Re: MPS prstack

["Gerd Möllmann" <gerd.moellmann@gmail.com>]

Helmut Eller <eller.helmut@gmail.com> writes:

> On Sat, May 25 2024, Gerd Möllmann wrote:
>
>> Moint Helmut. I've pushed another fix that prevents processing non-Lisp
>> objs in sub char-tables.
>
> Oops.  Would be nice if print-tests.el would cover some of the more
> exotic cases.
>
> Somewhat related: I wonder if this line in print_stack_push is safe:
>
>   prstack.stack[prstack.sp++] = e;
>
> This is not an atomic operation and prstack.sp gets updated before the
> new stack slot is initialized.  In theory, there is a small time window
> where scan_prstack can see the new prstack.sp but not the new value for
> the slot.  Do we need to worry about such issues?

Hm. No idea when the increment is sequenced. The side effect of postfix
++ could be applied anywhere before the next sequence point. But where
is that sequence point? There is one at the end of the whole expression
I think.

Reminds me why I avoid postfix ++ and -- :-)