Merge branch 'bpf_get_ns_current_pid_tgid'

Carlos Neira says:

====================
Currently bpf_get_current_pid_tgid(), is used to do pid filtering in bcc's
scripts but this helper returns the pid as seen by the root namespace which is
fine when a bcc script is not executed inside a container.
When the process of interest is inside a container, pid filtering will not work
if bpf_get_current_pid_tgid() is used.
This helper addresses this limitation returning the pid as it's seen by the current
namespace where the script is executing.

In the future different pid_ns files may belong to different devices, according to the
discussion between Eric Biederman and Yonghong in 2017 Linux plumbers conference.
To address that situation the helper requires inum and dev_t from /proc/self/ns/pid.
This helper has the same use cases as bpf_get_current_pid_tgid() as it can be
used to do pid filtering even inside a container.
====================

Signed-off-by: Alexei Starovoitov <ast@kernel.org>

5 files changed, 287 insertions, 1 deletions

diff --git a/tools/testing/selftests/bpf/.gitignore b/tools/testing/selftests/bpf/.gitignore
index ec464859c6b6..2198cd876675 100644
--- a/tools/testing/selftests/bpf/.gitignore
+++ b/tools/testing/selftests/bpf/.gitignore
@@ -31,6 +31,7 @@ test_tcp_check_syncookie_user
 test_sysctl
 test_hashmap
 test_btf_dump
+test_current_pid_tgid_new_ns
 xdping
 test_cpp
 *.skel.h
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index ee4ad34adb4a..da4389dde9f7 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -32,7 +32,8 @@ TEST_GEN_PROGS = test_verifier test_tag test_maps test_lru_map test_lpm_map test
 	test_sock test_btf test_sockmap get_cgroup_id_user test_socket_cookie \
 	test_cgroup_storage \
 	test_netcnt test_tcpnotify_user test_sock_fields test_sysctl test_hashmap \
-	test_progs-no_alu32
+	test_progs-no_alu32 \
+	test_current_pid_tgid_new_ns
 
