Skip to contentSkip to conversation

selftests/bpf: Fix cgroup_xattr/read_cgroupfs_xattr

PATCH Applied Song Liu 5 participants 13 messages Jun 27, 2025 – Jun 30, 2025

Subsystems (3)
lore
Share & tools
lore

Song Liu

author reviewer
#0 lore raw

PATCH v2 1/1 selftests/bpf: Fix cgroup_xattr/read_cgroupfs_xattr

cgroup_xattr/read_cgroupfs_xattr has two issues:

1. cgroup_xattr/read_cgroupfs_xattr messes up lo without creating a netns first. This causes issue with other tests.

Fix this by using a different hook (lsm.s/file_open) and not messing with lo.

2. cgroup_xattr/read_cgroupfs_xattr sets up cgroups without proper mount namespaces.

Fix this by using the existing cgroup helpers. A new helper set_cgroup_xattr() is added to set xattr on cgroup files.

---
Changes v1 => v2:
1. Add the second fix above.

v1: https://lore.kernel.org/bpf/[email protected]/

---
 tools/testing/selftests/bpf/cgroup_helpers.c  |  21 ++++
 tools/testing/selftests/bpf/cgroup_helpers.h  |   4 +
 .../selftests/bpf/prog_tests/cgroup_xattr.c   | 117 ++++--------------
 .../selftests/bpf/progs/read_cgroupfs_xattr.c |   4 +-
 4 files changed, 49 insertions(+), 97 deletions(-)
tools/testing/selftests/bpf/cgroup_helpers.c +21 -0
index e4535451322e..15f626014872 100644
@@ -4,6 +4,7 @@
 #include <sys/mount.h>
 #include <sys/stat.h>
 #include <sys/types.h>
+#include <sys/xattr.h>
 #include <linux/limits.h>
 #include <stdio.h>
 #include <stdlib.h>
@@ -318,6 +319,26 @@ int join_parent_cgroup(const char *relative_path)
 	return join_cgroup_from_top(cgroup_path);
 }
 
+/**
+ * set_cgroup_xattr() - Set xattr on a cgroup dir
+ * @relative_path: The cgroup path, relative to the workdir, to set xattr
+ * @name: xattr name
+ * @value: xattr value
+ *
+ * This function set xattr on cgroup dir.
+ *
+ * On success, it returns 0, otherwise on failure it returns -1.
+ */
+int set_cgroup_xattr(const char *relative_path,
+		     const char *name,
+		     const char *value)
+{
+	char cgroup_path[PATH_MAX + 1];
+
+	format_cgroup_path(cgroup_path, relative_path);
+	return setxattr(cgroup_path, name, value, strlen(value) + 1, 0);
+}
+
 /**
  * __cleanup_cgroup_environment() - Delete temporary cgroups
  *
tools/testing/selftests/bpf/cgroup_helpers.h +4 -0
index 502845160d88..182e1ac36c95 100644
@@ -26,6 +26,10 @@ int join_cgroup(const char *relative_path);
 int join_root_cgroup(void);
 int join_parent_cgroup(const char *relative_path);
 
+int set_cgroup_xattr(const char *relative_path,
+		     const char *name,
+		     const char *value);
+
 int setup_cgroup_environment(void);
 void cleanup_cgroup_environment(void);
 
tools/testing/selftests/bpf/prog_tests/cgroup_xattr.c +22 -95

155 lines not on this page · raw message

tools/testing/selftests/bpf/progs/read_cgroupfs_xattr.c +2 -2
index 855f85fc5522..405adbe5e8b0 100644
@@ -17,8 +17,8 @@ static const char expected_value_b[] = "bpf_selftest_value_b";
 bool found_value_a;
 bool found_value_b;
 
-SEC("lsm.s/socket_connect")
-int BPF_PROG(test_socket_connect)
+SEC("lsm.s/file_open")
+int BPF_PROG(test_file_open)
 {
 	u64 cgrp_id = bpf_get_current_cgroup_id();
 	struct cgroup_subsys_state *css, *tmp;
Signature · 1 line
2.47.1
maintainer Tested-by on v2 1/1
#1 lore raw
Quoted 19 lines · Song Liu

cgroup_xattr/read_cgroupfs_xattr has two issues:

1. cgroup_xattr/read_cgroupfs_xattr messes up lo without creating a netns first. This causes issue with other tests.

