mm: invoke oom-killer from page fault

Rather than have the pagefault handler kill a process directly if it gets
a VM_FAULT_OOM, have it call into the OOM killer.

With increasingly sophisticated oom behaviour (cpusets, memory cgroups,
oom killing throttling, oom priority adjustment or selective disabling,
panic on oom, etc), it's silly to unconditionally kill the faulting
process at page fault time.  Create a hook for pagefault oom path to call
into instead.

Only converted x86 and uml so far.

[akpm@linux-foundation.org: make __out_of_memory() static]
[akpm@linux-foundation.org: fix comment]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Jeff Dike <jdike@addtoit.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

4 files changed, 84 insertions, 63 deletions

diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c
index 44e4904..7384d8a 100644
--- a/arch/um/kernel/trap.c
+++ b/arch/um/kernel/trap.c
@@ -64,11 +64,10 @@ good_area:
 
 	do {
 		int fault;
-survive:
+
 		fault = handle_mm_fault(mm, vma, address, is_write);
 		if (unlikely(fault & VM_FAULT_ERROR)) {
 			if (fault & VM_FAULT_OOM) {
-				err = -ENOMEM;
 				goto out_of_memory;
 			} else if (fault & VM_FAULT_SIGBUS) {
 				err = -EACCES;
@@ -104,18 +103,14 @@ out:
 out_nosemaphore:
 	return err;
 
-/*
- * We ran out of memory, or some other thing happened to us that made
- * us unable to handle the page fault gracefully.
- */
 out_of_memory:
-	if (is_global_init(current)) {
-		up_read(&mm->mmap_sem);
-		yield();
-		down_read(&mm->mmap_sem);
-		goto survive;
-	}
-	goto out;
+	/*
+	 * We ran out of memory, call the OOM killer, and return the userspace
+	 * (which will retry the fault, or kill us if we got oom-killed).
+	 */
+	up_read(&mm->mmap_sem);
+	pagefault_out_of_memory();
+	return 0;
 }
 
 static void bad_segv(struct faultinfo fi, unsigned long ip)
@@ -214,9 +209,6 @@ unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user,
 		si.si_addr = (void __user *)address;
 		current->thread.arch.faultinfo = fi;
 		force_sig_info(SIGBUS, &si, current);
-	} else if (err == -ENOMEM) {
-		printk(KERN_INFO "VM: killing process %s\n", current->comm);
-		do_exit(SIGKILL);
 	} else {
 		BUG_ON(err != -EFAULT);
 		si.si_signo = SIGSEGV;
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 57ec8c8..9e268b6b 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -667,7 +667,6 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code)
 	if (unlikely(in_atomic() || !mm))
 		goto bad_area_nosemaphore;
 
-again:
 	/*
 	 * When running in the kernel we expect faults to occur only to
 	 * addresses in user space.  All other faults represent errors in the
@@ -859,25 +858,14 @@ no_context:
 	oops_end(flags, regs, sig);
 #endif
 
-/*
- * We ran out of memory, or some other thing happened to us that made
- * us unable to handle the page fault gracefully.
- */
 out_of_memory:
+	/*
+	 * We ran out of memory, call the OOM killer, and return the userspace
+	 * (which will retry the fault, or kill us if we got oom-killed).
+	 */
 	up_read(&mm->mmap_sem);
-	if (is_global_init(tsk)) {
-		yield();
-		/*
-		 * Re-lookup the vma - in theory the vma tree might
-		 * have changed:
-		 */
-		goto again;
-	}
-
-	printk("VM: killing process %s\n", tsk->comm);
-	if (error_code & PF_USER)
-		do_group_exit(SIGKILL);
-	goto no_context;
+	pagefault_out_of_memory();
+	return;
 
 do_sigbus:
 	up_read(&mm->mmap_sem);
diff --git a/include/linux/mm.h b/include/linux/mm.h
index aaa8b84..4a3d28c 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -717,6 +717,11 @@ static inline int page_mapped(struct page *page)
 
 #define VM_FAULT_ERROR	(VM_FAULT_OOM | VM_FAULT_SIGBUS)
 
