Skip to content

Instantly share code, notes, and snippets.

@pfactum

pfactum/pksm-v0.2-for-3.14.patch

Created May 27, 2014 19:33
Show Gist options
  • Save pfactum/e962b1a08431e342bbe7 to your computer and use it in GitHub Desktop.
Save pfactum/e962b1a08431e342bbe7 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
pksm-v0.2-for-3.14.patch
diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
index 136e548..e69cb2f 100644
--- a/fs/proc/meminfo.c
+++ b/fs/proc/meminfo.c
@@ -121,6 +121,9 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
"SUnreclaim: %8lu kB\n"
"KernelStack: %8lu kB\n"
"PageTables: %8lu kB\n"
+#ifdef CONFIG_PKSM
+ "KsmSharingPages:%8lu kB\n"
+#endif
#ifdef CONFIG_QUICKLIST
"Quicklists: %8lu kB\n"
#endif
@@ -175,6 +178,9 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
K(global_page_state(NR_SLAB_UNRECLAIMABLE)),
global_page_state(NR_KERNEL_STACK) * THREAD_SIZE / 1024,
K(global_page_state(NR_PAGETABLE)),
+#ifdef CONFIG_PKSM
+ K(global_page_state(NR_PKSM_SHARING_PAGES)),
+#endif
#ifdef CONFIG_QUICKLIST
K(quicklist_total_size()),
#endif
diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index 34c7bdc..55f0de3 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -6,6 +6,7 @@
#include <linux/mm_types.h>
#include <linux/bug.h>
+#include <linux/pksm.h>
/*
* On almost all architectures and configurations, 0 can be used as the
@@ -516,7 +517,7 @@ static inline int is_zero_pfn(unsigned long pfn)
static inline int is_zero_pfn(unsigned long pfn)
{
extern unsigned long zero_pfn;
- return pfn == zero_pfn;
+ return (pfn == zero_pfn) || (is_pksm_zero_pfn(pfn));
}
static inline unsigned long my_zero_pfn(unsigned long addr)
diff --git a/include/linux/ksm.h b/include/linux/ksm.h
index 3be6bb1..d423a38 100644
--- a/include/linux/ksm.h
+++ b/include/linux/ksm.h
@@ -19,21 +19,6 @@ struct mem_cgroup;
#ifdef CONFIG_KSM
int ksm_madvise(struct vm_area_struct *vma, unsigned long start,
unsigned long end, int advice, unsigned long *vm_flags);
-int __ksm_enter(struct mm_struct *mm);
-void __ksm_exit(struct mm_struct *mm);
-
-static inline int ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm)
-{
- if (test_bit(MMF_VM_MERGEABLE, &oldmm->flags))
- return __ksm_enter(mm);
- return 0;
-}
-
-static inline void ksm_exit(struct mm_struct *mm)
-{
- if (test_bit(MMF_VM_MERGEABLE, &mm->flags))
- __ksm_exit(mm);
-}
/*
* A KSM page is one of those write-protected "shared pages" or "merged pages"
@@ -76,6 +61,34 @@ struct page *ksm_might_need_to_copy(struct page *page,
int rmap_walk_ksm(struct page *page, struct rmap_walk_control *rwc);
void ksm_migrate_page(struct page *newpage, struct page *oldpage);
+#ifdef CONFIG_KSM_LEGACY
+int __ksm_enter(struct mm_struct *mm);
+void __ksm_exit(struct mm_struct *mm);
+static inline int ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm)
+{
+ if (test_bit(MMF_VM_MERGEABLE, &oldmm->flags))
+ return __ksm_enter(mm);
+ return 0;
+}
+
+static inline void ksm_exit(struct mm_struct *mm)
+{
+ if (test_bit(MMF_VM_MERGEABLE, &mm->flags))
+ __ksm_exit(mm);
+}
+
+#elif defined(CONFIG_PKSM)
+
+static inline int ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm)
+{
+ return 0;
+}
+
+static inline void ksm_exit(struct mm_struct *mm)
+{
+}
+#endif /* !CONFIG_UKSM */
+
#else /* !CONFIG_KSM */
static inline int ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm)
@@ -123,4 +136,6 @@ static inline void ksm_migrate_page(struct page *newpage, struct page *oldpage)
#endif /* CONFIG_MMU */
#endif /* !CONFIG_KSM */
+#include <linux/pksm.h>
+
#endif /* __LINUX_KSM_H */
diff --git a/include/linux/memory.h b/include/linux/memory.h
index bb7384e..93362c5 100644
--- a/include/linux/memory.h
+++ b/include/linux/memory.h
@@ -153,4 +153,6 @@ struct memory_accessor {
*/
extern struct mutex text_mutex;
+extern int is_pksm_zero_pfn(unsigned long pfn);
+
#endif /* _LINUX_MEMORY_H_ */
diff --git a/include/linux/mm.h b/include/linux/mm.h
index c1b7414..7eef94f 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -19,6 +19,10 @@
#include <linux/bit_spinlock.h>
#include <linux/shrinker.h>
+#ifdef CONFIG_PKSM
+#include <linux/pksm.h>
+#endif
+
struct mempolicy;
struct anon_vma;
struct anon_vma_chain;
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 290901a..e5ab02e 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -195,6 +195,9 @@ struct page {
#ifdef LAST_CPUPID_NOT_IN_PAGE_FLAGS
int _last_cpupid;
#endif
+#ifdef CONFIG_PKSM
+ void *pksm;
+#endif
}
/*
* The struct page can be forced to be double word aligned so that atomic ops
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 9b61b9b..e5b751e 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -143,6 +143,9 @@ enum zone_stat_item {
NUMA_OTHER, /* allocation from other node */
#endif
NR_ANON_TRANSPARENT_HUGEPAGES,
+#ifdef CONFIG_PKSM
+ NR_PKSM_SHARING_PAGES,
+#endif
NR_FREE_CMA_PAGES,
NR_VM_ZONE_STAT_ITEMS };
diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index d1fe1a7..eacc856 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -109,6 +109,9 @@ enum pageflags {
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
PG_compound_lock,
#endif
+#ifdef CONFIG_PKSM
+ PG_pksm,
+#endif
__NR_PAGEFLAGS,
/* Filesystems */
@@ -275,6 +278,11 @@ PAGEFLAG_FALSE(HWPoison)
#define __PG_HWPOISON 0
#endif
+#ifdef CONFIG_PKSM
+PAGEFLAG(PKSM, pksm) __CLEARPAGEFLAG(PKSM, pksm)
+TESTSCFLAG(PKSM, pksm)
+#endif
+
u64 stable_page_flags(struct page *page);
static inline int PageUptodate(struct page *page)
@@ -504,6 +512,7 @@ static inline void ClearPageSlabPfmemalloc(struct page *page)
1 << PG_writeback | 1 << PG_reserved | \
1 << PG_slab | 1 << PG_swapcache | 1 << PG_active | \
1 << PG_unevictable | __PG_MLOCKED | __PG_HWPOISON | \
+ 1 << PG_pksm | \
__PG_COMPOUND_LOCK)
/*
diff --git a/include/linux/pksm.h b/include/linux/pksm.h
new file mode 100644
index 0000000..ee9fc09
--- /dev/null
+++ b/include/linux/pksm.h
@@ -0,0 +1,37 @@
+#ifndef __LINUX_PKSM_H
+#define __LINUX_PKSM_H
+/*
+ * Memory merging support.
+ *
+ * This code enables dynamic sharing of identical pages found in different
+ * memory areas, even if they are not shared by fork().
+ */
+
+/* if !CONFIG_PKSM this file should not be compiled at all. */
+#ifdef CONFIG_PKSM
+
+extern unsigned long zero_pfn __read_mostly;
+extern unsigned long pksm_zero_pfn __read_mostly;
+extern struct page *empty_pksm_zero_page;
+
+/* The number of page slots additionally sharing those nodes */
+extern unsigned long ksm_pages_zero_sharing;
+
+/* must be done before linked to mm */
+
+struct rmap_item;
+
+extern struct rmap_item *pksm_alloc_rmap_item(void);
+extern void pksm_free_rmap_item(struct rmap_item *rmap_item);
+extern int pksm_add_new_anon_page(struct page *page, struct rmap_item *rmap_item, struct anon_vma *anon_vma);
+extern int pksm_del_anon_page(struct page *page);
+extern void pksm_unmap_sharing_page(struct page *page, struct mm_struct *mm, unsigned long address);
+extern void pksm_free_stable_anon(struct anon_vma *anon_vma);
+extern int is_pksm_zero_pfn(unsigned long pfn);
+extern void pksm_map_zero_page(pte_t pte);
+extern void pksm_unmap_zero_page(pte_t pte);
+
+#else
+
+#endif /* !CONFIG_PKSM */
+#endif /* __LINUX_PKSM_H */
diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index b66c211..d35a206 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -82,6 +82,13 @@ enum ttu_flags {
TTU_IGNORE_HWPOISON = (1 << 10),/* corrupted page is recoverable */
};
+struct page_referenced_arg {
+ int mapcount;
+ int referenced;
+ unsigned long vm_flags;
+ struct mem_cgroup *memcg;
+};
+
#ifdef CONFIG_MMU
static inline void get_anon_vma(struct anon_vma *anon_vma)
{
diff --git a/mm/Kconfig b/mm/Kconfig
index 2888024..f999ccfd 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -324,6 +324,28 @@ config KSM
See Documentation/vm/ksm.txt for more information: KSM is inactive
until a program has madvised that an area is MADV_MERGEABLE, and
root has set /sys/kernel/mm/ksm/run to 1 (if CONFIG_SYSFS is set).
+choice
+ prompt "Choose PKSM/KSM strategy"
+ default PKSM
+ depends on KSM
+ help
+ This option allows to select a PKSM/KSM stragety.
+
+config PKSM
+ bool "AnonPage-KSM for page merging"
+ depends on KSM
+ help
+ PKSM is inspired by the Linux kernel project -KSM, but with a fundamentally rewritten core algorithm. With
+ an advanced algorithm, PKSM now can transparently scans all anonymously
+ mapped pages. Since KVM is friendly to KSM, KVM can also benefit from
+ PKSM. Now PKSM has its first stable release and first real world enterprise user.
+
+config KSM_LEGACY
+ bool "Legacy KSM implementation"
+ depends on KSM
+ help
+ The legacy KSM implementation from Redhat.
