Re: [PATCH v4 50/66] mm/gup: Use maple tree navigation instead of linked list

From: Vlastimil Babka
Date: Wed Jan 19 2022 - 12:39:21 EST


On 12/1/21 15:30, Liam Howlett wrote:
> From: "Liam R. Howlett" <Liam.Howlett@xxxxxxxxxx>

Empty message.

> Signed-off-by: Liam R. Howlett <Liam.Howlett@xxxxxxxxxx>
> ---
> mm/gup.c | 7 +++----
> 1 file changed, 3 insertions(+), 4 deletions(-)
>
> diff --git a/mm/gup.c b/mm/gup.c
> index 2c51e9748a6a..60892e5df6a2 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -1570,7 +1570,6 @@ int __mm_populate(unsigned long start, unsigned long len, int ignore_errors)
> long ret = 0;
>
> end = start + len;
> -
> for (nstart = start; nstart < end; nstart = nend) {
> /*
> * We want to fault in pages for [nstart; end) address range.
> @@ -1579,10 +1578,10 @@ int __mm_populate(unsigned long start, unsigned long len, int ignore_errors)
> if (!locked) {
> locked = 1;
> mmap_read_lock(mm);
> - vma = find_vma(mm, nstart);
> + vma = find_vma_intersection(mm, start, end);

Is it ok to use start instead of nstart? At least theoretically,
populate_vma_page_range() called later can reset locked to 0, so !locked
doesn't guarantee we are in the first iteration?

> } else if (nstart >= vma->vm_end)
> - vma = vma->vm_next;
> - if (!vma || vma->vm_start >= end)
> + vma = find_vma(mm, vma->vm_end);
> + if (!vma)
> break;
> /*
> * Set [nstart; nend) to intersection of desired address