[PATCH] Short-term workaround for hidden x86 float precision in calculateSpillWeightAndHint()

Matt Arsenault arsenm2 at gmail.com
Mon Apr 21 11:32:07 PDT 2014

On Apr 21, 2014, at 10:19 AM, Hal Finkel <hfinkel at anl.gov> wrote:

> To be honest, I don't understand why you're seeing this problem. I know this was a problem pre-SSE, but I think that once SSE became available, the cost of forcing IEEE strict math became reasonable. IEEE arithmetic (although not the transcendental functions) should be strictly deterministic across platforms, and I don't see any reason we should support building LLVM in any other mode.

Even pre-SSE, you can change the default excess precision mode for x87 to force 32/64-bit rounding. The default is different on different platforms, and I thought I saw somewhere in LLVM the wrappers around all the different wrappers for fldcw before.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140421/ef933240/attachment.html>

More information about the llvm-commits mailing list