Re: [PATCH 3/4] mm: numa: Mark huge PTEs young when clearing NUMA hinting faults

2015-03-07 Thread Linus Torvalds
On Sat, Mar 7, 2015 at 10:33 AM, Linus Torvalds wrote: > > Completely untested, but that "just > or in the new protection bits" is what pnf_pte() does just a few lines > above this. Hmm. Looking at this, we do *not* want to set _PAGE_ACCESSED when we turn a page into PROT_NONE or mark

Re: [PATCH 3/4] mm: numa: Mark huge PTEs young when clearing NUMA hinting faults

2015-03-07 Thread Linus Torvalds
On Sat, Mar 7, 2015 at 7:20 AM, Mel Gorman wrote: > pmd = pmd_modify(pmd, vma->vm_page_prot); > + pmd = pmd_mkyoung(pmd); Hmm. I *thought* this should be unnecessary. vm_page_prot alreadty has the accessed bit set, and we kind of depend on the initial page table setup and mk_pte() a

[PATCH 3/4] mm: numa: Mark huge PTEs young when clearing NUMA hinting faults

2015-03-07 Thread Mel Gorman
Base PTEs are marked young when the NUMA hinting information is cleared but the same does not happen for huge pages which this patch addresses. Note that migrated pages are not marked young as the base page migration code does not assume that migrated pages have been referenced. This could be addre