.5049 x86_64 (Ryzen 9) | normal | 1016.51 | 296.939 x86_64 (Ryzen 9) | close-exponents | 18.4428 | 16.0244 aarch64 (N1) | subnormal | 11.153 | 6.81778 aarch64 (N1) | normal | 528.649 | 155.62 aarch64 (N1) | close-exponents | 11.4517 | 8.21306 I also see similar improvements on arm-linux-gnueabihf when running on the N1 aarch64 chips, where it a lot of soft-fp implementation (for modulo, clz, ctz, and multiplication): Architecture | Input | master | patch -----------------|-----------------|----------|-------- armhf (N1) | subnormal | 15.908 | 15.1083 armhf (N1) | normal | 837.525 | 244.833 armhf (N1) | close-exponents | 16.2111 | 21.8182 Instead of using the math_private.h definitions, I used the math_config.h instead which is used on newer math implementations. Co-authored-by: kirill <kirill.okhotnikov@gmail.com> [1] https://sourceware.org/pipermail/libc-alpha/2020-November/119794.html Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
* update auto-libm-test-out-hypotPaul Zimmermann2023-02-141-0/+25
| | | | | This change was forgotten in commit cf7ffdd. Reviewed-by: Florian Weimer <fweimer@redhat.com>
* added pair of inputs for hypotf in binary32Paul Zimmermann2023-02-141-0/+2
| | | | | | | | | | | | | | This pair yields an error of 1 ulp in binary32, whereas the current maximal known error for hypotf on x86_64 is zero: Checking hypot with glibc-2.37 hypot 0 -1 -0x1.003222p-20,-0x1.6a2d58p-32 [0.501] 0.500001 0.500000001392678 libm gives 0x1.003224p-20 mpfr gives 0x1.003222p-20 See https://sourceware.org/pipermail/libc-alpha/2023-February/145432.html and https://sourceware.org/pipermail/libc-alpha/2023-February/145442.html Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* Update copyright dates with scripts/update-copyrightsJoseph Myers2023-01-06390-390/+390
|
* C2x semantics for <tgmath.h>Joseph Myers2023-01-062-162/+297
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | <tgmath.h> implements semantics for integer generic arguments that handle cases involving _FloatN / _FloatNx types as specified in TS 18661-3 plus some defect fixes. C2x has further changes to the semantics for <tgmath.h> macros with such types, which should also be considered defect fixes (although handled through the integration of TS 18661-3 in C2x rather than through an issue tracking process). Specifically, the rules were changed because of problems raised with using the macros with the evaluation format types such as float_t and _Float32_t: the older version of the rules didn't allow passing _FloatN / _FloatNx types to the narrowing macros returning float or double, or passing float / double / long double to the narrowing macros returning _FloatN / _FloatNx, which was a problem with the evaluation format types which could be either kind of type depending on the value of FLT_EVAL_METHOD. Thus the new rules allow cases of mixing types which were not allowed before, and, as part of the changes, the handling of integer arguments was also changed: if there is any _FloatNx generic argument, integer generic arguments are treated as _Float32x (not double), while the rule about treating integer arguments to narrowing macros returning _FloatN or _FloatNx as _Float64 not double was removed (no longer needed now double is a valid argument to such macros). I've implemented the changes in GCC's __builtin_tgmath, which thus requires updates to glibc's test expectations so that the tests continue to build with GCC 13 (the test is also updated to test the argument types that weren't allowed before but are now valid under C2x rules). Given those test changes, it's then also necessary to fix the implementations in <tgmath.h> to have appropriate semantics with older GCC so that the tests pass with GCC versions before GCC 13 as well. For some cases (non-narrowing macros with two or three generic arguments; narrowing macros returning _Float32x), the older version of __builtin_tgmath doesn't correspond sufficiently well to C2x semantics, so in those cases <tgmath.h> is adjusted to use the older macro implementation instead of __builtin_tgmath. The older macro implementation is itself adjusted to give the desired semantics, with GCC 7 and later. (It's not possible to get the right semantics in all cases for the narrowing macros with GCC 6 and before when the _FloatN / _FloatNx names are typedefs rather than distinct types.) Tested as follows: with the full glibc testsuite for x86_64, GCC 6, 7, 11, 13; with execution of the math/tests for aarch64, arm, powerpc and powerpc64le, GCC 6, 7, 12 and 13 (powerpc64le only with GCC 12 and 13); with build-many-glibcs.py with GCC 6, 7, 12 and 13.
