Author: hans
Date: Thu Feb 22 05:24:27 2018
New Revision: 325776

ReleaseNotes: The fast-math-flags changes

By Sanjay Patel!


Modified: llvm/branches/release_60/docs/ReleaseNotes.rst
--- llvm/branches/release_60/docs/ReleaseNotes.rst (original)
+++ llvm/branches/release_60/docs/ReleaseNotes.rst Thu Feb 22 05:24:27 2018
@@ -74,6 +74,15 @@ Non-comprehensive list of changes in thi
 Changes to the LLVM IR
+* The fast-math-flags (FMF) have been updated. Previously, the 'fast' flag
+  indicated that floating-point reassociation was allowed and all other flags
+  were set too. The 'fast' flag still exists, but there is a new flag called
+  'reassoc' to indicate specifically that reassociation is allowed. A new bit
+  called 'afn' was also added to selectively allow approximations for common
+  mathlib functions like square-root. The new flags provide more flexibility
+  to enable/disable specific floating-point optimizations. Making the
+  optimizer respond appropriately to these flags is an ongoing effort.
 Changes to the AArch64 Target

llvm-branch-commits mailing list

Reply via email to