Subject: + mm-pagewalk-move-pmd_trans_huge_lock-from-callbacks-to-common-code.patch added to -mm tree
To: n-horiguchi@xxxxxxxxxxxxx,aneesh.kumar@xxxxxxxxxxxxxxxxxx,benh@xxxxxxxxxxxxxxxxxxx,dave.hansen@xxxxxxxxx,hannes@xxxxxxxxxxx,hughd@xxxxxxxxxx,kamezawa.hiroyu@xxxxxxxxxxxxxx,kirill@xxxxxxxxxxxxx,kosaki.motohiro@xxxxxxxxxxxxxx,mhocko@xxxxxxx,riel@xxxxxxxxxx,xemul@xxxxxxxxxxxxx
From: akpm@xxxxxxxxxxxxxxxxxxxx
Date: Wed, 11 Jun 2014 16:21:35 -0700
The patch titled
Subject: mm/pagewalk: move pmd_trans_huge_lock() from callbacks to common code
has been added to the -mm tree. Its filename is
mm-pagewalk-move-pmd_trans_huge_lock-from-callbacks-to-common-code.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/mm-pagewalk-move-pmd_trans_huge_lock-from-callbacks-to-common-code.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/mm-pagewalk-move-pmd_trans_huge_lock-from-callbacks-to-common-code.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: mm/pagewalk: move pmd_trans_huge_lock() from callbacks to common code
Now all of current users of page table walker are canonicalized, i.e.
pmd_entry() handles only trans_pmd entry, and pte_entry() handles pte entry.
So we can factorize common code more.
This patch moves pmd_trans_huge_lock() in each pmd_entry() to pagewalk core.
Signed-off-by: Naoya Horiguchi <n-horiguchi@xxxxxxxxxxxxx>
Cc: Dave Hansen <dave.hansen@xxxxxxxxx>
Cc: Hugh Dickins <hughd@xxxxxxxxxx>
Cc: "Kirill A. Shutemov" <kirill@xxxxxxxxxxxxx>
Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxx>
Cc: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@xxxxxxxxxxxxxxxxxx>
Cc: Pavel Emelyanov <xemul@xxxxxxxxxxxxx>
Cc: Rik van Riel <riel@xxxxxxxxxx>
Cc: Benjamin Herrenschmidt <benh@xxxxxxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
fs/proc/task_mmu.c | 67 ++++++++++++++-----------------------------
mm/memcontrol.c | 53 ++++++++++------------------------
mm/pagewalk.c | 25 +++++++++++++---
3 files changed, 60 insertions(+), 85 deletions(-)
diff -puN fs/proc/task_mmu.c~mm-pagewalk-move-pmd_trans_huge_lock-from-callbacks-to-common-code fs/proc/task_mmu.c
--- a/fs/proc/task_mmu.c~mm-pagewalk-move-pmd_trans_huge_lock-from-callbacks-to-common-code
+++ a/fs/proc/task_mmu.c
@@ -502,14 +502,8 @@ static int smaps_pmd(pmd_t *pmd, unsigne
struct mm_walk *walk)
{
struct mem_size_stats *mss = walk->private;
- spinlock_t *ptl;
-
- if (pmd_trans_huge_lock(pmd, walk->vma, &ptl) == 1) {
- smaps_pte((pte_t *)pmd, addr, addr + HPAGE_PMD_SIZE, walk);
- spin_unlock(ptl);
- mss->anonymous_thp += HPAGE_PMD_SIZE;
- } else
- walk->control = PTWALK_DOWN;
+ smaps_pte((pte_t *)pmd, addr, addr + HPAGE_PMD_SIZE, walk);
+ mss->anonymous_thp += HPAGE_PMD_SIZE;
return 0;
}
@@ -1043,31 +1037,22 @@ static int pagemap_pmd(pmd_t *pmd, unsig
struct vm_area_struct *vma = walk->vma;
struct pagemapread *pm = walk->private;
pagemap_entry_t pme = make_pme(PM_NOT_PRESENT(pm->v2));
- spinlock_t *ptl;
+ int pmd_flags2;
- if (!vma)
- return err;
- if (pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
- int pmd_flags2;
-
- if ((vma->vm_flags & VM_SOFTDIRTY) || pmd_soft_dirty(*pmd))
- pmd_flags2 = __PM_SOFT_DIRTY;
- else
- pmd_flags2 = 0;
-
- for (; addr != end; addr += PAGE_SIZE) {
- unsigned long offset;
-
- offset = (addr & ~PAGEMAP_WALK_MASK) >>
- PAGE_SHIFT;
- thp_pmd_to_pagemap_entry(&pme, pm, *pmd, offset, pmd_flags2);
- err = add_to_pagemap(addr, &pme, pm);
- if (err)
- break;
- }
- spin_unlock(ptl);
- } else
- walk->control = PTWALK_DOWN;
+ if ((vma->vm_flags & VM_SOFTDIRTY) || pmd_soft_dirty(*pmd))
+ pmd_flags2 = __PM_SOFT_DIRTY;
+ else
+ pmd_flags2 = 0;
+
+ for (; addr != end; addr += PAGE_SIZE) {
+ unsigned long offset;
+
+ offset = (addr & ~PAGEMAP_WALK_MASK) >> PAGE_SHIFT;
+ thp_pmd_to_pagemap_entry(&pme, pm, *pmd, offset, pmd_flags2);
+ err = add_to_pagemap(addr, &pme, pm);
+ if (err)
+ break;
+ }
return err;
}