* Disable use of -fsignaling-nans if compiler does not support itAdhemerval Zanella2022-11-019-10/+33
| | | | Reviewed-by: Fangrui Song <maskray@google.com>
* Fix build with GCC 13 _FloatN, _FloatNx built-in functionsJoseph Myers2022-10-311-7/+240
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | GCC 13 has added more _FloatN and _FloatNx versions of existing <math.h> and <complex.h> built-in functions, for use in libstdc++-v3. This breaks the glibc build because of how those functions are defined as aliases to functions with the same ABI but different types. Add appropriate -fno-builtin-* options for compiling relevant files, as already done for the case of long double functions aliasing double ones and based on the list of files used there. I fixed some mistakes in that list of double files that I noticed while implementing this fix, but there may well be more such (harmless) cases, in this list or the new one (files that don't actually exist or don't define the named functions as aliases so don't need the options). I did try to exclude cases where glibc doesn't define certain functions for _FloatN or _FloatNx types at all from the new uses of -fno-builtin-* options. As with the options for double files (see the commit message for commit 49348beafe9ba150c9bd48595b3f372299bddbb0, "Fix build with GCC 10 when long double = double."), it's deliberate that the options are used even if GCC currently doesn't have a built-in version of a given functions, so providing some level of future-proofing against more such built-in functions being added in future. Tested with build-many-glibcs.py for aarch64-linux-gnu powerpc-linux-gnu powerpc64le-linux-gnu x86_64-linux-gnu (compilers and glibcs builds) with GCC mainline.
* Fix iseqsig for _FloatN and _FloatNx in C++ with GCC 13Joseph Myers2022-09-302-1/+339
| | | | | | | | | | | | | | | | | | | | | | | | | | | With GCC 13, _FloatN and _FloatNx types, when they exist, are distinct types like they are in C with GCC 7 and later, rather than typedefs for types such as float, double or long double. This breaks the templated iseqsig implementation for C++ in <math.h>, when used with types that were previously implemented as aliases. Add the necessary definitions for _Float32, _Float64, _Float128 (when the same format as long double), _Float32x and _Float64x in this case, so that iseqsig can be used with such types in C++ with GCC 13 as it could with previous GCC versions. Also add tests for calling iseqsig in C++ with arguments of such types (more minimal than existing tests, so that they can work with older GCC versions and without relying on any C++ library support for the types or on hardcoding details of their formats). The LDBL_MANT_DIG != 106 conditionals on some tests are because the type-generic comparison macros have undefined behavior when neither argument has a type whose set of values is a subset of those for the type of the other argument, which applies when one argument is IBM long double and the other is an IEEE format wider than binary64. Tested with build-many-glibcs.py glibcs build for aarch64-linux-gnu i686-linux-gnu mips-linux-gnu mips64-linux-gnu-n32 powerpc-linux-gnu powerpc64le-linux-gnu x86_64-linux-gnu.
* Use '%z' instead of '%Z' on printf functionsAdhemerval Zanella Netto2022-09-221-1/+1
| | | | | | | | The Z modifier is a nonstandard synonymn for z (that predates z itself) and compiler might issue an warning for in invalid conversion specifier. Reviewed-by: Florian Weimer <fweimer@redhat.com>
* math: Add more input to atanh accuracy testsSunil K Pandey2022-02-242-0/+28
| | | | | | | | | | | This patch adds following input to atanh accuracy test. 0x1.f80094p-8 Tested on x86-64 and i686 platforms. Other platforms may have to regenerate ulps file. Reviewed-by: Paul Zimmermann <Paul.Zimmermann@inria.fr>
* math: Add more inputs to atan2 accuracy tests [BZ #28765]Sunil K Pandey2022-01-142-0/+796
| | | | | | | | | | | | | | | This patch adds following inputs: 0x1.bcab29da0e947p-54 0x1.bc41f4d2294b8p-54 0x1.a11891ec004d4p-348 0x1.814830510be26p-348 0x1.b836ed678be29p-588 0x1.b7be6f5a03a8cp-588 0x1.a83f842ef3f73p-633 0x1.a799d8a6677ep-633 to atan2 tests and updates x86_64 double atan2 ulps. This fixes BZ #28765. Reviewed-By: Paul Zimmermann <Paul.Zimmermann@inria.fr>
* math: Fix float conversion regressions with gcc-12 [BZ #28713]Szabolcs Nagy2022-01-107-10/+27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Converting double precision constants to float is now affected by the runtime dynamic rounding mode instead of being evaluated at compile time with default rounding mode (except static object initializers). This can change the computed result and cause performance regression. The known correctness issues (increased ulp errors) are already fixed, this patch fixes remaining cases of unnecessary runtime conversions. Add float M_* macros to math.h as new GNU extension API. To avoid conversions the new M_* macros are used and instead of casting double literals to float, use float literals (only required if the conversion is inexact). The patch was tested on aarch64 where the following symbols had new spurious conversion instructions that got fixed: __clog10f __gammaf_r_finite@GLIBC_2.17 __j0f_finite@GLIBC_2.17 __j1f_finite@GLIBC_2.17 __jnf_finite@GLIBC_2.17 __kernel_casinhf __lgamma_negf __log1pf __y0f_finite@GLIBC_2.17 __y1f_finite@GLIBC_2.17 cacosf cacoshf casinhf catanf catanhf clogf gammaf_positive Fixes bug 28713. Reviewed-by: Paul Zimmermann <Paul.Zimmermann@inria.fr>
* Update copyright dates with scripts/update-copyrightsPaul Eggert2022-01-01390-391/+391
| | | | | | | | | | | | | | | | | | | | | | | I used these shell commands: ../glibc/scripts/update-copyrights $PWD/../gnulib/build-aux/update-copyright (cd ../glibc && git commit -am"[this commit message]") and then ignored the output, which consisted lines saying "FOO: warning: copyright statement not found" for each of 7061 files FOO. I then removed trailing white space from math/tgmath.h, support/tst-support-open-dev-null-range.c, and sysdeps/x86_64/multiarch/strlen-vec.S, to work around the following obscure pre-commit check failure diagnostics from Savannah. I don't know why I run into these diagnostics whereas others evidently do not. remote: *** 912-#endif remote: *** 913: remote: *** 914- remote: *** error: lines with trailing whitespace found ... remote: *** error: sysdeps/unix/sysv/linux/statx_cp.c: trailing lines
