From: Jeff Xu <jeffxu@google.com>
To: jeffxu@chromium.org
Cc: akpm@linux-foundation.org, keescook@chromium.org,
torvalds@linux-foundation.org, usama.anjum@collabora.com,
corbet@lwn.net, Liam.Howlett@oracle.com,
lorenzo.stoakes@oracle.com, jorgelo@chromium.org,
groeck@chromium.org, linux-kernel@vger.kernel.org,
linux-kselftest@vger.kernel.org, linux-mm@kvack.org,
jannh@google.com, sroettger@google.com, pedro.falcato@gmail.com,
linux-hardening@vger.kernel.org, willy@infradead.org,
gregkh@linuxfoundation.org, deraadt@openbsd.org,
surenb@google.com, merimus@google.com, rdunlap@infradead.org
Subject: Re: [PATCH] munmap sealed memory cause memory to split (bug)
Date: Wed, 16 Oct 2024 19:38:08 -0700 [thread overview]
Message-ID: <CALmYWFsnx+QvkqLyYRSBW1ueJonEJeXROq6UM7Bbktn3bu+PMQ@mail.gmail.com> (raw)
In-Reply-To: <20241017022627.3112811-1-jeffxu@chromium.org>
On Wed, Oct 16, 2024 at 7:26 PM <jeffxu@chromium.org> wrote:
>
> From: Jeff Xu <jeffxu@google.com>
>
> It appears there is a regression on the latest mm,
> when munmap seals memory, it can cause an unexpected VMA split.
> E.g. repro use this test.
It appears that this test has some dependency tests that haven't been
merged, so can't be run as is.
This is the repro step:
- Allocate 12 pages (0-11).
- Seal middle 4 pages (4567)
- munmap (2345) - this will fail
Seeing VMA for page (0123) is split as 2 VMAs (01)-(23), those 2 VMA
have the same attribute, and should be merged as one.
> ---
> tools/testing/selftests/mm/mseal_test.c | 76 +++++++++++++++++++++++++
> 1 file changed, 76 insertions(+)
>
> diff --git a/tools/testing/selftests/mm/mseal_test.c b/tools/testing/selftests/mm/mseal_test.c
> index fa74dbe4a684..0af33e13b606 100644
> --- a/tools/testing/selftests/mm/mseal_test.c
> +++ b/tools/testing/selftests/mm/mseal_test.c
> @@ -1969,6 +1969,79 @@ static void test_madvise_filebacked_was_writable(bool seal)
> REPORT_TEST_PASS();
> }
>
> +static void test_munmap_free_multiple_ranges_with_split(bool seal)
> +{
> + void *ptr;
> + unsigned long page_size = getpagesize();
> + unsigned long size = 12 * page_size;
> + int ret;
> + int prot;
> +
> + setup_single_address(size, &ptr);
> + FAIL_TEST_IF_FALSE(ptr != (void *)-1);
> +
> + /* seal the middle 4 page */
> + if (seal) {
> + ret = sys_mseal(ptr + 4 * page_size, 4 * page_size);
> + FAIL_TEST_IF_FALSE(!ret);
> +
> + size = get_vma_size(ptr, &prot);
> + FAIL_TEST_IF_FALSE(size == 4 * page_size);
> + FAIL_TEST_IF_FALSE(prot == 4);
> +
> + size = get_vma_size(ptr + 4 * page_size, &prot);
> + FAIL_TEST_IF_FALSE(size == 4 * page_size);
> + FAIL_TEST_IF_FALSE(prot == 4);
> +
> + size = get_vma_size(ptr + 8 * page_size, &prot);
> + FAIL_TEST_IF_FALSE(size == 4 * page_size);
> + FAIL_TEST_IF_FALSE(prot == 4);
> + }
> +
> + /* munmap 4 pages from the third page */
> + ret = sys_munmap(ptr + 2 * page_size, 4 * page_size);
> + if (seal) {
