selinux.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH testsuite v3] tests/bpf: Add tests for SELinux BPF token access control
@ 2025-08-21 19:03 Eric Suen
  2025-08-25 14:21 ` Stephen Smalley
  0 siblings, 1 reply; 3+ messages in thread
From: Eric Suen @ 2025-08-21 19:03 UTC (permalink / raw)
  To: selinux; +Cc: paul, stephen.smalley.work, omosnace, danieldurning.work

This patch adds new tests to verify the SELinux support for BPF token
access control, as introduced in the corresponding kernel patch:
  https://lore.kernel.org/selinux/20250816201420.197-1-ericsu@linux.microsoft.com/

Four new tests are added to cover both positive and negative scenarios,
ensuring that the SELinux policy enforcement on BPF token usage behaves
as expected.
  - Successful map_create and prog_load when SELinux permissions are
    granted.
  - Enforcement of SELinux policy restrictions when access is denied.

For testing purposes, you can update the base policy by manually
modifying your base module and tweaking /usr/share/selinux/devel as
follows:
  sudo semodule -c -E base
  sudo cp base.cil base.cil.orig
  sudo sed -i "s/map_create/map_create map_create_as/" base.cil
  sudo sed -i "s/prog_load/prog_load prog_load_as/" base.cil
  sudo semodule -i base.cil
  echo "(policycap bpf_token_perms)" > bpf_token_perms.cil
  sudo semodule -i bpf_token_perms.cil
  sudo cp /usr/share/selinux/devel/include/support/all_perms.spt \
      /usr/share/selinux/devel/include/support/all_perms.spt.orig
  sudo sed -i "s/map_create/map_create map_create_as/" \
      /usr/share/selinux/devel/include/support/all_perms.spt
  sudo sed -i "s/prog_load/prog_load prog_load_as/" \
      /usr/share/selinux/devel/include/support/all_perms.spt

When finished testing, you can semodule -r base bpf_token_perms to
undo the two module changes and restore your all_perms.spt file from
the saved .orig file.

Changes in v2:
- Removed allow rule for 'kernel_t' in test_bpf.te which was added due
  to a bug in the kernel
- Cleaned up other unnecessary rules in test_bpf.te
- Added token_test.c which was missing from previous patch

Changes in v3:
- Added original license in 'token_test.c'
- Updated patch description to
    - replace 'base.sil' with 'base.cil'
    - Remove extra quotation mark in 'sudo 'sed -i "s/"map_create'

Signed-off-by: Eric Suen <ericsu@linux.microsoft.com>
Tested-by: Daniel Durning <danieldurning.work@gmail.com>
---
 policy/test_bpf.te     |  39 +++
 tests/bpf/Makefile     |   5 +-
 tests/bpf/bpf_common.h |  10 +
 tests/bpf/bpf_test.c   |  59 +++--
 tests/bpf/test         |  21 +-
 tests/bpf/token_test.c | 543 +++++++++++++++++++++++++++++++++++++++++
 6 files changed, 658 insertions(+), 19 deletions(-)
 create mode 100644 tests/bpf/token_test.c

diff --git a/policy/test_bpf.te b/policy/test_bpf.te
index 5eab0bd..45da852 100644
--- a/policy/test_bpf.te
+++ b/policy/test_bpf.te
@@ -57,3 +57,42 @@ typeattribute test_bpf_deny_prog_run_t bpfdomain;
 allow test_bpf_deny_prog_run_t self:process { setrlimit };
 allow test_bpf_deny_prog_run_t self:capability { sys_resource sys_admin };
 allow test_bpf_deny_prog_run_t self:bpf { map_create map_read map_write prog_load };
+
+################### Allow map_create_as and prog_load_as ###################
+fs_list_bpf_dirs(test_bpf_t)
+allow test_bpf_t bpf_t:filesystem mount;
+allow test_bpf_t root_t:dir mounton;
+allow test_bpf_t self:bpf { map_create_as prog_load_as };
+allow test_bpf_t self:cap2_userns { bpf perfmon };
+allow test_bpf_t self:cap_userns { net_admin setgid setuid sys_admin };
+allow test_bpf_t self:user_namespace create;
+
+############################ Deny map_create_as ############################
+type test_bpf_deny_map_create_as_t;
+testsuite_domain_type(test_bpf_deny_map_create_as_t)
+typeattribute test_bpf_deny_map_create_as_t bpfdomain;
+allow test_bpf_deny_map_create_as_t self:process { setrlimit };
+allow test_bpf_deny_map_create_as_t self:capability { sys_resource sys_admin };
+
+fs_list_bpf_dirs(test_bpf_deny_map_create_as_t)
+allow test_bpf_deny_map_create_as_t bpf_t:filesystem mount;
+allow test_bpf_deny_map_create_as_t root_t:dir mounton;
+allow test_bpf_deny_map_create_as_t self:bpf { map_create map_read map_write prog_load prog_load_as };
+allow test_bpf_deny_map_create_as_t self:cap2_userns { bpf };
+allow test_bpf_deny_map_create_as_t self:cap_userns { setgid setuid sys_admin };
+allow test_bpf_deny_map_create_as_t self:user_namespace create;
+
+############################ Deny prog_load_as #############################
+type test_bpf_deny_prog_load_as_t;
+testsuite_domain_type(test_bpf_deny_prog_load_as_t)
+typeattribute test_bpf_deny_prog_load_as_t bpfdomain;
+allow test_bpf_deny_prog_load_as_t self:process { setrlimit };
+allow test_bpf_deny_prog_load_as_t self:capability { sys_resource sys_admin };
+
+fs_list_bpf_dirs(test_bpf_deny_prog_load_as_t)
+allow test_bpf_deny_prog_load_as_t bpf_t:filesystem mount;
+allow test_bpf_deny_prog_load_as_t root_t:dir mounton;
+allow test_bpf_deny_prog_load_as_t self:bpf { map_create map_create_as map_read map_write prog_load };
+allow test_bpf_deny_prog_load_as_t self:cap2_userns { bpf perfmon };
+allow test_bpf_deny_prog_load_as_t self:cap_userns { net_admin setgid setuid sys_admin };
+allow test_bpf_deny_prog_load_as_t self:user_namespace create;
diff --git a/tests/bpf/Makefile b/tests/bpf/Makefile
index 1ae8ce9..cacefbe 100644
--- a/tests/bpf/Makefile
+++ b/tests/bpf/Makefile
@@ -1,5 +1,5 @@
 TARGETS = bpf_test
-DEPS = bpf_common.c bpf_common.h
+SRCS = bpf_test.c bpf_common.c token_test.c
 LDLIBS += -lselinux -lbpf
 
 # export so that BPF_ENABLED entries get built correctly on local build
@@ -14,4 +14,5 @@ clean:
 	rm -f $(TARGETS) test_sock flag *_flag
 	@set -e; for i in $(BPF_ENABLED); do $(MAKE) -C $$i clean ; done
 
-$(TARGETS): $(DEPS)
+$(TARGETS): $(SRCS)
+	$(CC) $(CFLAGS) -o $@ $^ $(LDLIBS)
\ No newline at end of file
diff --git a/tests/bpf/bpf_common.h b/tests/bpf/bpf_common.h
index 44ac28f..adba522 100644
--- a/tests/bpf/bpf_common.h
+++ b/tests/bpf/bpf_common.h
@@ -12,6 +12,8 @@
 extern int create_bpf_map(void);
 extern int create_bpf_prog(void);
 extern void bpf_setrlimit(void);
+extern int test_bpf_map_create(void);
+extern int test_bpf_prog_load(void);
 
 /* edited eBPF instruction library */
 /* Short form of mov, dst_reg = imm32 */
@@ -32,3 +34,11 @@ extern void bpf_setrlimit(void);
 					       .off   = 0,			\
 							.imm   = 0 })
 
+/* Raw code statement block */
+#define BPF_RAW_INSN(CODE, DST, SRC, OFF, IMM)			\
+	((struct bpf_insn) {					\
+		.code  = CODE,					\
+			.dst_reg = DST,					\
+				.src_reg = SRC,					\
+					.off   = OFF,					\
+						.imm   = IMM })
diff --git a/tests/bpf/bpf_test.c b/tests/bpf/bpf_test.c
index 3c6a29c..a8dc383 100644
--- a/tests/bpf/bpf_test.c
+++ b/tests/bpf/bpf_test.c
@@ -1,28 +1,38 @@
 #include "bpf_common.h"
 
+#define write_verbose(verbose, fmt, ...) \
+	do { \
+		if (verbose) \
+			printf(fmt "\n", ##__VA_ARGS__); \
+	} while (0)
+
 static void usage(char *progname)
 {
 	fprintf(stderr,
-		"usage:  %s -m|-p [-v]\n"
+		"usage:  %s -m|-p|-c|-l [-v]\n"
 		"Where:\n\t"
 		"-m    Create BPF map fd\n\t"
 		"-p    Create BPF prog fd\n\t"
+		"-c    Test BPF token map create\n\t"
+		"-l    Test BPF token program load\n\t"
 		"-v Print information.\n", progname);
 	exit(-1);
 }
 
 int main(int argc, char *argv[])
 {
-	int opt, result, fd;
-	bool verbose = false;
+	int opt, result, ret;
+	bool verbose = false, is_fd = true;
 	char *context;
 
 	enum {
 		MAP_FD = 1,
-		PROG_FD
+		PROG_FD,
+		MAP_CREATE,
+		PROG_LOAD,
 	} bpf_fd_type;
 
-	while ((opt = getopt(argc, argv, "mpv")) != -1) {
+	while ((opt = getopt(argc, argv, "mpclv")) != -1) {
 		switch (opt) {
 		case 'm':
 			bpf_fd_type = MAP_FD;
@@ -30,6 +40,12 @@ int main(int argc, char *argv[])
 		case 'p':
 			bpf_fd_type = PROG_FD;
 			break;
+		case 'c':
+			bpf_fd_type = MAP_CREATE;
+			break;
+		case 'l':
+			bpf_fd_type = PROG_LOAD;
+			break;
 		case 'v':
 			verbose = true;
 			break;
@@ -44,8 +60,7 @@ int main(int argc, char *argv[])
 		exit(-1);
 	}
 
