[PATCH] D71082: Allow system header to provide their own implementation of some builtin

Nico Weber via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Tue Feb 4 12:42:12 PST 2020


thakis added a comment.

This caused a regression when using `-fsanitize=memory -D_FORTIFY_SOURCE=2`. MemorySanitizer requires all code, including system libraries, to be instrumented. Else it reports false positives. For glibc, it has lots of interceptors to make sure glibc doesn't have to be instrumented (as far as I understand). So people usually don't have an instrumented glibc in their sysroot. But after this change, memcpy calls aren't always intercepted. I filed https://bugs.llvm.org/show_bug.cgi?id=44779 with a standalone repro.

I'm not sure what the best fix for that is. Unless it's something obvious, we should probably revert this while we figure things out.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D71082/new/

https://reviews.llvm.org/D71082





More information about the cfe-commits mailing list