On 2/16/06, Bob Ippolito <[EMAIL PROTECTED]> wrote: > On Feb 16, 2006, at 11:45 AM, Gustavo Sverzut Barbieri wrote: > > On 2/16/06, Mike Orr <[EMAIL PROTECTED]> wrote: > >> > >> On 2/16/06, Kevin Dangoor <[EMAIL PROTECTED]> wrote: > >>> > >>> On 2/15/06, Mike Orr <[EMAIL PROTECTED]> wrote: > >>>> Then I discovered if the server is not handling a request, pressing > >>>> ctrl-C quits all the threads. But if it is handling a request, the > >>>> threads remain. If I then press Stop in the browser, the threads > >>>> still remain, and I have to kill them. Whereas if I kill the > >>>> threads > >>>> with extreme prejudice, the browser suddenly times out. > >>>> > >>>> I'd expected the threads to just halt when I press ctrl-C and the > >>>> browser to time out immediately. But I haven't used a > >>>> multithreaded > >>>> framework for a while so I'm not sure this is possible. Still, the > >>>> main thread can tell the others to hurry up and exit, no? > >>> > >>> You can use a global variable that tells the other threads to exit > >>> when they get around to it. I don't know of a way for the "main > >>> thread" to "kill" the other threads. > >> > >> That's one thing I never understood. Java does it, but Python claims > >> it's impossible. > > > > Linux have pthread_kill(), but maybe it's too dependent on platform > > and Python doesn't support it. > > > > Java, AFAIR, does its own thread implementation.... but I may be > > wrong. > > > > I'm also looking for a solution in this area :-( > > You're out of luck, there is not and never will be a solution to this > "problem". > > It's simply not possible to safely kill a thread in Python without > the consent of that thread, because it's not designed in a way that > makes that a safe operation. Java was designed with that constraint, > but I'm sure that even Java fails to do this if the runaway thread is > stuck in native code. > > Using pthread_kill would be very bad -- imagine that the killed > thread was holding the GIL at the time. Assuming the platform > supported that function: you'd kill the thread, but you'd also > deadlock Python because every other thread would wait forever on the > GIL. > > IOTW, If you want killable execution contexts that don't require > instrumentation, you must use processes. You have no other option.
That was my guess... :-) -- Gustavo Sverzut Barbieri -------------------------------------- Computer Engineer 2001 - UNICAMP Mobile: +55 (19) 9165 8010 Phone: +1 (347) 624 6296; [EMAIL PROTECTED] Jabber: [EMAIL PROTECTED] ICQ#: 17249123 MSN: [EMAIL PROTECTED] GPG: 0xB640E1A2 @ wwwkeys.pgp.net --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "TurboGears" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears -~----------~----~----~----~------~----~------~--~---
