[LLVMbugs] [Bug 21385] optimize reciprocals with fast-math (x86)

[bugzilla-daemon at llvm.org]

http://llvm.org/bugs/show_bug.cgi?id=21385

Sanjay Patel <spatel+llvm at rotateright.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

--- Comment #19 from Sanjay Patel <spatel+llvm at rotateright.com> ---
Refinement accuracy made configurable via cl::opt:
http://llvm.org/viewvc/llvm-project?view=revision&revision=221819

With that, I think we can resolve this bug as fixed.

The customer can control generation of rcpps/rcpss with "-use-recip-est" and
control the speed vs. accuracy trade-off with "-x86-recip-refinement-steps".

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20141112/979b2c9b/attachment.html>