From 4617721c502b2ddaa4e324e86da4997edf738fa5 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 2 Sep 2026 09:55:01 -0400 Subject: ftrace: Synchronize the initialization of ftrace_ops There's some internal state that ftrace_ops needs to have set, but since it can be declared outside of the ftrace.c code, it calls ftrace_ops_init() on the ops in every global function. The issue is that if two tasks call it on the same ops at the same time it is possible to have the initialization of one corrupt the initialization of the other call. Create a ops_mutex to use to synchronize every initialization of the ftrace_ops. The mutex is taken within checking the ftrace_ops flag that states it was initializied but the flag is checked again after the mutex has been taken. Checking first outside the mutex allows it to shortcut having to take the mutex. But then the check needs to be done again after the mute is taken in case of races. Cc: stable@vger.kernel.org Link: https://patch.msgid.link/20260902095501.6b59af20@gandalf.local.home Fixes: f04f24fb7e48d ("ftrace, kprobes: Fix a deadlock on ftrace_regex_lock") Reported-by: sashiko-bot@kernel.org Close: https://lore.kernel.org/all/20260829025528.49A831F000E9@smtp.kernel.org/ Signed-off-by: Steven Rostedt --- tools/testing/selftests/clone3/.gitignore | 5 + tools/testing/selftests/clone3/Makefile | 8 + tools/testing/selftests/clone3/clone3.c | 342 +++++++++++++++++ .../clone3/clone3_cap_checkpoint_restore.c | 164 ++++++++ .../selftests/clone3/clone3_clear_sighand.c | 124 ++++++ tools/testing/selftests/clone3/clone3_selftests.h | 69 ++++ tools/testing/selftests/clone3/clone3_set_tid.c | 418 +++++++++++++++++++++ 7 files changed, 1130 insertions(+) create mode 100644 tools/testing/selftests/clone3/.gitignore create mode 100644 tools/testing/selftests/clone3/Makefile create mode 100644 tools/testing/selftests/clone3/clone3.c create mode 100644 tools/testing/selftests/clone3/clone3_cap_checkpoint_restore.c create mode 100644 tools/testing/selftests/clone3/clone3_clear_sighand.c create mode 100644 tools/testing/selftests/clone3/clone3_selftests.h create mode 100644 tools/testing/selftests/clone3/clone3_set_tid.c (limited to 'tools/testing/selftests/clone3') diff --git a/tools/testing/selftests/clone3/.gitignore b/tools/testing/selftests/clone3/.gitignore new file mode 100644 index 000000000..83c0f6246 --- /dev/null +++ b/tools/testing/selftests/clone3/.gitignore @@ -0,0 +1,5 @@ +# SPDX-License-Identifier: GPL-2.0-only +clone3 +clone3_clear_sighand +clone3_set_tid +clone3_cap_checkpoint_restore diff --git a/tools/testing/selftests/clone3/Makefile b/tools/testing/selftests/clone3/Makefile new file mode 100644 index 000000000..84832c369 --- /dev/null +++ b/tools/testing/selftests/clone3/Makefile @@ -0,0 +1,8 @@ +# SPDX-License-Identifier: GPL-2.0 +CFLAGS += -g -std=gnu99 $(KHDR_INCLUDES) +LDLIBS += -lcap + +TEST_GEN_PROGS := clone3 clone3_clear_sighand clone3_set_tid \ + clone3_cap_checkpoint_restore + +include ../lib.mk diff --git a/tools/testing/selftests/clone3/clone3.c b/tools/testing/selftests/clone3/clone3.c new file mode 100644 index 000000000..289e0c7c1 --- /dev/null +++ b/tools/testing/selftests/clone3/clone3.c @@ -0,0 +1,342 @@ +// SPDX-License-Identifier: GPL-2.0 + +/* Based on Christian Brauner's clone3() example */ + +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "kselftest.h" +#include "clone3_selftests.h" + +enum test_mode { + CLONE3_ARGS_NO_TEST, + CLONE3_ARGS_ALL_0, + CLONE3_ARGS_INVAL_EXIT_SIGNAL_BIG, + CLONE3_ARGS_INVAL_EXIT_SIGNAL_NEG, + CLONE3_ARGS_INVAL_EXIT_SIGNAL_CSIG, + CLONE3_ARGS_INVAL_EXIT_SIGNAL_NSIG, +}; + +static int call_clone3(uint64_t flags, size_t size, enum test_mode test_mode) +{ + struct __clone_args args = { + .flags = flags, + .exit_signal = SIGCHLD, + }; + + struct clone_args_extended { + struct __clone_args args; + __aligned_u64 excess_space[2]; + } args_ext; + + pid_t pid = -1; + int status; + + memset(&args_ext, 0, sizeof(args_ext)); + if (size > sizeof(struct __clone_args)) + args_ext.excess_space[1] = 1; + + if (size == 0) + size = sizeof(struct __clone_args); + + switch (test_mode) { + case CLONE3_ARGS_NO_TEST: + /* + * Uses default 'flags' and 'SIGCHLD' + * assignment. + */ + break; + case CLONE3_ARGS_ALL_0: + args.flags = 0; + args.exit_signal = 0; + break; + case CLONE3_ARGS_INVAL_EXIT_SIGNAL_BIG: + args.exit_signal = 0xbadc0ded00000000ULL; + break; + case CLONE3_ARGS_INVAL_EXIT_SIGNAL_NEG: + args.exit_signal = 0x0000000080000000ULL; + break; + case CLONE3_ARGS_INVAL_EXIT_SIGNAL_CSIG: + args.exit_signal = 0x0000000000000100ULL; + break; + case CLONE3_ARGS_INVAL_EXIT_SIGNAL_NSIG: + args.exit_signal = 0x00000000000000f0ULL; + break; + } + + memcpy(&args_ext.args, &args, sizeof(struct __clone_args)); + + pid = sys_clone3((struct __clone_args *)&args_ext, size); + if (pid < 0) { + ksft_print_msg("%s - Failed to create new process\n", + strerror(errno)); + return -errno; + } + + if (pid == 0) { + ksft_print_msg("I am the child, my PID is %d\n", getpid()); + _exit(EXIT_SUCCESS); + } + + ksft_print_msg("I am the parent (%d). My child's pid is %d\n", + getpid(), pid); + + if (waitpid(-1, &status, __WALL) < 0) { + ksft_print_msg("waitpid() returned %s\n", strerror(errno)); + return -errno; + } + if (!WIFEXITED(status)) { + ksft_print_msg("Child did not exit normally, status 0x%x\n", + status); + return EXIT_FAILURE; + } + if (WEXITSTATUS(status)) + return WEXITSTATUS(status); + + return 0; +} + +static bool test_clone3(uint64_t flags, size_t size, int expected, + enum test_mode test_mode) +{ + int ret; + + ksft_print_msg( + "[%d] Trying clone3() with flags %#" PRIx64 " (size %zu)\n", + getpid(), flags, size); + ret = call_clone3(flags, size, test_mode); + ksft_print_msg("[%d] clone3() with flags says: %d expected %d\n", + getpid(), ret, expected); + if (ret != expected) { + ksft_print_msg( + "[%d] Result (%d) is different than expected (%d)\n", + getpid(), ret, expected); + return false; + } + + return true; +} + +typedef bool (*filter_function)(void); +typedef size_t (*size_function)(void); + +static bool not_root(void) +{ + if (getuid() != 0) { + ksft_print_msg("Not running as root\n"); + return true; + } + + return false; +} + +static bool no_timenamespace(void) +{ + if (not_root()) + return true; + + if (!access("/proc/self/ns/time", F_OK)) + return false; + + ksft_print_msg("Time namespaces are not supported\n"); + return true; +} + +static size_t page_size_plus_8(void) +{ + return getpagesize() + 8; +} + +struct test { + const char *name; + uint64_t flags; + size_t size; + size_function size_function; + int expected; + enum test_mode test_mode; + filter_function filter; +}; + +static const struct test tests[] = { + { + .name = "simple clone3()", + .flags = 0, + .size = 0, + .expected = 0, + .test_mode = CLONE3_ARGS_NO_TEST, + }, + { + .name = "clone3() in a new PID_NS", + .flags = CLONE_NEWPID, + .size = 0, + .expected = 0, + .test_mode = CLONE3_ARGS_NO_TEST, + .filter = not_root, + }, + { + .name = "CLONE_ARGS_SIZE_VER0", + .flags = 0, + .size = CLONE_ARGS_SIZE_VER0, + .expected = 0, + .test_mode = CLONE3_ARGS_NO_TEST, + }, + { + .name = "CLONE_ARGS_SIZE_VER0 - 8", + .flags = 0, + .size = CLONE_ARGS_SIZE_VER0 - 8, + .expected = -EINVAL, + .test_mode = CLONE3_ARGS_NO_TEST, + }, + { + .name = "sizeof(struct clone_args) + 8", + .flags = 0, + .size = sizeof(struct __clone_args) + 8, + .expected = 0, + .test_mode = CLONE3_ARGS_NO_TEST, + }, + { + .name = "exit_signal with highest 32 bits non-zero", + .flags = 0, + .size = 0, + .expected = -EINVAL, + .test_mode = CLONE3_ARGS_INVAL_EXIT_SIGNAL_BIG, + }, + { + .name = "negative 32-bit exit_signal", + .flags = 0, + .size = 0, + .expected = -EINVAL, + .test_mode = CLONE3_ARGS_INVAL_EXIT_SIGNAL_NEG, + }, + { + .name = "exit_signal not fitting into CSIGNAL mask", + .flags = 0, + .size = 0, + .expected = -EINVAL, + .test_mode = CLONE3_ARGS_INVAL_EXIT_SIGNAL_CSIG, + }, + { + .name = "NSIG < exit_signal < CSIG", + .flags = 0, + .size = 0, + .expected = -EINVAL, + .test_mode = CLONE3_ARGS_INVAL_EXIT_SIGNAL_NSIG, + }, + { + .name = "Arguments sizeof(struct clone_args) + 8", + .flags = 0, + .size = sizeof(struct __clone_args) + 8, + .expected = 0, + .test_mode = CLONE3_ARGS_ALL_0, + }, + { + .name = "Arguments sizeof(struct clone_args) + 16", + .flags = 0, + .size = sizeof(struct __clone_args) + 16, + .expected = -E2BIG, + .test_mode = CLONE3_ARGS_ALL_0, + }, + { + .name = "Arguments sizeof(struct clone_arg) * 2", + .flags = 0, + .size = sizeof(struct __clone_args) + 16, + .expected = -E2BIG, + .test_mode = CLONE3_ARGS_ALL_0, + }, + { + .name = "Arguments > page size", + .flags = 0, + .size_function = page_size_plus_8, + .expected = -E2BIG, + .test_mode = CLONE3_ARGS_NO_TEST, + }, + { + .name = "CLONE_ARGS_SIZE_VER0 in a new PID NS", + .flags = CLONE_NEWPID, + .size = CLONE_ARGS_SIZE_VER0, + .expected = 0, + .test_mode = CLONE3_ARGS_NO_TEST, + .filter = not_root, + }, + { + .name = "CLONE_ARGS_SIZE_VER0 - 8 in a new PID NS", + .flags = CLONE_NEWPID, + .size = CLONE_ARGS_SIZE_VER0 - 8, + .expected = -EINVAL, + .test_mode = CLONE3_ARGS_NO_TEST, + }, + { + .name = "sizeof(struct clone_args) + 8 in a new PID NS", + .flags = CLONE_NEWPID, + .size = sizeof(struct __clone_args) + 8, + .expected = 0, + .test_mode = CLONE3_ARGS_NO_TEST, + .filter = not_root, + }, + { + .name = "Arguments > page size in a new PID NS", + .flags = CLONE_NEWPID, + .size_function = page_size_plus_8, + .expected = -E2BIG, + .test_mode = CLONE3_ARGS_NO_TEST, + }, + { + .name = "New time NS", + .flags = CLONE_NEWTIME, + .size = 0, + .expected = 0, + .test_mode = CLONE3_ARGS_NO_TEST, + .filter = no_timenamespace, + }, + { + .name = "exit signal (SIGCHLD) in flags", + .flags = SIGCHLD, + .size = 0, + .expected = -EINVAL, + .test_mode = CLONE3_ARGS_NO_TEST, + }, +}; + +int main(int argc, char *argv[]) +{ + size_t size; + int i; + + ksft_print_header(); + ksft_set_plan(ARRAY_SIZE(tests)); + test_clone3_supported(); + + for (i = 0; i < ARRAY_SIZE(tests); i++) { + if (tests[i].filter && tests[i].filter()) { + ksft_test_result_skip("%s\n", tests[i].name); + continue; + } + + if (tests[i].size_function) + size = tests[i].size_function(); + else + size = tests[i].size; + + ksft_print_msg("Running test '%s'\n", tests[i].name); + + ksft_test_result(test_clone3(tests[i].flags, size, + tests[i].expected, + tests[i].test_mode), + "%s\n", tests[i].name); + } + + ksft_finished(); +} diff --git a/tools/testing/selftests/clone3/clone3_cap_checkpoint_restore.c b/tools/testing/selftests/clone3/clone3_cap_checkpoint_restore.c new file mode 100644 index 000000000..ab62bcf41 --- /dev/null +++ b/tools/testing/selftests/clone3/clone3_cap_checkpoint_restore.c @@ -0,0 +1,164 @@ +// SPDX-License-Identifier: GPL-2.0 + +/* + * Based on Christian Brauner's clone3() example. + * These