Messages in this thread | | | Date | Wed, 5 Nov 2014 08:04:09 -0800 (PST) | From | Hugh Dickins <> | Subject | Re: [PATCH 10/10] mm/hugetlb: share the i_mmap_rwsem |
| |
On Tue, 4 Nov 2014, Davidlohr Bueso wrote: > 8<------------------------------------------------ > From: Davidlohr Bueso <dave@stgolabs.net> > Subject: [PATCH 11/10] mm/memory.c: share the i_mmap_rwsem > > The unmap_mapping_range family of functions do the unmapping > of user pages (ultimately via zap_page_range_single) without > touching the actual interval tree, thus share the lock. > > Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Acked-by: Hugh Dickins <hughd@google.com>
Yes, thanks, let's get this 11/10 into mmotm along with the rest, but put the hugetlb 10/10 on the shelf for now, until we've had time to contemplate it more deeply.
> --- > mm/memory.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/mm/memory.c b/mm/memory.c > index 2ca3105..06f2458 100644 > --- a/mm/memory.c > +++ b/mm/memory.c > @@ -2396,12 +2396,12 @@ void unmap_mapping_range(struct address_space *mapping, > details.last_index = ULONG_MAX; > > > - i_mmap_lock_write(mapping); > + i_mmap_lock_read(mapping); > if (unlikely(!RB_EMPTY_ROOT(&mapping->i_mmap))) > unmap_mapping_range_tree(&mapping->i_mmap, &details); > if (unlikely(!list_empty(&mapping->i_mmap_nonlinear))) > unmap_mapping_range_list(&mapping->i_mmap_nonlinear, &details); > - i_mmap_unlock_write(mapping); > + i_mmap_unlock_read(mapping); > } > EXPORT_SYMBOL(unmap_mapping_range); > > -- > 1.8.4.5
| |