Messages in this thread | | | Date | Tue, 27 Dec 2011 14:30:21 +0100 | From | Michal Hocko <> | Subject | Re: [PATCH] mm: hugetlb: add might_sleep() for gigantic page |
| |
On Tue 27-12-11 21:21:18, Hillf Danton wrote: > On Tue, Dec 27, 2011 at 8:59 PM, Michal Hocko <mhocko@suse.cz> wrote: > > On Fri 23-12-11 21:41:08, Hillf Danton wrote: > >> From: Hillf Danton <dhillf@gmail.com> > >> Subject: [PATCH] mm: hugetlb: add might_sleep() for gigantic page > >> > >> Like the case of huge page, might_sleep() is added for gigantic page, then > >> both are treated in same way. > > > > Why do we need to call might_sleep here? There is cond_resched in the > > loop... > > > > IIUC it is the reason to add... and the comment says
cond_resched calls __might_sleep so there is no reason to call might_sleep outside the loop as well.
-- Michal Hocko SUSE Labs SUSE LINUX s.r.o. Lihovarska 1060/12 190 00 Praha 9 Czech Republic
| |