-	if (verbose)
-		printf("Process context:\n\t%s\n", context);
+	write_verbose(verbose, "Process context:\n\n%s", context);
 
 	free(context);
 
@@ -54,24 +69,36 @@ int main(int argc, char *argv[])
 
 	switch (bpf_fd_type) {
 	case MAP_FD:
-		if (verbose)
-			printf("Creating BPF map\n");
+		write_verbose(verbose, "Creating BPF map");
 
-		fd = create_bpf_map();
+		ret = create_bpf_map();
 		break;
 	case PROG_FD:
-		if (verbose)
-			printf("Creating BPF prog\n");
+		write_verbose(verbose, "Creating BPF prog");
+
+		ret = create_bpf_prog();
+		break;
+	case MAP_CREATE:
+		is_fd = false;
+		write_verbose(verbose, "Testing BPF map create");
+
+		ret = test_bpf_map_create();
+		break;
+	case PROG_LOAD:
+		is_fd = false;
+		write_verbose(verbose, "Testing BPF prog load");
 
-		fd = create_bpf_prog();
+		ret = test_bpf_prog_load();
 		break;
 	default:
 		usage(argv[0]);
 	}
 
-	if (fd < 0)
-		return fd;
+	if (ret < 0)
+		return ret;
+
+	if (is_fd)
+		close(ret);
 
-	close(fd);
 	return 0;
 }
diff --git a/tests/bpf/test b/tests/bpf/test
index a3fd856..14bda32 100755
--- a/tests/bpf/test
+++ b/tests/bpf/test
@@ -9,8 +9,10 @@ BEGIN {
 
     $test_bpf_count       = 7;
     $test_fdreceive_count = 4;
+    $test_bpf_token_count = 4;
 
-    $test_count = $test_bpf_count + $test_fdreceive_count;
+    $test_count =
+      $test_bpf_count + $test_fdreceive_count + $test_bpf_token_count;
 
     # allow info to be shown during tests
     $v = $ARGV[0];
@@ -67,6 +69,13 @@ ok( $result eq 0 );
 $result = system "runcon -t test_bpf_t $basedir/bpf_test -p $v";
 ok( $result eq 0 );
 
+# BPF token - BPF_MAP_CREATE_AS, BPF_PROG_LOAD_AS
+$result = system "runcon -t test_bpf_t $basedir/bpf_test -c $v";
+ok( $result eq 0 );
+
+$result = system "runcon -t test_bpf_t $basedir/bpf_test -l $v";
+ok( $result eq 0 );
+
 # Deny map_create permission
 $result =
   system "runcon -t test_bpf_deny_map_create_t $basedir/bpf_test -m $v 2>&1";
@@ -92,6 +101,16 @@ $result =
   system "runcon -t test_bpf_deny_prog_run_t $basedir/bpf_test -p $v 2>&1";
 ok($result);
 
