diff --git a/src/core/bpf-devices.c b/src/core/bpf-devices.c
index adaa929364..4417ab58e9 100644
--- a/src/core/bpf-devices.c
+++ b/src/core/bpf-devices.c
@@ -252,51 +252,6 @@ int bpf_devices_apply_policy(
         return 0;
 }
 
-int bpf_devices_supported(void) {
-        const struct bpf_insn trivial[] = {
-                BPF_MOV64_IMM(BPF_REG_0, 1),
-                BPF_EXIT_INSN()
-        };
-
-        _cleanup_(bpf_program_freep) BPFProgram *program = NULL;
-        static int supported = -1;
-        int r;
-
-        /* Checks whether BPF device controller is supported. For this, we check two things:
-         *
-         * a) whether we are privileged
-         * b) the BPF implementation in the kernel supports BPF_PROG_TYPE_CGROUP_DEVICE programs, which we require
-         */
-
-        if (supported >= 0)
-                return supported;
-
-        if (geteuid() != 0) {
-                log_debug("Not enough privileges, BPF device control is not supported.");
-                return supported = 0;
-        }
-
-        r = bpf_program_new(BPF_PROG_TYPE_CGROUP_DEVICE, "sd_devices", &program);
-        if (r < 0) {
-                log_debug_errno(r, "Can't allocate CGROUP DEVICE BPF program, BPF device control is not supported: %m");
-                return supported = 0;
-        }
-
-        r = bpf_program_add_instructions(program, trivial, ELEMENTSOF(trivial));
-        if (r < 0) {
-                log_debug_errno(r, "Can't add trivial instructions to CGROUP DEVICE BPF program, BPF device control is not supported: %m");
-                return supported = 0;
-        }
-
-        r = bpf_program_load_kernel(program, NULL, 0);
-        if (r < 0) {
-                log_debug_errno(r, "Can't load kernel CGROUP DEVICE BPF program, BPF device control is not supported: %m");
-                return supported = 0;
-        }
-
-        return supported = 1;
-}
-
 static int allow_list_device_pattern(
                 BPFProgram *prog,
                 const char *path,
diff --git a/src/core/bpf-devices.h b/src/core/bpf-devices.h
index 5660e1a03a..a1d261548d 100644
--- a/src/core/bpf-devices.h
+++ b/src/core/bpf-devices.h
@@ -15,7 +15,6 @@ int bpf_devices_apply_policy(
                 const char *cgroup_path,
                 BPFProgram **prog_installed);
 
-int bpf_devices_supported(void);
 int bpf_devices_allow_list_device(BPFProgram *prog, const char *path, const char *node, CGroupDevicePermissions p);
 int bpf_devices_allow_list_major(BPFProgram *prog, const char *path, const char *name, char type, CGroupDevicePermissions p);
 int bpf_devices_allow_list_static(BPFProgram *prog, const char *path);
diff --git a/src/core/cgroup.c b/src/core/cgroup.c
index 896db2d647..dbab38c3c0 100644
--- a/src/core/cgroup.c
+++ b/src/core/cgroup.c
@@ -3259,18 +3259,12 @@ static int cg_bpf_mask_supported(CGroupMask *ret) {
         CGroupMask mask = 0;
         int r;
 
-        /* BPF-based firewall and pinned foreign prog */
+        /* BPF-based firewall, device access control, and pinned foreign prog */
         if (bpf_program_supported() > 0)
                 mask |= CGROUP_MASK_BPF_FIREWALL |
+                        CGROUP_MASK_BPF_DEVICES |
                         CGROUP_MASK_BPF_FOREIGN;
 
-        /* BPF-based device access control */
-        r = bpf_devices_supported();
-        if (r < 0)
-                return r;
-        if (r > 0)
-                mask |= CGROUP_MASK_BPF_DEVICES;
-
         /* BPF-based bind{4|6} hooks */
         r = bpf_socket_bind_supported();
         if (r < 0)
diff --git a/src/test/test-bpf-devices.c b/src/test/test-bpf-devices.c
index 8b0b744892..1642673e75 100644
--- a/src/test/test-bpf-devices.c
+++ b/src/test/test-bpf-devices.c
@@ -257,6 +257,11 @@ int main(int argc, char *argv[]) {
 
         test_setup_logging(LOG_DEBUG);
 
+        r = bpf_program_supported();
+        if (r < 0)
+                return log_tests_skipped_errno(r, "BPF device filter not supported");
+        ASSERT_TRUE(r);
+
         ASSERT_OK(getrlimit(RLIMIT_MEMLOCK, &rl));
         rl.rlim_cur = rl.rlim_max = MAX(rl.rlim_max, CAN_MEMLOCK_SIZE);
         (void) setrlimit(RLIMIT_MEMLOCK, &rl);
@@ -274,11 +279,6 @@ int main(int argc, char *argv[]) {
         if (r < 0)
                 return log_tests_skipped_errno(r, "Failed to prepare cgroup subtree");
 
-        r = bpf_devices_supported();
-        if (r == 0)
-                return log_tests_skipped("BPF device filter not supported");
-        ASSERT_EQ(r, 1);
-
         r = cg_get_path(SYSTEMD_CGROUP_CONTROLLER, cgroup, NULL, &controller_path);
         ASSERT_OK(r);