Subject: [merged] mm-hugetlb-initialize-pg_reserved-for-tail-pages-of-gigantig-compound-pages.patch removed from -mm tree To: aarcange@xxxxxxxxxx,ajs124.ajs124@xxxxxxxxx,aquini@xxxxxxxxxx,gleb@xxxxxxxxxx,hughd@xxxxxxxxxx,mgorman@xxxxxxx,riel@xxxxxxxxxx,mm-commits@xxxxxxxxxxxxxxx From: akpm@xxxxxxxxxxxxxxxxxxxx Date: Thu, 17 Oct 2013 11:13:11 -0700 The patch titled Subject: mm: hugetlb: initialize PG_reserved for tail pages of gigantic compound pages has been removed from the -mm tree. Its filename was mm-hugetlb-initialize-pg_reserved-for-tail-pages-of-gigantig-compound-pages.patch This patch was dropped because it was merged into mainline or a subsystem tree ------------------------------------------------------ From: Andrea Arcangeli <aarcange@xxxxxxxxxx> Subject: mm: hugetlb: initialize PG_reserved for tail pages of gigantic compound pages 11feeb498086 ("kvm: optimize away THP checks in kvm_is_mmio_pfn()") introduced a memory leak when KVM is run on gigantic compound pages. 11feeb498086 depends on the assumption that PG_reserved is identical for all head and tail pages of a compound page. So that if get_user_pages returns a tail page, we don't need to check the head page in order to know if we deal with a reserved page that requires different refcounting. The assumption that PG_reserved is the same for head and tail pages is certainly correct for THP and regular hugepages, but gigantic hugepages allocated through bootmem don't clear the PG_reserved on the tail pages (the clearing of PG_reserved is done later only if the gigantic hugepage is freed). This patch corrects the gigantic compound page initialization so that we can retain the optimization in 11feeb498086. The cacheline was already modified in order to set PG_tail so this won't affect the boot time of large memory systems. [akpm@xxxxxxxxxxxxxxxxxxxx: tweak comment layout and grammar] Signed-off-by: Andrea Arcangeli <aarcange@xxxxxxxxxx> Reported-by: andy123 <ajs124.ajs124@xxxxxxxxx> Acked-by: Rik van Riel <riel@xxxxxxxxxx> Cc: Gleb Natapov <gleb@xxxxxxxxxx> Cc: Mel Gorman <mgorman@xxxxxxx> Cc: Hugh Dickins <hughd@xxxxxxxxxx> Acked-by: Rafael Aquini <aquini@xxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- mm/hugetlb.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff -puN mm/hugetlb.c~mm-hugetlb-initialize-pg_reserved-for-tail-pages-of-gigantig-compound-pages mm/hugetlb.c --- a/mm/hugetlb.c~mm-hugetlb-initialize-pg_reserved-for-tail-pages-of-gigantig-compound-pages +++ a/mm/hugetlb.c @@ -696,8 +696,22 @@ static void prep_compound_gigantic_page( /* we rely on prep_new_huge_page to set the destructor */ set_compound_order(page, order); __SetPageHead(page); + __ClearPageReserved(page); for (i = 1; i < nr_pages; i++, p = mem_map_next(p, page, i)) { __SetPageTail(p); + /* + * For gigantic hugepages allocated through bootmem at + * boot, it's safer to be consistent with the not-gigantic + * hugepages and clear the PG_reserved bit from all tail pages + * too. Otherwse drivers using get_user_pages() to access tail + * pages may get the reference counting wrong if they see + * PG_reserved set on a tail page (despite the head page not + * having PG_reserved set). Enforcing this consistency between + * head and tail pages allows drivers to optimize away a check + * on the head page when they need know if put_page() is needed + * after get_user_pages(). + */ + __ClearPageReserved(p); set_page_count(p, 0); p->first_page = page; } @@ -1330,9 +1344,9 @@ static void __init gather_bootmem_preall #else page = virt_to_page(m); #endif - __ClearPageReserved(page); WARN_ON(page_count(page) != 1); prep_compound_huge_page(page, h->order); + WARN_ON(PageReserved(page)); prep_new_huge_page(h, page, page_to_nid(page)); /* * If we had gigantic hugepages allocated at boot time, we need _ Patches currently in -mm which might be from aarcange@xxxxxxxxxx are origin.patch mm-thp-cleanup-mv-alloc_hugepage-to-better-place.patch mm-thp-khugepaged-add-policy-for-finding-target-node.patch mm-thp-khugepaged-add-policy-for-finding-target-node-fix.patch mm-avoid-increase-sizeofstruct-page-due-to-split-page-table-lock.patch mm-rename-use_split_ptlocks-to-use_split_pte_ptlocks.patch mm-convert-mm-nr_ptes-to-atomic_long_t.patch mm-introduce-api-for-split-page-table-lock-for-pmd-level.patch mm-thp-change-pmd_trans_huge_lock-to-return-taken-lock.patch mm-thp-move-ptl-taking-inside-page_check_address_pmd.patch mm-thp-do-not-access-mm-pmd_huge_pte-directly.patch mm-hugetlb-convert-hugetlbfs-to-use-split-pmd-lock.patch mm-convert-the-rest-to-new-page-table-lock-api.patch mm-implement-split-page-table-lock-for-pmd-level.patch x86-mm-enable-split-page-table-lock-for-pmd-level.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html