+# BPF token - deny BPF_MAP_CREATE_AS
+$result =
+  system "runcon -t test_bpf_deny_map_create_as_t $basedir/bpf_test -c $v 2>&1";
+ok($result);
+
+# BPF token - deny BPF_PROG_LOAD_AS
+$result =
+  system "runcon -t test_bpf_deny_prog_load_as_t $basedir/bpf_test -l $v 2>&1";
+ok($result);
+
 #
 ################ BPF Tests for fdreceive #######################
 #
diff --git a/tests/bpf/token_test.c b/tests/bpf/token_test.c
new file mode 100644
index 0000000..8e2659b
--- /dev/null
+++ b/tests/bpf/token_test.c
@@ -0,0 +1,543 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Code derived from: linux/source/tools/testing/selftests/bpf/prog_tests/token.c
+ * Copyright (c) 2023 Meta Platforms, Inc. and affiliates.
+ */
+
+#include "bpf_common.h"
+#include "signal.h"
+#include "linux/mount.h"
+#include <linux/unistd.h>
+#include "sys/wait.h"
+#include "sys/socket.h"
+#include "fcntl.h"
+#include "sched.h"
+#include <bpf/btf.h>
+
+#define bit(n) (1ULL << (n))
+
+#define zclose(fd) do { if (fd >= 0) close(fd); fd = -1; } while (0)
+#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
+
+#define _CHECK(condition, format...) ({    \
+	int __ret = !!(condition); \
+	int __save_errno = errno;   \
+	if (__ret) {    \
+		fprintf(stderr, ##format);   \
+	}   \
+	errno = __save_errno;   \
+	__ret;  \
+})
+
+#define ASSERT_OK(res, name) ({     \
+	long long ___res = (res);       \
+	bool ___ok = ___res == 0;       \
+	_CHECK(!___ok, \
+			"%s failed. unexpected error: %lld (errno %d)\n",  \
+			 name, ___res, errno);  \
+	___ok;                          \
+})
+
+#define ASSERT_GT(actual, expected, name) ({    \
+	typeof(actual) ___act = (actual);   \
+	typeof(expected) ___exp = (expected);   \
+	bool ___ok = ___act > ___exp;       \
+	_CHECK(!___ok,  \
+			"unexpected %s: actual %lld <= expected %lld (errno %d)\n",   \
+			(name), (long long)(___act), (long long)(___exp), errno);  \
+	___ok;  \
+})
+
+#define ASSERT_GE(actual, expected, name) ({    \
+	typeof(actual) ___act = (actual);       \
+	typeof(expected) ___exp = (expected);   \
+	bool ___ok = ___act >= ___exp;          \
+	_CHECK(!___ok,  \
+			"unexpected %s: actual %lld < expected %lld (errno %d)\n",   \
+			(name), (long long)(___act), (long long)(___exp), errno);          \
+	___ok;  \
+})
+
+#define ASSERT_EQ(actual, expected, name) ({    \
+	typeof(actual) ___act = (actual);           \
+	typeof(expected) ___exp = (expected);       \
+	bool ___ok = ___act == ___exp;              \
+	_CHECK(!___ok,   \
+			"unexpected %s: actual %lld != expected %lld (errno %d)\n",   \
+			(name), (long long)(___act), (long long)(___exp), errno);          \
+	___ok;  \
+})
+
+#define ASSERT_OK_PTR(ptr, name) ({     \
+	const void *___res = (ptr);         \
+	int ___err = libbpf_get_error(___res);  \
+	bool ___ok = ___err == 0;           \
+	_CHECK(!___ok,  \
+			"%s unexpected error: %d\n", name, ___err);  \
+	___ok;      \
+})
+
+struct bpffs_opts {
+	__u64 cmds;
+	__u64 maps;
+	__u64 progs;
+	__u64 attachs;
+	const char *cmds_str;
+	const char *maps_str;
+	const char *progs_str;
+	const char *attachs_str;
+};
+
+typedef int (*child_callback_fn)(int bpffs_fd);
+
+static inline int sys_mount(const char *dev_name, const char *dir_name,
+			    const char *type, unsigned long flags,
+			    const void *data)
+{
+	return syscall(__NR_mount, dev_name, dir_name, type, flags, data);
+}
+
+static inline int sys_fsopen(const char *fsname, unsigned int flags)
+{
+	return syscall(__NR_fsopen, fsname, flags);
+}
+
+static inline int sys_fsconfig(int fs_fd, unsigned int cmd, const char *key,
+			       const void *val, int aux)
+{
+	return syscall(__NR_fsconfig, fs_fd, cmd, key, val, aux);
+}
+
+static inline int sys_fsmount(int fs_fd, unsigned int flags,
+			      unsigned int ms_flags)
+{
+	return syscall(__NR_fsmount, fs_fd, flags, ms_flags);
+}
+
+static int set_delegate_mask(int fs_fd, const char *key, __u64 mask,
+			     const char *mask_str)
+{
+	char buf[32];
+	int err;
+
+	if (!mask_str) {
+		if (mask == ~0ULL) {
+			mask_str = "any";
+		} else {
+			snprintf(buf, sizeof(buf), "0x%llx", (unsigned long long)mask);
+			mask_str = buf;
+		}
+	}
+
+	err = sys_fsconfig(fs_fd, FSCONFIG_SET_STRING, key,
+			   mask_str, 0);
+	if (err < 0)
+		err = -errno;
+	return err;
+}
+
+static int create_bpffs_fd(void)
+{
+	int fs_fd;
+
+	/* create VFS context */
+	fs_fd = sys_fsopen("bpf", 0);
