I imagine it discourages pathologically poor performing code as in the following:

// atomic<Foo*> foo;
foo->do_one_thing();
foo->do_another_thing();
foo->do_a_final_thing();

You have now done an atomic load 3 times and it is possible that stores or loads to foo have occurred in between. The pattern for atomic usage is to typically load once, perform a change, and possibly compare/swap in a loop in a manner that prevents ABA bugs.

On Mon, Oct 21, 2019 at 1:38 AM Giuseppe D'Angelo via Std-Discussion <std-discussion@lists.isocpp.org> wrote:
Hello,

I've been wondering why std::atomic<T*> has an implicit operator T*()
(that simply does a seq_cst load), but not an operator->. I'm not a big
fan of having the implicit conversion operator in the first place, but
since it's there, why not also providing operator* / operator->? Is it
by design?


In other words, if this code is "reasonable":

> std::atomic<S *> a;
> S *s = a;
> s->f();

then why shouldn't one also be allowed to simply write this:

> std::atomic<S *> a;
> a->f();

If, on the other hand, one is not supposed to use the conversion
operator: should it be deprecated / turned into explicit?

(No, I'm not even remotely proposing to change the rules for operator->
expressions to consider implicit conversions for the lhs.)

Thanks for reading,
--
Giuseppe D'Angelo | giuseppe.dangelo@kdab.com | Senior Software Engineer
KDAB (France) S.A.S., a KDAB Group company
Tel. France +33 (0)4 90 84 08 53, http://www.kdab.com
KDAB - The Qt, C++ and OpenGL Experts

--
Std-Discussion mailing list
Std-Discussion@lists.isocpp.org
https://lists.isocpp.org/mailman/listinfo.cgi/std-discussion