diff options
author | David S. Miller <davem@davemloft.net> | 2013-01-14 21:47:29 -0800 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2013-01-14 21:47:29 -0800 |
commit | 8b954ab9b808a51d8cf6a90c7d1d46a3366e3274 (patch) | |
tree | cbf6dba7b97106b9ce476078cda8e8f0a1e7bf7c /sysdeps/sparc | |
parent | 65a82e3dd538aabcdbeacf87fcee32f4e355e5b6 (diff) | |
download | glibc-8b954ab9b808a51d8cf6a90c7d1d46a3366e3274.zip glibc-8b954ab9b808a51d8cf6a90c7d1d46a3366e3274.tar.gz glibc-8b954ab9b808a51d8cf6a90c7d1d46a3366e3274.tar.bz2 |
Optimize sparc {ceil,floor}{,f} using vis2 'siam' instruction.
* sysdeps/sparc/sparc-ifunc.h (SPARC_ASM_IFUNC2): New macro.
(SPARC_ASM_VIS2_IFUNC): Likewise.
(SPARC_ASM_VIS3_VIS2_IFUNC): Likewise.
* sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil-vis3.S: Make
use of 'siam' instruction.
* sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf-vis3.S:
Likewise.
* sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor-vis3.S:
Likewise.
* sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf-vis3.S:
Likewise.
* sysdeps/sparc/sparc64/fpu/multiarch/s_ceil-vis3.S: Likewise.
* sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf-vis3.S: Likewise.
* sysdeps/sparc/sparc64/fpu/multiarch/s_floor-vis3.S: Likewise.
* sysdeps/sparc/sparc64/fpu/multiarch/s_floorf-vis3.S: Likewise.
* sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil-vis2.S: New
file.
* sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf-vis2.S: New
file.
* sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor-vis2.S: New
file.
* sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf-vis2.S: New
file.
* sysdeps/sparc/sparc64/fpu/multiarch/s_ceil-vis2.S: New file.
* sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf-vis2.S: New file.
* sysdeps/sparc/sparc64/fpu/multiarch/s_floor-vis2.S: New file.
* sysdeps/sparc/sparc64/fpu/multiarch/s_floorf-vis2.S: New file.
* sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil.S: Hook in
new VIS2 routines.
* sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf.S: Likewise.
* sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor.S: Likewise.
* sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf.S:
Likewise.
* sysdeps/sparc/sparc64/fpu/multiarch/s_ceil.S: Likewise.
* sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf.S: Likewise.
* sysdeps/sparc/sparc64/fpu/multiarch/s_floor.S: Likewise.
* sysdeps/sparc/sparc64/fpu/multiarch/s_floorf.S: Likewise.
* sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/Makefile: Add new VIS2
routines to libm-sysdep_routines.
* sysdeps/sparc/sparc64/fpu/multiarch/Makefile: Likewise.
Diffstat (limited to 'sysdeps/sparc')
27 files changed, 629 insertions, 223 deletions
diff --git a/sysdeps/sparc/sparc-ifunc.h b/sysdeps/sparc/sparc-ifunc.h index edff5c8..f68161f 100644 --- a/sysdeps/sparc/sparc-ifunc.h +++ b/sysdeps/sparc/sparc-ifunc.h @@ -51,6 +51,33 @@ ENTRY (__##name) \ mov %o1, %o0; \ END (__##name) +# define SPARC_ASM_IFUNC2(name, m1, f1, m2, f2, dflt) \ +ENTRY (__##name) \ + .type __##name, @gnu_indirect_function; \ + SETUP_PIC_REG_LEAF(o3, o5); \ + set m1, %o1; \ + andcc %o0, %o1, %g0; \ + be 8f; \ + nop; \ + sethi %gdop_hix22(f1), %o1; \ + xor %o1, %gdop_lox10(f1), %o1; \ + ba 10f; \ + nop; \ +8: set m2, %o1; \ + andcc %o0, %o1, %g0; \ + be 9f; \ + nop; \ + sethi %gdop_hix22(f2), %o1; \ + xor %o1, %gdop_lox10(f2), %o1; \ + ba 10f; \ + nop; \ +9: sethi %gdop_hix22(dflt), %o1; \ + xor %o1, %gdop_lox10(dflt), %o1; \ +10: add %o3, %o1, %o1; \ + retl; \ + mov %o1, %o0; \ +END (__##name) + # else /* SHARED */ # ifdef __arch64__ @@ -82,19 +109,54 @@ ENTRY (__##name) \ mov %o1, %o0; \ END (__##name) +# define SPARC_ASM_IFUNC2(name, m1, f1, m2, f2, dflt) \ +ENTRY (__##name) \ + .type __##name, @gnu_indirect_function; \ + set m1, %o1; \ + andcc %o0, %o1, %g0; \ + be 8f; \ + nop; \ + SET(f1, %g1, %o1); \ + ba 10f; \ + nop; \ +8: set m2, %o1; \ + andcc %o0, %o1, %g0; \ + be 9f; \ + nop; \ + SET(f2, %g1, %o1); \ + ba 10f; \ + nop; \ +9: SET(dflt, %g1, %o1); \ +10: retl; \ + mov %o1, %o0; \ +END (__##name) + # endif /* SHARED */ +#define SPARC_ASM_VIS2_IFUNC(name) \ + SPARC_ASM_IFUNC1(name, HWCAP_SPARC_VIS2, \ + __##name##_vis2, __##name##_generic) + # ifdef HAVE_AS_VIS3_SUPPORT #define SPARC_ASM_VIS3_IFUNC(name) \ SPARC_ASM_IFUNC1(name, HWCAP_SPARC_VIS3, \ __##name##_vis3, __##name##_generic) +#define SPARC_ASM_VIS3_VIS2_IFUNC(name) \ + SPARC_ASM_IFUNC2(name, HWCAP_SPARC_VIS3, \ + __##name##_vis3, \ + HWCAP_SPARC_VIS2, \ + __##name##_vis2, __##name##_generic) + # else /* HAVE_AS_VIS3_SUPPORT */ #define