Remove the slow paths from pow. Like several other double precision math
functions, pow is exactly rounded. This is not required from math functions
and causes major overheads as it requires multiple fallbacks using higher
precision arithmetic if a result is close to 0.5ULP. Ridiculous slowdowns
of up to 100000x have been reported when the highest precision path triggers.
All GLIBC math tests pass on AArch64 and x64 (with ULP of pow set to 1).
The worst case error is ~0.506ULP. A simple test over a few hundred million
values shows pow is 10% faster on average. This fixes BZ #13932.
[BZ #13932]
* sysdeps/ieee754/dbl-64/uexp.h (err_1): Remove.
* benchtests/pow-inputs: Update comment for slow path cases.
* manual/probes.texi (slowpow_p10): Delete removed probe.
(slowpow_p10): Likewise.
* math/Makefile: Remove halfulp.c and slowpow.c.
* sysdeps/aarch64/libm-test-ulps: Set ULP of pow to 1.
* sysdeps/generic/math_private.h (__exp1): Remove error argument.
(__halfulp): Remove.
(__slowpow): Remove.
* sysdeps/i386/fpu/halfulp.c: Delete file.
* sysdeps/i386/fpu/slowpow.c: Likewise.
* sysdeps/ia64/fpu/halfulp.c: Likewise.
* sysdeps/ia64/fpu/slowpow.c: Likewise.
* sysdeps/ieee754/dbl-64/e_exp.c (__exp1): Remove error argument,
improve comments and add error analysis.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Add error analysis.
(power1): Remove function:
(log1): Remove error argument, add error analysis.
(my_log2): Remove function.
* sysdeps/ieee754/dbl-64/halfulp.c: Delete file.
* sysdeps/ieee754/dbl-64/slowpow.c: Likewise.
* sysdeps/m68k/m680x0/fpu/halfulp.c: Likewise.
* sysdeps/m68k/m680x0/fpu/slowpow.c: Likewise.
* sysdeps/powerpc/power4/fpu/Makefile: Remove CPPFLAGS-slowpow.c.
* sysdeps/x86_64/fpu/libm-test-ulps: Set ULP of pow to 1.
* sysdeps/x86_64/fpu/multiarch/Makefile: Remove slowpow-fma.c,
slowpow-fma4.c, halfulp-fma.c, halfulp-fma4.c.
* sysdeps/x86_64/fpu/multiarch/e_pow-fma.c (__slowpow): Remove define.
* sysdeps/x86_64/fpu/multiarch/e_pow-fma4.c (__slowpow): Likewise.
* sysdeps/x86_64/fpu/multiarch/halfulp-fma.c: Delete file.
* sysdeps/x86_64/fpu/multiarch/halfulp-fma4.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/slowpow-fma.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/slowpow-fma4.c: Likewise.
This patch adds the narrowing add functions from TS 18661-1 to glibc's
libm: fadd, faddl, daddl, f32addf64, f32addf32x, f32xaddf64 for all
configurations; f32addf64x, f32addf128, f64addf64x, f64addf128,
f32xaddf64x, f32xaddf128, f64xaddf128 for configurations with
_Float64x and _Float128; __nldbl_daddl for ldbl-opt. As discussed for
the build infrastructure patch, tgmath.h support is deliberately
deferred, and FP_FAST_* macros are not applicable without optimized
function implementations.
Function implementations are added for all relevant pairs of formats
(including certain cases of a format and itself where more than one
type has that format). The main implementations use round-to-odd, or
a trivial computation in the case where both formats are the same or
where the wider format is IBM long double (in which case we don't
attempt to be correctly rounding). The sysdeps/ieee754/soft-fp
implementations use soft-fp, and are used automatically for
configurations without exceptions and rounding modes by virtue of
existing Implies files. As previously discussed, optimized versions
for particular architectures are possible, but not included.
i386 gets a special version of f32xaddf64 to avoid problems with
double rounding (similar to the existing fdim version), since this
function must round just once without an intermediate rounding to long
double. (No such special version is needed for any other function,
because the nontrivial functions use round-to-odd, which does the
intermediate computation with the rounding mode set to round-to-zero,
and double rounding is OK except in round-to-nearest mode, so is OK
for that intermediate round-to-zero computation.) mul and div will
need slightly different special versions for i386 (using round-to-odd
on long double instead of precision control) because of the
possibility of inexact intermediate results in the subnormal range for
double.
To reduce duplication among the different function implementations,
math-narrow.h gets macros CHECK_NARROW_ADD, NARROW_ADD_ROUND_TO_ODD
and NARROW_ADD_TRIVIAL.
In the trivial cases and for any architecture-specific optimized
implementations, the overhead of the errno setting might be
significant, but I think that's best handled through compiler built-in
functions rather than providing separate no-errno versions in glibc
(and likewise there are no __*_finite entry points for these function
provided, __*_finite effectively being no-errno versions at present in
most cases).
Tested for x86_64 and x86, with both GCC 6 and GCC 7. Tested for
mips64 (all three ABIs, both hard and soft float) and powerpc with GCC
7. Tested with build-many-glibcs.py with both GCC 6 and GCC 7.
* math/Makefile (libm-narrow-fns): Add add.
(libm-test-funcs-narrow): Likewise.
* math/Versions (GLIBC_2.28): Add narrowing add functions.
* math/bits/mathcalls-narrow.h (add): Use __MATHCALL_NARROW .
* math/gen-auto-libm-tests.c (test_functions): Add add.
* math/math-narrow.h (CHECK_NARROW_ADD): New macro.
(NARROW_ADD_ROUND_TO_ODD): Likewise.
(NARROW_ADD_TRIVIAL): Likewise.
* sysdeps/ieee754/float128/float128_private.h (__faddl): New
macro.
(__daddl): Likewise.
* sysdeps/ieee754/ldbl-opt/Makefile (libnldbl-calls): Add fadd and
dadd.
(CFLAGS-nldbl-dadd.c): New variable.
(CFLAGS-nldbl-fadd.c): Likewise.
* sysdeps/ieee754/ldbl-opt/Versions (GLIBC_2.28): Add
__nldbl_daddl.
* sysdeps/ieee754/ldbl-opt/nldbl-compat.h (__nldbl_daddl): New
prototype.
* manual/arith.texi (Misc FP Arithmetic): Document fadd, faddl,
daddl, fMaddfN, fMaddfNx, fMxaddfN and fMxaddfNx.
* math/auto-libm-test-in: Add tests of add.
* math/auto-libm-test-out-narrow-add: New generated file.
* math/libm-test-narrow-add.inc: New file.
* sysdeps/i386/fpu/s_f32xaddf64.c: Likewise.
* sysdeps/ieee754/dbl-64/s_f32xaddf64.c: Likewise.
* sysdeps/ieee754/dbl-64/s_fadd.c: Likewise.
* sysdeps/ieee754/float128/s_f32addf128.c: Likewise.
* sysdeps/ieee754/float128/s_f64addf128.c: Likewise.
* sysdeps/ieee754/float128/s_f64xaddf128.c: Likewise.
* sysdeps/ieee754/ldbl-128/s_daddl.c: Likewise.
* sysdeps/ieee754/ldbl-128/s_f64xaddf128.c: Likewise.
* sysdeps/ieee754/ldbl-128/s_faddl.c: Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_daddl.c: Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_faddl.c: Likewise.
* sysdeps/ieee754/ldbl-96/s_daddl.c: Likewise.
* sysdeps/ieee754/ldbl-96/s_faddl.c: Likewise.
* sysdeps/ieee754/ldbl-opt/nldbl-dadd.c: Likewise.
* sysdeps/ieee754/ldbl-opt/nldbl-fadd.c: Likewise.
* sysdeps/ieee754/soft-fp/s_daddl.c: Likewise.
* sysdeps/ieee754/soft-fp/s_fadd.c: Likewise.
* sysdeps/ieee754/soft-fp/s_faddl.c: Likewise.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/mach/hurd/i386/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/aarch64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/alpha/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/arm/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/hppa/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/i386/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/ia64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/m68k/coldfire/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/m68k/m680x0/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/microblaze/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips32/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/nios2/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc32/fpu/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc32/nofpu/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc64/libm-le.abilist: Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/riscv/rv64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/s390/s390-32/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/s390/s390-64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/sh/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc32/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/tile/tilegx32/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/tile/tilegx64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/x86_64/64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/x86_64/x32/libm.abilist: Likewise.
Testing narrowing functions for x86_64 with GCC 6 showed up a further
testsuite fix needed: there is no _Float128 sNaN support before GCC 7
on x86_64 / x86, and the existing tests of SNAN_TESTS only checked it
for the return type, not for the argument type. This patch fixes the
code to check SNAN_TESTS (ARG_FLOAT) as well (in a variable set in
libm-test-driver.c, since libm-test-support.c is compiled only once
for each choice of FLOAT).
Tested for x86_64 and x86 with GCC 6 in conjunction with the main
patch adding narrowing add functions.
* math/libm-test-driver.c (snan_tests_arg): New variable.
* math/libm-test-support.h (snan_tests_arg): New declaration.
