From: Emil Velikov <emil.l.velikov@gmail.com>
To: igt-dev@lists.freedesktop.org
Subject: [igt-dev] [PATCH i-g-t v2] ts/core_setmaster: new test for drop/set master semantics
Date: Wed, 19 Feb 2020 14:24:45 +0000 [thread overview]
Message-ID: <20200219142445.69061-1-emil.l.velikov@gmail.com> (raw)
From: Emil Velikov <emil.velikov@collabora.com>
This test adds three distinct subtests:
- drop/set master as root
- drop/set master as non-root
- drop/set master for a shared fd
Currently the second subtest will fail, with kernel patch to address
that has been submitted.
v2: Add to the autotools build
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
---
tests/Makefile.sources | 1 +
tests/core_setmaster.c | 182 +++++++++++++++++++++++++++++++++++++++++
tests/meson.build | 1 +
3 files changed, 184 insertions(+)
create mode 100644 tests/core_setmaster.c
diff --git a/tests/Makefile.sources b/tests/Makefile.sources
index b87d6333..5da36a91 100644
--- a/tests/Makefile.sources
+++ b/tests/Makefile.sources
@@ -18,6 +18,7 @@ TESTS_progs = \
core_getclient \
core_getstats \
core_getversion \
+ core_setmaster \
core_setmaster_vs_auth \
debugfs_test \
dmabuf \
diff --git a/tests/core_setmaster.c b/tests/core_setmaster.c
new file mode 100644
index 00000000..2447c077
--- /dev/null
+++ b/tests/core_setmaster.c
@@ -0,0 +1,182 @@
+/*
+ * Copyright © 2020 Collabora, Ltd.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Authors:
+ * Emil Velikov <emil.l.velikov@gmail.com>
+ *
+ */
+
+/*
+ * Testcase: Check that drop/setMaster behaves correctly wrt root/user access
+ *
+ * Test checks if the ioctls succeed or fail, depending if the applications was
+ * run with root, user privileges or if we have separate privileged arbitrator.
+ */
+
+#include "igt.h"
+#include <unistd.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include <sys/stat.h>
+
+IGT_TEST_DESCRIPTION("Check that drop/setMaster behaves correctly wrt root/user"
+ " access");
+
+static bool is_master(int fd)
+{
+ /* FIXME: replace with drmIsMaster once we bumped libdrm version */
+ return drmAuthMagic(fd, 0) != -EACCES;
+}
+
+static void check_drop_set(void)
+{
+ int master;
+
+ master = __drm_open_driver(DRIVER_ANY);
+
+ /* Double-check if open has failed */
+ igt_assert_neq(master, -1);
+
+ /* At this point we're master capable due to:
+ * - being root - always
+ * - normal user - as the only drm only drm client (on this VT)
+ */
+ igt_assert_eq(is_master(master), true);
+
+ /* If we have SYS_CAP_ADMIN we're in the textbook best-case scenario.
+ *
+ * Otherwise newer kernels allow the application to drop/revoke its
+ * master capability and request it again later.
+ *
+ * In this case, we address two types of issues:
+ * - the application no longer need suid-root (or equivalent) which
+ * was otherwise required _solely_ for these two ioctls
+ * - plenty of applications ignore (or discard) the result of the
+ * calls all together.
+ */
+ igt_assert_eq(drmDropMaster(master), 0);
+ igt_assert_eq(drmSetMaster(master), 0);
+
+ close(master);
+}
+
+static unsigned tweak_perm(uint8_t *saved_perm, unsigned max_perm, bool save)
+{
+ char path[256];
+ struct stat st;
+ unsigned i;
+
+ for (i = 0; i < max_perm; i++) {
+ snprintf(path, sizeof(path), "/dev/dri/card%u", i);
+
+ /* Existing userspace assumes there's no gaps, do the same. */
+ if (stat(path, &st) != 0)
+ break;
+
+ if (save) {
+ /* Save and toggle */
+ saved_perm[i] = st.st_mode & (S_IROTH | S_IWOTH);
+ st.st_mode |= S_IROTH | S_IWOTH;
+ } else {
+ /* Clear and restore */
+ st.st_mode &= ~(S_IROTH | S_IWOTH);
+ st.st_mode |= saved_perm[i];
+ }
+
+ /* In the extremely unlikely case of this failing, there't not
+ * much we can do.
