On Thu, Feb 25, 2010 at 03:39:41PM +0100, Oliver Hartkopp wrote: > Kurt Van Dijck wrote: > > Hello, > > > > The pcmcia_request_window() is postponed a little, as some flags must be > > set before this call (I believe). It is a small change, but with some > > noise in the patch. > > Same with softing_card_irq() > > > > The netif_start_queue() is omitted, as softing_cycle() does a > > netif_wake_queue() > > Looks good! No problems anymore. > > If it's ok for you, i can commit it to the SVN?! it's ok for me. It is a proper step. > > [...] > Mine is similar. But i'm also using a NO_HZ setup ... should not make a difference. I use 64bit (you might have noticed earlier). I noticed you compiled SMP too (as I did). That would have been my next guess. Now, I'm out of proper approaches. If I read well, &card->fw.lock 'depends' on s_active. My brain is too small to get the point. I just don't see what I could do to let a mutex depend on another one.
Kurt _______________________________________________ Socketcan-core mailing list [email protected] https://lists.berlios.de/mailman/listinfo/socketcan-core
