Jan Vesely 3b8b4eb64d half_powr: Implement using powr ...

v2: Use full precision implementation

Reviewer: Jeroen Ketema <j.ketema@xs4all.nl>
Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
llvm-svn: 323942

2018-02-01 03:00:35 +00:00

..

async

…

atomic

…

cl_khr_global_int32_base_atomics

…

cl_khr_global_int32_extended_atomics

…

cl_khr_int64_base_atomics

Implement cl_khr_int64_base_atomics builtins

2017-09-20 20:42:14 +00:00

cl_khr_int64_extended_atomics

Implement cl_khr_int64_extended_atomics builtins

2017-09-20 20:42:19 +00:00

cl_khr_local_int32_base_atomics

…

cl_khr_local_int32_extended_atomics

…

common

…

explicit_fence

amdgcn: Implement {read_,write_,}mem_fence builtin

2017-08-16 17:08:56 +00:00

float

math: Use precomputed constant for log2(10.0)

2017-11-13 18:28:45 +00:00

geometric

geometric: geometric functions are only supported for vector lengths <=4

2017-09-29 19:06:47 +00:00

image

…

integer

integer: Add popcount implementation using ctpop intrinsic

2017-09-09 02:23:54 +00:00

math

half_powr: Implement using powr

2018-02-01 03:00:35 +00:00

misc

relational: Implement shuffle2 builtin

2017-09-02 02:23:28 +00:00

relational

…

shared

shared: Implement aligned vector stores (vstorea_half)

2017-10-22 14:21:59 +00:00

synchronization

…

workitem

Let get_work_dim take exactly 0 arguments

2017-10-01 20:11:46 +00:00

as_type.h

…

clc.h

half_powr: Implement using powr

2018-02-01 03:00:35 +00:00

clcfunc.h

…

clcmacros.h

Do no circularly define NULL

2017-10-01 20:10:14 +00:00

clctypes.h

Add halfN types and enable fp16 when generating builtin declarations

2017-09-02 02:23:16 +00:00

convert.h

…