Comment author: @ygrek

There is CAMLdrop now, see #4891

Comment author: goswin

Yes, but that's only have the solution imho.

Comment author: @ygrek

You mean "half of the solution"?
I think it is adequate solution and nothing more to fix here.

Comment author: @mshinwell

@Goswin Please explain why you think CAMLdrop is insufficient.

I think Goswin's concern may be that the call to CAMLlocal1 needs to be preceded by a call to CAMLparam0, and the documentation for CAMLparam0 says:

   The function body must start with one of the [CAMLparam] macros.

so it appears that the body of inner can't be inlined into outer, since that would place CAMLparam0 after the call to caml_leave_blocking_section (i.e. not at the beginning of the function).

I think CAMLparam0 can probably be safely be placed after the call to caml_leave_blocking_section, and the documentation is a bit too restrictive. If that's true then we can fix this issue by clarifying the documentation.

If compiling with a C90 compiler, the CAMLparam0 must be at the start of a block but that need not be at the start of a function. So the following works:

void foo(const char *name) {
    caml_leave_blocking_section();
    {
        CAMLparam0();
        CAMLlocal1(_name);
        _name = caml_copy_string(name);
        caml_callback1(_func, _name);
        CAMLdrop;
    }
    caml_enter_blocking_section();
}

This issue has been open one year with no activity. Consequently, it is being marked with the "stale" label. What this means is that the issue will be automatically closed in 30 days unless more comments are added or the "stale" label is removed. Comments that provide new information on the issue are especially welcome: is it still reproducible? did it appear in other contexts? how critical is it? etc.