> + FAIL_TEST_IF_FALSE(ret);
> + FAIL_TEST_IF_FALSE(errno == EPERM);
> +
> + size = get_vma_size(ptr, &prot);
> + FAIL_TEST_IF_FALSE(size == 4 * page_size);
> + FAIL_TEST_IF_FALSE(prot == 4);
> +
> + size = get_vma_size(ptr + 4 * page_size, &prot);
> + FAIL_TEST_IF_FALSE(size == 4 * page_size);
> + FAIL_TEST_IF_FALSE(prot == 4);
> +
> + size = get_vma_size(ptr + 8 * page_size, &prot);
> + FAIL_TEST_IF_FALSE(size == 4 * page_size);
> + FAIL_TEST_IF_FALSE(prot == 4);
> + } else
> + FAIL_TEST_IF_FALSE(!ret);
> +
> + /* munmap 4 pages from the sealed page */
> + ret = sys_munmap(ptr + 6 * page_size, 4 * page_size);
> + if (seal) {
> + FAIL_TEST_IF_FALSE(ret);
> + FAIL_TEST_IF_FALSE(errno == EPERM);
> +
> + size = get_vma_size(ptr + 4 * page_size, &prot);
> + FAIL_TEST_IF_FALSE(size == 4 * page_size);
> + FAIL_TEST_IF_FALSE(prot == 4);
> +
> + size = get_vma_size(ptr + 4 * page_size, &prot);
> + FAIL_TEST_IF_FALSE(size == 4 * page_size);
> + FAIL_TEST_IF_FALSE(prot == 4);
> +
> + size = get_vma_size(ptr + 8 * page_size, &prot);
> + FAIL_TEST_IF_FALSE(size == 4 * page_size);
> + FAIL_TEST_IF_FALSE(prot == 4);
> + } else
> + FAIL_TEST_IF_FALSE(!ret);
> +
> + REPORT_TEST_PASS();
> +}
> +
> +
> int main(int argc, char **argv)
> {
> bool test_seal = seal_support();
> @@ -2099,5 +2172,8 @@ int main(int argc, char **argv)
> test_madvise_filebacked_was_writable(false);
> test_madvise_filebacked_was_writable(true);
>
> + test_munmap_free_multiple_ranges_with_split(false);
> + test_munmap_free_multiple_ranges_with_split(true);
> +
> ksft_finished();
> }
> --
> 2.47.0.rc1.288.g06298d1525-goog
>
next prev parent reply other threads:[~2024-10-17 2:38 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-17 2:26 jeffxu
2024-10-17 2:38 ` Jeff Xu [this message]
2024-10-17 6:03 ` Greg KH
2024-10-17 16:17 ` Jeff Xu
2024-10-17 8:18 ` Lorenzo Stoakes
2024-10-17 16:20 ` Jeff Xu
2024-10-17 19:14 ` Pedro Falcato
2024-10-17 19:16 ` Lorenzo Stoakes
2024-10-17 20:12 ` Jeff Xu
2024-10-17 9:46 ` Lorenzo Stoakes
2024-10-17 9:48 ` Lorenzo Stoakes
2024-10-17 9:59 ` Lorenzo Stoakes
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CALmYWFsnx+QvkqLyYRSBW1ueJonEJeXROq6UM7Bbktn3bu+PMQ@mail.gmail.com \
--to=jeffxu@google.com \
--cc=Liam.Howlett@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=corbet@lwn.net \
--cc=deraadt@openbsd.org \
--cc=gregkh@linuxfoundation.org \
--cc=groeck@chromium.org \
--cc=jannh@google.com \
--cc=jeffxu@chromium.org \
--cc=jorgelo@chromium.org \
--cc=keescook@chromium.org \
--cc=linux-hardening@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lorenzo.stoakes@oracle.com \
--cc=merimus@google.com \
--cc=pedro.falcato@gmail.com \
--cc=rdunlap@infradead.org \
--cc=sroettger@google.com \
--cc=surenb@google.com \
--cc=torvalds@linux-foundation.org \
--cc=usama.anjum@collabora.com \
--cc=willy@infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox