glibc/sysdeps/i386/fpu/e_atanh.S
Joseph Myers 8020a80887 Fix atanhl missing underflows (bug 16352).
Similar to various other bugs in this area, some atanh implementations
do not raise the underflow exception for subnormal arguments, when the
result is tiny and inexact.  This patch forces the exception in a
similar way to previous fixes.  (No change in this regard is needed
for the i386 implementation; special handling to force underflows in
these cases will only be needed there when the spurious underflows,
bug 18049, get fixed.)

Tested for x86_64, x86, powerpc and mips64.

	[BZ #16352]
	* sysdeps/i386/fpu/e_atanh.S (dbl_min): New object.
	(__ieee754_atanh): Force underflow exception for results with
	small absolute value.
	* sysdeps/i386/fpu/e_atanhf.S (flt_min): New object.
	(__ieee754_atanhf): Force underflow exception for results with
	small absolute value.
	* sysdeps/ieee754/dbl-64/e_atanh.c: Include <float.h>.
	(__ieee754_atanh): Force underflow exception for results with
	small absolute value.
	* sysdeps/ieee754/flt-32/e_atanhf.c: Include <float.h>.
	(__ieee754_atanhf): Force underflow exception for results with
	small absolute value.
	* sysdeps/ieee754/ldbl-128/e_atanhl.c: Include <float.h>.
	(__ieee754_atanhl): Force underflow exception for results with
	small absolute value.
	* sysdeps/ieee754/ldbl-128ibm/e_atanhl.c: Include <float.h>.
	(__ieee754_atanhl): Force underflow exception for results with
	small absolute value.
	* sysdeps/ieee754/ldbl-96/e_atanhl.c: Include <float.h>.
	(__ieee754_atanhl): Force underflow exception for results with
	small absolute value.
	* math/auto-libm-test-in: Do not allow missing underflow
	exceptions from atanh.
	* math/auto-libm-test-out: Regenerated.
2015-05-15 22:07:57 +00:00

131 lines
3.2 KiB
ArmAsm
Raw Blame History

/* ix87 specific implementation of arctanh function.
Copyright (C) 1996-2015 Free Software Foundation, Inc.
This file is part of the GNU C Library.
Contributed by Ulrich Drepper <drepper@cygnus.com>, 1996.
The GNU C Library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
The GNU C Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with the GNU C Library; if not, see
<http://www.gnu.org/licenses/>. */
#include <machine/asm.h>
.section .rodata
.align ALIGNARG(4)
.type half,@object
half: .double 0.5
ASM_SIZE_DIRECTIVE(half)
.type one,@object
one: .double 1.0
ASM_SIZE_DIRECTIVE(one)
.type limit,@object
limit: .double 0.29
ASM_SIZE_DIRECTIVE(limit)
.type ln2_2,@object
ln2_2: .tfloat 0.3465735902799726547086160
ASM_SIZE_DIRECTIVE(ln2_2)
.section .rodata.cst8,"aM",@progbits,8
.p2align 3
.type dbl_min,@object
dbl_min: .byte 0, 0, 0, 0, 0, 0, 0x10, 0
ASM_SIZE_DIRECTIVE(dbl_min)
#ifdef PIC
#define MO(op) op##@GOTOFF(%edx)
#else
#define MO(op) op
#endif
.text
ENTRY(__ieee754_atanh)
movl 8(%esp), %ecx
movl %ecx, %eax
andl $0x7fffffff, %eax
cmpl $0x7ff00000, %eax
jae 5f
7:
#ifdef PIC
call 1f
cfi_adjust_cfa_offset (4)
1: popl %edx
cfi_adjust_cfa_offset (-4)
addl $_GLOBAL_OFFSET_TABLE_+[.-1b], %edx
#endif
andl $0x80000000, %ecx // ECX == 0 iff X >= 0
fldt MO(ln2_2) // 0.5*ln2
xorl %ecx, 8(%esp)
fldl 4(%esp) // |x| : 0.5*ln2
fcoml MO(half) // |x| : 0.5*ln2
fld %st // |x| : |x| : 0.5*ln2
fnstsw // |x| : |x| : 0.5*ln2
sahf
jae 2f
fadd %st, %st(1) // |x| : 2*|x| : 0.5*ln2
fld %st // |x| : |x| : 2*|x| : 0.5*ln2
fsubrl MO(one) // 1-|x| : |x| : 2*|x| : 0.5*ln2
fxch // |x| : 1-|x| : 2*|x| : 0.5*ln2
fmul %st(2) // 2*|x|^2 : 1-|x| : 2*|x| : 0.5*ln2
fdivp // (2*|x|^2)/(1-|x|) : 2*|x| : 0.5*ln2
faddp // 2*|x|+(2*|x|^2)/(1-|x|) : 0.5*ln2
fcoml MO(limit) // 2*|x|+(2*|x|^2)/(1-|x|) : 0.5*ln2
fnstsw // 2*|x|+(2*|x|^2)/(1-|x|) : 0.5*ln2
sahf
jae 4f
fyl2xp1 // 0.5*ln2*ld(1+2*|x|+(2*|x|^2)/(1-|x|))
fcoml MO(dbl_min)
fnstsw
sahf
jae 8f
subl $8, %esp
cfi_adjust_cfa_offset (8)
fld %st(0)
fmul %st(0)
fstpl (%esp)
addl $8, %esp
cfi_adjust_cfa_offset (-8)
8: jecxz 3f
fchs // 0.5*ln2*ld(1+2*x+(2*x^2)/(1-x))
3: ret
.align ALIGNARG(4)
4: faddl MO(one) // 1+2*|x|+(2*|x|^2)/(1-|x|) : 0.5*ln2
fyl2x // 0.5*ln2*ld(1+2*|x|+(2*|x|^2)/(1-|x|))
jecxz 3f
fchs // 0.5*ln2*ld(1+2*x+(2*x^2)/(1-x))
3: ret
.align ALIGNARG(4)
2: faddl MO(one) // 1+|x| : |x| : 0.5*ln2
fxch // |x| : 1+|x| : 0.5*ln2
fsubrl MO(one) // 1-|x| : 1+|x| : 0.5*ln2
fdivrp // (1+|x|)/(1-|x|) : 0.5*ln2
fyl2x // 0.5*ln2*ld((1+|x|)/(1-|x|))
jecxz 3f
fchs // 0.5*ln2*ld((1+x)/(1-x))
3: ret
// x == NaN or <EFBFBD>Inf
5: ja 6f
cmpl $0, 4(%esp)
je 7b
6: fldl 4(%esp)
ret
END(__ieee754_atanh)
strong_alias (__ieee754_atanh, __atanh_finite)