about summary refs log tree commit diff
path: root/sysdeps/i386/fpu/e_scalb.S
Commit message (Collapse)AuthorAgeFilesLines
* Add libm_alias_finite for _finite symbolsWilco Dijkstra2020-01-031-1/+2
| | | | | | | | | | | | | | | | | | | This patch adds a new macro, libm_alias_finite, to define all _finite symbol. It sets all _finite symbol as compat symbol based on its first version (obtained from the definition at built generated first-versions.h). The <fn>f128_finite symbols were introduced in GLIBC 2.26 and so need special treatment in code that is shared between long double and float128. It is done by adding a list, similar to internal symbol redifinition, on sysdeps/ieee754/float128/float128_private.h. Alpha also needs some tricky changes to ensure we still emit 2 compat symbols for sqrt(f). Passes buildmanyglibc. Co-authored-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
* Avoid excess range in results from i386 scalb functions (bug 18981).Joseph Myers2015-09-181-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | i386 scalb / scalbn / scalbln (and thus ldexp) functions for float and double can return results with excess range (and consequently excess precision for subnormal results). As the results of these functions are fully determined by reference to IEEE 754 operations, this is unambiguously a bug, apart from the testsuite failures it causes. This patch makes those functions store their results on the stack and load them back to eliminate the excess range. Double rounding is not a problem, as the only cases where it could occur are when the result overflows or underflows for extended precision, and then the double-rounded results are the same as the single-rounded results. The new macros will be used for more functions, more such macros added, and existing code refactored to use such macros, in subsequent patches. Tested for x86. Committed. [BZ #18981] * sysdeps/i386/fpu/i386-math-asm.h: New file. * sysdeps/i386/fpu/e_scalb.S: Include <i386-math-asm.h>. (__ieee754_scalb): Use DBL_NARROW_EVAL. * sysdeps/i386/fpu/e_scalbf.S: Include <i386-math-asm.h>. (__ieee754_scalbf): Use FLT_NARROW_EVAL. * sysdeps/i386/fpu/s_scalbn.S: Include <i386-math-asm.h>. (__scalbn): Use DBL_NARROW_EVAL. * sysdeps/i386/fpu/s_scalbnf.S: Include <i386-math-asm.h>. (__scalbnf): Use FLT_NARROW_EVAL.
* Fix x86/x86_64 scalb (qNaN, -Inf) (bug 16783).Joseph Myers2015-02-241-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Various x86 / x86_64 versions of scalb / scalbf / scalbl produce spurious "invalid" exceptions for (qNaN, -Inf) arguments, because this is wrongly handled like (+/-Inf, -Inf) which *should* raise such an exception. (In fact the NaN case of the code determining whether to quietly return a zero or a NaN for second argument -Inf was accidentally dead since the code had been made to return a NaN with exception.) This patch fixes the code to do the proper test for an infinity as distinct from a NaN. (Since the existing code does nothing to distinguish qNaNs and sNaNs here, this patch doesn't either. If in future we systematically implement proper sNaN semantics following TS 18661-1:2014, there will be lots of bugs to address - Thomas found lots of issues with his patch <https://sourceware.org/ml/libc-ports/2013-04/msg00008.html> to add SNaN tests (which never went in and would now require significant reworking).) Tested for x86_64 and x86. Committed. [BZ #16783] * sysdeps/i386/fpu/e_scalb.S (__ieee754_scalb): Do not handle arguments (NaN, -Inf) the same as (+/-Inf, -Inf). * sysdeps/i386/fpu/e_scalbf.S (__ieee754_scalbf): Likewise. * sysdeps/i386/fpu/e_scalbl.S (__ieee754_scalbl): Likewise. * sysdeps/x86_64/fpu/e_scalbl.S (__ieee754_scalbl): Likewise. * math/libm-test.inc (scalb_test_data): Add more tests.
* Get rid of ASM_TYPE_DIRECTIVE{,_PREFIX}.Marek Polacek2012-08-021-1/+1
|
* Remove __ELF__ conditionalsMarek Polacek2012-02-071-4/+0
|
* Optimize libmUlrich Drepper2011-10-121-7/+5
| | | | | libm is now somewhat integrated with gcc's -ffinite-math-only option and lots of the wrapper functions have been optimized.
