2012-11-09 17:53:51 +00:00
|
|
|
# Begin of automatic generation
|
|
|
|
|
|
|
|
# Maximal error of functions:
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "acos":
|
2021-03-01 12:29:17 +00:00
|
|
|
double: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
|
2023-11-03 12:12:20 +00:00
|
|
|
Function: "acos_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "acos_downward":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
2023-11-03 12:12:20 +00:00
|
|
|
Function: "acos_sve":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "acos_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: "acos_upward":
|
2013-12-11 16:19:34 +00:00
|
|
|
double: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
2013-12-11 16:19:34 +00:00
|
|
|
|
2014-01-01 17:30:01 +00:00
|
|
|
Function: "acosh":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 2
|
2015-06-02 09:47:45 +00:00
|
|
|
float: 2
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 4
|
2014-01-01 17:30:01 +00:00
|
|
|
|
2024-02-20 16:59:40 +00:00
|
|
|
Function: "acosh_advsimd":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
|
2014-06-03 12:25:34 +00:00
|
|
|
Function: "acosh_downward":
|
2016-07-18 10:42:52 +00:00
|
|
|
double: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
2014-06-03 12:25:34 +00:00
|
|
|
|
2024-02-20 16:59:40 +00:00
|
|
|
Function: "acosh_sve":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
|
2014-06-03 12:25:34 +00:00
|
|
|
Function: "acosh_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
2014-06-03 12:25:34 +00:00
|
|
|
|
|
|
|
Function: "acosh_upward":
|
2015-06-02 09:47:45 +00:00
|
|
|
double: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
float: 2
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 3
|
2015-05-06 13:00:15 +00:00
|
|
|
|
|
|
|
Function: "asin":
|
2021-03-01 12:29:17 +00:00
|
|
|
double: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 1
|
2014-06-03 12:25:34 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
2023-11-03 12:12:19 +00:00
|
|
|
Function: "asin_advsimd":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "asin_downward":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2023-11-03 12:12:19 +00:00
|
|
|
Function: "asin_sve":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "asin_towardzero":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 1
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 1
|
2013-12-11 16:19:34 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "asin_upward":
|
2021-03-10 12:39:56 +00:00
|
|
|
double: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "asinh":
|
2020-12-21 16:39:20 +00:00
|
|
|
double: 2
|
2020-04-08 11:50:15 +00:00
|
|
|
float: 2
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2024-02-20 16:59:41 +00:00
|
|
|
Function: "asinh_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "asinh_downward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 4
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2024-02-20 16:59:41 +00:00
|
|
|
Function: "asinh_sve":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "asinh_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: "asinh_upward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 4
|
2015-05-06 13:00:15 +00:00
|
|
|
|
|
|
|
Function: "atan":
|
|
|
|
double: 1
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 1
|
2014-01-01 17:30:01 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "atan2":
|
|
|
|
float: 1
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2023-11-03 12:12:22 +00:00
|
|
|
Function: "atan2_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "atan2_downward":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2023-11-03 12:12:22 +00:00
|
|
|
Function: "atan2_sve":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "atan2_towardzero":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "atan2_upward":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
ldouble: 2
|
|
|
|
|
2023-11-03 12:12:21 +00:00
|
|
|
Function: "atan_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "atan_downward":
|
|
|
|
double: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2023-11-03 12:12:21 +00:00
|
|
|
Function: "atan_sve":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "atan_towardzero":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: "atan_upward":
|
|
|
|
double: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "atanh":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 2
|
2015-06-02 09:47:45 +00:00
|
|
|
float: 2
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 4
|
2015-05-06 13:00:15 +00:00
|
|
|
|
2024-04-03 11:13:53 +00:00
|
|
|
Function: "atanh_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "atanh_downward":
|
2015-06-02 09:47:45 +00:00
|
|
|
double: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2024-04-03 11:13:53 +00:00
|
|
|
Function: "atanh_sve":
|
|
|
|
double: 2
|
|
|
|
float: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "atanh_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "atanh_upward":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 3
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 4
|
2015-05-06 13:00:15 +00:00
|
|
|
|
|
|
|
Function: "cabs":
|
|
|
|
double: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
|
|
|
|
Function: "cabs_downward":
|
|
|
|
double: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
|
|
|
|
Function: "cabs_towardzero":
|
|
|
|
double: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
|
|
|
|
Function: "cabs_upward":
|
2014-04-16 21:37:32 +00:00
|
|
|
double: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 1
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: Real part of "cacos":
|
|
|
|
double: 1
|
2013-05-21 08:01:19 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "cacos":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 2
|
2013-05-21 08:01:19 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-06-03 12:25:34 +00:00
|
|
|
Function: Real part of "cacos_downward":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 3
|
2014-06-03 12:25:34 +00:00
|
|
|
float: 2
|
2017-03-27 11:01:34 +00:00
|
|
|
ldouble: 3
|
2014-06-03 12:25:34 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "cacos_downward":
|
|
|
|
double: 5
|
|
|
|
float: 3
|
2017-03-27 11:01:34 +00:00
|
|
|
ldouble: 6
|
2014-06-03 12:25:34 +00:00
|
|
|
|
|
|
|
Function: Real part of "cacos_towardzero":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 3
|
2014-06-03 12:25:34 +00:00
|
|
|
float: 2
|
2017-03-27 11:01:34 +00:00
|
|
|
ldouble: 3
|
2014-06-03 12:25:34 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "cacos_towardzero":
|
|
|
|
double: 5
|
|
|
|
float: 3
|
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: Real part of "cacos_upward":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Imaginary part of "cacos_upward":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 5
|
|
|
|
float: 5
|
|
|
|
ldouble: 7
|
2014-06-03 12:25:34 +00:00
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: Real part of "cacosh":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 2
|
2013-05-21 08:01:19 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "cacosh":
|
|
|
|
double: 1
|
2013-05-21 08:01:19 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Real part of "cacosh_downward":
|
|
|
|
double: 5
|
|
|
|
float: 3
|
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: Imaginary part of "cacosh_downward":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 3
|
|
|
|
float: 3
|
|
|
|
ldouble: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Real part of "cacosh_towardzero":
|
|
|
|
double: 5
|
|
|
|
float: 3
|
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: Imaginary part of "cacosh_towardzero":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 2