+endchoice
config DEFAULT_MMAP_MIN_ADDR
int "Low address space to protect from user allocation"
diff --git a/mm/Makefile b/mm/Makefile
index 310c90a..091ecc8 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -40,7 +40,8 @@ obj-$(CONFIG_SPARSEMEM) += sparse.o
obj-$(CONFIG_SPARSEMEM_VMEMMAP) += sparse-vmemmap.o
obj-$(CONFIG_SLOB) += slob.o
obj-$(CONFIG_MMU_NOTIFIER) += mmu_notifier.o
-obj-$(CONFIG_KSM) += ksm.o
+obj-$(CONFIG_KSM_LEGACY) += ksm.o
+obj-$(CONFIG_PKSM) += pksm.o
obj-$(CONFIG_PAGE_POISONING) += debug-pagealloc.o
obj-$(CONFIG_SLAB) += slab.o
obj-$(CONFIG_SLUB) += slub.o
diff --git a/mm/internal.h b/mm/internal.h
index 29e1e76..2856874 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -229,6 +229,12 @@ extern pmd_t maybe_pmd_mkwrite(pmd_t pmd, struct vm_area_struct *vma);
extern unsigned long vma_address(struct page *page,
struct vm_area_struct *vma);
#endif
+
+#ifdef CONFIG_PKSM
+extern unsigned long vma_address(struct page *page,
+ struct vm_area_struct *vma);
+#endif
+
#else /* !CONFIG_MMU */
static inline int mlocked_vma_newpage(struct vm_area_struct *v, struct page *p)
{
diff --git a/mm/memory.c b/mm/memory.c
index 22dfa61..c4ca6f9 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -127,7 +127,6 @@ static int __init init_zero_pfn(void)
}
core_initcall(init_zero_pfn);
-
#if defined(SPLIT_RSS_COUNTING)
void sync_mm_rss(struct mm_struct *mm)
@@ -871,7 +870,8 @@ copy_one_pte(struct mm_struct *dst_mm, struct mm_struct *src_mm,
rss[MM_ANONPAGES]++;
else
rss[MM_FILEPAGES]++;
- }
+ } else
+ pksm_map_zero_page(pte);
out_set_pte:
set_pte_at(dst_mm, addr, dst_pte, pte);
@@ -1113,8 +1113,10 @@ again:
ptent = ptep_get_and_clear_full(mm, addr, pte,
tlb->fullmm);
tlb_remove_tlb_entry(tlb, pte, addr);
- if (unlikely(!page))
+ if (unlikely(!page)) {
+ pksm_unmap_zero_page(ptent);
continue;
+ }
if (unlikely(details) && details->nonlinear_vma
&& linear_page_index(details->nonlinear_vma,
addr) != page->index) {
@@ -1133,6 +1135,9 @@ again:
mark_page_accessed(page);
rss[MM_FILEPAGES]--;
}
+#ifdef CONFIG_PKSM
+ pksm_unmap_sharing_page(page, mm, addr);
+#endif
page_remove_rmap(page);
if (unlikely(page_mapcount(page) < 0))
print_bad_pte(vma, addr, ptent, page);
@@ -2815,8 +2820,10 @@ gotten:
dec_mm_counter_fast(mm, MM_FILEPAGES);
inc_mm_counter_fast(mm, MM_ANONPAGES);
}
- } else
+ } else {
+ pksm_unmap_zero_page(orig_pte);
inc_mm_counter_fast(mm, MM_ANONPAGES);
+ }
flush_cache_page(vma, address, pte_pfn(orig_pte));
entry = mk_pte(new_page, vma->vm_page_prot);
entry = maybe_mkwrite(pte_mkdirty(entry), vma);
@@ -2858,6 +2865,9 @@ gotten:
* mapcount is visible. So transitively, TLBs to
* old page will be flushed before it can be reused.
*/
+#ifdef CONFIG_PKSM
+ pksm_unmap_sharing_page(old_page, mm, address);
+#endif
page_remove_rmap(old_page);
}
@@ -3226,6 +3236,10 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
spinlock_t *ptl;
pte_t entry;
+#ifdef CONFIG_PKSM
+ struct rmap_item *rmap = NULL;
+#endif
+
pte_unmap(page_table);
/* Check if we need to add a guard page to the stack */
@@ -3262,6 +3276,10 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
if (vma->vm_flags & VM_WRITE)
entry = pte_mkwrite(pte_mkdirty(entry));
+#ifdef CONFIG_PKSM
+ rmap = pksm_alloc_rmap_item();
+#endif
+
page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
if (!pte_none(*page_table))
goto release;
@@ -3273,13 +3291,23 @@ setpte:
/* No need to invalidate - it was non-present before */
update_mmu_cache(vma, address, page_table);
+
unlock:
pte_unmap_unlock(page_table, ptl);
+#ifdef CONFIG_PKSM
+ if (rmap)
+ pksm_add_new_anon_page(page, rmap, vma->anon_vma);
+#endif
return 0;
release:
mem_cgroup_uncharge_page(page);
page_cache_release(page);
+#ifdef CONFIG_PKSM
+ if (rmap)
+ pksm_free_rmap_item(rmap);
+#endif
goto unlock;
+
oom_free_page:
page_cache_release(page);
oom:
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 3bac76a..fb771ca 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1376,9 +1376,13 @@ void free_hot_cold_page(struct page *page, int cold)
unsigned long flags;
int migratetype;
+#ifdef CONFIG_PKSM
+ if (PagePKSM(page))
+ pksm_del_anon_page(page);
+#endif
+
if (!free_pages_prepare(page, 0))
return;
-
migratetype = get_pageblock_migratetype(page);
set_freepage_migratetype(page, migratetype);
local_irq_save(flags);
@@ -6511,6 +6515,9 @@ static const struct trace_print_flags pageflag_names[] = {
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
{1UL << PG_compound_lock, "compound_lock" },
#endif
+#ifdef CONFIG_PKSM
+ {1UL << PG_pksm, "pksm" },
+#endif
};
static void dump_page_flags(unsigned long flags)
diff --git a/mm/pksm.c b/mm/pksm.c
new file mode 100644
index 0000000..9544c1e
--- /dev/null
+++ b/mm/pksm.c
@@ -0,0 +1,2378 @@
+/*
+ * PKSM (Anon-page KSM).
+ *Copyright (C) 2012-2013
+ * Authors:
+ * Figo.zhang (figo1802@gmail.com)
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.
+ *
+ * This is an improvement upon KSM/UKSM. Some basic data structures and routines
+ * are borrowed from ksm.c and uksm.c .
+ *
+ * Its new features:
+ * 1. Full system scan:
+ * It automatically scans all user processes' anonymous pages. Kernel-user
+ * interaction to submit a memory area to KSM is no longer needed.
+ *
+ * 2. High efficiency for anonymous page detection:
+ * It automatically detects anonymous page is created and freed, It will use a new
+ * algorithm and mechanism that it directly handle the anonymous page when it be created/freed
+ * by the linux kernel. It is no need to waster CPU time to traversing all of the VMA
+ * areas to find valid anonymous pages. KSM/UKSM will waster a lot of CPU time to traversing VMAs
+ * to find a valid anonymous page, PKSM no need considering it.
+ *
+ * 3. Full Zero Page consideration
+ * Now pksmd consider full zero pages as special pages and merge them to an
+ * special unswappable pksm zero page.
+ *
+ * 4. Check page content periodically
+ * Pksm considers unstable page contents are volatile, which are add the a FIFO list. It check page
+ * content by a hash value periodically. The default scanning cycles is 20 minutes.
+ *
+ */
+
+#include <linux/errno.h>
+#include <linux/mm.h>
+#include <linux/fs.h>
+#include <linux/mman.h>
+#include <linux/sched.h>
+#include <linux/rwsem.h>
+#include <linux/pagemap.h>
+#include <linux/rmap.h>
+#include <linux/spinlock.h>
+#include <linux/jhash.h>
+#include <linux/delay.h>
+#include <linux/kthread.h>
+#include <linux/wait.h>
+#include <linux/slab.h>
+#include <linux/rbtree.h>
+#include <linux/memory.h>
+#include <linux/mmu_notifier.h>
+#include <linux/swap.h>
+#include <linux/ksm.h>
+#include <linux/hash.h>
+#include <linux/freezer.h>
+#include <linux/oom.h>
+#include <linux/random.h>
+
+#include <asm/tlbflush.h>
+#include "internal.h"
+
+#define CONFIG_PKSM_RHASH
+
+#ifdef CONFIG_X86
+#undef memcmp
+
+#ifdef CONFIG_X86_32
+#define memcmp memcmpx86_32
+/*
+ * Compare 4-byte-aligned address s1 and s2, with length n
+ */
+int memcmpx86_32(void *s1, void *s2, size_t n)
+{
+ size_t num = n / 4;
+ register int res;
+
+ __asm__ __volatile__
+ (
+ "testl %3,%3\n\t"
+ "repe; cmpsd\n\t"
+ "je 1f\n\t"
+ "sbbl %0,%0\n\t"
+ "orl $1,%0\n"
+ "1:"
+ : "=&a" (res), "+&S" (s1), "+&D" (s2), "+&c" (num)
+ : "0" (0)
+ : "cc");
+
+ return res;
+}
+
+/*
+ * Check the page is all zero ?
+ */
+static int is_full_zero(const void *s1, size_t len)
+{
+ unsigned char same;
+
+ len /= 4;
+
+ __asm__ __volatile__
+ ("repe; scasl;"
+ "sete %0"
+ : "=qm" (same), "+D" (s1), "+c" (len)
+ : "a" (0)
+ : "cc");
+
+ return same;
+}
+
+
+#elif defined(CONFIG_X86_64)
+#define memcmp memcmpx86_64
+/*
+ * Compare 8-byte-aligned address s1 and s2, with length n
+ */
+int memcmpx86_64(void *s1, void *s2, size_t n)
+{
+ size_t num = n / 8;
+ register int res;
+
+ __asm__ __volatile__
+ (
+ "testq %q3,%q3\n\t"
+ "repe; cmpsq\n\t"
+ "je 1f\n\t"
+ "sbbq %q0,%q0\n\t"
+ "orq $1,%q0\n"
+ "1:"
+ : "=&a" (res), "+&S" (s1), "+&D" (s2), "+&c" (num)
+ : "0" (0)
+ : "cc");
+
+ return res;
+}
+
+static int is_full_zero(const void *s1, size_t len)
+{
+ unsigned char same;
+
+ len /= 8;
+
+ __asm__ __volatile__
+ ("repe; scasq;"
+ "sete %0"
+ : "=qm" (same), "+D" (s1), "+c" (len)
+ : "a" (0)
+ : "cc");
+
+ return same;
+}
+
+#endif
+#else
+static int is_full_zero(const void *s1, size_t len)
+{
+ unsigned long *src = s1;
+ int i;
+
+ len /= sizeof(*src);
+
+ for (i = 0; i < len; i++) {
+ if (src[i])
+ return 0;
+ }
+
+ return 1;
+}
+#endif
+
+struct stable_node_anon {
+ struct hlist_node hlist;
+ struct anon_vma *anon_vma;
+};
+
+/**
+ * struct rmap_item - reverse mapping item for virtual addresses
+ * @rmap_list: next rmap_item in mm_slot's singly-linked rmap_list
+ * @anon_vma: pointer to anon_vma for this mm,address, when in stable tree
+ * @mm: the memory structure this rmap_item is pointing into
+ * @address: the virtual address this rmap_item tracks (+ flags in low bits)
+ * @oldchecksum: previous checksum of the page at that virtual address
+ * @node: rb node of this rmap_item in the unstable tree
+ * @head: pointer to stable_node heading this list in the stable tree
+ * @hlist: link into hlist of rmap_items hanging off that stable_node
+ */
+struct rmap_item {
+ struct anon_vma *anon_vma; /* point to the page anon_page */
+ struct page *page;
+ unsigned long address; /* + low bits used for flags below */
+ struct list_head list; /*list for add new page(rmap_item)*/
+ struct list_head del_list; /*list for del a page(rmap_item)*/
+ struct hlist_head hlist; /*list for stable anon */
+ union {
+ struct rb_node node; /* when node of unstable tree */
+ };
+ atomic_t _mapcount;
+ unsigned long checksum;
+ struct list_head update_list; /*list for unstable page checksum update*/
+};
+
+
+/**
+ * struct mm_slot - ksm information per mm that is being scanned
+ * @link: link to the mm_slots hash list
+ * @mm_list: link into the mm_slots list, rooted in ksm_mm_head
+ * @rmap_list: head for this mm_slot's singly-linked list of rmap_items
+ * @mm: the mm that this information is valid for
+ */
+struct mm_slot {
+ struct hlist_node link;
+ struct list_head mm_list;
+ struct rmap_item *rmap_list;
+ struct mm_struct *mm;
+};
+
+/**
+ * struct ksm_scan - cursor for scanning
+ * @mm_slot: the current mm_slot we are scanning
+ * @address: the next address inside that to be scanned
+ * @rmap_list: link to the next rmap to be scanned in the rmap_list
+ * @seqnr: count of completed full scans (needed when removing unstable node)
+ *
+ * There is only the one ksm_scan instance of this cursor structure.
