The patch titled
Subject: mincore: apply page table walker on do_mincore()
has been added to the -mm tree. Its filename is
mincore-apply-page-table-walker-on-do_mincore.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/mincore-apply-page-table-walker-on-do_mincore.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/mincore-apply-page-table-walker-on-do_mincore.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Naoya Horiguchi <n-horiguchi@xxxxxxxxxxxxx>
Subject: mincore: apply page table walker on do_mincore()
This patch makes do_mincore() use walk_page_vma(), which reduces many
lines of code by using common page table walk code.
Signed-off-by: Naoya Horiguchi <n-horiguchi@xxxxxxxxxxxxx>
Cc: "Kirill A. Shutemov" <kirill.shutemov@xxxxxxxxxxxxxxx>
Cc: Andrea Arcangeli <aarcange@xxxxxxxxxx>
Cc: Cyrill Gorcunov <gorcunov@xxxxxxxxxx>
Cc: Dave Hansen <dave.hansen@xxxxxxxxx>
Cc: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx>
Cc: Pavel Emelyanov <xemul@xxxxxxxxxxxxx>
Cc: Benjamin Herrenschmidt <benh@xxxxxxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/huge_memory.c | 20 -----
mm/mincore.c | 169 +++++++++++++++------------------------------
2 files changed, 59 insertions(+), 130 deletions(-)
diff -puN mm/huge_memory.c~mincore-apply-page-table-walker-on-do_mincore mm/huge_memory.c
--- a/mm/huge_memory.c~mincore-apply-page-table-walker-on-do_mincore
+++ a/mm/huge_memory.c
@@ -1415,26 +1415,6 @@ int zap_huge_pmd(struct mmu_gather *tlb,
return ret;
}
-int mincore_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
- unsigned long addr, unsigned long end,
- unsigned char *vec)
-{
- spinlock_t *ptl;
- int ret = 0;
-
- if (__pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
- /*
- * All logical pages in the range are present
- * if backed by a huge page.
- */
- spin_unlock(ptl);
- memset(vec, 1, (end - addr) >> PAGE_SHIFT);
- ret = 1;
- }
-
- return ret;
-}
-
int move_huge_pmd(struct vm_area_struct *vma, struct vm_area_struct *new_vma,
unsigned long old_addr,
unsigned long new_addr, unsigned long old_end,
diff -puN mm/mincore.c~mincore-apply-page-table-walker-on-do_mincore mm/mincore.c
--- a/mm/mincore.c~mincore-apply-page-table-walker-on-do_mincore
+++ a/mm/mincore.c
@@ -19,38 +19,26 @@
#include <asm/uaccess.h>
#include <asm/pgtable.h>
-static void mincore_hugetlb_page_range(struct vm_area_struct *vma,
- unsigned long addr, unsigned long end,
- unsigned char *vec)
+static int mincore_hugetlb(pte_t *pte, unsigned long hmask, unsigned long addr,
+ unsigned long end, struct mm_walk *walk)
{
+ int err = 0;
#ifdef CONFIG_HUGETLB_PAGE
- struct hstate *h;
+ unsigned char present;
+ unsigned char *vec = walk->private;
- h = hstate_vma(vma);
- while (1) {
- unsigned char present;
- pte_t *ptep;
- /*
- * Huge pages are always in RAM for now, but
- * theoretically it needs to be checked.
- */
- ptep = huge_pte_offset(current->mm,
- addr & huge_page_mask(h));
- present = ptep && !huge_pte_none(huge_ptep_get(ptep));
- while (1) {
- *vec = present;
- vec++;
- addr += PAGE_SIZE;
- if (addr == end)
- return;
- /* check hugepage border */
- if (!(addr & ~huge_page_mask(h)))
- break;
- }
- }
+ /*
+ * Hugepages under user process are always in RAM and never
+ * swapped out, but theoretically it needs to be checked.
