lkml.org 
[lkml]   [2022]   [Aug]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH] hugetlb: freeze allocated pages before creating hugetlb pages
From
Date
On 2022/8/9 5:28, Mike Kravetz wrote:
> When creating hugetlb pages, the hugetlb code must first allocate
> contiguous pages from a low level allocator such as buddy, cma or
> memblock. The pages returned from these low level allocators are
> ref counted. This creates potential issues with other code taking
> speculative references on these pages before they can be transformed to
> a hugetlb page. This issue has been addressed with methods and code
> such as that provided in [1].
>
> Recent discussions about vmemmap freeing [2] have indicated that it
> would be beneficial to freeze all sub pages, including the head page
> of pages returned from low level allocators before converting to a
> hugetlb page. This helps avoid races if want to replace the page
> containing vmemmap for the head page.
>
> There have been proposals to change at least the buddy allocator to
> return frozen pages as described at [3]. If such a change is made, it
> can be employed by the hugetlb code. However, as mentioned above
> hugetlb uses several low level allocators so each would need to be
> modified to return frozen pages. For now, we can manually freeze the
> returned pages. This is done in two places:
> 1) alloc_buddy_huge_page, only the returned head page is ref counted.
> We freeze the head page, retrying once in the VERY rare case where
> there may be an inflated ref count.
> 2) prep_compound_gigantic_page, for gigantic pages the current code
> freezes all pages except the head page. New code will simply freeze
> the head page as well.
>
> In a few other places, code checks for inflated ref counts on newly
> allocated hugetlb pages. With the modifications to freeze after
> allocating, this code can be removed.
>
> After hugetlb pages are freshly allocated, they are often added to the
> hugetlb free lists. Since these pages were previously ref counted, this
> was done via put_page() which would end up calling the hugetlb
> destructor: free_huge_page. With changes to freeze pages, we simply
> call free_huge_page directly to add the pages to the free list.
>
> In a few other places, freshly allocated hugetlb pages were immediately
> put into use, and the expectation was they were already ref counted. In
> these cases, we must manually ref count the page.
>
> [1] https://lore.kernel.org/linux-mm/20210622021423.154662-3-mike.kravetz@oracle.com/
> [2] https://lore.kernel.org/linux-mm/20220802180309.19340-1-joao.m.martins@oracle.com/
> [3] https://lore.kernel.org/linux-mm/20220531150611.1303156-1-willy@infradead.org/
>
> Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
> ---
> mm/hugetlb.c | 97 +++++++++++++++++++---------------------------------
> 1 file changed, 35 insertions(+), 62 deletions(-)
>
> diff --git a/mm/hugetlb.c b/mm/hugetlb.c
> index 28516881a1b2..6b90d85d545b 100644
> --- a/mm/hugetlb.c
> +++ b/mm/hugetlb.c
> @@ -1769,13 +1769,12 @@ static bool __prep_compound_gigantic_page(struct page *page, unsigned int order,
> {
> int i, j;
> int nr_pages = 1 << order;
> - struct page *p = page + 1;
> + struct page *p = page;
>
> /* we rely on prep_new_huge_page to set the destructor */
> set_compound_order(page, order);
> - __ClearPageReserved(page);
> __SetPageHead(page);
> - for (i = 1; i < nr_pages; i++, p = mem_map_next(p, page, i)) {
> + for (i = 0; i < nr_pages; i++, p = mem_map_next(p, page, i)) {
> /*
> * For gigantic hugepages allocated through bootmem at
> * boot, it's safer to be consistent with the not-gigantic
> @@ -1814,7 +1813,8 @@ static bool __prep_compound_gigantic_page(struct page *page, unsigned int order,
> } else {
> VM_BUG_ON_PAGE(page_count(p), p);
> }
> - set_compound_head(p, page);
> + if (i != 0)
> + set_compound_head(p, page);

It seems we forget to unfreeze the head page in out_error path. If unexpected inflated ref count occurs,
the ref count of head page will become negative in free_gigantic_page?

Thanks for your patch, Mike. I hope this will help solve the races with memory failure. ;) And I will take
a more close review when I have enough time.

\
 
 \ /
  Last update: 2022-08-09 04:50    [W:0.176 / U:0.320 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site