+	ASSERT_GE(fs_fd, 0, "fs_fd");
+
+	return fs_fd;
+}
+
+static int materialize_bpffs_fd(int fs_fd, struct bpffs_opts *opts)
+{
+	int mnt_fd, err;
+
+	/* set up token delegation mount options */
+	err = set_delegate_mask(fs_fd, "delegate_cmds", opts->cmds, opts->cmds_str);
+	if (!ASSERT_OK(err, "fs_cfg_cmd"))
+		return err;
+	err = set_delegate_mask(fs_fd, "delegate_maps", opts->maps, opts->maps_str);
+	if (!ASSERT_OK(err, "fs_cfg_maps"))
+		return err;
+	err = set_delegate_mask(fs_fd, "delegate_progs", opts->progs, opts->progs_str);
+	if (!ASSERT_OK(err, "fs_cfg_progs"))
+		return err;
+	err = set_delegate_mask(fs_fd, "delegate_attachs", opts->attachs,
+				opts->attachs_str);
+	if (!ASSERT_OK(err, "fs_cfg_attachs"))
+		return err;
+
+	/* instantiate FS object */
+	err = sys_fsconfig(fs_fd, FSCONFIG_CMD_CREATE, NULL, NULL, 0);
+	if (err < 0)
+		return -errno;
+
+	/* create O_PATH fd for detached mount */
+	mnt_fd = sys_fsmount(fs_fd, 0, 0);
+	if (err < 0)
+		return -errno;
+
+	return mnt_fd;
+}
+
+static ssize_t write_nointr(int fd, const void *buf, size_t count)
+{
+	ssize_t ret;
+
+	do {
+		ret = write(fd, buf, count);
+	} while (ret < 0 && errno == EINTR);
+
+	return ret;
+}
+
+static int write_file(const char *path, const void *buf, size_t count)
+{
+	int fd;
+	ssize_t ret;
+
+	fd = open(path, O_WRONLY | O_CLOEXEC | O_NOCTTY | O_NOFOLLOW);
+	if (fd < 0)
+		return -1;
+
+	ret = write_nointr(fd, buf, count);
+	close(fd);
+	if (ret < 0 || (size_t)ret != count)
+		return -1;
+
+	return 0;
+}
+
+static int create_and_enter_userns(void)
+{
+	uid_t uid;
+	gid_t gid;
+	char map[100];
+
+	uid = getuid();
+	gid = getgid();
+
+	if (unshare(CLONE_NEWUSER))
+		return -1;
+
+	if (write_file("/proc/self/setgroups", "deny", sizeof("deny") - 1) &&
+	    errno != ENOENT)
+		return -1;
+
+	snprintf(map, sizeof(map), "0 %d 1", uid);
+	if (write_file("/proc/self/uid_map", map, strlen(map)))
+		return -1;
+
+
+	snprintf(map, sizeof(map), "0 %d 1", gid);
+	if (write_file("/proc/self/gid_map", map, strlen(map)))
+		return -1;
+
+	if (setgid(0))
+		return -1;
+
+	if (setuid(0))
+		return -1;
+
+	return 0;
+}
+
+static int sendfd(int sockfd, int fd)
+{
+	struct msghdr msg = {};
+	struct cmsghdr *cmsg;
+	int fds[1] = { fd }, err;
+	char iobuf[1];
+	struct iovec io = {
+		.iov_base = iobuf,
+		.iov_len = sizeof(iobuf),
+	};
+	union {
+		char buf[CMSG_SPACE(sizeof(fds))];
+		struct cmsghdr align;
+	} u;
+
+	msg.msg_iov = &io;
+	msg.msg_iovlen = 1;
+	msg.msg_control = u.buf;
+	msg.msg_controllen = sizeof(u.buf);
+	cmsg = CMSG_FIRSTHDR(&msg);
+	cmsg->cmsg_level = SOL_SOCKET;
+	cmsg->cmsg_type = SCM_RIGHTS;
+	cmsg->cmsg_len = CMSG_LEN(sizeof(fds));
+	memcpy(CMSG_DATA(cmsg), fds, sizeof(fds));
+
+	err = sendmsg(sockfd, &msg, 0);
+	if (err < 0)
+		err = -errno;
+	if (!ASSERT_EQ(err, 1, "sendmsg"))
+		return -EINVAL;
+
+	return 0;
+}
+
+static int recvfd(int sockfd, int *fd)
+{
+	struct msghdr msg = {};
+	struct cmsghdr *cmsg;
+	int fds[1], err;
+	char iobuf[1];
+	struct iovec io = {
+		.iov_base = iobuf,
+		.iov_len = sizeof(iobuf),
+	};
+	union {
+		char buf[CMSG_SPACE(sizeof(fds))];
+		struct cmsghdr align;
+	} u;
+
+	msg.msg_iov = &io;
+	msg.msg_iovlen = 1;
+	msg.msg_control = u.buf;
+	msg.msg_controllen = sizeof(u.buf);
+
+	err = recvmsg(sockfd, &msg, 0);
+	if (err < 0)
+		err = -errno;
+	if (!ASSERT_EQ(err, 1, "recvmsg"))
+		return -EINVAL;
+
+	cmsg = CMSG_FIRSTHDR(&msg);
+	if (!ASSERT_OK_PTR(cmsg, "cmsg_null") ||
+	    !ASSERT_EQ(cmsg->cmsg_len, CMSG_LEN(sizeof(fds)), "cmsg_len") ||
+	    !ASSERT_EQ(cmsg->cmsg_level, SOL_SOCKET, "cmsg_level") ||
+	    !ASSERT_EQ(cmsg->cmsg_type, SCM_RIGHTS, "cmsg_type"))
+		return -EINVAL;
+
+	memcpy(fds, CMSG_DATA(cmsg), sizeof(fds));
+	*fd = fds[0];
+
+	return 0;
+}
+
+static int wait_for_pid(pid_t pid)
+{
+	int status, ret;
+
+again:
+	ret = waitpid(pid, &status, 0);
+	if (ret == -1) {
+		if (errno == EINTR)
+			goto again;
+
+		return -1;
+	}
+
+	if (!WIFEXITED(status))
+		return -1;
+
+	return WEXITSTATUS(status);
+}
+
+static int child(int sock_fd, struct bpffs_opts *bpffs_opts,
+		 child_callback_fn callback)
+{
+	int mnt_fd = -1, fs_fd = -1, err = 0, bpffs_fd = -1, token_fd = -1;
+
+	err = create_and_enter_userns();
+	if (!ASSERT_OK(err, "create_and_enter_userns"))
+		goto cleanup;
+
+	err = unshare(CLONE_NEWNS);
+	if (!ASSERT_OK(err, "create_mountns"))
+		goto cleanup;
