Peter Collingbourne 95fd2ca69f Annotate imprecise FP division with fpaccuracy metadata ... The OpenCL single precision division operation is only required to be accurate to 2.5ulp. Annotate the fdiv instruction with metadata which signals to the backend that an imprecise divide instruction may be used. llvm-svn: 143136		2011-10-27 19:19:51 +00:00
..
2011-04-15-vec-init-from-vec.cl	Do not violate the opencl casting rules. This test case still illustrates the problem. In the future, we should throw an error when doing invalid casting.	2011-07-07 00:12:54 +00:00
address-spaces.cl	Add support for language-specific address spaces. On top of that,	2011-03-18 22:38:29 +00:00
ext-vector-shuffle.cl	…
fpaccuracy.cl	Annotate imprecise FP division with fpaccuracy metadata	2011-10-27 19:19:51 +00:00
kernel-metadata.cl	…
local.cl	OpenCL: introduce support for function scope __local variables	2011-09-19 21:14:35 +00:00
ptx-calls.cl	PTX: Set proper calling conventions for PTX in OpenCL mode.	2011-10-05 17:58:44 +00:00
ptx-kernels.cl	PTX: Set proper calling conventions for PTX in OpenCL mode.	2011-10-05 17:58:44 +00:00
single-precision-constant.cl	…
vector_literals_valid.cl	This handles the missing cases of opencl vector literals.	2011-07-15 23:07:01 +00:00