* + mm-save-soft-dirty-bits-on-file-pages.patch added to -mm tree
@ 2013-08-07 20:28 akpm
0 siblings, 0 replies; only message in thread
From: akpm @ 2013-08-07 20:28 UTC (permalink / raw)
To: mm-commits, xiaoguangrong, xemul, sfr, peterz, mtosatti, mpm,
minchan, luto, liwanp, kosaki.motohiro, gorcunov, aneesh.kumar,
gorcunov
Subject: + mm-save-soft-dirty-bits-on-file-pages.patch added to -mm tree
To: gorcunov@gmail.com,aneesh.kumar@linux.vnet.ibm.com,gorcunov@openvz.org,kosaki.motohiro@gmail.com,liwanp@linux.vnet.ibm.com,luto@amacapital.net,minchan@kernel.org,mpm@selenic.com,mtosatti@redhat.com,peterz@infradead.org,sfr@canb.auug.org.au,xemul@parallels.com,xiaoguangrong@linux.vnet.ibm.com
From: akpm@linux-foundation.org
Date: Wed, 07 Aug 2013 13:28:24 -0700
The patch titled
Subject: mm: save soft-dirty bits on file pages
has been added to the -mm tree. Its filename is
mm-save-soft-dirty-bits-on-file-pages.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/mm-save-soft-dirty-bits-on-file-pages.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/mm-save-soft-dirty-bits-on-file-pages.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: Cyrill Gorcunov <gorcunov@gmail.com>
Subject: 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>
---
arch/x86/include/asm/pgtable-2level.h | 48 +++++++++++++++++++++++-
arch/x86/include/asm/pgtable-3level.h | 3 +
arch/x86/include/asm/pgtable.h | 15 +++++++
arch/x86/include/asm/pgtable_types.h | 4 +-
fs/proc/task_mmu.c | 2 +
include/asm-generic/pgtable.h | 15 +++++++
mm/fremap.c | 11 ++++-
mm/memory.c | 11 ++++-
mm/rmap.c | 8 +++-
9 files changed, 107 insertions(+), 10 deletions(-)
diff -puN arch/x86/include/asm/pgtable-2level.h~mm-save-soft-dirty-bits-on-file-pages arch/x86/include/asm/pgtable-2level.h
--- a/arch/x86/include/asm/pgtable-2level.h~mm-save-soft-dirty-bits-on-file-pages
+++ a/arch/x86/include/asm/pgtable-2level.h
@@ -55,9 +55,53 @@ static inline pmd_t native_pmdp_get_and_
#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_
<< 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 -puN arch/x86/include/asm/pgtable-3level.h~mm-save-soft-dirty-bits-on-file-pages arch/x86/include/asm/pgtable-3level.h
--- a/arch/x86/include/asm/pgtable-3level.h~mm-save-soft-dirty-bits-on-file-pages
+++ a/arch/x86/include/asm/pgtable-3level.h
@@ -179,6 +179,9 @@ static inline pmd_t native_pmdp_get_and_
/*
* 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 -puN arch/x86/include/asm/pgtable.h~mm-save-soft-dirty-bits-on-file-pages arch/x86/include/asm/pgtable.h
--- a/arch/x86/include/asm/pgtable.h~mm-save-soft-dirty-bits-on-file-pages
+++ a/arch/x86/include/asm/pgtable.h
@@ -329,6 +329,21 @@ static inline pte_t pte_swp_clear_soft_d
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 -puN arch/x86/include/asm/pgtable_types.h~mm-save-soft-dirty-bits-on-file-pages arch/x86/include/asm/pgtable_types.h
--- a/arch/x86/include/asm/pgtable_types.h~mm-save-soft-dirty-bits-on-file-pages
+++ a/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 -puN fs/proc/task_mmu.c~mm-save-soft-dirty-bits-on-file-pages fs/proc/task_mmu.c
--- a/fs/proc/task_mmu.c~mm-save-soft-dirty-bits-on-file-pages
+++ a/fs/proc/task_mmu.c
@@ -736,6 +736,8 @@ static inline void clear_soft_dirty(stru