* x86-64: Add vector tan/tanf implementation to libmvecSunil K Pandey2021-12-301-1/+1
| | | | | | | | Implement vectorized tan/tanf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector tan/tanf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector erfc/erfcf implementation to libmvecSunil K Pandey2021-12-301-1/+1
| | | | | | | | Implement vectorized erfc/erfcf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector erfc/erfcf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector asinh/asinhf implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized asinh/asinhf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector asinh/asinhf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector tanh/tanhf implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized tanh/tanhf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector tanh/tanhf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector erf/erff implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized erf/erff containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector erf/erff with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector acosh/acoshf implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized acosh/acoshf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector acosh/acoshf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector atanh/atanhf implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized atanh/atanhf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector atanh/atanhf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector log1p/log1pf implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized log1p/log1pf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector log1p/log1pf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector log2/log2f implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized log2/log2f containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector log2/log2f with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector log10/log10f implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized log10/log10f containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector log10/log10f with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector atan2/atan2f implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized atan2/atan2f containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector atan2/atan2f with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector cbrt/cbrtf implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized cbrt/cbrtf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector cbrt/cbrtf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector sinh/sinhf implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized sinh/sinhf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector sinh/sinhf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector expm1/expm1f implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized expm1/expm1f containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector expm1/expm1f with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector cosh/coshf implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized cosh/coshf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector cosh/coshf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector exp10/exp10f implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized exp10/exp10f containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector exp10/exp10f with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector exp2/exp2f implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized exp2/exp2f containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector exp2/exp2f with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector hypot/hypotf implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized hypot/hypotf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector hypot/hypotf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector asin/asinf implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized asin/asinf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector asin/asinf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86-64: Add vector atan/atanf implementation to libmvecSunil K Pandey2021-12-291-1/+1
| | | | | | | | Implement vectorized atan/atanf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector atan/atanf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* math: Properly cast X_TLOSS to float [BZ #28713]H.J. Lu2021-12-234-6/+11
| | | | | | | | | | | | | | | | | | | | | | | | Add #define AS_FLOAT_CONSTANT_1(x) x##f #define AS_FLOAT_CONSTANT(x) AS_FLOAT_CONSTANT_1(x) to cast X_TLOSS to float at compile-time to fix: FAIL: math/test-float-j0 FAIL: math/test-float-jn FAIL: math/test-float-y0 FAIL: math/test-float-y1 FAIL: math/test-float-yn FAIL: math/test-float32-j0 FAIL: math/test-float32-jn FAIL: math/test-float32-y0 FAIL: math/test-float32-y1 FAIL: math/test-float32-yn when compiling with GCC 12. Reviewed-by: Paul Zimmermann <Paul.Zimmermann@inria.fr>
* x86-64: Add vector acos/acosf implementation to libmvecSunil K Pandey2021-12-221-1/+1
| | | | | | | | Implement vectorized acos/acosf containing SSE, AVX, AVX2 and AVX512 versions for libmvec as per vector ABI. It also contains accuracy and ABI tests for vector acos/acosf with regenerated ulps. Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* math: Remove the error handling wrapper from hypot and hypotfAdhemerval Zanella2021-12-135-6/+31
| | | | | | | | | | | | The error handling is moved to sysdeps/ieee754 version with no SVID support. The compatibility symbol versions still use the wrapper with SVID error handling around the new code. There is no new symbol version nor compatibility code on !LIBM_SVID_COMPAT targets (e.g. riscv). Only ia64 is unchanged, since it still uses the arch specific __libm_error_region on its implementation. Checked on x86_64-linux-gnu, i686-linux-gnu, and aarch64-linux-gnu.