+
+	err = sys_mount(NULL, "/", NULL, MS_REC | MS_PRIVATE, 0);
+	if (!ASSERT_OK(err, "remount_root"))
+		goto cleanup;
+
+	fs_fd = create_bpffs_fd();
+	if (!ASSERT_GE(fs_fd, 0, "create_bpffs_fd")) {
+		err = -EINVAL;
+		goto cleanup;
+	}
+
+	err = sendfd(sock_fd, fs_fd);
+	if (!ASSERT_OK(err, "send_fs_fd"))
+		goto cleanup;
+	zclose(fs_fd);
+
+	err = recvfd(sock_fd, &mnt_fd);
+	if (!ASSERT_OK(err, "recv_mnt_fd"))
+		goto cleanup;
+
+	bpffs_fd = openat(mnt_fd, ".", 0, O_RDWR);
+	if (!ASSERT_GE(bpffs_fd, 0, "bpffs_open")) {
+		err = -EINVAL;
+		goto cleanup;
+	}
+
+	err = callback(bpffs_fd);
+	if (!ASSERT_OK(err, "test_callback"))
+		goto cleanup;
+
+	err = 0;
+
+cleanup:
+	zclose(sock_fd);
+	zclose(mnt_fd);
+	zclose(fs_fd);
+	zclose(bpffs_fd);
+	zclose(token_fd);
+
+	exit(-err);
+}
+
+static int parent(int child_pid, struct bpffs_opts *bpffs_opts, int sock_fd)
+{
+	int fs_fd = -1, mnt_fd = -1, token_fd = -1, err;
+
+	err = recvfd(sock_fd, &fs_fd);
+	if (!ASSERT_OK(err, "recv_bpffs_fd"))
+		goto cleanup;
+
+	mnt_fd = materialize_bpffs_fd(fs_fd, bpffs_opts);
+	if (!ASSERT_GE(mnt_fd, 0, "materialize_bpffs_fd")) {
+		err = -EINVAL;
+		goto cleanup;
+	}
+	zclose(fs_fd);
+
+	err = sendfd(sock_fd, mnt_fd);
+	if (!ASSERT_OK(err, "send_mnt_fd"))
+		goto cleanup;
+	zclose(mnt_fd);
+
+	err = wait_for_pid(child_pid);
+	if (!ASSERT_OK(err, "waitpid_child")) {
+		err = -EINVAL;
+		goto cleanup;
+	}
+
+cleanup:
+	zclose(sock_fd);
+	zclose(fs_fd);
+	zclose(mnt_fd);
+	zclose(token_fd);
+
+	if (child_pid > 0)
+		(void)kill(child_pid, SIGKILL);
+
+	return err;
+}
+
+static int subtest(struct bpffs_opts *bpffs_opts, child_callback_fn child_cb)
+{
+	int sock_fds[2] = { -1, -1 };
+	int child_pid = 0, err;
+
+	err = socketpair(AF_UNIX, SOCK_STREAM, 0, sock_fds);
+	if (!ASSERT_OK(err, "socketpair"))
+		goto cleanup;
+
+	child_pid = fork();
+	if (!ASSERT_GE(child_pid, 0, "fork"))
+		goto cleanup;
+
+	if (child_pid == 0) {
+		zclose(sock_fds[0]);
+		return child(sock_fds[1], bpffs_opts, child_cb);
+	} else {
+		zclose(sock_fds[1]);
+		return parent(child_pid, bpffs_opts, sock_fds[0]);
+	}
+
+cleanup:
+	zclose(sock_fds[0]);
+	zclose(sock_fds[1]);
+	if (child_pid > 0)
+		(void)kill(child_pid, SIGKILL);
+
+	return -err;
+}
+
+static int userns_map_create(int mnt_fd)
+{
+	LIBBPF_OPTS(bpf_map_create_opts, map_opts);
+	int err = 0, token_fd = -1, map_fd = -1;
+
+	/* create BPF token from BPF FS mount */
+	token_fd = bpf_token_create(mnt_fd, NULL);
+	if (!ASSERT_GT(token_fd, 0, "userns_map_create/token_create")) {
+		err = -EINVAL;
+		goto cleanup;
+	}
+
+	map_opts.map_flags = BPF_F_TOKEN_FD;
+	map_opts.token_fd = token_fd;
+	map_fd = bpf_map_create(BPF_MAP_TYPE_STACK, "userns_map_create", 0, 8, 1,
+				&map_opts);
+	if (!ASSERT_GT(map_fd, 0, "userns_map_create/bpf_map_create")) {
+		err = -EINVAL;
+		goto cleanup;
+	}
+
+cleanup:
+	zclose(token_fd);
+	zclose(map_fd);
+
+	return err;
+}
+
+static int userns_prog_load(int mnt_fd)
+{
+	LIBBPF_OPTS(bpf_prog_load_opts, prog_opts);
+	int err, token_fd = -1, prog_fd = -1;
+	struct bpf_insn insns[] = {
+		/* bpf_jiffies64() requires CAP_BPF */
+		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_jiffies64),
+		/* bpf_get_current_task() requires CAP_PERFMON */
+		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_current_task),
+		/* r0 = 0; exit; */
+		BPF_MOV64_IMM(BPF_REG_0, 0),
+		BPF_EXIT_INSN(),
+	};
+	size_t insn_cnt = ARRAY_SIZE(insns);
+
+	token_fd = bpf_token_create(mnt_fd, NULL);
+	if (!ASSERT_GT(token_fd, 0, "userns_prog_load/token_create")) {
+		err = -EINVAL;
+		goto cleanup;
+	}
+
+	prog_opts.prog_flags = BPF_F_TOKEN_FD;
+	prog_opts.token_fd = token_fd;
+	prog_opts.expected_attach_type = BPF_XDP;
+	prog_fd = bpf_prog_load(BPF_PROG_TYPE_XDP, "token_prog", "GPL",
+				insns, insn_cnt, &prog_opts);
+	if (!ASSERT_GT(prog_fd, 0, "userns_prog_load/bpf_prog_load")) {
+		err = -EPERM;
+		goto cleanup;
+	}
+
+	err = 0;
+
+cleanup:
+	zclose(prog_fd);
+	zclose(token_fd);
+	return err;
+}
+
+int test_bpf_map_create(void)
+{
+	struct bpffs_opts opts = {
+		.cmds_str = "map_create",
+		.maps_str = "stack"
+	};
+
+	return subtest(&opts, userns_map_create);
+}
+
+int test_bpf_prog_load(void)
+{
+	struct bpffs_opts opts = {
+		.cmds_str = "prog_load",
+		.progs_str = "XDP",
+		.attachs_str = "xdp",
+	};
+
+	return subtest(&opts, userns_prog_load);
+}
-- 
2.50.1.windows.1