+ */
+struct ksm_scan {
+ struct mm_slot *mm_slot;
+ unsigned long address;
+ struct rmap_item **rmap_list;
+ unsigned long seqnr;
+};
+
+#define SEQNR_MASK 0x0ff /* low bits of unstable tree seqnr */
+
+#define NEWLIST_FLAG (1<<0) /* rmap_item is at new_anon_page_list */
+#define DELLIST_FLAG (1<<1) /*rmap_item at del_anon_page_list*/
+#define INKSM_FLAG (1<<2) /*this page add to ksm subsystem*/
+#define UNSTABLE_FLAG (1<<3) /* is a node of the unstable tree */
+#define STABLE_FLAG (1<<4) /* is listed from the stable tree */
+#define CHECKSUM_LIST_FLAG (1<<5) /*rmap_item in checksum list*/
+#define INITCHECKSUM_FLAG (1<<6)
+#define RESCAN_LIST_FLAG (1<<7) /*rmap_item in pksm_rescan_page_list*/
+
+
+#define PKSM_FAULT_SUCCESS 0
+#define PKSM_FAULT_DROP 1 /*drop this rmap_item*/
+#define PKSM_FAULT_TRY 2 /*retry this rmap_item*/
+#define PKSM_FAULT_KEEP 3 /*keep this rmap_item*/
+
+/* The stable and unstable tree heads */
+static struct rb_root root_stable_tree = RB_ROOT;
+static struct rb_root root_unstable_tree = RB_ROOT;
+
+
+#define MM_SLOTS_HASH_SHIFT 10
+#define MM_SLOTS_HASH_HEADS (1 << MM_SLOTS_HASH_SHIFT)
+static struct hlist_head mm_slots_hash[MM_SLOTS_HASH_HEADS];
+
+static struct mm_slot ksm_mm_head = {
+ .mm_list = LIST_HEAD_INIT(ksm_mm_head.mm_list),
+};
+static struct ksm_scan ksm_scan = {
+ .mm_slot = &ksm_mm_head,
+};
+
+static struct kmem_cache *rmap_item_cache;
+static struct kmem_cache *stable_anon_cache;
+static struct kmem_cache *mm_slot_cache;
+
+/* The number of nodes in the stable tree */
+static unsigned long ksm_pages_shared;
+
+/* The number of page slots additionally sharing those nodes */
+static unsigned long ksm_pages_sharing;
+
+/* The number of nodes in the unstable tree */
+static unsigned long ksm_pages_unshared;
+
+/* The number of rmap_items in use*/
+static unsigned long ksm_rmap_items;
+
+static unsigned long ksm_stable_nodes;
+
+unsigned long ksm_pages_zero_sharing;
+
+/* Number of pages ksmd should scan in one batch */
+static unsigned int ksm_thread_pages_to_scan = 1000;
+
+/* Milliseconds ksmd should sleep between batches */
+static unsigned int ksm_thread_sleep_millisecs = 20;
+
+/*Seconds pksm should update all unshared_pages by one period*/
+static unsigned int pksm_unshared_page_update_period = 10;
+
+#define KSM_RUN_STOP 0
+#define KSM_RUN_MERGE 1
+#define KSM_RUN_UNMERGE 2
+static unsigned int ksm_run = KSM_RUN_MERGE;
+
+/* The hash strength needed to hash a full page */
+#define RSAD_STRENGTH_FULL (PAGE_SIZE / sizeof(u32))
+
+/* The random offsets in a page */
+static u32 *pksm_random_table;
+static u32 pksm_zero_random_checksum;
+
+/* 32/3 < they < 32/2 */
+#define shiftl 8
+#define shiftr 12
+
+#define HASH_FROM_TO(from, to) \
+for (index = from; index < to; index++) { \
+ pos = pksm_random_table[index]; \
+ hash += key[pos]; \
+ hash += (hash << shiftl); \
+ hash ^= (hash >> shiftr); \
+}
+
+#define ZERO_HASH_FROM_TO(from, to) \
+for (index = from; index < to; index++) { \
+ pos = pksm_random_table[index]; \
+ hash += 0; \
+ hash += (hash << shiftl); \
+ hash ^= (hash >> shiftr); \
+}
+
+static DECLARE_WAIT_QUEUE_HEAD(ksm_thread_wait);
+static DEFINE_MUTEX(ksm_thread_mutex);
+static DEFINE_SPINLOCK(ksm_mmlist_lock);
+
+//add by figo
+static DEFINE_SPINLOCK(pksm_np_list_lock);
+struct list_head new_anon_page_list = LIST_HEAD_INIT(new_anon_page_list);
+struct list_head del_anon_page_list = LIST_HEAD_INIT(del_anon_page_list);
+struct list_head pksm_rescan_page_list = LIST_HEAD_INIT(pksm_rescan_page_list);
+
+struct list_head unstabletree_checksum_list = LIST_HEAD_INIT(unstabletree_checksum_list);
+struct list_head pksm_volatile_page_list = LIST_HEAD_INIT(pksm_volatile_page_list);
+
+#define KSM_KMEM_CACHE(__struct, __flags) kmem_cache_create("ksm_"#__struct,\
+ sizeof(struct __struct), __alignof__(struct __struct),\
+ (__flags), NULL)
+
+static int __init ksm_slab_init(void)
+{
+ rmap_item_cache = KSM_KMEM_CACHE(rmap_item, 0);
+ if (!rmap_item_cache)
+ goto out;
+
+ stable_anon_cache = KSM_KMEM_CACHE(stable_node_anon, 0);
+ if (!stable_anon_cache)
+ goto out_free1;
+
+ mm_slot_cache = KSM_KMEM_CACHE(mm_slot, 0);
+ if (!mm_slot_cache)
+ goto out_free2;
+
+ return 0;
+
+out_free2:
+ kmem_cache_destroy(stable_anon_cache);
+out_free1:
+ kmem_cache_destroy(rmap_item_cache);
+out:
+ return -ENOMEM;
+}
+
+static void __init ksm_slab_free(void)
+{
+ kmem_cache_destroy(mm_slot_cache);
+ kmem_cache_destroy(stable_anon_cache);
+ kmem_cache_destroy(rmap_item_cache);
+ mm_slot_cache = NULL;
+}
+
+static inline struct stable_node_anon *alloc_stable_anon(void)
+{
+ struct stable_node_anon *node =NULL;
+
+ node = kmem_cache_alloc(stable_anon_cache, GFP_KERNEL);
+ if (node) {
+ ksm_stable_nodes++;
+ }
+
+ return node;
+}
+
+static inline void free_stable_anon(struct stable_node_anon *stable_anon)
+{
+ if (stable_anon) {
+ kmem_cache_free(stable_anon_cache, stable_anon);
+ ksm_stable_nodes--;
+ }
+}
+
+struct rmap_item *pksm_alloc_rmap_item(void)
+{
+ struct rmap_item *rmap_item;
+
+ rmap_item = kmem_cache_zalloc(rmap_item_cache, GFP_KERNEL);
+ if (rmap_item) {
+ INIT_HLIST_HEAD(&rmap_item->hlist);
+ INIT_LIST_HEAD(&rmap_item->list);
+ INIT_LIST_HEAD(&rmap_item->del_list);
+ INIT_LIST_HEAD(&rmap_item->update_list);
+ RB_CLEAR_NODE(&rmap_item->node);
+ ksm_rmap_items++;
+ atomic_set(&rmap_item->_mapcount, 0);
+ }
+ return rmap_item;
+}
+
+void pksm_free_rmap_item(struct rmap_item *rmap_item)
+{
+ if (rmap_item) {
+ ksm_rmap_items--;
+ kmem_cache_free(rmap_item_cache, rmap_item);
+ }
+}
+
+static struct page *page_trans_compound_anon(struct page *page)
+{
+ if (PageTransCompound(page)) {
+ struct page *head = compound_head(page);
+ /*
+ * head may actually be splitted and freed from under
+ * us but it's ok here.
+ */
+ if (PageAnon(head))
+ return head;
+ }
+ return NULL;
+}
+
+static int check_valid_rmap_item(struct rmap_item *rmap_item)
+{
+ if (!rmap_item || !rmap_item->page || !PagePKSM(rmap_item->page) ||
+ !(rmap_item->address & INKSM_FLAG))
+ return 0;
+ else
+ return 1;
+}
+
+static inline struct rmap_item *page_stable_rmap_item(struct page *page)
+{
+ //return PageKsm(page) ? ((unsigned long)page->mapping & ~PAGE_MAPPING_FLAGS) : NULL;
+
+ if (!PageKsm(page))
+ return NULL;
+
+ if (!PagePKSM(page))
+ return NULL;
+
+ return page->pksm;
+}
+
+static inline struct mm_slot *alloc_mm_slot(void)
+{
+ if (!mm_slot_cache) /* initialization failed */
+ return NULL;
+ return kmem_cache_zalloc(mm_slot_cache, GFP_KERNEL);
+}
+
+static inline void free_mm_slot(struct mm_slot *mm_slot)
+{
+ kmem_cache_free(mm_slot_cache, mm_slot);
+}
+
+static struct mm_slot *get_mm_slot(struct mm_struct *mm)
+{
+ struct mm_slot *mm_slot;
+ struct hlist_head *bucket;
+
+ bucket = &mm_slots_hash[hash_ptr(mm, MM_SLOTS_HASH_SHIFT)];
+ hlist_for_each_entry(mm_slot, bucket, link) {
+ if (mm == mm_slot->mm)
+ return mm_slot;
+ }
+ return NULL;
+}
+
+static void insert_to_mm_slots_hash(struct mm_struct *mm,
+ struct mm_slot *mm_slot)
+{
+ struct hlist_head *bucket;
+
+ bucket = &mm_slots_hash[hash_ptr(mm, MM_SLOTS_HASH_SHIFT)];
+ mm_slot->mm = mm;
+ hlist_add_head(&mm_slot->link, bucket);
+}
+
+static inline int in_stable_tree(struct rmap_item *rmap_item)
+{
+ return rmap_item->address & STABLE_FLAG;
+}
+
+static inline int in_unstable_tree(struct rmap_item *rmap_item)
+{
+ return rmap_item->address & UNSTABLE_FLAG;
+}
+
+unsigned long pksm_zero_pfn __read_mostly;
+struct page *empty_pksm_zero_page;
+
+static int __init setup_pksm_zero_page(void)
+{
+ unsigned long addr;
+ addr = __get_free_pages(GFP_KERNEL | __GFP_ZERO, 0);
+ if (!addr)
+ panic("Oh boy, that early out of memory?");
+
+ empty_pksm_zero_page = virt_to_page((void *) addr);
+ SetPageReserved(empty_pksm_zero_page);
+
+ pksm_zero_pfn = page_to_pfn(empty_pksm_zero_page);
+
+ return 0;
+}
+core_initcall(setup_pksm_zero_page);
+
+inline int is_pksm_zero_pfn(unsigned long pfn)
+{
+ return pfn == pksm_zero_pfn;
+}
+
+inline void pksm_unmap_zero_page(pte_t pte)
+{
+ if (pte_pfn(pte) == pksm_zero_pfn) {
+ ksm_pages_zero_sharing--;
+ __dec_zone_page_state(empty_pksm_zero_page, NR_PKSM_SHARING_PAGES);
+ }
+}
+
+inline void pksm_map_zero_page(pte_t pte)
+{
+ if (pte_pfn(pte) == pksm_zero_pfn) {
+ ksm_pages_zero_sharing++;
+ __inc_zone_page_state(empty_pksm_zero_page, NR_PKSM_SHARING_PAGES);
+ }
+}
+
+static void pksm_del_sharing_page_counter(struct page *page, int n)
+{
+ int i;
+ struct rmap_item * rmap_item = (struct rmap_item *)(page->pksm);
+
+ for (i = 0; i < n; i++) {
+ ksm_pages_sharing--;
+ atomic_dec(&rmap_item->_mapcount);
+ __dec_zone_page_state(page, NR_PKSM_SHARING_PAGES);
+ }
+}
+
+static void pksm_add_sharing_page_counter(struct page *page, int n)
+{
+ int i;
+ struct rmap_item *rmap_item;
+ if (!PageKsm(page))
+ return;
+
+ rmap_item = (struct rmap_item *)(page->pksm);
+
+ for (i = 0; i<n; i++) {
+ ksm_pages_sharing++;
+ atomic_inc(&rmap_item->_mapcount);
+ __inc_zone_page_state(page, NR_PKSM_SHARING_PAGES);
+ }
+}
+
+void pksm_unmap_sharing_page(struct page *page, struct mm_struct *mm, unsigned long address)
+{
+ int ksm_map, mapcount;
+ struct rmap_item *rmap_item;
+
+ if (!(PageKsm(page) && PagePKSM(page)))
+ return;
+
+ rmap_item = (struct rmap_item *)(page->pksm);
+
+ if (!check_valid_rmap_item(rmap_item))
+ return;
+
+ ksm_map = atomic_read(&rmap_item->_mapcount);
+ mapcount = page_mapcount(page);
+ if (mapcount > ksm_map)
+ return;
+ else
+ goto free;
+
+free:
+ if (ksm_map > 0)
+ pksm_del_sharing_page_counter(page, 1);
+}
+
+/*
+ * ksmd must not touch an mm's
+ * page tables after it has passed through ksm_exit() - which, if necessary,
+ * takes mmap_sem briefly to serialize against them. ksm_exit() does not set
+ * a special flag: they can just back out as soon as mm_users goes to zero.