Fix this by using a different hook (lsm.s/file_open) and not messing with lo.

2. cgroup_xattr/read_cgroupfs_xattr sets up cgroups without proper mount namespaces.

Fix this by using the existing cgroup helpers. A new helper set_cgroup_xattr() is added to set xattr on cgroup files.

Quoted 6 lines

---
Changes v1 => v2:
1. Add the second fix above.

v1: https://lore.kernel.org/bpf/[email protected]/
---

This fixes ./test_progs -j -t cgroup for me. Changes seem to reflect logic of the original test.

[...]

reviewer Tested-by on v2 1/1
#2 lore raw
In reply to Song Liu
Quoted 23 lines · Song Liu

cgroup_xattr/read_cgroupfs_xattr has two issues:

1. cgroup_xattr/read_cgroupfs_xattr messes up lo without creating a netns first. This causes issue with other tests.

Fix this by using a different hook (lsm.s/file_open) and not messing with lo.

2. cgroup_xattr/read_cgroupfs_xattr sets up cgroups without proper mount namespaces.

Fix this by using the existing cgroup helpers. A new helper set_cgroup_xattr() is added to set xattr on cgroup files.

---
Changes v1 => v2:
1. Add the second fix above.

Quoted 7 lines

v1: https://lore.kernel.org/bpf/[email protected]/

---
  tools/testing/selftests/bpf/cgroup_helpers.c  |  21 ++++
  tools/testing/selftests/bpf/cgroup_helpers.h  |   4 +
  .../selftests/bpf/prog_tests/cgroup_xattr.c   | 117 ++++--------------
  .../selftests/bpf/progs/read_cgroupfs_xattr.c |   4 +-
  4 files changed, 49 insertions(+), 97 deletions(-)

Hi Song.

I tried this patch on BPF CI, and it appears it fixes the hanging failure we've been seeing today on bpf-next and netdev. I am going to add it to ci/diffs.

https://github.com/kernel-patches/vmtest/actions/runs/15935456082

Thanks!

Quoted 230 lines
tools/testing/selftests/bpf/cgroup_helpers.c +21 -0
index e4535451322e..15f626014872 100644
@@ -4,6 +4,7 @@
  #include <sys/mount.h>
  #include <sys/stat.h>
  #include <sys/types.h>
+#include <sys/xattr.h>
  #include <linux/limits.h>
  #include <stdio.h>
  #include <stdlib.h>
@@ -318,6 +319,26 @@ int join_parent_cgroup(const char *relative_path)
  	return join_cgroup_from_top(cgroup_path);
  }
 
+/**
+ * set_cgroup_xattr() - Set xattr on a cgroup dir
+ * @relative_path: The cgroup path, relative to the workdir, to set xattr
+ * @name: xattr name
+ * @value: xattr value
+ *
+ * This function set xattr on cgroup dir.
+ *
+ * On success, it returns 0, otherwise on failure it returns -1.
+ */
+int set_cgroup_xattr(const char *relative_path,
+		     const char *name,
+		     const char *value)
+{
+	char cgroup_path[PATH_MAX + 1];
+
+	format_cgroup_path(cgroup_path, relative_path);
+	return setxattr(cgroup_path, name, value, strlen(value) + 1, 0);
+}
+
  /**
   * __cleanup_cgroup_environment() - Delete temporary cgroups
   *
tools/testing/selftests/bpf/cgroup_helpers.h +4 -0
index 502845160d88..182e1ac36c95 100644
@@ -26,6 +26,10 @@ int join_cgroup(const char *relative_path);
  int join_root_cgroup(void);
  int join_parent_cgroup(const char *relative_path);
 
+int set_cgroup_xattr(const char *relative_path,
+		     const char *name,
+		     const char *value);
+
  int setup_cgroup_environment(void);
  void cleanup_cgroup_environment(void);
 
tools/testing/selftests/bpf/prog_tests/cgroup_xattr.c +22 -95
index 87978a0f7eb7..e0dd966e4a3e 100644
@@ -7,133 +7,60 @@
  #include <string.h>
  #include <unistd.h>
  #include <sys/socket.h>
-#include <sys/xattr.h>
-
  #include <test_progs.h>
+#include "cgroup_helpers.h"
 
  #include "read_cgroupfs_xattr.skel.h"
  #include "cgroup_read_xattr.skel.h"
 
