pagemap: fix pfn calculation for hugepage

When we look into pagemap using page-types with option -p, the value of
pfn for hugepages looks wrong (see below.) This is because pte was
evaluated only once for one vma although it should be updated for each
hugepage.  This patch fixes it.

  $ page-types -p 3277 -Nl -b huge
  voffset   offset  len     flags
  7f21e8a00 11e400  1       ___U___________H_G________________
  7f21e8a01 11e401  1ff     ________________TG________________
               ^^^
  7f21e8c00 11e400  1       ___U___________H_G________________
  7f21e8c01 11e401  1ff     ________________TG________________
               ^^^

One hugepage contains 1 head page and 511 tail pages in x86_64 and each
two lines represent each hugepage.  Voffset and offset mean virtual
address and physical address in the page unit, respectively.  The
different hugepages should not have the same offset value.

With this patch applied:

  $ page-types -p 3386 -Nl -b huge
  voffset   offset   len    flags
  7fec7a600 112c00   1      ___UD__________H_G________________
  7fec7a601 112c01   1ff    ________________TG________________
               ^^^
  7fec7a800 113200   1      ___UD__________H_G________________
  7fec7a801 113201   1ff    ________________TG________________
               ^^^
               OK

More info:

- This patch modifies walk_page_range()'s hugepage walker.  But the
  change only affects pagemap_read(), which is the only caller of hugepage
  callback.

- Without this patch, hugetlb_entry() callback is called per vma, that
  doesn't match the natural expectation from its name.

- With this patch, hugetlb_entry() is called per hugepte entry and the
  callback can become much simpler.

Signed-off-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Matt Mackall <mpm@selenic.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index a05a669510a44c1d52540a520e7dc809b610cb95..070553427dd570176253d83b99fe91d7a6ec0c89 100644 (file)
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -662,31 +662,18 @@ static u64 huge_pte_to_pagemap_entry(pte_t pte, int offset)
        return pme;
 }
 
-static int pagemap_hugetlb_range(pte_t *pte, unsigned long addr,
-                                unsigned long end, struct mm_walk *walk)
+/* This function walks within one hugetlb entry in the single call */
+static int pagemap_hugetlb_range(pte_t *pte, unsigned long hmask,
+                                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;
+       u64 pfn;
 
-       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);
-               }
+               int offset = (addr & ~hmask) >> PAGE_SHIFT;
+               pfn = huge_pte_to_pagemap_entry(*pte, offset);
                err = add_to_pagemap(addr, pfn, pm);
                if (err)
                        return err;

diff --git a/include/linux/mm.h b/include/linux/mm.h
index e70f21beb4b4db9ec2944ca2bf0a9768e42af2b5..462acaf36f3a3959cca19e2f3c2e66cc01fe5379 100644 (file)
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -783,8 +783,8 @@ 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 *);
+       int (*hugetlb_entry)(pte_t *, unsigned long,
+                            unsigned long, unsigned long, struct mm_walk *);
        struct mm_struct *mm;
        void *private;
 };

diff --git a/mm/pagewalk.c b/mm/pagewalk.c
index 7b47a57b66462dae3d9ae42929a1bba543d2c290..8b1a2ce21ee53f3a6ea598759faf02c21c4c5dd2 100644 (file)
--- a/mm/pagewalk.c
+++ b/mm/pagewalk.c
@@ -80,6 +80,37 @@ static int walk_pud_range(pgd_t *pgd, unsigned long addr, unsigned long end,
        return err;
 }
 
+#ifdef CONFIG_HUGETLB_PAGE
+static unsigned long hugetlb_entry_end(struct hstate *h, unsigned long addr,
+                                      unsigned long end)
+{
+       unsigned long boundary = (addr & huge_page_mask(h)) + huge_page_size(h);
+       return boundary < end ? boundary : end;
+}
+
+static int walk_hugetlb_range(struct vm_area_struct *vma,
+                             unsigned long addr, unsigned long end,
+                             struct mm_walk *walk)
+{
+       struct hstate *h = hstate_vma(vma);
+       unsigned long next;
+       unsigned long hmask = huge_page_mask(h);
+       pte_t *pte;
+       int err = 0;
+
+       do {
+               next = hugetlb_entry_end(h, addr, end);
+               pte = huge_pte_offset(walk->mm, addr & hmask);
+               if (pte && walk->hugetlb_entry)
+                       err = walk->hugetlb_entry(pte, hmask, addr, next, walk);
+               if (err)
+                       return err;
+       } while (addr = next, addr != end);
+
+       return 0;
+}
+#endif
+
 /**
  * walk_page_range - walk a memory map's page tables with a callback
  * @mm: memory map to walk
@@ -128,20 +159,16 @@ int walk_page_range(unsigned long addr, unsigned long end,
                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);
+                       /*
+                        * Hugepage is very tightly coupled with vma, so
+                        * walk through hugetlb entries within a given vma.
+                        */
+                       err = walk_hugetlb_range(vma, addr, next, walk);
                        if (err)
                                break;
+                       pgd = pgd_offset(walk->mm, next);
                        continue;
                }
 #endif