mm: save soft-dirty bits on file pages

Andy reported that if file page get reclaimed we lose the soft-dirty bit
if it was there, so save _PAGE_BIT_SOFT_DIRTY bit when page address get
encoded into pte entry.  Thus when #pf happens on such non-present pte
we can restore it back.

Reported-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Acked-by: Pavel Emelyanov <xemul@parallels.com>
Cc: Matt Mackall <mpm@selenic.com>
Cc: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@gmail.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

9 files changed, 107 insertions, 10 deletions

diff --git a/arch/x86/include/asm/pgtable-2level.h b/arch/x86/include/asm/pgtable-2level.h
index f2b489cf1602..3bf2dd0cf61f 100644
--- a/arch/x86/include/asm/pgtable-2level.h
+++ b/arch/x86/include/asm/pgtable-2level.h
@@ -55,9 +55,53 @@ static inline pmd_t native_pmdp_get_and_clear(pmd_t *xp)
 #define native_pmdp_get_and_clear(xp) native_local_pmdp_get_and_clear(xp)
 #endif
 
+#ifdef CONFIG_MEM_SOFT_DIRTY
+
+/*
+ * Bits _PAGE_BIT_PRESENT, _PAGE_BIT_FILE, _PAGE_BIT_SOFT_DIRTY and
+ * _PAGE_BIT_PROTNONE are taken, split up the 28 bits of offset
+ * into this range.
+ */
+#define PTE_FILE_MAX_BITS       28
+#define PTE_FILE_SHIFT1         (_PAGE_BIT_PRESENT + 1)
+#define PTE_FILE_SHIFT2         (_PAGE_BIT_FILE + 1)
+#define PTE_FILE_SHIFT3         (_PAGE_BIT_PROTNONE + 1)
+#define PTE_FILE_SHIFT4         (_PAGE_BIT_SOFT_DIRTY + 1)
+#define PTE_FILE_BITS1          (PTE_FILE_SHIFT2 - PTE_FILE_SHIFT1 - 1)
+#define PTE_FILE_BITS2          (PTE_FILE_SHIFT3 - PTE_FILE_SHIFT2 - 1)
+#define PTE_FILE_BITS3          (PTE_FILE_SHIFT4 - PTE_FILE_SHIFT3 - 1)
+
+#define pte_to_pgoff(pte)                                               \
+        ((((pte).pte_low >> (PTE_FILE_SHIFT1))                          \
+          & ((1U << PTE_FILE_BITS1) - 1)))                              \
+        + ((((pte).pte_low >> (PTE_FILE_SHIFT2))                        \
+            & ((1U << PTE_FILE_BITS2) - 1))                             \
+           << (PTE_FILE_BITS1))                                         \
+        + ((((pte).pte_low >> (PTE_FILE_SHIFT3))                        \
+            & ((1U << PTE_FILE_BITS3) - 1))                             \
+           << (PTE_FILE_BITS1 + PTE_FILE_BITS2))                        \
+        + ((((pte).pte_low >> (PTE_FILE_SHIFT4)))                       \
+            << (PTE_FILE_BITS1 + PTE_FILE_BITS2 + PTE_FILE_BITS3))
+
+#define pgoff_to_pte(off)                                               \
+        ((pte_t) { .pte_low =                                           \
+         ((((off)) & ((1U << PTE_FILE_BITS1) - 1)) << PTE_FILE_SHIFT1)  \
+         + ((((off) >> PTE_FILE_BITS1)                                  \
+             & ((1U << PTE_FILE_BITS2) - 1))                            \
+            << PTE_FILE_SHIFT2)                                         \
+         + ((((off) >> (PTE_FILE_BITS1 + PTE_FILE_BITS2))               \
+             & ((1U << PTE_FILE_BITS3) - 1))                            \
+            << PTE_FILE_SHIFT3)                                         \
+         + ((((off) >>                                                  \
+              (PTE_FILE_BITS1 + PTE_FILE_BITS2 + PTE_FILE_BITS3)))      \
+            << PTE_FILE_SHIFT4)                                         \
+         + _PAGE_FILE })
+
+#else /* CONFIG_MEM_SOFT_DIRTY */
+
 /*
  * Bits _PAGE_BIT_PRESENT, _PAGE_BIT_FILE and _PAGE_BIT_PROTNONE are taken,
- * split up the 29 bits of offset into this range:
+ * split up the 29 bits of offset into this range.
  */
 #define PTE_FILE_MAX_BITS       29
 #define PTE_FILE_SHIFT1         (_PAGE_BIT_PRESENT + 1)
@@ -88,6 +132,8 @@ static inline pmd_t native_pmdp_get_and_clear(pmd_t *xp)
             << PTE_FILE_SHIFT3)                                         \
          + _PAGE_FILE })
 
