Resguardo inicial

linux-7.2.3/tools/testing/selftests/pid_namespace/.gitignore

@@ -0,0 +1,3 @@
+pid_max
+pidns_init_via_setns
+regression_enomem

linux-7.2.3/tools/testing/selftests/pid_namespace/Makefile

@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0
+CFLAGS += -g $(KHDR_INCLUDES)
+
+TEST_GEN_PROGS = regression_enomem pid_max pidns_init_via_setns
+
+LOCAL_HDRS += $(selfdir)/pidfd/pidfd.h
+
+include ../lib.mk

linux-7.2.3/tools/testing/selftests/pid_namespace/config

@@ -0,0 +1,2 @@
+CONFIG_PID_NS=y
+CONFIG_USER_NS=y

linux-7.2.3/tools/testing/selftests/pid_namespace/pid_max.c

@@ -0,0 +1,451 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#define _GNU_SOURCE
+#include <assert.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <linux/types.h>
+#include <sched.h>
+#include <signal.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <syscall.h>
+#include <sys/mount.h>
+#include <sys/wait.h>
+#include <unistd.h>
+
+#include "kselftest_harness.h"
+#include "../pidfd/pidfd.h"
+
+/*
+ * The kernel computes the minimum allowed pid_max as:
+ *   max(RESERVED_PIDS + 1, PIDS_PER_CPU_MIN * num_possible_cpus())
+ * Mirror that here so the test values are always valid.
+ *
+ * Note: glibc's get_nprocs_conf() returns the number of *configured*
+ * (present) CPUs, not *possible* CPUs.  The kernel uses
+ * num_possible_cpus() which corresponds to /sys/devices/system/cpu/possible.
+ * These can differ significantly (e.g. 16 configured vs 128 possible).
+ */
+#define RESERVED_PIDS		300
+#define PIDS_PER_CPU_MIN	8
+
+/* Count CPUs from a range list like "0-31" or "0-15,32-47". */
+static int num_possible_cpus(void)
+{
+	FILE *f;
+	int count = 0;
+	int lo, hi;
+
+	f = fopen("/sys/devices/system/cpu/possible", "r");
+	if (!f)
+		return 0;
+
+	while (fscanf(f, "%d", &lo) == 1) {
+		if (fscanf(f, "-%d", &hi) == 1)
+			count += hi - lo + 1;
+		else
+			count++;
+		/* skip comma separator */
+		fscanf(f, ",");
+	}
+
+	fclose(f);
+	return count;
+}
+
+static int pid_min(void)
+{
+	int cpu_min = PIDS_PER_CPU_MIN * num_possible_cpus();
+
+	return cpu_min > (RESERVED_PIDS + 1) ? cpu_min : (RESERVED_PIDS + 1);
+}
+
+/*
+ * Outer and inner pid_max limits used by the tests.  The outer limit is
+ * the more restrictive ancestor; the inner limit is set higher in a
+ * nested namespace but must still be capped by the outer limit.
+ * Both are derived from the kernel's minimum so they are always writable.
+ *
+ * Global so that clone callbacks can access them without parameter plumbing.
+ */
+static int outer_limit;