* math/libm-test-support.c (enable_test): Check snan_tests_arg.
This patch continues preparations for adding TS 18661-1 narrowing libm
functions by adding the required testsuite infrastructure to test such
functions through the libm-test infrastructure.
That infrastructure is based around testing for a single type, FLOAT.
For the narrowing functions, FLOAT, the "main" type for testing, is
the function return type; the argument type is ARG_FLOAT. This is
consistent with how the code built once for each type,
libm-test-support.c, depends on FLOAT for such things as calculating
ulps errors in results but can already handle different argument types
(pointers, integers, long double for nexttoward).
Makefile machinery is added to handle building tests for all pairs of
types for which there are narrowing functions (as with non-narrowing
functions, aliases are tested just the same as the functions they
alias). gen-auto-libm-tests gains a --narrow option for building
outputs for narrowing functions (so narrowing sqrt and fma will share
the same inputs as non-narrowing, but gen-auto-libm-tests will be run
with and without that option to generate different output files). In
the narrowing case, the auto-libm-test-out-narrow-* files include
annotations for each test about what properties ARG_FLOAT must have to
be able to represent all the inputs for that test; those annotations
result in calls to the TEST_COND_arg_fmt macro.
gen-libm-test.pl has some minor updates to handle narrowing tests (for
example, arguments in such tests must be surrounded by ARG_LIT calls
instead of LIT calls). Various new macros are added to the C test
support code (for example, sNaN initializers need to be properly
typed, so arg_snan_value is added; other such arg_* macros are added
as it seems cleanest to do so, though some are not strictly required).
Special-casing of the ibm128 format to allow for its limitations is
adjusted to handle it as the argument format as well as as the result
format; thus, the tests of the new functions allow nonzero ulps only
in the case where ibm128 is the argument format, as otherwise the
functions correspond to fully-defined IEEE operations. The ulps in
question appear as e.g. 'Function: "add_ldouble"' in libm-test-ulps
(with 1ulp errors then listed for double and float for that function
in powerpc); no support is added to generate corresponding faddl /
daddl ulps listings in the ulps table in the manual.
For the previous patch, I noted the need to avoid spurious macro
expansions of identifiers such as "add". A test test-narrow-macros.c
is added to verify such macro expansions are successfully avoided, and
there is also a -mlong-double-64 version of that test for ldbl-opt.
This test is set up to cover the full set of relevant identifiers from
the start rather than adding functions one at a time as each function
group is added.
Tested for x86_64 (this patch in isolation, as well as testing for
various configurations in conjunction with the actual addition of
"add" functions).
* math/Makefile (test-type-pairs): New variable.
(test-type-pairs-f64xf128-yes): Likewise.
(tests): Add test-narrow-macros.
(libm-test-funcs-narrow): New variable.
(libm-test-c-narrow): Likewise.
(generated): Add $(libm-test-c-narrow).
(libm-tests-base-narrow): New variable.
(libm-tests-narrow): Likewise.
(libm-tests): Add $(libm-tests-narrow).
(libm-tests-for-type): Handle $(libm-tests-narrow).
(libm-test-c-narrow-obj): New variable.
($(libm-test-c-narrow-obj)): New rule.
($(foreach t,$(libm-tests-narrow),$(objpfx)$(t).c)): Likewise.
($(foreach f,$(libm-test-funcs-narrow),$(objpfx)$(o)-$(f).o)): Use
$(o-iterator) to set dependencies and CFLAGS.
* math/gen-auto-libm-tests.c: Document use for narrowing
functions.
(output_for_one_input_case): Take argument NARROW.
(generate_output): Likewise. Update call to
output_for_one_input_case.
(main): Take --narrow option. Update call to generate_output.
* math/gen-libm-test.pl (_apply_lit): Take macro name as argument.
(apply_lit): Update call to _apply_lit.
(apply_arglit): New function.
(parse_args): Handle "a" arguments.
(parse_auto_input): Handle format names using ":".
* math/README.libm-test: Document "a" parameter type.
* math/libm-test-support.h (ARG_TYPE_MIN): New macro.
(ARG_TYPE_TRUE_MIN): Likewise.
(ARG_TYPE_MAX): Likwise.
(ARG_MIN_EXP): Likewise.
(ARG_MAX_EXP): Likewise.
(ARG_MANT_DIG): Likewise.
(TEST_COND_arg_ibm128): Likewise.
(TEST_COND_ibm128_libgcc): Define conditional on [ARG_FLOAT].
(TEST_COND_arg_fmt): New macro.
(init_max_error): Update prototype.
* math/libm-test-support.c (test_ibm128): New variable.
(init_max_error): Take argument testing_ibm128 and set test_ibm128
instead of using [TEST_COND_ibm128] conditional.
(test_exceptions): Use test_ibm128 instead of TEST_COND_ibm128.
* math/libm-test-driver.c (STR_ARG_FLOAT): New macro.
[TEST_NARROW] (TEST_MSG): New definition.
(arg_plus_zero): New macro.
(arg_minus_zero): Likewise.
(arg_plus_infty): Likewise.
(arg_minus_infty): Likewise.
(arg_qnan_value_pl): Likewise.
(arg_qnan_value): Likewise.
(arg_snan_value_pl): Likewise.
(arg_snan_value): Likewise.
(arg_max_value): Likewise.
(arg_min_value): Likewise.
(arg_min_subnorm_value): Likewise.
[ARG_FLOAT] (struct test_aa_f_data): New struct type.
(RUN_TEST_LOOP_aa_f): New macro.
(TEST_SUFF): New macro.
(TEST_SUFF_STR): Likewise.
[!TEST_MATHVEC] (VEC_SUFF): Don't define.
(TEST_COND_any_ibm128): New macro.
(START): Use TEST_SUFF and TEST_SUFF_STR in initializer for
this_func. Update call to init_max_error.
* math/test-double.h (FUNC_NARROW_PREFIX): New macro.
* math/test-float.h (FUNC_NARROW_PREFIX): Likewise.
* math/test-float128.h (FUNC_NARROW_PREFIX): Likewise.
* math/test-float32.h (FUNC_NARROW_PREFIX): Likewise.
* math/test-float32x.h (FUNC_NARROW_PREFIX): Likewise.
* math/test-float64.h (FUNC_NARROW_PREFIX): Likewise.
* math/test-float64x.h (FUNC_NARROW_PREFIX): Likewise.
* math/test-math-scalar.h (TEST_NARROW): Likewise.
* math/test-math-vector.h (TEST_NARROW): Likewise.
* math/test-arg-double.h: New file.
* math/test-arg-float128.h: Likewise.
* math/test-arg-float32x.h: Likewise.
* math/test-arg-float64.h: Likewise.
* math/test-arg-float64x.h: Likewise.
* math/test-arg-ldouble.h: Likewise.
* math/test-math-narrow.h: Likewise.
* math/test-narrow-macros.c: Likewise.
* sysdeps/ieee754/ldbl-opt/test-narrow-macros-ldbl-64.c: Likewise.
* sysdeps/ieee754/ldbl-opt/Makefile (tests): Add
test-narrow-macros-ldbl-64.
(CFLAGS-test-narrow-macros-ldbl-64.c): New variable.
TS 18661-1 defines libm functions that carry out an operation (+ - * /
sqrt fma) on their arguments and return a result rounded to a
(usually) narrower type, as if the original result were computed to
infinite precision and then rounded directly to the result type
without any intermediate rounding to the argument type. For example,
fadd, faddl and daddl for addition. These are the last remaining TS
18661-1 functions left to be added to glibc. TS 18661-3 extends this
to corresponding functions for _FloatN and _FloatNx types.
As functions parametrized by two rather than one varying
floating-point types, these functions require infrastructure in glibc
that was not required for previous libm functions. This patch
provides such infrastructure - excluding test support, and actual
function implementations, which will be in subsequent patches.
Declaring the functions uses a header bits/mathcalls-narrow.h, which
is included many times, for each relevant pair of types. This will
end up containing macro calls of the form
__MATHCALL_NARROW (__MATHCALL_NAME (add), __MATHCALL_REDIR_NAME (add), 2);
for each family of narrowing functions. (The structure of this macro
call, with the calls to __MATHCALL_NAME and __MATHCALL_REDIR_NAME
there rather than in the definition of __MATHCALL_NARROW, arises from
the names such as "add" *not* themselves being reserved identifiers -
meaning it's necessary to avoid any indirection that would result in a
user-defined "add" macro being expanded.) Whereas for existing
functions declaring long double functions is disabled if _LIBC in the
case where they alias double functions, to facilitate defining the
long double functions as aliases of the double ones, there is no such
logic for the narrowing functions in this patch. Rather, the files
defining such functions are expected to use #define to hide the
original declarations of the alias names, to avoid errors about
defining aliases with incompatible types.
math/Makefile support is added for building the functions (listed in
libm-narrow-fns, currently empty) for all relevant pairs of types. An
internal header math-narrow.h is added for macros shared between
multiple function implementations - currently a ROUND_TO_ODD macro to
facilitate writing functions using the round-to-odd implementation
approach, and alias macros to create all the required function
aliases. libc_feholdexcept_setroundf128 and libc_feupdateenv_testf128
are added for use when required (only for x86_64). float128_private.h
support is added for ldbl-128 narrowing functions to be used for
_Float128.
