Re: question on irqs and memory ordering
Oliver Neukum <[email protected]>
| Newsgroups | gmane.linux.usb.devel |
|---|---|
| Message-ID | <[email protected]> |
Am Mittwoch, 14. März 2007 22:21 schrieb Pete Zaitcev: > On Wed, 14 Mar 2007 20:59:29 +0000, David Howells <[email protected]> wrote: > > > Oliver Neukum <[email protected]> wrote: > > > > > It is called in interrupt and uses no locking. What happens if the next > > > irq is processed on another cpu? Is that cpu guaranteed to see the updates > > > to the incremented variables? > > > > I thought that possibility was prevented by IRQ_INPROGRESS. > > As far as I can tell, Oliver is not talking about two CPUs executing > the same IRQ handler simultaneously (even the same chain, actually). > We know that it's impossible. I think he is concerned about CPU A > executing an interrupt handler, its stores getting stuck in its store > buffer or its write-back cache, the IRQ finished, IRQ get migrated > to CPU B, CPU B taking next interrupt and seeing old RAM state. Yes, exactly. > I don't see this possible, because we take too many spinlocks > when IRQ is processed and they definitely drain store buffers > and caches. Not to mention the IRQ migration from A to B... I am paranoid. But am I paranoid enough? Regards Oliver ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ [email protected] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel