Re: multithreading status

[<don...@is...>]

Sam Steingold writes:
 > >   > Vladimir, do you agree with this "warning":
 > >   > http://clisp.podval.org/impnotes/mt.html#mt-unsafe
 > > I notice setf aref in this list.  Why is that unsafe?
 > it is not.
Then I guess that should be removed.

 > >  I would appreciate an explanation (which I suggest adding to this
 > >  section of the doc) of what exactly can go wrong in these cases
 > >  and how that leads to segfaults.
 > I am not interested in documenting bugs in detail.
You mean you don't want to spend your time on that or you don't think
it should be in the impnotes?
Is there an explanation in some clisp-devel message?
If you could point me to an existing explanation I'd like to read it
and would even try to supply some text for impnotes.
If there is no existing explanation, I'd appreciate a little help in
understanding where the segfaults come from.  (Is it only segfaults?)

 > >  Also, I'd have thought it would be possible to write hashtable code so
 > >  that mt causes incorrect results but no segfaults (or similar).
 > http://www.cygwin.com/acronyms/#PTC
After I understand what causes the segfaults then maybe I can
suggest a patch to avoid them.