+ */
+ present = pte && !huge_pte_none(huge_ptep_get(pte));
+ for (; addr != end; vec++, addr += PAGE_SIZE)
+ *vec = present;
+ walk->private = vec;
#else
BUG();
#endif
+ return err;
}
/*
@@ -94,14 +82,15 @@ static unsigned char mincore_page(struct
return present;
}
-static void mincore_unmapped_range(struct vm_area_struct *vma,
- unsigned long addr, unsigned long end,
- unsigned char *vec)
+static int mincore_unmapped_range(unsigned long addr, unsigned long end,
+ struct mm_walk *walk)
{
+ struct vm_area_struct *vma = walk->vma;
+ unsigned char *vec = walk->private;
unsigned long nr = (end - addr) >> PAGE_SHIFT;
int i;
- if (vma->vm_file) {
+ if (vma && vma->vm_file) {
pgoff_t pgoff;
pgoff = linear_page_index(vma, addr);
@@ -111,24 +100,39 @@ static void mincore_unmapped_range(struc
for (i = 0; i < nr; i++)
vec[i] = 0;
}
+ walk->private += nr;
+ return 0;
}
-static void mincore_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
- unsigned long addr, unsigned long end,
- unsigned char *vec)
+static int mincore_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
+ struct mm_walk *walk)
{
- unsigned long next;
spinlock_t *ptl;
+ struct vm_area_struct *vma = walk->vma;
pte_t *ptep;
- ptep = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
- do {
+ if (pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
+ memset(walk->private, 1, (end - addr) >> PAGE_SHIFT);
+ walk->private += (end - addr) >> PAGE_SHIFT;
+ spin_unlock(ptl);
+ return 0;
+ }
+
+ if (pmd_trans_unstable(pmd)) {
+ mincore_unmapped_range(addr, end, walk);
+ return 0;
+ }
+
+ ptep = pte_offset_map_lock(walk->mm, pmd, addr, &ptl);
+ for (; addr != end; ptep++, addr += PAGE_SIZE) {
pte_t pte = *ptep;
+ unsigned char *vec = walk->private;
- next = addr + PAGE_SIZE;
- if (pte_none(pte))
- mincore_unmapped_range(vma, addr, next, vec);
- else if (pte_present(pte))
+ if (pte_none(pte)) {
+ mincore_unmapped_range(addr, addr + PAGE_SIZE, walk);
+ continue;
+ }
+ if (pte_present(pte))
*vec = 1;
else { /* pte is a swap entry */
swp_entry_t entry = pte_to_swp_entry(pte);
@@ -149,70 +153,11 @@ static void mincore_pte_range(struct vm_
#endif
}
}
- vec++;
- } while (ptep++, addr = next, addr != end);
+ walk->private++;
+ }
pte_unmap_unlock(ptep - 1, ptl);
-}
-
-static void mincore_pmd_range(struct vm_area_struct *vma, pud_t *pud,
- unsigned long addr, unsigned long end,
- unsigned char *vec)
-{
- unsigned long next;
- pmd_t *pmd;
-
- pmd = pmd_offset(pud, addr);
- do {
- next = pmd_addr_end(addr, end);
- if (pmd_trans_huge(*pmd)) {
- if (mincore_huge_pmd(vma, pmd, addr, next, vec)) {
- vec += (next - addr) >> PAGE_SHIFT;
- continue;
- }
- /* fall through */
- }
- if (pmd_none_or_trans_huge_or_clear_bad(pmd))
- mincore_unmapped_range(vma, addr, next, vec);
- else
- mincore_pte_range(vma, pmd, addr, next, vec);
- vec += (next - addr) >> PAGE_SHIFT;
- } while (pmd++, addr = next, addr != end);
-}
-
-static void mincore_pud_range(struct vm_area_struct *vma, pgd_t *pgd,
- unsigned long addr, unsigned long end,
- unsigned char *vec)
-{
- unsigned long next;
- pud_t *pud;
-
