[PATCH] D103907: [LV] Parallel annotated loop does not imply all loads can be hoisted.

[Michael Kruse via Phabricator via llvm-commits]

Meinersbur added inline comments.


================
Comment at: llvm/test/Transforms/LoopVectorize/X86/force-ifcvt.ll:1-6
+; This test is no longer valid, as !"llvm.loop.parallel_accesses" does not imply if-conversion safety anymore
+; The issues is explained here: https://bugs.llvm.org/show_bug.cgi?id=46666
+; XFAIL: *
 ; RUN: opt -loop-vectorize -S < %s | FileCheck %s
 target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
 target triple = "x86_64-unknown-linux-gnu"
----------------
I suggest to remove broken-by-design tests entirely.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D103907/new/

https://reviews.llvm.org/D103907