On Saturday 21 May 2005 12:14 am, Glenn Maynard wrote: > On Sat, May 21, 2005 at 12:07:15AM -0700, Pete Zaitcev wrote: > > On Sat, 21 May 2005 02:08:47 -0400, Glenn Maynard <[EMAIL PROTECTED]> wrote: > > > > > [...] It's not a correct fix, but due to time constraints > > > (and paranoia) I'm opting for the simplest fix that works. > > > > > @@ -2791,6 +2791,7 @@ > > > usb_unlock_device(hdev); > > > usb_put_intf(intf); > > > > > > + msleep(25); > > > } /* end while (1) */ > > > } > > > > I'm glad this works for you, though the remaining problem is that if > > we do not get a fix in mainline, you're bound to hit it again... > > I'm more concerned that others will hit it. I hope that somebody that knows > the code better than I finds a real fix, or that the code evolves beyond it > (not holding the hub lock for a second or more, etc). I just can't risk > regressions right now ...
What happens if you make that "while(1)" just be a single pass through the loop ... any improvements? - Dave ------------------------------------------------------- This SF.Net email is sponsored by Yahoo. Introducing Yahoo! Search Developer Network - Create apps using Yahoo! Search APIs Find out how you can build Yahoo! directly into your own Applications - visit http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005 _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
