The patch titled mm hugetlb: add hugepage support to pagemap has been removed from the -mm tree. Its filename was mm-hugetlb-add-hugepage-support-to-pagemap.patch This patch was dropped because it was merged into mainline or a subsystem tree The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: mm hugetlb: add hugepage support to pagemap From: Naoya Horiguchi <n-horiguchi@xxxxxxxxxxxxx> This patch enables extraction of the pfn of a hugepage from /proc/pid/pagemap in an architecture independent manner. Details ------- My test program (leak_pagemap) works as follows: - creat() and mmap() a file on hugetlbfs (file size is 200MB == 100 hugepages,) - read()/write() something on it, - call page-types with option -p, - munmap() and unlink() the file on hugetlbfs Without my patches ------------------ $ ./leak_pagemap flags page-count MB symbolic-flags long-symbolic-flags 0x0000000000000000 1 0 __________________________________ 0x0000000000000804 1 0 __R________M______________________ referenced,mmap 0x000000000000086c 81 0 __RU_lA____M______________________ referenced,uptodate,lru,active,mmap 0x0000000000005808 5 0 ___U_______Ma_b___________________ uptodate,mmap,anonymous,swapbacked 0x0000000000005868 12 0 ___U_lA____Ma_b___________________ uptodate,lru,active,mmap,anonymous,swapbacked 0x000000000000586c 1 0 __RU_lA____Ma_b___________________ referenced,uptodate,lru,active,mmap,anonymous,swapbacked total 101 0 The output of page-types don't show any hugepage. With my patches --------------- $ ./leak_pagemap flags page-count MB symbolic-flags long-symbolic-flags 0x0000000000000000 1 0 __________________________________ 0x0000000000030000 51100 199 ________________TG________________ compound_tail,huge 0x0000000000028018 100 0 ___UD__________H_G________________ uptodate,dirty,compound_head,huge 0x0000000000000804 1 0 __R________M______________________ referenced,mmap 0x000000000000080c 1 0 __RU_______M______________________ referenced,uptodate,mmap 0x000000000000086c 80 0 __RU_lA____M______________________ referenced,uptodate,lru,active,mmap 0x0000000000005808 4 0 ___U_______Ma_b___________________ uptodate,mmap,anonymous,swapbacked 0x0000000000005868 12 0 ___U_lA____Ma_b___________________ uptodate,lru,active,mmap,anonymous,swapbacked 0x000000000000586c 1 0 __RU_lA____Ma_b___________________ referenced,uptodate,lru,active,mmap,anonymous,swapbacked total 51300 200 The output of page-types shows 51200 pages contributing to hugepages, containing 100 head pages and 51100 tail pages as expected. [akpm@xxxxxxxxxxxxxxxxxxxx: build fix] Signed-off-by: Naoya Horiguchi <n-horiguchi@xxxxxxxxxxxxx> Cc: Andi Kleen <ak@xxxxxxxxxxxxxxx> Cc: Wu Fengguang <fengguang.wu@xxxxxxxxx> Cc: Hugh Dickins <hugh.dickins@xxxxxxxxxxxxx> Cc: Mel Gorman <mel@xxxxxxxxx> Cc: Lee Schermerhorn <lee.schermerhorn@xxxxxx> Cc: Andy Whitcroft <apw@xxxxxxxxxxxxx> Cc: David Rientjes <rientjes@xxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- fs/proc/task_mmu.c | 45 +++++++++++++++++++++++++++++++++++++++++++ include/linux/mm.h | 3 ++ mm/pagewalk.c | 22 ++++++++++++++++++--- 3 files changed, 67 insertions(+), 3 deletions(-) diff -puN fs/proc/task_mmu.c~mm-hugetlb-add-hugepage-support-to-pagemap fs/proc/task_mmu.c --- a/fs/proc/task_mmu.c~mm-hugetlb-add-hugepage-support-to-pagemap +++ a/fs/proc/task_mmu.c @@ -650,6 +650,50 @@ static int pagemap_pte_range(pmd_t *pmd, return err; } +static u64 huge_pte_to_pagemap_entry(pte_t pte, int offset) +{ + u64 pme = 0; + if (pte_present(pte)) + pme = PM_PFRAME(pte_pfn(pte) + offset) + | PM_PSHIFT(PAGE_SHIFT) | PM_PRESENT; + return pme; +} + +static int pagemap_hugetlb_range(pte_t *pte, unsigned long addr, + unsigned long end, struct mm_walk *walk) +{ + struct vm_area_struct *vma; + struct pagemapread *pm = walk->private; + struct hstate *hs = NULL; + int err = 0; + + vma = find_vma(walk->mm, addr); + if (vma) + hs = hstate_vma(vma); + for (; addr != end; addr += PAGE_SIZE) { + u64 pfn = PM_NOT_PRESENT; + + if (vma && (addr >= vma->vm_end)) { + vma = find_vma(walk->mm, addr); + if (vma) + hs = hstate_vma(vma); + } + + if (vma && (vma->vm_start <= addr) && is_vm_hugetlb_page(vma)) { + /* calculate pfn of the "raw" page in the hugepage. */ + int offset = (addr & ~huge_page_mask(hs)) >> PAGE_SHIFT; + pfn = huge_pte_to_pagemap_entry(*pte, offset); + } + err = add_to_pagemap(addr, pfn, pm); + if (err) + return err; + } + + cond_resched(); + + return err; +} + /* * /proc/pid/pagemap - an array mapping virtual pages to pfns * @@ -742,6 +786,7 @@ static ssize_t pagemap_read(struct file pagemap_walk.pmd_entry = pagemap_pte_range; pagemap_walk.pte_hole = pagemap_pte_hole; + pagemap_walk.hugetlb_entry = pagemap_hugetlb_range; pagemap_walk.mm = mm; pagemap_walk.private = ± diff -puN include/linux/mm.h~mm-hugetlb-add-hugepage-support-to-pagemap include/linux/mm.h --- a/include/linux/mm.h~mm-hugetlb-add-hugepage-support-to-pagemap +++ a/include/linux/mm.h @@ -770,6 +770,7 @@ unsigned long unmap_vmas(struct mmu_gath * @pmd_entry: if set, called for each non-empty PMD (3rd-level) entry * @pte_entry: if set, called for each non-empty PTE (4th-level) entry * @pte_hole: if set, called for each hole at all levels + * @hugetlb_entry: if set, called for each hugetlb entry * * (see walk_page_range for more details) */ @@ -779,6 +780,8 @@ struct mm_walk { int (*pmd_entry)(pmd_t *, unsigned long, unsigned long, struct mm_walk *); int (*pte_entry)(pte_t *, unsigned long, unsigned long, struct mm_walk *); int (*pte_hole)(unsigned long, unsigned long, struct mm_walk *); + int (*hugetlb_entry)(pte_t *, unsigned long, unsigned long, + struct mm_walk *); struct mm_struct *mm; void *private; }; diff -puN mm/pagewalk.c~mm-hugetlb-add-hugepage-support-to-pagemap mm/pagewalk.c --- a/mm/pagewalk.c~mm-hugetlb-add-hugepage-support-to-pagemap +++ a/mm/pagewalk.c @@ -120,15 +120,31 @@ int walk_page_range(unsigned long addr, do { next = pgd_addr_end(addr, end); - /* skip hugetlb vma to avoid hugepage PMD being cleared - * in pmd_none_or_clear_bad(). */ + /* + * handle hugetlb vma individually because pagetable walk for + * the hugetlb page is dependent on the architecture and + * we can't handled it in the same manner as non-huge pages. + */ vma = find_vma(walk->mm, addr); +#ifdef CONFIG_HUGETLB_PAGE if (vma && is_vm_hugetlb_page(vma)) { + pte_t *pte; + struct hstate *hs; + if (vma->vm_end < next) next = vma->vm_end; + hs = hstate_vma(vma); + pte = huge_pte_offset(walk->mm, + addr & huge_page_mask(hs)); + if (pte && !huge_pte_none(huge_ptep_get(pte)) + && walk->hugetlb_entry) + err = walk->hugetlb_entry(pte, addr, + next, walk); + if (err) + break; continue; } - +#endif if (pgd_none_or_clear_bad(pgd)) { if (walk->pte_hole) err = walk->pte_hole(addr, next, walk); _ Patches currently in -mm which might be from n-horiguchi@xxxxxxxxxxxxx are origin.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