-#define CGROUP_FS_ROOT "/sys/fs/cgroup/"
-#define CGROUP_FS_PARENT CGROUP_FS_ROOT "foo/"
+#define CGROUP_FS_PARENT "foo/"
  #define CGROUP_FS_CHILD CGROUP_FS_PARENT "bar/"
-
-static int move_pid_to_cgroup(const char *cgroup_folder, pid_t pid)
-{
-	char filename[128];
-	char pid_str[64];
-	int procs_fd;
-	int ret;
-
-	snprintf(filename, sizeof(filename), "%scgroup.procs", cgroup_folder);
-	snprintf(pid_str, sizeof(pid_str), "%d", pid);
-
-	procs_fd = open(filename, O_WRONLY | O_APPEND);
-	if (!ASSERT_OK_FD(procs_fd, "open"))
-		return -1;
-
-	ret = write(procs_fd, pid_str, strlen(pid_str));
-	close(procs_fd);
-	if (!ASSERT_GT(ret, 0, "write cgroup.procs"))
-		return -1;
-	return 0;
-}
-
-static void reset_cgroups_and_lo(void)
-{
-	rmdir(CGROUP_FS_CHILD);
-	rmdir(CGROUP_FS_PARENT);
-	system("ip addr del 1.1.1.1/32 dev lo");
-	system("ip link set dev lo down");
-}
+#define TMP_FILE "/tmp/selftests_cgroup_xattr"
 
  static const char xattr_value_a[] = "bpf_selftest_value_a";
  static const char xattr_value_b[] = "bpf_selftest_value_b";
  static const char xattr_name[] = "user.bpf_test";
 
-static int setup_cgroups_and_lo(void)
-{
-	int err;
-
-	err = mkdir(CGROUP_FS_PARENT, 0755);
-	if (!ASSERT_OK(err, "mkdir 1"))
-		goto error;
-	err = mkdir(CGROUP_FS_CHILD, 0755);
-	if (!ASSERT_OK(err, "mkdir 2"))
-		goto error;
-
-	err = setxattr(CGROUP_FS_PARENT, xattr_name, xattr_value_a,
-		       strlen(xattr_value_a) + 1, 0);
-	if (!ASSERT_OK(err, "setxattr 1"))
-		goto error;
-
-	err = setxattr(CGROUP_FS_CHILD, xattr_name, xattr_value_b,
-		       strlen(xattr_value_b) + 1, 0);
-	if (!ASSERT_OK(err, "setxattr 2"))
-		goto error;
-
-	err = system("ip link set dev lo up");
-	if (!ASSERT_OK(err, "lo up"))
-		goto error;
-
-	err = system("ip addr add 1.1.1.1 dev lo");
-	if (!ASSERT_OK(err, "lo addr v4"))
-		goto error;
-
-	err = write_sysctl("/proc/sys/net/ipv4/ping_group_range", "0 0");
-	if (!ASSERT_OK(err, "write_sysctl"))
-		goto error;
-
-	return 0;
-error:
-	reset_cgroups_and_lo();
-	return err;
-}
-
  static void test_read_cgroup_xattr(void)
  {
-	struct sockaddr_in sa4 = {
-		.sin_family = AF_INET,
-		.sin_addr.s_addr = htonl(INADDR_LOOPBACK),
-	};
+	int tmp_fd, parent_cgroup_fd = -1, child_cgroup_fd = -1;
  	struct read_cgroupfs_xattr *skel = NULL;
-	pid_t pid = gettid();
-	int sock_fd = -1;
-	int connect_fd = -1;
 
-	if (!ASSERT_OK(setup_cgroups_and_lo(), "setup_cgroups_and_lo"))
+	parent_cgroup_fd = test__join_cgroup(CGROUP_FS_PARENT);
+	if (!ASSERT_OK_FD(parent_cgroup_fd, "create parent cgroup"))
  		return;
-	if (!ASSERT_OK(move_pid_to_cgroup(CGROUP_FS_CHILD, pid),
-		       "move_pid_to_cgroup"))
+	if (!ASSERT_OK(set_cgroup_xattr(CGROUP_FS_PARENT, xattr_name, xattr_value_a),
+		       "set parent xattr"))
+		goto out;
+
+	child_cgroup_fd = test__join_cgroup(CGROUP_FS_CHILD);
+	if (!ASSERT_OK_FD(child_cgroup_fd, "create child cgroup"))
+		goto out;
+	if (!ASSERT_OK(set_cgroup_xattr(CGROUP_FS_CHILD, xattr_name, xattr_value_b),
+		       "set child xattr"))
  		goto out;
 
  	skel = read_cgroupfs_xattr__open_and_load();
  	if (!ASSERT_OK_PTR(skel, "read_cgroupfs_xattr__open_and_load"))
  		goto out;
 