Certain things are specifically omitted from this patch and the
immediate followups. tgmath.h support is deferred; there remain
unresolved questions about how the type-generic macros for these
functions are supposed to work, especially in the case of arguments of
integer type. The math.h / bits/mathcalls-narrow.h logic, and the
logic for determining what functions / aliases to define, will need
some adjustments to support the sqrt and fma functions, where
e.g. f32xsqrtf64 can just be an alias for sqrt rather than a separate
function. TS 18661-1 defines FP_FAST_* macros but no support is
included for defining them (they won't in general be true without
architecture-specific optimized function versions).
For each of the function groups (add sub mul div sqrt fma) there are
always six functions present (e.g. fadd, faddl, daddl, f32addf64,
f32addf32x, f32xaddf64). When _Float64x and _Float128 are supported,
there are seven more (e.g. f32addf64x, f32addf128, f64addf64x,
f64addf128, f32xaddf64x, f32xaddf128, f64xaddf128). In addition, in
the ldbl-opt case there are function names such as __nldbl_daddl (an
alias for f32xaddf64, which is not a reserved name in TS 18661-1, only
in TS 18661-3), for calls to daddl to be mapped to in the
-mlong-double-64 case. (Calls to faddl just get mapped to fadd, and
for sqrt and fma there won't be __nldbl_* functions because dsqrtl and
dfmal can just be mapped to sqrt and fma with -mlong-double-64.)
While there are six or thirteen functions present in each group (plus
__nldbl_* names only as an ABI, not an API), not all are distinct;
they fall in various groups of aliases. There are two distinct
versions built if long double has the same format as double; four if
they have distinct formats but there is no _Float64x or _Float128
support; five if long double has binary128 format; seven when
_Float128 is distinct from long double.
Architecture-specific optimized versions are possible, but not
included in my patches. For example, IA64 generally supports
narrowing the result of most floating-point instructions; Power ISA
2.07 (POWER8) supports double values as arguments to float
instructions, with the results narrowed as expected; Power ISA 3
(POWER9) supports round-to-odd for float128 instructions, so meaning
that approach can be used without needing to set and restore the
rounding mode and test "inexact". I intend to leave any such
optimized versions to the architecture maintainers. Generally in such
cases it would also make sense for calls to these functions to be
expanded inline (given -fno-math-errno); I put a suggestion for TS
18661-1 built-in functions at <https://gcc.gnu.org/wiki/SummerOfCode>.
Tested for x86_64 (this patch in isolation, as well as testing for
various configurations in conjunction with further patches).
* math/bits/mathcalls-narrow.h: New file.
* include/bits/mathcalls-narrow.h: Likewise.
* math/math-narrow.h: Likewise.
* math/math.h (__MATHCALL_NARROW_ARGS_1): New macro.
(__MATHCALL_NARROW_ARGS_2): Likewise.
(__MATHCALL_NARROW_ARGS_3): Likewise.
(__MATHCALL_NARROW_NORMAL): Likewise.
(__MATHCALL_NARROW_REDIR): Likewise.
(__MATHCALL_NARROW): Likewise.
[__GLIBC_USE (IEC_60559_BFP_EXT)]: Repeatedly include
<bits/mathcalls-narrow.h> with _Mret_, _Marg_ and __MATHCALL_NAME
defined.
[__GLIBC_USE (IEC_60559_TYPES_EXT)]: Likewise.
* math/Makefile (headers): Add bits/mathcalls-narrow.h.
(libm-narrow-fns): New variable.
(libm-narrow-types-basic): Likewise.
(libm-narrow-types-ldouble-yes): Likewise.
(libm-narrow-types-float128-yes): Likewise.
(libm-narrow-types-float128-alias-yes): Likewise.
(libm-narrow-types): Likewise.
(libm-routines): Add narrowing functions.
* sysdeps/i386/fpu/fenv_private.h [__x86_64__]
(libc_feholdexcept_setroundf128): New macro.
[__x86_64__] (libc_feupdateenv_testf128): Likewise.
* sysdeps/ieee754/float128/float128_private.h: Include
<math/math-narrow.h>.
[libc_feholdexcept_setroundf128] (libc_feholdexcept_setroundl):
Undefine and redefine.
[libc_feupdateenv_testf128] (libc_feupdateenv_testl): Likewise.
(libm_alias_float_ldouble): Undefine and redefine.
(libm_alias_double_ldouble): Likewise.
The math/Makefile variable libm-test-incs was formerly used, but no
longer is. This patch removes it.
Tested for x86_64.
* math/Makefile [$(PERL) != no] (libm-test-incs): Remove variable.
Currently math_errhandling is always set to MATH_ERRNO | MATH_ERREXCEPT
even if -fno-math-errno is used. It is not defined at all when fast-math
is used. Set it to 0 with fast-math - this is noncomforming but more
useful than not define math_errhandling at all. Also take __NO_MATH_ERRNO__
into account and update comment.
* math/math.h (math_errhandling): Set to 0 with __FAST_MATH__.
Add __NO_MATH_ERRNO__ check.
I found that "make regen-ulps" failed when building with unmodified
GNU make 4.1, and an objdir /some/where/math/ longer than about 37
characters, because the list of tests in the "for run in $^" loop
exceeded the Linux kernel's MAX_ARG_STRLEN limit (131072 bytes) on the
length of a single argument passed to a command.
Some GNU/Linux distributions have a patch to make to work around this
limit (see e.g. Debian bug 688601), but clearly this ought to work
without needing such a patch. This patch arranges for the shell loop
to be over the test names without a $(objdir) prefix, which reduces
the space used to less than half MAX_ARG_STRLEN.
(I think we ought to aim to get rid of bits/mathinline.h completely -
filing GCC bugs for any optimizations GCC can't currently do with
-ffast-math - which would mean we could halve the number of libm tests
run because separate inline function tests would no longer be needed.
However, with a long directory name even half the number of tests
could make this command exceed MAX_ARG_STRLEN without my patch.)
Tested regen-ulps on a system where it failed before this patch.
* math/Makefile (run-regen-ulps): Add $(objpfx) to test name here.
(regen-ulps): Use $(libm-tests) not $^ in shell loop.
In C++ mode, __MATH_TG cannot be used for defining iseqsig, because
__MATH_TG relies on __builtin_types_compatible_p, which is a C-only
builtin. This is true when float128 is provided as an ABI-distinct type
from long double.
Moreover, the comparison macros from ISO C take two floating-point
arguments, which need not have the same type. Choosing what underlying
function to call requires evaluating the formats of the arguments, then
selecting which is wider. The macro __MATH_EVAL_FMT2 provides this
information, however, only the type of the macro expansion is relevant
(actually evaluating the expression would be incorrect).
This patch provides a C++ version of iseqsig, in which only the type of
__MATH_EVAL_FMT2 (__typeof or decltype) is used as a template parameter
for __iseqsig_type. This function calls the appropriate underlying
function.
Tested for powerpc64le and x86_64.
[BZ #22377]
* math/Makefile [C++] (tests): Add test for iseqsig.
* math/math.h [C++] (iseqsig): New implementation, which does
not rely on __MATH_TG/__builtin_types_compatible_p.
* math/test-math-iseqsig.cc: New file.
* sysdeps/powerpc/powerpc64le/Makefile
(CFLAGS-test-math-iseqsig.cc): New variable.
These changes will be active for all platforms that don't provide
their own exp() routines. They will also be active for ieee754
versions of ccos, ccosh, cosh, csin, csinh, sinh, exp10, gamma, and
erf.
Typical performance gains is typically around 5x when measured on
Sparc s7 for common values between exp(1) and exp(40).
Using the glibc perf tests on sparc,
sparc (nsec) x86 (nsec)
old new old new
max 17629 395 5173 144
min 399 54 15 13
mean 5317 200 1349 23
The extreme max times for the old (ieee754) exp are due to the
multiprecision computation in the old algorithm when the true value is
very near 0.5 ulp away from an value representable in double
precision. The new algorithm does not take special measures for those
cases. The current glibc exp perf tests overrepresent those values.
Informal testing suggests approximately one in 200 cases might
invoke the high cost computation. The performance advantage of the new
algorithm for other values is still large but not as large as indicated
by the chart above.
Glibc correctness tests for exp() and expf() were run. Within the
test suite 3 input values were found to cause 1 bit differences (ulp)
when "FE_TONEAREST" rounding mode is set. No differences in exp() were
seen for the tested values for the other rounding modes.
Typical example:
exp(-0x1.760cd2p+0) (-1.46113312244415283203125)
new code: 2.31973271630014299393707e-01 0x1.db14cd799387ap-3
old code: 2.31973271630014271638132e-01 0x1.db14cd7993879p-3
exp = 2.31973271630014285508337 (high precision)
Old delta: off by 0.49 ulp
New delta: off by 0.51 ulp
In addition, because ieee754_exp() is used by other routines, cexp()
showed test results with very small imaginary input values where the
imaginary portion of the result was off by 3 ulp when in upward
rounding mode, but not in the other rounding modes. For x86, tgamma
showed a few values where the ulp increased to 6 (max ulp for tgamma
is 5). Sparc tgamma did not show these failures. I presume the tgamma
differences are due to compiler optimization differences within the
gamma function.The gamma function is known to be difficult to compute
accurately.