+ * ksm_test_exit() is used throughout to make this test for exit: in some
+ * places for correctness, in some places just to avoid unnecessary work.
+ */
+static inline bool ksm_test_exit(struct mm_struct *mm)
+{
+ return atomic_read(&mm->mm_users) == 0;
+}
+
+/*
+ * Check that no O_DIRECT or similar I/O is in progress on the
+ * page
+ */
+static int check_page_dio(struct page *page)
+{
+ int swapped = PageSwapCache(page);
+ return (page_mapcount(page) +1+ swapped != page_count(page));
+}
+
+static struct page *get_ksm_page(struct rmap_item *rmap_item)
+{
+ struct page *page;
+ void *expected_mapping;
+
+ if (!check_valid_rmap_item(rmap_item))
+ goto out;
+
+ if (!(rmap_item->address & STABLE_FLAG))
+ goto out;
+
+ page = rmap_item->page;
+ if (!page || !PageKsm(page))
+ goto out;
+
+ expected_mapping = (void *)rmap_item +
+ (PAGE_MAPPING_ANON | PAGE_MAPPING_KSM);
+ rcu_read_lock();
+ if (page->mapping != expected_mapping) {
+ goto stale;
+ }
+ if (!get_page_unless_zero(page))
+ goto stale;
+ if (!check_valid_rmap_item(rmap_item)) {
+ put_page(page);
+ goto stale;
+ }
+ if (page->mapping != expected_mapping) {
+ put_page(page);
+ goto stale;
+ }
+ rcu_read_unlock();
+ return page;
+stale:
+ rcu_read_unlock();
+out:
+ return NULL;
+}
+
+static struct page *get_mergeable_page(struct rmap_item *rmap_item)
+{
+
+ struct page *page =NULL;
+
+ if (!check_valid_rmap_item(rmap_item))
+ goto out;
+
+ if (!(rmap_item->address & UNSTABLE_FLAG))
+ goto out;
+
+ page = rmap_item->page;
+ if (IS_ERR_OR_NULL(page))
+ goto out;
+
+ rcu_read_lock();
+ if (!get_page_unless_zero(page))
+ goto unlock;
+
+ if (!check_valid_rmap_item(rmap_item))
+ goto unlock;
+
+ if (PageAnon(page)) {
+ flush_dcache_page(page);
+ } else {
+ put_page(page);
+ goto unlock;
+ }
+
+ rcu_read_unlock();
+ return page;
+
+unlock:
+ rcu_read_unlock();
+out:
+ return NULL;
+}
+
+/*
+ * Removing rmap_item from stable or unstable tree.
+ * This function will clean the information from the stable/unstable tree.
+ */
+static void remove_rmap_item_from_tree(struct rmap_item *rmap_item, int free_anon)
+{
+
+ struct stable_node_anon *stable_anon;
+ struct hlist_node *tmp;
+
+ if (!rmap_item)
+ return;
+
+ if (rmap_item->address & STABLE_FLAG) {
+ WARN_ON(RB_EMPTY_NODE(&rmap_item->node));
+ if (!RB_EMPTY_NODE(&rmap_item->node)) {
+ rmap_item->address &=~ STABLE_FLAG;
+ rb_erase(&rmap_item->node, &root_stable_tree);
+ RB_CLEAR_NODE(&rmap_item->node);
+ ksm_pages_shared--;
+ }
+ } else if (rmap_item->address & UNSTABLE_FLAG) {
+ WARN_ON(RB_EMPTY_NODE(&rmap_item->node));
+ if (!RB_EMPTY_NODE(&rmap_item->node)) {
+ rmap_item->address &=~ UNSTABLE_FLAG;
+ rb_erase(&rmap_item->node, &root_unstable_tree);
+ RB_CLEAR_NODE(&rmap_item->node);
+ ksm_pages_unshared--;
+ }
+
+ if (rmap_item->address & CHECKSUM_LIST_FLAG) {
+ list_del_init(&rmap_item->update_list);
+ rmap_item->address &=~ CHECKSUM_LIST_FLAG;
+ }
+ }
+
+ /*free stable_anon_node*/
+ if (free_anon && !hlist_empty(&rmap_item->hlist)) {
+ hlist_for_each_entry_safe(stable_anon, tmp, &rmap_item->hlist, hlist) {
+ if (!stable_anon)
+ continue;
+ hlist_del(&stable_anon->hlist);
+ put_anon_vma(stable_anon->anon_vma);
+ free_stable_anon(stable_anon);
+ cond_resched();
+ }
+ }
+}
+
+void pksm_clean_all_rmap_items(struct list_head *list)
+{
+ struct rmap_item *rmap_item , *n_item;
+
+ list_for_each_entry_safe(rmap_item, n_item, list, del_list) {
+ list_del(&rmap_item->del_list);
+ remove_rmap_item_from_tree(rmap_item, 1);
+ rmap_item->address &=~ INKSM_FLAG;
+ rmap_item->address &=~ DELLIST_FLAG;
+ pksm_free_rmap_item(rmap_item);
+ cond_resched();
+ }
+}
+
+static void pksm_free_all_rmap_items(void)
+{
+ struct rmap_item *rmap_item , *n_item;
+ LIST_HEAD(l_del);
+
+ spin_lock_irq(&pksm_np_list_lock);
+ list_for_each_entry_safe(rmap_item, n_item, &del_anon_page_list, del_list) {
+ if (!rmap_item)
+ continue;
+ list_move(&rmap_item->del_list, &l_del);
+ }
+ spin_unlock_irq(&pksm_np_list_lock);
+
+ pksm_clean_all_rmap_items(&l_del);
+}
+
+static u32 pksm_calc_checksum(void *addr, u32 hash_strength)
+{
+ u32 hash = 0xdeadbeef;
+ int index, pos;
+ u32 *key = (u32 *)addr;
+ HASH_FROM_TO(0, hash_strength);
+ return hash;
+}
+
+static u32 pksm_calc_zero_page_checksum(u32 hash_strength)
+{
+ u32 hash = 0xdeadbeef;
+ int index, pos;
+ ZERO_HASH_FROM_TO(0, hash_strength);
+ return hash;
+}
+
+static u32 calc_checksum(struct page *page)
+{
+ u32 checksum;
+ void *addr = kmap_atomic(page);
+ //checksum = jhash2(addr, PAGE_SIZE / 4, 17);
+ checksum = pksm_calc_checksum(addr, RSAD_STRENGTH_FULL >> 4);
+ kunmap_atomic(addr);
+ return checksum;
+}
+
+
+static int memcmp_pages(struct page *page1, struct page *page2)
+{
+ char *addr1, *addr2;
+ int ret;
+
+ addr1 = kmap_atomic(page1);
+ addr2 = kmap_atomic(page2);
+ ret = memcmp(addr1, addr2, PAGE_SIZE);
+ kunmap_atomic(addr2);
+ kunmap_atomic(addr1);
+ return ret;
+}
+
+static inline int pages_identical(struct page *page1, struct page *page2)
+{
+ return !memcmp_pages(page1, page2);
+}
+
+static int write_protect_page(struct vm_area_struct *vma, struct page *page,
+ pte_t *orig_pte)
+{
+ struct mm_struct *mm = vma->vm_mm;
+ unsigned long addr;
+ pte_t *ptep;
+ spinlock_t *ptl;
+ int swapped;
+ int err = PKSM_FAULT_DROP;
+
+ addr = page_address_in_vma(page, vma);
+ if (addr == -EFAULT)
+ goto out;
+
+ BUG_ON(PageTransCompound(page));
+ ptep = page_check_address(page, mm, addr, &ptl, 0);
+ if (!ptep)
+ goto out;
+
+ if (pte_write(*ptep) || pte_dirty(*ptep)) {
+ pte_t entry;
+
+ swapped = PageSwapCache(page);
+ flush_cache_page(vma, addr, page_to_pfn(page));
+ /*
+ * Ok this is tricky, when get_user_pages_fast() run it doesn't
+ * take any lock, therefore the check that we are going to make
+ * with the pagecount against the mapcount is racey and
+ * O_DIRECT can happen right after the check.
+ * So we clear the pte and flush the tlb before the check
+ * this assure us that no O_DIRECT can happen after the check
+ * or in the middle of the check.
+ */
+ entry = ptep_clear_flush(vma, addr, ptep);
+ /*
+ * Check that no O_DIRECT or similar I/O is in progress on the
+ * page
+ */
+ if (page_mapcount(page) +1+ swapped != page_count(page)) {
+ set_pte_at(mm, addr, ptep, entry);
+ err = PKSM_FAULT_TRY;
+ goto out_unlock;
+ }
+
+ if (pte_dirty(entry))
+ set_page_dirty(page);
+ entry = pte_mkclean(pte_wrprotect(entry));
+ set_pte_at_notify(mm, addr, ptep, entry);
+ }
+ *orig_pte = *ptep;
+ err = 0;
+
+out_unlock:
+ pte_unmap_unlock(ptep, ptl);
+out:
+ return err;
+}
+
+/**
+ * replace_page - replace page in vma by new ksm page
+ * @vma: vma that holds the pte pointing to page
+ * @page: the page we are replacing by kpage
+ * @kpage: the ksm page we replace page by
+ * @orig_pte: the original value of the pte
+ *
+ * Returns 0 on success, -EFAULT on failure.
