On Thu, Apr 16, 2020 at 11:29:48AM -0700, Paul E. McKenney wrote: > On Thu, Apr 16, 2020 at 08:23:09PM +0200, Sebastian Andrzej Siewior wrote: > > On 2020-04-16 10:29:53 [-0700], Paul E. McKenney wrote: > > > > migrate_disable() would work. > > > > > > Or moving to a raw spinlock, correct? > > > > I'm currently in favour of that migrate_disable(). raw spinlock does not > > work because there is memory allocation via get_page() later on. > > Fair enough! IIRC, get_page() uses non-raw spinlocks, so... Oh, I see Paul made the same suggestion I did for raw spinlocks. migrate_disable() is fine, but note that in the later code we will not be sleeping even during the get_page(), so I don't see a problem with raw spinlock... did I miss something? thanks, - Joel