 # Also test bpf-gcc, if present
 ifneq ($(BPF_GCC),)
diff --git a/tools/testing/selftests/bpf/prog_tests/ns_current_pid_tgid.c b/tools/testing/selftests/bpf/prog_tests/ns_current_pid_tgid.c
new file mode 100644
index 000000000000..542240e16564
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/ns_current_pid_tgid.c
@@ -0,0 +1,88 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2020 Carlos Neira cneirabustos@gmail.com */
+#include <test_progs.h>
+#include <sys/stat.h>
+#include <sys/types.h>
+#include <unistd.h>
+#include <sys/syscall.h>
+
+struct bss {
+	__u64 dev;
+	__u64 ino;
+	__u64 pid_tgid;
+	__u64 user_pid_tgid;
+};
+
+void test_ns_current_pid_tgid(void)
+{
+	const char *probe_name = "raw_tracepoint/sys_enter";
+	const char *file = "test_ns_current_pid_tgid.o";
+	int err, key = 0, duration = 0;
+	struct bpf_link *link = NULL;
+	struct bpf_program *prog;
+	struct bpf_map *bss_map;
+	struct bpf_object *obj;
+	struct bss bss;
+	struct stat st;
+	__u64 id;
+
+	obj = bpf_object__open_file(file, NULL);
+	if (CHECK(IS_ERR(obj), "obj_open", "err %ld\n", PTR_ERR(obj)))
+		return;
+
+	err = bpf_object__load(obj);
+	if (CHECK(err, "obj_load", "err %d errno %d\n", err, errno))
+		goto cleanup;
+
+	bss_map = bpf_object__find_map_by_name(obj, "test_ns_.bss");
+	if (CHECK(!bss_map, "find_bss_map", "failed\n"))
+		goto cleanup;
+
+	prog = bpf_object__find_program_by_title(obj, probe_name);
+	if (CHECK(!prog, "find_prog", "prog '%s' not found\n",
+		  probe_name))
+		goto cleanup;
+
+	memset(&bss, 0, sizeof(bss));
+	pid_t tid = syscall(SYS_gettid);
+	pid_t pid = getpid();
+
+	id = (__u64) tid << 32 | pid;
+	bss.user_pid_tgid = id;
+
+	if (CHECK_FAIL(stat("/proc/self/ns/pid", &st))) {
+		perror("Failed to stat /proc/self/ns/pid");
+		goto cleanup;
+	}
+
+	bss.dev = st.st_dev;
+	bss.ino = st.st_ino;
+
+	err = bpf_map_update_elem(bpf_map__fd(bss_map), &key, &bss, 0);
+	if (CHECK(err, "setting_bss", "failed to set bss : %d\n", err))
+		goto cleanup;
+
+	link = bpf_program__attach_raw_tracepoint(prog, "sys_enter");
+	if (CHECK(IS_ERR(link), "attach_raw_tp", "err %ld\n",
+		  PTR_ERR(link))) {
+		link = NULL;
+		goto cleanup;
+	}
+
+	/* trigger some syscalls */
+	usleep(1);
+
+	err = bpf_map_lookup_elem(bpf_map__fd(bss_map), &key, &bss);
+	if (CHECK(err, "set_bss", "failed to get bss : %d\n", err))
+		goto cleanup;
+
+	if (CHECK(id != bss.pid_tgid, "Compare user pid/tgid vs. bpf pid/tgid",
+		  "User pid/tgid %llu BPF pid/tgid %llu\n", id, bss.pid_tgid))
+		goto cleanup;
+cleanup:
+	if (!link) {
+		bpf_link__destroy(link);
+		link = NULL;
+	}
+	bpf_object__close(obj);
+}
diff --git a/tools/testing/selftests/bpf/progs/test_ns_current_pid_tgid.c b/tools/testing/selftests/bpf/progs/test_ns_current_pid_tgid.c
new file mode 100644
index 000000000000..1dca70a6de2f
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_ns_current_pid_tgid.c
@@ -0,0 +1,37 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Carlos Neira cneirabustos@gmail.com */
+
+#include <linux/bpf.h>
+#include <stdint.h>
+#include <bpf/bpf_helpers.h>
+
+static volatile struct {
+	__u64 dev;
+	__u64 ino;
+	__u64 pid_tgid;
+	__u64 user_pid_tgid;
+} res;
+
+SEC("raw_tracepoint/sys_enter")
+int trace(void *ctx)
+{
+	__u64  ns_pid_tgid, expected_pid;
+	struct bpf_pidns_info nsdata;
+	__u32 key = 0;
+
+	if (bpf_get_ns_current_pid_tgid(res.dev, res.ino, &nsdata,
+		   sizeof(struct bpf_pidns_info)))
+		return 0;
+
+	ns_pid_tgid = (__u64)nsdata.tgid << 32 | nsdata.pid;
+	expected_pid = res.user_pid_tgid;
+
+	if (expected_pid != ns_pid_tgid)
+		return 0;
+
+	res.pid_tgid = ns_pid_tgid;
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/test_current_pid_tgid_new_ns.c b/tools/testing/selftests/bpf/test_current_pid_tgid_new_ns.c
new file mode 100644
index 000000000000..ed253f252cd0
--- /dev/null
+++ b/tools/testing/selftests/bpf/test_current_pid_tgid_new_ns.c
@@ -0,0 +1,159 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2020 Carlos Neira cneirabustos@gmail.com */
+#define _GNU_SOURCE
