Re: race condition in run-time system
Taylor R Campbell <[email protected]>
| Newsgroups | gmane.lisp.scheme.scheme48 |
|---|---|
| Message-ID | <[email protected]> |
It occurs to me that in the current implementation of the compiler and virtual machine, no interrupts will ever be delivered between the STORED-OBJECT-REF and STORED-OBJECT-SET! instructions in the code for PERIODICALLY-FORCE-OUTPUT!. This means that adding the clause (INTEGRATE #T) to the I/O package will change the semantics of the procedure and introduce the race condition I described. (Any other changes that cause the procedure to involve procedure calls will have the same effect.) This is surely not a desirable state of affairs.