- pud = pud_offset(pgd, addr);
- do {
- next = pud_addr_end(addr, end);
- if (pud_none_or_clear_bad(pud))
- mincore_unmapped_range(vma, addr, next, vec);
- else
- mincore_pmd_range(vma, pud, addr, next, vec);
- vec += (next - addr) >> PAGE_SHIFT;
- } while (pud++, addr = next, addr != end);
-}
-
-static void mincore_page_range(struct vm_area_struct *vma,
- unsigned long addr, unsigned long end,
- unsigned char *vec)
-{
- unsigned long next;
- pgd_t *pgd;
-
- pgd = pgd_offset(vma->vm_mm, addr);
- do {
- next = pgd_addr_end(addr, end);
- if (pgd_none_or_clear_bad(pgd))
- mincore_unmapped_range(vma, addr, next, vec);
- else
- mincore_pud_range(vma, pgd, addr, next, vec);
- vec += (next - addr) >> PAGE_SHIFT;
- } while (pgd++, addr = next, addr != end);
+ cond_resched();
+ return 0;
}
/*
@@ -224,18 +169,22 @@ static long do_mincore(unsigned long add
{
struct vm_area_struct *vma;
unsigned long end;
+ int err;
+ struct mm_walk mincore_walk = {
+ .pmd_entry = mincore_pte_range,
+ .pte_hole = mincore_unmapped_range,
+ .hugetlb_entry = mincore_hugetlb,
+ .private = vec,
+ };
vma = find_vma(current->mm, addr);
if (!vma || addr < vma->vm_start)
return -ENOMEM;
-
+ mincore_walk.mm = vma->vm_mm;
end = min(vma->vm_end, addr + (pages << PAGE_SHIFT));
-
- if (is_vm_hugetlb_page(vma))
- mincore_hugetlb_page_range(vma, addr, end, vec);
- else
- mincore_page_range(vma, addr, end, vec);
-
+ err = walk_page_range(addr, end, &mincore_walk);
+ if (err < 0)
+ return err;
return (end - addr) >> PAGE_SHIFT;
}
_
Patches currently in -mm which might be from n-horiguchi@xxxxxxxxxxxxx are
mm-pagewalk-call-pte_hole-for-vm_pfnmap-during-walk_page_range.patch
mm-add-kpf_zero_page-flag-for-proc-kpageflags.patch
mm-hugetlb-reduce-arch-dependent-code-around-follow_huge_.patch
mm-hugetlb-pmd_huge-returns-true-for-non-present-hugepage.patch
mm-hugetlb-take-page-table-lock-in-follow_huge_pmd.patch
mm-hugetlb-fix-getting-refcount-0-page-in-hugetlb_fault.patch
mm-hugetlb-add-migration-hwpoisoned-entry-check-in-hugetlb_change_protection.patch
mm-hugetlb-add-migration-entry-check-in-__unmap_hugepage_range.patch
mm-hugetlb-fix-suboptimal-migration-hwpoisoned-entry-check.patch
mm-hugetlb-cleanup-and-rename-is_hugetlb_entry_migrationhwpoisoned.patch
mm-pagewalk-remove-pgd_entry-and-pud_entry.patch
revert-mm-pagewalk-call-pte_hole-for-vm_pfnmap-during-walk_page_range.patch
pagewalk-improve-vma-handling.patch
pagewalk-add-walk_page_vma.patch
smaps-remove-mem_size_stats-vma-and-use-walk_page_vma.patch
clear_refs-remove-clear_refs_private-vma-and-introduce-clear_refs_test_walk.patch
pagemap-use-walk-vma-instead-of-calling-find_vma.patch
numa_maps-fix-typo-in-gather_hugetbl_stats.patch
numa_maps-remove-numa_maps-vma.patch
memcg-cleanup-preparation-for-page-table-walk.patch
arch-powerpc-mm-subpage-protc-use-walk-vma-and-walk_page_vma.patch
mempolicy-apply-page-table-walker-on-queue_pages_range.patch
mm-proc-pid-clear_refs-avoid-split_huge_page.patch
mincore-apply-page-table-walker-on-do_mincore.patch
do_shared_fault-check-that-mmap_sem-is-held.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