+#include <sys/stat.h>
+#include <sys/types.h>
+#include <unistd.h>
+#include <sys/syscall.h>
+#include <sched.h>
+#include <sys/wait.h>
+#include <sys/mount.h>
+#include "test_progs.h"
+
+#define CHECK_NEWNS(condition, tag, format...) ({		\
+	int __ret = !!(condition);			\
+	if (__ret) {					\
+		printf("%s:FAIL:%s ", __func__, tag);	\
+		printf(format);				\
+	} else {					\
+		printf("%s:PASS:%s\n", __func__, tag);	\
+	}						\
+	__ret;						\
+})
+
+struct bss {
+	__u64 dev;
+	__u64 ino;
+	__u64 pid_tgid;
+	__u64 user_pid_tgid;
+};
+
+int main(int argc, char **argv)
+{
+	pid_t pid;
+	int exit_code = 1;
+	struct stat st;
+
+	printf("Testing bpf_get_ns_current_pid_tgid helper in new ns\n");
+
+	if (stat("/proc/self/ns/pid", &st)) {
+		perror("stat failed on /proc/self/ns/pid ns\n");
+		printf("%s:FAILED\n", argv[0]);
+		return exit_code;
+	}
+
+	if (CHECK_NEWNS(unshare(CLONE_NEWPID | CLONE_NEWNS),
+			"unshare CLONE_NEWPID | CLONE_NEWNS", "error errno=%d\n", errno))
+		return exit_code;
+
+	pid = fork();
+	if (pid == -1) {
+		perror("Fork() failed\n");
+		printf("%s:FAILED\n", argv[0]);
+		return exit_code;
+	}
+
+	if (pid > 0) {
+		int status;
+
+		usleep(5);
+		waitpid(pid, &status, 0);
+		return 0;
+	} else {
+
+		pid = fork();
+		if (pid == -1) {
+			perror("Fork() failed\n");
+			printf("%s:FAILED\n", argv[0]);
+			return exit_code;
+		}
+
+		if (pid > 0) {
+			int status;
+			waitpid(pid, &status, 0);
+			return 0;
+		} else {
+			if (CHECK_NEWNS(mount("none", "/proc", NULL, MS_PRIVATE|MS_REC, NULL),
+				"Unmounting proc", "Cannot umount proc! errno=%d\n", errno))
+				return exit_code;
+
+			if (CHECK_NEWNS(mount("proc", "/proc", "proc", MS_NOSUID|MS_NOEXEC|MS_NODEV, NULL),
+				"Mounting proc", "Cannot mount proc! errno=%d\n", errno))
+				return exit_code;
+
+			const char *probe_name = "raw_tracepoint/sys_enter";
+			const char *file = "test_ns_current_pid_tgid.o";
+			struct bpf_link *link = NULL;
+			struct bpf_program *prog;
+			struct bpf_map *bss_map;
+			struct bpf_object *obj;
+			int exit_code = 1;
+			int err, key = 0;
+			struct bss bss;
+			struct stat st;
+			__u64 id;
+
+			obj = bpf_object__open_file(file, NULL);
+			if (CHECK_NEWNS(IS_ERR(obj), "obj_open", "err %ld\n", PTR_ERR(obj)))
+				return exit_code;
+
+			err = bpf_object__load(obj);
+			if (CHECK_NEWNS(err, "obj_load", "err %d errno %d\n", err, errno))
+				goto cleanup;
+
+			bss_map = bpf_object__find_map_by_name(obj, "test_ns_.bss");
+			if (CHECK_NEWNS(!bss_map, "find_bss_map", "failed\n"))
+				goto cleanup;
+
+			prog = bpf_object__find_program_by_title(obj, probe_name);
+			if (CHECK_NEWNS(!prog, "find_prog", "prog '%s' not found\n",
+						probe_name))
+				goto cleanup;
+
+			memset(&bss, 0, sizeof(bss));
+			pid_t tid = syscall(SYS_gettid);
+			pid_t pid = getpid();
+
+			id = (__u64) tid << 32 | pid;
+			bss.user_pid_tgid = id;
+
+			if (CHECK_NEWNS(stat("/proc/self/ns/pid", &st),
+				"stat new ns", "Failed to stat /proc/self/ns/pid errno=%d\n", errno))
+				goto cleanup;
+
+			bss.dev = st.st_dev;
+			bss.ino = st.st_ino;
+
+			err = bpf_map_update_elem(bpf_map__fd(bss_map), &key, &bss, 0);
+			if (CHECK_NEWNS(err, "setting_bss", "failed to set bss : %d\n", err))
+				goto cleanup;
+
+			link = bpf_program__attach_raw_tracepoint(prog, "sys_enter");
+			if (CHECK_NEWNS(IS_ERR(link), "attach_raw_tp", "err %ld\n",
+						PTR_ERR(link))) {
+				link = NULL;
+				goto cleanup;
+			}
+
+			/* trigger some syscalls */
+			usleep(1);
+
+			err = bpf_map_lookup_elem(bpf_map__fd(bss_map), &key, &bss);
+			if (CHECK_NEWNS(err, "set_bss", "failed to get bss : %d\n", err))
+				goto cleanup;
+
+			if (CHECK_NEWNS(id != bss.pid_tgid, "Compare user pid/tgid vs. bpf pid/tgid",
+						"User pid/tgid %llu BPF pid/tgid %llu\n", id, bss.pid_tgid))
+				goto cleanup;
+
+			exit_code = 0;
+			printf("%s:PASS\n", argv[0]);
+cleanup:
+			if (!link) {
+				bpf_link__destroy(link);
+				link = NULL;
+			}
+			bpf_object__close(obj);
+		}
+	}
+}