linux-kernel - [PATCH v3 7/8] tools/testing/selftests/mm: add MADV

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <41d851aea2975ab54ddaafa3a132e2fd6f8c5582.1762531708.git.lorenzo.stoakes@oracle.com>
Date: Fri,  7 Nov 2025 16:11:52 +0000
From: Lorenzo Stoakes <lorenzo.stoakes@...cle.com>
To: Andrew Morton <akpm@...ux-foundation.org>
Cc: Jonathan Corbet <corbet@....net>, David Hildenbrand <david@...hat.com>,
        "Liam R . Howlett" <Liam.Howlett@...cle.com>,
        Vlastimil Babka <vbabka@...e.cz>, Mike Rapoport <rppt@...nel.org>,
        Suren Baghdasaryan <surenb@...gle.com>, Michal Hocko <mhocko@...e.com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Masami Hiramatsu <mhiramat@...nel.org>,
        Mathieu Desnoyers <mathieu.desnoyers@...icios.com>,
        Jann Horn <jannh@...gle.com>, Pedro Falcato <pfalcato@...e.de>,
        Zi Yan <ziy@...dia.com>, Baolin Wang <baolin.wang@...ux.alibaba.com>,
        Nico Pache <npache@...hat.com>, Ryan Roberts <ryan.roberts@....com>,
        Dev Jain <dev.jain@....com>, Barry Song <baohua@...nel.org>,
        Lance Yang <lance.yang@...ux.dev>, linux-kernel@...r.kernel.org,
        linux-fsdevel@...r.kernel.org, linux-doc@...r.kernel.org,
        linux-mm@...ck.org, linux-trace-kernel@...r.kernel.org,
        linux-kselftest@...r.kernel.org, Andrei Vagin <avagin@...il.com>
Subject: [PATCH v3 7/8] tools/testing/selftests/mm: add MADV_COLLAPSE test case

To ensure the retract_page_tables() logic functions correctly with the
introduction of VM_MAYBE_GUARD, add a test to assert that madvise collapse
fails when guard regions are established in the collapsed range in all
cases.

Unfortunately we cannot differentiate between
e.g. CONFIG_READ_ONLY_THP_FOR_FS not being set vs. a file-backed VMA having
collapse correctly disallowed, so in each instance we will get an assert
pass here.

We add an additional check to see whether guard regions are preserved
across collapse in case of a bug causing the collapse to succeed, which
will give us more data to debug with should this occur in future.

Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@...cle.com>
---
 tools/testing/selftests/mm/guard-regions.c | 65 ++++++++++++++++++++++
 1 file changed, 65 insertions(+)

diff --git a/tools/testing/selftests/mm/guard-regions.c b/tools/testing/selftests/mm/guard-regions.c
index 8dd81c0a4a5a..c549bcd6160b 100644
--- a/tools/testing/selftests/mm/guard-regions.c
+++ b/tools/testing/selftests/mm/guard-regions.c
@@ -2138,4 +2138,69 @@ TEST_F(guard_regions, pagemap_scan)
 	ASSERT_EQ(munmap(ptr, 10 * page_size), 0);
 }
 
+TEST_F(guard_regions, collapse)
+{
+	const unsigned long page_size = self->page_size;
+	const unsigned long size = 2 * HPAGE_SIZE;
+	const unsigned long num_pages = size / page_size;
+	char *ptr;
+	int i;
+
+	/* Need file to be correct size for tests for non-anon. */
+	if (variant->backing != ANON_BACKED)
+		ASSERT_EQ(ftruncate(self->fd, size), 0);
+
+	/*
+	 * We must close and re-open local-file backed as read-only for
+	 * CONFIG_READ_ONLY_THP_FOR_FS to work.
+	 */
+	if (variant->backing == LOCAL_FILE_BACKED) {
+		ASSERT_EQ(close(self->fd), 0);
+
+		self->fd = open(self->path, O_RDONLY);
+		ASSERT_GE(self->fd, 0);
+	}
+
+	ptr = mmap_(self, variant, NULL, size, PROT_READ, 0, 0);
+	ASSERT_NE(ptr, MAP_FAILED);
+
+	/* Prevent being faulted-in as huge. */
+	ASSERT_EQ(madvise(ptr, size, MADV_NOHUGEPAGE), 0);
+	/* Fault in. */
+	ASSERT_EQ(madvise(ptr, size, MADV_POPULATE_READ), 0);
+
+	/* Install guard regions in ever other page. */
+	for (i = 0; i < num_pages; i += 2) {
+		char *ptr_page = &ptr[i * page_size];
+
+		ASSERT_EQ(madvise(ptr_page, page_size, MADV_GUARD_INSTALL), 0);
+		/* Accesses should now fail. */
+		ASSERT_FALSE(try_read_buf(ptr_page));
+	}
+
+	/* Allow huge page throughout region. */
+	ASSERT_EQ(madvise(ptr, size, MADV_HUGEPAGE), 0);
+
+	/*
+	 * Now collapse the entire region. This should fail in all cases.
+	 *
+	 * The madvise() call will also fail if CONFIG_READ_ONLY_THP_FOR_FS is
+	 * not set for the local file case, but we can't differentiate whether
+	 * this occurred or if the collapse was rightly rejected.
+	 */
+	EXPECT_NE(madvise(ptr, size, MADV_COLLAPSE), 0);
+
+	/*
+	 * If we introduce a bug that causes the collapse to succeed, gather
+	 * data on whether guard regions are at least preserved. The test will
+	 * fail at this point in any case.
+	 */
+	for (i = 0; i < num_pages; i += 2) {
+		char *ptr_page = &ptr[i * page_size];
+
+		/* Accesses should still fail. */
+		ASSERT_FALSE(try_read_buf(ptr_page));
+	}
+}
+
 TEST_HARNESS_MAIN
-- 
2.51.0