+static int inner_limit;
+
+static int write_int_to_fd(int fd, int val)
+{
+	char buf[12];
+	int len = snprintf(buf, sizeof(buf), "%d", val);
+
+	return write(fd, buf, len);
+}
+
+#define __STACK_SIZE (8 * 1024 * 1024)
+static pid_t do_clone(int (*fn)(void *), void *arg, int flags)
+{
+	char *stack;
+	pid_t ret;
+
+	stack = malloc(__STACK_SIZE);
+	if (!stack)
+		return -ENOMEM;
+
+#ifdef __ia64__
+	ret = __clone2(fn, stack, __STACK_SIZE, flags | SIGCHLD, arg);
+#else
+	ret = clone(fn, stack + __STACK_SIZE, flags | SIGCHLD, arg);
+#endif
+	free(stack);
+	return ret;
+}
+
+static int pid_max_cb(void *data)
+{
+	int fd, ret;
+	pid_t pid;
+
+	ret = mount("", "/", NULL, MS_PRIVATE | MS_REC, 0);
+	if (ret) {
+		fprintf(stderr, "%m - Failed to make rootfs private mount\n");
+		return -1;
+	}
+
+	umount2("/proc", MNT_DETACH);
+
+	ret = mount("proc", "/proc", "proc", 0, NULL);
+	if (ret) {
+		fprintf(stderr, "%m - Failed to mount proc\n");
+		return -1;
+	}
+
+	fd = open("/proc/sys/kernel/pid_max", O_RDWR | O_CLOEXEC | O_NOCTTY);
+	if (fd < 0) {
+		fprintf(stderr, "%m - Failed to open pid_max\n");
+		return -1;
+	}
+
+	ret = write_int_to_fd(fd, inner_limit);
+	if (ret < 0) {
+		fprintf(stderr, "%m - Failed to write pid_max\n");
+		return -1;
+	}
+
+	for (int i = 0; i < inner_limit + 1; i++) {
+		pid = fork();
+		if (pid == 0)
+			exit(EXIT_SUCCESS);
+		wait_for_pid(pid);
+		if (pid > inner_limit) {
+			fprintf(stderr, "Managed to create pid number beyond limit\n");
+			return -1;
+		}
+	}
+
+	return 0;
+}
+
+static int pid_max_nested_inner(void *data)
+{
+	int fret = -1;
+	pid_t pids[2];
+	int fd, i, ret;
+
+	ret = mount("", "/", NULL, MS_PRIVATE | MS_REC, 0);
+	if (ret) {
+		fprintf(stderr, "%m - Failed to make rootfs private mount\n");
+		return fret;
+	}
+
+	umount2("/proc", MNT_DETACH);
+
+	ret = mount("proc", "/proc", "proc", 0, NULL);
+	if (ret) {
+		fprintf(stderr, "%m - Failed to mount proc\n");
+		return fret;
+	}
+
+	fd = open("/proc/sys/kernel/pid_max", O_RDWR | O_CLOEXEC | O_NOCTTY);
+	if (fd < 0) {
+		fprintf(stderr, "%m - Failed to open pid_max\n");
+		return fret;
+	}
+
+	ret = write_int_to_fd(fd, inner_limit);
+	close(fd);
+	if (ret < 0) {
+		fprintf(stderr, "%m - Failed to write pid_max\n");
+		return fret;
+	}
+
+	pids[0] = fork();
+	if (pids[0] < 0) {
+		fprintf(stderr, "Failed to create first new process\n");
+		return fret;
+	}
+
+	if (pids[0] == 0)
+		exit(EXIT_SUCCESS);
+
+	pids[1] = fork();
+	wait_for_pid(pids[0]);
+	if (pids[1] >= 0) {
+		if (pids[1] == 0)
+			exit(EXIT_SUCCESS);
+		wait_for_pid(pids[1]);
+
+		fprintf(stderr, "Managed to create process even though ancestor pid namespace had a limit\n");
+		return fret;
+	}
+
+	/* Now make sure that we wrap pids at outer_limit. */
+	for (i = 0; i < inner_limit + 10; i++) {
+		pid_t pid;
+
+		pid = fork();
+		if (pid < 0)
+			return fret;
+
+		if (pid == 0)
+			exit(EXIT_SUCCESS);
+
+		wait_for_pid(pid);
+		if (pid >= inner_limit) {
+			fprintf(stderr, "Managed to create process with pid %d beyond configured limit\n", pid);
+			return fret;
+		}
+	}
+
+	return 0;
+}
+
+static int pid_max_nested_outer(void *data)
+{
+	int fret = -1, nr_procs = 0;
+	pid_t *pids;
+	int fd, ret;
+	pid_t pid;
+
+	pids = malloc(outer_limit * sizeof(pid_t));
+	if (!pids)
+		return -1;
+
+	ret = mount("", "/", NULL, MS_PRIVATE | MS_REC, 0);
+	if (ret) {
+		fprintf(stderr, "%m - Failed to make rootfs private mount\n");
+		goto out;
+	}
+
+	umount2("/proc", MNT_DETACH);
+
+	ret = mount("proc", "/proc", "proc", 0, NULL);
+	if (ret) {
+		fprintf(stderr, "%m - Failed to mount proc\n");
+		goto out;
+	}
+
+	fd = open("/proc/sys/kernel/pid_max", O_RDWR | O_CLOEXEC | O_NOCTTY);
+	if (fd < 0) {
+		fprintf(stderr, "%m - Failed to open pid_max\n");
+		goto out;
+	}
+
+	ret = write_int_to_fd(fd, outer_limit);
+	close(fd);
+	if (ret < 0) {
+		fprintf(stderr, "%m - Failed to write pid_max\n");
+		goto out;
+	}
+
+	/*
+	 * Create (outer_limit - 4) processes. This leaves room for
+	 * do_clone() and one more. So creating another process needs
+	 * to fail.
+	 */
+	for (nr_procs = 0; nr_procs < outer_limit - 4; nr_procs++) {
+		pid = fork();
+		if (pid < 0)
+			goto reap;
+
+		if (pid == 0)
+			exit(EXIT_SUCCESS);
+
+		pids[nr_procs] = pid;
+	}
+
+	pid = do_clone(pid_max_nested_inner, NULL, CLONE_NEWPID | CLONE_NEWNS);
+	if (pid < 0) {
+		fprintf(stderr, "%m - Failed to clone nested pidns\n");
+		goto reap;
+	}
+
+	if (wait_for_pid(pid)) {
+		fprintf(stderr, "%m - Nested pid_max failed\n");
+		goto reap;
+	}
+
+	fret = 0;
+
+reap:
+	for (int i = 0; i < nr_procs; i++)
+		wait_for_pid(pids[i]);
+
+out:
+	free(pids);
+	return fret;
+}
+
+static int pid_max_nested_limit_inner(void *data)
+{
+	int fret = -1, nr_procs = 0;
+	int fd, ret;
+	pid_t pid;
+	pid_t *pids;
+
+	pids = malloc(inner_limit * sizeof(pid_t));
+	if (!pids)
+		return -1;
+
+	ret = mount("", "/", NULL, MS_PRIVATE | MS_REC, 0);
+	if (ret) {
+		fprintf(stderr, "%m - Failed to make rootfs private mount\n");
+		goto out;
+	}
+
+	umount2("/proc", MNT_DETACH);
+
+	ret = mount("proc", "/proc", "proc", 0, NULL);
+	if (ret) {
