From 495fd99f3a119e5c0c542ccc6cf9c93b1fb9e892 Mon Sep 17 00:00:00 2001 From: Joseph Myers Date: Mon, 7 May 2012 19:13:08 +0000 Subject: Fix x86/x86_64 expm1l inaccuracy and exceptions (bugs 13885, 13923). --- sysdeps/i386/fpu/e_expl.S | 45 +++++++++++++++++++- sysdeps/i386/fpu/libm-test-ulps | 64 ++++++++++------------------- sysdeps/i386/fpu/s_expm1l.S | 91 +---------------------------------------- 3 files changed, 68 insertions(+), 132 deletions(-) (limited to 'sysdeps/i386') diff --git a/sysdeps/i386/fpu/e_expl.S b/sysdeps/i386/fpu/e_expl.S index 9adf2a489e..bab0a081b8 100644 --- a/sysdeps/i386/fpu/e_expl.S +++ b/sysdeps/i386/fpu/e_expl.S @@ -28,6 +28,10 @@ # define IEEE754_EXPL __ieee754_exp10l # define EXPL_FINITE __exp10l_finite # define FLDLOG fldl2t +#elif defined USE_AS_EXPM1L +# define IEEE754_EXPL __expm1l +# undef EXPL_FINITE +# define FLDLOG fldl2e #else # define IEEE754_EXPL __ieee754_expl # define EXPL_FINITE __expl_finite @@ -69,6 +73,12 @@ csat: .byte 0, 0, 0, 0, 0, 0, 0, 0x80, 0x0e, 0x40 .text ENTRY(IEEE754_EXPL) +#ifdef USE_AS_EXPM1L + movzwl 4+8(%esp), %eax + xorb $0x80, %ah // invert sign bit (now 1 is "positive") + cmpl $0xc006, %eax // is num positive and exp >= 6 (number is >= 128.0)? + jae HIDDEN_JUMPTARGET (__expl) // (if num is denormal, it is at least >= 64.0) +#endif fldt 4(%esp) /* I added the following ugly construct because expl(+-Inf) resulted in NaN. The ugliness results from the bright minds at Intel. @@ -78,7 +88,9 @@ ENTRY(IEEE754_EXPL) #ifdef PIC LOAD_PIC_REG (cx) #endif +#ifndef USE_AS_EXPM1L movzwl 4+8(%esp), %eax +#endif andl $0x7fff, %eax cmpl $0x400d, %eax jle 3f @@ -96,7 +108,16 @@ ENTRY(IEEE754_EXPL) andb $2, %ah jz 3f fchs -3: FLDLOG /* 1 log2(base) */ +3: +#ifdef USE_AS_EXPM1L + /* Test for +-0 as argument. */ + fstsw %ax + movb $0x45, %dh + andb %ah, %dh + cmpb $0x40, %dh + je 2f +#endif + FLDLOG /* 1 log2(base) */ fmul %st(1), %st /* 1 x log2(base) */ frndint /* 1 i */ fld %st(1) /* 2 x */ @@ -114,17 +135,39 @@ ENTRY(IEEE754_EXPL) fmul %st(4), %st /* 4 c1 * x */ faddp %st, %st(1) /* 3 f = f + c1 * x */ f2xm1 /* 3 2^(fract(x * log2(base))) - 1 */ +#ifdef USE_AS_EXPM1L + fstp %st(1) /* 2 */ + fscale /* 2 scale factor is st(1); base^x - 2^i */ + fxch /* 2 i */ + fld1 /* 3 1.0 */ + fscale /* 3 2^i */ + fld1 /* 4 1.0 */ + fsubrp %st, %st(1) /* 3 2^i - 1.0 */ + fstp %st(1) /* 2 */ + faddp %st, %st(1) /* 1 base^x - 1.0 */ +#else fld1 /* 4 1.0 */ faddp /* 3 2^(fract(x * log2(base))) */ fstp %st(1) /* 2 */ fscale /* 2 scale factor is st(1); base^x */ fstp %st(1) /* 1 */ +#endif fstp %st(1) /* 0 */ jmp 2f 1: testl $0x200, %eax /* Test sign. */ jz 2f /* If positive, jump. */ fstp %st +#ifdef USE_AS_EXPM1L + fld1 + fchs +#else