[PATCH] D95763: [MemorySSA] Don't treat lifetime.end as NoAlias

[Nikita Popov via Phabricator via llvm-commits]

nikic created this revision.
nikic added reviewers: asbirlea, fhahn.
Herald added subscribers: george.burgess.iv, hiraditya, Prazek.
nikic requested review of this revision.
Herald added a project: LLVM.
Herald added a subscriber: llvm-commits.

MemorySSA currently treats lifetime.end intrinsics as not aliasing anything. This breaks MemorySSA-based MemCpyOpt, because we'll happily move a read of a pointer below a lifetime.end intrinsic, as no clobber is reported.

I think the MemorySSA modelling here isn't correct. lifetime.end(p) has approximately the same effect as doing a memcpy(p, undef), and should be treated as a clobber.

This patch removes the special handling of lifetime.end, leaving alias analysis to handle it appropriately.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D95763

Files:
  llvm/lib/Analysis/MemorySSA.cpp
  llvm/test/Analysis/MemorySSA/lifetime-simple.ll
  llvm/test/Transforms/MemCpyOpt/lifetime.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D95763.320359.patch
Type: text/x-patch
Size: 7358 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210131/f6f07184/attachment.bin>