rsmith added inline comments.
================
Comment at: test/SemaCXX/source_location.cpp:10-35
+struct source_location {
+private:
+ unsigned int __m_line = 0;
+ unsigned int __m_col = 0;
+ const char *__m_file = nullptr;
+ const char *__m_func = nullptr;
+public:
----------------
This seems suboptimal. It would seem better for the compiler to generate a
global containing the relevant data and to represent a `source_location` as a
pointer to such a value. We should also try to minimize the number of
relocations necessary to build a `source_location` object, since such
constructions are likely to be extremely common in some codebases. We should
also keep in mind that we're likely to want to add fields to `source_location`
in future, so designing it in a way that avoids an ABI break for such cases
would be preferable.
How long has GCC supported this?
https://reviews.llvm.org/D37035
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