+		fprintf(stderr, "%m - Failed to mount proc\n");
+		goto out;
+	}
+
+	fd = open("/proc/sys/kernel/pid_max", O_RDWR | O_CLOEXEC | O_NOCTTY);
+	if (fd < 0) {
+		fprintf(stderr, "%m - Failed to open pid_max\n");
+		goto out;
+	}
+
+	ret = write_int_to_fd(fd, inner_limit);
+	close(fd);
+	if (ret < 0) {
+		fprintf(stderr, "%m - Failed to write pid_max\n");
+		goto out;
+	}
+
+	for (nr_procs = 0; nr_procs < inner_limit; nr_procs++) {
+		pid = fork();
+		if (pid < 0)
+			break;
+
+		if (pid == 0)
+			exit(EXIT_SUCCESS);
+
+		pids[nr_procs] = pid;
+	}
+
+	if (nr_procs >= outer_limit) {
+		fprintf(stderr, "Managed to create processes beyond the configured outer limit\n");
+		goto reap;
+	}
+
+	fret = 0;
+
+reap:
+	for (int i = 0; i < nr_procs; i++)
+		wait_for_pid(pids[i]);
+
+out:
+	free(pids);
+	return fret;
+}
+
+static int pid_max_nested_limit_outer(void *data)
+{
+	int fd, ret;
+	pid_t pid;
+
+	ret = mount("", "/", NULL, MS_PRIVATE | MS_REC, 0);
+	if (ret) {
+		fprintf(stderr, "%m - Failed to make rootfs private mount\n");
+		return -1;
+	}
+
+	umount2("/proc", MNT_DETACH);
+
+	ret = mount("proc", "/proc", "proc", 0, NULL);
+	if (ret) {
+		fprintf(stderr, "%m - Failed to mount proc\n");
+		return -1;
+	}
+
+	fd = open("/proc/sys/kernel/pid_max", O_RDWR | O_CLOEXEC | O_NOCTTY);
+	if (fd < 0) {
+		fprintf(stderr, "%m - Failed to open pid_max\n");
+		return -1;
+	}
+
+	ret = write_int_to_fd(fd, outer_limit);
+	close(fd);
+	if (ret < 0) {
+		fprintf(stderr, "%m - Failed to write pid_max\n");
+		return -1;
+	}
+
+	pid = do_clone(pid_max_nested_limit_inner, NULL, CLONE_NEWPID | CLONE_NEWNS);
+	if (pid < 0) {
+		fprintf(stderr, "%m - Failed to clone nested pidns\n");
+		return -1;
+	}
+
+	if (wait_for_pid(pid)) {
+		fprintf(stderr, "%m - Nested pid_max failed\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+FIXTURE(pid_max) {
+	int dummy;
+};
+
+FIXTURE_SETUP(pid_max)
+{
+	int min = pid_min();
+
+	outer_limit = min + 100;
+	inner_limit = min + 200;
+}
+
+FIXTURE_TEARDOWN(pid_max)
+{
+}
+
+TEST_F(pid_max, simple)
+{
+	pid_t pid;
+
+	pid = do_clone(pid_max_cb, NULL, CLONE_NEWPID | CLONE_NEWNS);
+	ASSERT_GT(pid, 0);
+	ASSERT_EQ(0, wait_for_pid(pid));
+}
+
+TEST_F(pid_max, nested_limit)
+{
+	pid_t pid;
+
+	pid = do_clone(pid_max_nested_limit_outer, NULL, CLONE_NEWPID | CLONE_NEWNS);
+	ASSERT_GT(pid, 0);
+	ASSERT_EQ(0, wait_for_pid(pid));
+}
+
+TEST_F(pid_max, nested)
+{
+	pid_t pid;
+
+	pid = do_clone(pid_max_nested_outer, NULL, CLONE_NEWPID | CLONE_NEWNS);
+	ASSERT_GT(pid, 0);
+	ASSERT_EQ(0, wait_for_pid(pid));
+}
+
+TEST_HARNESS_MAIN