* sysdeps/ieee754/dbl-64/e_exp.c: Include <math-svid-compat.h> and
<errno.h>. Include "eexp.tbl".
(half): New constant.
(one): Likewise.
(__ieee754_exp): Rewrite.
(__slowexp): Remove prototype.
* sysdeps/ieee754/dbl-64/eexp.tbl: New file.
* sysdeps/ieee754/dbl-64/slowexp.c: Remove file.
* sysdeps/i386/fpu/slowexp.c: Likewise.
* sysdeps/ia64/fpu/slowexp.c: Likewise.
* sysdeps/m68k/m680x0/fpu/slowexp.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/slowexp-avx.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/slowexp-fma.c: Likewise.
* sysdeps/x86_64/fpu/multiarch/slowexp-fma4.c: Likewise.
* sysdeps/generic/math_private.h (__slowexp): Remove prototype.
* sysdeps/ieee754/dbl-64/e_pow.c: Remove mention of slowexp.c in
comment.
* sysdeps/powerpc/power4/fpu/Makefile [$(subdir) = math]
(CPPFLAGS-slowexp.c): Remove variable.
* sysdeps/x86_64/fpu/multiarch/Makefile (libm-sysdep_routines):
Remove slowexp-fma, slowexp-fma4 and slowexp-avx.
(CFLAGS-slowexp-fma.c): Remove variable.
(CFLAGS-slowexp-fma4.c): Likewise.
(CFLAGS-slowexp-avx.c): Likewise.
* sysdeps/x86_64/fpu/multiarch/e_exp-avx.c (__slowexp): Do not
define as macro.
* sysdeps/x86_64/fpu/multiarch/e_exp-fma.c (__slowexp): Likewise.
* sysdeps/x86_64/fpu/multiarch/e_exp-fma4.c (__slowexp): Likewise.
* math/Makefile (type-double-routines): Remove slowexp.
* manual/probes.texi (slowexp_p6): Remove.
(slowexp_p32): Likewise.
These functions set errno and thus the const attribute was wrong.
[BZ #22593]
* math/bits/mathcalls.h (nextafter): Remove const.
(nexttoward): Likewise.
As per C11 DR#471, ctanh (0 + i NaN) and ctanh (0 + i Inf) should
return 0 + i NaN (with "invalid" exception in the second case but not
the first), not NaN + i NaN. This has corresponding implications for
ctan since its special cases are defined by ctan (z) = -i ctanh (iz).
This patch implements these cases for ctanh and ctan, updating
tests accordingly.
Tested for x86_64.
[BZ #22568]
* math/s_ctan_template.c (M_DECL_FUNC (__ctan)): Set imaginary
part of result to imaginary part of argument if it is zero and the
real part of the argument is not finite.
* math/s_ctanh_template.c (M_DECL_FUNC (__ctanh)): Set real part
of result to real part of argument if it is zero and the imaginary
part of the argument is not finite.
This patch adds the header required for testing _Float32 function
aliases, using float ulps. The corresponding makefile support will be
included in the main patch that enables those aliases.
In conjunction with other _Float32 changes, tested for x86_64 and with
build-many-glibcs.py.
* math/test-float32.h: New file.
As per C11 DR#471 (adjusted resolution accepted for C17), cacosh (0 +
iNaN) should return NaN +/- i pi/2, not NaN + iNaN.
This patch fixes the code accordingly. The test has hardcoded the
result with positive sign of the imaginary part (with an associated
comment), since the unspecified sign for a result other than 0 or
infinity isn't currently supported by the test infrastructure.
Tested for x86_64.
[BZ #22561]
* math/s_cacosh_template.c (M_DECL_FUNC (__cacosh)): Use pi/2 for
real part of result for argument 0 + i * NaN.
* math/libm-test-cacosh.inc (cacosh_test_data): Update expected
results for tests of 0 + i * NaN.
This patch continues filling out TS 18661-3 support by adding *f64 and
*f32x function aliases, supporting _Float64 and _Float32x, as aliases
for double functions. These types are supported for all glibc
configurations. The API corresponds exactly to that for _Float128 and
_Float64x. _Float32 aliases to float functions remain to be added in
subsequent patches to complete this process (then there are a few
miscellaneous functions in TS 18661-3 to implement that aren't simply
versions of existing functions for new types).
The patch enables the feature in bits/floatn-common.h, adds symbol
versions and documentation with updates to ABI baselines, and arranges
for the libm functions for the new types to be tested. As with the
_Float64x changes there are some x86 ulps updates because of header
inlines not used for the new types (and one other change to the
non-multiarch libm-test-ulps, which I suppose comes from using a
different compiler version / configuration from when it was last
regenerated).
Tested for x86_64 and x86, and with build-many-glibcs.py, with both
GCC 6 and GCC 7.
* bits/floatn-common.h (__HAVE_FLOAT64): Define to 1.
(__HAVE_FLOAT32X): Likewise.
* manual/math.texi (Mathematics): Document support for _Float64
and _Float32x.
* math/Makefile (test-types): Add float64 and float32x.
* math/Versions (GLIBC_2.27): Add _Float64 and _Float32x
functions.
* stdlib/Versions (GLIBC_2.27): Likewise.
* wcsmbs/Versions (GLIBC_2.27): Likewise.
* sysdeps/unix/sysv/linux/aarch64/libc.abilist: Update.
* sysdeps/unix/sysv/linux/aarch64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/alpha/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/alpha/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/arm/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/arm/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/hppa/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/hppa/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/i386/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/i386/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/ia64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/ia64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/m68k/coldfire/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/m68k/coldfire/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/m68k/m680x0/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/m68k/m680x0/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/microblaze/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/microblaze/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips32/fpu/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips32/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips32/nofpu/libc.abilist:
Likewise.
* sysdeps/unix/sysv/linux/mips/mips64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips64/n32/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips64/n64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/nios2/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/nios2/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc32/fpu/libc.abilist:
Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc32/fpu/libm.abilist:
Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc32/nofpu/libc.abilist:
Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc32/nofpu/libm.abilist:
Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc64/libc-le.abilist:
Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc64/libc.abilist:
Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc64/libm-le.abilist:
Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc64/libm.abilist:
Likewise.
* sysdeps/unix/sysv/linux/s390/s390-32/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/s390/s390-32/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/s390/s390-64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/s390/s390-64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/sh/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/sh/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc32/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc32/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/tile/tilegx/tilegx32/libc.abilist:
Likewise.
* sysdeps/unix/sysv/linux/tile/tilegx/tilegx32/libm.abilist:
Likewise.
* sysdeps/unix/sysv/linux/tile/tilegx/tilegx64/libc.abilist:
Likewise.
* sysdeps/unix/sysv/linux/tile/tilegx/tilegx64/libm.abilist:
Likewise.
* sysdeps/unix/sysv/linux/tile/tilepro/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/tile/tilepro/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/x86_64/64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/x86_64/64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/x86_64/x32/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/x86_64/x32/libm.abilist: Likewise.
* sysdeps/i386/fpu/libm-test-ulps: Likewise.
* sysdeps/i386/i686/fpu/multiarch/libm-test-ulps: Likewise.
This patch adds the headers required for testing _Float64 and
_Float32x function aliases (using double ulps). The corresponding
makefile support will be included in the patch that actually adds
those aliases; there doesn't seem much point in adding makefile
conditionals for testing something that will be available
unconditionally.
In conjunction with other _Float64 / _Float32x changes, test for
x86_64 and with build-many-glibcs.py.
* math/test-float32x.h: New file.
* math/test-float64.h: Likewise.
This patch continues filling out TS 18661-3 support by adding *f64x
function aliases on platforms with _Float64x support. (It so happens
the set of such platforms is exactly the same as the set of platforms
with _Float128 support, although on x86_64, x86 and ia32 the _Float64x
format is Intel extended rather than binary128.) The API provided
corresponds exactly to that provided for _Float128, mostly coming from
TS 18661-3. As these functions always alias those for another type
(long double, _Float128 or both), __* function names are not provided,
as in other cases of alias types.
Given the preparation done in previous patches, this one just enables
the feature via Makeconfig and bits/floatn.h, adds symbol versions,
and updates documentation and ABI baselines. The symbol versions are
present unconditionally as GLIBC_2.27 in the relevant Versions files,
as it's OK for those to specify versions for functions that may not be
present in some configurations; no additional complexity is needed
unless in future some configuration gains support for this type that
didn't have such support in 2.27. The Makeconfig additions for ia64
and x86 aren't strictly needed, as those configurations also get
float64x-alias-fcts definitions from
sysdeps/ieee754/float128/Makeconfig, but still seem appropriate given
that _Float64x is not _Float128 for those configurations.