SPARC_ASM_VIS3_IFUNC(name) \ SPARC_ASM_IFUNC_DFLT(name, __##name##_generic) +#define SPARC_ASM_VIS3_VIS2_IFUNC(name) \ + SPARC_ASM_VIS2_IFUNC(name) + # endif /* HAVE_AS_VIS3_SUPPORT */ diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/Makefile b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/Makefile index 6349681..561b0ee 100644 --- a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/Makefile +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/Makefile @@ -1,4 +1,6 @@ ifeq ($(subdir),math) +libm-sysdep_routines += s_ceil-vis2 s_ceilf-vis2 \ + s_floor-vis2 s_floorf-vis2 ifeq ($(have-as-vis3),yes) libm-sysdep_routines += m_copysignf-vis3 m_copysign-vis3 s_ceilf-vis3 \ s_ceil-vis3 s_fabs-vis3 s_fabsf-vis3 s_floor-vis3 \ diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil-vis2.S b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil-vis2.S new file mode 100644 index 0000000..9438800 --- /dev/null +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil-vis2.S @@ -0,0 +1,61 @@ +/* ceil function, sparc32 v9 vis2 version. + Copyright (C) 2013 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by David S. Miller <davem@davemloft.net>, 2013. + + 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 <sysdep.h> + + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. + + We add then subtract (or subtract than add if the initial + value was negative) 2**23 to the value, then subtract it + back out. + + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. */ + +#define TWO_FIFTYTWO 0x43300000 /* 2**52 */ + +#define ZERO %f10 /* 0.0 */ +#define SIGN_BIT %f12 /* -0.0 */ + +ENTRY (__ceil_vis2) + sethi %hi(TWO_FIFTYTWO), %o2 + sllx %o0, 32, %o0 + or %o0, %o1, %o0 + stx %o0, [%sp + 72] + sllx %o2, 32, %o2 + fzero ZERO + ldd [%sp + 72], %f0 + fnegd ZERO, SIGN_BIT + stx %o2, [%sp + 72] + fabsd %f0, %f14 + ldd [%sp + 72], %f16 + fcmpd %fcc3, %f14, %f16 + fmovduge %fcc3, ZERO, %f16 + fand %f0, SIGN_BIT, SIGN_BIT + for %f16, SIGN_BIT, %f16 + siam (1 << 2) | 2 + faddd %f0, %f16, %f18 + siam (1 << 2) | 0 + fsubd %f18, %f16, %f18 + siam (0 << 2) + retl + for %f18, SIGN_BIT, %f0 +END (__ceil_vis2) diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil-vis3.S b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil-vis3.S index 0c2140d..aebff5c 100644 --- a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil-vis3.S +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil-vis3.S @@ -19,27 +19,21 @@ #include <sysdep.h> - /* Since changing the rounding mode is extremely expensive, we - try to round up using a method that is rounding mode - agnostic. + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. We add then subtract (or subtract than add if the initial value was negative) 2**23 to the value, then subtract it back out. - This will clear out the fractional portion of the value. - One of two things will happen for non-whole initial values. - Either the rounding mode will round it up, or it will be - rounded down. If the value started out whole, it will be - equal after the addition and subtraction. This means we - can accurately detect with one test whether we need to add - another 1.0 to round it up properly. + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. - VIS instructions are used to facilitate the formation of - easier constants, and the propagation of the sign bit. */ + We also use VIS3 moves to avoid using the stack to transfer + values between float and integer registers. */ #define TWO_FIFTYTWO 0x43300000 /* 2**52 */ -#define ONE_DOT_ZERO 0x3ff00000 /* 1.0 */ #define ZERO %f10 /* 0.0 */ #define SIGN_BIT %f12 /* -0.0 */ @@ -47,32 +41,22 @@ ENTRY (__ceil_vis3) sethi %hi(TWO_FIFTYTWO), %o2 sllx %o0, 32, %o0 - sethi %hi(ONE_DOT_ZERO), %o3 + sllx %o2, 32, %o2 or %o0, %o1, %o0 movxtod %o0, %f0 - sllx %o2, 32, %o2 fzero ZERO - sllx %o3, 32, %o3 - fnegd ZERO, SIGN_BIT - movxtod %o2, %f16 fabsd %f0, %f14 - fcmpd %fcc3, %f14, %f16 - fmovduge %fcc3, ZERO, %f16 fand %f0, SIGN_BIT, SIGN_BIT - for %f16, SIGN_BIT, %f16 + siam (1 << 2) | 2 faddd %f0, %f16, %f18 + siam (1 << 2) | 0 fsubd %f18, %f16, %f18 - fcmpd %fcc2, %f18, %f0 - movxtod %o3, %f20 - - fmovduge %fcc2, ZERO, %f20 - faddd %f18, %f20, %f0 - fabsd %f0, %f0 + siam (0 << 2) retl - for %f0, SIGN_BIT, %f0 + for %f18, SIGN_BIT, %f0 END (__ceil_vis3) diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil.S b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil.S index 835703f..efc8d49 100644 --- a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil.S +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceil.S @@ -1,7 +1,7 @@ #include <sparc-ifunc.h> #include <math_ldbl_opt.h> -SPARC_ASM_VIS3_IFUNC(ceil) +SPARC_ASM_VIS3_VIS2_IFUNC(ceil) weak_alias (__ceil, ceil) diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf-vis2.S b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf-vis2.S new file mode 100644 index 0000000..bc51676 --- /dev/null +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf-vis2.S @@ -0,0 +1,58 @@ +/* Float ceil function, sparc32 v9 vis2 version. + Copyright (C) 2013 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by David S. Miller <davem@davemloft.net>, 2013. + + 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 <sysdep.h> + + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. + + We add then subtract (or subtract than add if the initial + value was negative) 2**23 to the value, then subtract it + back out. + + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. */ + +#define TWO_TWENTYTHREE 0x4b000000 /* 2**23 */ + +#define ZERO %f10 /* 0.0 */ +#define SIGN_BIT %f12 /* -0.0 */ + +ENTRY (__ceilf_vis2) + st %o0, [%sp + 68] + sethi %hi(TWO_TWENTYTHREE), %o2 + fzeros ZERO + ld [%sp + 68], %f0 + fnegs ZERO, SIGN_BIT + st %o2, [%sp + 68] + fabss %f0, %f14 + ld [%sp + 68], %f16 + fcmps %fcc3, %f14, %f16 + fmovsuge %fcc3, ZERO, %f16 + fands %f0, SIGN_BIT, SIGN_BIT + fors %f16, SIGN_BIT, %f16 + siam (1 << 2) | 2 + fadds %f0, %f16, %f1 + siam (1 << 2) | 0 + fsubs %f1, %f16, %f1 + siam (0 << 2) + retl + fors %f1, SIGN_BIT, %f0 +END (__ceilf_vis2) diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf-vis3.S b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf-vis3.S index 7d30c0b..0a6768c 100644 --- a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf-vis3.S +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf-vis3.S @@ -19,27 +19,21 @@ #include <sysdep.h> - /* Since changing the rounding mode is extremely expensive, we - try to round up using a method that is rounding mode - agnostic. + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. We add then subtract (or subtract than add if the initial value was negative) 2**23 to the value, then subtract it back out. - This will clear out the fractional portion of the value. - One of two things will happen for non-whole initial values. - Either the rounding mode will round it up, or it will be - rounded down. If the value started out whole, it will be - equal after the addition and subtraction. This means we - can accurately detect with one test whether we need to add - another 1.0 to round it up properly. + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. - VIS instructions are used to facilitate the formation of - easier constants, and the propagation of the sign bit. */ + We also use VIS3 moves to avoid using the stack to transfer + values between float and integer registers. */ #define TWO_TWENTYTHREE 0x4b000000 /* 2**23 */ -#define ONE_DOT_ZERO 0x3f800000 /* 1.0 */ #define ZERO %f10 /* 0.0 */ #define SIGN_BIT %f12 /* -0.0 */ @@ -47,28 +41,19 @@ ENTRY (__ceilf_vis3) movwtos %o0, %f0 sethi %hi(TWO_TWENTYTHREE), %o2 - sethi %hi(ONE_DOT_ZERO), %o3 fzeros ZERO - fnegs ZERO, SIGN_BIT - movwtos %o2, %f16 fabss %f0, %f14 - fcmps %fcc3, %f14, %f16 - fmovsuge %fcc3, ZERO, %f16 fands %f0, SIGN_BIT, SIGN_BIT - fors %f16, SIGN_BIT, %f16 + siam (1 << 2) | 2 fadds %f0, %f16, %f1 + siam (1 << 2) | 0 fsubs %f1, %f16, %f1 - fcmps %fcc2, %f1, %f0 - movwtos %o3, %f9 - - fmovsuge %fcc2, ZERO, %f9 - fadds %f1, %f9, %f0 - fabss %f0, %f0 + siam (0 << 2) retl - fors %f0, SIGN_BIT, %f0 + fors %f1, SIGN_BIT, %f0 END (__ceilf_vis3) diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf.S b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf.S index 3047dd8..1c72a57 100644 --- a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf.S +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_ceilf.S @@ -1,6 +1,6 @@ #include <sparc-ifunc.h> -SPARC_ASM_VIS3_IFUNC(ceilf) +SPARC_ASM_VIS3_VIS2_IFUNC(ceilf) weak_alias (__ceilf, ceilf) diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor-vis2.S b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor-vis2.S new file mode 100644 index 0000000..3b5e8fd --- /dev/null +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor-vis2.S @@ -0,0 +1,61 @@ +/* floor function, sparc32 v9 vis2 version. + Copyright (C) 2013 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by David S. Miller <davem@davemloft.net>, 2013. + + 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 <sysdep.h> + + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. + + We add then subtract (or subtract than