| Message ID | 20210104081631.2921415-1-lixinhai.lxh@gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [V2] mm/hugetlb.c: fix unnecessary address expansion of pmd sharing | expand |
On 1/4/21 12:16 AM, Li Xinhai wrote: > The current code would unnecessarily expand the address range. Consider > one example, (start, end) = (1G-2M, 3G+2M), and (vm_start, vm_end) = > (1G-4M, 3G+4M), the expected adjustment should be keep (1G-2M, 3G+2M) > without expand. But the current result will be (1G-4M, 3G+4M). Actually, > the range (1G-4M, 1G) and (3G, 3G+4M) would never been involved in pmd > sharing. > > After this patch, we will check that the vma span at least one PUD > aligned size and the start,end range overlap the aligned range of vma. > > With above example, the aligned vma range is (1G, 3G), so if (start, end) > range is within (1G-4M, 1G), or within (3G, 3G+4M), then no adjustment > to both start and end. Otherwise, we will have chance to adjust start > downwards or end upwards without exceeding (vm_start, vm_end). > > Fixes: 75802ca66354 ("mm/hugetlb: fix calculation of adjust_range_if_pmd_sharing_possible") > Suggested-by: Mike Kravetz <mike.kravetz@oracle.com> > Cc: Mike Kravetz <mike.kravetz@oracle.com> > Cc: Peter Xu <peterx@redhat.com> > Signed-off-by: Li Xinhai <lixinhai.lxh@gmail.com> > --- Thanks again. Comments and discussion in previous thread. Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
On 1/6/21 2:23 AM, Mike Kravetz wrote: > On 1/4/21 12:16 AM, Li Xinhai wrote: >> The current code would unnecessarily expand the address range. Consider >> one example, (start, end) = (1G-2M, 3G+2M), and (vm_start, vm_end) = >> (1G-4M, 3G+4M), the expected adjustment should be keep (1G-2M, 3G+2M) >> without expand. But the current result will be (1G-4M, 3G+4M). Actually, >> the range (1G-4M, 1G) and (3G, 3G+4M) would never been involved in pmd >> sharing. >> >> After this patch, we will check that the vma span at least one PUD >> aligned size and the start,end range overlap the aligned range of vma. >> >> With above example, the aligned vma range is (1G, 3G), so if (start, end) >> range is within (1G-4M, 1G), or within (3G, 3G+4M), then no adjustment >> to both start and end. Otherwise, we will have chance to adjust start >> downwards or end upwards without exceeding (vm_start, vm_end). >> >> Fixes: 75802ca66354 ("mm/hugetlb: fix calculation of adjust_range_if_pmd_sharing_possible") >> Suggested-by: Mike Kravetz <mike.kravetz@oracle.com> >> Cc: Mike Kravetz <mike.kravetz@oracle.com> >> Cc: Peter Xu <peterx@redhat.com> >> Signed-off-by: Li Xinhai <lixinhai.lxh@gmail.com> >> --- > > Thanks again. Comments and discussion in previous thread. > > Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com> > Thanks for review.
diff --git a/mm/hugetlb.c b/mm/hugetlb.c index cbf32d2824fd..a41ce2b22275 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -5239,21 +5239,23 @@ static bool vma_shareable(struct vm_area_struct *vma, unsigned long addr) void adjust_range_if_pmd_sharing_possible(struct vm_area_struct *vma, unsigned long *start, unsigned long *end) { - unsigned long a_start, a_end; + unsigned long v_start = ALIGN(vma->vm_start, PUD_SIZE), + v_end = ALIGN_DOWN(vma->vm_end, PUD_SIZE); - if (!(vma->vm_flags & VM_MAYSHARE)) + /* + * vma need span at least one aligned PUD size and the start,end range + * must at least partialy within it. + */ + if (!(vma->vm_flags & VM_MAYSHARE) || !(v_end > v_start) || + (*end <= v_start) || (*start >= v_end)) return; /* Extend the range to be PUD aligned for a worst case scenario */ - a_start = ALIGN_DOWN(*start, PUD_SIZE); - a_end = ALIGN(*end, PUD_SIZE); + if (*start > v_start) + *start = ALIGN_DOWN(*start, PUD_SIZE); - /* - * Intersect the range with the vma range, since pmd sharing won't be - * across vma after all - */ - *start = max(vma->vm_start, a_start); - *end = min(vma->vm_end, a_end); + if (*end < v_end) + *end = ALIGN(*end, PUD_SIZE); } /*
The current code would unnecessarily expand the address range. Consider one example, (start, end) = (1G-2M, 3G+2M), and (vm_start, vm_end) = (1G-4M, 3G+4M), the expected adjustment should be keep (1G-2M, 3G+2M) without expand. But the current result will be (1G-4M, 3G+4M). Actually, the range (1G-4M, 1G) and (3G, 3G+4M) would never been involved in pmd sharing. After this patch, we will check that the vma span at least one PUD aligned size and the start,end range overlap the aligned range of vma. With above example, the aligned vma range is (1G, 3G), so if (start, end) range is within (1G-4M, 1G), or within (3G, 3G+4M), then no adjustment to both start and end. Otherwise, we will have chance to adjust start downwards or end upwards without exceeding (vm_start, vm_end). Fixes: 75802ca66354 ("mm/hugetlb: fix calculation of adjust_range_if_pmd_sharing_possible") Suggested-by: Mike Kravetz <mike.kravetz@oracle.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: Peter Xu <peterx@redhat.com> Signed-off-by: Li Xinhai <lixinhai.lxh@gmail.com> --- v2<-v1: Mike found that previous patch didn't fix all possible cases, and proposed new patch to fully fix the existing issue, and also suggested to consider checking the vma's size at the begining of the function. Yes, we can simplify the code a bit wiht less comparison by aligning vma's size first and then do the checking with start,end. Hope this new one is also easier to understand. mm/hugetlb.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-)