Skip to contentSkip to conversation

bpf: Fix iter/task tid filtering

PATCH Superseded Jordan Rome 2 participants 4 messages Oct 15, 2024 – Oct 16, 2024

Subsystems (4)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/2 bpf: Fix iter/task tid filtering

In userspace, you can add a tid filter by setting the "task.tid" field for "bpf_iter_link_info". However, get_pid_task when called for the BPF_TASK_ITER_TID type should have been using PIDTYPE_PID (tid) instead of PIDTYPE_TGID (pid).

Fixes f0d74c4da1f0 ("bpf: Parameterize task iterators.")
---
 kernel/bpf/task_iter.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/bpf/task_iter.c +1 -1
index 02aa9db8d796..5af9e130e500 100644
@@ -99,7 +99,7 @@ static struct task_struct *task_seq_get_next(struct bpf_iter_seq_task_common *co
 		rcu_read_lock();
 		pid = find_pid_ns(common->pid, common->ns);
 		if (pid) {
-			task = get_pid_task(pid, PIDTYPE_TGID);
+			task = get_pid_task(pid, PIDTYPE_PID);
 			*tid = common->pid;
 		}
 		rcu_read_unlock();

--
2.43.5

#1 lore raw

PATCH 2/2 bpf: properly test iter/task tid filtering

Previously test_task_tid was setting linfo.task.tid to getpid() which is the same as gettid() for the parent process. Instead create a new child thread and set linfo.task.tid to gettid() to make sure the tid filtering logic is working as expected.

---
 .../selftests/bpf/prog_tests/bpf_iter.c       | 26 +++++++++++++++----
 1 file changed, 21 insertions(+), 5 deletions(-)
tools/testing/selftests/bpf/prog_tests/bpf_iter.c +21 -5
index 52e6f7570475..5b056eb5d166 100644
@@ -226,7 +226,7 @@ static void test_task_common_nocheck(struct bpf_iter_attach_opts *opts,
 	ASSERT_OK(pthread_create(&thread_id, NULL, &do_nothing_wait, NULL),
 		  "pthread_create");
 
-	skel->bss->tid = getpid();
+	skel->bss->tid = gettid();
 
 	do_dummy_read_opts(skel->progs.dump_task, opts);
 
@@ -249,25 +249,41 @@ static void test_task_common(struct bpf_iter_attach_opts *opts, int num_unknown,
 	ASSERT_EQ(num_known_tid, num_known, "check_num_known_tid");
 }
 
-static void test_task_tid(void)
+static void *run_test_task_tid(void *arg)
 {
+	ASSERT_NEQ(getpid(), gettid(), "check_new_thread_id");
 	LIBBPF_OPTS(bpf_iter_attach_opts, opts);
 	union bpf_iter_link_info linfo;
 	int num_unknown_tid, num_known_tid;
 
 	memset(&linfo, 0, sizeof(linfo));
-	linfo.task.tid = getpid();
+	linfo.task.tid = gettid();
 	opts.link_info = &linfo;
 	opts.link_info_len = sizeof(linfo);
 	test_task_common(&opts, 0, 1);
 
 	linfo.task.tid = 0;
 	linfo.task.pid = getpid();
-	test_task_common(&opts, 1, 1);
+	// This includes the parent thread, this thread, and the do_nothing_wait thread
+	test_task_common(&opts, 2, 1);
 
 	test_task_common_nocheck(NULL, &num_unknown_tid, &num_known_tid);
-	ASSERT_GT(num_unknown_tid, 1, "check_num_unknown_tid");
+	ASSERT_GT(num_unknown_tid, 2, "check_num_unknown_tid");
 	ASSERT_EQ(num_known_tid, 1, "check_num_known_tid");
+
+	pthread_exit(arg);
+}
+
+static void test_task_tid(void)
+{
+	pthread_t thread_id;
+	void *ret;
+
+	// Create a new thread so pid and tid aren't the same
+	ASSERT_OK(pthread_create(&thread_id, NULL, &run_test_task_tid, NULL),
+		  "pthread_create");
+	ASSERT_FALSE(pthread_join(thread_id, &ret) || ret != NULL,
+		     "pthread_join");
 }
 
 static void test_task_pid(void)

--
2.43.5

#2 lore raw

Re: [bpf-next v1 2/2] bpf: properly test iter/task tid filtering

Quoted 26 lines · Jordan Rome

Previously test_task_tid was setting linfo.task.tid to getpid() which is the same as gettid() for the parent process. Instead create a new child thread and set linfo.task.tid to gettid() to make sure the tid filtering logic is working as expected.

---
 .../selftests/bpf/prog_tests/bpf_iter.c       | 26 +++++++++++++++----
 1 file changed, 21 insertions(+), 5 deletions(-)
tools/testing/selftests/bpf/prog_tests/bpf_iter.c +1 -1
index 52e6f7570475..5b056eb5d166 100644
@@ -226,7 +226,7 @@ static void test_task_common_nocheck(struct bpf_iter_attach_opts *opts,
        ASSERT_OK(pthread_create(&thread_id, NULL, &do_nothing_wait, NULL),
                  "pthread_create");
 
-       skel->bss->tid = getpid();
+       skel->bss->tid = gettid();
 
        do_dummy_read_opts(skel->progs.dump_task, opts);
 

tools/testing/selftests/bpf/prog_tests/bpf_iter.c static void test_task_common(struct bpf_iter_attach_opts *opts, int num_unknown, · line 254

Quoted 7 lines
tools/testing/selftests/bpf/prog_tests/bpf_iter.c +2 -1
@@ -249,25 +249,41 @@ static void test_task_common(struct bpf_iter_attach_opts *opts, int num_unknown,
        ASSERT_EQ(num_known_tid, num_known, "check_num_known_tid");
 }
 
-static void test_task_tid(void)
+static void *run_test_task_tid(void *arg)
 {
+       ASSERT_NEQ(getpid(), gettid(), "check_new_thread_id");

this is variable declaration block, move assertion after it (and empty line)

tools/testing/selftests/bpf/prog_tests/bpf_iter.c static void test_task_common(struct bpf_iter_attach_opts *opts, int num_unknown, · line 258

Quoted 4 lines
tools/testing/selftests/bpf/prog_tests/bpf_iter.c +0 -0
        LIBBPF_OPTS(bpf_iter_attach_opts, opts);
        union bpf_iter_link_info linfo;
        int num_unknown_tid, num_known_tid;
 

here

tools/testing/selftests/bpf/prog_tests/bpf_iter.c static void test_task_common(struct bpf_iter_attach_opts *opts, int num_unknown, · line 267

Quoted 11 lines
tools/testing/selftests/bpf/prog_tests/bpf_iter.c +2 -2
        memset(&linfo, 0, sizeof(linfo));
-       linfo.task.tid = getpid();
+       linfo.task.tid = gettid();
        opts.link_info = &linfo;
        opts.link_info_len = sizeof(linfo);
        test_task_common(&opts, 0, 1);
 
        linfo.task.tid = 0;
        linfo.task.pid = getpid();
-       test_task_common(&opts, 1, 1);
+       // This includes the parent thread, this thread, and the do_nothing_wait thread

we don't use C++-style comments in C code base, please use /* */

tools/testing/selftests/bpf/prog_tests/bpf_iter.c static void test_task_common(struct bpf_iter_attach_opts *opts, int num_unknown, · line 274

Quoted 8 lines
tools/testing/selftests/bpf/prog_tests/bpf_iter.c +4 -1
+       test_task_common(&opts, 2, 1);
 
        test_task_common_nocheck(NULL, &num_unknown_tid, &num_known_tid);
-       ASSERT_GT(num_unknown_tid, 1, "check_num_unknown_tid");
+       ASSERT_GT(num_unknown_tid, 2, "check_num_unknown_tid");
        ASSERT_EQ(num_known_tid, 1, "check_num_known_tid");
+
+       pthread_exit(arg);

nit: wouldn't return arg; do the same?

tools/testing/selftests/bpf/prog_tests/bpf_iter.c static void test_task_common(struct bpf_iter_attach_opts *opts, int num_unknown, · line 282

Quoted 8 lines
tools/testing/selftests/bpf/prog_tests/bpf_iter.c +8 -0
+}
+
+static void test_task_tid(void)
+{
+       pthread_t thread_id;
+       void *ret;
+
+       // Create a new thread so pid and tid aren't the same

C++ comment

tools/testing/selftests/bpf/prog_tests/bpf_iter.c static void test_task_common(struct bpf_iter_attach_opts *opts, int num_unknown, · line 285

Quoted 3 lines
tools/testing/selftests/bpf/prog_tests/bpf_iter.c +3 -0
+       ASSERT_OK(pthread_create(&thread_id, NULL, &run_test_task_tid, NULL),
+                 "pthread_create");
+       ASSERT_FALSE(pthread_join(thread_id, &ret) || ret != NULL,

it's best to avoid combining two check in single ASSERT_*(), so

ASSERT_OK(pthread_join(...), ...);
ASSERT_NULL(ret, ...);

is way easier to follow and debug, if something breaks

But also, why do we check ret? Do we ever return non-NULL?

pw-bot: cr

Quoted 7 lines
tools/testing/selftests/bpf/prog_tests/bpf_iter.c +1 -1
+                    "pthread_join");
 }
 
 static void test_task_pid(void)
--

2.43.5

#3 lore raw
In reply to Jordan Rome
Quoted 7 lines · Jordan Rome

In userspace, you can add a tid filter by setting the "task.tid" field for "bpf_iter_link_info". However, get_pid_task when called for the BPF_TASK_ITER_TID type should have been using PIDTYPE_PID (tid) instead of PIDTYPE_TGID (pid).

Quoted 6 lines
Fixes f0d74c4da1f0 ("bpf: Parameterize task iterators.")
---
 kernel/bpf/task_iter.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

This change is an important fix, so it has to target bpf tree, please use [PATCH bpf] subject prefix

Quoted 16 lines
kernel/bpf/task_iter.c +1 -2
index 02aa9db8d796..5af9e130e500 100644
@@ -99,7 +99,7 @@ static struct task_struct *task_seq_get_next(struct bpf_iter_seq_task_common *co
                rcu_read_lock();
                pid = find_pid_ns(common->pid, common->ns);
                if (pid) {
-                       task = get_pid_task(pid, PIDTYPE_TGID);
+                       task = get_pid_task(pid, PIDTYPE_PID);
                        *tid = common->pid;
                }
                rcu_read_unlock();
--

2.43.5