linux-7.2.3/tools/testing/selftests/pid_namespace/pidns_init_via_setns.c

@@ -0,0 +1,238 @@
+// SPDX-License-Identifier: GPL-2.0
+#define _GNU_SOURCE
+#include <fcntl.h>
+#include <sched.h>
+#include <stdio.h>
+#include <sys/types.h>
+#include <unistd.h>
+
+#include "kselftest_harness.h"
+#include "../pidfd/pidfd.h"
+
+/*
+ * Test that a process can become PID 1 (init) in a new PID namespace
+ * created via unshare() and joined via setns().
+ *
+ * Flow:
+ *  1. Parent creates a pipe for synchronization.
+ *  2. Parent forks a child.
+ *  3. Parent calls unshare(CLONE_NEWPID) to create a new PID namespace.
+ *  4. Parent signals the child via the pipe.
+ *  5. Child opens parent's /proc/<ppid>/ns/pid_for_children and calls
+ *     setns(fd, CLONE_NEWPID) to join the new namespace.
+ *  6. Child forks a grandchild.
+ *  7. Grandchild verifies getpid() == 1.
+ */
+TEST(pidns_init_via_setns)
+{
+	pid_t child, parent_pid;
+	int pipe_fd[2];
+	char buf;
+
+	if (geteuid())
+		ASSERT_EQ(0, unshare(CLONE_NEWUSER));
+
+	parent_pid = getpid();
+
+	ASSERT_EQ(0, pipe(pipe_fd));
+
+	child = fork();
+	ASSERT_GE(child, 0);
+
+	if (child == 0) {
+		char path[256];
+		int nsfd;
+		pid_t grandchild;
+
+		close(pipe_fd[1]);
+
+		/* Wait for parent to complete unshare */
+		ASSERT_EQ(1, read_nointr(pipe_fd[0], &buf, 1));
+		close(pipe_fd[0]);
+
+		snprintf(path, sizeof(path),
+			 "/proc/%d/ns/pid_for_children", parent_pid);
+		nsfd = open(path, O_RDONLY);
+		ASSERT_GE(nsfd, 0);
+
+		ASSERT_EQ(0, setns(nsfd, CLONE_NEWPID));
+		close(nsfd);
+
+		grandchild = fork();
+		ASSERT_GE(grandchild, 0);
+
+		if (grandchild == 0) {
+			/* Should be init (PID 1) in the new namespace */
+			if (getpid() != 1)
+				_exit(1);
+			_exit(0);
+		}
+
+		ASSERT_EQ(0, wait_for_pid(grandchild));
+		_exit(0);
+	}
+
+	close(pipe_fd[0]);
+
+	ASSERT_EQ(0, unshare(CLONE_NEWPID));
+
+	/* Signal child that the new PID namespace is ready */
+	buf = 0;
+	ASSERT_EQ(1, write_nointr(pipe_fd[1], &buf, 1));
+	close(pipe_fd[1]);
+
+	ASSERT_EQ(0, wait_for_pid(child));
+}
+
+/*
+ * Similar to pidns_init_via_setns, but:
+ *  1. Parent enters a new PID namespace right from the start to be able to
+ *     later freely use pid 1001 in it.
+ *  2. After forking child, parent also calls unshare(CLONE_NEWUSER)
+ *     before unshare(CLONE_NEWPID) so that new old and new pid namespaces have
+ *     different user namespace owners.
+ *  3. Child uses clone3() with set_tid={1, 1001} instead of fork() and
+ *     grandchild checks that it gets desired pids .
+ *
+ * Flow:
+ *  1. Test process creates a new PID namespace and forks a wrapper
+ *     (PID 1 in the outer namespace).
+ *  2. Wrapper forks a child.
+ *  3. Wrapper calls unshare(CLONE_NEWUSER) + unshare(CLONE_NEWPID)
+ *     to create an inner PID namespace.
+ *  4. Wrapper signals the child via pipe.
+ *  5. Child opens wrapper's /proc/<pid>/ns/pid_for_children and calls
+ *     setns(fd, CLONE_NEWPID) to join the inner namespace.
+ *  6. Child calls clone3() with set_tid={1, 1001}.
+ *  7. Grandchild verifies its NSpid ends with "1001 1".
+ */
+
+pid_t set_tid[] = {1, 1001};
+
+static int pidns_init_via_setns_set_tid_grandchild(struct __test_metadata *_metadata)
+{
+	char *line = NULL;
+	size_t len = 0;
+	int found = 0;
+	FILE *gf;
+
+	gf = fopen("/proc/self/status", "r");
+	ASSERT_NE(gf, NULL);
+
