Re: [PATCH] mm/page_alloc: fix a crash in free_pages_prepare()

From: Andrew Morton
Date: Fri Sep 27 2019 - 17:59:48 EST


On Fri, 27 Sep 2019 17:28:06 -0400 Qian Cai <cai@xxxxxx> wrote:

> >
> > So I think you've moved the arch_free_page() to be after the final
> > thing which can access page contents, yes? If so, we should have a
> > comment in free_pages_prepare() to attmept to prevent this problem from
> > reoccurring as the code evolves?
>
> Right, something like this above arch_free_page() there?
>
> /*
> * It needs to be just above kernel_map_pages(), as s390 could mark those
> * pages unused and then trigger a fault when accessing.
> */

I did this.

--- a/mm/page_alloc.c~mm-page_alloc-fix-a-crash-in-free_pages_prepare-fix
+++ a/mm/page_alloc.c
@@ -1179,7 +1179,13 @@ static __always_inline bool free_pages_p
kernel_init_free_pages(page, 1 << order);

kernel_poison_pages(page, 1 << order, 0);
+ /*
+ * arch_free_page() can make the page's contents inaccessible. s390
+ * does this. So nothing which can access the page's contents should
+ * happen after this.
+ */
arch_free_page(page, order);
+
if (debug_pagealloc_enabled())
kernel_map_pages(page, 1 << order, 0);

_