-	skel->bss->target_pid = pid;
+	skel->bss->target_pid = gettid();
 
  	if (!ASSERT_OK(read_cgroupfs_xattr__attach(skel), "read_cgroupfs_xattr__attach"))
  		goto out;
 
-	sock_fd = socket(AF_INET, SOCK_DGRAM, IPPROTO_ICMP);
-	if (!ASSERT_OK_FD(sock_fd, "sock create"))
-		goto out;
-
-	connect_fd = connect(sock_fd, &sa4, sizeof(sa4));
-	if (!ASSERT_OK_FD(connect_fd, "connect 1"))
-		goto out;
-	close(connect_fd);
+	tmp_fd = open(TMP_FILE, O_RDONLY | O_CREAT);
+	ASSERT_OK_FD(tmp_fd, "open tmp file");
+	close(tmp_fd);
 
  	ASSERT_TRUE(skel->bss->found_value_a, "found_value_a");
  	ASSERT_TRUE(skel->bss->found_value_b, "found_value_b");
 
  out:
-	close(connect_fd);
-	close(sock_fd);
+	close(child_cgroup_fd);
+	close(parent_cgroup_fd);
  	read_cgroupfs_xattr__destroy(skel);
-	move_pid_to_cgroup(CGROUP_FS_ROOT, pid);
-	reset_cgroups_and_lo();
+	unlink(TMP_FILE);
  }
 
  void test_cgroup_xattr(void)
tools/testing/selftests/bpf/progs/read_cgroupfs_xattr.c +2 -2
index 855f85fc5522..405adbe5e8b0 100644
@@ -17,8 +17,8 @@ static const char expected_value_b[] = "bpf_selftest_value_b";
  bool found_value_a;
  bool found_value_b;
 
-SEC("lsm.s/socket_connect")
-int BPF_PROG(test_socket_connect)
+SEC("lsm.s/file_open")
+int BPF_PROG(test_file_open)
  {
  	u64 cgrp_id = bpf_get_current_cgroup_id();
  	struct cgroup_subsys_state *css, *tmp;
#3 lore raw
Quoted 33 lines · Ihor Solodrai
Quoted 23 lines · Song Liu

cgroup_xattr/read_cgroupfs_xattr has two issues:

1. cgroup_xattr/read_cgroupfs_xattr messes up lo without creating a netns first. This causes issue with other tests.

Fix this by using a different hook (lsm.s/file_open) and not messing with lo.

2. cgroup_xattr/read_cgroupfs_xattr sets up cgroups without proper mount namespaces.

Fix this by using the existing cgroup helpers. A new helper set_cgroup_xattr() is added to set xattr on cgroup files.

---
Changes v1 => v2:
1. Add the second fix above.

Quoted 7 lines

v1: https://lore.kernel.org/bpf/[email protected]/

---
  tools/testing/selftests/bpf/cgroup_helpers.c  |  21 ++++
  tools/testing/selftests/bpf/cgroup_helpers.h  |   4 +
  .../selftests/bpf/prog_tests/cgroup_xattr.c   | 117 ++++--------------
  .../selftests/bpf/progs/read_cgroupfs_xattr.c |   4 +-
  4 files changed, 49 insertions(+), 97 deletions(-)
Quoted 5 lines

Hi Song.

I tried this patch on BPF CI, and it appears it fixes the hanging failure we've been seeing today on bpf-next and netdev. I am going to add it to ci/diffs.

Applied to bpf-next already.

reviewer
#4 lore raw
Quoted 34 lines · Alexei Starovoitov
Quoted 33 lines · Ihor Solodrai
Quoted 30 lines · Song Liu

cgroup_xattr/read_cgroupfs_xattr has two issues:

1. cgroup_xattr/read_cgroupfs_xattr messes up lo without creating a netns first. This causes issue with other tests.

Fix this by using a different hook (lsm.s/file_open) and not messing with lo.