A libm-test-ulps update for x86 is included. This is because
bits/mathinline.h does not have _Float64x support added and for two
functions the use of out-of-line functions results in increased ulps
(ifloat64x shares ulps with ildouble / ifloat128 as appropriate).
Given that we'd like generally to eliminate bits/mathinline.h
optimizations, preferring to have such optimizations in GCC instead,
it seems reasonable not to add such support there for new types. GCC
support for _FloatN / _FloatNx built-in functions is limited, but has
been improved in GCC 8, and at some point I hope the full set of libm
built-in functions in GCC, and other optimizations with
per-floating-type aspects, will be enabled for all _FloatN / _FloatNx
types.
Tested for x86_64 and x86, and with build-many-glibcs.py, with both
GCC 6 and GCC 7.
* sysdeps/ia64/Makeconfig (float64x-alias-fcts): New variable.
* sysdeps/ieee754/float128/Makeconfig (float64x-alias-fcts):
Likewise.
* sysdeps/ieee754/ldbl-128/Makeconfig (float64x-alias-fcts):
Likewise.
* sysdeps/x86/Makeconfig: New file.
* bits/floatn-common.h (__HAVE_FLOAT64X): Remove macro.
(__HAVE_FLOAT64X_LONG_DOUBLE): Likewise.
* bits/floatn.h (__HAVE_FLOAT64X): New macro.
(__HAVE_FLOAT64X_LONG_DOUBLE): Likewise.
* sysdeps/ia64/bits/floatn.h (__HAVE_FLOAT64X): Likewise.
(__HAVE_FLOAT64X_LONG_DOUBLE): Likewise.
* sysdeps/ieee754/ldbl-128/bits/floatn.h (__HAVE_FLOAT64X):
Likewise.
(__HAVE_FLOAT64X_LONG_DOUBLE): Likewise.
* sysdeps/mips/ieee754/bits/floatn.h (__HAVE_FLOAT64X): Likewise.
(__HAVE_FLOAT64X_LONG_DOUBLE): Likewise.
* sysdeps/powerpc/bits/floatn.h (__HAVE_FLOAT64X): Likewise.
(__HAVE_FLOAT64X_LONG_DOUBLE): Likewise.
* sysdeps/x86/bits/floatn.h (__HAVE_FLOAT64X): Likewise.
(__HAVE_FLOAT64X_LONG_DOUBLE): Likewise.
* manual/math.texi (Mathematics): Document support for _Float64x.
* math/Versions (GLIBC_2.27): Add _Float64x functions.
* stdlib/Versions (GLIBC_2.27): Likewise.
* wcsmbs/Versions (GLIBC_2.27): Likewise.
* sysdeps/unix/sysv/linux/aarch64/libc.abilist: Update.
* sysdeps/unix/sysv/linux/aarch64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/alpha/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/alpha/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/i386/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/i386/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/ia64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/ia64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips64/n32/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/mips/mips64/n64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc64/libc-le.abilist:
Likewise.
* sysdeps/unix/sysv/linux/powerpc/powerpc64/libm-le.abilist:
Likewise.
* sysdeps/unix/sysv/linux/s390/s390-32/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/s390/s390-32/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/s390/s390-64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/s390/s390-64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc32/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc32/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/sparc/sparc64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/x86_64/64/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/x86_64/64/libm.abilist: Likewise.
* sysdeps/unix/sysv/linux/x86_64/x32/libc.abilist: Likewise.
* sysdeps/unix/sysv/linux/x86_64/x32/libm.abilist: Likewise.
* sysdeps/i386/fpu/libm-test-ulps: Likewise.
* sysdeps/i386/i686/fpu/multiarch/libm-test-ulps: Likewise.
This patch adds support for testing _Float64x libm functions. A
configuration with such functions sets float64x-alias-fcts = yes in a
sysdeps Makeconfig file; until such settings are added, this test
support is inactive.
Tested for x86_64, including in conjunction with _Float64x support
patches.
* math/test-float64x.h: New file.
* math/Makefile (type-float64x-yes): New variable.
(test-types): Add $(type-float64x-$(float64x-alias-fcts)).
math/gen-tgmath-tests.py was missing a create_type argument when
creating the internal types for combinations of long double with
_Float64 and _Float64x, so resulting in output that did not compile
when glibc support for those types was enabled. This patch adds the
missing argument so that the tests properly compile in that case.
Tested for x86_64, including in conjunction with _Float64x support
patches.
* math/gen-tgmath-tests.py (Type.init_types): Pass suffix argument
for combinations of long double with _Float64 and _Float64x.
GCC mainline now supports __builtin_tgmath to allow <tgmath.h> macro
implementations that expand their arguments only once, so avoiding
exponential blowup in the size of macro expansions when calls to those
macros are nested in arguments to those macros.
This patch makes glibc's tgmath.h support using __builtin_tgmath, as a
much simpler and more efficient alternative to the existing
implementation. (As a side effect, the new feature would make it much
more practical to support decimal floating point in <tgmath.h> with
new compilers; currently, libdfp does not provide a <tgmath.h>
implementation, and making decimal arguments cause integer arguments
to be considered of type _Decimal64 instead of double would have been
very problematic in the old implementation.)
Tested for x86_64 (with GCC mainline).
[BZ #21660]
* math/tgmath.h (__HAVE_BUILTIN_TGMATH): New macro.
[__HAVE_BUILTIN_TGMATH] (__TG_F16_ARG): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TG_F32_ARG): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TG_F64_ARG): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TG_F128_ARG): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TG_F32X_ARG): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TG_F64X_ARG): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TG_F128X_ARG): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_FUNCS): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_RCFUNCS): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_1): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_2): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_2STD): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_3): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_1C): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_2C): Likewise.
(__tgml): Make conditional on [!__HAVE_BUILTIN_TGMATH].
(__floating_type): Likewise.
(__real_integer_type): Likewise.
(__complex_integer_type): Likewise.
(__expr_is_real): Likewise.
(__tgmath_real_type_sub): Likewise.
(__tgmath_real_type): Likewise.
(__tgmath_complex_type_sub): Likewise.
(__tgmath_complex_type): Likewise.
(__TGMATH_F128): Likewise.
(__TGMATH_CF128): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_UNARY_REAL_ONLY): Define using
new macros.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_UNARY_REAL_RET_ONLY): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_BINARY_FIRST_REAL_ONLY):
Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_BINARY_FIRST_REAL_STD_ONLY):
Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_BINARY_REAL_ONLY): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_BINARY_REAL_STD_ONLY): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_BINARY_REAL_RET_ONLY): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_TERNARY_FIRST_SECOND_REAL_ONLY):
Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_TERNARY_REAL_ONLY): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_TERNARY_FIRST_REAL_RET_ONLY):
Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_UNARY_REAL_IMAG): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_UNARY_IMAG): Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_UNARY_REAL_IMAG_RET_REAL):
Likewise.
[__HAVE_BUILTIN_TGMATH] (__TGMATH_BINARY_REAL_IMAG): Likewise.
(__TGMATH_UNARY_REAL_IMAG_RET_REAL_SAME): New macro.
(carg): Use __TGMATH_UNARY_REAL_IMAG_RET_REAL_SAME.
(cimag): Likewise.
(creal): Likewise.
Continuing the preparation for additional _FloatN / _FloatNx type
support, this patch prepares __MATH_TG to handle more such types.
Various unhandled cases, which do not correspond to any current glibc
configuration, have explicit #errors added. _Float32 and _Float64x
are then handled appropriately in the _Generic case, which is the only
one, other than the cases where use of sizeof is sufficient, where
they should ever be explicit types at the language level instead of
typedefs. There is no need to handle _Float64 or _Float32x explicitly
there because the default case calling a double function is correct
for those types.
Tested for x86_64.
* math/math.h [__HAVE_DISTINCT_FLOAT16
|| __HAVE_DISTINCT_FLOAT32 || __HAVE_DISTINCT_FLOAT64
|| __HAVE_DISTINCT_FLOAT32X || __HAVE_DISTINCT_FLOAT64X
|| __HAVE_DISTINCT_FLOAT128X]: Use #error.
[__NO_LONG_DOUBLE_MATH && __HAVE_DISTINCT_FLOAT128]: Likewise.
[__HAVE_DISTINCT_FLOAT128 && !__HAVE_GENERIC_SELECTION
&& __HAVE_FLOATN_NOT_TYPEDEF]: Likewise.
[__HAVE_DISTINCT_FLOAT128 && __HAVE_GENERIC_SELECTION]
(__MATH_TG_F32): New macro.
[__HAVE_DISTINCT_FLOAT128 && __HAVE_GENERIC_SELECTION]
(__MATH_TG_F64X): Likewise.
[__HAVE_DISTINCT_FLOAT128 && __HAVE_GENERIC_SELECTION]
(__MATH_TG): Use __MATH_TG_F32 and __MATH_TG_F64X.
Continuing the preparation for additional _FloatN / _FloatNx type
support, this patch improves how <tgmath.h> handles such types.
