Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

The C++ stdlib versions of vectors have different design goals, including a bunch of ABI constraints for very old code (things like the vector<char> specializations). Likewise there are questionable spec behaviors like unordered_map that specifies suboptimal implementation.


Follow on thought - what happens for operator new[]() in kernel space? Does it depend on a malloc (vs. traditional kmalloc)?


It's currently ending up in kmalloc(). The compiler handles the new[]/delete[] element count and I don't ask any questions about how.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: