On Wed, May 29, 2019 at 11:05:48AM -0700, Ira Weiny wrote:

On Fri, May 24, 2019 at 01:50:45PM -0400, Daniel Jordan wrote:

quoted

+static inline int account_locked_vm(struct mm_struct *mm, unsigned long pages,
+				    bool inc)
+{
+	int ret;
+
+	if (pages == 0 || !mm)
+		return 0;
+
+	down_write(&mm->mmap_sem);
+	ret = __account_locked_vm(mm, pages, inc, current,
+				  capable(CAP_IPC_LOCK));
+	up_write(&mm->mmap_sem);
+
+	return ret;
+}
+

...snip...

quoted

+/**
+ * __account_locked_vm - account locked pages to an mm's locked_vm
+ * @mm:          mm to account against, may be NULL

This kernel doc is wrong.  You dereference mm straight away...

...snip...

quoted

+
+	locked_vm = mm->locked_vm;

here...

Perhaps the comment was meant to document account_locked_vm()?

Yes, the comment got out of sync when I moved the !mm check outside
__account_locked_vm.  Thanks for catching, will fix.