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH testsuite v3] tests/bpf: Add tests for SELinux BPF token access control
  2025-08-21 19:03 [PATCH testsuite v3] tests/bpf: Add tests for SELinux BPF token access control Eric Suen
@ 2025-08-25 14:21 ` Stephen Smalley
  2025-08-29 16:39   ` Eric Suen
  0 siblings, 1 reply; 3+ messages in thread
From: Stephen Smalley @ 2025-08-25 14:21 UTC (permalink / raw)
  To: Eric Suen; +Cc: selinux, paul, omosnace, danieldurning.work

On Thu, Aug 21, 2025 at 3:03 PM Eric Suen <ericsu@linux.microsoft.com> wrote:
>
> This patch adds new tests to verify the SELinux support for BPF token
> access control, as introduced in the corresponding kernel patch:
>   https://lore.kernel.org/selinux/20250816201420.197-1-ericsu@linux.microsoft.com/
>

No need to re-spin just for this, but this also depends on the
corresponding userspace patch for libsepol.

> Four new tests are added to cover both positive and negative scenarios,
> ensuring that the SELinux policy enforcement on BPF token usage behaves
> as expected.
>   - Successful map_create and prog_load when SELinux permissions are
>     granted.
>   - Enforcement of SELinux policy restrictions when access is denied.
>
> For testing purposes, you can update the base policy by manually
> modifying your base module and tweaking /usr/share/selinux/devel as
> follows:
>   sudo semodule -c -E base
>   sudo cp base.cil base.cil.orig
>   sudo sed -i "s/map_create/map_create map_create_as/" base.cil
>   sudo sed -i "s/prog_load/prog_load prog_load_as/" base.cil
>   sudo semodule -i base.cil
>   echo "(policycap bpf_token_perms)" > bpf_token_perms.cil
>   sudo semodule -i bpf_token_perms.cil
>   sudo cp /usr/share/selinux/devel/include/support/all_perms.spt \
>       /usr/share/selinux/devel/include/support/all_perms.spt.orig
>   sudo sed -i "s/map_create/map_create map_create_as/" \
>       /usr/share/selinux/devel/include/support/all_perms.spt
>   sudo sed -i "s/prog_load/prog_load prog_load_as/" \
>       /usr/share/selinux/devel/include/support/all_perms.spt
>
> When finished testing, you can semodule -r base bpf_token_perms to
> undo the two module changes and restore your all_perms.spt file from
> the saved .orig file.
>
> Changes in v2:
> - Removed allow rule for 'kernel_t' in test_bpf.te which was added due
>   to a bug in the kernel
> - Cleaned up other unnecessary rules in test_bpf.te
> - Added token_test.c which was missing from previous patch
>
> Changes in v3:
> - Added original license in 'token_test.c'
> - Updated patch description to
>     - replace 'base.sil' with 'base.cil'
>     - Remove extra quotation mark in 'sudo 'sed -i "s/"map_create'
>
> Signed-off-by: Eric Suen <ericsu@linux.microsoft.com>
> Tested-by: Daniel Durning <danieldurning.work@gmail.com>
> ---
>  policy/test_bpf.te     |  39 +++
>  tests/bpf/Makefile     |   5 +-
>  tests/bpf/bpf_common.h |  10 +
>  tests/bpf/bpf_test.c   |  59 +++--
>  tests/bpf/test         |  21 +-
>  tests/bpf/token_test.c | 543 +++++++++++++++++++++++++++++++++++++++++
>  6 files changed, 658 insertions(+), 19 deletions(-)
>  create mode 100644 tests/bpf/token_test.c

If I run the tests manually ala "sudo ./tests/bpf/test", I see the
following output during the new tests:
ok 9
unexpected userns_map_create/token_create: actual -13 <= expected 0 (errno 13)
test_callback failed. unexpected error: -22 (errno 13)
waitpid_child failed. unexpected error: 22 (errno 2)
ok 10
unexpected userns_prog_load/token_create: actual -13 <= expected 0 (errno 13)
test_callback failed. unexpected error: -22 (errno 13)
waitpid_child failed. unexpected error: 22 (errno 2)
ok 11

Generally we try to avoid/minimize spurious output from the tests for
legitimate errors, so ideally you would eliminate this noise if these
are in fact operating as intended.

