Messages in this thread | | | Date | Tue, 10 Dec 2013 17:38:25 +0900 | From | Joonsoo Kim <> | Subject | Re: [PATCH v2 2/7] mm/migrate: correct failure handling if !hugepage_migration_support() |
| |
On Mon, Dec 09, 2013 at 04:17:32PM +0000, Christoph Lameter wrote: > On Mon, 9 Dec 2013, Joonsoo Kim wrote: > > > We should remove the page from the list if we fail without ENOSYS, > > since migrate_pages() consider error cases except -ENOMEM and -EAGAIN > > as permanent failure and it assumes that the page would be removed from > > the list. Without this patch, we could overcount number of failure. > > Ok what does the patch do about this? I dont see any modifications. Remove > this part of the description?
Description is slightly wrong. Following is correct one.
"We should remove the page from the list if we fail *with* ENOSYS,"
And this patch do this by adding putback_active_hugepage(hpage) on ENOSYS case.
> > > In addition, we should put back the new hugepage if > > !hugepage_migration_support(). If not, we would leak hugepage memory. > > Ok looks like that is fixed by this patch. > > Acked-by: Christoph Lameter <cl@linux.com>
Thanks.
| |