[PATCH] D67645: [aarch64] add def-pats for dot product
Sebastian Pop via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Mon Sep 16 22:28:25 PDT 2019
sebpop planned changes to this revision.
sebpop marked an inline comment as done.
sebpop added inline comments.
================
Comment at: llvm/lib/Target/AArch64/AArch64InstrInfo.td:7080
+ (i32 (ADDVv4i32v
+ (DOT (DUPv4i32gpr node:$Vo), node:$Vm, node:$Vn)))>;
+
----------------
I think there is an error in this case: as we duplicate the original value $Vo across the 4 lanes of the dot product, and then in the end we do the ADDV reduction across lanes, we end up with 4 times the original value.
I will prepare an updated patch to fix this.
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D67645/new/
https://reviews.llvm.org/D67645
More information about the llvm-commits
mailing list