fldz /* Set result to 0. */ +#endif 2: ret END(IEEE754_EXPL) +#ifdef USE_AS_EXPM1L +libm_hidden_def (__expm1l) +weak_alias (__expm1l, expm1l) +#else strong_alias (IEEE754_EXPL, EXPL_FINITE) +#endif diff --git a/sysdeps/i386/fpu/libm-test-ulps b/sysdeps/i386/fpu/libm-test-ulps index 63235537e0..cdc26a0c63 100644 --- a/sysdeps/i386/fpu/libm-test-ulps +++ b/sysdeps/i386/fpu/libm-test-ulps @@ -1008,10 +1008,6 @@ Test "cos_upward (9) == -0.9111302618846769883682947111811653112463": ildouble: 1 ldouble: 1 -# cosh -Test "cosh (0.75) == 1.29468328467684468784170818539018176": -ildouble: 1 - # cosh_downward Test "cosh_downward (22) == 1792456423.065795780980053377632656584997": double: 1 @@ -1899,29 +1895,20 @@ double: 1 float: 1 idouble: 1 ifloat: 1 -ildouble: 4 -ldouble: 4 +ildouble: 2 +ldouble: 2 Test "sinh_downward (23) == 4872401723.124451299966006944252978187305": double: 1 float: 1 idouble: 1 ifloat: 1 +ildouble: 2 +ldouble: 2 Test "sinh_downward (24) == 13244561064.92173614705070540368454568168": float: 1 ifloat: 1 -ildouble: 5 -ldouble: 5 - -# sinh_tonearest -Test "sinh_tonearest (22) == 1792456423.065795780701106568345764104225": -ildouble: 3 -ldouble: 3 -Test "sinh_tonearest (23) == 4872401723.124451299966006944252978187305": -ildouble: 1 -ldouble: 1 -Test "sinh_tonearest (24) == 13244561064.92173614705070540368454568168": -ildouble: 6 -ldouble: 6 +ildouble: 2 +ldouble: 2 # sinh_towardzero Test "sinh_towardzero (22) == 1792456423.065795780701106568345764104225": @@ -1929,31 +1916,31 @@ double: 1 float: 1 idouble: 1 ifloat: 1 -ildouble: 4 -ldouble: 4 +ildouble: 2 +ldouble: 2 Test "sinh_towardzero (23) == 4872401723.124451299966006944252978187305": double: 1 float: 1 idouble: 1 ifloat: 1 +ildouble: 2 +ldouble: 2 Test "sinh_towardzero (24) == 13244561064.92173614705070540368454568168": float: 1 ifloat: 1 -ildouble: 5 -ldouble: 5 +ildouble: 2 +ldouble: 2 # sinh_upward Test "sinh_upward (22) == 1792456423.065795780701106568345764104225": -ildouble: 16 -ldouble: 16 +ildouble: 1 +ldouble: 1 Test "sinh_upward (23) == 4872401723.124451299966006944252978187305": -ildouble: 27 -ldouble: 27 +ildouble: 1 +ldouble: 1 Test "sinh_upward (24) == 13244561064.92173614705070540368454568168": double: 1 idouble: 1 -ildouble: 7 -ldouble: 7 # tan Test "tan (0x1p16383) == 0.422722393732022337800504160054440141575": @@ -2587,9 +2574,6 @@ ifloat: 1 ildouble: 1 ldouble: 1 -Function: "cosh": -ildouble: 1 - Function: "cosh_downward": double: 1 float: 1 @@ -2862,26 +2846,22 @@ double: 1 float: 1 idouble: 1 ifloat: 1 -ildouble: 5 -ldouble: 5 - -Function: "sinh_tonearest": -ildouble: 6 -ldouble: 6 +ildouble: 2 +ldouble: 2 Function: "sinh_towardzero": double: 1 float: 1 idouble: 1 ifloat: 1 -ildouble: 