+#endif /* CONFIG_MEM_SOFT_DIRTY */
+
 /* Encode and de-code a swap entry */
 #if _PAGE_BIT_FILE < _PAGE_BIT_PROTNONE
 #define SWP_TYPE_BITS (_PAGE_BIT_FILE - _PAGE_BIT_PRESENT - 1)
diff --git a/arch/x86/include/asm/pgtable-3level.h b/arch/x86/include/asm/pgtable-3level.h
index 4cc9f2b7cdc3..81bb91b49a88 100644
--- a/arch/x86/include/asm/pgtable-3level.h
+++ b/arch/x86/include/asm/pgtable-3level.h
@@ -179,6 +179,9 @@ static inline pmd_t native_pmdp_get_and_clear(pmd_t *pmdp)
 /*
  * Bits 0, 6 and 7 are taken in the low part of the pte,
  * put the 32 bits of offset into the high part.
+ *
+ * For soft-dirty tracking 11 bit is taken from
+ * the low part of pte as well.
  */
 #define pte_to_pgoff(pte) ((pte).pte_high)
 #define pgoff_to_pte(off)                                               \
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index bd0518a7f197..1c00631164c2 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -329,6 +329,21 @@ static inline pte_t pte_swp_clear_soft_dirty(pte_t pte)
         return pte_clear_flags(pte, _PAGE_SWP_SOFT_DIRTY);
 }
 
+static inline pte_t pte_file_clear_soft_dirty(pte_t pte)
+{
+        return pte_clear_flags(pte, _PAGE_SOFT_DIRTY);
+}
+
+static inline pte_t pte_file_mksoft_dirty(pte_t pte)
+{
+        return pte_set_flags(pte, _PAGE_SOFT_DIRTY);
+}
+
+static inline int pte_file_soft_dirty(pte_t pte)
+{
+        return pte_flags(pte) & _PAGE_SOFT_DIRTY;
+}
+
 /*
  * Mask out unsupported bits in a present pgprot.  Non-present pgprots
  * can use those bits for other purposes, so leave them be.
diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h
index 5e8442f178f9..f4843e031131 100644
--- a/arch/x86/include/asm/pgtable_types.h
+++ b/arch/x86/include/asm/pgtable_types.h
@@ -61,8 +61,10 @@
  * they do not conflict with each other.
  */
 
+#define _PAGE_BIT_SOFT_DIRTY    _PAGE_BIT_HIDDEN
+
 #ifdef CONFIG_MEM_SOFT_DIRTY
-#define _PAGE_SOFT_DIRTY        (_AT(pteval_t, 1) << _PAGE_BIT_HIDDEN)
+#define _PAGE_SOFT_DIRTY        (_AT(pteval_t, 1) << _PAGE_BIT_SOFT_DIRTY)
 #else
 #define _PAGE_SOFT_DIRTY        (_AT(pteval_t, 0))
 #endif
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index e2d9bdce5e7e..a11720767abc 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -736,6 +736,8 @@ static inline void clear_soft_dirty(struct vm_area_struct *vma,
                 ptent = pte_clear_flags(ptent, _PAGE_SOFT_DIRTY);
         } else if (is_swap_pte(ptent)) {
                 ptent = pte_swp_clear_soft_dirty(ptent);
+        } else if (pte_file(ptent)) {
+                ptent = pte_file_clear_soft_dirty(ptent);
         }
 
         set_pte_at(vma->vm_mm, addr, pte, ptent);
diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index 2a7e0d10ad9a..0807ddf97b05 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -432,6 +432,21 @@ static inline pte_t pte_swp_clear_soft_dirty(pte_t pte)
 {
         return pte;
 }
+
+static inline pte_t pte_file_clear_soft_dirty(pte_t pte)
+{
+       return pte;
+}
+
+static inline pte_t pte_file_mksoft_dirty(pte_t pte)
+{
+       return pte;
+}
+
+static inline int pte_file_soft_dirty(pte_t pte)
+{
+       return 0;
+}
 #endif
 
 #ifndef __HAVE_PFNMAP_TRACKING
diff --git a/mm/fremap.c b/mm/fremap.c
index 87da3590c61e..5bff08147768 100644
--- a/mm/fremap.c
+++ b/mm/fremap.c
@@ -57,17 +57,22 @@ static int install_file_pte(struct mm_struct *mm, struct vm_area_struct *vma,