mm: convert mm->cpu_vm_cpumask into cpumask_var_t

cpumask_t is very big struct and cpu_vm_mask is placed wrong position.
It might lead to reduce cache hit ratio.

This patch has two change.
1) Move the place of cpumask into last of mm_struct. Because usually cpumask
   is accessed only front bits when the system has cpu-hotplug capability
2) Convert cpu_vm_mask into cpumask_var_t. It may help to reduce memory
   footprint if cpumask_size() will use nr_cpumask_bits properly in future.

In addition, this patch change the name of cpu_vm_mask with cpu_vm_mask_var.
It may help to detect out of tree cpu_vm_mask users.

This patch has no functional change.

[[email protected]: build fix]
[[email protected]: coding-style fixes]
Signed-off-by: KOSAKI Motohiro <[email protected]>
Cc: David Howells <[email protected]>
Cc: Koichi Yasutake <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Chris Metcalf <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

Documentation/cachetlb.txt

@@ -16,7 +16,7 @@ on all processors in the system.  Don't let this scare you into
 thinking SMP cache/tlb flushing must be so inefficient, this is in
 fact an area where many optimizations are possible.  For example,
 if it can be proven that a user address space has never executed
-on a cpu (see vma->cpu_vm_mask), one need not perform a flush
+on a cpu (see mm_cpumask()), one need not perform a flush
 for this address space on that cpu.
 
 First, the TLB flushing interfaces, since they are the simplest.  The

arch/x86/kernel/tboot.c

@@ -110,7 +110,6 @@ static struct mm_struct tboot_mm = {
 	.mmap_sem       = __RWSEM_INITIALIZER(init_mm.mmap_sem),
 	.page_table_lock =  __SPIN_LOCK_UNLOCKED(init_mm.page_table_lock),
 	.mmlist         = LIST_HEAD_INIT(init_mm.mmlist),
-	.cpu_vm_mask    = CPU_MASK_ALL,
 };
 
 static inline void switch_to_tboot_pt(void)

include/linux/mm_types.h

@@ -265,8 +265,6 @@ struct mm_struct {
 
 	struct linux_binfmt *binfmt;
 
-	cpumask_t cpu_vm_mask;
-
 	/* Architecture-specific MM context */
 	mm_context_t context;
 
@@ -316,9 +314,14 @@ struct mm_struct {
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
 	pgtable_t pmd_huge_pte; /* protected by page_table_lock */
 #endif
+
+	cpumask_var_t cpu_vm_mask_var;
 };
 
 /* Future-safe accessor for struct mm_struct's cpu_vm_mask. */
-#define mm_cpumask(mm) (&(mm)->cpu_vm_mask)
+static inline cpumask_t *mm_cpumask(struct mm_struct *mm)
+{
+	return mm->cpu_vm_mask_var;
+}
 
 #endif /* _LINUX_MM_TYPES_H */

include/linux/sched.h

@@ -2176,6 +2176,7 @@ static inline void mmdrop(struct mm_struct * mm)
 	if (unlikely(atomic_dec_and_test(&mm->mm_count)))
 		__mmdrop(mm);
 }
+extern int mm_init_cpumask(struct mm_struct *mm, struct mm_struct *oldmm);
 
 /* mmput gets rid of the mappings and all user-space */
 extern void mmput(struct mm_struct *);

init/main.c

@@ -509,6 +509,8 @@ asmlinkage void __init start_kernel(void)
 	sort_main_extable();
 	trap_init();
 	mm_init();
+	BUG_ON(mm_init_cpumask(&init_mm, 0));
+
 	/*
 	 * Set up the scheduler prior starting any interrupts (such as the
 	 * timer interrupt). Full topology setup happens at smp_init()

kernel/fork.c

@@ -485,6 +485,20 @@ static void mm_init_aio(struct mm_struct *mm)
 #endif
 }
 
+int mm_init_cpumask(struct mm_struct *mm, struct mm_struct *oldmm)
+{
+#ifdef CONFIG_CPUMASK_OFFSTACK
+	if (!alloc_cpumask_var(&mm->cpu_vm_mask_var, GFP_KERNEL))
+		return -ENOMEM;
+
+	if (oldmm)
+		cpumask_copy(mm_cpumask(mm), mm_cpumask(oldmm));
+	else
+		memset(mm_cpumask(mm), 0, cpumask_size());
+#endif
+	return 0;
+}
+
 static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p)
 {
 	atomic_set(&mm->mm_users, 1);
@@ -521,10 +535,20 @@ struct mm_struct * mm_alloc(void)
 	struct mm_struct * mm;
 
 	mm = allocate_mm();
-	if (mm) {
-		memset(mm, 0, sizeof(*mm));
-		mm = mm_init(mm, current);
+	if (!mm)
+		return NULL;
+
+	memset(mm, 0, sizeof(*mm));
+	mm = mm_init(mm, current);
+	if (!mm)
+		return NULL;
+
+	if (mm_init_cpumask(mm, NULL)) {
+		mm_free_pgd(mm);
+		free_mm(mm);
+		return NULL;
 	}
+
 	return mm;
 }
 
@@ -536,6 +560,7 @@ struct mm_struct * mm_alloc(void)
 void __mmdrop(struct mm_struct *mm)
 {
 	BUG_ON(mm == &init_mm);
+	free_cpumask_var(mm->cpu_vm_mask_var);
 	mm_free_pgd(mm);
 	destroy_context(mm);
 	mmu_notifier_mm_destroy(mm);
@@ -690,6 +715,9 @@ struct mm_struct *dup_mm(struct task_struct *tsk)
 	if (!mm_init(mm, tsk))
 		goto fail_nomem;
 
+	if (mm_init_cpumask(mm, oldmm))
+		goto fail_nocpumask;
+
 	if (init_new_context(tsk, mm))
 		goto fail_nocontext;
 
@@ -716,6 +744,9 @@ struct mm_struct *dup_mm(struct task_struct *tsk)
 	return NULL;
 
 fail_nocontext:
+	free_cpumask_var(mm->cpu_vm_mask_var);
+
+fail_nocpumask:
 	/*
 	 * If init_new_context() failed, we cannot use mmput() to free the mm
 	 * because it calls destroy_context()

mm/init-mm.c

@@ -21,6 +21,5 @@ struct mm_struct init_mm = {
 	.mmap_sem	= __RWSEM_INITIALIZER(init_mm.mmap_sem),
 	.page_table_lock =  __SPIN_LOCK_UNLOCKED(init_mm.page_table_lock),
 	.mmlist		= LIST_HEAD_INIT(init_mm.mmlist),
-	.cpu_vm_mask	= CPU_MASK_ALL,
 	INIT_MM_CONTEXT(init_mm)
 };