@@ -1336,19 +1321,13 @@ static int gather_pmd_stats(pmd_t *pmd,
{
struct numa_maps *md = walk->private;
struct vm_area_struct *vma = walk->vma;
- spinlock_t *ptl;
+ pte_t huge_pte = *(pte_t *)pmd;
+ struct page *page;
- if (pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
- pte_t huge_pte = *(pte_t *)pmd;
- struct page *page;
-
- page = can_gather_numa_stats(huge_pte, vma, addr);
- if (page)
- gather_stats(page, md, pte_dirty(huge_pte),
- HPAGE_PMD_SIZE/PAGE_SIZE);
- spin_unlock(ptl);
- } else
- walk->control = PTWALK_DOWN;
+ page = can_gather_numa_stats(huge_pte, vma, addr);
+ if (page)
+ gather_stats(page, md, pte_dirty(huge_pte),
+ HPAGE_PMD_SIZE/PAGE_SIZE);
return 0;
}
#ifdef CONFIG_HUGETLB_PAGE
diff -puN mm/memcontrol.c~mm-pagewalk-move-pmd_trans_huge_lock-from-callbacks-to-common-code mm/memcontrol.c
--- a/mm/memcontrol.c~mm-pagewalk-move-pmd_trans_huge_lock-from-callbacks-to-common-code
+++ a/mm/memcontrol.c
@@ -6664,14 +6664,9 @@ static int mem_cgroup_count_precharge_pm
struct mm_walk *walk)
{
struct vm_area_struct *vma = walk->vma;
- spinlock_t *ptl;
- if (pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
- if (get_mctgt_type_thp(vma, addr, *pmd, NULL) == MC_TARGET_PAGE)
- mc.precharge += HPAGE_PMD_NR;
- spin_unlock(ptl);
- } else
- skip->control = PTWALK_DOWN;
+ if (get_mctgt_type_thp(vma, addr, *pmd, NULL) == MC_TARGET_PAGE)
+ mc.precharge += HPAGE_PMD_NR;
return 0;
}
@@ -6892,38 +6887,22 @@ static int mem_cgroup_move_charge_pmd(pm
struct page *page;
struct page_cgroup *pc;
- /*
- * We don't take compound_lock() here but no race with splitting thp
- * happens because:
- * - if pmd_trans_huge_lock() returns 1, the relevant thp is not
- * under splitting, which means there's no concurrent thp split,
- * - if another thread runs into split_huge_page() just after we
- * entered this if-block, the thread must wait for page table lock
- * to be unlocked in __split_huge_page_splitting(), where the main
- * part of thp split is not executed yet.
- */
- if (pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
- if (mc.precharge < HPAGE_PMD_NR) {
- spin_unlock(ptl);
- return 0;
- }
- target_type = get_mctgt_type_thp(vma, addr, *pmd, &target);
- if (target_type == MC_TARGET_PAGE) {
- page = target.page;
- if (!isolate_lru_page(page)) {
- pc = lookup_page_cgroup(page);
- if (!mem_cgroup_move_account(page, HPAGE_PMD_NR,
- pc, mc.from, mc.to)) {
- mc.precharge -= HPAGE_PMD_NR;
- mc.moved_charge += HPAGE_PMD_NR;
- }
- putback_lru_page(page);
+ if (mc.precharge < HPAGE_PMD_NR)
+ return 0;
+ target_type = get_mctgt_type_thp(vma, addr, *pmd, &target);
+ if (target_type == MC_TARGET_PAGE) {
+ page = target.page;
+ if (!isolate_lru_page(page)) {
+ pc = lookup_page_cgroup(page);
+ if (!mem_cgroup_move_account(page, HPAGE_PMD_NR,
+ pc, mc.from, mc.to)) {
+ mc.precharge -= HPAGE_PMD_NR;
+ mc.moved_charge += HPAGE_PMD_NR;
}
- put_page(page);
+ putback_lru_page(page);
}
- spin_unlock(ptl);
- } else
- walk->control = PTWALK_DOWN;
+ put_page(page);
+ }
return 0;
}
diff -puN mm/pagewalk.c~mm-pagewalk-move-pmd_trans_huge_lock-from-callbacks-to-common-code mm/pagewalk.c
--- a/mm/pagewalk.c~mm-pagewalk-move-pmd_trans_huge_lock-from-callbacks-to-common-code
+++ a/mm/pagewalk.c
@@ -61,6 +61,7 @@ static int walk_pmd_range(pud_t *pud, un
pmd_t *pmd;
unsigned long next;
int err = 0;
+ spinlock_t *ptl;
pmd = pmd_offset(pud, addr);
do {
@@ -75,8 +76,22 @@ again:
continue;
}
+ /*
+ * We don't take compound_lock() here but no race with splitting
+ * thp happens because:
+ * - if pmd_trans_huge_lock() returns 1, the relevant thp is
+ * not under splitting, which means there's no concurrent
+ * thp split,
+ * - if another thread runs into split_huge_page() just after
+ * we entered this if-block, the thread must wait for page
+ * table lock to be unlocked in __split_huge_page_splitting(),
+ * where the main part of thp split is not executed yet.
