[llvm-bugs] [Bug 37890] Equivalent vector shuffles emit widely different x86 avx2 code

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=37890

Simon Pilgrim <llvm-dev at redking.me.uk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
 Fixed By Commit(s)|                            |56a8a5c9fe24
         Resolution|---                         |FIXED

--- Comment #13 from Simon Pilgrim <llvm-dev at redking.me.uk> ---
(In reply to Simon Pilgrim from comment #12)
> The next step on this is probably to add support for EXTRACT_SUBVECTOR
> patterns further up a reduction chain in SelectionDAG::matchBinOpReduction

In conjunction with the add(extract(x),extract(x)) ->
extract(add(x,shuffle(x))) combine in the vector-combiner pass, this was fixed
by https://reviews.llvm.org/rG56a8a5c9fe24

-- 
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/20200705/9868e6f2/attachment.html>