|
2017-03-27 11:01:34 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Real part of "cacosh_upward":
|
|
|
|
double: 4
|
|
|
|
float: 4
|
2017-03-27 11:01:34 +00:00
|
|
|
ldouble: 6
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "cacosh_upward":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 2
|
2017-03-27 11:01:34 +00:00
|
|
|
ldouble: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "carg":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2015-05-06 13:00:15 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "carg_downward":
|
|
|
|
double: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "carg_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 1
|
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "carg_upward":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: Real part of "casin":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
2013-05-21 08:01:19 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "casin":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 2
|
2013-05-21 08:01:19 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Real part of "casin_downward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
2017-03-27 11:01:34 +00:00
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Imaginary part of "casin_downward":
|
|
|
|
double: 5
|
|
|
|
float: 3
|
2017-03-27 11:01:34 +00:00
|
|
|
ldouble: 6
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Real part of "casin_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 1
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Imaginary part of "casin_towardzero":
|
|
|
|
double: 5
|
|
|
|
float: 3
|
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: Real part of "casin_upward":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 3
|
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Imaginary part of "casin_upward":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 5
|
|
|
|
float: 5
|
|
|
|
ldouble: 7
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: Real part of "casinh":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 2
|
2013-05-21 08:01:19 +00:00
|
|
|
float: 2
|
2013-12-17 13:03:01 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "casinh":
|
2013-12-17 13:03:01 +00:00
|
|
|
double: 1
|
2013-05-21 08:01:19 +00:00
|
|
|
float: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 2
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Real part of "casinh_downward":
|
|
|
|
double: 5
|
|
|
|
float: 3
|
2017-03-27 11:01:34 +00:00
|
|
|
ldouble: 6
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "casinh_downward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
2017-03-27 11:01:34 +00:00
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Real part of "casinh_towardzero":
|
|
|
|
double: 5
|
|
|
|
float: 3
|
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: Imaginary part of "casinh_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 1
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Real part of "casinh_upward":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 5
|
|
|
|
float: 5
|
|
|
|
ldouble: 7
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "casinh_upward":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
|
|
|
|
2013-05-21 08:01:19 +00:00
|
|
|
Function: Real part of "catan":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 1
|
2013-05-21 08:01:19 +00:00
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: Imaginary part of "catan":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Real part of "catan_downward":
|
|
|
|
double: 1
|
2017-03-27 11:01:34 +00:00
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Imaginary part of "catan_downward":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Real part of "catan_towardzero":
|
|
|
|
double: 1
|
2017-03-27 11:01:34 +00:00
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Imaginary part of "catan_towardzero":
|
|
|
|
double: 2
|
2017-03-27 11:01:34 +00:00
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Real part of "catan_upward":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 1
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 1
|
2017-03-27 11:01:34 +00:00
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "catan_upward":
|
|
|
|
double: 3
|
|
|
|
float: 3
|
|
|
|
ldouble: 3
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: Real part of "catanh":
|
2013-12-17 13:03:01 +00:00
|
|
|
double: 1
|
2013-05-21 08:01:19 +00:00
|
|
|
float: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: Imaginary part of "catanh":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 1
|
2013-05-21 08:01:19 +00:00
|
|
|
float: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Real part of "catanh_downward":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Imaginary part of "catanh_downward":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "catanh_towardzero":
|
|
|
|
double: 2
|
2017-03-27 11:01:34 +00:00
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Imaginary part of "catanh_towardzero":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "catanh_upward":
|
|
|
|
double: 4
|
2017-03-27 11:01:34 +00:00
|
|
|
float: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 4
|
|
|
|
|
|
|
|
Function: Imaginary part of "catanh_upward":
|
2017-03-27 11:01:34 +00:00
|
|
|
double: 1
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 1
|
2017-03-27 11:01:34 +00:00
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "cbrt":
|
2020-12-21 16:39:20 +00:00
|
|
|
double: 4
|
2013-12-11 16:19:34 +00:00
|
|
|
float: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
ldouble: 1
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2024-04-30 12:49:59 +00:00
|
|
|
Function: "cbrt_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "cbrt_downward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
2024-04-30 12:49:59 +00:00
|
|
|
Function: "cbrt_sve":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "cbrt_towardzero":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 3
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 1
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: "cbrt_upward":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 5
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: Real part of "ccos":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: Imaginary part of "ccos":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Real part of "ccos_downward":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Imaginary part of "ccos_downward":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "ccos_towardzero":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Imaginary part of "ccos_towardzero":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "ccos_upward":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Imaginary part of "ccos_upward":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: Real part of "ccosh":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: Imaginary part of "ccosh":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Real part of "ccosh_downward":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Imaginary part of "ccosh_downward":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "ccosh_towardzero":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Imaginary part of "ccosh_towardzero":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "ccosh_upward":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Imaginary part of "ccosh_upward":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "cexp":
|
|
|
|
double: 2
|
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: Imaginary part of "cexp":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
ldouble: 1
|
|
|
|
|
