Encountering flt_rounds_ in llvm3.3 for newlib compilation

I made the switch to llvm3.3, and encountered a flt_rounds

I'm using a soft float architecture and hopefully people have some
ideas on how to help:

I received:

i32 = flt_rounds

"Do not know how to promote this operator!"

I currently do not have any custom setting for the FLT_ROUNDS_

I'd like to just replace the FLT_ROUNDS_ with a "1" value.

Any thoughts on how to do this or mimic this from other places?

I see that other targets have hardware support for getting the actual
hardware setting of rounding mode, but like I said, I'm using a soft
float architecture
and this didn't come up during 3.2 compilation.