Commit 2c2365e2 authored by Nhat Pham's avatar Nhat Pham Committed by Jinjiang Tu
Browse files

mm: cachestat: fix folio read-after-free in cache walk 

stable inclusion
from stable-v6.6.21
commit ba60fdf75e89ea762bb617be578dc47f27655117
category: bugfix
bugzilla: https://gitee.com/src-openeuler/kernel/issues/I98BXJ
CVE: CVE-2024-26630

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=ba60fdf75e89ea762bb617be578dc47f27655117

--------------------------------

commit 3a75cb05d53f4a6823a32deb078de1366954a804 upstream.

In cachestat, we access the folio from the page cache's xarray to compute
its page offset, and check for its dirty and writeback flags.  However, we
do not hold a reference to the folio before performing these actions,
which means the folio can concurrently be released and reused as another
folio/page/slab.

Get around this altogether by just using xarray's existing machinery for
the folio page offsets and dirty/writeback states.

This changes behavior for tmpfs files to now always report zeroes in their
dirty and writeback counters.  This is okay as tmpfs doesn't follow
conventional writeback cache behavior: its pages get "cleaned" during
swapout, after which they're no longer resident etc.

Link: https://lkml.kernel.org/r/20240220153409.GA216065@cmpxchg.org


Fixes: cf264e13 ("cachestat: implement cachestat syscall")
Reported-by: default avatarJann Horn <jannh@google.com>
Suggested-by: default avatarMatthew Wilcox <willy@infradead.org>
Signed-off-by: default avatarNhat Pham <nphamcs@gmail.com>
Signed-off-by: default avatarJohannes Weiner <hannes@cmpxchg.org>
Tested-by: default avatarJann Horn <jannh@google.com>
Cc: <stable@vger.kernel.org>	[6.4+]
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: default avatarJinjiang Tu <tujinjiang@huawei.com>
parent 2173cb4a

mm/filemap.c

+26 −25
Original line number Diff line number Diff line
@@ -4139,18 +4139,25 @@ static void filemap_cachestat(struct address_space *mapping, 


	rcu_read_lock();

	xas_for_each(&xas, folio, last_index) {

		int order;

		unsigned long nr_pages;

		pgoff_t folio_first_index, folio_last_index;



		/*

		 * Don't deref the folio. It is not pinned, and might

		 * get freed (and reused) underneath us.

		 *

		 * We *could* pin it, but that would be expensive for

		 * what should be a fast and lightweight syscall.

		 *

		 * Instead, derive all information of interest from

		 * the rcu-protected xarray.

		 */



		if (xas_retry(&xas, folio))

			continue;



		if (xa_is_value(folio)) {

			/* page is evicted */

			void *shadow = (void *)folio;

			bool workingset; /* not used */

			int order = xa_get_order(xas.xa, xas.xa_index);



		order = xa_get_order(xas.xa, xas.xa_index);

		nr_pages = 1 << order;

		folio_first_index = round_down(xas.xa_index, 1 << order);

		folio_last_index = folio_first_index + nr_pages - 1;
@@ -4162,6 +4169,11 @@ static void filemap_cachestat(struct address_space *mapping, 
		if (folio_last_index > last_index)

			nr_pages -= folio_last_index - last_index;



		if (xa_is_value(folio)) {

			/* page is evicted */

			void *shadow = (void *)folio;

			bool workingset; /* not used */



			cs->nr_evicted += nr_pages;



#ifdef CONFIG_SWAP /* implies CONFIG_MMU */
@@ -4178,24 +4190,13 @@ static void filemap_cachestat(struct address_space *mapping, 
			goto resched;

		}



		nr_pages = folio_nr_pages(folio);

		folio_first_index = folio_pgoff(folio);

		folio_last_index = folio_first_index + nr_pages - 1;



		/* Folios might straddle the range boundaries, only count covered pages */

		if (folio_first_index < first_index)

			nr_pages -= first_index - folio_first_index;



		if (folio_last_index > last_index)

			nr_pages -= folio_last_index - last_index;



		/* page is in cache */

		cs->nr_cache += nr_pages;



		if (folio_test_dirty(folio))

		if (xas_get_mark(&xas, PAGECACHE_TAG_DIRTY))

			cs->nr_dirty += nr_pages;



		if (folio_test_writeback(folio))

		if (xas_get_mark(&xas, PAGECACHE_TAG_WRITEBACK))

			cs->nr_writeback += nr_pages;



resched: