Re: [PATCH bpf-next 3/3] selftests/bpf: Add a test for arena fault-in under memory.max

["Emil Tsalapatis" <emil@etsalapatis.com>]

On Mon Jul 27, 2026 at 2:24 AM EDT, Jiayuan Chen wrote:
> A child joins a memcg capped at 64M and faults an arena in until it runs
> out of the budget. Without the kernel fix the child dies with SIGSEGV on
> a valid arena address; with it, the child is killed by the memcg OOM
> killer.
>
> With the fix:
>
>   serial_test_arena_memcg:PASS:child killed by signal
>   serial_test_arena_memcg:PASS:not killed by SIGSEGV
>   #5       arena_memcg:OK
>
>   # dmesg
>    arena_vm_fault+0x655/0xa90
>   Memory cgroup out of memory: Killed process 512, file-rss:67920kB
>
> Without the fix:
>
>   serial_test_arena_memcg:PASS:child killed by signal
>   serial_test_arena_memcg:FAIL:not killed by SIGSEGV: actual 11
>   #5       arena_memcg:FAIL
>
>   # dmesg
>   test_progs[508]: segfault at 100004025000 ...
>
> Signed-off-by: Jiayuan Chen <jiayuan.chen@linux.dev>
> ---
>  .../selftests/bpf/prog_tests/arena_memcg.c    | 158 ++++++++++++++++++
>  .../testing/selftests/bpf/progs/arena_memcg.c |  24 +++
>  2 files changed, 182 insertions(+)
>  create mode 100644 tools/testing/selftests/bpf/prog_tests/arena_memcg.c
>  create mode 100644 tools/testing/selftests/bpf/progs/arena_memcg.c
>
> diff --git a/tools/testing/selftests/bpf/prog_tests/arena_memcg.c b/tools/testing/selftests/bpf/prog_tests/arena_memcg.c
> new file mode 100644
> index 000000000000..9665946fa29e
> --- /dev/null
> +++ b/tools/testing/selftests/bpf/prog_tests/arena_memcg.c
> @@ -0,0 +1,158 @@
> +// SPDX-License-Identifier: GPL-2.0
> +
> +#include <test_progs.h>
> +#include <fcntl.h>
> +#include <signal.h>
> +#include <sys/mman.h>
> +#include <sys/wait.h>
> +#include <unistd.h>
> +#include <sys/user.h>
> +#ifndef PAGE_SIZE /* on some archs it comes in sys/user.h */
> +#include <unistd.h>
> +#define PAGE_SIZE getpagesize()
> +#endif
> +
> +#include "cgroup_helpers.h"
> +#include "arena_memcg.skel.h"
> +
> +#define CG_PATH		"/arena_memcg"
> +
> +/* Budget the arena gets on top of whatever is already charged after load. */
> +#define ARENA_BUDGET	(64 * 1024 * 1024)
> +
> +/*
> + * cgroup_helpers builds paths from getpid(), but the work dir belongs to the
> + * process that set the environment up. The child references it through that
> + * pid, so build the path explicitly.
> + */
> +static void cg_file_path(char *buf, size_t sz, pid_t owner, const char *file)
> +{
> +	snprintf(buf, sz, "/mnt/cgroup-test-work-dir%d%s/%s", owner, CG_PATH, file);

These are copied over from cgroup_helpers.c, but it's not obvious they
originate from there. Maybe let's move them to cgroup_helpers.h where we
can use them everywhere for consistency.

The main issue I see is that this only triggers consistently with PREEMPT_RT, correct?
I tried with the default vmtest config we have but it does not trigger
at all. More importantly, it doesn't trigger reliably with PREEMPT_RT,
either. Does it for you? Can we cycle forks/frees multiple times to try
and trigger this more reliably?

