From: Albert Esteve <aesteve@redhat.com>
To: Sumit Semwal <sumit.semwal@linaro.org>,
Christian König <christian.koenig@amd.com>,
Benjamin Gaignard <benjamin.gaignard@collabora.com>,
Brian Starkey <Brian.Starkey@arm.com>,
John Stultz <jstultz@google.com>,
"T.J. Mercier" <tjmercier@google.com>,
Shuah Khan <shuah@kernel.org>
Cc: linux-media@vger.kernel.org, dri-devel@lists.freedesktop.org,
linaro-mm-sig@lists.linaro.org, linux-kernel@vger.kernel.org,
linux-kselftest@vger.kernel.org,
Albert Esteve <aesteve@redhat.com>,
mripard@kernel.org
Subject: [PATCH 2/2] selftests: dma-buf: add DERIVE ioctl tests
Date: Thu, 21 May 2026 11:10:15 +0200 [thread overview]
Message-ID: <20260521-dmabuf-limit-access-v1-2-26c01e27365a@redhat.com> (raw)
In-Reply-To: <20260521-dmabuf-limit-access-v1-0-26c01e27365a@redhat.com>
dma-buf now supports aliasing an existing file descriptor with reduced
access permissions via DMA_BUF_IOCTL_DERIVE ioctl, and enforces that a
shared writable mapping cannot be created through a read-only file
descriptor.
Add two tests to the dmabuf-heaps selftest to exercise this behaviour.
The positive test allocates a buffer, derives its file descriptor as
O_RDONLY through the ioctl, confirms that a read-only shared mapping
succeeds, and verifies that data written through the original read-write
file descriptor is visible through the derived one.
The negative test confirms that attempting a DMA_BUF_IOCTL_DERIVE ioctl
call with RW flags on a read-only file descriptor is rejected with EACCES.
Same for mmap() escalation attempt.
Signed-off-by: Albert Esteve <aesteve@redhat.com>
---
tools/testing/selftests/dmabuf-heaps/dmabuf-heap.c | 114 ++++++++++++++++++++-
1 file changed, 113 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/dmabuf-heaps/dmabuf-heap.c b/tools/testing/selftests/dmabuf-heaps/dmabuf-heap.c
index fc9694fc4e89e..c3856189200be 100644
--- a/tools/testing/selftests/dmabuf-heaps/dmabuf-heap.c
+++ b/tools/testing/selftests/dmabuf-heaps/dmabuf-heap.c
@@ -390,6 +390,116 @@ static void test_alloc_errors(char *heap_name)
close(heap_fd);
}
+static int setup_ro_derive(int heap_fd, int *dmabuf_fd, int *ro_fd)
+{
+ struct dma_buf_derive params = {
+ .flags = O_RDONLY | O_CLOEXEC,
+ };
+ int ret;
+
+ ret = dmabuf_heap_alloc(heap_fd, ONE_MEG, 0, dmabuf_fd);
+ ksft_test_result(!ret, "Allocate RW buffer\n");
+ if (ret)
+ return -1;
+
+ ret = ioctl(*dmabuf_fd, DMA_BUF_IOCTL_DERIVE, ¶ms);
+ ksft_test_result(!ret, "Derive as O_RDONLY %s\n",
+ ret < 0 ? strerror(errno) : "OK");
+ if (ret < 0) {
+ close(*dmabuf_fd);
+ *dmabuf_fd = -1;
+ return -1;
+ }
+
+ *ro_fd = params.fd;
+ return 0;
+}
+
+static void test_ro_derive(char *heap_name)
+{
+ int heap_fd = -1, dmabuf_fd = -1, ro_fd = -1;
+ void *rw_map = MAP_FAILED, *ro_map = MAP_FAILED;
+ int ret;
+
+ heap_fd = dmabuf_heap_open(heap_name);
+
+ ksft_print_msg("Testing read-only derive with mmap:\n");
+
+ if (setup_ro_derive(heap_fd, &dmabuf_fd, &ro_fd))
+ goto out;
+
+ rw_map = mmap(NULL, ONE_MEG, PROT_READ | PROT_WRITE, MAP_SHARED,
+ dmabuf_fd, 0);
+ ksft_test_result(rw_map != MAP_FAILED, "RW mmap on RW fd %s\n",
+ rw_map == MAP_FAILED ? strerror(errno) : "OK");
+ if (rw_map == MAP_FAILED)
+ goto out;
+
+ dmabuf_sync(dmabuf_fd, DMA_BUF_SYNC_START);
+ memset(rw_map, 0xab, ONE_MEG);
+ dmabuf_sync(dmabuf_fd, DMA_BUF_SYNC_END);
+
+ ro_map = mmap(NULL, ONE_MEG, PROT_READ, MAP_SHARED, ro_fd, 0);
+ ksft_test_result(ro_map != MAP_FAILED, "RO mmap on RO fd %s\n",
+ ro_map == MAP_FAILED ? strerror(errno) : "OK");
+ if (ro_map == MAP_FAILED)
+ goto out;
+
+ dmabuf_sync(ro_fd, DMA_BUF_SYNC_START);
+ ret = memcmp(rw_map, ro_map, ONE_MEG);
+ dmabuf_sync(ro_fd, DMA_BUF_SYNC_END);
+ ksft_test_result(!ret, "Data written via RW fd visible through RO fd\n");
+
+out:
+ if (ro_map != MAP_FAILED)
+ munmap(ro_map, ONE_MEG);
+ if (rw_map != MAP_FAILED)
+ munmap(rw_map, ONE_MEG);
+ if (ro_fd >= 0)
+ close(ro_fd);
+ if (dmabuf_fd >= 0)
+ close(dmabuf_fd);
+ close(heap_fd);
+}
+
+static void test_ro_derive_escalation(char *heap_name)
+{
+ struct dma_buf_derive params = {
+ .flags = O_RDWR | O_CLOEXEC,
+ };
+ int heap_fd = -1, dmabuf_fd = -1, ro_fd = -1, ret = 0;
+ void *bad_map;
+
+ heap_fd = dmabuf_heap_open(heap_name);
+
+ ksft_print_msg("Testing read-only derive with escalation attempt:\n");
+
+ if (setup_ro_derive(heap_fd, &dmabuf_fd, &ro_fd))
+ goto out;
+
+ ret = ioctl(ro_fd, DMA_BUF_IOCTL_DERIVE, ¶ms);
+ ksft_test_result(ret < 0 && errno == EACCES,
+ "O_RDWR derive on RO fd correctly rejected (errno=%d)\n",
+ errno);
+ if (!ret)
+ close(params.fd);
+
+ bad_map = mmap(NULL, ONE_MEG, PROT_READ | PROT_WRITE, MAP_SHARED,
+ ro_fd, 0);
+ ksft_test_result(bad_map == MAP_FAILED && errno == EACCES,
+ "RW shared mmap on RO fd correctly rejected (errno=%d)\n",
+ errno);
+ if (bad_map != MAP_FAILED)
+ munmap(bad_map, ONE_MEG);
+
+out:
+ if (ro_fd >= 0)
+ close(ro_fd);
+ if (dmabuf_fd >= 0)
+ close(dmabuf_fd);
+ close(heap_fd);
+}
+
static int numer_of_heaps(void)
{
DIR *d = opendir(DEVPATH);
@@ -420,7 +530,7 @@ int main(void)
return KSFT_SKIP;
}
- ksft_set_plan(11 * numer_of_heaps());
+ ksft_set_plan(20 * numer_of_heaps());
while ((dir = readdir(d))) {
if (!strncmp(dir->d_name, ".", 2))
@@ -435,6 +545,8 @@ int main(void)
test_alloc_zeroed(dir->d_name, ONE_MEG);
test_alloc_compat(dir->d_name);
test_alloc_errors(dir->d_name);
+ test_ro_derive(dir->d_name);
+ test_ro_derive_escalation(dir->d_name);
}
closedir(d);
--
2.53.0
next prev parent reply other threads:[~2026-05-21 9:11 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-21 9:10 [PATCH 0/2] dma-buf: add DMA_BUF_IOCTL_DERIVE for reduced-permission aliases Albert Esteve
2026-05-21 9:10 ` [PATCH 1/2] " Albert Esteve
2026-05-21 12:30 ` Christian König
2026-05-25 10:27 ` Claude review: " Claude Code Review Bot
2026-05-21 9:10 ` Albert Esteve [this message]
2026-05-25 10:27 ` Claude review: selftests: dma-buf: add DERIVE ioctl tests Claude Code Review Bot
2026-05-21 12:28 ` [PATCH 0/2] dma-buf: add DMA_BUF_IOCTL_DERIVE for reduced-permission aliases Christian König
2026-05-21 13:01 ` Albert Esteve
2026-05-21 13:14 ` Christian König
2026-05-25 10:27 ` Claude review: " Claude Code Review Bot
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=20260521-dmabuf-limit-access-v1-2-26c01e27365a@redhat.com \
--to=aesteve@redhat.com \
--cc=Brian.Starkey@arm.com \
--cc=benjamin.gaignard@collabora.com \
--cc=christian.koenig@amd.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=jstultz@google.com \
--cc=linaro-mm-sig@lists.linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=mripard@kernel.org \
--cc=shuah@kernel.org \
--cc=sumit.semwal@linaro.org \
--cc=tjmercier@google.com \
/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