Open nickdesaulniers opened 1 month ago
Hi!
This issue may be a good introductory issue for people new to working on LLVM. If you would like to work on this issue, your first steps are:
test/
create fine-grained testing targets, so you can e.g. use make check-clang-ast
to only run Clang's AST tests.git clang-format HEAD~1
to format your changes.If you have any further questions about this issue, don't hesitate to ask via a comment in the thread below.
@llvm/issue-subscribers-good-first-issue
Author: Nick Desaulniers (nickdesaulniers)
@llvm/issue-subscribers-libc
Author: Nick Desaulniers (nickdesaulniers)
I can take a look at this if possible 👍. Can you please assign this to me @nickdesaulniers
done!
@nickdesaulniers I have a quick question in regards to memcpy and memcmp. The Standard does not define that either of the pointers shouldn't be null and does not specify if there is undefined behavior if either is null.
Should I still add nullchecks to both ? Here's the Standard I'm currently referencing
@michaelrj-google @SchrodingerZhu any ideas on the question above here. Nick seems OOO for the week.
1896ee38898a73ea9c2894e848884c8999884ab1 added a config option to check for nullptr args to math.h functions. I think we could expand the usage of this macro to additional functions that accept pointer args that are defined by the standard to be non-null (or UB occurs in the case of NULL) (example:
memcpy
). This would allow us to provide a hardened config.