+/*
+ * Can be called by the pagefault handler when it gets a VM_FAULT_OOM.
+ */
+extern void pagefault_out_of_memory(void);
+
 #define offset_in_page(p)	((unsigned long)(p) & ~PAGE_MASK)
 
 extern void show_free_areas(void);
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 558f9af..c592965 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -509,6 +509,69 @@ void clear_zonelist_oom(struct zonelist *zonelist, gfp_t gfp_mask)
 	spin_unlock(&zone_scan_mutex);
 }
 
+/*
+ * Must be called with tasklist_lock held for read.
+ */
+static void __out_of_memory(gfp_t gfp_mask, int order)
+{
+	if (sysctl_oom_kill_allocating_task) {
+		oom_kill_process(current, gfp_mask, order, 0, NULL,
+				"Out of memory (oom_kill_allocating_task)");
+
+	} else {
+		unsigned long points;
+		struct task_struct *p;
+
+retry:
+		/*
+		 * Rambo mode: Shoot down a process and hope it solves whatever
+		 * issues we may have.
+		 */
+		p = select_bad_process(&points, NULL);
+
+		if (PTR_ERR(p) == -1UL)
+			return;
+
+		/* Found nothing?!?! Either we hang forever, or we panic. */
+		if (!p) {
+			read_unlock(&tasklist_lock);
+			panic("Out of memory and no killable processes...\n");
+		}
+
+		if (oom_kill_process(p, gfp_mask, order, points, NULL,
+				     "Out of memory"))
+			goto retry;
+	}
+}
+
+/*
+ * pagefault handler calls into here because it is out of memory but
+ * doesn't know exactly how or why.
+ */
+void pagefault_out_of_memory(void)
+{
+	unsigned long freed = 0;
+
+	blocking_notifier_call_chain(&oom_notify_list, 0, &freed);
+	if (freed > 0)
+		/* Got some memory back in the last second. */
+		return;
+
+	if (sysctl_panic_on_oom)
+		panic("out of memory from page fault. panic_on_oom is selected.\n");
+
+	read_lock(&tasklist_lock);
+	__out_of_memory(0, 0); /* unknown gfp_mask and order */
+	read_unlock(&tasklist_lock);
+
+	/*
+	 * Give "p" a good chance of killing itself before we
+	 * retry to allocate memory.
+	 */
+	if (!test_thread_flag(TIF_MEMDIE))
+		schedule_timeout_uninterruptible(1);
+}
+
 /**
  * out_of_memory - kill the "best" process when we run out of memory
  * @zonelist: zonelist pointer
@@ -522,8 +585,6 @@ void clear_zonelist_oom(struct zonelist *zonelist, gfp_t gfp_mask)
  */
 void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order)
 {
-	struct task_struct *p;
-	unsigned long points = 0;
 	unsigned long freed = 0;
 	enum oom_constraint constraint;
 
@@ -544,7 +605,7 @@ void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order)
 
 	switch (constraint) {
 	case CONSTRAINT_MEMORY_POLICY:
-		oom_kill_process(current, gfp_mask, order, points, NULL,
+		oom_kill_process(current, gfp_mask, order, 0, NULL,
 				"No available memory (MPOL_BIND)");
 		break;
 
@@ -553,35 +614,10 @@ void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order)
 			panic("out of memory. panic_on_oom is selected\n");
 		/* Fall-through */
 	case CONSTRAINT_CPUSET:
-		if (sysctl_oom_kill_allocating_task) {
-			oom_kill_process(current, gfp_mask, order, points, NULL,
-					"Out of memory (oom_kill_allocating_task)");
-			break;
-		}
-retry:
-		/*
-		 * Rambo mode: Shoot down a process and hope it solves whatever
-		 * issues we may have.
-		 */
-		p = select_bad_process(&points, NULL);
-
-		if (PTR_ERR(p) == -1UL)
-			goto out;
-
-		/* Found nothing?!?! Either we hang forever, or we panic. */
-		if (!p) {
-			read_unlock(&tasklist_lock);
-			panic("Out of memory and no killable processes...\n");
-		}
-
-		if (oom_kill_process(p, gfp_mask, order, points, NULL,
-				     "Out of memory"))
-			goto retry;
-
+		__out_of_memory(gfp_mask, order);
 		break;
 	}
 
-out:
 	read_unlock(&tasklist_lock);
 
 	/*