aaron.ballman requested changes to this revision.
aaron.ballman added a comment.
This revision now requires changes to proceed.
While I agree with the security aspects of this in principle, it is not a
conforming behavior in C and it runs significant risk of breaking existing code
such that it introduces new security issues. This behavior needs to be opt-in
(either through the existing attribute or through a feature flag of some kind).
It also points out that we really, really should have a diagnostic for
structure initialization using positional initialization instead of designated
initialization if the structure is randomized. (I think that case should be an
error, perhaps with an exception if the initialization is empty `{}`.)
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D123544/new/
https://reviews.llvm.org/D123544
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
