[llvm-bugs] [Bug 40053] _mm_adds_epu8 and _mm_subs_epu8 with constant operand don't generate paddusb/psubusb

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=40053

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|CONFIRMED                   |RESOLVED
 Fixed By Commit(s)|                            |r349630, r349631

--- Comment #13 from Sanjay Patel <spatel+llvm at rotateright.com> ---
This got fixed in a different way with:
https://reviews.llvm.org/rL349630
https://reviews.llvm.org/rL349631

...so now we take the x86 intrinsics in C, translate them to LLVM intrinsics in
IR, translate those to generic DAG nodes, and isel those as the expected x86
instructions.

Should we still be able to recognize the raw IR patterns that correspond to
psubus and friends though because we may not always be able to canonicalize IR
to the LLVM intrinsics?

-- 
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/20181219/2d102d93/attachment.html>