Merge tag 'seccomp-v6.9-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux

Pull seccomp updates from Kees Cook:
 "There are no core kernel changes here; it's entirely selftests and
  samples:

   - Improve reliability of selftests (Terry Tritton, Kees Cook)

   - Fix strict-aliasing warning in samples (Arnd Bergmann)"

* tag 'seccomp-v6.9-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
  samples: user-trap: fix strict-aliasing warning
  selftests/seccomp: Pin benchmark to single CPU
  selftests/seccomp: user_notification_addfd check nextfd is available
  selftests/seccomp: Change the syscall used in KILL_THREAD test
  selftests/seccomp: Handle EINVAL on unshare(CLONE_NEWPID)

Author: torvalds

samples/seccomp/user-trap.c

@@ -33,6 +33,7 @@ static int send_fd(int sock, int fd)
 {
 	struct msghdr msg = {};
 	struct cmsghdr *cmsg;
+	int *fd_ptr;
 	char buf[CMSG_SPACE(sizeof(int))] = {0}, c = 'c';
 	struct iovec io = {
 		.iov_base = &c,
@@ -47,7 +48,8 @@ static int send_fd(int sock, int fd)
 	cmsg->cmsg_level = SOL_SOCKET;
 	cmsg->cmsg_type = SCM_RIGHTS;
 	cmsg->cmsg_len = CMSG_LEN(sizeof(int));
-	*((int *)CMSG_DATA(cmsg)) = fd;
+	fd_ptr = (int *)CMSG_DATA(cmsg);
+	*fd_ptr = fd;
 	msg.msg_controllen = cmsg->cmsg_len;
 
 	if (sendmsg(sock, &msg, 0) < 0) {
@@ -62,6 +64,7 @@ static int recv_fd(int sock)
 {
 	struct msghdr msg = {};
 	struct cmsghdr *cmsg;
+	int *fd_ptr;
 	char buf[CMSG_SPACE(sizeof(int))] = {0}, c = 'c';
 	struct iovec io = {
 		.iov_base = &c,
@@ -79,8 +82,9 @@ static int recv_fd(int sock)
 	}
 
 	cmsg = CMSG_FIRSTHDR(&msg);
+	fd_ptr = (int *)CMSG_DATA(cmsg);
 
-	return *((int *)CMSG_DATA(cmsg));
+	return *fd_ptr;
 }
 
 static int user_trap_syscall(int nr, unsigned int flags)

tools/testing/selftests/seccomp/seccomp_benchmark.c

@@ -4,7 +4,9 @@
  */
 #define _GNU_SOURCE
 #include <assert.h>
+#include <err.h>
 #include <limits.h>
+#include <sched.h>
 #include <stdbool.h>
 #include <stddef.h>
 #include <stdio.h>
@@ -76,8 +78,12 @@ unsigned long long calibrate(void)
 
 bool approx(int i_one, int i_two)
 {
-	double one = i_one, one_bump = one * 0.01;
-	double two = i_two, two_bump = two * 0.01;
+	/*
+	 * This continues to be a noisy test. Instead of a 1% comparison
+	 * go with 10%.
+	 */
+	double one = i_one, one_bump = one * 0.1;
+	double two = i_two, two_bump = two * 0.1;
 
 	one_bump = one + MAX(one_bump, 2.0);
 	two_bump = two + MAX(two_bump, 2.0);
@@ -131,6 +137,32 @@ long compare(const char *name_one, const char *name_eval, const char *name_two,
 	return good ? 0 : 1;
 }
 
+/* Pin to a single CPU so the benchmark won't bounce around the system. */
+void affinity(void)
+{
+	long cpu;
+	ulong ncores = sysconf(_SC_NPROCESSORS_CONF);
+	cpu_set_t *setp = CPU_ALLOC(ncores);
+	ulong setsz = CPU_ALLOC_SIZE(ncores);
+
+	/*
+	 * Totally unscientific way to avoid CPUs that might be busier:
+	 * choose the highest CPU instead of the lowest.
+	 */
+	for (cpu = ncores - 1; cpu >= 0; cpu--) {
+		CPU_ZERO_S(setsz, setp);
+		CPU_SET_S(cpu, setsz, setp);
+		if (sched_setaffinity(getpid(), setsz, setp) == -1)
+			continue;
+		printf("Pinned to CPU %lu of %lu\n", cpu + 1, ncores);
+		goto out;
+	}
+	fprintf(stderr, "Could not set CPU affinity -- calibration may not work well");
+
+out:
+	CPU_FREE(setp);
+}
+
 int main(int argc, char *argv[])
 {
 	struct sock_filter bitmap_filter[] = {
@@ -172,6 +204,8 @@ int main(int argc, char *argv[])
 	ksft_print_msg("");
 	system("grep -H . /proc/sys/net/core/bpf_jit_harden");
 
+	affinity();
+
 	if (argc > 1)
 		samples = strtoull(argv[1], NULL, 0);
 	else

tools/testing/selftests/seccomp/seccomp_bpf.c

@@ -784,7 +784,7 @@ void *kill_thread(void *data)
 	bool die = (bool)data;
 
 	if (die) {
-		prctl(PR_GET_SECCOMP, 0, 0, 0, 0);
+		syscall(__NR_getpid);
 		return (void *)SIBLING_EXIT_FAILURE;
 	}
 