+ */
+static int replace_page(struct vm_area_struct *vma, struct page *page,
+ struct page *kpage, pte_t orig_pte)
+{
+ struct mm_struct *mm = vma->vm_mm;
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *ptep;
+ spinlock_t *ptl;
+ unsigned long addr;
+ int err = PKSM_FAULT_DROP;
+ pte_t entry;
+
+ addr = page_address_in_vma(page, vma);
+ if (addr == -EFAULT)
+ goto out;
+
+ pgd = pgd_offset(mm, addr);
+ if (!pgd_present(*pgd))
+ goto out;
+
+ pud = pud_offset(pgd, addr);
+ if (!pud_present(*pud))
+ goto out;
+
+ pmd = pmd_offset(pud, addr);
+ BUG_ON(pmd_trans_huge(*pmd));
+ if (!pmd_present(*pmd))
+ goto out;
+
+ ptep = pte_offset_map_lock(mm, pmd, addr, &ptl);
+ if (!pte_same(*ptep, orig_pte)) {
+ pte_unmap_unlock(ptep, ptl);
+ goto out;
+ }
+
+ flush_cache_page(vma, addr, pte_pfn(*ptep));
+ ptep_clear_flush(vma, addr, ptep);
+ entry = mk_pte(kpage, vma->vm_page_prot);
+
+ /* special treatment is needed for zero_page */
+ if ((page_to_pfn(kpage) == pksm_zero_pfn) ||
+ (page_to_pfn(kpage) == zero_pfn)) {
+ entry = pte_mkspecial(entry);
+ ksm_pages_zero_sharing++;
+ __inc_zone_page_state(kpage, NR_PKSM_SHARING_PAGES);
+ dec_mm_counter(mm, MM_ANONPAGES);
+ } else {
+ get_page(kpage);
+ page_add_anon_rmap(kpage, vma, addr);
+ }
+
+ set_pte_at_notify(mm, addr, ptep, entry);
+
+ page_remove_rmap(page);
+ if (!page_mapped(page))
+ try_to_free_swap(page);
+ page_cache_release(page);
+
+ pte_unmap_unlock(ptep, ptl);
+
+ /*add ksm_pages_sharing++*/
+ pksm_add_sharing_page_counter(kpage, 1);
+
+ err = 0;
+
+out:
+ return err;
+}
+
+static int page_trans_compound_anon_split(struct page *page)
+{
+ int ret = 0;
+ struct page *transhuge_head = page_trans_compound_anon(page);
+ if (transhuge_head) {
+ /* Get the reference on the head to split it. */
+ if (get_page_unless_zero(transhuge_head)) {
+ /*
+ * Recheck we got the reference while the head
+ * was still anonymous.
+ */
+ if (PageAnon(transhuge_head))
+ ret = split_huge_page(transhuge_head);
+ else
+ /*
+ * Retry later if split_huge_page run
+ * from under us.
+ */
+ ret = 1;
+ put_page(transhuge_head);
+ } else
+ /* Retry later if split_huge_page run from under us. */
+ ret = 1;
+ }
+ return ret;
+}
+
+static inline void set_page_stable_ksm(struct page *page,
+ struct rmap_item *rmap_item)
+{
+ page->mapping = (void *)rmap_item +
+ (PAGE_MAPPING_ANON | PAGE_MAPPING_KSM);
+}
+
+static inline int pksm_flags_can_scan(unsigned long vm_flags)
+{
+ return !(vm_flags & (VM_PFNMAP | VM_IO | VM_DONTEXPAND |
+ VM_DONTDUMP | VM_HUGETLB |
+ VM_NONLINEAR | VM_MIXEDMAP | VM_PAT |
+ VM_SHARED | VM_MAYSHARE | VM_GROWSUP
+ | VM_GROWSDOWN));
+}
+
+/*
+ * What kind of VMA is considered ?
+ */
+static inline int vma_can_enter(struct vm_area_struct *vma)
+{
+ return pksm_flags_can_scan(vma->vm_flags);
+}
+
+static int pksm_rmap_walk(struct page *page, int (*rmap_one)(struct page *,
+ struct vm_area_struct *, unsigned long, void *), void *arg)
+{
+ struct anon_vma *anon_vma;
+ struct anon_vma_chain *avc;
+ unsigned long address;
+ int ret = PKSM_FAULT_DROP;
+
+ VM_BUG_ON(!PageLocked(page));
+
+ if (!PageAnon(page))
+ goto out;
+
+ anon_vma = page_lock_anon_vma_read(page);
+ if (!anon_vma)
+ goto out;
+
+ anon_vma_interval_tree_foreach(avc, &anon_vma->rb_root, 0, ULONG_MAX) {
+ struct vm_area_struct *vma = avc->vma;
+ if (!vma_can_enter(vma))
+ break;
+ address = vma_address(page, vma);
+ if (address == -EFAULT)
+ break;
+ ret = rmap_one(page, vma, address, arg);
+ if (ret != PKSM_FAULT_SUCCESS)
+ break;
+ }
+ page_unlock_anon_vma_read(anon_vma);
+out:
+ return ret;
+}
+
+static int pksm_writepect_pte(struct page *page, struct vm_area_struct *vma,
+ unsigned long addr, void *kpage)
+{
+ int err = PKSM_FAULT_DROP;
+ pte_t orig_pte = __pte(0);
+
+ if ((err = write_protect_page(vma, page, &orig_pte)) == 0) {
+ if (!kpage) {
+ set_page_stable_ksm(page, NULL);
+ mark_page_accessed(page);
+ err = 0;
+ } else if (pages_identical(page, kpage)) {
+ err = replace_page(vma, page, kpage, orig_pte);
+ }
+ }
+
+ if ((vma->vm_flags & VM_LOCKED) && kpage && !err) {
+ munlock_vma_page(page);
+ if (!PageMlocked(kpage)) {
+ unlock_page(page);
+ lock_page(kpage);
+ mlock_vma_page(kpage);
+ page = kpage; /* for final unlock */
+ }
+ }
+
+ return err;
+}
+
+static int try_to_merge_one_anon_page(struct page *page, struct page *kpage)
+{
+ int err =PKSM_FAULT_DROP;
+
+ if (page == kpage) /* ksm page forked */
+ return 0;
+
+ if (PageTransCompound(page) && page_trans_compound_anon_split(page))
+ goto out;
+ BUG_ON(PageTransCompound(page));
+ if (!PageAnon(page))
+ goto out;
+
+ /*
+ * We need the page lock to read a stable PageSwapCache in
+ * write_protect_page(). We use trylock_page() instead of
+ * lock_page() because we don't want to wait here - we
+ * prefer to continue scanning and merging different pages,
+ * then come back to this page when it is unlocked.
+ */
+ if (!trylock_page(page)) {
+ err = PKSM_FAULT_TRY;
+ goto out;
+ }
+
+ /*set write-protect and migrate pte*/
+ err = pksm_rmap_walk(page, pksm_writepect_pte, kpage);
+ if (err != PKSM_FAULT_SUCCESS)
+ goto unlock;
+
+ err = 0;
+
+unlock:
+ unlock_page(page);
+out:
+ return err;
+}
+
+static int try_to_merge_with_pksm_page(struct rmap_item *rmap_item,
+ struct page *page, struct page *kpage)
+{
+ int err = PKSM_FAULT_DROP;
+ BUG_ON (rmap_item != page->pksm);
+ err = try_to_merge_one_anon_page(page, kpage);
+ return err;
+}
+
+/*
+ * try_to_merge_two_pages - take two identical pages and prepare them
+ * to be merged into one page.
+ *
+ * This function returns the kpage if we successfully merged two identical
+ * pages into one ksm page, NULL otherwise.
+ *
+ * Note that this function upgrades page to ksm page: if one of the pages
+ * is already a ksm page, try_to_merge_with_ksm_page should be used.
+ */
+static int try_to_merge_two_pages(struct rmap_item *rmap_item,
+ struct page *page,
+ struct rmap_item *tree_rmap_item,
+ struct page *tree_page)
+{
+ int err = PKSM_FAULT_DROP;
+
+ BUG_ON(!rmap_item);
+ BUG_ON(!page);
+ BUG_ON(!tree_rmap_item);
+ BUG_ON(!tree_page);
+ BUG_ON (rmap_item != page->pksm);
+ BUG_ON (tree_rmap_item != tree_page->pksm);
+
+ err = try_to_merge_with_pksm_page(rmap_item, page, NULL);
+ if (!err) {
+ err = try_to_merge_with_pksm_page(tree_rmap_item,
+ tree_page, page);
+ }
+
+ return err;
+}
+
+static inline int hash_cmp(u32 new_val, u32 node_val)
+{
+ if (new_val > node_val)
+ return 1;
+ else if (new_val < node_val)
+ return -1;
+ else
+ return 0;
+}
+
+/*
+ * stable_tree_search - search for page inside the stable tree
+ *
+ * This function checks if there is a page inside the stable tree
+ * with identical content to the page that we are scanning right now.
+ *
+ * This function returns the stable tree node of identical content if found,
+ * NULL otherwise.
+ */
+static struct page *stable_tree_search(struct page *page)
+{
+ struct rmap_item *rmap_item = (struct rmap_item *)(page->pksm);
+ struct rb_node *parent = NULL;
+ struct rb_node **new;
+
+ if (PageKsm(page))
+ return NULL;
+
+retry:
+ new = &root_stable_tree.rb_node;
+
+ while (*new) {
+ struct rmap_item *tree_rmap_item;
+ struct page *tree_page;
+ int ret;
+
+ cond_resched();
+
+ tree_rmap_item = rb_entry(*new, struct rmap_item, node);
+
+ if ((tree_rmap_item->address & DELLIST_FLAG) ||
+ !tree_rmap_item->page) {
+ remove_rmap_item_from_tree(tree_rmap_item, 0);
+ //printk("%s:%d, page:0x%x:retry...\n",__func__, __LINE__, page);
+ goto retry;
+ }
+
+ /*inc tree_page->_count ref*/
+ tree_page = get_ksm_page(tree_rmap_item);
+ if (!tree_page) {
+ //printk("%s: get_ksm_page error, tree_page=0x%x, mapp=0x%x\n", __func__, tree_page, tree_page->mapping);
+ return NULL;
+ }
+
+#ifdef CONFIG_PKSM_RHASH
+ ret = hash_cmp(rmap_item->checksum, tree_rmap_item->checksum);
+#else
+ ret = memcmp_pages(page, tree_page);
+#endif
+ parent = *new;
+ if (ret < 0) {
+ put_page(tree_page);
+ new = &parent->rb_left;
+ } else if (ret > 0) {
+ put_page(tree_page);
+ new = &parent->rb_right;
+ } else
+ return tree_page;
+ }
+
+ return NULL;
+}
+
+/*
+ * stable_tree_insert - insert rmap_item pointing to new ksm page
+ * into the stable tree.
+ *
+ * This function returns the stable tree node just allocated on success,
+ * NULL otherwise.
+ */
+static int stable_tree_insert(struct rmap_item *rmap_item, struct page *kpage)
+{
+
+ struct rb_node *parent = NULL;
+ struct rb_node **new;
+
+retry:
+ new = &root_stable_tree.rb_node;
+
+ while (*new) {
+ struct rmap_item *tree_rmap_item;
+ struct page *tree_page;
+ int ret;
+
+ cond_resched();
+ tree_rmap_item = rb_entry(*new, struct rmap_item, node);
+
+ /*rmap_item have add to dellist, drop it*/
+ if ((tree_rmap_item->address & DELLIST_FLAG) ||
+ !tree_rmap_item->page) {
+ remove_rmap_item_from_tree(tree_rmap_item, 0);
+ //printk("%s:%d, page:0x%x:retry...\n",__func__, __LINE__, kpage);
+ goto retry;
+ }
+
+ tree_page = get_ksm_page(tree_rmap_item);
+ if (!tree_page)
+ return PKSM_FAULT_DROP;
+
+#ifdef CONFIG_PKSM_RHASH
+ ret = hash_cmp(rmap_item->checksum, tree_rmap_item->checksum);
+#else
+ ret = memcmp_pages(kpage, tree_page);
+#endif
+
+ put_page(tree_page);
+
+ parent = *new;
+ if (ret < 0)
+ new = &parent->rb_left;
+ else if (ret > 0)
+ new = &parent->rb_right;
+ else {
+ /*
+ * It is not a bug that stable_tree_search() didn't
+ * find this node: because at that time our page was
+ * not yet write-protected, so may have changed since.
+ */
+ return PKSM_FAULT_TRY;
+ }
+ }
+
+ BUG_ON (rmap_item->address & UNSTABLE_FLAG);
+ BUG_ON (rmap_item->address & STABLE_FLAG);
+
+ rb_link_node(&rmap_item->node, parent, new);
+ rb_insert_color(&rmap_item->node, &root_stable_tree);
+ set_page_stable_ksm(kpage, rmap_item);
+ rmap_item->address |= STABLE_FLAG;
+
+ return 0;
+}
+
+/*
+ * unstable_tree_search_insert - search for identical page,
+ * else insert rmap_item into the unstable tree.