2015-07-02 13:58:12 +00:00
|
|
|
Function: Real part of "cexp_downward":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 2
|
2015-07-02 13:58:12 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Imaginary part of "cexp_downward":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 3
|
2015-07-02 13:58:12 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "cexp_towardzero":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 2
|
2015-07-02 13:58:12 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Imaginary part of "cexp_towardzero":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 3
|
2015-07-02 13:58:12 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "cexp_upward":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Imaginary part of "cexp_upward":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 3
|
2015-07-02 13:58:12 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Real part of "clog":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
|
|
|
float: 3
|
2015-12-01 12:57:16 +00:00
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "clog":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: Real part of "clog10":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
2015-12-01 12:57:16 +00:00
|
|
|
float: 4
|
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "clog10":
|
2016-11-10 11:22:35 +00:00
|
|
|
double: 2
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "clog10_downward":
|
2015-12-01 12:57:16 +00:00
|
|
|
double: 5
|
2017-09-18 12:37:08 +00:00
|
|
|
float: 5
|
2015-12-01 12:57:16 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "clog10_downward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
|
|
|
float: 4
|
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Real part of "clog10_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 5
|
2015-12-01 12:57:16 +00:00
|
|
|
float: 5
|
|
|
|
ldouble: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "clog10_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
|
|
|
float: 4
|
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Real part of "clog10_upward":
|
2015-12-01 12:57:16 +00:00
|
|
|
double: 6
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 5
|
2015-12-01 12:57:16 +00:00
|
|
|
ldouble: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "clog10_upward":
|
2012-11-09 17:53:51 +00:00
|
|
|
double: 2
|
2015-12-01 12:57:16 +00:00
|
|
|
float: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Real part of "clog_downward":
|
2015-12-01 12:57:16 +00:00
|
|
|
double: 4
|
|
|
|
float: 3
|
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Imaginary part of "clog_downward":
|
2012-11-09 17:53:51 +00:00
|
|
|
double: 1
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Real part of "clog_towardzero":
|
2015-12-01 12:57:16 +00:00
|
|
|
double: 4
|
|
|
|
float: 4
|
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Imaginary part of "clog_towardzero":
|
2012-11-09 17:53:51 +00:00
|
|
|
double: 1
|
2015-12-01 12:57:16 +00:00
|
|
|
float: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "clog_upward":
|
2015-12-01 12:57:16 +00:00
|
|
|
double: 4
|
|
|
|
float: 3
|
|
|
|
ldouble: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "clog_upward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 1
|
|
|
|
float: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: "cos":
|
2018-04-03 15:24:29 +00:00
|
|
|
double: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 1
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2023-04-12 13:37:49 +00:00
|
|
|
Function: "cos_advsimd":
|
2023-06-28 11:19:36 +00:00
|
|
|
double: 2
|
2023-04-12 13:37:49 +00:00
|
|
|
float: 1
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "cos_downward":
|
2013-12-11 16:19:34 +00:00
|
|
|
double: 1
|
2018-08-14 09:45:59 +00:00
|
|
|
float: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2023-04-12 13:37:49 +00:00
|
|
|
Function: "cos_sve":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "cos_towardzero":
|
2013-12-11 16:19:34 +00:00
|
|
|
double: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 1
|
2013-12-17 13:03:01 +00:00
|
|
|
ldouble: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: "cos_upward":
|
2013-12-11 16:19:34 +00:00
|
|
|
double: 1
|
2018-08-14 09:45:59 +00:00
|
|
|
float: 1
|
2013-12-11 16:19:34 +00:00
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: "cosh":
|
2020-12-21 16:39:20 +00:00
|
|
|
double: 2
|
2020-04-08 11:50:15 +00:00
|
|
|
float: 2
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2024-02-20 16:59:39 +00:00
|
|
|
Function: "cosh_advsimd":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "cosh_downward":
|
2020-12-21 16:39:20 +00:00
|
|
|
double: 3
|
2014-01-01 17:30:01 +00:00
|
|
|
float: 1
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2024-02-20 16:59:39 +00:00
|
|
|
Function: "cosh_sve":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "cosh_towardzero":
|
2020-12-21 16:39:20 +00:00
|
|
|
double: 3
|
2014-01-01 17:30:01 +00:00
|
|
|
float: 1
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: "cosh_upward":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 2
|
2014-01-01 17:30:01 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: Real part of "cpow":
|
|
|
|
double: 2
|
|
|
|
float: 5
|
|
|
|
ldouble: 4
|
|
|
|
|
|
|
|
Function: Imaginary part of "cpow":
|
|
|
|
float: 2
|
|
|
|
ldouble: 1
|
|
|
|
|
2014-07-11 17:56:42 +00:00
|
|
|
Function: Real part of "cpow_downward":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 5
|
2014-07-11 17:56:42 +00:00
|
|
|
float: 8
|
|
|
|
ldouble: 6
|
|
|
|
|
|
|
|
Function: Imaginary part of "cpow_downward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 1
|
2014-07-11 17:56:42 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "cpow_towardzero":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 5
|
2014-07-11 17:56:42 +00:00
|
|
|
float: 8
|
|
|
|
ldouble: 6
|
|
|
|
|
|
|
|
Function: Imaginary part of "cpow_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 1
|
2014-07-11 17:56:42 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "cpow_upward":
|
|
|
|
double: 4
|
|
|
|
float: 1
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Imaginary part of "cpow_upward":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: Real part of "csin":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: Imaginary part of "csin":
|
|
|
|
ldouble: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Real part of "csin_downward":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Imaginary part of "csin_downward":
|
|
|
|
double: 1
|
2015-07-02 13:58:12 +00:00
|
|
|
float: 1
|
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Real part of "csin_towardzero":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Imaginary part of "csin_towardzero":
|
2015-07-02 13:58:12 +00:00
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "csin_upward":
|
2014-04-16 21:37:32 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Imaginary part of "csin_upward":
|
|
|
|
double: 1
|
2015-07-02 13:58:12 +00:00
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 3
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: Real part of "csinh":
|
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: Imaginary part of "csinh":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Real part of "csinh_downward":
|
2015-07-02 13:58:12 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "csinh_downward":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: Real part of "csinh_towardzero":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
2015-07-02 13:58:12 +00:00
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "csinh_towardzero":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 3
|
2015-07-02 13:58:12 +00:00