add if the initial + value was negative) 2**23 to the value, then subtract it + back out. + + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. */ + +#define TWO_FIFTYTWO 0x43300000 /* 2**52 */ + +#define ZERO %f10 /* 0.0 */ +#define SIGN_BIT %f12 /* -0.0 */ + +ENTRY (__floor_vis2) + sethi %hi(TWO_FIFTYTWO), %o2 + sllx %o0, 32, %o0 + or %o0, %o1, %o0 + stx %o0, [%sp + 72] + sllx %o2, 32, %o2 + fzero ZERO + ldd [%sp + 72], %f0 + fnegd ZERO, SIGN_BIT + stx %o2, [%sp + 72] + fabsd %f0, %f14 + ldd [%sp + 72], %f16 + fcmpd %fcc3, %f14, %f16 + fmovduge %fcc3, ZERO, %f16 + fand %f0, SIGN_BIT, SIGN_BIT + for %f16, SIGN_BIT, %f16 + siam (1 << 2) | 3 + faddd %f0, %f16, %f18 + siam (1 << 2) | 0 + fsubd %f18, %f16, %f18 + siam (0 << 2) + retl + for %f18, SIGN_BIT, %f0 +END (__floor_vis2) diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor-vis3.S b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor-vis3.S index 8445f1d..41fdfac 100644 --- a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor-vis3.S +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor-vis3.S @@ -19,27 +19,21 @@ #include <sysdep.h> - /* Since changing the rounding mode is extremely expensive, we - try to round up using a method that is rounding mode - agnostic. + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. We add then subtract (or subtract than add if the initial value was negative) 2**23 to the value, then subtract it back out. - This will clear out the fractional portion of the value. - One of two things will happen for non-whole initial values. - Either the rounding mode will round it up, or it will be - rounded down. If the value started out whole, it will be - equal after the addition and subtraction. This means we - can accurately detect with one test whether we need to add - another 1.0 to round it up properly. + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. - VIS instructions are used to facilitate the formation of - easier constants, and the propagation of the sign bit. */ + We also use VIS3 moves to avoid using the stack to transfer + values between float and integer registers. */ #define TWO_FIFTYTWO 0x43300000 /* 2**52 */ -#define ONE_DOT_ZERO 0x3ff00000 /* 1.0 */ #define ZERO %f10 /* 0.0 */ #define SIGN_BIT %f12 /* -0.0 */ @@ -47,32 +41,22 @@ ENTRY (__floor_vis3) sethi %hi(TWO_FIFTYTWO), %o2 sllx %o0, 32, %o0 - sethi %hi(ONE_DOT_ZERO), %o3 + sllx %o2, 32, %o2 or %o0, %o1, %o0 movxtod %o0, %f0 - sllx %o2, 32, %o2 fzero ZERO - sllx %o3, 32, %o3 - fnegd ZERO, SIGN_BIT - movxtod %o2, %f16 fabsd %f0, %f14 - fcmpd %fcc3, %f14, %f16 - fmovduge %fcc3, ZERO, %f16 fand %f0, SIGN_BIT, SIGN_BIT - for %f16, SIGN_BIT, %f16 + siam (1 << 2) | 3 faddd %f0, %f16, %f18 + siam (1 << 2) | 0 fsubd %f18, %f16, %f18 - fcmpd %fcc2, %f18, %f0 - movxtod %o3, %f20 - - fmovdule %fcc2, ZERO, %f20 - fsubd %f18, %f20, %f0 - fabsd %f0, %f0 + siam (0 << 2) retl - for %f0, SIGN_BIT, %f0 + for %f18, SIGN_BIT, %f0 END (__floor_vis3) diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor.S b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor.S index 37aeb43..1fe4b95 100644 --- a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor.S +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floor.S @@ -1,7 +1,7 @@ #include <sparc-ifunc.h> #include <math_ldbl_opt.h> -SPARC_ASM_VIS3_IFUNC(floor) +SPARC_ASM_VIS3_VIS2_IFUNC(floor) weak_alias (__floor, floor) diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf-vis2.S b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf-vis2.S new file mode 100644 index 0000000..4f73121 --- /dev/null +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf-vis2.S @@ -0,0 +1,58 @@ +/* Float floor function, sparc32 v9 vis2 version. + Copyright (C) 2013 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by David S. Miller <davem@davemloft.net>, 2013. + + 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 <sysdep.h> + + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. + + We add then subtract (or subtract than add if the initial + value was negative) 2**23 to the value, then subtract it + back out. + + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. */ + +#define TWO_TWENTYTHREE 0x4b000000 /* 2**23 */ + +#define ZERO %f10 /* 0.0 */ +#define SIGN_BIT %f12 /* -0.0 */ + +ENTRY (__floorf_vis2) + st %o0, [%sp + 68] + sethi %hi(TWO_TWENTYTHREE), %o2 + fzeros ZERO + ld [%sp + 68], %f0 + fnegs ZERO, SIGN_BIT + st %o2, [%sp + 68] + fabss %f0, %f14 + ld [%sp + 68], %f16 + fcmps %fcc3, %f14, %f16 + fmovsuge %fcc3, ZERO, %f16 + fands %f0, SIGN_BIT, SIGN_BIT + fors %f16, SIGN_BIT, %f16 + siam (1 << 2) | 3 + fadds %f0, %f16, %f1 + siam (1 << 2) | 0 + fsubs %f1, %f16, %f1 + siam (0 << 2) + retl + fors %f1, SIGN_BIT, %f0 +END (__floorf_vis2) diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf-vis3.S b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf-vis3.S index 133a0a4..fe2d2da 100644 --- a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf-vis3.S +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf-vis3.S @@ -19,27 +19,21 @@ #include <sysdep.h> - /* Since changing the rounding mode is extremely expensive, we - try to round up using a method that is rounding mode - agnostic. + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. We add then subtract (or subtract than add if the initial value was negative) 2**23 to the value, then subtract it back out. - This will clear out the fractional portion of the value. - One of two things will happen for non-whole initial values. - Either the rounding mode will round it up, or it will be - rounded down. If the value started out whole, it will be - equal after the addition and subtraction. This means we - can accurately detect with one test whether we need to add - another 1.0 to round it up properly. + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. - VIS instructions are used to facilitate the formation of - easier constants, and the propagation of the sign bit. */ + We also use VIS3 moves to avoid using the stack to transfer + values between float and integer registers. */ #define TWO_TWENTYTHREE 0x4b000000 /* 2**23 */ -#define ONE_DOT_ZERO 0x3f800000 /* 1.0 */ #define ZERO %f10 /* 0.0 */ #define SIGN_BIT %f12 /* -0.0 */ @@ -47,28 +41,19 @@ ENTRY (__floorf_vis3) movwtos %o0, %f0 sethi %hi(TWO_TWENTYTHREE), %o2 - sethi %hi(ONE_DOT_ZERO), %o3 fzeros ZERO - fnegs ZERO, SIGN_BIT - movwtos %o2, %f16 fabss %f0, %f14 - fcmps %fcc3, %f14, %f16 - fmovsuge %fcc3, ZERO, %f16 fands %f0, SIGN_BIT, SIGN_BIT - fors %f16, SIGN_BIT, %f16 + siam (1 << 2) | 3 fadds %f0, %f16, %f1 + siam (1 << 2) | 0 fsubs %f1, %f16, %f1 - fcmps %fcc2, %f1, %f0 - movwtos %o3, %f9 - - fmovsule %fcc2, ZERO, %f9 - fsubs %f1, %f9, %f0 - fabss %f0, %f0 + siam (0 << 2) retl - fors %f0, SIGN_BIT, %f0 + fors %f1, SIGN_BIT, %f0 END (__floorf_vis3) diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf.S b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf.S index 31cda38..d2a83cb 100644 --- a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf.S +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_floorf.S @@ -1,6 +1,6 @@ #include <sparc-ifunc.h> -SPARC_ASM_VIS3_IFUNC(floorf) +SPARC_ASM_VIS3_VIS2_IFUNC(floorf) weak_alias (__floorf, floorf) diff --git a/sysdeps/sparc/sparc64/fpu/multiarch/Makefile b/sysdeps/sparc/sparc64/fpu/multiarch/Makefile index 8345848..7a5a9dd 100644 --- a/sysdeps/sparc/sparc64/fpu/multiarch/Makefile +++ b/sysdeps/sparc/sparc64/fpu/multiarch/Makefile @@ -1,4 +1,6 @@ ifeq ($(subdir),math) +libm-sysdep_routines += s_ceil-vis2 s_ceilf-vis2 \ + s_floor-vis2 s_floorf-vis2 ifeq ($(have-as-vis3),yes) libm-sysdep_routines += m_signbitf-vis3 m_signbit-vis3 s_ceilf-vis3 \ s_ceil-vis3 m_finitef-vis3 m_finite-vis3 \ diff --git a/sysdeps/sparc/sparc64/fpu/multiarch/s_ceil-vis2.S b/sysdeps/sparc/sparc64/fpu/multiarch/s_ceil-vis2.S new file mode 100644 index 0000000..50d96a6 --- /dev/null +++ b/sysdeps/sparc/sparc64/fpu/multiarch/s_ceil-vis2.S @@ -0,0 +1,57 @@ +/* ceil function, sparc64 vis2 version. + Copyright (C) 2012-2013 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by David S. Miller <davem@davemloft.net>, 2012. + + 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 <sysdep.h> + + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. + + We add then subtract (or subtract than add if the initial + value was negative) 2**23 to the value, then subtract it + back out. + + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. */ + +#define TWO_FIFTYTWO 0x43300000 /* 2**52 */ + +#define ZERO %f10 /* 0.0 */ +#define SIGN_BIT %f12 /* -0.0 */ + +ENTRY (__ceil_vis2) + sethi %hi(TWO_FIFTYTWO), %o2 + fzero ZERO + sllx %o2, 32, %o2 + fnegd ZERO, SIGN_BIT + stx %o2, [%sp + STACK_BIAS + 128] + fabsd %f0, %f14 + ldd [%sp + STACK_BIAS + 128], %f16 + fcmpd %fcc3, %f14, %f16 + fmovduge %fcc3, ZERO, %f16 + fand %f0, SIGN_BIT, SIGN_BIT + for %f16, SIGN_BIT, %f16 + siam (1 << 2) | 2 + faddd %f0, %f16, %f18 + siam (1 << 2) | 0 + fsubd %f18, %f16, %f18 + siam (0 << 2) + retl + for %f18, SIGN_BIT, %f0 +END (__ceil_vis2) diff --git a/sysdeps/sparc/sparc64/fpu/multiarch/s_ceil-vis3.S b/sysdeps/sparc/sparc64/fpu/multiarch/s_ceil-vis3.S index 