@@ -803,11 +803,11 @@ void kill_thread_or_group(struct __test_metadata *_metadata,
 {
 	pthread_t thread;
 	void *status;
-	/* Kill only when calling __NR_prctl. */
+	/* Kill only when calling __NR_getpid. */
 	struct sock_filter filter_thread[] = {
 		BPF_STMT(BPF_LD|BPF_W|BPF_ABS,
 			offsetof(struct seccomp_data, nr)),
-		BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_prctl, 0, 1),
+		BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_getpid, 0, 1),
 		BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_KILL_THREAD),
 		BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_ALLOW),
 	};
@@ -819,7 +819,7 @@ void kill_thread_or_group(struct __test_metadata *_metadata,
 	struct sock_filter filter_process[] = {
 		BPF_STMT(BPF_LD|BPF_W|BPF_ABS,
 			offsetof(struct seccomp_data, nr)),
-		BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_prctl, 0, 1),
+		BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_getpid, 0, 1),
 		BPF_STMT(BPF_RET|BPF_K, kill),
 		BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_ALLOW),
 	};
@@ -3709,7 +3709,12 @@ TEST(user_notification_sibling_pid_ns)
 	ASSERT_GE(pid, 0);
 
 	if (pid == 0) {
-		ASSERT_EQ(unshare(CLONE_NEWPID), 0);
+		ASSERT_EQ(unshare(CLONE_NEWPID), 0) {
+			if (errno == EPERM)
+				SKIP(return, "CLONE_NEWPID requires CAP_SYS_ADMIN");
+			else if (errno == EINVAL)
+				SKIP(return, "CLONE_NEWPID is invalid (missing CONFIG_PID_NS?)");
+		}
 
 		pid2 = fork();
 		ASSERT_GE(pid2, 0);
@@ -3727,6 +3732,8 @@ TEST(user_notification_sibling_pid_ns)
 	ASSERT_EQ(unshare(CLONE_NEWPID), 0) {
 		if (errno == EPERM)
 			SKIP(return, "CLONE_NEWPID requires CAP_SYS_ADMIN");
+		else if (errno == EINVAL)
+			SKIP(return, "CLONE_NEWPID is invalid (missing CONFIG_PID_NS?)");
 	}
 	ASSERT_EQ(errno, 0);
 
@@ -4037,6 +4044,16 @@ TEST(user_notification_filter_empty_threaded)
 	EXPECT_GT((pollfd.revents & POLLHUP) ?: 0, 0);
 }
 
+
+int get_next_fd(int prev_fd)
+{
+	for (int i = prev_fd + 1; i < FD_SETSIZE; ++i) {
+		if (fcntl(i, F_GETFD) == -1)
+			return i;
+	}
+	_exit(EXIT_FAILURE);
+}
+
 TEST(user_notification_addfd)
 {
 	pid_t pid;
@@ -4053,7 +4070,7 @@ TEST(user_notification_addfd)
 	/* There may be arbitrary already-open fds at test start. */
 	memfd = memfd_create("test", 0);
 	ASSERT_GE(memfd, 0);
-	nextfd = memfd + 1;
+	nextfd = get_next_fd(memfd);
 
 	ret = prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0);
 	ASSERT_EQ(0, ret) {
@@ -4064,7 +4081,8 @@ TEST(user_notification_addfd)
 	/* Check that the basic notification machinery works */
 	listener = user_notif_syscall(__NR_getppid,
 				      SECCOMP_FILTER_FLAG_NEW_LISTENER);
-	ASSERT_EQ(listener, nextfd++);
+	ASSERT_EQ(listener, nextfd);
+	nextfd = get_next_fd(nextfd);
 
 	pid = fork();
 	ASSERT_GE(pid, 0);
@@ -4119,14 +4137,16 @@ TEST(user_notification_addfd)
 
 	/* Verify we can set an arbitrary remote fd */
 	fd = ioctl(listener, SECCOMP_IOCTL_NOTIF_ADDFD, &addfd);
-	EXPECT_EQ(fd, nextfd++);
+	EXPECT_EQ(fd, nextfd);
+	nextfd = get_next_fd(nextfd);
 	EXPECT_EQ(filecmp(getpid(), pid, memfd, fd), 0);
 
 	/* Verify we can set an arbitrary remote fd with large size */
 	memset(&big, 0x0, sizeof(big));
 	big.addfd = addfd;
 	fd = ioctl(listener, SECCOMP_IOCTL_NOTIF_ADDFD_BIG, &big);
-	EXPECT_EQ(fd, nextfd++);
+	EXPECT_EQ(fd, nextfd);
+	nextfd = get_next_fd(nextfd);
 
 	/* Verify we can set a specific remote fd */
 	addfd.newfd = 42;
@@ -4164,7 +4184,8 @@ TEST(user_notification_addfd)
 	 * Child has earlier "low" fds and now 42, so we expect the next
 	 * lowest available fd to be assigned here.
 	 */
-	EXPECT_EQ(fd, nextfd++);
+	EXPECT_EQ(fd, nextfd);
+	nextfd = get_next_fd(nextfd);
 	ASSERT_EQ(filecmp(getpid(), pid, memfd, fd), 0);
 
 	/*