[llvm-bugs] [Bug 46115] -instcombine with fast flag moves div outside of loop into loop, dramatically regressing performance

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=46115

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

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

--- Comment #8 from Sanjay Patel <spatel+llvm at rotateright.com> ---
(In reply to David Bolvansky from comment #7)
> Is there any testcase in llvm tree to ensure that things will not regress if
> someone in the future tries to add instcombine after licm?

Good point - added a test based on the example here:
https://github.com/llvm/llvm-project/commit/91b45fb527af8f1f83b51e7238642d3af011bad4

That shows the CSE opportunity. This bug is about fdiv sinking/hoisting, so
resolving since there's a Julia bug for that now.

-- 
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/20200531/709201d9/attachment.html>