* math: Add math-use-builtinds-fmin.hAdhemerval Zanella2021-12-131-0/+5
| | | | | It allows the architecture to use the builtin instead of generic implementation.
* math: Add math-use-builtinds-fmax.hAdhemerval Zanella2021-12-131-0/+5
| | | | | It allows the architecture to use the builtin instead of generic implementation.
* math: Also xfail the new j0f tests for ibm128-libgccAdhemerval Zanella2021-10-062-1144/+1144
| | | | | | From commit 6bbf7298323bf31b. Checked on powerpc64-linux-gnu.
* Fixed inaccuracy of j0f (BZ #28185)Paul Zimmermann2021-10-052-0/+1145
| | | | | | | | | | | | | The largest errors over the full binary32 range are after this patch (on x86_64): RNDN: libm wrong by up to 9.00e+00 ulp(s) [9] for x=0x1.04c39cp+6 RNDZ: libm wrong by up to 9.00e+00 ulp(s) [9] for x=0x1.04c39cp+6 RNDU: libm wrong by up to 9.00e+00 ulp(s) [9] for x=0x1.04c39cp+6 RNDD: libm wrong by up to 8.98e+00 ulp(s) [9] for x=0x1.4b7066p+7 Inputs that were yielding huge errors have been added to "make check". Reviewed-by: Adhemeral Zanella <adhemerval.zanella@linaro.org>
* Add exp10 macro to <tgmath.h> (bug 26108)Joseph Myers2021-09-304-4/+21
| | | | | | | | | | | | | glibc has had exp10 functions since long before they were standardized; now they are standardized in TS 18661-4 and C2X, they are also specified there to have a corresponding type-generic macro. Add one to <tgmath.h>, so fixing bug 26108. glibc doesn't have other functions from TS 18661-4 yet, but when added, it will be natural to add the type-generic macro for each function family at the same time as the functions. Tested for x86_64.
* Do not declare fmax, fmin _FloatN, _FloatNx versions for C2XJoseph Myers2021-09-292-2/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | At the last WG14 meeting, <http://www.open-std.org/jtc1/sc22/wg14/www/docs/n2711.htm> was accepted, which places more emphasis on the new fmaximum / fminimum functions and less on the old fmax / fmin functions. Some of the changes are to examples, notes or otherwise don't require implementation changes. However, the changes include removing the _FloatN / _FloatNx versions of the fmax and fmin functions that came from TS 18661-3. Thus, those function versions should only be declared under similar conditions to the _FloatN / _FloatNx versions of fmaxmag and fminmag: for _GNU_SOURCE and pre-C2X use of __STDC_WANT_IEC_60559_TYPES_EXT__, but not for C2X without _GNU_SOURCE. In turn this requires a tgmath.h change so that the corresponding tgmath.h macros, for C2X with __STDC_WANT_IEC_60559_TYPES_EXT__ but without _GNU_SOURCE, don't try to use function variants that aren't declared. (That issue doesn't arise for the tgmath.h macros for fmaxmag and fminmag, because those aren't defined at all in those circumstances unless __STDC_WANT_IEC_60559_BFP_EXT__ (from TS 18661-1 and not specified at all by C2X) is also defined, and in that case the _FloatN / _FloatNx versions of fmaxmag and fminmag get declared - this is only ever an issue when it's possible for some functions corresponding to a type-generic-macro to be declared, and for _FloatN / _FloatNx functions in general to be declared, but without the _FloatN / _FloatNx functions corresponding to that particular macro being declared.) Tested for x86_64.
* Do not define tgmath.h fmaxmag, fminmag macros for C2X (bug 28397)Joseph Myers2021-09-291-0/+2
| | | | | | | | | | | | | C2X does not include fmaxmag and fminmag. When I updated feature test macro handling accordingly (commit 858045ad1c5ac1682288bbcb3676632b97a21ddf, "Update floating-point feature test macro handling for C2X", included in 2.34), I missed updating tgmath.h so it doesn't define the corresponding type-generic macros unless __STDC_WANT_IEC_60559_BFP_EXT__ is defined; I've now reported this as bug 28397. Adjust the conditionals in tgmath.h accordingly. Tested for x86_64.