2. cgroup_xattr/read_cgroupfs_xattr sets up cgroups without proper mount namespaces.

Fix this by using the existing cgroup helpers. A new helper set_cgroup_xattr() is added to set xattr on cgroup files.

---
Changes v1 => v2:
1. Add the second fix above.

v1: https://lore.kernel.org/bpf/[email protected]/

---
   tools/testing/selftests/bpf/cgroup_helpers.c  |  21 ++++
   tools/testing/selftests/bpf/cgroup_helpers.h  |   4 +
   .../selftests/bpf/prog_tests/cgroup_xattr.c   | 117 ++++--------------
   .../selftests/bpf/progs/read_cgroupfs_xattr.c |   4 +-
   4 files changed, 49 insertions(+), 97 deletions(-)
Quoted 7 lines
Quoted 5 lines

Hi Song.

I tried this patch on BPF CI, and it appears it fixes the hanging failure we've been seeing today on bpf-next and netdev. I am going to add it to ci/diffs.

Applied to bpf-next already.

CI patches apply to all base branches. My understanding is, it's needed at least for netdev too.

#5 lore raw
Quoted 38 lines · Ihor Solodrai
Quoted 36 lines · Alexei Starovoitov
Quoted 35 lines · Ihor Solodrai
Quoted 23 lines · Song Liu

cgroup_xattr/read_cgroupfs_xattr has two issues:

1. cgroup_xattr/read_cgroupfs_xattr messes up lo without creating a netns first. This causes issue with other tests.

Fix this by using a different hook (lsm.s/file_open) and not messing with lo.

2. cgroup_xattr/read_cgroupfs_xattr sets up cgroups without proper mount namespaces.

Fix this by using the existing cgroup helpers. A new helper set_cgroup_xattr() is added to set xattr on cgroup files.

---
Changes v1 => v2:
1. Add the second fix above.

Quoted 7 lines

v1: https://lore.kernel.org/bpf/[email protected]/

---
   tools/testing/selftests/bpf/cgroup_helpers.c  |  21 ++++
   tools/testing/selftests/bpf/cgroup_helpers.h  |   4 +
   .../selftests/bpf/prog_tests/cgroup_xattr.c   | 117 ++++--------------
   .../selftests/bpf/progs/read_cgroupfs_xattr.c |   4 +-
   4 files changed, 49 insertions(+), 97 deletions(-)

Hi Song.

Quoted 8 lines
Quoted 5 lines
Quoted 3 lines

I tried this patch on BPF CI, and it appears it fixes the hanging failure we've been seeing today on bpf-next and netdev. I am going to add it to ci/diffs.

Applied to bpf-next already.

CI patches apply to all base branches. My understanding is, it's needed at least for netdev too.

How is that possible?

The offending commit is only in /master and in /for-next branches, while /for-next is there for linux-next only.

reviewer
#6 lore raw
Quoted 45 lines · Alexei Starovoitov
Quoted 44 lines · Ihor Solodrai
Quoted 41 lines · Alexei Starovoitov
Quoted 33 lines · Ihor Solodrai
Quoted 23 lines · Song Liu

cgroup_xattr/read_cgroupfs_xattr has two issues:

1. cgroup_xattr/read_cgroupfs_xattr messes up lo without creating a netns first. This causes issue with other tests.

Fix this by using a different hook (lsm.s/file_open) and not messing with lo.

2. cgroup_xattr/read_cgroupfs_xattr sets up cgroups without proper mount namespaces.

Fix this by using the existing cgroup helpers. A new helper set_cgroup_xattr() is added to set xattr on cgroup files.

---
Changes v1 => v2:
1. Add the second fix above.

Quoted 7 lines

v1: https://lore.kernel.org/bpf/[email protected]/

---
    tools/testing/selftests/bpf/cgroup_helpers.c  |  21 ++++
    tools/testing/selftests/bpf/cgroup_helpers.h  |   4 +
    .../selftests/bpf/prog_tests/cgroup_xattr.c   | 117 ++++--------------
    .../selftests/bpf/progs/read_cgroupfs_xattr.c |   4 +-
    4 files changed, 49 insertions(+), 97 deletions(-)
Quoted 5 lines

Hi Song.

I tried this patch on BPF CI, and it appears it fixes the hanging failure we've been seeing today on bpf-next and netdev. I am going to add it to ci/diffs.

Applied to bpf-next already.

Quoted 7 lines
Quoted 2 lines

CI patches apply to all base branches. My understanding is, it's needed at least for netdev too.

How is that possible?

The offending commit is only in /master and in /for-next branches, while /for-next is there for linux-next only.

I am not sure.

I compared CI logs between bpf-next and netdev runs that both were cancelled due to 100min job timeout, and they are very similar.

netdev: https://github.com/kernel-patches/bpf/actions/runs/15932863319/job/44946276955
bpf-next: https://github.com/kernel-patches/bpf/actions/runs/15934258609/job/44950981852

So the root cause is likely the same.

And most recent netdev (with this patch applied) is green:
https://github.com/kernel-patches/bpf/actions/runs/15936292169

CC Jakub

reviewer
#7 lore raw
Quoted 63 lines · Ihor Solodrai
Quoted 46 lines · Alexei Starovoitov
Quoted 45 lines · Ihor Solodrai
Quoted 42 lines · Alexei Starovoitov
Quoted 34 lines · Ihor Solodrai
Quoted 23 lines · Song Liu

cgroup_xattr/read_cgroupfs_xattr has two issues:

1. cgroup_xattr/read_cgroupfs_xattr messes up lo without creating a netns first. This causes issue with other tests.

      Fix this by using a different hook (lsm.s/file_open) and not messing with lo.

2. cgroup_xattr/read_cgroupfs_xattr sets up cgroups without proper mount namespaces.

      Fix this by using the existing cgroup helpers. A new helper set_cgroup_xattr() is added to set xattr on cgroup files.

Fixes f4fba2d6d282 ("selftests/bpf: Add tests for bpf_cgroup_read_xattr")
Closes https://lore.kernel.org/bpf/ CAADnVQ+iqMi2HEj_iH7hsx+XJAsqaMWqSDe4tzcGAnehFWA9Sw@mail.gmail.com/

---
Changes v1 => v2:
1. Add the second fix above.

Quoted 8 lines

v1: https://lore.kernel.org/bpf/20250627165831.2979022-1- [email protected]/
---
    tools/testing/selftests/bpf/cgroup_helpers.c  |  21 ++++ tools/testing/selftests/bpf/cgroup_helpers.h  |   4 + .../selftests/bpf/prog_tests/cgroup_xattr.c   | 117 +++

+--------------
    .../selftests/bpf/progs/read_cgroupfs_xattr.c |   4 +-
    4 files changed, 49 insertions(+), 97 deletions(-)
Quoted 5 lines

Hi Song.

I tried this patch on BPF CI, and it appears it fixes the hanging failure we've been seeing today on bpf-next and netdev. I am going to add it to ci/diffs.

Applied to bpf-next already.

Quoted 7 lines
Quoted 2 lines

CI patches apply to all base branches. My understanding is, it's needed at least for netdev too.

How is that possible?

The offending commit is only in /master and in /for-next branches, while /for-next is there for linux-next only.

I am not sure.

I compared CI logs between bpf-next and netdev runs that both were cancelled due to 100min job timeout, and they are very similar.

netdev: https://github.com/kernel-patches/bpf/actions/runs/15932863319/ job/44946276955
bpf-next: https://github.com/kernel-patches/bpf/actions/ runs/15934258609/job/44950981852

Quoted 8 lines

So the root cause is likely the same.

And most recent netdev (with this patch applied) is green:
https://github.com/kernel-patches/bpf/actions/runs/15936292169

CC Jakub

Apparently offending patches were merged by Christian Brauner:

https://github.com/linux-netdev/testing-bpf-ci/commit/13b0cce9e294f8ddf228b9db3e01d76ac29872f2

#8 lore raw
Quoted 79 lines · Ihor Solodrai
Quoted 77 lines · Ihor Solodrai
Quoted 53 lines · Alexei Starovoitov
Quoted 51 lines · Ihor Solodrai
Quoted 48 lines · Alexei Starovoitov
Quoted 39 lines · Ihor Solodrai
Quoted 29 lines · Song Liu

cgroup_xattr/read_cgroupfs_xattr has two issues:

1. cgroup_xattr/read_cgroupfs_xattr messes up lo without creating a netns
first. This causes issue with other tests.

Fix this by using a different hook (lsm.s/file_open) and not
messing
with lo.

2. cgroup_xattr/read_cgroupfs_xattr sets up cgroups without proper mount namespaces.

Fix this by using the existing cgroup helpers. A new helper set_cgroup_xattr() is added to set xattr on cgroup files.

Fixes f4fba2d6d282 ("selftests/bpf: Add tests for

bpf_cgroup_read_xattr")

CAADnVQ+iqMi2HEj_iH7hsx+XJAsqaMWqSDe4tzcGAnehFWA9Sw@mail.gmail.com/

---
Changes v1 => v2:
1. Add the second fix above.

v1: https://lore.kernel.org/bpf/20250627165831.2979022-1-
[email protected]/

Quoted 7 lines
---
    tools/testing/selftests/bpf/cgroup_helpers.c  |  21 ++++
    tools/testing/selftests/bpf/cgroup_helpers.h  |   4 +
    .../selftests/bpf/prog_tests/cgroup_xattr.c   | 117 +++
+--------------
    .../selftests/bpf/progs/read_cgroupfs_xattr.c |   4 +-
    4 files changed, 49 insertions(+), 97 deletions(-)
Quoted 5 lines

Hi Song.

I tried this patch on BPF CI, and it appears it fixes the hanging failure we've been seeing today on bpf-next and netdev. I am going to add it to ci/diffs.

Applied to bpf-next already.

Quoted 7 lines
Quoted 2 lines

CI patches apply to all base branches. My understanding is, it's needed at least for netdev too.

How is that possible?

The offending commit is only in /master and in /for-next branches, while /for-next is there for linux-next only.

I am not sure.

I compared CI logs between bpf-next and netdev runs that both were cancelled due to 100min job timeout, and they are very similar.

netdev: https://github.com/kernel-patches/bpf/actions/runs/15932863319/
job/44946276955
bpf-next: https://github.com/kernel-patches/bpf/actions/
runs/15934258609/job/44950981852

So the root cause is likely the same.

And most recent netdev (with this patch applied) is green:
https://github.com/kernel-patches/bpf/actions/runs/15936292169

Quoted 7 lines
Quoted 3 lines

CC Jakub

Apparently offending patches were merged by Christian Brauner:

https://github.com/linux-netdev/testing-bpf-ci/commit/13b0cce9e294f8ddf228b9db3e01d76ac29872f2

Yes. We're aware.

re: why netdev CI was failing.
I bet it's bpf-next + net-next auto-merge one. It's a special ephemeral branch that netdev CI is doing. It should be green already without CI extra patch, because bpf-next is green.

reviewer
#9 lore raw
In reply to Alexei Starovoitov
Quoted 45 lines · Alexei Starovoitov
Quoted 44 lines · Ihor Solodrai
Quoted 41 lines · Alexei Starovoitov
Quoted 33 lines · Ihor Solodrai
Quoted 23 lines · Song Liu

cgroup_xattr/read_cgroupfs_xattr has two issues:

1. cgroup_xattr/read_cgroupfs_xattr messes up lo without creating a netns first. This causes issue with other tests.

Fix this by using a different hook (lsm.s/file_open) and not messing with lo.

2. cgroup_xattr/read_cgroupfs_xattr sets up cgroups without proper mount namespaces.

Fix this by using the existing cgroup helpers. A new helper set_cgroup_xattr() is added to set xattr on cgroup files.

---
Changes v1 => v2:
1. Add the second fix above.

Quoted 7 lines

v1: https://lore.kernel.org/bpf/[email protected]/

---
    tools/testing/selftests/bpf/cgroup_helpers.c  |  21 ++++
    tools/testing/selftests/bpf/cgroup_helpers.h  |   4 +
    .../selftests/bpf/prog_tests/cgroup_xattr.c   | 117 ++++--------------
    .../selftests/bpf/progs/read_cgroupfs_xattr.c |   4 +-
    4 files changed, 49 insertions(+), 97 deletions(-)
Quoted 5 lines

Hi Song.

I tried this patch on BPF CI, and it appears it fixes the hanging failure we've been seeing today on bpf-next and netdev. I am going to add it to ci/diffs.

Applied to bpf-next already.

Quoted 7 lines
Quoted 2 lines

CI patches apply to all base branches. My understanding is, it's needed at least for netdev too.

How is that possible?

The offending commit is only in /master and in /for-next branches, while /for-next is there for linux-next only.

Alexei, for-next contains offending commit, but does not have Song's fix. Right now it's the only base branch on BPF CI that uses the temp patch.

We do run tests on for-next, so I suppose the patch should remain in ci/diffs until it's committed into for-next?

#10 lore raw
Quoted 50 lines · Ihor Solodrai
Quoted 48 lines · Alexei Starovoitov
Quoted 46 lines · Ihor Solodrai
Quoted 34 lines · Alexei Starovoitov
Quoted 33 lines · Ihor Solodrai
Quoted 30 lines · Song Liu

cgroup_xattr/read_cgroupfs_xattr has two issues:

1. cgroup_xattr/read_cgroupfs_xattr messes up lo without creating a netns first. This causes issue with other tests.

Fix this by using a different hook (lsm.s/file_open) and not messing with lo.

2. cgroup_xattr/read_cgroupfs_xattr sets up cgroups without proper mount namespaces.

Fix this by using the existing cgroup helpers. A new helper set_cgroup_xattr() is added to set xattr on cgroup files.

---
Changes v1 => v2:
1. Add the second fix above.

v1: https://lore.kernel.org/bpf/[email protected]/

---
    tools/testing/selftests/bpf/cgroup_helpers.c  |  21 ++++
    tools/testing/selftests/bpf/cgroup_helpers.h  |   4 +
    .../selftests/bpf/prog_tests/cgroup_xattr.c   | 117 ++++--------------
    .../selftests/bpf/progs/read_cgroupfs_xattr.c |   4 +-
    4 files changed, 49 insertions(+), 97 deletions(-)
Quoted 7 lines
Quoted 5 lines

Hi Song.

I tried this patch on BPF CI, and it appears it fixes the hanging failure we've been seeing today on bpf-next and netdev. I am going to add it to ci/diffs.

Applied to bpf-next already.

CI patches apply to all base branches. My understanding is, it's needed at least for netdev too.

Quoted 8 lines
Quoted 4 lines

How is that possible?

The offending commit is only in /master and in /for-next branches, while /for-next is there for linux-next only.

Alexei, for-next contains offending commit, but does not have Song's fix. Right now it's the only base branch on BPF CI that uses the temp patch.

ok. updated /for-next

Quoted 2 lines

We do run tests on for-next, so I suppose the patch should remain in ci/diffs until it's committed into for-next?

It's news to me that we run BPF CI on /for-next.
I thought we only do it on /master and /net.

reviewer
#11 lore raw
Quoted 11 lines · Alexei Starovoitov
Quoted 9 lines · Ihor Solodrai

[...]

Quoted 3 lines

The offending commit is only in /master and in /for-next branches, while /for-next is there for linux-next only.

Alexei, for-next contains offending commit, but does not have Song's fix. Right now it's the only base branch on BPF CI that uses the temp patch.

Quoted 7 lines

ok. updated /for-next

Quoted 2 lines

We do run tests on for-next, so I suppose the patch should remain in ci/diffs until it's committed into for-next?

It's news to me that we run BPF CI on /for-next.
I thought we only do it on /master and /net.

Currently we have 4 base branches, for which CI runs on push:
* bpf = bpf/master
* bpf-next = bpf-next/master
* bpf-net = bpf-next/net
* for-next = bpf-next/for-next

I added bpf-net and for-next in March, but Manu opened a PR [1] for that last year, so it looks like we planned to do it a while ago.

The netdev is done via PR from [2] on github side, not managed by KPD.

[1] https://github.com/linux-netdev/testing-bpf-ci/tree/to-test
[2] https://github.com/kernel-patches/vmtest/pull/286

#12 lore raw
In reply to Song Liu
This patch was applied to bpf/bpf-next.git (master) by Alexei Starovoitov <[email protected]>:

Hello:

This patch was applied to bpf/bpf-next.git (master) by Alexei Starovoitov <[email protected]>:

Quoted 9 lines · On Fri, 27 Jun 2025 12:12:21 -0700 you wrote:

cgroup_xattr/read_cgroupfs_xattr has two issues:

1. cgroup_xattr/read_cgroupfs_xattr messes up lo without creating a netns first. This causes issue with other tests.

Fix this by using a different hook (lsm.s/file_open) and not messing with lo.

[...]

Here is the summary with links:
- [v2,bpf-next] selftests/bpf: Fix cgroup_xattr/read_cgroupfs_xattr

    https://git.kernel.org/bpf/bpf-next/c/bacdf5a0e69d

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html