From: Andrey Konovalov <[email protected]>
The implementation of page_alloc poisoning sampling assumed that
tag_clear_highpage resets page tags for __GFP_ZEROTAGS allocations.
However, this is no longer the case since commit 70c248aca9e7
("mm: kasan: Skip unpoisoning of user pages").
This leads to kernel crashes when MTE-enabled userspace mappings are
used with Hardware Tag-Based KASAN enabled.
Reset page tags for __GFP_ZEROTAGS allocations in post_alloc_hook().
Also clarify and fix related comments.
Reported-by: Peter Collingbourne <[email protected]>
Tested-by: Peter Collingbourne <[email protected]>
Signed-off-by: Andrey Konovalov <[email protected]>
---
mm/page_alloc.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 5514d84cc712..370d4f2c0276 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2471,7 +2471,7 @@ inline void post_alloc_hook(struct page *page, unsigned int order,
bool init = !want_init_on_free() && want_init_on_alloc(gfp_flags) &&
!should_skip_init(gfp_flags);
bool zero_tags = init && (gfp_flags & __GFP_ZEROTAGS);
- bool reset_tags = !zero_tags;
+ bool reset_tags = true;
int i;
set_page_private(page, 0);
@@ -2498,7 +2498,7 @@ inline void post_alloc_hook(struct page *page, unsigned int order,
* (which happens only when memory should be initialized as well).
*/
if (zero_tags) {
- /* Initialize both memory and tags. */
+ /* Initialize both memory and memory tags. */
for (i = 0; i != 1 << order; ++i)
tag_clear_highpage(page + i);
@@ -2516,14 +2516,14 @@ inline void post_alloc_hook(struct page *page, unsigned int order,
} else {
/*
* KASAN decided to exclude this allocation from being
- * poisoned due to sampling. Skip poisoning as well.
+ * unpoisoned due to sampling. Skip poisoning as well.
*/
SetPageSkipKASanPoison(page);
}
}
/*
- * If memory tags have not been set, reset the page tags to ensure
- * page_address() dereferencing does not fault.
+ * If memory tags have not been set by KASAN, reset the page tags to
+ * ensure page_address() dereferencing does not fault.
*/
if (reset_tags) {
for (i = 0; i != 1 << order; ++i)
--
2.25.1
On Tue, 24 Jan 2023 21:35:26 +0100 [email protected] wrote:
> The implementation of page_alloc poisoning sampling assumed that
> tag_clear_highpage resets page tags for __GFP_ZEROTAGS allocations.
> However, this is no longer the case since commit 70c248aca9e7
> ("mm: kasan: Skip unpoisoning of user pages").
>
> This leads to kernel crashes when MTE-enabled userspace mappings are
> used with Hardware Tag-Based KASAN enabled.
>
> Reset page tags for __GFP_ZEROTAGS allocations in post_alloc_hook().
>
> Also clarify and fix related comments.
I assume this is a fix against 44383cef54c0 ("kasan: allow sampling
page_alloc allocations for HW_TAGS") which is presently in mm-stable,
yes?
On Tue, Jan 24, 2023 at 9:45 PM Andrew Morton <[email protected]> wrote:
>
> On Tue, 24 Jan 2023 21:35:26 +0100 [email protected] wrote:
>
> > The implementation of page_alloc poisoning sampling assumed that
> > tag_clear_highpage resets page tags for __GFP_ZEROTAGS allocations.
> > However, this is no longer the case since commit 70c248aca9e7
> > ("mm: kasan: Skip unpoisoning of user pages").
> >
> > This leads to kernel crashes when MTE-enabled userspace mappings are
> > used with Hardware Tag-Based KASAN enabled.
> >
> > Reset page tags for __GFP_ZEROTAGS allocations in post_alloc_hook().
> >
> > Also clarify and fix related comments.
>
> I assume this is a fix against 44383cef54c0 ("kasan: allow sampling
> page_alloc allocations for HW_TAGS") which is presently in mm-stable,
> yes?
Correct. I assumed I shouldn't include a Fixes tag, as the patch is
not in the mainline.
On Tue, 24 Jan 2023 21:46:51 +0100 Andrey Konovalov <[email protected]> wrote:
> On Tue, Jan 24, 2023 at 9:45 PM Andrew Morton <[email protected]> wrote:
> >
> > On Tue, 24 Jan 2023 21:35:26 +0100 [email protected] wrote:
> >
> > > The implementation of page_alloc poisoning sampling assumed that
> > > tag_clear_highpage resets page tags for __GFP_ZEROTAGS allocations.
> > > However, this is no longer the case since commit 70c248aca9e7
> > > ("mm: kasan: Skip unpoisoning of user pages").
> > >
> > > This leads to kernel crashes when MTE-enabled userspace mappings are
> > > used with Hardware Tag-Based KASAN enabled.
> > >
> > > Reset page tags for __GFP_ZEROTAGS allocations in post_alloc_hook().
> > >
> > > Also clarify and fix related comments.
> >
> > I assume this is a fix against 44383cef54c0 ("kasan: allow sampling
> > page_alloc allocations for HW_TAGS") which is presently in mm-stable,
> > yes?
>
> Correct. I assumed I shouldn't include a Fixes tag, as the patch is
> not in the mainline.
I think it's best to add the Fixes: if it's known. If the patch was in
mm-unstable then I'd just fold the fix into the base patch, but a
Fixes: is still helpful because it tells people (especially me) which
patch needs the fix.
If the patch is in mm-stable then the SHA is stable and the Fixes: is
desirable for people who are backporting the base patch into earlier
kernels - hopefully when doing this they know to search the tree for
other patches which fix the patch which they are backporting.