Use of #error is added for cases of distinct types that are not
supported by the header, to indicate that additional work on the
header would be needed if, for example, _Float16 support were added to
glibc. Given that #error, types with the same format as other types
are handled automatically by the sizeof-based logic, so the only case
needing special handling is that where _Float64x exists, has the same
format as _Float128, does not have the same format as long double, and
is not a typedef for _Float128. In this case (which will apply for
powerpc64le once _Float64x support is added to glibc), the
__builtin_types_compatible_p calls testing for _Float128 need
corresponding calls testing for _Float64x, which this patch adds.
Tested for x86_64.
* math/tgmath.h [__HAVE_DISTINCT_FLOAT16
|| __HAVE_DISTINCT_FLOAT32 || __HAVE_DISTINCT_FLOAT64
|| __HAVE_DISTINCT_FLOAT32X || __HAVE_DISTINCT_FLOAT64X
|| __HAVE_DISTINCT_FLOAT128X]: Use #error.
[__HAVE_DISTINCT_FLOAT128 && __GLIBC_USE (IEC_60559_TYPES_EXT)
&& __HAVE_FLOAT64X && !__HAVE_FLOAT64X_LONG_DOUBLE
&& __HAVE_FLOATN_NOT_TYPEDEF] (__TGMATH_F128): Handle _Float64x
the same as _Float128.
[__HAVE_DISTINCT_FLOAT128 && __GLIBC_USE (IEC_60559_TYPES_EXT)
&& __HAVE_FLOAT64X && !__HAVE_FLOAT64X_LONG_DOUBLE
&& __HAVE_FLOATN_NOT_TYPEDEF] (__TGMATH_CF128): Likewise.
Continuing the preparation for additional _FloatN / _FloatNx type
support, this patch arranges for <bits/cmathcalls.h> to be included by
<complex.h> for each such type under conditions and with macros
defined corresponding to those used for _Float128.
Tested for x86_64.
* math/complex.h
[(__HAVE_DISTINCT_FLOAT16 || (__HAVE_FLOAT16 && !_LIBC))
&& __GLIBC_USE (IEC_60559_TYPES_EXT)]: Include <bits/cmathcalls.h>
with appropriate macros defined and undefined.
[(__HAVE_DISTINCT_FLOAT32 || (__HAVE_FLOAT32 && !_LIBC))
&& __GLIBC_USE (IEC_60559_TYPES_EXT)]: Likewise.
[(__HAVE_DISTINCT_FLOAT64 || (__HAVE_FLOAT64 && !_LIBC))
&& __GLIBC_USE (IEC_60559_TYPES_EXT)]: Likewise.
[(__HAVE_DISTINCT_FLOAT32X || (__HAVE_FLOAT32X && !_LIBC))
&& __GLIBC_USE (IEC_60559_TYPES_EXT)]: Likewise.
[(__HAVE_DISTINCT_FLOAT64X || (__HAVE_FLOAT64X && !_LIBC))
&& __GLIBC_USE (IEC_60559_TYPES_EXT)]: Likewise.
[(__HAVE_DISTINCT_FLOAT128X || (__HAVE_FLOAT128X && !_LIBC))
&& __GLIBC_USE (IEC_60559_TYPES_EXT)]: Likewise.
This patch cleans up the way complex.h handles inclusion of
bits/cmathcalls.h for float128. The inclusion was between those for
the types float and long double; the patch moves it after that for
long double, matching how bits/mathcalls.h and bits/math-finite.h
inclusions are ordered. There is no need for the undefine and define
of _Mdouble_complex_ to be conditional, since __CFLOAT128 is always
defined by bits/floatn.h when _Float128 is supported, so the patch
removes the unnecessary conditionals.
Tested for x86_64.
* math/complex.h
[(__HAVE_DISTINCT_FLOAT128 || (__HAVE_FLOAT128 && !LIBC))
&& __GLIBC_USE (IEC_60559_TYPES_EXT)]: Move conditional code after
that for long double. Do not condition define and undefine of
_Mdouble_complex_ on [__CFLOAT128].
Continuing the preparation for additional _FloatN / _FloatNx type
support, this patch adds an additional case in the definition of
__MATH_EVAL_FMT2, as used in defining iseqsig: when
__FLT_EVAL_METHOD__ is 0 or 32, it adds 0.0f to the arguments, so that
the correct function would be selected in the case of _Float16
arguments with excess precision (were glibc to support _Float16, which
of course __MATH_TG and other facilities do not at present - and
_Float16 support is not part of what this patch series is aiming for,
but this particular fix is simple so is included anyway).
Tested for x86_64.
* math/math.h
[__FLT_EVAL_METHOD__ == 0 || __FLT_EVAL_METHOD__ == 32]
(__MATH_EVAL_FMT2): Define to add 0.0f.
Continuing the preparation for additional _FloatN / _FloatNx type
support, this patch extends the includes of <bits/math-finite.h> to
cover all such types, under conditions analogous to those for
_Float128.
Tested for x86_64.
* math/math.h [__HAVE_DISTINCT_FLOAT16 || (__HAVE_FLOAT16 && !_LIBC)]:
Include <bits/math-finite.h> with appropriate macros defined and
undefined.
[__HAVE_DISTINCT_FLOAT32 || (__HAVE_FLOAT32 && !_LIBC)]: Likewise.
[__HAVE_DISTINCT_FLOAT64 || (__HAVE_FLOAT64 && !_LIBC)]: Likewise.
[__HAVE_DISTINCT_FLOAT32X || (__HAVE_FLOAT32X && !_LIBC)]: Likewise.
[__HAVE_DISTINCT_FLOAT64X || (__HAVE_FLOAT64X && !_LIBC)]: Likewise.
[__HAVE_DISTINCT_FLOAT128X || (__HAVE_FLOAT128X && !_LIBC)]: Likewise.
math.h has a macro _Mlong_double_ for the type to use when declaring
long double functions, and similar macros for other types.
math/Makefile uses -D_Mlong_double_=double in the case of long double
having the same ABI as double.
This originates with:
Mon Jul 8 13:37:40 1996 Roland McGrath <roland@delasyd.gnu.ai.mit.edu>
* math/math.h (_Mfloat_, _Mlong_double_): New macros, defined iff not
already defined to float, long double. Use those macros for _Mdouble_
defns when including mathcalls.h.
* math/Makefile [$(long-double-fcts) != yes] (CPPFLAGS): Append
-D_Mlong_double_=double.
However, math.h stopped declaring long double functions in the case of
long double having the same ABI as double (and thus probably stopped
actually needing the Makefile definition of _Mlong_double_) with:
1998-11-05 Ulrich Drepper <drepper@cygnus.com>
* math/math.h: Unconditionally include bits/mathdef.h. Declare
long double functions only if __NO_LONG_DOUBLE_MATH is not
defined.
* sysdeps/generic/bits/mathdef.h: Define only if __USE_ISOC9X.
Define __NO_LONG_DOUBLE_MATH.
* sysdeps/m68k/fpu/bits/mathdef.h: Define only if __USE_ISOC9X.
* sysdeps/i386/fpu/bits/mathdef.h: Likewise.
The declarations were since restored for compiling user code, but
remain absent when _LIBC is defined, which is sufficient to avoid
problems declaring function aliases of incompatible types. Thus the
indirection through the _Mlong_double_ macro is not needed (probably
since that 1998 patch), and this patch removes _Mlong_double_ and
associated macros for other types, leaving only the macro _Mdouble_
which is actually used as the type for which a given inclusion of
<bits/mathcalls.h> should declared functions.
Tested for x86_64, and tested with build-many-glibcs.py that installed
stripped shared libraries are unchanged by this patch.
* math/math.h [!_Mfloat_] (_Mfloat_): Do not define.
[!_Mlong_double_] (_Mlong_double_): Likewise.
[!_Mfloat16_] (_Mfloat16_): Likewise.
[!_Mfloat32_] (_Mfloat32_): Likewise.
[!_Mfloat64_] (_Mfloat64_): Likewise.
[!_Mfloat128_] (_Mfloat128_): Likewise.
[!_Mfloat32x_] (_Mfloat32x_): Likewise.
[!_Mfloat64x_] (_Mfloat64x_): Likewise.
[!_Mfloat128x_] (_Mfloat128x_): Likewise.
(_Mdouble_): Define without indirection through those macros.
* math/complex.h [!_Mfloat_] (_Mfloat_): Do not define.
[!_Mfloat128_] (_Mfloat128_): Likewise.
[_Mlong_double_] (_Mlong_double_): Likewise.
(_Mdouble_): Define without indirection through those macros.
* math/Makefile [$(long-double-fcts) != yes] (math-CPPFLAGS): Do
not add -D_Mlong_double_=double.
* include/math.h [_ISOMAC] (_Mlong_double_): Do not undefine.
* math/test-signgam-finite-c99.c (_Mlong_double_): Likewise.
Continuing the preparation for additional _FloatN / _FloatNx type
support, this patch arranges for <bits/mathcalls.h> and
<bits/mathcalls-helper-functions.h> to be included for each such type
under conditions and with macros defined corresponding to those
already present for _Float128.
Tested for x86_64.
* math/math.h [__HAVE_DISTINCT_FLOAT16 || (__HAVE_FLOAT16 && !_LIBC)]:
Include <bits/mathcalls-helper-functions.h> and <bits/mathcalls.h>
with appropriate macros defined and undefined.