+	while (getline(&line, &len, gf) != -1) {
+		if (strncmp(line, "NSpid:", 6) != 0)
+			continue;
+
+		for (int i = 0; i < 2; i++) {
+			char *last = strrchr(line, '\t');
+			pid_t pid;
+
+			ASSERT_NE(last, NULL);
+			ASSERT_EQ(sscanf(last, "%d", &pid), 1);
+			ASSERT_EQ(pid, set_tid[i]);
+			*last = '\0';
+		}
+
+		found = true;
+		break;
+	}
+
+	free(line);
+	fclose(gf);
+	ASSERT_TRUE(found);
+	return 0;
+}
+
+static int pidns_init_via_setns_set_tid_child(struct __test_metadata *_metadata,
+					      pid_t parent_pid, int pipe_fd[2])
+{
+	struct __clone_args args = {
+		.exit_signal	= SIGCHLD,
+		.set_tid	= ptr_to_u64(set_tid),
+		.set_tid_size	= 2,
+	};
+	pid_t grandchild;
+	char path[256];
+	char buf;
+	int nsfd;
+
+	close(pipe_fd[1]);
+
+	ASSERT_EQ(1, read_nointr(pipe_fd[0], &buf, 1));
+	close(pipe_fd[0]);
+
+	snprintf(path, sizeof(path),
+		 "/proc/%d/ns/pid_for_children", parent_pid);
+	nsfd = open(path, O_RDONLY);
+	ASSERT_GE(nsfd, 0);
+
+	ASSERT_EQ(0, setns(nsfd, CLONE_NEWPID));
+	close(nsfd);
+
+	grandchild = sys_clone3(&args, sizeof(args));
+	ASSERT_GE(grandchild, 0);
+
+	if (grandchild == 0)
+		_exit(pidns_init_via_setns_set_tid_grandchild(_metadata));
+
+	ASSERT_EQ(0, wait_for_pid(grandchild));
+	return 0;
+}
+
+static int pidns_init_via_setns_set_tid_wrapper(struct __test_metadata *_metadata)
+{
+	int pipe_fd[2];
+	pid_t child, parent_pid;
+	char buf;
+	FILE *f;
+
+	/*
+	 * We are PID 1 inside the new namespace, but /proc is
+	 * mounted from the host.  Read our host-visible PID so
+	 * the child can reach our pid_for_children via /proc.
+	 */
+	f = fopen("/proc/self/stat", "r");
+	ASSERT_NE(f, NULL);
+	ASSERT_EQ(fscanf(f, "%d", &parent_pid), 1);
+	ASSERT_EQ(0, pipe(pipe_fd));
+
+	child = fork();
+	ASSERT_GE(child, 0);
+
+	if (child == 0)
+		_exit(pidns_init_via_setns_set_tid_child(_metadata, parent_pid, pipe_fd));
+
+	close(pipe_fd[0]);
+
+	ASSERT_EQ(0, unshare(CLONE_NEWUSER));
+	ASSERT_EQ(0, unshare(CLONE_NEWPID));
+
+	buf = 0;
+	ASSERT_EQ(1, write_nointr(pipe_fd[1], &buf, 1));
+	close(pipe_fd[1]);
+
+	ASSERT_EQ(0, wait_for_pid(child));
+
+	fclose(f);
+	return 0;
+}
+
+TEST(pidns_init_via_setns_set_tid)
+{
+	pid_t wrapper;
+
+	if (geteuid())
+		SKIP(return, "This test needs root to run!");
+
+	ASSERT_EQ(0, unshare(CLONE_NEWPID));
+
+	wrapper = fork();
+	ASSERT_GE(wrapper, 0);
+
+	if (wrapper == 0)
+		_exit(pidns_init_via_setns_set_tid_wrapper(_metadata));
+
+	ASSERT_EQ(0, wait_for_pid(wrapper));
+}
+
+TEST_HARNESS_MAIN

linux-7.2.3/tools/testing/selftests/pid_namespace/regression_enomem.c

@@ -0,0 +1,44 @@
+#define _GNU_SOURCE
+#include <assert.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <linux/types.h>
+#include <sched.h>
+#include <signal.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <syscall.h>
+#include <sys/wait.h>
+
+#include "kselftest_harness.h"
+#include "../pidfd/pidfd.h"
+
+/*
+ * Regression test for:
+ * 35f71bc0a09a ("fork: report pid reservation failure properly")
+ * b26ebfe12f34 ("pid: Fix error return value in some cases")
+ */
+TEST(regression_enomem)
+{
+	pid_t pid;
+
+	if (geteuid())
+		EXPECT_EQ(0, unshare(CLONE_NEWUSER));
+
+	EXPECT_EQ(0, unshare(CLONE_NEWPID));
+
+	pid = fork();
+	ASSERT_GE(pid, 0);
+
+	if (pid == 0)
+		exit(EXIT_SUCCESS);
+
+	EXPECT_EQ(0, wait_for_pid(pid));
+
+	pid = fork();
+	ASSERT_LT(pid, 0);
+	ASSERT_EQ(errno, ENOMEM);
+}
+
+TEST_HARNESS_MAIN