Re: Another VM race? (was: page_launder() bug)

[Mikulas Patocka]

> > CPU 0				CPU 1
> > is executing the code marked	is executing try_to_free_buffers on
> > above with ^^^^^^^:		the same page (it can be, because CPU 0
> > 				did not lock the page)
> > 
> > (page->buffers &&
> > 
> > 				page->buffers = NULL
> > 
> > MAJOR(page->buffers->b_dev) == 
> > 	RAMDISK_MAJOR)) ===> Oops, NULL pointer dereference!
> > 
> > 
> > 
> > Maybe compiler CSE optimization will eliminate the double load of
> > page->buffers, but we must not rely on it. If the compiler doesn't
> > optimize it, it can produce random oopses.
> 
> You're right, this should be fixed. Do you happen to have a
> patch ? ;)

You can apply this one.

Mikulas

--- linux/mm/vmscan.c_	Mon May 14 11:41:42 2001
+++ linux/mm/vmscan.c	Mon May 14 11:44:54 2001
@@ -454,8 +454,7 @@
 
 		/* Page is or was in use?  Move it to the active list. */
 		if (PageTestandClearReferenced(page) || page->age > 0 ||
-				(!page->buffers && page_count(page) > 1) ||
-				page_ramdisk(page)) {
+				(!page->buffers && page_count(page) > 1)) {
 			del_page_from_inactive_dirty_list(page);
 			add_page_to_active_list(page);
 			continue;
@@ -470,6 +469,9 @@
 			list_add(page_lru, &inactive_dirty_list);
 			continue;
 		}
+
+		/* M.P.: We must not call page_ramdisk for unlocked page */
+		if (page_ramdisk(page)) goto page_active;
 
 		/*
 		 * Dirty swap-cache page? Write it out if