Displaying 1 result from an estimated 1 matches for "fp_rounding_sensit".
2014 Mar 26
3
[LLVMdev] [cfe-dev] computing a conservatively rounded square of a double
...know we don't
currently support floating point context and flags, but not being able
to support standard idioms from numerical computation seems very
problematic.
Here's one proposal to get the conversation started - please don't take
it too seriously.
We could introduce an "fp_rounding_sensitive" annotation (flag?) on the
instruction level which disables optimizations sensative to floating
point rounding. This could be parsed by clang as either a function
attribute or statement attribute.
Alternatively, we could introduce an "fp_rounding_mode(x)" annotation
with the...