Mikhail R. Gadelha
Mikhail R. Gadelha
I think the new vector extension is reason enough to make a new release! I'd say it's a quite big extension. off-topic: Rafael, do you think we can extend your...
Can we extend it to support https://en.cppreference.com/w/cpp/numeric/complex? The operations should be distributed between the real and img parts, just like vectors. Em seg., 25 de jul. de 2022 às 05:14,...
Merging the IR has been a pain for a long time :( The best solution IMHO is to use clang to merge to ASTs before we convert them to our...
Do you have a simple test that doesn't work because of the model? Does it work with the system's cstring?
No, what I meant to ask is if you are using the system's cstring/string headers in whatever test you're running, i.e, run the tests without the "-I ". That should...
The gcc is our fault: The error: /../lib/gcc/x86_64-linux-gnu/9/../../../../include/c++/9/bits/char_traits.h:372:33: error: use of undeclared identifier 'memcpy'; did you mean 'wmemcpy'? return static_cast(__builtin_memcpy(__s1, __s2, __n)); ^ Our call: clang Invocation: "clang-tool" "-cc1" "-triple"...
The `"__builtin_memcpy=memcpy"* "-D"` should've been bold in the previous message. Should we get rid of it, update `clang_c_adjust::do_special_functions` to remove the `__builtin_` prefix from functions that we don't handle and...
> src/c2goto/library/string.c and src/cpp/library/cstring --- they look > very similar. Some functions have sutble differences, but are functionally > same, e.g. strlen and memmove. Perhaps, the latter is a copy...
Ops. [1] https://gist.github.com/mikhailramalho/0472f7e2e407082248e7
Maybe duplicate of #92?