[PATCH] D24681: Optimize patterns of vectorized interleaved memory accesses for X86.
Simon Pilgrim via llvm-commits
llvm-commits at lists.llvm.org
Fri Sep 30 06:49:18 PDT 2016
RKSimon added a comment.
What's the reasoning behind adding X86InterleavedAccess.cpp instead of including inside X86ISelLowering.cpp?
> X86InterleavedAccess.cpp:150
> +
> + if (Shuffles.size() != UniqueIndices.size())
> + return false;
Move both the tests to the top and test (Shuffles.size() != Indices.size()) directly?
> X86InterleavedAccess.cpp:155
> + return false;
> + }
> +
Remove braces (style).
> x86-interleaved-access.ll:1
> +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
> +; RUN: llc -mtriple=x86_64-pc-linux -mattr=+avx < %s | FileCheck %s --check-prefix=AVX --check-prefix=AVX1
Is this actually true? The checks below don't look like what the script would generate.
https://reviews.llvm.org/D24681
More information about the llvm-commits
mailing list