5 -ldouble: 5 +ildouble: 2 +ldouble: 2 Function: "sinh_upward": double: 1 idouble: 1 -ildouble: 27 -ldouble: 27 +ildouble: 1 +ldouble: 1 Function: "tan": double: 1 diff --git a/sysdeps/i386/fpu/s_expm1l.S b/sysdeps/i386/fpu/s_expm1l.S index e91f18b694..7fbd99b0db 100644 --- a/sysdeps/i386/fpu/s_expm1l.S +++ b/sysdeps/i386/fpu/s_expm1l.S @@ -1,89 +1,2 @@ -/* ix87 specific implementation of exp(x)-1. - Copyright (C) 1996-1997, 2002, 2005, 2008, 2012 Free Software Foundation, Inc. - This file is part of the GNU C Library. - Contributed by Ulrich Drepper , 1996. - Based on code by John C. Bowman . - Corrections by H.J. Lu (hjl@gnu.ai.mit.edu), 1997. - - 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 - . */ - - /* Using: e^x - 1 = 2^(x * log2(e)) - 1 */ - -#include -#include - - .section .rodata - - .align ALIGNARG(4) - ASM_TYPE_DIRECTIVE(minus1,@object) -minus1: .double -1.0 - ASM_SIZE_DIRECTIVE(minus1) - ASM_TYPE_DIRECTIVE(one,@object) -one: .double 1.0 - ASM_SIZE_DIRECTIVE(one) - ASM_TYPE_DIRECTIVE(l2e,@object) -l2e: .tfloat 1.442695040888963407359924681002 - ASM_SIZE_DIRECTIVE(l2e) - -#ifdef PIC -#define MO(op) op##@GOTOFF(%edx) -#else -#define MO(op) op -#endif - - .text -ENTRY(__expm1l) - movzwl 4+8(%esp), %eax // load sign bit and 15-bit exponent - xorb $0x80, %ah // invert sign bit (now 1 is "positive") - cmpl $0xc006, %eax // is num positive and exp >= 6 (number is >= 128.0)? - jae HIDDEN_JUMPTARGET (__expl) // (if num is denormal, it is at least >= 64.0) - - fldt 4(%esp) // x - fxam // Is NaN or +-Inf? - fstsw %ax - movb $0x45, %ch - andb %ah, %ch - cmpb $0x40, %ch - je 3f // If +-0, jump. -#ifdef PIC - LOAD_PIC_REG (dx) -#endif - cmpb $0x05, %ch - je 2f // If +-Inf, jump. - - fldt MO(l2e) // log2(e) : x - fmulp // log2(e)*x - fld %st // log2(e)*x : log2(e)*x - frndint // int(log2(e)*x) : log2(e)*x - fsubr %st, %st(1) // int(log2(e)*x) : fract(log2(e)*x) - fxch // fract(log2(e)*x) : int(log2(e)*x) - f2xm1 // 2^fract(log2(e)*x)-1 : int(log2(e)*x) - fscale // 2^(log2(e)*x)-2^int(log2(e)*x) : int(log2(e)*x) - fxch // int(log2(e)*x) : 2^(log2(e)*x)-2^int(log2(e)*x) - fldl MO(one) // 1 : int(log2(e)*x) : 2^(log2(e)*x)-2^int(log2(e)*x) - fscale // 2^int(log2(e)*x) : int(log2(e)*x) : 2^(log2(e)*x)-2^int(log2(e)*x) - fsubrl MO(one) // 1-2^int(log2(e)*x) : int(log2(e)*x) : 2^(log2(e)*x)-2^int(log2(e)*x) - fstp %st(1) // 1-2^int(log2(e)*x) : 2^(log2(e)*x)-2^int(log2(e)*x) - fsubrp %st, %st(1) // 2^(log2(e)*x) - ret - -2: testl $0x200, %eax // Test sign. - jz 3f // If positive, jump. - fstp %st - fldl MO(minus1) // Set result to -1.0. -3: ret -END(__expm1l) -libm_hidden_def (__expm1l) -weak_alias (__expm1l, expm1l) +#define USE_AS_EXPM1L +#include -- cgit 1.4.1