9a598ea..6acff09 100644 --- a/sysdeps/sparc/sparc64/fpu/multiarch/s_ceil-vis3.S +++ b/sysdeps/sparc/sparc64/fpu/multiarch/s_ceil-vis3.S @@ -19,57 +19,41 @@ #include <sysdep.h> - /* Since changing the rounding mode is extremely expensive, we - try to round up using a method that is rounding mode - agnostic. + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. We add then subtract (or subtract than add if the initial value was negative) 2**23 to the value, then subtract it back out. - This will clear out the fractional portion of the value. - One of two things will happen for non-whole initial values. - Either the rounding mode will round it up, or it will be - rounded down. If the value started out whole, it will be - equal after the addition and subtraction. This means we - can accurately detect with one test whether we need to add - another 1.0 to round it up properly. + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. - VIS instructions are used to facilitate the formation of - easier constants, and the propagation of the sign bit. */ + We also use VIS3 moves to avoid using the stack to transfer + values between float and integer registers. */ #define TWO_FIFTYTWO 0x43300000 /* 2**52 */ -#define ONE_DOT_ZERO 0x3ff00000 /* 1.0 */ #define ZERO %f10 /* 0.0 */ #define SIGN_BIT %f12 /* -0.0 */ ENTRY (__ceil_vis3) sethi %hi(TWO_FIFTYTWO), %o2 - sethi %hi(ONE_DOT_ZERO), %o3 fzero ZERO - sllx %o2, 32, %o2 fnegd ZERO, SIGN_BIT - - sllx %o3, 32, %o3 movxtod %o2, %f16 fabsd %f0, %f14 - fcmpd %fcc3, %f14, %f16 - fmovduge %fcc3, ZERO, %f16 fand %f0, SIGN_BIT, SIGN_BIT - for %f16, SIGN_BIT, %f16 + siam (1 << 2) | 2 faddd %f0, %f16, %f18 + siam (1 << 2) | 0 fsubd %f18, %f16, %f18 - fcmpd %fcc2, %f18, %f0 - movxtod %o3, %f20 - - fmovduge %fcc2, ZERO, %f20 - faddd %f18, %f20, %f0 - fabsd %f0, %f0 + siam (0 << 2) retl - for %f0, SIGN_BIT, %f0 + for %f18, SIGN_BIT, %f0 END (__ceil_vis3) diff --git a/sysdeps/sparc/sparc64/fpu/multiarch/s_ceil.S b/sysdeps/sparc/sparc64/fpu/multiarch/s_ceil.S index fa4c7c3..e7822bc 100644 --- a/sysdeps/sparc/sparc64/fpu/multiarch/s_ceil.S +++ b/sysdeps/sparc/sparc64/fpu/multiarch/s_ceil.S @@ -1,6 +1,6 @@ #include <sparc-ifunc.h> -SPARC_ASM_VIS3_IFUNC(ceil) +SPARC_ASM_VIS3_VIS2_IFUNC(ceil) weak_alias (__ceil, ceil) diff --git a/sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf-vis2.S b/sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf-vis2.S new file mode 100644 index 0000000..cd5937b --- /dev/null +++ b/sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf-vis2.S @@ -0,0 +1,56 @@ +/* Float ceil function, sparc64 vis2 version. + Copyright (C) 2012-2013 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by David S. Miller <davem@davemloft.net>, 2012. + + 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 <sysdep.h> + + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. + + We add then subtract (or subtract than add if the initial + value was negative) 2**23 to the value, then subtract it + back out. + + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. */ + +#define TWO_TWENTYTHREE 0x4b000000 /* 2**23 */ + +#define ZERO %f10 /* 0.0 */ +#define SIGN_BIT %f12 /* -0.0 */ + +ENTRY (__ceilf_vis2) + sethi %hi(TWO_TWENTYTHREE), %o2 + fzeros ZERO + fnegs ZERO, SIGN_BIT + st %o2, [%sp + STACK_BIAS + 128] + fabss %f1, %f14 + ld [%sp + STACK_BIAS + 128], %f16 + fcmps %fcc3, %f14, %f16 + fmovsuge %fcc3, ZERO, %f16 + fands %f1, SIGN_BIT, SIGN_BIT + fors %f16, SIGN_BIT, %f16 + siam (1 << 2) | 2 + fadds %f1, %f16, %f5 + siam (1 << 2) | 0 + fsubs %f5, %f16, %f5 + siam (0 << 2) + retl + fors %f5, SIGN_BIT, %f0 +END (__ceilf_vis2) diff --git a/sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf-vis3.S b/sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf-vis3.S index 8c63590..b3ec348 100644 --- a/sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf-vis3.S +++ b/sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf-vis3.S @@ -19,55 +19,40 @@ #include <sysdep.h> - /* Since changing the rounding mode is extremely expensive, we - try to round up using a method that is rounding mode - agnostic. + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. We add then subtract (or subtract than add if the initial value was negative) 2**23 to the value, then subtract it back out. - This will clear out the fractional portion of the value. - One of two things will happen for non-whole initial values. - Either the rounding mode will round it up, or it will be - rounded down. If the value started out whole, it will be - equal after the addition and subtraction. This