+ *
+ * This function searches for a page in the unstable tree identical to the
+ * page currently being scanned; and if no identical page is found in the
+ * tree, we insert rmap_item as a new object into the unstable tree.
+ *
+ * This function returns pointer to rmap_item found to be identical
+ * to the currently scanned page, NULL otherwise.
+ *
+ * This function does both searching and inserting, because they share
+ * the same walking algorithm in an rbtree.
+ */
+static
+struct rmap_item *unstable_tree_search_insert(struct rmap_item *rmap_item,
+ struct page *page,
+ struct page **tree_pagep)
+
+{
+
+ struct rb_node *parent = NULL;
+ struct rb_node **new;
+
+ BUG_ON (rmap_item != page->pksm);
+
+retry:
+ new = &root_unstable_tree.rb_node;
+
+ while (*new) {
+ struct rmap_item *tree_rmap_item;
+ struct page *tree_page;
+ int ret;
+
+ cond_resched();
+ tree_rmap_item = rb_entry(*new, struct rmap_item, node);
+
+ BUG_ON (rmap_item != page->pksm);
+
+ if ((tree_rmap_item->address & DELLIST_FLAG) ||
+ !tree_rmap_item->page) {
+ remove_rmap_item_from_tree(tree_rmap_item, 0);
+ //printk("%s:%d, page:0x%x:retry...\n",__func__, __LINE__, page);
+ goto retry;
+ }
+
+ tree_page = get_mergeable_page(tree_rmap_item);
+ if (IS_ERR_OR_NULL(tree_page))
+ return NULL;
+
+ BUG_ON (tree_rmap_item != tree_page->pksm);
+
+ /*
+ * Don't substitute a ksm page for a forked page.
+ */
+ if (page == tree_page) {
+ put_page(tree_page);
+ return NULL;
+ }
+
+#ifdef CONFIG_PKSM_RHASH
+ ret = hash_cmp(rmap_item->checksum, tree_rmap_item->checksum);
+#else
+ ret = memcmp_pages(page, tree_page);
+#endif
+
+ parent = *new;
+ if (ret < 0) {
+ put_page(tree_page);
+ new = &parent->rb_left;
+ } else if (ret > 0) {
+ put_page(tree_page);
+ new = &parent->rb_right;
+ } else {
+ *tree_pagep = tree_page;
+ return tree_rmap_item;
+ }
+ }
+
+ BUG_ON (rmap_item != page->pksm);
+
+ BUG_ON (rmap_item->address & UNSTABLE_FLAG);
+ BUG_ON (rmap_item->address & STABLE_FLAG);
+
+ if (!(rmap_item->address & UNSTABLE_FLAG)) {
+ rmap_item->address |= UNSTABLE_FLAG;
+ rb_link_node(&rmap_item->node, parent, new);
+ rb_insert_color(&rmap_item->node, &root_unstable_tree);
+ ksm_pages_unshared++;
+
+ list_add_tail(&rmap_item->update_list, &unstabletree_checksum_list);
+ rmap_item->address |= CHECKSUM_LIST_FLAG;
+ }
+
+ return NULL;
+}
+
+/*
+ * stable_tree_append - add another rmap_item to the linked list of
+ * rmap_items hanging off a given node of the stable tree, all sharing
+ * the same ksm page.
+ */
+static void stable_tree_append(struct rmap_item *rmap_head, struct page *page)
+{
+
+ struct stable_node_anon *anon_node;
+ struct rmap_item *rmap =(struct rmap_item *)(page->pksm);
+
+ if (!check_valid_rmap_item(rmap_head))
+ return;
+ if (!check_valid_rmap_item(rmap))
+ return;
+
+ anon_node = alloc_stable_anon();
+
+ if (PageKsm(page)) {
+ anon_node->anon_vma = rmap->anon_vma;
+ } else
+ anon_node->anon_vma = page_rmapping(page);
+
+ get_anon_vma(anon_node->anon_vma);
+
+ /*add anon_node to rmap_head->hlist*/
+ hlist_add_head(&anon_node->hlist, &rmap_head->hlist);
+
+ if (!anon_node->hlist.next)
+ ksm_pages_shared++;
+
+}
+
+/*---------------------------- zero page----------------------------*/
+static inline int is_page_full_zero(struct page *page)
+{
+ char *addr;
+ int ret;
+
+ addr = kmap_atomic(page);
+ ret = is_full_zero(addr, PAGE_SIZE);
+ kunmap_atomic(addr);
+
+ return ret;
+}
+
+static inline int find_zero_page_hash(struct page *page)
+{
+#ifdef CONFIG_PKSM_RHASH
+ struct rmap_item *rmap_item= (struct rmap_item *)page->pksm;
+ return (rmap_item->checksum == pksm_zero_random_checksum);
+#else
+ return is_page_full_zero(page);
+#endif
+}
+
+static int pksm_merge_zero_page(struct page *page, struct vm_area_struct *vma,
+ unsigned long addr, void *kpage)
+{
+ int err = PKSM_FAULT_DROP;
+ pte_t orig_pte = __pte(0);
+
+ if ((err = write_protect_page(vma, page, &orig_pte)) == 0) {
+ if (is_page_full_zero(page))
+ err = replace_page(vma, page, kpage, orig_pte);
+ }
+
+ return err;
+}
+
+
+static
+int try_to_merge_zero_page(struct page *page)
+{
+ struct page *zero_page = empty_pksm_zero_page;
+ int err = PKSM_FAULT_DROP;
+
+ if (PageTransCompound(page) && page_trans_compound_anon_split(page))
+ goto out;
+ BUG_ON(PageTransCompound(page));
+ if (!PageAnon(page))
+ goto out;
+
+ /*
+ * We need the page lock to read a stable PageSwapCache in
+ * write_protect_page(). We use trylock_page() instead of
+ * lock_page() because we don't want to wait here - we
+ * prefer to continue scanning and merging different pages,
+ * then come back to this page when it is unlocked.
+ */
+ if (!trylock_page(page))
+ goto out;
+
+ /*set write-protect and migrate pte*/
+ err = pksm_rmap_walk(page, pksm_merge_zero_page, zero_page);
+ if (err != PKSM_FAULT_SUCCESS)
+ goto unlock;
+
+ err = 0;
+
+unlock:
+ unlock_page(page);
+out:
+ return err;
+}
+
+int cmp_and_merge_zero_page(struct page *page)
+{
+ if (find_zero_page_hash(page)) {
+ if (!try_to_merge_zero_page(page)) {
+ return 0;
+ }
+ }
+
+ return -EFAULT;
+}
+
+/*-------------------------------------------------------------------*/
+
+/*
+ * cmp_and_merge_page - first see if page can be merged into the stable tree;
+ * if not, compare checksum to previous and if it's the same, see if page can
+ * be inserted into the unstable tree, or merged with a page already there and
+ * both transferred to the stable tree.
+ *
+ * @page: the page that we are searching identical page to.
+ * @rmap_item: the reverse mapping into the virtual address of this page
+ */
+static int cmp_and_merge_page(struct page *page, struct rmap_item *rmap_item,
+ int init_checksum)
+{
+ struct rmap_item *tree_rmap_item = NULL;
+ struct page *tree_page = NULL;
+ struct page *kpage;
+ int err = PKSM_FAULT_SUCCESS;
+
+ if (!check_valid_rmap_item(rmap_item)) {
+ err = PKSM_FAULT_DROP;
+ goto out;
+ }
+
+ if (PageKsm(page) || in_stable_tree(rmap_item)) {
+ err = PKSM_FAULT_DROP;
+ goto out;
+ }
+
+ remove_rmap_item_from_tree(rmap_item, 0);
+
+ if (init_checksum)
+ rmap_item->checksum = calc_checksum(page);
+
+ if (!cmp_and_merge_zero_page(page))
+ return 0;
+
+ /* We first start with searching the page inside the stable tree */
+ kpage = stable_tree_search(page);
+ if (kpage) {
+ BUG_ON (rmap_item != page->pksm);
+ err = try_to_merge_with_pksm_page(rmap_item, page, kpage);
+ if (!err) {
+ lock_page(kpage);
+ stable_tree_append(page_stable_rmap_item(kpage), page);
+ unlock_page(kpage);
+ err = 0;
+ }
+ put_page(kpage);
+ goto out;
+ }
+
+ tree_rmap_item =
+ unstable_tree_search_insert(rmap_item, page, &tree_page);
+ if (tree_rmap_item) { /*if find the same page on unstable_tree*/
+ /*unstable_rmap page map to the new page*/
+ err = try_to_merge_two_pages(rmap_item, page,
+ tree_rmap_item, tree_page);
+ /*
+ * As soon as we merge this page, we want to remove the
+ * rmap_item of the page we have merged with from the unstable
+ * tree, and insert it instead as new node in the stable tree.
+ */
+ if (!err) {
+ kpage = page;
+ BUG_ON (rmap_item != page->pksm);
+ BUG_ON(rmap_item->page != kpage);
+
+ remove_rmap_item_from_tree(tree_rmap_item, 0);
+
+ /*insert new page to stable tree*/
+ lock_page(kpage);
+ err = stable_tree_insert(rmap_item, kpage);
+ if (!err) {
+ stable_tree_append(rmap_item, kpage);
+
+ lock_page(tree_page);
+ stable_tree_append(rmap_item, tree_page);
+ unlock_page(tree_page);
+ err = 0;
+ }
+ unlock_page(kpage);
+ }
+ put_page(tree_page);
+ }
+
+out:
+ return err;
+}
+
+static unsigned int pksm_calc_update_pages_num(void)
+{
+ unsigned int need_scan = 0;
+
+ if (ksm_pages_unshared < ksm_thread_pages_to_scan)
+ need_scan = ksm_pages_unshared;
+ else
+ need_scan = (ksm_pages_unshared * ksm_thread_sleep_millisecs) /
+ (pksm_unshared_page_update_period *1000);
+
+ return need_scan;
+}
+static void pksm_update_unstable_page_checksum(void)
+{
+ int scan = 0;
+ unsigned long checksum;
+ struct page *page;
+ struct rmap_item *rmap_item, *n_item;
+ LIST_HEAD(l_add);
+ unsigned int need_scan = 0;
+
+ need_scan = pksm_calc_update_pages_num();
+ if (need_scan <= 0)
+ return;
+
+ need_scan = min(need_scan, ksm_thread_pages_to_scan);
+
+ list_for_each_entry_safe(rmap_item, n_item, &unstabletree_checksum_list, update_list) {
+ if (!rmap_item)
+ continue;
+
+ BUG_ON(rmap_item->address & NEWLIST_FLAG);
+ BUG_ON(!(rmap_item->address & CHECKSUM_LIST_FLAG));
+ BUG_ON((rmap_item->address & STABLE_FLAG));
+
+ page = rmap_item->page;
+ if (!check_valid_rmap_item(rmap_item))
+ goto out;
+
+ if (!get_page_unless_zero(page))
+ goto out;
+
+ if (PageLocked(page))
+ goto putpage;
+ if (check_page_dio(page))
+ goto putpage;
+
+ checksum = calc_checksum(page);
+ /*page content change? */
+ if (rmap_item->checksum != checksum) {
+ rmap_item->checksum = checksum;
+ goto re_cmp;
+ } else
+ goto putpage;
+
+re_cmp:
+ remove_rmap_item_from_tree(rmap_item, 0);
+ spin_lock(&pksm_np_list_lock);
+ rmap_item->address &=~ INITCHECKSUM_FLAG;
+ rmap_item->address |= RESCAN_LIST_FLAG;
+ list_add_tail(&rmap_item->list, &pksm_rescan_page_list);
+ spin_unlock(&pksm_np_list_lock);
+
+putpage:
+ put_page(page);
+out:
+ if (scan++ > need_scan)
+ break;
+ cond_resched();
+ }
+}
+
+static void pksm_drop_rmap_item(struct rmap_item *rmap_item)
+{
+ struct page *page = (struct page *)(rmap_item->page);
+
+ if (page && PagePKSM(page))
+ __ClearPagePKSM(page);
+
+ rmap_item->address = 0;
+ remove_rmap_item_from_tree(rmap_item, 1);
+ if (page)
+ page->pksm = NULL;
+ rmap_item->page = NULL;
+ pksm_free_rmap_item(rmap_item);
+}
+
+/**
+ * ksm_do_scan - the ksm scanner main worker function.
