[PATCH] D24159: Fix PR30202 - notify_all_at_thread_exit seg faults if run from a raw pthread context.

[Ben Craig via cfe-commits]

bcraig added a comment.

LGTM


================
Comment at: src/condition_variable.cpp:86
@@ +85,3 @@
+    if (tl_ptr.get() == nullptr) {
+        tl_ptr.set_pointer(new __thread_struct);
+    }
----------------
EricWF wrote:
> bcraig wrote:
> > The standard synopsis of notify_all_at_thread_exit in the standard doesn't have a "Throws:" clause.  I think that means that this is non-conformant.
> > 
> > There is some phrasing in 30.2.2 [thread.req.exception]:
> >    Failure to allocate storage shall be reported as described in 17.6.5.12 [res.on.exception.handling].
> > But I don't see wording in that section that gives permission to throw from functions without Throws clauses.
> > 
> > From a user perspective, I generally expect my synchronization primitives to be no-throw.  I don't know if it is really possible to provide that guarantee with notify_all_at_thread_exit though, especially if we want it to work outside of a std::thread.
> [res.on.exception.handling]/p4
> > 
> > Destructor operations defined in the C ++ standard library shall not throw exceptions. Every destructor
> > in the C ++ standard library shall behave as if it had a non-throwing exception specification. Any other
> > functions defined in the C ++ standard library that do not have an exception-specification may throw
> > implementation-defined exceptions unless otherwise specified. [188] An implementation may strengthen this
> > implicit exception-specification by adding an explicit one.
> 
> And the relevant note:
> 
> > 188) In particular, they can report a failure to allocate storage by throwing an exception of type bad_alloc, or a class derived
> > from bad_alloc (18.6.3.1). Library implementations should report errors by throwing exceptions of or derived from the standard
> > exception classes (18.6.3.1, 18.8, 19.2).
> 
> So throwing `std::bad_alloc` is allowed, as it is almost everywhere in the STL. Also the function already [allocates memory to store the new entry](https://github.com/llvm-mirror/libcxx/blob/master/src/thread.cpp#L202).
> 
> 
> 
> 
> 
> 
> 
> 
Yet again, I find that reading comprehension with regards to the standard isn't my strong suit :).  I got to "Destructor operations" in p4 and stopped reading.

I will still wish for a noexcept version of this function, but it's a pretty impractical wish.  The implementation you have here is fine.


https://reviews.llvm.org/D24159