Regression in ash (crashing)

[Leonid Lisovskiy]

On Thu Dec 13 19:49:39 UTC 2012, Rich Felker wrote:

> I can try investigating further using them, but if you read my
> most recent post to this thread, you'll see that I'm fairly certain
> the problem is just that the stunalloc operation is fundamentally
> impossible at the point it was added in
> 4ba6c5d3ba3d2c7922aff6b5c2e73b8325f1cf17 due to the nature of the
> stack allocator. I don't think this bug is any deeper than that.

Yes, I agree with you. But copynode() should be improved in any case, IMHO.
Unfortunately, I still can't find proper place for it(deallocator).

Probably we should create bug report with reproducible minimal test-cases?

Leonid