+ * @scan_npages - number of pages we want to scan before we return.
+ */
+static void ksm_do_scan(unsigned int scan_npages)
+{
+ struct rmap_item *rmap_item, *n_item;
+ struct page *page;
+ int init_checksum = 0;
+ LIST_HEAD(l_add);
+ int scan = 0;
+
+ spin_lock_irq(&pksm_np_list_lock);
+ list_for_each_entry_safe(rmap_item, n_item, &new_anon_page_list, list) {
+ if (!rmap_item)
+ continue;
+ list_move(&rmap_item->list, &l_add);
+ rmap_item->address &=~ NEWLIST_FLAG;
+ rmap_item->address |= INKSM_FLAG;
+ if (scan++ > scan_npages)
+ break;
+ }
+ spin_unlock_irq(&pksm_np_list_lock);
+
+ scan = 0;
+
+ spin_lock(&pksm_np_list_lock);
+ list_for_each_entry_safe(rmap_item, n_item, &pksm_rescan_page_list, list) {
+ list_del_init(&rmap_item->list);
+ rmap_item->address &=~RESCAN_LIST_FLAG;
+
+ /*page have add del_list? free rmap_item later */
+ if (rmap_item->address & DELLIST_FLAG)
+ continue;
+ list_add_tail(&rmap_item->list, &l_add);
+ if (scan++ > scan_npages)
+ break;
+ }
+ spin_unlock(&pksm_np_list_lock);
+
+ list_for_each_entry_safe(rmap_item, n_item, &l_add, list) {
+ list_del_init(&rmap_item->list);
+
+ /*page have add del_list? free rmap_item later */
+ if (rmap_item->address & DELLIST_FLAG)
+ goto out;
+
+ if (rmap_item->address & INITCHECKSUM_FLAG) {
+ rmap_item->address &=~ INITCHECKSUM_FLAG;
+ init_checksum = 1;
+ } else
+ init_checksum = 0;
+
+ page = rmap_item->page;
+ if (!check_valid_rmap_item(rmap_item))
+ goto out;
+
+ if (!(PageAnon(page)))
+ goto out;
+
+ flush_dcache_page(page);
+
+ /*get ref for new page*/
+ if (!get_page_unless_zero(page))
+ goto out;
+
+ if (PageLocked(page))
+ goto rescan;
+
+ if (check_page_dio(page))
+ goto rescan;
+
+ switch (cmp_and_merge_page(page, rmap_item, init_checksum)) {
+ case PKSM_FAULT_SUCCESS:
+ case PKSM_FAULT_KEEP:
+ goto putpage;
+ case PKSM_FAULT_DROP:
+ pksm_drop_rmap_item(rmap_item);
+ goto putpage;
+ case PKSM_FAULT_TRY:
+ goto rescan;
+ }
+
+rescan:
+ spin_lock(&pksm_np_list_lock);
+ rmap_item->address |= INITCHECKSUM_FLAG;
+ rmap_item->address |= RESCAN_LIST_FLAG;
+ list_add_tail(&rmap_item->list, &pksm_rescan_page_list);
+ spin_unlock(&pksm_np_list_lock);
+
+putpage:
+ put_page(page);
+out:
+ cond_resched();
+ }
+
+ pksm_free_all_rmap_items();
+ pksm_update_unstable_page_checksum();
+}
+
+static int ksmd_should_run(void)
+{
+ return (ksm_run & KSM_RUN_MERGE);
+}
+
+static int ksm_scan_thread(void *nothing)
+{
+ set_freezable();
+ set_user_nice(current, 5);
+
+ while (!kthread_should_stop()) {
+ mutex_lock(&ksm_thread_mutex);
+ if (ksmd_should_run())
+ ksm_do_scan(ksm_thread_pages_to_scan);
+ mutex_unlock(&ksm_thread_mutex);
+
+ try_to_freeze();
+
+ if (ksmd_should_run()) {
+ schedule_timeout_interruptible(
+ msecs_to_jiffies(ksm_thread_sleep_millisecs));
+ } else {
+ wait_event_freezable(ksm_thread_wait,
+ ksmd_should_run() || kthread_should_stop());
+ }
+ }
+ return 0;
+}
+
+int ksm_madvise(struct vm_area_struct *vma, unsigned long start,
+ unsigned long end, int advice, unsigned long *vm_flags)
+{
+
+ switch (advice) {
+ case MADV_MERGEABLE:
+ return 0; /* just ignore the advice */
+ break;
+
+ case MADV_UNMERGEABLE:
+ return 0;
+ break;
+ }
+
+ return 0;
+}
+
+int __ksm_enter(struct mm_struct *mm)
+{
+ struct mm_slot *mm_slot;
+ int needs_wakeup;
+
+ mm_slot = alloc_mm_slot();
+ if (!mm_slot)
+ return -ENOMEM;
+
+ /* Check ksm_run too? Would need tighter locking */
+ needs_wakeup = list_empty(&ksm_mm_head.mm_list);
+
+ spin_lock(&ksm_mmlist_lock);
+ insert_to_mm_slots_hash(mm, mm_slot);
+ /*
+ * Insert just behind the scanning cursor, to let the area settle
+ * down a little; when fork is followed by immediate exec, we don't
+ * want ksmd to waste time setting up and tearing down an rmap_list.
+ */
+ list_add_tail(&mm_slot->mm_list, &ksm_scan.mm_slot->mm_list);
+ spin_unlock(&ksm_mmlist_lock);
+
+ set_bit(MMF_VM_MERGEABLE, &mm->flags);
+ atomic_inc(&mm->mm_count);
+
+ if (needs_wakeup)
+ wake_up_interruptible(&ksm_thread_wait);
+
+ return 0;
+}
+
+void __ksm_exit(struct mm_struct *mm)
+{
+ struct mm_slot *mm_slot;
+ int easy_to_free = 0;
+
+ /*
+ * This process is exiting: if it's straightforward (as is the
+ * case when ksmd was never running), free mm_slot immediately.
+ * Beware: ksm may already have noticed it exiting and freed the slot.
+ */
+
+ spin_lock(&ksm_mmlist_lock);
+ mm_slot = get_mm_slot(mm);
+ if (mm_slot && ksm_scan.mm_slot != mm_slot) {
+ if (!mm_slot->rmap_list) {
+ hlist_del(&mm_slot->link);
+ list_del(&mm_slot->mm_list);
+ easy_to_free = 1;
+ } else {
+ list_move(&mm_slot->mm_list,
+ &ksm_scan.mm_slot->mm_list);
+ }
+ }
+ spin_unlock(&ksm_mmlist_lock);
+
+ if (easy_to_free) {
+ free_mm_slot(mm_slot);
+ clear_bit(MMF_VM_MERGEABLE, &mm->flags);
+ mmdrop(mm);
+ } else if (mm_slot) {
+ down_write(&mm->mmap_sem);
+ up_write(&mm->mmap_sem);
+ }
+}
+
+int pksm_add_new_anon_page(struct page *page, struct rmap_item *rmap_item, struct anon_vma *anon_vma)
+{
+ rmap_item = (struct rmap_item *)rmap_item;
+
+ if (!rmap_item)
+ return -EFAULT;
+ if (!page || !anon_vma)
+ return -EFAULT;
+
+ /*is pksm page ? */
+ if (PagePKSM(page))
+ return -EFAULT;
+
+ if (!(PageAnon(page)))
+ return -EFAULT;
+
+ if (PageKsm(page))
+ return -EFAULT;
+
+ anon_vma = (struct anon_vma *)((unsigned long)anon_vma & ~PAGE_MAPPING_FLAGS);
+ rmap_item->anon_vma = anon_vma;
+
+ SetPagePKSM(page);
+ rmap_item->address |= NEWLIST_FLAG;
+ rmap_item->address |= INITCHECKSUM_FLAG;
+ page->pksm = rmap_item;
+ rmap_item->page = page;
+
+ spin_lock_irq(&pksm_np_list_lock);
+ list_add_tail(&rmap_item->list, &new_anon_page_list);
+ spin_unlock_irq(&pksm_np_list_lock);
+
+ return 0;
+}
+
+int pksm_del_anon_page(struct page *page)
+{
+
+ struct rmap_item *rmap_item = NULL;
+ int map;
+
+ if (!PagePKSM(page))
+ return -EFAULT;
+
+ __ClearPagePKSM(page);
+
+ rmap_item = (struct rmap_item *)(page->pksm);
+ if (!rmap_item)
+ return -EFAULT;
+
+ if (page != rmap_item->page)
+ return -EFAULT;
+
+ map = atomic_read(&rmap_item->_mapcount);
+ if (map > 0) {
+ pksm_del_sharing_page_counter(page, map);
+ }
+
+ page->pksm = NULL;
+ rmap_item->page = NULL;
+
+ spin_lock_irq(&pksm_np_list_lock);
+ if (rmap_item->address & (NEWLIST_FLAG | RESCAN_LIST_FLAG)) {
+ /*rmap_item still at new list, have not added to pksm, directly free rmap_item*/
+ list_del(&rmap_item->list);
+ rmap_item->address = 0;
+ pksm_free_rmap_item(rmap_item);
+ } else {
+ /*rmap_item have added to pksm*/
+ rmap_item->address |= DELLIST_FLAG;
+ list_add_tail(&rmap_item->del_list, &del_anon_page_list);
+ }
+ spin_unlock_irq(&pksm_np_list_lock);
+
+ return 0;
+}
+
+struct page *ksm_does_need_to_copy(struct page *page,
+ struct vm_area_struct *vma, unsigned long address)
+{
+ struct page *new_page;
+
+ new_page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, address);
+ if (new_page) {
+ copy_user_highpage(new_page, page, address, vma);
+
+ SetPageDirty(new_page);
+ __SetPageUptodate(new_page);
+ SetPageSwapBacked(new_page);
+ __set_page_locked(new_page);
+
+ if (page_evictable(new_page))
+ lru_cache_add_anon(new_page);
+ else
+ add_page_to_unevictable_list(new_page);
+ }
+
+ return new_page;
+}
+
+int page_referenced_ksm(struct page *page, struct mem_cgroup *memcg,
+ void *arg)
+{
+ struct stable_node_anon *stable_anon;
+ struct rmap_item *rmap_item;
+ struct page_referenced_arg *pra = arg;
+ int referenced = 0;
+ int search_new_forks = 0;
+
+ VM_BUG_ON(!PageKsm(page));
+ VM_BUG_ON(!PageLocked(page));
+
+ rmap_item = page_stable_rmap_item(page);
+ if (!rmap_item)
+ return 0;
+again:
+ hlist_for_each_entry(stable_anon, &rmap_item->hlist, hlist) {
+ struct anon_vma *anon_vma = stable_anon->anon_vma;
+ struct anon_vma_chain *vmac;
+ struct vm_area_struct *vma;
+
+ anon_vma_lock_write(anon_vma);
+ anon_vma_interval_tree_foreach(vmac, &anon_vma->rb_root, 0, ULONG_MAX) {
+ vma = vmac->vma;
+
+ if (memcg && !mm_match_cgroup(vma->vm_mm, memcg))
+ continue;
+
+ referenced += page_referenced_one(page, vma,
+ rmap_item->address, pra);
+ if (!search_new_forks || !pra->mapcount)
+ break;
+ }
+ anon_vma_unlock_write(anon_vma);
+ if (!pra->mapcount)
+ goto out;
+ }
+ if (!search_new_forks++)
+ goto again;
+out:
+ return referenced;
+}
+
+int try_to_unmap_ksm(struct page *page, void *arg)
+{
+ struct stable_node_anon *stable_anon;
+ struct rmap_item *rmap_item;
+ int ret = SWAP_AGAIN;
+ int search_new_forks = 0;
+
+ VM_BUG_ON(!PageKsm(page));
+ VM_BUG_ON(!PageLocked(page));
+
+ rmap_item = page_stable_rmap_item(page);
+ if (!rmap_item)
+ return SWAP_FAIL;
+again:
+ hlist_for_each_entry(stable_anon, &rmap_item->hlist, hlist) {
+ struct anon_vma *anon_vma = stable_anon->anon_vma;
+ struct anon_vma_chain *vmac;
+ struct vm_area_struct *vma;
+
+ anon_vma_lock_write(anon_vma);
+ anon_vma_interval_tree_foreach(vmac, &anon_vma->rb_root, 0, ULONG_MAX) {
+ vma = vmac->vma;
+
+ ret = try_to_unmap_one(page, vma,
+ rmap_item->address, arg);
+ if (ret != SWAP_AGAIN || !page_mapped(page)) {
+ anon_vma_unlock_write(anon_vma);
+ goto out;
+ }
+ }
+ anon_vma_unlock_write(anon_vma);
+ }
+ if (!search_new_forks++)
+ goto again;
+out:
+ return ret;
+}
+
+#ifdef CONFIG_MEMORY_HOTREMOVE
+static struct stable_node *ksm_check_stable_tree(unsigned long start_pfn,
+ unsigned long end_pfn)
+{
+
+}
+
+static int ksm_memory_callback(struct notifier_block *self,
+ unsigned long action, void *arg)
+{
+ struct memory_notify *mn = arg;
+ struct stable_node *stable_node;
+
+ switch (action) {
+ case MEM_GOING_OFFLINE:
+ /*
+ * Keep it very simple for now: just lock out ksmd and
+ * MADV_UNMERGEABLE while any memory is going offline.
+ * mutex_lock_nested() is necessary because lockdep was alarmed
+ * that here we take ksm_thread_mutex inside notifier chain
+ * mutex, and later take notifier chain mutex inside
+ * ksm_thread_mutex to unlock it. But that's safe because both
+ * are inside mem_hotplug_mutex.
+ */
+ mutex_lock_nested(&ksm_thread_mutex, SINGLE_DEPTH_NESTING);
+ break;
+
+ case MEM_OFFLINE:
+ /*
+ * Most of the work is done by page migration; but there might
+ * be a few stable_nodes left over, still pointing to struct
+ * pages which have been offlined: prune those from the tree.
+ */
+#if 0
+ while ((stable_node = ksm_check_stable_tree(mn->start_pfn,
+ mn->start_pfn + mn->nr_pages)) != NULL)
+ remove_node_from_stable_tree(stable_node);
+ /* fallthrough */
+#endif
+
+ case MEM_CANCEL_OFFLINE:
+ mutex_unlock(&ksm_thread_mutex);
+ break;
+ }
+ return NOTIFY_OK;
+}
+#endif /* CONFIG_MEMORY_HOTREMOVE */
+
+#ifdef CONFIG_SYSFS
+/*
+ * This all compiles without CONFIG_SYSFS, but is a waste of space.
+ */
+
+#define KSM_ATTR_RO(_name) \
+ static struct kobj_attribute _name##_attr = __ATTR_RO(_name)
+#define KSM_ATTR(_name) \
+ static struct kobj_attribute _name##_attr = \
+ __ATTR(_name, 0644, _name##_show, _name##_store)
+
+static ssize_t sleep_millisecs_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return sprintf(buf, "%u\n", ksm_thread_sleep_millisecs);
+}
+
+static ssize_t sleep_millisecs_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ unsigned long msecs;
+ int err;
+
+ err = strict_strtoul(buf, 10, &msecs);
+ if (err || msecs > UINT_MAX)
+ return -EINVAL;
+
+ ksm_thread_sleep_millisecs = msecs;
+
+ return count;
+}
+KSM_ATTR(sleep_millisecs);
+
+static ssize_t period_seconds_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return sprintf(buf, "%u\n", pksm_unshared_page_update_period);
+}
+
+static ssize_t period_seconds_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ unsigned long secs;
+ int err;
+
+ err = strict_strtoul(buf, 10, &secs);
+ if (err || secs > UINT_MAX)
+ return -EINVAL;
+
+ pksm_unshared_page_update_period= secs;
+
+ return count;
+}
+KSM_ATTR(period_seconds);
+
+static ssize_t pages_to_scan_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return sprintf(buf, "%u\n", ksm_thread_pages_to_scan);
+}
+
+static ssize_t pages_to_scan_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ int err;
+ unsigned long nr_pages;
+
+ err = strict_strtoul(buf, 10, &nr_pages);
+ if (err || nr_pages > UINT_MAX)
+ return -EINVAL;
+
+ ksm_thread_pages_to_scan = nr_pages;
+
+ return count;
+}
+KSM_ATTR(pages_to_scan);
+
+static ssize_t run_show(struct kobject *kobj, struct kobj_attribute *attr,
+ char *buf)
+{
+ return sprintf(buf, "%u\n", ksm_run);
+}
+
+static ssize_t run_store(struct kobject *kobj, struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ int err;
+ unsigned long flags;
+
+ err = strict_strtoul(buf, 10, &flags);
+ if (err || flags > UINT_MAX)
+ return -EINVAL;
+ if (flags > KSM_RUN_UNMERGE)
+ return -EINVAL;
+
+ mutex_lock(&ksm_thread_mutex);
+ if (ksm_run != flags) {
+ ksm_run = flags;
+ }
+ mutex_unlock(&ksm_thread_mutex);
+
+ if (flags & KSM_RUN_MERGE)
+ wake_up_interruptible(&ksm_thread_wait);
+
+ return count;
+}
+KSM_ATTR(run);
+
+static ssize_t pages_shared_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return sprintf(buf, "%lu\n", ksm_pages_shared);
+}
+KSM_ATTR_RO(pages_shared);
+
+static ssize_t pages_sharing_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return sprintf(buf, "%lu\n", ksm_pages_sharing+ksm_pages_zero_sharing);
+}
+KSM_ATTR_RO(pages_sharing);
+
+static ssize_t pages_zero_sharing_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return sprintf(buf, "%lu\n", ksm_pages_zero_sharing);
+}
+KSM_ATTR_RO(pages_zero_sharing);
+
+static ssize_t pages_unshared_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return sprintf(buf, "%lu\n", ksm_pages_unshared);
+}
+KSM_ATTR_RO(pages_unshared);
+
+static ssize_t full_scans_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return sprintf(buf, "%lu\n", ksm_scan.seqnr);
+}
+KSM_ATTR_RO(full_scans);
+
+static ssize_t stable_nodes_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return sprintf(buf, "%lu\n", ksm_stable_nodes);
+}
+KSM_ATTR_RO(stable_nodes);
+
+static ssize_t rmap_items_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return sprintf(buf, "%lu\n", ksm_rmap_items);
+}
+KSM_ATTR_RO(rmap_items);
+
+static struct attribute *ksm_attrs[] = {
+ &sleep_millisecs_attr.attr,
+ &period_seconds_attr.attr,
+ &pages_to_scan_attr.attr,
+ &run_attr.attr,
+ &pages_shared_attr.attr,
+ &pages_sharing_attr.attr,
+ &pages_zero_sharing_attr.attr,
+ &pages_unshared_attr.attr,
+ &full_scans_attr.attr,
+ &stable_nodes_attr.attr,
+ &rmap_items_attr.attr,
+ NULL,
+};
+
+static struct attribute_group ksm_attr_group = {
+ .attrs = ksm_attrs,
+ .name = "pksm",
+};
+#endif /* CONFIG_SYSFS */
+
+static inline int init_random_sampling(void)
+{
+ unsigned long i;
+ pksm_random_table = kmalloc(PAGE_SIZE, GFP_KERNEL);
+ if (!pksm_random_table)
+ return -ENOMEM;
+
+ for (i = 0; i < RSAD_STRENGTH_FULL; i++)
+ pksm_random_table[i] = i;
+
+ for (i = 0; i < RSAD_STRENGTH_FULL; i++) {
+ /*init pksm_random_table*/
+ unsigned long rand_range, swap_index, tmp;
+ rand_range = RSAD_STRENGTH_FULL - i;
+ swap_index = i + prandom_u32() % rand_range;
+ tmp = pksm_random_table[i];
+ pksm_random_table[i] = pksm_random_table[swap_index];
+ pksm_random_table[swap_index] = tmp;
+ }
+
+ pksm_zero_random_checksum = pksm_calc_zero_page_checksum(RSAD_STRENGTH_FULL >> 4);
+
+ return 0;
+}
+
+static int __init ksm_init(void)
+{
+ struct task_struct *ksm_thread;
+ int err;
+
+ err = ksm_slab_init();
+ if (err)
+ goto out;
+
+ err = init_random_sampling();
+ if (err)
+ goto out_random;
+
+ ksm_thread = kthread_run(ksm_scan_thread, NULL, "pksmd");
+ if (IS_ERR(ksm_thread)) {
+ printk(KERN_ERR "pksm: creating kthread failed\n");
+ err = PTR_ERR(ksm_thread);
+ goto out_free;
+ }
+
+#ifdef CONFIG_SYSFS
+ err = sysfs_create_group(mm_kobj, &ksm_attr_group);
+ if (err) {
+ printk(KERN_ERR "ksm: register sysfs failed\n");
+ kthread_stop(ksm_thread);
+ goto out_free;
+ }
+#else
+ ksm_run = KSM_RUN_MERGE; /* no way for user to start it */
+
+#endif /* CONFIG_SYSFS */
+
+#ifdef CONFIG_MEMORY_HOTREMOVE
+ /*
+ * Choose a high priority since the callback takes ksm_thread_mutex:
+ * later callbacks could only be taking locks which nest within that.
+ */
+ hotplug_memory_notifier(ksm_memory_callback, 100);
+#endif
+ return 0;
+
+out_free:
+ kfree(pksm_random_table);
+out_random:
+ ksm_slab_free();
+out:
+ return err;
+}
+module_init(ksm_init)
diff --git a/mm/rmap.c b/mm/rmap.c
index 8fc049f..066570d 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -660,12 +660,6 @@ int page_mapped_in_vma(struct page *page, struct vm_area_struct *vma)
return 1;
}
-struct page_referenced_arg {
- int mapcount;
- int referenced;
- unsigned long vm_flags;
- struct mem_cgroup *memcg;
-};
/*
* arg: page_referenced_arg will be passed
*/
diff --git a/mm/swap.c b/mm/swap.c
index 0092097..6a9e2bd 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -37,6 +37,10 @@
#define CREATE_TRACE_POINTS
#include <trace/events/pagemap.h>
+#ifdef CONFIG_PKSM
+#include <linux/ksm.h>
+#endif
+
/* How many pages do we try to swap or page in/out together? */
int page_cluster;
diff --git a/mm/vmstat.c b/mm/vmstat.c
index def5dd2..afde0d7 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -771,6 +771,9 @@ const char * const vmstat_text[] = {
"numa_other",
#endif
"nr_anon_transparent_hugepages",
+#ifdef CONFIG_PKSM
+ "nr_pksm_sharing_pages",
+#endif
"nr_free_cma",
"nr_dirty_threshold",
"nr_dirty_background_threshold",
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment