Jan Vesely 973c1fa5f5 math: Use single precision fmax in sp path ...

Fixes fdim piglit on Turks

v2: use CL fmax instead of __builtin

Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Reviewed-by: Tom Stellard <tom.stellard@amd.com>
llvm-svn: 269807

2016-05-17 19:44:01 +00:00

..

acos.cl

…

acos.inc

Fix compilation warnings without cl_khr_fp64

2015-04-24 19:54:17 +00:00

acosh.cl

Implement acosh builtin

2015-04-07 16:20:20 +00:00

acospi.cl

Implement acospi builtin

2015-04-02 17:01:52 +00:00

asin.cl

…

asin.inc

Fix compilation warnings without cl_khr_fp64

2015-04-24 19:54:17 +00:00

asinh.cl

Implement asinh builtin

2015-04-02 17:01:54 +00:00

asinpi.cl

Implement asinpi builtin

2015-04-02 17:01:56 +00:00

atan2.cl

Implement atan2 for doubles

2015-05-12 13:48:51 +00:00

atan2pi.cl

Implement atan2pi builtin

2015-05-12 14:48:26 +00:00

atan.cl

…

atanh.cl

Implement atanh builtin

2015-04-07 16:20:22 +00:00

atanpi.cl

Implement atanpi builtin

2015-04-02 17:01:58 +00:00

binary_impl.inc

…

clc_ldexp.cl

Move ldexp soft implementation to a separate file

2015-05-06 21:59:29 +00:00

clc_nextafter.cl

…

clc_sqrt_impl.inc

Fix implementation of sqrt v2

2015-07-10 13:37:07 +00:00

clc_sqrt.cl

Fix implementation of sqrt v2

2015-07-10 13:37:07 +00:00

copysign.cl

…

cos.cl

Implement cos for double types

2015-05-12 17:18:46 +00:00

cospi.cl

…

ep_log.cl

Implement asinh builtin

2015-04-02 17:01:54 +00:00

ep_log.h

Implement asinh builtin

2015-04-02 17:01:54 +00:00

erf.cl

math: Add erf ported from amd-builtins

2016-05-06 18:02:30 +00:00

erfc.cl

math: Implement erfc

2015-03-18 21:52:07 +00:00

exp2.cl

Implement exp2 using OpenCL C rather than using an intrinsic

2015-05-13 03:55:07 +00:00

exp10.cl

…

exp10.inc

…

exp_helper.cl

Implement exp2 using OpenCL C rather than using an intrinsic

2015-05-13 03:55:07 +00:00

exp_helper.h

Implement exp2 using OpenCL C rather than using an intrinsic

2015-05-13 03:55:07 +00:00

exp.cl

Use a more accurate implementation for exp

2015-05-13 03:55:09 +00:00

fdim.cl

math: Add fdim implementation

2016-05-06 03:34:45 +00:00

fdim.inc

math: Use single precision fmax in sp path

2016-05-17 19:44:01 +00:00

fmax.cl

Implement fmax using __builtin_fmax

2015-03-31 16:59:23 +00:00

fmax.inc

Implement fmax using __builtin_fmax

2015-03-31 16:59:23 +00:00

fmin.cl

Implement fmin using __builtin_fmin

2015-03-31 16:59:21 +00:00

fmin.inc

Implement fmin using __builtin_fmin

2015-03-31 16:59:21 +00:00

fmod.cl

…

fract.cl

Implement fract builtin

2015-04-23 18:50:14 +00:00

fract.inc

Implement fract builtin

2015-04-23 18:50:14 +00:00

frexp.cl

math: Add frexp ported from amd-builtins

2016-02-08 17:07:21 +00:00

frexp.inc

math: Add frexp ported from amd-builtins

2016-02-08 17:07:21 +00:00

half_rsqrt.cl

Implement half_rsqrt builtin v3

2015-05-08 23:28:44 +00:00

half_rsqrt.inc

Implement half_rsqrt builtin v3

2015-05-08 23:28:44 +00:00

half_sqrt.cl

math: limit half_sqrt to single precision

2015-05-09 22:31:03 +00:00

half_sqrt.inc

Implement half_sqrt builtin v2

2015-03-23 17:01:37 +00:00

hypot.cl

…

hypot.inc

…

ilogb.cl

math: Fix ilogb(double) return type

2016-02-24 00:52:15 +00:00

ldexp.cl

Fix ldexp fp64 build error

2015-05-09 22:30:59 +00:00

ldexp.inc

math: Add ldexp implementation

2015-05-06 20:53:32 +00:00

log1p.cl

…

log2.cl

math: Fix log2 vectorization on non-fp64 hw

2016-02-09 22:17:42 +00:00

log10.cl

…

log10.inc

…

log_base.h

Implement accurate log2 function

2015-07-24 18:07:12 +00:00

log.cl

Fix double implementation of log

2015-07-24 18:07:14 +00:00

mad.cl

…

mad.inc

…

math.h

…

modf.cl

Implement modf math builtin

2016-01-27 14:52:10 +00:00

modf.inc

Implement modf math builtin

2016-01-27 14:52:10 +00:00

native_log2.cl

Use llvm intrinsics for native_log and native_log2

2015-07-24 18:07:06 +00:00

native_log2.inc

Use llvm intrinsics for native_log and native_log2

2015-07-24 18:07:06 +00:00

native_log.cl

Use llvm intrinsics for native_log and native_log2

2015-07-24 18:07:06 +00:00

native_log.inc

Use llvm intrinsics for native_log and native_log2

2015-07-24 18:07:06 +00:00

nextafter.cl

…

pown.cl

…

sin.cl

Implement sin for double types

2015-05-12 17:18:47 +00:00

sincos_helpers.cl

Implement cos for double types

2015-05-12 17:18:46 +00:00

sincos_helpers.h

Implement cos for double types

2015-05-12 17:18:46 +00:00

sincos.cl

…

sincos.inc

…

sincosD_piby4.h

…

sincospiF_piby4.h

…

sinpi.cl

Implement sinpi builtin

2015-05-06 21:59:26 +00:00

sqrt.cl

Fix implementation of sqrt v2

2015-07-10 13:37:07 +00:00

tables.cl

Implement accurate log2 function

2015-07-24 18:07:12 +00:00

tables.h

Implement accurate log2 function

2015-07-24 18:07:12 +00:00

tan.cl

…

tan.inc

Fix compilation warnings without cl_khr_fp64

2015-04-24 19:54:17 +00:00

tanh.cl

Implement tanh builtin

2015-09-29 06:39:09 +00:00