[PATCH] D143565: [Asan] Ensure unpoisonning doesn't get inlined unnecessarily due to small holes in the mask

[Sylvain Audi via Phabricator via llvm-commits]

saudi created this revision.
saudi added reviewers: rnk, steveire, rsmith.
saudi added projects: LLVM, Sanitizers, All.
Herald added subscribers: Enna1, hiraditya.
saudi requested review of this revision.
Herald added a subscriber: llvm-commits.

In the case of allocas introduced by the Address Sanitizer, the shadow bytes may contain sparse zeros, missing opportunities for grouping into __asan_set_shadow_00 calls, causing inlining despite the "asan-max-inline-poisoning-size" argument value.

This fix ignores small holes in the shadow bytes when unpoisonning.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D143565

Files:
  llvm/lib/Transforms/Instrumentation/AddressSanitizer.cpp
  llvm/test/Instrumentation/AddressSanitizer/asan-optimize-inline-poisoning.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D143565.495775.patch
Type: text/x-patch
Size: 4592 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230208/2e947e4c/attachment.bin>