+ */
if (walk->pmd_entry) {
- err = walk->pmd_entry(pmd, addr, next, walk);
+ if (pmd_trans_huge_lock(pmd, walk->vma, &ptl) == 1) {
+ err = walk->pmd_entry(pmd, addr, next, walk);
+ spin_unlock(ptl);
+ }
if (err)
break;
switch (get_reset_walk_control(walk)) {
@@ -286,9 +301,11 @@ static int __walk_page_range(unsigned lo
* outside a vma. If you want to access to some caller-specific data from
* callbacks, @walk->private should be helpful.
*
- * The callers should hold @walk->mm->mmap_sem. Note that the lower level
- * iterators can take page table lock in lowest level iteration and/or
- * in split_huge_page_pmd().
+ * Locking:
+ * Callers of walk_page_range() and walk_page_vma() should hold
+ * @walk->mm->mmap_sem, because these function traverse vma list and/or
+ * access to vma's data. And page table lock is held during running
+ * pmd_entry() and pte_entry().
*/
int walk_page_range(unsigned long start, unsigned long end,
struct mm_walk *walk)
_
Patches currently in -mm which might be from n-horiguchi@xxxxxxxxxxxxx are
origin.patch
hwpoison-fix-the-handling-path-of-the-victimized-page-frame-that-belong-to-non-lur.patch
mm-hugetlbfs-fix-rmapping-for-anonymous-hugepages-with-page_pgoff.patch
mm-hugetlbfs-fix-rmapping-for-anonymous-hugepages-with-page_pgoff-v2.patch
mm-hugetlbfs-fix-rmapping-for-anonymous-hugepages-with-page_pgoff-v3.patch
mm-hugetlbfs-fix-rmapping-for-anonymous-hugepages-with-page_pgoff-v3-fix.patch
pagewalk-update-page-table-walker-core.patch
pagewalk-update-page-table-walker-core-fix-end-address-calculation-in-walk_page_range.patch
pagewalk-update-page-table-walker-core-fix-end-address-calculation-in-walk_page_range-fix.patch
pagewalk-update-page-table-walker-core-fix.patch
pagewalk-add-walk_page_vma.patch
smaps-redefine-callback-functions-for-page-table-walker.patch
clear_refs-redefine-callback-functions-for-page-table-walker.patch
pagemap-redefine-callback-functions-for-page-table-walker.patch
pagemap-redefine-callback-functions-for-page-table-walker-fix.patch
numa_maps-redefine-callback-functions-for-page-table-walker.patch
memcg-redefine-callback-functions-for-page-table-walker.patch
arch-powerpc-mm-subpage-protc-use-walk_page_vma-instead-of-walk_page_range.patch
pagewalk-remove-argument-hmask-from-hugetlb_entry.patch
pagewalk-remove-argument-hmask-from-hugetlb_entry-fix.patch
pagewalk-remove-argument-hmask-from-hugetlb_entry-fix-fix.patch
mempolicy-apply-page-table-walker-on-queue_pages_range.patch
mm-pagewalkc-move-pte-null-check.patch
mm-prom-pid-clear_refs-avoid-split_huge_page.patch
mm-pagewalk-remove-pgd_entry-and-pud_entry.patch
mm-pagewalk-replace-mm_walk-skip-with-more-general-mm_walk-control.patch
madvise-cleanup-swapin_walk_pmd_entry.patch
memcg-separate-mem_cgroup_move_charge_pte_range.patch
arch-powerpc-mm-subpage-protc-cleanup-subpage_walk_pmd_entry.patch
mm-pagewalk-move-pmd_trans_huge_lock-from-callbacks-to-common-code.patch
mincore-apply-page-table-walker-on-do_mincore.patch
mm-introduce-do_shared_fault-and-drop-do_fault-fix-fix.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
