[PATCH] D48920: [NEON] Fix combining of vldx_dup intrinsics with updating of base addresses
Ivan Kosarev via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Wed Jul 4 02:51:41 PDT 2018
kosarev added inline comments.
================
Comment at: lib/Target/ARM/ARMISelLowering.cpp:11492
+ // combining base updates for such intrinsics.
+ continue;
case Intrinsic::arm_neon_vld2lane: NewOpc = ARMISD::VLD2LN_UPD;
----------------
dnsampaio wrote:
> Perhaps I got something wrong, but the desired actions for these 3 intrinsics aren't already defined in the "else" statement below, where it already receives the desired node type? Just need to define isLane=true and NumVecs and NewOpc accordingly.
Yes, we have the VLDxDUP_UPD nodes, but they are not supported for 256-bit vectors (q versions of the intrinsics) yet.
https://reviews.llvm.org/D48920
More information about the llvm-commits
mailing list