Re: [RFC PATCH 1/4] mm/hugetlb: Simplify hugetlb unmap

2016-06-02 Thread Aneesh Kumar K.V
Andrew Morton writes: > On Mon, 30 May 2016 11:14:19 +0530 "Aneesh Kumar K.V" > wrote: > >> For hugetlb like THP (and unlike regular page), we do tlb flush after >> dropping ptl. Because of the above, we don't need to track

Re: [RFC PATCH 1/4] mm/hugetlb: Simplify hugetlb unmap

2016-06-02 Thread Aneesh Kumar K.V
Andrew Morton writes: > On Mon, 30 May 2016 11:14:19 +0530 "Aneesh Kumar K.V" > wrote: > >> For hugetlb like THP (and unlike regular page), we do tlb flush after >> dropping ptl. Because of the above, we don't need to track force_flush >> like we do now. Instead we can simply call

Re: [RFC PATCH 1/4] mm/hugetlb: Simplify hugetlb unmap

2016-06-01 Thread Andrew Morton
On Mon, 30 May 2016 11:14:19 +0530 "Aneesh Kumar K.V" wrote: > For hugetlb like THP (and unlike regular page), we do tlb flush after > dropping ptl. Because of the above, we don't need to track force_flush > like we do now. Instead we can simply call

Re: [RFC PATCH 1/4] mm/hugetlb: Simplify hugetlb unmap

2016-06-01 Thread Andrew Morton
On Mon, 30 May 2016 11:14:19 +0530 "Aneesh Kumar K.V" wrote: > For hugetlb like THP (and unlike regular page), we do tlb flush after > dropping ptl. Because of the above, we don't need to track force_flush > like we do now. Instead we can simply call tlb_remove_page() which > will do the flush

Re: [RFC PATCH 1/4] mm/hugetlb: Simplify hugetlb unmap

2016-05-30 Thread Hillf Danton
> @@ -3157,19 +3156,22 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, > struct vm_area_struct *vma, > tlb_start_vma(tlb, vma); > mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end); > address = start; > -again: > for (; address < end; address += sz) {

Re: [RFC PATCH 1/4] mm/hugetlb: Simplify hugetlb unmap

2016-05-30 Thread Hillf Danton
> @@ -3157,19 +3156,22 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, > struct vm_area_struct *vma, > tlb_start_vma(tlb, vma); > mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end); > address = start; > -again: > for (; address < end; address += sz) {

[RFC PATCH 1/4] mm/hugetlb: Simplify hugetlb unmap

2016-05-29 Thread Aneesh Kumar K.V
For hugetlb like THP (and unlike regular page), we do tlb flush after dropping ptl. Because of the above, we don't need to track force_flush like we do now. Instead we can simply call tlb_remove_page() which will do the flush if needed. No functionality change in this patch. Signed-off-by:

[RFC PATCH 1/4] mm/hugetlb: Simplify hugetlb unmap

2016-05-29 Thread Aneesh Kumar K.V
For hugetlb like THP (and unlike regular page), we do tlb flush after dropping ptl. Because of the above, we don't need to track force_flush like we do now. Instead we can simply call tlb_remove_page() which will do the flush if needed. No functionality change in this patch. Signed-off-by: