---

yaml
---
r: 164402
b: refs/heads/master
c: f8af4da
h: refs/heads/master
v: v3

[refs]

@@ -1,2 +1,2 @@
 ---
-refs/heads/master: d19f352484467a5e518639ddff0554669c10ffab
+refs/heads/master: f8af4da3b4c14e7267c4ffb952079af3912c51c5

trunk/include/linux/ksm.h

@@ -0,0 +1,50 @@
+#ifndef __LINUX_KSM_H
+#define __LINUX_KSM_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().
+ */
+
+#include <linux/bitops.h>
+#include <linux/mm.h>
+#include <linux/sched.h>
+
+#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);
+}
+#else  /* !CONFIG_KSM */
+
+static inline int ksm_madvise(struct vm_area_struct *vma, unsigned long start,
+		unsigned long end, int advice, unsigned long *vm_flags)
+{
+	return 0;
+}
+
+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_KSM */
+
+#endif

trunk/include/linux/mm.h

@@ -103,6 +103,7 @@ extern unsigned int kobjsize(const void *objp);
 #define VM_MIXEDMAP	0x10000000	/* Can contain "struct page" and pure PFN pages */
 #define VM_SAO		0x20000000	/* Strong Access Ordering (powerpc) */
 #define VM_PFN_AT_MMAP	0x40000000	/* PFNMAP vma that is fully mapped at mmap time */
+#define VM_MERGEABLE	0x80000000	/* KSM may merge identical pages */
 
 #ifndef VM_STACK_DEFAULT_FLAGS		/* arch can override this */
 #define VM_STACK_DEFAULT_FLAGS VM_DATA_DEFAULT_FLAGS

trunk/include/linux/sched.h

@@ -434,7 +434,9 @@ extern int get_dumpable(struct mm_struct *mm);
 /* dumpable bits */
 #define MMF_DUMPABLE      0  /* core dump is permitted */
 #define MMF_DUMP_SECURELY 1  /* core file is readable only by root */
+
 #define MMF_DUMPABLE_BITS 2
+#define MMF_DUMPABLE_MASK ((1 << MMF_DUMPABLE_BITS) - 1)
 
 /* coredump filter bits */
 #define MMF_DUMP_ANON_PRIVATE	2
@@ -444,6 +446,7 @@ extern int get_dumpable(struct mm_struct *mm);
 #define MMF_DUMP_ELF_HEADERS	6
 #define MMF_DUMP_HUGETLB_PRIVATE 7
 #define MMF_DUMP_HUGETLB_SHARED  8
+
 #define MMF_DUMP_FILTER_SHIFT	MMF_DUMPABLE_BITS
 #define MMF_DUMP_FILTER_BITS	7
 #define MMF_DUMP_FILTER_MASK \
@@ -457,6 +460,10 @@ extern int get_dumpable(struct mm_struct *mm);
 #else
 # define MMF_DUMP_MASK_DEFAULT_ELF	0
 #endif
+					/* leave room for more dump flags */
+#define MMF_VM_MERGEABLE	16	/* KSM may merge identical pages */
+
+#define MMF_INIT_MASK		(MMF_DUMPABLE_MASK | MMF_DUMP_FILTER_MASK)
 
 struct sighand_struct {
 	atomic_t		count;

trunk/kernel/fork.c

@@ -49,6 +49,7 @@
 #include <linux/ftrace.h>
 #include <linux/profile.h>
 #include <linux/rmap.h>
+#include <linux/ksm.h>
 #include <linux/acct.h>
 #include <linux/tsacct_kern.h>
 #include <linux/cn_proc.h>
@@ -299,6 +300,9 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm)
 	rb_link = &mm->mm_rb.rb_node;
 	rb_parent = NULL;
 	pprev = &mm->mmap;
+	retval = ksm_fork(mm, oldmm);
+	if (retval)
+		goto out;
 
 	for (mpnt = oldmm->mmap; mpnt; mpnt = mpnt->vm_next) {
 		struct file *file;
@@ -435,7 +439,8 @@ static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p)
 	atomic_set(&mm->mm_count, 1);
 	init_rwsem(&mm->mmap_sem);
 	INIT_LIST_HEAD(&mm->mmlist);
-	mm->flags = (current->mm) ? current->mm->flags : default_dump_filter;
+	mm->flags = (current->mm) ?
+		(current->mm->flags & MMF_INIT_MASK) : default_dump_filter;
 	mm->core_state = NULL;
 	mm->nr_ptes = 0;
 	set_mm_counter(mm, file_rss, 0);
@@ -496,6 +501,7 @@ void mmput(struct mm_struct *mm)
 
 	if (atomic_dec_and_test(&mm->mm_users)) {
 		exit_aio(mm);
+		ksm_exit(mm);
 		exit_mmap(mm);
 		set_mm_exe_file(mm, NULL);
 		if (!list_empty(&mm->mmlist)) {