means we - can accurately detect with one test whether we need to add - another 1.0 to round it up properly. + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. - VIS instructions are used to facilitate the formation of - easier constants, and the propagation of the sign bit. */ + We also use VIS3 moves to avoid using the stack to transfer + values between float and integer registers. */ #define TWO_TWENTYTHREE 0x4b000000 /* 2**23 */ -#define ONE_DOT_ZERO 0x3f800000 /* 1.0 */ #define ZERO %f10 /* 0.0 */ #define SIGN_BIT %f12 /* -0.0 */ ENTRY (__ceilf_vis3) sethi %hi(TWO_TWENTYTHREE), %o2 - sethi %hi(ONE_DOT_ZERO), %o3 fzeros ZERO - fnegs ZERO, SIGN_BIT - movwtos %o2, %f16 fabss %f1, %f14 - fcmps %fcc3, %f14, %f16 - fmovsuge %fcc3, ZERO, %f16 fands %f1, SIGN_BIT, SIGN_BIT - fors %f16, SIGN_BIT, %f16 + siam (1 << 2) | 2 fadds %f1, %f16, %f5 + siam (1 << 2) | 0 fsubs %f5, %f16, %f5 - fcmps %fcc2, %f5, %f1 - movwtos %o3, %f9 - - fmovsuge %fcc2, ZERO, %f9 - fadds %f5, %f9, %f0 - fabss %f0, %f0 + siam (0 << 2) retl - fors %f0, SIGN_BIT, %f0 + fors %f5, SIGN_BIT, %f0 END (__ceilf_vis3) diff --git a/sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf.S b/sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf.S index 3047dd8..1c72a57 100644 --- a/sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf.S +++ b/sysdeps/sparc/sparc64/fpu/multiarch/s_ceilf.S @@ -1,6 +1,6 @@ #include <sparc-ifunc.h> -SPARC_ASM_VIS3_IFUNC(ceilf) +SPARC_ASM_VIS3_VIS2_IFUNC(ceilf) weak_alias (__ceilf, ceilf) diff --git a/sysdeps/sparc/sparc64/fpu/multiarch/s_floor-vis2.S b/sysdeps/sparc/sparc64/fpu/multiarch/s_floor-vis2.S new file mode 100644 index 0000000..5479ced --- /dev/null +++ b/sysdeps/sparc/sparc64/fpu/multiarch/s_floor-vis2.S @@ -0,0 +1,57 @@ +/* floor function, sparc64 vis2 version. + Copyright (C) 2012-2013 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by David S. Miller <davem@davemloft.net>, 2012. + + 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 <sysdep.h> + + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. + + We add then subtract (or subtract than add if the initial + value was negative) 2**23 to the value, then subtract it + back out. + + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. */ + +#define TWO_FIFTYTWO 0x43300000 /* 2**52 */ + +#define ZERO %f10 /* 0.0 */ +#define SIGN_BIT %f12 /* -0.0 */ + +ENTRY (__floor_vis2) + sethi %hi(TWO_FIFTYTWO), %o2 + fzero ZERO + sllx %o2, 32, %o2 + fnegd ZERO, SIGN_BIT + stx %o2, [%sp + STACK_BIAS + 128] + fabsd %f0, %f14 + ldd [%sp + STACK_BIAS + 128], %f16 + fcmpd %fcc3, %f14, %f16 + fmovduge %fcc3, ZERO, %f16 + fand %f0, SIGN_BIT, SIGN_BIT + for %f16, SIGN_BIT, %f16 + siam (1 << 2) | 3 + faddd %f0, %f16, %f18 + siam (1 << 2) | 0 + fsubd %f18, %f16, %f18 + siam (0 << 2) + retl + for %f18, SIGN_BIT, %f0 +END (__floor_vis2) diff --git a/sysdeps/sparc/sparc64/fpu/multiarch/s_floor-vis3.S b/sysdeps/sparc/sparc64/fpu/multiarch/s_floor-vis3.S index 169d206..c2ffe9f 100644 --- a/sysdeps/sparc/sparc64/fpu/multiarch/s_floor-vis3.S +++ b/sysdeps/sparc/sparc64/fpu/multiarch/s_floor-vis3.S @@ -19,57 +19,41 @@ #include <sysdep.h> - /* Since changing the rounding mode is extremely expensive, we - try to round up using a method that is rounding mode - agnostic. + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. We add then subtract (or subtract than add if the initial value was negative) 2**23 to the value, then subtract it back out. - This will clear out the fractional portion of the value. - One of two things will happen for non-whole initial values. - Either the rounding mode will round it up, or it will be - rounded down. If the value started out whole, it will be - equal after the addition and subtraction. This means we - can accurately detect with one test whether we need to add - another 1.0 to round it up properly. + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. - VIS instructions are used to facilitate the formation of - easier constants, and the propagation of the sign bit. */ + We also use VIS3 moves to avoid using the stack to transfer + values between float and integer registers. */ #define TWO_FIFTYTWO 0x43300000 /* 2**52 */ -#define ONE_DOT_ZERO 0x3ff00000 /* 1.0 */ #define ZERO %f10 /* 0.0 */ #define SIGN_BIT %f12 /* -0.0 */ ENTRY (__floor_vis3) sethi %hi(TWO_FIFTYTWO), %o2 - sethi %hi(ONE_DOT_ZERO), %o3 fzero ZERO - sllx %o2, 32, %o2 fnegd ZERO, SIGN_BIT - - sllx %o3, 32, %o3 movxtod %o2, %f16 fabsd %f0, %f14 - fcmpd %fcc3, %f14, %f16 - fmovduge %fcc3, ZERO, %f16 fand %f0, SIGN_BIT, SIGN_BIT - for %f16, SIGN_BIT, %f16 + siam (1 << 2) | 3 faddd %f0, %f16, %f18 + siam (1 << 2) | 0 fsubd %f18, %f16, %f18 - fcmpd %fcc2, %f18, %f0 - movxtod %o3, %f20 - - fmovdule %fcc2, ZERO, %f20 - fsubd %f18, %f20, %f0 - fabsd %f0, %f0 + siam (0 << 2) retl - for %f0, SIGN_BIT, %f0 + for %f18, SIGN_BIT, %f0 END (__floor_vis3) diff --git a/sysdeps/sparc/sparc64/fpu/multiarch/s_floor.S b/sysdeps/sparc/sparc64/fpu/multiarch/s_floor.S index 6ae9947..989ccab 100644 --- a/sysdeps/sparc/sparc64/fpu/multiarch/s_floor.S +++ b/sysdeps/sparc/sparc64/fpu/multiarch/s_floor.S @@ -1,6 +1,6 @@ #include <sparc-ifunc.h> -SPARC_ASM_VIS3_IFUNC(floor) +SPARC_ASM_VIS3_VIS2_IFUNC(floor) weak_alias (__floor, floor) diff --git a/sysdeps/sparc/sparc64/fpu/multiarch/s_floorf-vis2.S b/sysdeps/sparc/sparc64/fpu/multiarch/s_floorf-vis2.S new file mode 100644 index 0000000..935fa85 --- /dev/null +++ b/sysdeps/sparc/sparc64/fpu/multiarch/s_floorf-vis2.S @@ -0,0 +1,56 @@ +/* Float floor function, sparc64 vis2 version. + Copyright (C) 2012-2013 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by David S. Miller <davem@davemloft.net>, 2012. + + 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 <sysdep.h> + + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. + + We add then subtract (or subtract than add if the initial + value was negative) 2**23 to the value, then subtract it + back out. + + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. */ + +#define TWO_TWENTYTHREE 0x4b000000 /* 2**23 */ + +#define ZERO %f10 /* 0.0 */ +#define SIGN_BIT %f12 /* -0.0 */ + +ENTRY (__floorf_vis2) + sethi %hi(TWO_TWENTYTHREE), %o2 + fzeros ZERO + fnegs ZERO, SIGN_BIT + st %o2, [%sp + STACK_BIAS + 128] + fabss %f1, %f14 + ld [%sp + STACK_BIAS + 128], %f16 + fcmps %fcc3, %f14, %f16 + fmovsuge %fcc3, ZERO, %f16 + fands %f1, SIGN_BIT, SIGN_BIT + fors %f16, SIGN_BIT, %f16 + siam (1 << 2) | 3 + fadds %f1, %f16, %f5 + siam (1 << 2) | 0 + fsubs %f5, %f16, %f5 + siam (0 << 2) + retl + fors %f5, SIGN_BIT, %f0 +END (__floorf_vis2) diff --git a/sysdeps/sparc/sparc64/fpu/multiarch/s_floorf-vis3.S b/sysdeps/sparc/sparc64/fpu/multiarch/s_floorf-vis3.S index 65be297..225e17e 100644 --- a/sysdeps/sparc/sparc64/fpu/multiarch/s_floorf-vis3.S +++ b/sysdeps/sparc/sparc64/fpu/multiarch/s_floorf-vis3.S @@ -19,55 +19,40 @@ #include <sysdep.h> - /* Since changing the rounding mode is extremely expensive, we - try to round up using a method that is rounding mode - agnostic. + /* 'siam' (Set Interval Arithmetic Mode) is used to quickly override + the rounding mode during this routine. We add then subtract (or subtract than add if the initial value was negative) 2**23 to the value, then subtract it back out. - This will clear out the fractional portion of the value. - One of two things will happen for non-whole initial values. - Either the rounding mode will round it up, or it will be - rounded down. If the value started out whole, it will be - equal after the addition and subtraction. This means we - can accurately detect with one test whether we need to add - another 1.0 to round it up properly. + This will clear out the fractional portion of the value and, + with suitable 'siam' initiated rouding mode settings, round + the final result in the proper direction. - VIS instructions are used to facilitate the formation of - easier constants, and the propagation of the sign bit. */ + We also use VIS3 moves to avoid using the stack to transfer + values between float and integer registers. */ #define TWO_TWENTYTHREE 0x4b000000 /* 2**23 */ -#define ONE_DOT_ZERO 0x3f800000 /* 1.0 */ #define ZERO %f10 /* 0.0 */ #define SIGN_BIT %f12 /* -0.0 */ ENTRY (__floorf_vis3) sethi %hi(TWO_TWENTYTHREE), %o2 - sethi %hi(ONE_DOT_ZERO), %o3 fzeros ZERO - fnegs ZERO, SIGN_BIT - - movwtos %o2, %f16 + movwtos %o2, %f16 fabss %f1, %f14 - fcmps %fcc3, %f14, %f16 - fmovsuge %fcc3, ZERO, %f16 fands %f1, SIGN_BIT, SIGN_BIT - fors %f16, SIGN_BIT, %f16 + siam (1 << 2) | 3 fadds %f1, %f16, %f5 + siam (1 << 2) | 0 fsubs %f5, %f16, %f5 - fcmps %fcc2, %f5, %f1 - movwtos %o3, %f9 - - fmovsule %fcc2, ZERO, %f9 - fsubs %f5, %f9, %f0 - fabss %f0, %f0 + siam (0 << 2) retl - fors %f0, SIGN_BIT, %f0 + fors %f5, SIGN_BIT, %f0 END (__floorf_vis3) diff --git a/sysdeps/sparc/sparc64/fpu/multiarch/s_floorf.S b/sysdeps/sparc/sparc64/fpu/multiarch/s_floorf.S index 31cda38..d2a83cb 100644 --- a/sysdeps/sparc/sparc64/fpu/multiarch/s_floorf.S +++ b/sysdeps/sparc/sparc64/fpu/multiarch/s_floorf.S @@ -1,6 +1,6 @@ #include <sparc-ifunc.h> -SPARC_ASM_VIS3_IFUNC(floorf) +SPARC_ASM_VIS3_VIS2_IFUNC(floorf) weak_alias (__floorf, floorf) |