tests are assuming to be running in the host's + * PID namespace. + */ + +/* capabilities related code based on selftests/bpf/test_verifier.c */ + +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "kselftest_harness.h" +#include "clone3_selftests.h" + +static void child_exit(int ret) +{ + fflush(stdout); + fflush(stderr); + _exit(ret); +} + +static int call_clone3_set_tid(struct __test_metadata *_metadata, + pid_t *set_tid, size_t set_tid_size) +{ + int status; + pid_t pid = -1; + + struct __clone_args args = { + .exit_signal = SIGCHLD, + .set_tid = ptr_to_u64(set_tid), + .set_tid_size = set_tid_size, + }; + + pid = sys_clone3(&args, sizeof(args)); + if (pid < 0) { + TH_LOG("%s - Failed to create new process", strerror(errno)); + return -errno; + } + + if (pid == 0) { + TH_LOG("I am the child, my PID is %d (expected %d)", getpid(), set_tid[0]); + + if (set_tid[0] != getpid()) + child_exit(EXIT_FAILURE); + child_exit(EXIT_SUCCESS); + } + + TH_LOG("I am the parent (%d). My child's pid is %d", getpid(), pid); + + if (waitpid(pid, &status, 0) < 0) { + TH_LOG("Child returned %s", strerror(errno)); + return -errno; + } + + if (!WIFEXITED(status)) + return -1; + + return WEXITSTATUS(status); +} + +static int test_clone3_set_tid(struct __test_metadata *_metadata, + pid_t *set_tid, size_t set_tid_size) +{ + int ret; + + TH_LOG("[%d] Trying clone3() with CLONE_SET_TID to %d", getpid(), set_tid[0]); + ret = call_clone3_set_tid(_metadata, set_tid, set_tid_size); + TH_LOG("[%d] clone3() with CLONE_SET_TID %d says:%d", getpid(), set_tid[0], ret); + return ret; +} + +static int set_capability(void) +{ + cap_value_t cap_values[] = { + CAP_SETUID, CAP_SETGID, CAP_CHECKPOINT_RESTORE + }; + int ret = -1; + cap_t caps; + + caps = cap_get_proc(); + if (!caps) { + perror("cap_get_proc"); + return -1; + } + + /* Drop all capabilities */ + if (cap_clear(caps)) { + perror("cap_clear"); + goto out; + } + + cap_set_flag(caps, CAP_EFFECTIVE, 3, cap_values, CAP_SET); + cap_set_flag(caps, CAP_PERMITTED, 3, cap_values, CAP_SET); + + if (cap_set_proc(caps)) { + perror("cap_set_proc"); + goto out; + } + ret = 0; +out: + if (cap_free(caps)) + perror("cap_free"); + return ret; +} + +TEST(clone3_cap_checkpoint_restore) +{ + pid_t pid; + int status; + pid_t set_tid[1]; + + test_clone3_supported(); + + EXPECT_EQ(getuid(), 0) + SKIP(return, "Skipping all tests as non-root"); + + memset(&set_tid, 0, sizeof(set_tid)); + + /* Find the current active PID */ + pid = fork(); + if (pid == 0) { + TH_LOG("Child has PID %d", getpid()); + child_exit(EXIT_SUCCESS); + } + ASSERT_GT(waitpid(pid, &status, 0), 0) + TH_LOG("Waiting for child %d failed", pid); + + /* After the child has finished, its PID should be free. */ + set_tid[0] = pid; + + ASSERT_EQ(set_capability(), 0) + TH_LOG("Could not set CAP_CHECKPOINT_RESTORE"); + + ASSERT_EQ(prctl(PR_SET_KEEPCAPS, 1, 0, 0, 0), 0); + + EXPECT_EQ(setgid(65534), 0) + TH_LOG("Failed to setgid(65534)"); + ASSERT_EQ(setuid(65534), 0); + + set_tid[0] = pid; + /* This would fail without CAP_CHECKPOINT_RESTORE */ + ASSERT_EQ(test_clone3_set_tid(_metadata, set_tid, 1), -EPERM); + ASSERT_EQ(set_capability(), 0) + TH_LOG("Could not set CAP_CHECKPOINT_RESTORE"); + /* This should work as we have CAP_CHECKPOINT_RESTORE as non-root */ + ASSERT_EQ(test_clone3_set_tid(_metadata, set_tid, 1), 0); +} + +TEST_HARNESS_MAIN diff --git a/tools/testing/selftests/clone3/clone3_clear_sighand.c b/tools/testing/selftests/clone3/clone3_clear_sighand.c new file mode 100644 index 000000000..de0c9d620 --- /dev/null +++ b/tools/testing/selftests/clone3/clone3_clear_sighand.c @@ -0,0 +1,124 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "kselftest.h" +#include "clone3_selftests.h" + +static void nop_handler(int signo) +{ +} + +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 void test_clone3_clear_sighand(void) +{ + int ret; + pid_t pid; + struct __clone_args args = {}; + struct sigaction act; + + /* + * Check that CLONE_CLEAR_SIGHAND and CLONE_SIGHAND are mutually + * exclusive. + */ + args.flags |= CLONE_CLEAR_SIGHAND | CLONE_SIGHAND; + args.exit_signal = SIGCHLD; + pid = sys_clone3(&args, sizeof(args)); + if (pid > 0) + ksft_exit_fail_msg( + "clone3(CLONE_CLEAR_SIGHAND | CLONE_SIGHAND) succeeded\n"); + + act.sa_handler = nop_handler; + ret = sigemptyset(&act.sa_mask); + if (ret < 0) + ksft_exit_fail_msg("%s - sigemptyset() failed\n", + strerror(errno)); + + act.sa_flags = 0; + + /* Register signal handler for SIGUSR1 */ + ret = sigaction(SIGUSR1, &act, NULL); + if (ret < 0) + ksft_exit_fail_msg( + "%s - sigaction(SIGUSR1, &act, NULL) failed\n", + strerror(errno)); + + /* Register signal handler for SIGUSR2 */ + ret = sigaction(SIGUSR2, &act, NULL); + if (ret < 0) + ksft_exit_fail_msg( + "%s - sigaction(SIGUSR2, &act, NULL) failed\n", + strerror(errno)); + + /* Check that CLONE_CLEAR_SIGHAND works. */ + args.flags = CLONE_CLEAR_SIGHAND; + pid = sys_clone3(&args, sizeof(args)); + if (pid < 0) + ksft_exit_fail_msg("%s - clone3(CLONE_CLEAR_SIGHAND) failed\n", + strerror(errno)); + + if (pid == 0) { + ret = sigaction(SIGUSR1, NULL, &act); + if (ret < 0) + exit(EXIT_FAILURE); + + if (act.sa_handler != SIG_DFL) + exit(EXIT_FAILURE); + + ret = sigaction(SIGUSR2, NULL, &act); + if (ret < 0) + exit(EXIT_FAILURE); + + if (act.sa_handler != SIG_DFL) + exit(EXIT_FAILURE); + + exit(EXIT_SUCCESS); + } + + ret = wait_for_pid(pid); + if (ret) + ksft_exit_fail_msg( + "Failed to clear signal handler for child process\n"); + + ksft_test_result_pass("Cleared signal handlers for child process\n"); +} + +int main(int argc, char **argv) +{ + ksft_print_header(); + ksft_set_plan(1); + test_clone3_supported(); + + test_clone3_clear_sighand(); + + ksft_exit_pass(); +} diff --git a/tools/testing/selftests/clone3/clone3_selftests.h b/tools/testing/selftests/clone3/clone3_selftests.h new file mode 100644 index 000000000..a0593e895 --- /dev/null +++ b/tools/testing/selftests/clone3/clone3_selftests.h @@ -0,0 +1,69 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef _CLONE3_SELFTESTS_H +#define _CLONE3_SELFTESTS_H + +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include + +#include "kselftest.h" + +#define ptr_to_u64(ptr) ((__u64)((uintptr_t)(ptr))) + +#ifndef __NR_clone3 +#define __NR_clone3 435 +#endif + +struct __clone_args { + __aligned_u64 flags; + __aligned_u64 pidfd; + __aligned_u64 child_tid; + __aligned_u64 parent_tid; + __aligned_u64 exit_signal; + __aligned_u64 stack; + __aligned_u64 stack_size; + __aligned_u64 tls; + __aligned_u64 set_tid; + __aligned_u64 set_tid_size; + __aligned_u64 cgroup; +}; + +static pid_t sys_clone3(struct __clone_args *args, size_t size) +{ + fflush(stdout); + fflush(stderr); + return syscall(__NR_clone3, args, size); +} + +static inline void test_clone3_supported(void) +{ + pid_t pid; + struct __clone_args args = {}; + + if (__NR_clone3 < 0) + ksft_exit_skip("clone3() syscall is not supported\n"); + + /* Set to something that will always cause EINVAL. */ + args.exit_signal = -1; + pid = sys_clone3(&args, sizeof(args)); + if (!pid) + exit(EXIT_SUCCESS); + + if (pid > 0) { + wait(NULL); + ksft_exit_fail_msg( + "Managed to create child process with invalid exit_signal\n"); + } + + if (errno == ENOSYS) + ksft_exit_skip("clone3() syscall is not supported\n"); + + ksft_print_msg("clone3() syscall supported\n"); +} + +#endif /* _CLONE3_SELFTESTS_H */ diff --git a/tools/testing/selftests/clone3/clone3_set_tid.c b/tools/testing/selftests/clone3/clone3_set_tid.c new file mode 100644 index 000000000..485efa7c9 --- /dev/null +++ b/tools/testing/selftests/clone3/clone3_set_tid.c @@ -0,0 +1,418 @@ +// SPDX-License-Identifier: GPL-2.0 + +/* + * Based on Christian Brauner's clone3() example. + * These tests are assuming to be running in the host's + * PID namespace. + */ + +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "kselftest.h" +#include "clone3_selftests.h" + +#define MAX_PID_NS_LEVEL 32 + +static int pipe_1[2]; +static int pipe_2[2]; + +static void child_exit(int ret) +{ + fflush(stdout); + fflush(stderr); + _exit(ret); +} + +static int call_clone3_set_tid(pid_t *set_tid, + size_t set_tid_size, + int flags, + int expected_pid, + bool wait_for_it) +{ + int status; + pid_t pid = -1; + + struct __clone_args args = { + .flags = flags, + .exit_signal = SIGCHLD, + .set_tid = ptr_to_u64(set_tid), + .set_tid_size = set_tid_size, + }; + + pid = sys_clone3(&args, sizeof(args)); + if (pid < 0) { + ksft_print_msg("%s - Failed to create new process\n", + strerror(errno)); + return -errno; + } + + if (pid == 0) { + int ret; + char tmp = 0; + int exit_code = EXIT_SUCCESS; + + ksft_print_msg("I am the child, my PID is %d (expected %d)\n", + getpid(), set_tid[0]); + if (wait_for_it) { + ksft_print_msg("[%d] Child is ready and waiting\n", + getpid()); + + /* Signal the parent that the child is ready */ + close(pipe_1[0]); + ret = write(pipe_1[1], &tmp, 1); + if (ret != 1) { + ksft_print_msg( + "Writing to pipe returned %d", ret); + exit_code = EXIT_FAILURE; + } + close(pipe_1[1]); + close(pipe_2[1]); + ret = read(pipe_2[0], &tmp, 1); + if (ret != 1) { + ksft_print_msg( + "Reading from pipe returned %d", ret); + exit_code = EXIT_FAILURE; + } + close(pipe_2[0]); + } + + if (set_tid[0] != getpid()) + child_exit(EXIT_FAILURE); + child_exit(exit_code); + } + + if (expected_pid == 0 || expected_pid == pid) { + ksft_print_msg("I am the parent (%d). My child's pid is %d\n", + getpid(), pid); + } else { + ksft_print_msg( + "Expected child pid %d does not match actual pid %d\n", + expected_pid, pid); + return -1; + } + + if (waitpid(pid, &status, 0) < 0) { + ksft_print_msg("Child returned %s\n", strerror(errno)); + return -errno; + } + + if (!WIFEXITED(status)) + return -1; + + return WEXITSTATUS(status); +} + +static void test_clone3_set_tid(const char *desc, + pid_t *set_tid, + size_t set_tid_size, + int flags, + int expected, + int expected_pid, + bool wait_for_it) +{ + int ret; + + ksft_print_msg( + "[%d] Trying clone3() with CLONE_SET_TID to %d and 0x%x\n", + getpid(), set_tid[0], flags); + ret = call_clone3_set_tid(set_tid, set_tid_size, flags, expected_pid, + wait_for_it); + ksft_print_msg( + "[%d] clone3() with CLONE_SET_TID %d says: %d - expected %d\n", + getpid(), set_tid[0], ret, expected); + + ksft_test_result(ret == expected, "%s with %zu TIDs and flags 0x%x\n", + desc, set_tid_size, flags); +} + +int main(int argc, char *argv[]) +{ + FILE *f; + char buf; + char *line = NULL; + int status; + int ret = -1; + size_t len = 0; + int pid_max = 0; + uid_t uid = getuid(); + char proc_path[100] = {0}; + pid_t pid, ns1, ns2, ns3, ns_pid; + pid_t set_tid[MAX_PID_NS_LEVEL * 2]; + + ksft_print_header(); + ksft_set_plan(29); + test_clone3_supported(); + + if (pipe(pipe_1) < 0 || pipe(pipe_2) < 0) + ksft_exit_fail_msg("pipe() failed\n"); + + f = fopen("/proc/sys/kernel/pid_max", "r"); + if (f == NULL) + ksft_exit_fail_msg( + "%s - Could not open /proc/sys/kernel/pid_max\n", + strerror(errno)); + fscanf(f, "%d", &pid_max); + fclose(f); + ksft_print_msg("/proc/sys/kernel/pid_max %d\n", pid_max); + + /* Try invalid settings */ + memset(&set_tid, 0, sizeof(set_tid)); + test_clone3_set_tid("invalid size, 0 TID", + set_tid, MAX_PID_NS_LEVEL + 1, 0, -EINVAL, 0, 0); + + test_clone3_set_tid("invalid size, 0 TID", + set_tid, MAX_PID_NS_LEVEL * 2, 0, -EINVAL, 0, 0); + + test_clone3_set_tid("invalid size, 0 TID", + set_tid, MAX_PID_NS_LEVEL * 2 + 1, 0, + -EINVAL, 0, 0); + + test_clone3_set_tid("invalid size, 0 TID", + set_tid, MAX_PID_NS_LEVEL * 42, 0, -EINVAL, 0, 0); + + /* + * This can actually work if this test running in a MAX_PID_NS_LEVEL - 1 + * nested PID namespace. + */ + test_clone3_set_tid("invalid size, 0 TID", + set_tid, MAX_PID_NS_LEVEL - 1, 0, -EINVAL, 0, 0); + + memset(&set_tid, 0xff, sizeof(set_tid)); + test_clone3_set_tid("invalid size, TID all 1s", + set_tid, MAX_PID_NS_LEVEL + 1, 0, -EINVAL, 0, 0); + + test_clone3_set_tid("invalid size, TID all 1s", + set_tid, MAX_PID_NS_LEVEL * 2, 0, -EINVAL, 0, 0); + + test_clone3_set_tid("invalid size, TID all 1s", + set_tid, MAX_PID_NS_LEVEL * 2 + 1, 0, + -EINVAL, 0, 0); + + test_clone3_set_tid("invalid size, TID all 1s", + set_tid, MAX_PID_NS_LEVEL * 42, 0, -EINVAL, 0, 0); + + /* + * This can actually work if this test running in a MAX_PID_NS_LEVEL - 1 + * nested PID namespace. + */ + test_clone3_set_tid("invalid size, TID all 1s", + set_tid, MAX_PID_NS_LEVEL - 1, 0, -EINVAL, 0, 0); + + memset(&set_tid, 0, sizeof(set_tid)); + /* Try with an invalid PID */ + set_tid[0] = 0; + test_clone3_set_tid("valid size, 0 TID", + set_tid, 1, 0, -EINVAL, 0, 0); + + set_tid[0] = -1; + test_clone3_set_tid("valid size, -1 TID", + set_tid, 1, 0, -EINVAL, 0, 0); + + /* Claim that the set_tid array actually contains 2 elements. */ + test_clone3_set_tid("2 TIDs, -1 and 0", + set_tid, 2, 0, -EINVAL, 0, 0); + + /* Try it in a new PID namespace */ + if (uid == 0) + test_clone3_set_tid("valid size, -1 TID", + set_tid, 1, CLONE_NEWPID, -EINVAL, 0, 0); + else + ksft_test_result_skip("Clone3() with set_tid requires root\n"); + + /* Try with a valid PID (1) this should return -EEXIST. */ + set_tid[0] = 1; + if (uid == 0) + test_clone3_set_tid("duplicate PID 1", + set_tid, 1, 0, -EEXIST, 0, 0); + else + ksft_test_result_skip("Clone3() with set_tid requires root\n"); + + /* Try it in a new PID namespace */ + if (uid == 0) + test_clone3_set_tid("duplicate PID 1", + set_tid, 1, CLONE_NEWPID, 0, 0, 0); + else + ksft_test_result_skip("Clone3() with set_tid requires root\n"); + + /* pid_max should fail everywhere */ + set_tid[0] = pid_max; + test_clone3_set_tid("set TID to maximum", + set_tid, 1, 0, -EINVAL, 0, 0); + + if (uid == 0) + test_clone3_set_tid("set TID to maximum", + set_tid, 1, CLONE_NEWPID, -EINVAL, 0, 0); + else + ksft_test_result_skip("Clone3() with set_tid requires root\n"); + + if (uid != 0) { + /* + * All remaining tests require root. Tell the framework + * that all those tests are skipped as non-root. + */ + ksft_cnt.ksft_xskip += ksft_plan - ksft_test_num(); + goto out; + } + + /* Find the current active PID */ + pid = fork(); + if (pid == 0) { + ksft_print_msg("Child has PID %d\n", getpid()); + child_exit(EXIT_SUCCESS); + } + if (waitpid(pid, &status, 0) < 0) + ksft_exit_fail_msg("Waiting for child %d failed", pid); + + /* After the child has finished, its PID should be free. */ + set_tid[0] = pid; + test_clone3_set_tid("reallocate child TID", + set_tid, 1, 0, 0, 0, 0); + + /* This should fail as there is no PID 1 in that namespace */ + test_clone3_set_tid("duplicate child TID", + set_tid, 1, CLONE_NEWPID, -EINVAL, 0, 0); + + /* + * Creating a process with PID 1 in the newly created most nested + * PID namespace and PID 'pid' in the parent PID namespace. This + * needs to work. + */ + set_tid[0] = 1; + set_tid[1] = pid; + test_clone3_set_tid("create PID 1 in new NS", + set_tid, 2, CLONE_NEWPID, 0, pid, 0); + + ksft_print_msg("unshare PID namespace\n"); + if (unshare(CLONE_NEWPID) == -1) + ksft_exit_fail_msg("unshare(CLONE_NEWPID) failed: %s\n", + strerror(errno)); + + set_tid[0] = pid; + + /* This should fail as there is no PID 1 in that namespace */ + test_clone3_set_tid("duplicate PID 1", + set_tid, 1, 0, -EINVAL, 0, 0); + + /* Let's create a PID 1 */ + ns_pid = fork(); + if (ns_pid == 0) { + /* + * This and the next test cases check that all pid-s are + * released on error paths. + */ + set_tid[0] = 43; + set_tid[1] = -1; + test_clone3_set_tid("check leak on invalid TID -1", + set_tid, 2, 0, -EINVAL, 0, 0); + + set_tid[0] = 43; + set_tid[1] = pid; + test_clone3_set_tid("check leak on invalid specific TID", + set_tid, 2, 0, 0, 43, 0); + + ksft_print_msg("Child in PID namespace has PID %d\n", getpid()); + set_tid[0] = 2; + test_clone3_set_tid("create PID 2 in child NS", + set_tid, 1, 0, 0, 2, 0); + + set_tid[0] = 1; + set_tid[1] = -1; + set_tid[2] = pid; + /* This should fail as there is invalid PID at level '1'. */ + test_clone3_set_tid("fail due to invalid TID at level 1", + set_tid, 3, CLONE_NEWPID, -EINVAL, 0, 0); + + set_tid[0] = 1; + set_tid[1] = 42; + set_tid[2] = pid; + /* + * This should fail as there are not enough active PID + * namespaces. Again assuming this is running in the host's + * PID namespace. Not yet nested. + */ + test_clone3_set_tid("fail due to too few active PID NSs", + set_tid, 4, CLONE_NEWPID, -EINVAL, 0, 0); + + /* + * This should work and from the parent we should see + * something like 'NSpid: pid 42 1'. + */ + test_clone3_set_tid("verify that we have 3 PID NSs", + set_tid, 3, CLONE_NEWPID, 0, 42, true); + + child_exit(ksft_cnt.ksft_fail); + } + + close(pipe_1[1]); + close(pipe_2[0]); + while (read(pipe_1[0], &buf, 1) > 0) { + ksft_print_msg("[%d] Child is ready and waiting\n", getpid()); + break; + } + + snprintf(proc_path, sizeof(proc_path), "/proc/%d/status", pid); + f = fopen(proc_path, "r"); + if (f == NULL) + ksft_exit_fail_msg( + "%s - Could not open %s\n", + strerror(errno), proc_path); + + while (getline(&line, &len, f) != -1) { + if (strstr(line, "NSpid")) { + int i; + + /* Verify that all generated PIDs are as expected. */ + i = sscanf(line, "NSpid:\t%d\t%d\t%d", + &ns3, &ns2, &ns1); + if (i != 3) { + ksft_print_msg( + "Unexpected 'NSPid:' entry: %s", + line); + ns1 = ns2 = ns3 = 0; + } + break; + } + } + fclose(f); + free(line); + close(pipe_2[0]); + + /* Tell the clone3()'d child to finish. */ + write(pipe_2[1], &buf, 1); + close(pipe_2[1]); + + if (waitpid(ns_pid, &status, 0) < 0) { + ksft_print_msg("Child returned %s\n", strerror(errno)); + ret = -errno; + goto out; + } + + if (!WIFEXITED(status)) + ksft_test_result_fail("Child error\n"); + + ksft_cnt.ksft_pass += 6 - (ksft_cnt.ksft_fail - WEXITSTATUS(status)); + ksft_cnt.ksft_fail = WEXITSTATUS(status); + + ksft_print_msg("Expecting PIDs %d, 42, 1\n", pid); + ksft_print_msg("Have PIDs in namespaces: %d, %d, %d\n", ns3, ns2, ns1); + ksft_test_result(ns3 == pid && ns2 == 42 && ns1 == 1, + "PIDs in all namespaces as expected\n"); +out: + ret = 0; + + if (ret) + ksft_exit_fail(); + ksft_exit_pass(); +} -- cgit v1.3.1