From: Feng Zhou <[email protected]>
Trace sched related functions, such as enqueue_task_fair, it is necessary to
specify a task instead of the current task which within a given cgroup.
Feng Zhou (2):
bpf: Add bpf_task_under_cgroup() kfunc
selftests/bpf: Add testcase for bpf_task_under_cgroup
Changelog:
v2->v3: Addressed comments from Alexei Starovoitov
- Modify the comment information of the function.
- Narrow down the testcase's hook point
Details in here:
https://lore.kernel.org/all/[email protected]/
v1->v2: Addressed comments from Alexei Starovoitov
- Add kfunc instead.
Details in here:
https://lore.kernel.org/all/[email protected]/
kernel/bpf/helpers.c | 20 ++++++++
tools/testing/selftests/bpf/DENYLIST.s390x | 1 +
.../bpf/prog_tests/task_under_cgroup.c | 47 +++++++++++++++++++
.../selftests/bpf/progs/cgrp_kfunc_common.h | 1 +
.../bpf/progs/test_task_under_cgroup.c | 37 +++++++++++++++
5 files changed, 106 insertions(+)
create mode 100644 tools/testing/selftests/bpf/prog_tests/task_under_cgroup.c
create mode 100644 tools/testing/selftests/bpf/progs/test_task_under_cgroup.c
--
2.20.1
From: Feng Zhou <[email protected]>
Add a kfunc that's similar to the bpf_current_task_under_cgroup.
The difference is that it is a designated task.
When hook sched related functions, sometimes it is necessary to
specify a task instead of the current task.
Signed-off-by: Feng Zhou <[email protected]>
---
kernel/bpf/helpers.c | 20 ++++++++++++++++++++
1 file changed, 20 insertions(+)
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index bb6b4637ebf2..453cbd312366 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2149,6 +2149,25 @@ __bpf_kfunc struct cgroup *bpf_cgroup_from_id(u64 cgid)
return NULL;
return cgrp;
}
+
+/**
+ * bpf_task_under_cgroup - wrap task_under_cgroup_hierarchy() as a kfunc, test
+ * task's membership of cgroup ancestry.
+ * @task: the task to be tested
+ * @ancestor: possible ancestor of @task's cgroup
+ *
+ * Tests whether @task's default cgroup hierarchy is a descendant of @ancestor.
+ * It follows all the same rules as cgroup_is_descendant, and only applies
+ * to the default hierarchy.
+ */
+__bpf_kfunc long bpf_task_under_cgroup(struct task_struct *task,
+ struct cgroup *ancestor)
+{
+ if (unlikely(!ancestor || !task))
+ return -EINVAL;
+
+ return task_under_cgroup_hierarchy(task, ancestor);
+}
#endif /* CONFIG_CGROUPS */
/**
@@ -2400,6 +2419,7 @@ BTF_ID_FLAGS(func, bpf_cgroup_acquire, KF_ACQUIRE | KF_RCU | KF_RET_NULL)
BTF_ID_FLAGS(func, bpf_cgroup_release, KF_RELEASE)
BTF_ID_FLAGS(func, bpf_cgroup_ancestor, KF_ACQUIRE | KF_RCU | KF_RET_NULL)
BTF_ID_FLAGS(func, bpf_cgroup_from_id, KF_ACQUIRE | KF_RET_NULL)
+BTF_ID_FLAGS(func, bpf_task_under_cgroup, KF_RCU)
#endif
BTF_ID_FLAGS(func, bpf_task_from_pid, KF_ACQUIRE | KF_RET_NULL)
BTF_SET8_END(generic_btf_ids)
--
2.20.1
From: Feng Zhou <[email protected]>
test_progs:
Tests new kfunc bpf_task_under_cgroup().
The bpf program saves the pid which call the getpgid syscall within a
given cgroup to the remote_pid, which is convenient for the user-mode
program to verify the test correctness.
The user-mode program creates its own mount namespace, and mounts the
cgroupsv2 hierarchy in there, call the getpgid syscall, then check if
remote_pid and local_pid are equal.