* * intl/Makefile (tst-gettext[45].out): Pass also $(run-program-prefix) cvs/fedora-glibc-20050504T1818Ulrich Drepper2005-05-041-6/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | as argument to the scripts. * intl/tst-gettext2.sh: Use mkdir -p instead of test -d + mkdir. * intl/tst-gettext4.sh: Likewise. Use run_program_prefix argument. * intl/tst-gettext5.sh: Likewise. * intl/tst-translit.sh: Add mkdir -p. * sysdeps/unix/sysv/linux/i386/sysdep.h (SETUP_PIC_REG): Use .ifndef/.endif to allow use of this macro more than once per .S file. (LOAD_PIC_REG): New macro. * sysdeps/unix/sysv/linux/i386/makecontext.S: Add call frame information. * sysdeps/unix/sysv/linux/i386/getcontext.S: Likewise. * sysdeps/unix/sysv/linux/i386/vfork.S: Likewise. * sysdeps/i386/fpu/s_asinh.S: Use LOAD_PIC_REG macro. Add call frame information. * sysdeps/i386/fpu/e_log10f.S: Likewise. * sysdeps/i386/fpu/s_expm1.S: Likewise. * sysdeps/i386/fpu/e_acoshf.S: Likewise. * sysdeps/i386/fpu/e_log2l.S: Likewise. * sysdeps/i386/fpu/s_log1pf.S: Likewise. * sysdeps/i386/fpu/s_lrint.S: Likewise. * sysdeps/i386/fpu/s_llrint.S: Likewise. * sysdeps/i386/fpu/s_ilogbf.S: Likewise. * sysdeps/i386/fpu/s_cbrtl.S: Likewise. * sysdeps/i386/fpu/s_asinhf.S: Likewise. * sysdeps/i386/fpu/e_log.S: Likewise. * sysdeps/i386/fpu/e_atanhf.S: Likewise. * sysdeps/i386/fpu/s_lrintl.S: Likewise. * sysdeps/i386/fpu/e_atanhl.S: Likewise. * sysdeps/i386/fpu/e_scalb.S: Likewise. * sysdeps/i386/fpu/s_log1p.S: Likewise. * sysdeps/i386/fpu/s_nearbyintl.S: Likewise. * sysdeps/i386/fpu/s_frexpl.S: Likewise. * sysdeps/i386/fpu/s_log1pl.S: Likewise. * sysdeps/i386/fpu/s_nearbyintf.S: Likewise. * sysdeps/i386/fpu/s_cbrt.S: Likewise. * sysdeps/i386/fpu/s_expm1l.S: Likewise. * sysdeps/i386/fpu/s_lrintf.S: Likewise. * sysdeps/i386/fpu/e_acosh.S: Likewise. * sysdeps/i386/fpu/s_cexp.S: Likewise. * sysdeps/i386/fpu/s_ilogbl.S: Likewise. * sysdeps/i386/fpu/s_expm1f.S: Likewise. * sysdeps/i386/fpu/e_powl.S: Likewise. * sysdeps/i386/fpu/e_powf.S: Likewise. * sysdeps/i386/fpu/e_scalbf.S: Likewise. * sysdeps/i386/fpu/e_logl.S: Likewise. * sysdeps/i386/fpu/e_acoshl.S: Likewise. * sysdeps/i386/fpu/s_frexp.S: Likewise. * sysdeps/i386/fpu/e_pow.S: Likewise. * sysdeps/i386/fpu/e_logf.S: Likewise. * sysdeps/i386/fpu/e_log2.S: Likewise. * sysdeps/i386/fpu/s_frexpf.S: Likewise. * sysdeps/i386/fpu/s_cexpl.S: Likewise. * sysdeps/i386/fpu/s_llrintf.S: Likewise. * sysdeps/i386/fpu/s_ilogb.S: Likewise. * sysdeps/i386/fpu/e_scalbl.S: Likewise. * sysdeps/i386/fpu/e_atanh.S: Likewise. * sysdeps/i386/fpu/e_log10.S: Likewise. * sysdeps/i386/fpu/s_cbrtf.S: Likewise. * sysdeps/i386/fpu/s_cexpf.S: Likewise. * sysdeps/i386/fpu/s_llrintl.S: Likewise. * sysdeps/i386/fpu/e_log10l.S: Likewise. * sysdeps/i386/fpu/s_nearbyint.S: Likewise. * sysdeps/i386/fpu/s_asinhl.S: Likewise. * sysdeps/i386/fpu/e_log2f.S: Likewise. * sysdeps/i386/addmul_1.S: Various fixes to cfi handling. * sysdeps/i386/mul_1.S: Likewise. * sysdeps/i386/strtok.S: Likewise. * sysdeps/i386/sub_n.S: Likewise. * sysdeps/i386/submul_1.S: Likewise. * sysdeps/i386/i586/addmul_1.S: Likewise. * sysdeps/i386/i586/memcpy.S: Likewise. * sysdeps/i386/i586/mul_1.S: Likewise. * sysdeps/i386/i586/rshift.S: Likewise. * sysdeps/i386/i586/sub_n.S: Likewise. * sysdeps/i386/i586/submul_1.S: Likewise. * sysdeps/i386/i686/memcmp.S: Likewise. * sysdeps/i386/i686/memmove.S: Likewise. * sysdeps/unix/sysv/linux/i386/clone.S: Likewise. * sysdeps/unix/sysv/linux/i386/mmap.S: Likewise. * sysdeps/unix/sysv/linux/i386/mmap64.S: Likewise. * sysdeps/unix/sysv/linux/i386/setcontext.S: Likewise.
* (CFLAGS-tst-align.c): Add -mpreferred-stack-boundary=4.Ulrich Drepper2004-12-221-2/+6
|
* 2.5-18.1Jakub Jelinek2007-07-121-6/+2
|
* Update.Ulrich Drepper2001-02-181-1/+9
| | | | | | | | | | 2001-02-18 Ulrich Drepper <drepper@redhat.com> * math/libm-test.inc (scalb_test): Require invalid exception being raised for invalid parameters. * sysdeps/i386/fpu/e_scalb.S: Raise invalid exception if necessary. * sysdeps/i386/fpu/e_scalbf.S: Likewise. * sysdeps/i386/fpu/e_scalbl.S: Likewise.
* Update.Ulrich Drepper2000-12-051-1/+5
| | | | | | | | | | | | | * sysdeps/i386/fpu/e_scalb.S: Handle NaN as first parameter correctly. * sysdeps/i386/fpu/e_scalbf.S: Likewise. * sysdeps/i386/fpu/e_scalbl.S: Likewise. * math/w_scalb.c: Don't use matherr except in SVID mode. * math/w_scalbf.c: Likewise. * math/w_scalbl.c: Likewise. * math/test-misc.c: Add test for NaN and scalbl. Reported by Fred J. Tydeman <tydeman@tybor.com>. 2000-12-04 Ulrich Drepper <drepper@redhat.com>
* Update.Ulrich Drepper1999-07-141-0/+94