+ */
+ chmod(path, st.st_mode);
+ }
+ return i;
+}
+
+
+igt_main
+{
+ igt_subtest("master-drop-set-root") {
+ check_drop_set();
+ }
+
+
+ igt_subtest("master-drop-set-user") {
+ uint8_t saved_perm[255];
+ unsigned num;
+
+ /* Upon dropping root we end up as random user, which
+ * a) is not in the video group, and
+ * b) lacks logind ACL, thus
+ * any open() fill fail.
+ *
+ * As such, save the state of original other rw permissions
+ * and toggle them on.
+ */
+ num = tweak_perm(saved_perm, ARRAY_SIZE(saved_perm), true);
+
+ igt_fork(child, 1) {
+ igt_drop_root();
+ check_drop_set();
+ }
+ igt_waitchildren();
+
+ /* Restore the orignal permissions */
+ tweak_perm(saved_perm, num, false);
+ }
+
+ igt_subtest("master-drop-set-shared-fd") {
+ int master;
+
+ master = __drm_open_driver(DRIVER_ANY);
+
+ /* Double-check if open has failed */
+ igt_assert_neq(master, -1);
+
+ igt_assert_eq(is_master(master), true);
+ igt_fork(child, 1) {
+ igt_drop_root();
+
+ /* Dropping root privileges should not alter the
+ * master capability of the fd */
+ igt_assert_eq(is_master(master), true);
+
+ /* Even though we've got the master capable fd, we're
+ * a different process (kernel struct pid *) than the
+ * one which opened the device node.
+ *
+ * This ensures that existing workcases of separate
+ * (privileged) arbitrator still work. For example:
+ * - logind + X/Wayland compositor
+ * - weston-launch + weston
+ */
+ igt_assert_eq(drmDropMaster(master), -1);
+ igt_assert_eq(errno, EACCES);
+ igt_assert_eq(drmSetMaster(master), -1);
+ igt_assert_eq(errno, EACCES);
+
+ close(master);
+ }
+ igt_waitchildren();
+
+ close(master);
+ }
+}
diff --git a/tests/meson.build b/tests/meson.build
index fa0103e3..d8fb9f3e 100644
--- a/tests/meson.build
+++ b/tests/meson.build
@@ -3,6 +3,7 @@ test_progs = [
'core_getclient',
'core_getstats',
'core_getversion',
+ 'core_setmaster',
'core_setmaster_vs_auth',
'debugfs_test',
'dmabuf',
--
2.25.0
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev
next reply other threads:[~2020-02-19 14:24 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-19 14:24 Emil Velikov [this message]
2020-02-19 15:41 ` [igt-dev] ✗ GitLab.Pipeline: failure for ts/core_setmaster: new test for drop/set master semantics (rev2) Patchwork
2020-02-19 15:59 ` [igt-dev] ✓ Fi.CI.BAT: success " Patchwork
2020-02-19 16:20 ` [igt-dev] [PATCH i-g-t v2] ts/core_setmaster: new test for drop/set master semantics Emil Velikov
2020-02-21 9:30 ` [igt-dev] ✓ Fi.CI.IGT: success for ts/core_setmaster: new test for drop/set master semantics (rev2) Patchwork
2020-02-21 11:36 ` Emil Velikov
2020-02-21 11:58 ` [igt-dev] [PATCH i-g-t v2] ts/core_setmaster: new test for drop/set master semantics Petri Latvala
2020-02-21 12:19 ` Petri Latvala
2020-02-21 13:55 ` Emil Velikov
2020-02-24 8:49 ` Petri Latvala
2020-02-24 11:57 ` Emil Velikov
2020-02-25 9:10 ` Petri Latvala
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=20200219142445.69061-1-emil.l.velikov@gmail.com \
--to=emil.l.velikov@gmail.com \
--cc=igt-dev@lists.freedesktop.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