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Real part of "csinh_upward":
|
|
|
|
double: 1
|
2015-07-02 13:58:12 +00:00
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Imaginary part of "csinh_upward":
|
|
|
|
double: 2
|
2015-07-02 13:58:12 +00:00
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 2
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: Real part of "csqrt":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "csqrt":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: Real part of "csqrt_downward":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 5
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 4
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "csqrt_downward":
|
2015-06-02 09:47:45 +00:00
|
|
|
double: 4
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Real part of "csqrt_towardzero":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 4
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "csqrt_towardzero":
|
2015-06-02 09:47:45 +00:00
|
|
|
double: 4
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Real part of "csqrt_upward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 5
|
|
|
|
float: 4
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "csqrt_upward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: Real part of "ctan":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
2014-01-01 17:30:01 +00:00
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "ctan":
|
|
|
|
double: 2
|
2017-12-18 14:40:36 +00:00
|
|
|
float: 2
|
2014-01-01 17:30:01 +00:00
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: Real part of "ctan_downward":
|
|
|
|
double: 6
|
|
|
|
float: 5
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 4
|
|
|
|
|
|
|
|
Function: Imaginary part of "ctan_downward":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 2
|
2017-12-18 14:40:36 +00:00
|
|
|
float: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: Real part of "ctan_towardzero":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 5
|
2016-07-21 08:48:45 +00:00
|
|
|
float: 3
|
2014-01-01 17:30:01 +00:00
|
|
|
ldouble: 4
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "ctan_towardzero":
|
2016-07-21 08:48:45 +00:00
|
|
|
double: 2
|
2014-01-01 17:30:01 +00:00
|
|
|
float: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: Real part of "ctan_upward":
|
|
|
|
double: 2
|
2017-12-18 14:40:36 +00:00
|
|
|
float: 4
|
2014-01-01 17:30:01 +00:00
|
|
|
ldouble: 5
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "ctan_upward":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 2
|
|
|
|
float: 3
|
2014-07-11 17:56:42 +00:00
|
|
|
ldouble: 5
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: Real part of "ctanh":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 2
|
2014-01-01 17:30:01 +00:00
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: Imaginary part of "ctanh":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 1
|
2014-01-01 17:30:01 +00:00
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: Real part of "ctanh_downward":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 4
|
2017-12-18 14:40:36 +00:00
|
|
|
float: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: Imaginary part of "ctanh_downward":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 6
|
|
|
|
float: 5
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 4
|
|
|
|
|
|
|
|
Function: Real part of "ctanh_towardzero":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: Imaginary part of "ctanh_towardzero":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 5
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
|
|
|
|
2014-01-01 17:30:01 +00:00
|
|
|
Function: Real part of "ctanh_upward":
|
|
|
|
double: 2
|
|
|
|
float: 3
|
2014-07-11 17:56:42 +00:00
|
|
|
ldouble: 5
|
2014-01-01 17:30:01 +00:00
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: Imaginary part of "ctanh_upward":
|
|
|
|
double: 2
|
2014-01-01 17:30:01 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 5
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: "erf":
|
|
|
|
double: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 1
|
2013-12-11 16:19:34 +00:00
|
|
|
ldouble: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2024-02-20 16:59:38 +00:00
|
|
|
Function: "erf_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "erf_downward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 1
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2024-02-20 16:59:38 +00:00
|
|
|
Function: "erf_sve":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "erf_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 1
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: "erf_upward":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: "erfc":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 4
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2024-02-20 16:59:45 +00:00
|
|
|
Function: "erfc_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "erfc_downward":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 4
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 4
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 5
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2024-02-20 16:59:45 +00:00
|
|
|
Function: "erfc_sve":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "erfc_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "erfc_upward":
|
Add new exp and exp2 implementations
Optimized exp and exp2 implementations using a lookup table for
fractional powers of 2. There are several variants, see e_exp_data.c,
they can be selected by modifying math_config.h allowing different
tradeoffs.
The default selection should be acceptable as generic libm code.
Worst case error is 0.509 ULP for exp and 0.507 ULP for exp2, on
aarch64 the rodata size is 2160 bytes, shared between exp and exp2.
On aarch64 .text + .rodata size decreased by 24912 bytes.
The non-nearest rounding error is less than 1 ULP even on targets
without efficient round implementation (although the error rate is
higher in that case). Targets with single instruction, rounding mode
independent, to nearest integer rounding and conversion can use them
by setting TOINT_INTRINSICS and adding the necessary code to their
math_private.h.
The __exp1 code uses the same algorithm, so the error bound of pow
increased a bit.
New double precision error handling code was added following the
style of the single precision error handling code.
Improvements on Cortex-A72 compared to current glibc master:
exp thruput: 1.61x in [-9.9 9.9]
exp latency: 1.53x in [-9.9 9.9]
exp thruput: 1.13x in [0.5 1]
exp latency: 1.30x in [0.5 1]
exp2 thruput: 2.03x in [-9.9 9.9]
exp2 latency: 1.64x in [-9.9 9.9]
For small (< 1) inputs the current exp code uses a separate algorithm
so the speed up there is less.
Was tested on
aarch64-linux-gnu (TOINT_INTRINSICS, fma contraction) and
arm-linux-gnueabihf (!TOINT_INTRINSICS, no fma contraction) and
x86_64-linux-gnu (!TOINT_INTRINSICS, no fma contraction) and
powerpc64le-linux-gnu (!TOINT_INTRINSICS, fma contraction) targets,
only non-nearest rounding ulp errors increase and they are within
acceptable bounds (ulp updates are in separate patches).
* NEWS: Mention exp and exp2 improvements.
* math/Makefile (libm-support): Remove t_exp.
(type-double-routines): Add math_err and e_exp_data.
* sysdeps/aarch64/libm-test-ulps: Update.
* sysdeps/arm/libm-test-ulps: Update.
* sysdeps/i386/fpu/e_exp_data.c: New file.
* sysdeps/i386/fpu/math_err.c: New file.
* sysdeps/i386/fpu/t_exp.c: Remove.
* sysdeps/ia64/fpu/e_exp_data.c: New file.
* sysdeps/ia64/fpu/math_err.c: New file.
* sysdeps/ia64/fpu/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/e_exp.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp2.c: Rewrite.
* sysdeps/ieee754/dbl-64/e_exp_data.c: New file.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Update error bound.
* sysdeps/ieee754/dbl-64/eexp.tbl: Remove.
* sysdeps/ieee754/dbl-64/math_config.h: New file.
* sysdeps/ieee754/dbl-64/math_err.c: New file.
* sysdeps/ieee754/dbl-64/t_exp.c: Remove.
* sysdeps/ieee754/dbl-64/t_exp2.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.h: Remove.
* sysdeps/ieee754/dbl-64/uexp.tbl: Remove.
* sysdeps/m68k/m680x0/fpu/e_exp_data.c: New file.
* sysdeps/m68k/m680x0/fpu/math_err.c: New file.
* sysdeps/m68k/m680x0/fpu/t_exp.c: Remove.
* sysdeps/powerpc/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
2018-02-12 18:16:03 +00:00
|
|
|
double: 4
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 4
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: "exp":
|
2020-12-21 16:39:20 +00:00
|
|
|
double: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
float: 1
|
2015-12-01 12:57:16 +00:00
|
|
|
ldouble: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
|
|
|
|
Function: "exp10":
|
2014-04-16 21:37:32 +00:00
|
|
|
double: 2
|
2020-04-08 11:50:15 +00:00
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2015-05-06 13:00:15 +00:00
|
|
|
|
2023-10-05 16:10:52 +00:00
|
|
|
Function: "exp10_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "exp10_downward":
|
|
|
|
double: 2
|
|
|
|
float: 1
|
|
|
|
ldouble: 3
|
|
|
|
|
2023-10-05 16:10:52 +00:00
|
|
|
Function: "exp10_sve":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "exp10_towardzero":
|
|
|
|
double: 2
|
|
|
|
float: 1
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: "exp10_upward":
|
|
|
|
double: 2
|
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2024-06-18 20:01:40 +00:00
|
|
|
Function: "exp10m1":
|
2024-08-07 13:27:04 +00:00
|
|
|
double: 4
|
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
2024-06-18 20:01:40 +00:00
|
|
|
|
|
|
|
Function: "exp10m1_downward":
|
2024-08-07 13:27:04 +00:00
|
|
|
double: 3
|
|
|
|
float: 3
|
|
|
|
ldouble: 6
|
2024-06-18 20:01:40 +00:00
|
|
|
|
|
|
|
Function: "exp10m1_towardzero":
|
2024-08-07 13:27:04 +00:00
|
|
|
double: 2
|
|
|
|
float: 3
|
|
|
|
ldouble: 6
|
2024-06-18 20:01:40 +00:00
|
|
|
|
|
|
|
Function: "exp10m1_upward":
|
2024-08-07 13:27:04 +00:00
|
|
|
double: 5
|
|
|
|
float: 3
|
|
|
|
ldouble: 6
|
2024-06-18 20:01:40 +00:00
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "exp2":
|
2012-11-09 17:53:51 +00:00
|
|
|
double: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
float: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
2023-10-05 16:10:49 +00:00
|
|
|
Function: "exp2_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "exp2_downward":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 1
|
2014-07-11 17:56:42 +00:00
|
|
|
float: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
ldouble: 1
|
2014-01-01 17:30:01 +00:00
|
|
|
|
2023-10-05 16:10:49 +00:00
|
|
|
Function: "exp2_sve":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "exp2_towardzero":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 1
|
2014-07-11 17:56:42 +00:00
|
|
|
float: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
ldouble: 1
|
2014-01-01 17:30:01 +00:00
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "exp2_upward":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 1
|
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2024-06-18 20:01:40 +00:00
|
|
|
Function: "exp2m1":
|
2024-08-07 13:27:04 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
2024-06-18 20:01:40 +00:00
|
|
|
|
|
|
|
Function: "exp2m1_downward":
|
2024-08-07 13:27:04 +00:00
|
|
|
double: 3
|
|
|
|
float: 3
|
|
|
|
ldouble: 3
|
2024-06-18 20:01:40 +00:00
|
|
|
|
|
|
|
Function: "exp2m1_towardzero":
|
2024-08-07 13:27:04 +00:00
|
|
|
double: 3
|
|
|
|
float: 2
|
|
|
|
ldouble: 4
|
2024-06-18 20:01:40 +00:00
|
|
|
|
|
|
|
Function: "exp2m1_upward":
|
2024-08-07 13:27:04 +00:00
|
|
|
double: 3
|
|
|
|
float: 3
|
|
|
|
ldouble: 5
|
2024-06-18 20:01:40 +00:00
|
|
|
|
2023-06-28 11:19:39 +00:00
|
|
|
Function: "exp_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "exp_downward":
|
2013-12-11 16:19:34 +00:00
|
|
|
double: 1
|
2017-09-18 12:37:08 +00:00
|
|
|
float: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2023-06-28 11:19:39 +00:00
|
|
|
Function: "exp_sve":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "exp_towardzero":
|
2013-12-11 16:19:34 +00:00
|
|
|
double: 1
|
2017-09-18 12:37:08 +00:00
|
|
|
float: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: "exp_upward":
|
2013-12-11 16:19:34 +00:00
|
|
|
double: 1
|
2017-09-18 12:37:08 +00:00
|
|
|
float: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: "expm1":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2023-11-16 13:24:18 +00:00
|
|
|
Function: "expm1_advsimd":
|
|
|
|
double: 2
|
|
|
|
float: 1
|
|
|
|
|
2014-01-01 17:30:01 +00:00
|
|
|
Function: "expm1_downward":
|
2013-07-05 08:30:52 +00:00
|
|
|
double: 1
|
2013-12-17 13:03:01 +00:00
|
|
|
float: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2023-11-16 13:24:18 +00:00
|
|
|
Function: "expm1_sve":
|
|
|
|
double: 2
|
|
|
|
float: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "expm1_towardzero":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 4
|
2014-01-01 17:30:01 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "expm1_upward":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
ldouble: 3
|
2014-01-01 17:30:01 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "gamma":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
float: 4
|
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: "gamma_downward":
|
|
|
|
double: 4
|
|
|
|
float: 4
|
|
|
|
ldouble: 8
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "gamma_towardzero":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 4
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 5
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "gamma_upward":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 4
|
|
|
|
float: 5
|
|
|
|
ldouble: 8
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "hypot":
|
|
|
|
double: 1
|
2023-02-22 11:23:49 +00:00
|
|
|
float: 1
|
2014-01-01 17:30:01 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
2024-04-30 12:49:58 +00:00
|
|
|
Function: "hypot_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "hypot_downward":
|
|
|
|
double: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
2024-04-30 12:49:58 +00:00
|
|
|
Function: "hypot_sve":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "hypot_towardzero":
|
|
|
|
double: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: "hypot_upward":
|
2014-01-01 17:30:01 +00:00
|
|
|
double: 1
|
|
|
|
ldouble: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: "j0":
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
double: 3
|
|
|
|
float: 9
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 2
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "j0_downward":
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
double: 6
|
|
|
|
float: 9
|
|
|
|
ldouble: 9
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "j0_towardzero":
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
double: 7
|
|
|
|
float: 9
|
|
|
|
ldouble: 9
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "j0_upward":
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
double: 9
|
2021-10-05 12:43:26 +00:00
|
|
|
float: 9
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
ldouble: 7
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "j1":
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
double: 4
|
|
|
|
float: 9
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 4
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "j1_downward":
|
|
|
|
double: 3
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
float: 8
|
|
|
|
ldouble: 6
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "j1_towardzero":
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
double: 4
|
|
|
|
float: 8
|
|
|
|
ldouble: 9
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "j1_upward":
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
double: 9
|
|
|
|
float: 9
|
|
|
|
ldouble: 9
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "jn":
|
|
|
|
double: 4
|
2014-01-01 17:30:01 +00:00
|
|
|
float: 4
|
|
|
|
ldouble: 7
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2015-07-02 13:58:12 +00:00
|
|
|
Function: "jn_downward":
|
|
|
|
double: 4
|
|
|
|
float: 5
|
|
|
|
ldouble: 8
|
|
|
|
|
|
|
|
Function: "jn_towardzero":
|
|
|
|
double: 4
|
|
|
|
float: 5
|
|
|
|
ldouble: 8
|
|
|
|
|
|
|
|
Function: "jn_upward":
|
|
|
|
double: 5
|
|
|
|
float: 4
|
|
|
|
ldouble: 7
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "lgamma":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
float: 4
|
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: "lgamma_downward":
|
|
|
|
double: 4
|
|
|
|
float: 4
|
|
|
|
ldouble: 8
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "lgamma_towardzero":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 4
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 5
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "lgamma_upward":
|
2015-09-24 13:22:31 +00:00
|
|
|
double: 4
|
|
|
|
float: 5
|
|
|
|
ldouble: 8
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2013-12-11 16:19:34 +00:00
|
|
|
Function: "log":
|
2020-12-21 16:39:20 +00:00
|
|
|
double: 1
|
2013-12-11 16:19:34 +00:00
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "log10":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 2
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2023-10-05 16:10:51 +00:00
|
|
|
Function: "log10_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
|
2014-06-03 12:25:34 +00:00
|
|
|
Function: "log10_downward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
2015-06-02 09:47:45 +00:00
|
|
|
float: 3
|
2014-06-03 12:25:34 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
2023-10-05 16:10:51 +00:00
|
|
|
Function: "log10_sve":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
|
2014-06-03 12:25:34 +00:00
|
|
|
Function: "log10_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
2014-06-03 12:25:34 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: "log10_upward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
2014-06-03 12:25:34 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
2024-06-18 20:01:40 +00:00
|
|
|
Function: "log10p1":
|
2024-08-07 13:27:04 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
2024-06-18 20:01:40 +00:00
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: "log10p1_downward":
|
|
|
|
double: 2
|
2024-08-07 13:27:04 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 4
|
2024-06-18 20:01:40 +00:00
|
|
|
|
|
|
|
Function: "log10p1_towardzero":
|
2024-08-07 13:27:04 +00:00
|
|
|
double: 3
|
2024-06-18 20:01:40 +00:00
|
|
|
float: 2
|
2024-08-07 13:27:04 +00:00
|
|
|
ldouble: 3
|
2024-06-18 20:01:40 +00:00
|
|
|
|
|
|
|
Function: "log10p1_upward":
|
|
|
|
double: 2
|
2024-08-07 13:27:04 +00:00
|
|
|
float: 3
|
|
|
|
ldouble: 4
|
2024-06-18 20:01:40 +00:00
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "log1p":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 1
|
2013-12-11 16:19:34 +00:00
|
|
|
float: 1
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2023-11-03 12:12:23 +00:00
|
|
|
Function: "log1p_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "log1p_downward":
|
|
|
|
double: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2023-11-03 12:12:23 +00:00
|
|
|
Function: "log1p_sve":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "log1p_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "log1p_upward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2015-05-06 13:00:15 +00:00
|
|
|
|
|
|
|
Function: "log2":
|
2014-04-16 21:37:32 +00:00
|
|
|
double: 1
|
|
|
|
float: 1
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2023-10-05 16:10:50 +00:00
|
|
|
Function: "log2_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
|
2014-07-11 17:56:42 +00:00
|
|
|
Function: "log2_downward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2015-05-06 13:00:15 +00:00
|
|
|
|
2023-10-05 16:10:50 +00:00
|
|
|
Function: "log2_sve":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "log2_towardzero":
|
2014-07-11 17:56:42 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 1
|
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "log2_upward":
|
|
|
|
double: 3
|
|
|
|
float: 3
|
2014-07-11 17:56:42 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
2024-05-20 15:52:18 +00:00
|
|
|
Function: "log2p1":
|
2024-07-25 13:40:32 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
2024-05-20 15:52:18 +00:00
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: "log2p1_downward":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: "log2p1_towardzero":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
|
|
|
|
|
|
|
Function: "log2p1_upward":
|
2024-07-25 13:40:32 +00:00
|
|
|
double: 2
|
2024-05-20 15:52:18 +00:00
|
|
|
float: 2
|
2024-07-25 13:40:32 +00:00
|
|
|
ldouble: 3
|
2024-05-20 15:52:18 +00:00
|
|
|
|
2023-06-28 11:19:38 +00:00
|
|
|
Function: "log_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 3
|
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "log_downward":
|
2014-07-11 17:56:42 +00:00
|
|
|
float: 2
|
|
|
|
ldouble: 1
|
|
|
|
|
2023-06-28 11:19:38 +00:00
|
|
|
Function: "log_sve":
|
|
|
|
double: 1
|
|
|
|
float: 3
|
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "log_towardzero":
|
2016-07-18 10:42:52 +00:00
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "log_upward":
|
2015-12-01 12:57:16 +00:00
|
|
|
double: 1
|
2016-07-18 10:42:52 +00:00
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
Implement C23 logp1
C23 adds various <math.h> function families originally defined in TS
18661-4. Add the logp1 functions (aliases for log1p functions - the
name is intended to be more consistent with the new log2p1 and
log10p1, where clearly it would have been very confusing to name those
functions log21p and log101p). As aliases rather than new functions,
the content of this patch is somewhat different from those actually
adding new functions.
Tests are shared with log1p, so this patch *does* mechanically update
all affected libm-test-ulps files to expect the same errors for both
functions.
The vector versions of log1p on aarch64 and x86_64 are *not* updated
to have logp1 aliases (and thus there are no corresponding header,
tests, abilist or ulps changes for vector functions either). It would
be reasonable for such vector aliases and corresponding changes to
other files to be made separately. For now, the log1p tests instead
avoid testing logp1 in the vector case (a Makefile change is needed to
avoid problems with grep, used in generating the .c files for vector
function tests, matching more than one ALL_RM_TEST line in a file
testing multiple functions with the same inputs, when it assumes that
the .inc file only has a single such line).
Tested for x86_64 and x86, and with build-many-glibcs.py.
2024-06-17 13:47:09 +00:00
|
|
|
Function: "logp1":
|
|
|
|
double: 1
|
|
|
|
float: 1
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: "logp1_downward":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: "logp1_towardzero":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: "logp1_upward":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
ldouble: 2
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "pow":
|
2018-02-12 10:42:42 +00:00
|
|
|
double: 1
|
2015-12-01 12:57:16 +00:00
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2024-05-16 08:21:24 +00:00
|
|
|
Function: "pow_advsimd":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "pow_downward":
|
2014-07-11 17:56:42 +00:00
|
|
|
double: 1
|
2015-12-01 12:57:16 +00:00
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2024-05-16 08:21:24 +00:00
|
|
|
Function: "pow_sve":
|
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "pow_towardzero":
|
2014-07-11 17:56:42 +00:00
|
|
|
double: 1
|
2015-12-01 12:57:16 +00:00
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: "pow_upward":
|
2014-07-11 17:56:42 +00:00
|
|
|
double: 1
|
2015-12-01 12:57:16 +00:00
|
|
|
float: 1
|
2014-07-11 17:56:42 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2013-12-11 16:19:34 +00:00
|
|
|
Function: "sin":
|
2018-04-03 15:24:29 +00:00
|
|
|
double: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 1
|
2021-03-01 12:29:17 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2023-06-28 11:19:37 +00:00
|
|
|
Function: "sin_advsimd":
|
|
|
|
double: 2
|
|
|
|
float: 1
|
|
|
|
|
2013-12-11 16:19:34 +00:00
|
|
|
Function: "sin_downward":
|
|
|
|
double: 1
|
2018-08-14 09:45:59 +00:00
|
|
|
float: 1
|
2015-06-02 09:47:45 +00:00
|
|
|
ldouble: 3
|
2013-12-11 16:19:34 +00:00
|
|
|
|
2023-06-28 11:19:37 +00:00
|
|
|
Function: "sin_sve":
|
|
|
|
double: 2
|
|
|
|
float: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "sin_towardzero":
|
|
|
|
double: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 1
|
2015-06-02 09:47:45 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "sin_upward":
|
2013-12-11 16:19:34 +00:00
|
|
|
double: 1
|
2018-08-14 09:45:59 +00:00
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "sincos":
|
2018-04-03 15:24:29 +00:00
|
|
|
double: 1
|
2014-04-16 21:37:32 +00:00
|
|
|
float: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 1
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "sincos_downward":
|
2013-12-11 16:19:34 +00:00
|
|
|
double: 1
|
2018-08-14 09:45:59 +00:00
|
|
|
float: 1
|
2015-05-06 13:00:15 +00:00
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "sincos_towardzero":
|
|
|
|
double: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 1
|
2015-06-02 09:47:45 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "sincos_upward":
|
|
|
|
double: 1
|
2018-08-14 09:45:59 +00:00
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "sinh":
|
2015-06-02 09:47:45 +00:00
|
|
|
double: 2
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2024-04-03 11:15:41 +00:00
|
|
|
Function: "sinh_advsimd":
|
|
|
|
double: 2
|
|
|
|
float: 1
|
|
|
|
|
2015-05-06 13:00:15 +00:00
|
|
|
Function: "sinh_downward":
|
|
|
|
double: 3
|
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2015-05-06 13:00:15 +00:00
|
|
|
|
2024-04-03 11:15:41 +00:00
|
|
|
Function: "sinh_sve":
|
|
|
|
double: 2
|
|
|
|
float: 1
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "sinh_towardzero":
|
2020-12-21 16:39:20 +00:00
|
|
|
double: 3
|
2015-05-06 13:00:15 +00:00
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: "sinh_upward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 4
|
2015-05-06 13:00:15 +00:00
|
|
|
|
|
|
|
Function: "tan":
|
2013-12-11 16:19:34 +00:00
|
|
|
float: 1
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 1
|
2013-12-11 16:19:34 +00:00
|
|
|
|
2023-10-05 16:10:48 +00:00
|
|
|
Function: "tan_advsimd":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "tan_downward":
|
2013-12-11 16:19:34 +00:00
|
|
|
double: 1
|
|
|
|
float: 2
|
|
|
|
ldouble: 1
|
|
|
|
|
2023-10-05 16:10:48 +00:00
|
|
|
Function: "tan_sve":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "tan_towardzero":
|
2013-12-11 16:19:34 +00:00
|
|
|