Signed-off-by: Feng Zhou <[email protected]>
---
tools/testing/selftests/bpf/DENYLIST.s390x | 1 +
.../bpf/prog_tests/task_under_cgroup.c | 47 +++++++++++++++++++
.../selftests/bpf/progs/cgrp_kfunc_common.h | 1 +
.../bpf/progs/test_task_under_cgroup.c | 37 +++++++++++++++
4 files changed, 86 insertions(+)
create mode 100644 tools/testing/selftests/bpf/prog_tests/task_under_cgroup.c
create mode 100644 tools/testing/selftests/bpf/progs/test_task_under_cgroup.c
diff --git a/tools/testing/selftests/bpf/DENYLIST.s390x b/tools/testing/selftests/bpf/DENYLIST.s390x
index c7463f3ec3c0..5061d9e24c16 100644
--- a/tools/testing/selftests/bpf/DENYLIST.s390x
+++ b/tools/testing/selftests/bpf/DENYLIST.s390x
@@ -26,3 +26,4 @@ user_ringbuf # failed to find kernel BTF type ID of
verif_stats # trace_vprintk__open_and_load unexpected error: -9 (?)
xdp_bonding # failed to auto-attach program 'trace_on_entry': -524 (trampoline)
xdp_metadata # JIT does not support calling kernel function (kfunc)
+test_task_under_cgroup # JIT does not support calling kernel function (kfunc)
diff --git a/tools/testing/selftests/bpf/prog_tests/task_under_cgroup.c b/tools/testing/selftests/bpf/prog_tests/task_under_cgroup.c
new file mode 100644
index 000000000000..6d5709a8203d
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/task_under_cgroup.c
@@ -0,0 +1,47 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Bytedance */
+
+#include <sys/syscall.h>
+#include <test_progs.h>
+#include <cgroup_helpers.h>
+#include "test_task_under_cgroup.skel.h"
+
+#define FOO "/foo"
+
+void test_task_under_cgroup(void)
+{
+ struct test_task_under_cgroup *skel;
+ int ret, foo = -1;
+
+ foo = test__join_cgroup(FOO);
+ if (!ASSERT_OK(foo < 0, "cgroup_join_foo"))
+ return;
+
+ skel = test_task_under_cgroup__open();
+ if (!ASSERT_OK_PTR(skel, "test_task_under_cgroup__open"))
+ goto cleanup;
+
+ skel->rodata->local_pid = getpid();
+ skel->rodata->cgid = get_cgroup_id(FOO);
+
+ ret = test_task_under_cgroup__load(skel);
+ if (!ASSERT_OK(ret, "test_task_under_cgroup__load"))
+ goto cleanup;
+
+ ret = test_task_under_cgroup__attach(skel);
+ if (!ASSERT_OK(ret, "test_task_under_cgroup__attach"))
+ goto cleanup;
+
+ syscall(SYS_getpgid);
+
+ test_task_under_cgroup__detach(skel);
+
+ ASSERT_EQ(skel->bss->remote_pid, skel->rodata->local_pid,
+ "test task_under_cgroup");
+
+cleanup:
+ if (foo >= 0)
+ close(foo);
+
+ test_task_under_cgroup__destroy(skel);
+}
diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h b/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
index 22914a70db54..001c416b42bc 100644
--- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
+++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
@@ -26,6 +26,7 @@ struct cgroup *bpf_cgroup_ancestor(struct cgroup *cgrp, int level) __ksym;
struct cgroup *bpf_cgroup_from_id(u64 cgid) __ksym;
void bpf_rcu_read_lock(void) __ksym;
void bpf_rcu_read_unlock(void) __ksym;
+int bpf_task_under_cgroup(struct task_struct *task, struct cgroup *ancestor) __ksym;
static inline struct __cgrps_kfunc_map_value *cgrps_kfunc_map_value_lookup(struct cgroup *cgrp)
{
diff --git a/tools/testing/selftests/bpf/progs/test_task_under_cgroup.c b/tools/testing/selftests/bpf/progs/test_task_under_cgroup.c
new file mode 100644
index 000000000000..8f23a2933fde
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_task_under_cgroup.c
@@ -0,0 +1,37 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2023 Bytedance */
+
+#include <vmlinux.h>
+#include <bpf/bpf_tracing.h>
+#include <bpf/bpf_helpers.h>
+
+#include "bpf_misc.h"
+#include "cgrp_kfunc_common.h"
+
+const volatile int local_pid;
+const volatile long cgid;
+int remote_pid;
+
+SEC("fentry/" SYS_PREFIX "sys_getpgid")
+int sys_getpgid(void *ctx)
+{
+ struct cgroup *cgrp;
+
+ if (local_pid != (bpf_get_current_pid_tgid() >> 32))
+ return 0;
+
+ cgrp = bpf_cgroup_from_id(cgid);
+ if (!cgrp)
+ return 0;
+
+ if (!bpf_task_under_cgroup(bpf_get_current_task_btf(), cgrp))
+ goto out;
+
+ remote_pid = local_pid;
+
+out:
+ bpf_cgroup_release(cgrp);
+ return 0;
+}
+
+char _license[] SEC("license") = "GPL";
--
2.20.1
On 4/26/23 7:30 PM, Feng zhou wrote:
> From: Feng Zhou <[email protected]>
>
> test_progs:
> Tests new kfunc bpf_task_under_cgroup().
>
> The bpf program saves the pid which call the getpgid syscall within a
> given cgroup to the remote_pid, which is convenient for the user-mode
> program to verify the test correctness.
>
> The user-mode program creates its own mount namespace, and mounts the
> cgroupsv2 hierarchy in there, call the getpgid syscall, then check if
> remote_pid and local_pid are equal.
>
> Signed-off-by: Feng Zhou <[email protected]>
> ---
> tools/testing/selftests/bpf/DENYLIST.s390x | 1 +
> .../bpf/prog_tests/task_under_cgroup.c | 47 +++++++++++++++++++
> .../selftests/bpf/progs/cgrp_kfunc_common.h | 1 +
> .../bpf/progs/test_task_under_cgroup.c | 37 +++++++++++++++
> 4 files changed, 86 insertions(+)
> create mode 100644 tools/testing/selftests/bpf/prog_tests/task_under_cgroup.c
> create mode 100644 tools/testing/selftests/bpf/progs/test_task_under_cgroup.c
>
> diff --git a/tools/testing/selftests/bpf/DENYLIST.s390x b/tools/testing/selftests/bpf/DENYLIST.s390x
> index c7463f3ec3c0..5061d9e24c16 100644
> --- a/tools/testing/selftests/bpf/DENYLIST.s390x
> +++ b/tools/testing/selftests/bpf/DENYLIST.s390x
> @@ -26,3 +26,4 @@ user_ringbuf # failed to find kernel BTF type ID of
> verif_stats # trace_vprintk__open_and_load unexpected error: -9 (?)
> xdp_bonding # failed to auto-attach program 'trace_on_entry': -524 (trampoline)
> xdp_metadata # JIT does not support calling kernel function (kfunc)
> +test_task_under_cgroup # JIT does not support calling kernel function (kfunc)
> diff --git a/tools/testing/selftests/bpf/prog_tests/task_under_cgroup.c b/tools/testing/selftests/bpf/prog_tests/task_under_cgroup.c
> new file mode 100644
> index 000000000000..6d5709a8203d
> --- /dev/null
> +++ b/tools/testing/selftests/bpf/prog_tests/task_under_cgroup.c
> @@ -0,0 +1,47 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/* Copyright (c) 2023 Bytedance */
> +
> +#include <sys/syscall.h>
> +#include <test_progs.h>
> +#include <cgroup_helpers.h>
> +#include "test_task_under_cgroup.skel.h"
> +
> +#define FOO "/foo"
> +
> +void test_task_under_cgroup(void)
> +{
> + struct test_task_under_cgroup *skel;
> + int ret, foo = -1;
> +
> + foo = test__join_cgroup(FOO);
> + if (!ASSERT_OK(foo < 0, "cgroup_join_foo"))
> + return;
> +
> + skel = test_task_under_cgroup__open();
> + if (!ASSERT_OK_PTR(skel, "test_task_under_cgroup__open"))
> + goto cleanup;
> +
> + skel->rodata->local_pid = getpid();
> + skel->rodata->cgid = get_cgroup_id(FOO);
> +
> + ret = test_task_under_cgroup__load(skel);
> + if (!ASSERT_OK(ret, "test_task_under_cgroup__load"))
> + goto cleanup;
> +
> + ret = test_task_under_cgroup__attach(skel);
> + if (!ASSERT_OK(ret, "test_task_under_cgroup__attach"))
> + goto cleanup;
> +
> + syscall(SYS_getpgid);
> +
> + test_task_under_cgroup__detach(skel);
> +
> + ASSERT_EQ(skel->bss->remote_pid, skel->rodata->local_pid,
> + "test task_under_cgroup");
> +
> +cleanup:
> + if (foo >= 0)
> + close(foo);
> +
> + test_task_under_cgroup__destroy(skel);
> +}
> diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h b/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
> index 22914a70db54..001c416b42bc 100644
> --- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
> +++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_common.h
> @@ -26,6 +26,7 @@ struct cgroup *bpf_cgroup_ancestor(struct cgroup *cgrp, int level) __ksym;
> struct cgroup *bpf_cgroup_from_id(u64 cgid) __ksym;
> void bpf_rcu_read_lock(void) __ksym;
> void bpf_rcu_read_unlock(void) __ksym;
> +int bpf_task_under_cgroup(struct task_struct *task, struct cgroup *ancestor) __ksym;
return type 'long'?
>
> static inline struct __cgrps_kfunc_map_value *cgrps_kfunc_map_value_lookup(struct cgroup *cgrp)
> {
> diff --git a/tools/testing/selftests/bpf/progs/test_task_under_cgroup.c b/tools/testing/selftests/bpf/progs/test_task_under_cgroup.c
> new file mode 100644
> index 000000000000..8f23a2933fde
> --- /dev/null
> +++ b/tools/testing/selftests/bpf/progs/test_task_under_cgroup.c
> @@ -0,0 +1,37 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/* Copyright (c) 2023 Bytedance */
> +
> +#include <vmlinux.h>
> +#include <bpf/bpf_tracing.h>
> +#include <bpf/bpf_helpers.h>
> +
> +#include "bpf_misc.h"
> +#include "cgrp_kfunc_common.h"
> +
> +const volatile int local_pid;
> +const volatile long cgid;
> +int remote_pid;
> +
> +SEC("fentry/" SYS_PREFIX "sys_getpgid")
> +int sys_getpgid(void *ctx)
> +{
> + struct cgroup *cgrp;
> +
> + if (local_pid != (bpf_get_current_pid_tgid() >> 32))
> + return 0;
> +
> + cgrp = bpf_cgroup_from_id(cgid);
> + if (!cgrp)
> + return 0;
> +
> + if (!bpf_task_under_cgroup(bpf_get_current_task_btf(), cgrp))
> + goto out;
The whole point of using bpf_task_under_cgroup() is to test a
non-current task is under a particular cgroup.
Tracing kernel function enqueue_task_fair() is an option, but it
may be inlined with certain compilers as enqueue_task_fair() is
a static function. Also, the 'task' argument in enqueue_task_fair()
is not trusted, could be null ptr and not rcu safe as well.
We could promote 'task' in enqueue_task_fair() by:
- checking NULL ptr, if not
- increase reference count by 1 if current reference count is not 0
(using bpf_task_acquire kfunc)
- if reference count can be increased, it becomes trusted, and
then bpf_task_under_cgroup() can be used.
Could you try this approach or even better if there is another
non-static function also having a task (not the current task)?
> +
> + remote_pid = local_pid;
> +
> +out:
> + bpf_cgroup_release(cgrp);
> + return 0;
> +}
> +
> +char _license[] SEC("license") = "GPL";