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 -puN include/asm-generic/pgtable.h~mm-save-soft-dirty-bits-on-file-pages include/asm-generic/pgtable.h
--- a/include/asm-generic/pgtable.h~mm-save-soft-dirty-bits-on-file-pages
+++ a/include/asm-generic/pgtable.h
@@ -432,6 +432,21 @@ static inline pte_t pte_swp_clear_soft_d
{
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 -puN mm/fremap.c~mm-save-soft-dirty-bits-on-file-pages mm/fremap.c
--- a/mm/fremap.c~mm-save-soft-dirty-bits-on-file-pages
+++ a/mm/fremap.c
@@ -57,17 +57,22 @@ static int install_file_pte(struct mm_st
unsigned long addr, unsigned long pgoff, pgprot_t prot)
{
int err = -ENOMEM;
- pte_t *pte;
+ pte_t *pte, ptfile;
spinlock_t *ptl;
pte = get_locked_pte(mm, addr, &ptl);
if (!pte)
goto out;
- if (!pte_none(*pte))
+ ptfile = pgoff_to_pte(pgoff);
+
+ if (!pte_none(*pte)) {
+ if (pte_present(*pte) && pte_soft_dirty(*pte))
+ pte_file_mksoft_dirty(ptfile);
zap_pte(mm, vma, addr, pte);
+ }
- set_pte_at(mm, addr, pte, pgoff_to_pte(pgoff));
+ set_pte_at(mm, addr, pte, ptfile);
/*
* We don't need to run update_mmu_cache() here because the "file pte"
* being installed by install_file_pte() is not a real pte - it's a
diff -puN mm/memory.c~mm-save-soft-dirty-bits-on-file-pages mm/memory.c
--- a/mm/memory.c~mm-save-soft-dirty-bits-on-file-pages
+++ a/mm/memory.c
@@ -1141,9 +1141,12 @@ again:
continue;
if (unlikely(details) && details->nonlinear_vma
&& linear_page_index(details->nonlinear_vma,
- addr) != page->index)
- set_pte_at(mm, addr, pte,
- pgoff_to_pte(page->index));
+ addr) != page->index) {
+ pte_t ptfile = pgoff_to_pte(page->index);
+ if (pte_soft_dirty(ptent))
+ pte_file_mksoft_dirty(ptfile);
+ set_pte_at(mm, addr, pte, ptfile);
+ }
if (PageAnon(page))
rss[MM_ANONPAGES]--;
else {
@@ -3410,6 +3413,8 @@ static int __do_fault(struct mm_struct *
entry = mk_pte(page, vma->vm_page_prot);
if (flags & FAULT_FLAG_WRITE)
entry = maybe_mkwrite(pte_mkdirty(entry), vma);
+ else if (pte_file(orig_pte) && pte_file_soft_dirty(orig_pte))
+ pte_mksoft_dirty(entry);
if (anon) {
inc_mm_counter_fast(mm, MM_ANONPAGES);
page_add_new_anon_rmap(page, vma, address);
diff -puN mm/rmap.c~mm-save-soft-dirty-bits-on-file-pages mm/rmap.c
--- a/mm/rmap.c~mm-save-soft-dirty-bits-on-file-pages
+++ a/mm/rmap.c
@@ -1405,8 +1405,12 @@ static int try_to_unmap_cluster(unsigned
pteval = ptep_clear_flush(vma, address, pte);
/* If nonlinear, store the file page offset in the pte. */
- if (page->index != linear_page_index(vma, address))
- set_pte_at(mm, address, pte, pgoff_to_pte(page->index));
+ if (page->index != linear_page_index(vma, address)) {
+ pte_t ptfile = pgoff_to_pte(page->index);
+ if (pte_soft_dirty(pteval))
+ pte_file_mksoft_dirty(ptfile);
+ set_pte_at(mm, address, pte, ptfile);
+ }
/* Move the dirty bit to the physical page now the pte is gone. */
if (pte_dirty(pteval))
_
Patches currently in -mm which might be from gorcunov@gmail.com are
mm-save-soft-dirty-bits-on-swapped-pages.patch
mm-save-soft-dirty-bits-on-file-pages.patch
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2013-08-07 20:28 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-08-07 20:28 + mm-save-soft-dirty-bits-on-file-pages.patch added to -mm tree akpm
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.