double: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 1
|
|
|
|
ldouble: 1
|
|
|
|
|
|
|
|
Function: "tan_upward":
|
2013-12-11 16:19:34 +00:00
|
|
|
double: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
float: 1
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 1
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: "tanh":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 2
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2024-02-20 16:59:44 +00:00
|
|
|
Function: "tanh_advsimd":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "tanh_downward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
|
|
|
float: 3
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 4
|
2024-02-20 16:59:44 +00:00
|
|
|
|
|
|
|
Function: "tanh_sve":
|
|
|
|
double: 2
|
|
|
|
float: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "tanh_towardzero":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 2
|
|
|
|
float: 2
|
2015-09-24 13:22:31 +00:00
|
|
|
ldouble: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "tanh_upward":
|
2015-05-06 13:00:15 +00:00
|
|
|
double: 3
|
2013-05-21 08:01:19 +00:00
|
|
|
float: 3
|
2015-05-06 13:00:15 +00:00
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: "tgamma":
|
2020-12-21 16:39:20 +00:00
|
|
|
double: 9
|
2020-04-08 11:50:15 +00:00
|
|
|
float: 8
|
2015-07-02 13:58:12 +00:00
|
|
|
ldouble: 4
|
|
|
|
|
|
|
|
Function: "tgamma_downward":
|
2021-04-08 07:22:39 +00:00
|
|
|
double: 9
|
2020-04-08 11:50:15 +00:00
|
|
|
float: 7
|
2015-07-02 13:58:12 +00:00
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: "tgamma_towardzero":
|
2020-12-21 16:39:20 +00:00
|
|
|
double: 9
|
2020-04-08 11:50:15 +00:00
|
|
|
float: 7
|
2015-07-02 13:58:12 +00:00
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: "tgamma_upward":
|
2020-12-21 16:39:20 +00:00
|
|
|
double: 9
|
2020-04-08 11:50:15 +00:00
|
|
|
float: 8
|
2013-05-21 08:01:19 +00:00
|
|
|
ldouble: 4
|
2012-11-09 17:53:51 +00:00
|
|
|
|
|
|
|
Function: "y0":
|
|
|
|
double: 2
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
float: 8
|
2012-11-09 17:53:51 +00:00
|
|
|
ldouble: 3
|
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "y0_downward":
|
|
|
|
double: 3
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
float: 8
|
|
|
|
ldouble: 7
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "y0_towardzero":
|
|
|
|
double: 3
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
float: 8
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 3
|
|
|
|
|
|
|
|
Function: "y0_upward":
|
|
|
|
double: 2
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
float: 8
|
|
|
|
ldouble: 4
|
2014-04-16 21:37:32 +00:00
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "y1":
|
|
|
|
double: 3
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
float: 9
|
|
|
|
ldouble: 5
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-04-16 21:37:32 +00:00
|
|
|
Function: "y1_downward":
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
double: 6
|
|
|
|
float: 8
|
|
|
|
ldouble: 5
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "y1_towardzero":
|
|
|
|
double: 3
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
float: 9
|
2015-05-06 13:00:15 +00:00
|
|
|
ldouble: 2
|
2014-04-16 21:37:32 +00:00
|
|
|
|
|
|
|
Function: "y1_upward":
|
Fix the inaccuracy of j0f/j1f/y0f/y1f [BZ #14469, #14470, #14471, #14472]
For j0f/j1f/y0f/y1f, the largest error for all binary32
inputs is reduced to at most 9 ulps for all rounding modes.
The new code is enabled only when there is a cancellation at the very end of
the j0f/j1f/y0f/y1f computation, or for very large inputs, thus should not
give any visible slowdown on average. Two different algorithms are used:
* around the first 64 zeros of j0/j1/y0/y1, approximation polynomials of
degree 3 are used, computed using the Sollya tool (https://www.sollya.org/)
* for large inputs, an asymptotic formula from [1] is used
[1] Fast and Accurate Bessel Function Computation,
John Harrison, Proceedings of Arith 19, 2009.
Inputs yielding the new largest errors are added to auto-libm-test-in,
and ulps are regenerated for various targets (thanks Adhemerval Zanella).
Tested on x86_64 with --disable-multi-arch and on powerpc64le-linux-gnu.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-04-01 06:14:10 +00:00
|
|
|
double: 6
|
|
|
|
float: 9
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 5
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
Function: "yn":
|
|
|
|
double: 3
|
2017-12-18 14:40:36 +00:00
|
|
|
float: 3
|
2014-04-16 21:37:32 +00:00
|
|
|
ldouble: 5
|
2012-11-09 17:53:51 +00:00
|
|
|
|
2014-07-11 17:56:42 +00:00
|
|
|
Function: "yn_downward":
|
|
|
|
double: 3
|
2017-12-18 14:40:36 +00:00
|
|
|
float: 4
|
2014-07-11 17:56:42 +00:00
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: "yn_towardzero":
|
|
|
|
double: 3
|
|
|
|
float: 3
|
|
|
|
ldouble: 5
|
|
|
|
|
|
|
|
Function: "yn_upward":
|
|
|
|
double: 4
|
2017-12-18 14:40:36 +00:00
|
|
|
float: 5
|
2014-07-11 17:56:42 +00:00
|
|
|
ldouble: 5
|
|
|
|
|
2012-11-09 17:53:51 +00:00
|
|
|
# end of automatic generation
|