Re: [PATCH v2 0/2] Lock and Pointer guards

[Paolo Bonzini]

On 6/8/23 22:14, Nick Desaulniers wrote:
> Here, we're talking about using __attribute__((cleanup())) to DTR
> locally, but then we return a "raw" pointer to a caller. What cleanup
> function should the caller run, implicitly, if at all?  If we use
> __attribute__((cleanup())) that saves us a few gotos locally, but the
> caller perhaps now needs the same treatment.

But this is only a problem when you return a void*; and in general in C 
you will return a struct more often than a raw pointer (and in C++ you 
also have the issue of delete vs. delete[], that does not exist in C).

Returning a struct doesn't protect against use-after-free bugs in the 
way std::unique_ptr<> or Rust lifetimes do, but it at least tries to 
protect against calling the wrong cleanup function if you provide a 
typed "destructor" function that does the right thing---for example by 
handling reference counting or by freeing sub-structs before calling 
kfree/vfree.

Of course it's not a silver bullet, but then that's why people are 
looking into Rust for Linux.

Paolo