[__HAVE_DISTINCT_FLOAT32 || (__HAVE_FLOAT32 && !_LIBC)]: Likewise.
[__HAVE_DISTINCT_FLOAT64 || (__HAVE_FLOAT64 && !_LIBC)]: Likewise.
[__HAVE_DISTINCT_FLOAT32X || (__HAVE_FLOAT32X && !_LIBC)]: Likewise.
[__HAVE_DISTINCT_FLOAT64X || (__HAVE_FLOAT64X && !_LIBC)]: Likewise.
[__HAVE_DISTINCT_FLOAT128X || (__HAVE_FLOAT128X && !_LIBC)]: Likewise.
The bits/floatn.h header currently only has defines relating to
_Float128. This patch adds defines relating to other _FloatN /
_FloatNx types.
The approach taken is to add defines for all _FloatN / _FloatNx types
known to GCC, and to put them in a common bits/floatn-common.h header
included at the end of all the individual bits/floatn.h headers. If
in future some defines become different for different glibc
configurations, they will move out into the separate bits/floatn.h
headers.
Some defines are expected always to be the same across glibc ports.
Corresponding defines are nevertheless put in this header. The intent
is that where there are conditionals (in headers or in non-installed
files) that can just repeat the same or nearly the same logic for each
floating-point type, they should do so, even if in fact the cases for
some types could be unconditionally present or absent because the same
conditionals are true or false for all glibc configurations. This
should make the glibc code with such conditionals easier to read,
because the reader can just see that the same conditionals are
repeated for each type, rather than seeing different conditionals for
different types and needing to reason, at each location with such
differences, why those differences are indeed correct there. (Cases
involving per-format rather than per-type logic are more likely still
to need differences in how they handle different types.)
Having such defines and conditionals also helps in incremental
preparation for adding _Float32 / _Float64 / _Float32x / _Float64x
function aliases. I intend subsequent patches to add such
conditionals corresponding to those already present for _Float128, as
well as making more architecture-specific function implementations use
common macros to define aliases in preparation for adding such _FloatN
/ _FloatNx aliases.
Tested for x86_64.
* bits/floatn-common.h: New file.
* math/Makefile (headers): Add bits/floatn-common.h.
* bits/floatn.h: Include <bits/floatn-common.h>.
* sysdeps/ia64/bits/floatn.h: Likewise.
* sysdeps/ieee754/ldbl-128/bits/floatn.h: Likewise.
* sysdeps/mips/ieee754/bits/floatn.h: Likewise.
* sysdeps/powerpc/bits/floatn.h: Likewise.
* sysdeps/x86/bits/floatn.h: Likewise.
When using gcc < 6.x, signbit does not use the type-generic
__builtin_signbit builtin, instead it uses __MATH_TG.
However, when library support for float128 is available, __MATH_TG uses
__builtin_types_compatible_p, which is not available in C++ mode.
On the other hand, libstdc++ undefines (in cmath) many macros from
math.h, including signbit, so that it can provide its own functions.
However, during its configure tests, libstdc++ just tests for the
availability of the macros (it does not undefine them, nor does it
provide its own functions).
Finally, libstdc++ configure tests include math.h and get the definition
of signbit that uses __MATH_TG (and __builtin_types_compatible_p).
Since libstdc++ does not undefine the macros during its configure
tests, they fail.
This patch lets signbit use the builtin in C++ mode when gcc < 6.x is
used. This allows the configure test in libstdc++ to work.
Tested for x86_64.
[BZ #22296]
* math/math.h: Let signbit use the builtin in C++ mode with gcc
< 6.x
Cc: Gabriel F. T. Gomes <gftg@linux.vnet.ibm.com>
Cc: Joseph Myers <joseph@codesourcery.com>
With support for _Float128 functions on platforms where that type has
the same ABI as long double, as well as on platforms where it is
ABI-distinct, those functions will need to be exported from glibc's
shared libraries at appropriate symbol versions in each case.
This patch avoids duplication of lists of symbols to export by moving
the symbols other than __* to math/Versions and stdlib/Versions.
There, they are conditional on <float128-abi.h> defining
FLOAT128_VERSION and a default version of that header is added that
does not define that macro. Enabling the float128 function aliases
will then include adding a sysdeps/ieee754/ldbl-128/float128-abi.h
that defines FLOAT128_VERSION to GLIBC_2.27. Symbols __* remain in
sysdeps/ieee754/float128/Versions; those symbols should be present
only once per floating-point format, not once per type.
Note that if any platforms currently lacking support for a type with
binary128 format get glibc support for such a type in future (whether
only as _Float128, or also as a new long double format), and new libm
functions (present for all types) have been added by then, additional
macros will be needed to allow such functions to get a version of the
form "GLIBC_2.28 if the platform had _Float128 support by then, or the
later version at which that platform had _Float128 support added".
This is not however a preexisting condition, but would have applied
equally to the existing support for _Float128 as an ABI-distinct
type. New all-type libm functions should just be added to the
appropriate symbol version (currently GLIBC_2.27) for all types, with
such special-case handling for _Float128 versions (and _Float64x as
well in future) waiting until someone actually wants to add support
for _Float128 to an existing platform after a release in which that
platform and a post-2.26 libm function had support but that platform
lacked _Float128 support.
Tested with build-many-glibcs.py that installed stripped shared
libraries are unchanged by this patch. Also tested in conjunction
with the remaining changes to enable float128 aliases.
* sysdeps/generic/float128-abi.h: New file.
* sysdeps/ieee754/float128/Versions (FLOAT128_VERSION): Move
non-__prefixed symbols to ....
* math/Versions: ... here. Include <float128-abi.h>.
* stdlib/Versions ... and here. Include <float128-abi.h>
This patch adds support for running libm tests for float128 in the
case where the float128 functions are aliases of long double
functions. In this case, the sysdeps Makeconfig file
(i.e. sysdeps/ieee754/ldbl-128/Makeconfig) will need to define
"float128-alias-fcts = yes" to enable the tests.
Tested for x86_64. Also tested with build-many-glibcs.py; installed
stripped shared libraries are unchanged by the patch. Also tested
together with changes to enable the float128 aliases.
* math/Makefile (test-types): Add
$(type-float128-$(float128-alias-fcts)).
* math/test-float128.h (TYPE_STR): Define conditional on
[FLT128_MANT_DIG == LDBL_MANT_DIG].
(ULP_IDX): Likewise.
(ULP_I_IDX): Likewise.
All interfaces in the glibc API ought to be covered by the testsuite,
even where they alias other interfaces. This patch arranges for libm
tests to be run for long double when it aliases double; previously
those tests were run only for the floating-point types with distinct
formats. The long double tests are made to use the double ulps values
in this case, as having a separate duplicate set of ulps for them
seems unnecessary; to accommodate that, the test-<type>.h headers now
specify the macro indexing into the ulps array explicitly instead of
having it computed from PREFIX.
Nothing special is done about vector function tests. None are
supported for any long double = double platforms, and supporting
vector functions for a type alias such as _Float32 would not simply
fall out of adding the scalar aliases for that type - it would require
vector function wrappers like those for *_finite (or, better, a new
GCC feature to allow specifying the asm name for vector functions
independently of that for scalar ones, as previously discussed), so it
seems reasonable to require the sysdeps makefile setting of
libmvec-tests to be updated if any such tests are to be run for type
aliases.
Tested for x86_64 and arm.
* math/Makefile (test-types-basic): New variable.
(test-types): Likewise.
(libm-test-support): Use $(test-types) instead of $(types).
(libm-tests-base-normal): Likewise.
(libm-tests-base-finite): Likewise.
(libm-tests-base-inline): Likewise.
(generated): Likewise.
($(objpfx)libm-test-support-$(t).c): Likewise.
(libm-tests-for-type iterator): Likewise.
(libm-test-support iterator): Likewise.
* math/libm-test-support.c (ulp_i_idx): Use ULP_I_IDX.
(ulp_idx): Use ULP_IDX.
* math/test-ldouble.h: Include <float.h>.
(TYPE_STR): Define conditional on [LDBL_MANT_DIG == DBL_MANT_DIG].
(ULP_IDX): New macro.
(ULP_I_IDX): Likewise.
* math/test-double.h (ULP_IDX): Likewise.
(ULP_I_IDX): Likewise.
* math/test-float.h (ULP_IDX): Likewise.
(ULP_I_IDX): Likewise.
* math/test-float128.h (ULP_IDX): Likewise.
(ULP_I_IDX): Likewise.
Some libm functions are unable to use the generic alias macros such as
libm_alias_double because they have special symbol versioning
requirements for the main float, double or long double public names.
To facilitate adding _FloatN / _FloatNx function aliases in future,
it's still desirable to have generic macros those functions can use as
far as possible. This patch adds macros such as
libm_alias_double_other, which only define names for _FloatN /
_FloatNx aliases, not for float / double / long double. As present,
all these new macros do nothing, but they are called in the
appropriate places in macros such as libm_alias_double. This patch
also arranges for lgamma implementations, and the recently added
optimized float function implementations, to use the new macros to
make them ready for addition of _FloatN / _FloatNx aliases.
Tested for x86_64, and tested with build-many-glibcs.py that installed
stripped shared libraries are unchanged by this patch.
* sysdeps/generic/libm-alias-double.h (libm_alias_double_other_r):
New macro.
(libm_alias_double_other): Likewise.
(libm_alias_double_r): Use libm_alias_double_other_r.
* sysdeps/generic/libm-alias-float.h (libm_alias_float_other_r):
New macro.
(libm_alias_float_other): Likewise.
(libm_alias_float_r): Use libm_alias_float_other_r.
* sysdeps/generic/libm-alias-float128.h
(libm_alias_float128_other_r): New macro.
(libm_alias_float128_other): Likewise.
(libm_alias_float128_r): Use libm_alias_float128_other_r.
* sysdeps/generic/libm-alias-ldouble.h
(libm_alias_ldouble_other_r): New macro.
(libm_alias_ldouble_other): Likewise.
(libm_alias_ldouble_r): Use libm_alias_ldouble_other_r.
* sysdeps/ieee754/ldbl-opt/libm-alias-double.h
(libm_alias_double_other_r): New macro.
(libm_alias_double_other): Likewise.
(libm_alias_double_r): Use libm_alias_double_other_r.
* sysdeps/ieee754/ldbl-opt/libm-alias-ldouble.h
(libm_alias_ldouble_other_r): New macro.
(libm_alias_ldouble_other): Likewise.
(libm_alias_ldouble_r): Use libm_alias_ldouble_other_r.
* math/w_lgamma_main.c: Include <libm-alias-double.h>.
[!USE_AS_COMPAT]: Use libm_alias_double_other.
* math/w_lgammaf_main.c: Include <libm-alias-float.h>.
[!USE_AS_COMPAT]: Use libm_alias_float_other.
* math/w_lgammal_main.c: Include <libm-alias-ldouble.h>.
[!USE_AS_COMPAT]: Use libm_alias_ldouble_other.
* math/w_exp2f.c: Use libm_alias_float_other.
* math/w_expf.c: Likewise.
* math/w_log2f.c: Likewise.
* math/w_logf.c: Likewise.
* math/w_powf.c: Likewise.
* sysdeps/ieee754/flt-32/e_exp2f.c: Include <libm-alias-float.h>.
[!__exp2f]: Use libm_alias_float_other.
* sysdeps/ieee754/flt-32/e_expf.c: Include <libm-alias-float.h>.
[!__expf]: Use libm_alias_float_other.
* sysdeps/ieee754/flt-32/e_log2f.c: Include <libm-alias-float.h>.
[!__log2f]: Use libm_alias_float_other.
* sysdeps/ieee754/flt-32/e_logf.c: Include <libm-alias-float.h>.
[!__logf]: Use libm_alias_float_other.
* sysdeps/ieee754/flt-32/e_powf.c: Include <libm-alias-float.h>.
[!__powf]: Use libm_alias_float_other.
Continuing the use of generic macros for defining libm function
aliases, in preparation for adding more _FloatN / _FloatNx function
names, this patch makes the lgamma_r functions use such macros.
declare_mgen_alias_r becomes a standard macro in math-type-macros.h
instead of being locally defined in w_lgamma_r_templace.c. This in
turn must be defined by each math-type-macros-<type>.h. Rather than
providing an unused default in math-type-macros.h, that header is made
to give an error if math-type-macros-<type>.h failed to define
declare_mgen_alias or declare_mgen_alias_r. The compat lgamma_r
wrappers are updated similarly. The ldbl-opt versions are removed as
no longer needed.
Tested for x86_64, and with build-many-glibcs.py. Installed stripped
shared libraries are unchanged except for powerpc64le (where the usual
issue applies that an ldbl-opt long double function previously used
long_double_symbol unconditionally and now the symbol versions on
powerpc64le mean weak_alias is used instead, resulting in the same
symbol versions in the final shared library but still enough
difference in the input objects for that library not to be
byte-identical).
* sysdeps/generic/math-type-macros.h [!declare_mgen_alias]: Give
error. Remove default definition of declare_mgen_alias.
[!declare_mgen_alias_r]: Likewise.
* sysdeps/generic/math-type-macros-double.h
[!declare_mgen_alias_r] (declare_mgen_alias_r): New macro.
* sysdeps/generic/math-type-macros-float.h [!declare_mgen_alias_r]
(declare_mgen_alias_r): Likewise.
* sysdeps/generic/math-type-macros-float128.h
[!declare_mgen_alias_r] (declare_mgen_alias_r): Likewise.
* sysdeps/generic/math-type-macros-ldouble.h
[!declare_mgen_alias_r] (declare_mgen_alias_r): Likewise.
* math/w_lgamma_r_template.c (declare_mgen_alias_r_x): Remove
macro.
(declare_mgen_alias_r_s): Likewise.
(declare_mgen_alias_r): Likewise.
* math/w_lgamma_r_compat.c: Include <libm-alias-double.h>.
(lgamma_r): Define using libm_alias_double_r.
* math/w_lgammaf_r_compat.c: Include <libm-alias-float.h>.
(lgammaf_r): Define using libm_alias_float_r.
* math/w_lgammal_r_compat.c: Include <libm-alias-ldouble.h>.
(lgammal_r): Define using libm_alias_ldouble_r.
* sysdeps/ieee754/ldbl-opt/w_lgamma_r_compat.c: Remove file.
* sysdeps/ieee754/ldbl-opt/w_lgammal_r_compat.c: Likewise.
This patch ensures that the C99 and C11 features required by C++ are
defined according to the value of the __cplusplus macro, and not just
because G++ always defines _GNU_SOURCE. This will allow G++ to stop
defining _GNU_SOURCE some day, without causing the C99 and C11
interfaces to disappear for C++ programs.
[BZ #21326]
* include/features.h [__cplusplus >= 201103] (__USE_ISOC99): Define.
[__cplusplus >= 201703] (__USE_ISOCXX17, __USE_ISOC11): Define.
* math/Makefile (test-math-cxx11): New test.
* math/test-math-cxx11.cc: New file.
Fix GCC 7 compilation error:
test-math-iscanonical.cc: In function ‘void check_type()’:
test-math-iscanonical.cc:33:11: error: use of an operand of type ‘bool’ in ‘operator++’ is deprecated [-Werror=deprecated]
errors++;
^~
* math/test-math-iscanonical.cc (error): Replace bool with int.
All representations of floating-point numbers in types with IEC 60559
binary exchange format are canonical. On the other hand, types with IEC
60559 extended formats, such as those implemented under ldbl-96 and
ldbl-128ibm, contain representations that are not canonical.
TS 18661-1 introduced the type-generic macro iscanonical, which returns
whether a floating-point value is canonical or not. In Glibc, this
type-generic macro is implemented using the macro __MATH_TG, which, when
support for float128 is enabled, relies on __builtin_types_compatible_p
to select between floating-point types. However, this use of
iscanonical breaks C++ applications, because the builtin is only
available in C mode.
This patch provides a C++ implementation of iscanonical that relies on
function overloading, rather than builtins, to select between
floating-point types.
Unlike the C++ implementations for iszero and issignaling, this
implementation ignores __NO_LONG_DOUBLE_MATH. The double type always
matches IEC 60559 double format, which is always canonical. Thus, when
double and long double are the same (__NO_LONG_DOUBLE_MATH), iscanonical
always returns 1 and is not implemented with __MATH_TG.
Tested for powerpc64, powerpc64le and x86_64.
[BZ #22235]
* math/math.h: Trivial fix for unbalanced parentheses in comment.
* math/Makefile [CXX] (tests): Add test-math-iscanonical.cc.
(CFLAGS-test-math-iscanonical.cc): New variable.
* math/test-math-iscanonical.cc: New file.
* sysdeps/ieee754/ldbl-96/bits/iscanonical.h (iscanonical):
Provide a C++ implementation based on function overloading,
rather than using __MATH_TG, which uses C-only builtins.
* sysdeps/ieee754/ldbl-128ibm/bits/iscanonical.h (iscanonical):
Likewise.
* sysdeps/powerpc/powerpc64le/Makefile
(CFLAGS-test-math-iscanonical.cc): New variable.
The compat remainder code is no longer built for !LIBM_SVID_COMPAT targets,
but the legacy drem, dremf and dreml symbols should be still defined since
removing them may break existing code, so keep them as aliases in the
non-compat remainder code.
* math/w_remainder.c: New file.
* math/w_remainderf.c: New file.
* math/w_remainderl.c: New file.
The lgamma compat code is no longer built for !LIBM_SVID_COMPAT targets,
but the legacy gamma, gammaf and gammal symbols should be still defined,
so make them aliases to the non-compat lgamma code.
* math/w_lgamma.c: New file.
* math/w_lgammaf.c: New file.
* math/w_lgammal.c: New file.
Wilco Dijkstra
Joseph Myers
Gabriel F. T. Gomes
Patrick McGehearty
Szabolcs Nagy
H.J. Lu
Romain Naour
Jonathan Wakely