Re: [PATCH] mm: hugetlb: initialize PG_reserved for tail pages of gigantig compound pages

2013-10-10 Thread Rafael Aquini
On Thu, Oct 10, 2013 at 06:12:41PM +0200, Andrea Arcangeli wrote: > 11feeb498086a3a5907b8148bdf1786a9b18fc55 introduced a memory leak when > KVM is run on gigantic compound pages. > > 11feeb498086a3a5907b8148bdf1786a9b18fc55 depends on the assumption > that PG_reserved is identical for all head an

Re: [PATCH] mm: hugetlb: initialize PG_reserved for tail pages of gigantig compound pages

2013-10-10 Thread Rik van Riel
On 10/10/2013 12:12 PM, Andrea Arcangeli wrote: 11feeb498086a3a5907b8148bdf1786a9b18fc55 introduced a memory leak when KVM is run on gigantic compound pages. 11feeb498086a3a5907b8148bdf1786a9b18fc55 depends on the assumption that PG_reserved is identical for all head and tail pages of a compound

[PATCH] mm: hugetlb: initialize PG_reserved for tail pages of gigantig compound pages

2013-10-10 Thread Andrea Arcangeli
11feeb498086a3a5907b8148bdf1786a9b18fc55 introduced a memory leak when KVM is run on gigantic compound pages. 11feeb498086a3a5907b8148bdf1786a9b18fc55 depends on the assumption that PG_reserved is identical for all head and tail pages of a compound page. So that if get_user_pages returns a tail pa