From: Emil Velikov <emil.l.velikov@gmail.com>
To: igt-dev@lists.freedesktop.org
Cc: emil.l.velikov@gmail.com
Subject: [igt-dev] [PATCH i-g-t] tests/core_unauth_vs_render: new test for the relaxed DRM_AUTH handling
Date: Fri, 8 Feb 2019 18:03:30 +0000 [thread overview]
Message-ID: <20190208180330.30125-1-emil.l.velikov@gmail.com> (raw)
From: Emil Velikov <emil.velikov@collabora.com>
As the inline comment says, this test checks that the kernel allows
unauthenticated master with render capable, RENDER_ALLOW ioctls.
The kernel commit has extra details why.
v2:
- drop RUN_AS_ROOT guard
- call check_auth() on the {,un}authenticated device
- check the device is PRIME (import) capable
- check the device has render node
- tweak expectations based on above three
- elaborate why we care only about -EACCES
v3:
- fold into existing core_auth.c
- move igt_assert within the subtest
- make has_prime_import() an igt_require()
- check for BADF before and after, as requested. Not strictly needed.
- swap igt_info+drm_open_driver with comment + __drm_open_driver
Former calls igt_skip() which is problematic with igt_fork().
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
---
Hope that covers everything Dan. Alternatively I would prefer to keep
polishing this, while the kernel fix is merged. This way existing users
would work :-)
---
tests/core_auth.c | 109 ++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 109 insertions(+)
diff --git a/tests/core_auth.c b/tests/core_auth.c
index 0c016a37..0b9073cb 100644
--- a/tests/core_auth.c
+++ b/tests/core_auth.c
@@ -1,5 +1,6 @@
/*
* Copyright 2015 David Herrmann <dh.herrmann@gmail.com>
+ * Copyright 2018 Collabora, Ltd
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
@@ -40,6 +41,7 @@
#include <sys/time.h>
#include <sys/poll.h>
#include <sys/resource.h>
+#include <sys/sysmacros.h>
#include "drm.h"
#ifdef __linux__
@@ -190,6 +192,98 @@ static void test_basic_auth(int master)
close(slave);
}
+static bool has_prime_import(int fd)
+{
+ uint64_t value;
+
+ if (drmGetCap(fd, DRM_CAP_PRIME, &value))
+ return false;
+
+ return value & DRM_PRIME_CAP_IMPORT;
+}
+
+static void check_auth_sanity(int master)
+{
+ uint32_t handle;
+
+ igt_assert(check_auth(master) == true);
+ igt_require(has_prime_import(master));
+
+ igt_assert(drmPrimeFDToHandle(master, -1, &handle) < 0);
+
+ /* IOCTL requires authenticated master as done in drm_permit.
+ * As we get past that, we'll fail due to the invalid FD.
+ *
+ * Note: strictly speaking this is unrelated to the goal of
+ * the test, although danvet requested it.
+ */
+ igt_assert(errno == EBADF);
+}
+
+static bool has_render_node(int fd)
+{
+ char node_name[80];
+ struct stat sbuf;
+
+ if (fstat(fd, &sbuf))
+ return false;
+
+ sprintf(node_name, "/dev/dri/renderD%d", minor(sbuf.st_rdev) | 0x80);
+ if (stat(node_name, &sbuf))
+ return false;
+
+ return true;
+}
+
+/*
+ * Testcase: Render capable, unauthenticated master doesn't throw -EACCES for
+ * DRM_RENDER_ALLOW ioctls.
+ */
+static void test_unauth_vs_render(int master)
+{
+ int slave;
+ uint32_t handle;
+
+ /*
+ * FIXME: when drm_open_driver() fails to open() a node (insufficient
+ * permissions or otherwise, it will igt_skip.
+ * As of today, igt_skip and igt_fork do not work together.
+ */
+ slave = __drm_open_driver(DRIVER_ANY);
+ /*
+ * FIXME: relate to the master fd passed with the above open and fix
+ * all of IGT.
+ */
+
+ igt_assert(slave >= 0);
+
+ /*
+ * The second open() happens without CAP_SYS_ADMIN, thus it will NOT
+ * be authenticated.
+ */
+ igt_assert(check_auth(slave) == false);
+
+ /* Issuing the following ioctl will fail, no doubt about it. */
+ igt_assert(drmPrimeFDToHandle(slave, -1, &handle) < 0);
+
+ /*
+ * Updated kernels allow render capable, unauthenticated master to
+ * issue DRM_AUTH ioctls (like FD2HANDLE above), as long as they are
+ * annotated as DRM_RENDER_ALLOW.
+ *
+ * Otherwise, errno is set to -EACCES
+ *
+ * Note: We are _not_ interested in the FD2HANDLE specific errno,
+ * yet the EBADF check is added on the explicit request by danvet.
+ */
+ if (has_render_node(slave))
+ igt_assert(errno == EBADF);
+ else
+ igt_assert(errno == EACCES);
+
+ close(slave);
+}
+
igt_main
{
int master;
@@ -228,4 +322,19 @@ igt_main
igt_subtest("many-magics")
test_many_magics(master);
}
+
+ igt_subtest_group {
+ igt_fixture
+ master = drm_open_driver(DRIVER_ANY);
+
+ igt_subtest("unauth-vs-render") {
+ check_auth_sanity(master);
+
+ igt_fork(child, 1) {
+ igt_drop_root();
+ test_unauth_vs_render(master);
+ }
+ igt_waitchildren();
+ }
+ }
}
--
2.20.1
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev
next reply other threads:[~2019-02-08 18:07 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-08 18:03 Emil Velikov [this message]
2019-02-08 18:47 ` [igt-dev] ✗ Fi.CI.BAT: failure for tests/core_unauth_vs_render: new test for the relaxed DRM_AUTH handling (rev3) Patchwork
-- strict thread matches above, loose matches on Subject: below --
2019-02-06 13:18 [igt-dev] [PATCH i-g-t] tests/core_unauth_vs_render: new test for the relaxed DRM_AUTH handling Emil Velikov
2019-02-07 8:59 ` Petri Latvala
2019-02-07 12:00 ` Emil Velikov
2019-02-07 12:08 ` Petri Latvala
2019-02-07 14:17 ` Daniel Vetter
2019-02-07 17:08 ` Emil Velikov
2019-02-07 17:47 ` Daniel Vetter
2019-01-14 8:39 Emil Velikov
2019-01-14 10:58 ` Petri Latvala
2019-01-14 11:15 ` Emil Velikov
2019-01-18 15:58 ` Daniel Vetter
2019-01-22 17:44 ` Emil Velikov
2019-01-23 11:18 ` Daniel Vetter
2019-01-23 11:33 ` Daniel Vetter
2019-01-23 15:55 ` Emil Velikov
2019-01-23 16:43 ` Daniel Vetter
2019-01-23 20:01 ` Emil Velikov
2019-01-24 8:41 ` Petri Latvala
2019-01-24 10:56 ` Daniel Vetter
2019-01-24 11:03 ` Daniel Vetter
2019-01-24 13:55 ` Emil Velikov
2019-01-23 11:42 ` Petri Latvala
2019-01-23 12:08 ` Daniel Vetter
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=20190208180330.30125-1-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