trunk/mm/Kconfig

@@ -214,6 +214,17 @@ config HAVE_MLOCKED_PAGE_BIT
 config MMU_NOTIFIER
 	bool
 
+config KSM
+	bool "Enable KSM for page merging"
+	depends on MMU
+	help
+	  Enable Kernel Samepage Merging: KSM periodically scans those areas
+	  of an application's address space that an app has advised may be
+	  mergeable.  When it finds pages of identical content, it replaces
+	  the many instances by a single resident page with that content, so
+	  saving memory until one or another app needs to modify the content.
+	  Recommended for use with KVM, or with other duplicative applications.
+
 config DEFAULT_MMAP_MIN_ADDR
         int "Low address space to protect from user allocation"
         default 4096

trunk/mm/Makefile

@@ -25,6 +25,7 @@ obj-$(CONFIG_SPARSEMEM_VMEMMAP) += sparse-vmemmap.o
 obj-$(CONFIG_TMPFS_POSIX_ACL) += shmem_acl.o
 obj-$(CONFIG_SLOB) += slob.o
 obj-$(CONFIG_MMU_NOTIFIER) += mmu_notifier.o
+obj-$(CONFIG_KSM) += ksm.o
 obj-$(CONFIG_PAGE_POISONING) += debug-pagealloc.o
 obj-$(CONFIG_SLAB) += slab.o
 obj-$(CONFIG_SLUB) += slub.o

trunk/mm/ksm.c

@@ -0,0 +1,56 @@
+/*
+ * Initial dummy version just to illustrate KSM's interface to other files.
+ */
+
+#include <linux/errno.h>
+#include <linux/mman.h>
+#include <linux/ksm.h>
+
+int ksm_madvise(struct vm_area_struct *vma, unsigned long start,
+		unsigned long end, int advice, unsigned long *vm_flags)
+{
+	struct mm_struct *mm = vma->vm_mm;
+
+	switch (advice) {
+	case MADV_MERGEABLE:
+		/*
+		 * Be somewhat over-protective for now!
+		 */
+		if (*vm_flags & (VM_MERGEABLE | VM_SHARED  | VM_MAYSHARE   |
+				 VM_PFNMAP    | VM_IO      | VM_DONTEXPAND |
+				 VM_RESERVED  | VM_HUGETLB | VM_INSERTPAGE |
+				 VM_MIXEDMAP  | VM_SAO))
+			return 0;		/* just ignore the advice */
+
+		if (!test_bit(MMF_VM_MERGEABLE, &mm->flags))
+			if (__ksm_enter(mm) < 0)
+				return -EAGAIN;
+
+		*vm_flags |= VM_MERGEABLE;
+		break;
+
+	case MADV_UNMERGEABLE:
+		if (!(*vm_flags & VM_MERGEABLE))
+			return 0;		/* just ignore the advice */
+
+		/* Unmerge any merged pages here */
+
+		*vm_flags &= ~VM_MERGEABLE;
+		break;
+	}
+
+	return 0;
+}
+
+int __ksm_enter(struct mm_struct *mm)
+{
+	/* Allocate a structure to track mm and link it into KSM's list */
+	set_bit(MMF_VM_MERGEABLE, &mm->flags);
+	return 0;
+}
+
+void __ksm_exit(struct mm_struct *mm)
+{
+	/* Unlink and free all KSM's structures which track this mm */
+	clear_bit(MMF_VM_MERGEABLE, &mm->flags);
+}

trunk/mm/madvise.c

@@ -11,6 +11,7 @@
 #include <linux/mempolicy.h>
 #include <linux/hugetlb.h>
 #include <linux/sched.h>
+#include <linux/ksm.h>
 
 /*
  * Any behaviour which results in changes to the vma->vm_flags needs to
@@ -63,6 +64,12 @@ static long madvise_behavior(struct vm_area_struct * vma,
 		}
 		new_flags &= ~VM_DONTCOPY;
 		break;
+	case MADV_MERGEABLE:
+	case MADV_UNMERGEABLE:
+		error = ksm_madvise(vma, start, end, behavior, &new_flags);
+		if (error)
+			goto out;
+		break;
 	}
 
 	if (new_flags == vma->vm_flags) {
@@ -239,6 +246,10 @@ madvise_behavior_valid(int behavior)
 	case MADV_REMOVE:
 	case MADV_WILLNEED:
 	case MADV_DONTNEED:
+#ifdef CONFIG_KSM
+	case MADV_MERGEABLE:
+	case MADV_UNMERGEABLE:
+#endif
 		return 1;
 
 	default:
@@ -273,6 +284,9 @@ madvise_behavior_valid(int behavior)
  *  MADV_DONTFORK - omit this area from child's address space when forking:
  *		typically, to avoid COWing pages pinned by get_user_pages().
  *  MADV_DOFORK - cancel MADV_DONTFORK: no longer omit this area when forking.
+ *  MADV_MERGEABLE - the application recommends that KSM try to merge pages in
+ *		this area with pages of identical content from other such areas.
+ *  MADV_UNMERGEABLE- cancel MADV_MERGEABLE: no longer merge pages with others.
  *
  * return values:
  *  zero    - success