Messages in this thread | | | Date | Fri, 9 Jul 2010 18:50:47 +0300 | From | Gleb Natapov <> | Subject | Re: [PATCH v4 08/12] Inject asynchronous page fault into a guest if page is swapped out. |
| |
On Thu, Jul 08, 2010 at 08:10:31PM +0200, Peter Zijlstra wrote: > On Thu, 2010-07-08 at 20:09 +0200, Peter Zijlstra wrote: > > On Thu, 2010-07-08 at 21:05 +0300, Gleb Natapov wrote: > > > > > + /* do alloc atomic since if we are going to sleep anyway we > > > > > + may as well sleep faulting in page */ > > > > > + work = kmem_cache_zalloc(async_pf_cache, GFP_ATOMIC); > > > > > + if (!work) > > > > > + return 0; > > > > > > > > GFP_KERNEL is fine for this context. > > > But it can sleep, no? The comment explains why I don't want to sleep > > > here. > > > > In that case, use 0, no use wasting __GFP_HIGH on something that doesn't > > actually need it. > > Ah, I just saw we have GFP_NOWAIT for that. Indeed. Will use GFP_NOWAIT.
-- Gleb.
| |