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH testsuite v3] tests/bpf: Add tests for SELinux BPF token access control
  2025-08-25 14:21 ` Stephen Smalley
@ 2025-08-29 16:39   ` Eric Suen
  0 siblings, 0 replies; 3+ messages in thread
From: Eric Suen @ 2025-08-29 16:39 UTC (permalink / raw)
  To: Stephen Smalley; +Cc: selinux, paul, omosnace, danieldurning.work

On 8/25/2025 7:21 AM, Stephen Smalley wrote:
> On Thu, Aug 21, 2025 at 3:03 PM Eric Suen <ericsu@linux.microsoft.com> wrote:
>> This patch adds new tests to verify the SELinux support for BPF token
>> access control, as introduced in the corresponding kernel patch:
>>    https://lore.kernel.org/selinux/20250816201420.197-1-ericsu@linux.microsoft.com/
>>
> No need to re-spin just for this, but this also depends on the
> corresponding userspace patch for libsepol.
>
>> Four new tests are added to cover both positive and negative scenarios,
>> ensuring that the SELinux policy enforcement on BPF token usage behaves
>> as expected.
>>    - Successful map_create and prog_load when SELinux permissions are
>>      granted.
>>    - Enforcement of SELinux policy restrictions when access is denied.
>>
>> For testing purposes, you can update the base policy by manually
>> modifying your base module and tweaking /usr/share/selinux/devel as
>> follows:
>>    sudo semodule -c -E base
>>    sudo cp base.cil base.cil.orig
>>    sudo sed -i "s/map_create/map_create map_create_as/" base.cil
>>    sudo sed -i "s/prog_load/prog_load prog_load_as/" base.cil
>>    sudo semodule -i base.cil
>>    echo "(policycap bpf_token_perms)" > bpf_token_perms.cil
>>    sudo semodule -i bpf_token_perms.cil
>>    sudo cp /usr/share/selinux/devel/include/support/all_perms.spt \
>>        /usr/share/selinux/devel/include/support/all_perms.spt.orig
>>    sudo sed -i "s/map_create/map_create map_create_as/" \
>>        /usr/share/selinux/devel/include/support/all_perms.spt
>>    sudo sed -i "s/prog_load/prog_load prog_load_as/" \
>>        /usr/share/selinux/devel/include/support/all_perms.spt
>>
>> When finished testing, you can semodule -r base bpf_token_perms to
>> undo the two module changes and restore your all_perms.spt file from
>> the saved .orig file.
>>
>> Changes in v2:
>> - Removed allow rule for 'kernel_t' in test_bpf.te which was added due
>>    to a bug in the kernel
>> - Cleaned up other unnecessary rules in test_bpf.te
>> - Added token_test.c which was missing from previous patch
>>
>> Changes in v3:
>> - Added original license in 'token_test.c'
>> - Updated patch description to
>>      - replace 'base.sil' with 'base.cil'
>>      - Remove extra quotation mark in 'sudo 'sed -i "s/"map_create'
>>
>> Signed-off-by: Eric Suen <ericsu@linux.microsoft.com>
>> Tested-by: Daniel Durning <danieldurning.work@gmail.com>
>> ---
>>   policy/test_bpf.te     |  39 +++
>>   tests/bpf/Makefile     |   5 +-
>>   tests/bpf/bpf_common.h |  10 +
>>   tests/bpf/bpf_test.c   |  59 +++--
>>   tests/bpf/test         |  21 +-
>>   tests/bpf/token_test.c | 543 +++++++++++++++++++++++++++++++++++++++++
>>   6 files changed, 658 insertions(+), 19 deletions(-)
>>   create mode 100644 tests/bpf/token_test.c
> If I run the tests manually ala "sudo ./tests/bpf/test", I see the
> following output during the new tests:
> ok 9
> unexpected userns_map_create/token_create: actual -13 <= expected 0 (errno 13)
> test_callback failed. unexpected error: -22 (errno 13)
> waitpid_child failed. unexpected error: 22 (errno 2)
> ok 10
> unexpected userns_prog_load/token_create: actual -13 <= expected 0 (errno 13)
> test_callback failed. unexpected error: -22 (errno 13)
> waitpid_child failed. unexpected error: 22 (errno 2)
> ok 11
>
> Generally we try to avoid/minimize spurious output from the tests for
> legitimate errors, so ideally you would eliminate this noise if these
> are in fact operating as intended.

Stephen - apologies for the delayed response. Yes, that's a great 
feedback. Let me look into it and update in next patch.


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2025-08-29 16:39 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-08-21 19:03 [PATCH testsuite v3] tests/bpf: Add tests for SELinux BPF token access control Eric Suen
2025-08-25 14:21 ` Stephen Smalley
2025-08-29 16:39   ` Eric Suen

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).