[cfe-commits] r143136 - in /cfe/trunk: lib/CodeGen/CGExpr.cpp lib/CodeGen/CGExprScalar.cpp lib/CodeGen/CodeGenFunction.h test/CodeGenOpenCL/fpaccuracy.cl
clattner at apple.com
Thu Oct 27 14:58:31 CDT 2011
On Oct 27, 2011, at 12:19 PM, Peter Collingbourne wrote:
> Author: pcc
> Date: Thu Oct 27 14:19:51 2011
> New Revision: 143136
> URL: http://llvm.org/viewvc/llvm-project?rev=143136&view=rev
> 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.
Please make sure that these annotations are documented in LangRef.
More information about the cfe-commits