* Add fmaximum, fminimum functionsJoseph Myers2021-09-2822-6/+1554
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | C2X adds new <math.h> functions for floating-point maximum and minimum, corresponding to the new operations that were added in IEEE 754-2019 because of concerns about the old operations not being associative in the presence of signaling NaNs. fmaximum and fminimum handle NaNs like most <math.h> functions (any NaN argument means the result is a quiet NaN). fmaximum_num and fminimum_num handle both quiet and signaling NaNs the way fmax and fmin handle quiet NaNs (if one argument is a number and the other is a NaN, return the number), but still raise "invalid" for a signaling NaN argument, making them exceptions to the normal rule that a function with a floating-point result raising "invalid" also returns a quiet NaN. fmaximum_mag, fminimum_mag, fmaximum_mag_num and fminimum_mag_num are corresponding functions returning the argument with greatest or least absolute value. All these functions also treat +0 as greater than -0. There are also corresponding <tgmath.h> type-generic macros. Add these functions to glibc. The implementations use type-generic templates based on those for fmax, fmin, fmaxmag and fminmag, and test inputs are based on those for those functions with appropriate adjustments to the expected results. The RISC-V maintainers might wish to add optimized versions of fmaximum_num and fminimum_num (for float and double), since RISC-V (F extension version 2.2 and later) provides instructions corresponding to those functions - though it might be at least as useful to add architecture-independent built-in functions to GCC and teach the RISC-V back end to expand those functions inline, which is what you generally want for functions that can be implemented with a single instruction. Tested for x86_64 and x86, and with build-many-glibcs.py.
* Add narrowing fma functionsJoseph Myers2021-09-2213-308/+36075
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch adds the narrowing fused multiply-add functions from TS 18661-1 / TS 18661-3 / C2X to glibc's libm: ffma, ffmal, dfmal, f32fmaf64, f32fmaf32x, f32xfmaf64 for all configurations; f32fmaf64x, f32fmaf128, f64fmaf64x, f64fmaf128, f32xfmaf64x, f32xfmaf128, f64xfmaf128 for configurations with _Float64x and _Float128; __f32fmaieee128 and __f64fmaieee128 aliases in the powerpc64le case (for calls to ffmal and dfmal when long double is IEEE binary128). Corresponding tgmath.h macro support is also added. The changes are mostly similar to those for the other narrowing functions previously added, especially that for sqrt, so the description of those generally applies to this patch as well. As with sqrt, I reused the same test inputs in auto-libm-test-in as for non-narrowing fma rather than adding extra or separate inputs for narrowing fma. The tests in libm-test-narrow-fma.inc also follow those for non-narrowing fma. The non-narrowing fma has a known bug (bug 6801) that it does not set errno on errors (overflow, underflow, Inf * 0, Inf - Inf). Rather than fixing this or having narrowing fma check for errors when non-narrowing does not (complicating the cases when narrowing fma can otherwise be an alias for a non-narrowing function), this patch does not attempt to check for errors from narrowing fma and set errno; the CHECK_NARROW_FMA macro is still present, but as a placeholder that does nothing, and this missing errno setting is considered to be covered by the existing bug rather than needing a separate open bug. missing-errno annotations are duly added to many of the auto-libm-test-in test inputs for fma. This completes adding all the new functions from TS 18661-1 to glibc, so will be followed by corresponding stdc-predef.h changes to define __STDC_IEC_60559_BFP__ and __STDC_IEC_60559_COMPLEX__, as the support for TS 18661-1 will be at a similar level to that for C standard floating-point facilities up to C11 (pragmas not implemented, but library functions done). (There are still further changes to be done to implement changes to the types of fromfp functions from N2548.) Tested as followed: natively with the full glibc testsuite for x86_64 (GCC 11, 7, 6) and x86 (GCC 11); with build-many-glibcs.py with GCC 11, 7 and 6; cross testing of math/ tests for powerpc64le, powerpc32 hard float, mips64 (all three ABIs, both hard and soft float). The different GCC versions are to cover the different cases in tgmath.h and tgmath.h tests properly (GCC 6 has _Float* only as typedefs in glibc headers, GCC 7 has proper _Float* support, GCC 8 adds __builtin_tgmath).
* Adjust new narrowing div/mul tests for IBM long double, update powerpc ULPsJoseph Myers2021-09-223-3664/+3664
| | | | | | Testing for powerpc shows some of the new narrowing div/mul tests need XFAILing for IBM long double and some ULPs updates are needed for those tests.