> +}
> +
> +static long cg_read_ulong(pid_t owner, const char *file)
> +{
> +	char path[PATH_MAX], buf[64];
> +	long val = -1;
> +	FILE *f;
> +
> +	cg_file_path(path, sizeof(path), owner, file);
> +	f = fopen(path, "r");
> +	if (!f)
> +		return -1;
> +	if (fgets(buf, sizeof(buf), f))
> +		val = strtol(buf, NULL, 10);
> +	fclose(f);
> +	return val;
> +}
> +
> +static int cg_write(pid_t owner, const char *file, const char *val)
> +{
> +	char path[PATH_MAX];
> +	int fd, len, ret = -1;
> +
> +	cg_file_path(path, sizeof(path), owner, file);
> +	fd = open(path, O_WRONLY);
> +	if (fd < 0)
> +		return -1;
> +	len = strlen(val);
> +	if (write(fd, val, len) == len)
> +		ret = 0;
> +	close(fd);
> +	return ret;
> +}
> +
> +void serial_test_arena_memcg(void)
> +{
> +	int cgroup_fd = -1, status;
> +	const long ps = PAGE_SIZE;
> +	pid_t owner, pid;
> +
> +	if (setup_cgroup_environment())
> +		return;
> +	owner = getpid();
> +
> +	cgroup_fd = create_and_get_cgroup(CG_PATH);
> +	if (!ASSERT_OK_FD(cgroup_fd, "create_and_get_cgroup"))
> +		goto out;
> +
> +	/* No memory controller -> nothing to test. */
> +	if (cg_read_ulong(owner, "memory.current") < 0) {
> +		test__skip();
> +		goto out;
> +	}
> +
> +	pid = fork();
> +	if (!ASSERT_GE(pid, 0, "fork"))
> +		goto out;
> +	if (pid == 0) {
> +		struct arena_memcg *cskel;
> +		__u32 i, npages;
> +		char buf[32];
> +		char *base;
> +		size_t sz;
> +		long cur;
> +
> +		/*
> +		 * Do everything from the child: the arena vma is VM_DONTCOPY so
> +		 * it would not survive fork(), only the child should be under the
> +		 * limit so that a memcg OOM cannot pick test_progs, and a map is
> +		 * charged to the memcg of the task that creates it - so join
> +		 * before load. Errors are reported to the parent through the exit
> +		 * code, since ASSERT_* in a forked child does not reach it.
> +		 */
> +		snprintf(buf, sizeof(buf), "%d", getpid());
> +		if (cg_write(owner, "cgroup.procs", buf))
> +			_exit(2);
> +
> +		cskel = arena_memcg__open_and_load();
> +		if (!cskel)
> +			_exit(3);
> +
> +		base = bpf_map__initial_value(cskel->maps.arena, &sz);
> +		if (!base)
> +			_exit(4);
> +		npages = bpf_map__max_entries(cskel->maps.arena);
> +
> +		/*
> +		 * Cap only now, after load: everything but the fault-in is
> +		 * charged, so the arena gets a fixed budget regardless of what
> +		 * the load itself cost, and the load can never hit the limit.
> +		 */
> +		cur = cg_read_ulong(owner, "memory.current");
> +		if (cur < 0)
> +			_exit(5);
> +		snprintf(buf, sizeof(buf), "%ld", cur + ARENA_BUDGET);
> +		if (cg_write(owner, "memory.max", buf))
> +			_exit(6);
> +
> +		for (i = 0; i < npages; i++)
> +			base[(size_t)i * ps] = 1;
> +		_exit(0);
> +	}
> +
> +	if (!ASSERT_EQ(waitpid(pid, &status, 0), pid, "waitpid"))
> +		goto out;
> +
> +	/* A non-zero exit means the child failed to set up; the code says where. */
> +	if (WIFEXITED(status) && WEXITSTATUS(status)) {
> +		ASSERT_OK(WEXITSTATUS(status), "child setup");
> +		goto out;
> +	}
> +
> +	/*
> +	 * Faulting a valid arena address until memory.max is hit must not look
> +	 * like an invalid access. Without the fix the fault path allocated with
> +	 * the non-blocking allocator, turned its -ENOMEM into VM_FAULT_SIGSEGV,
> +	 * and the child died with SIGSEGV on a valid address; now it is handled
> +	 * by the memcg OOM path and the child is killed by SIGKILL instead. A
> +	 * clean exit means the child failed to set up (see the exit codes).
> +	 */
> +	if (!ASSERT_TRUE(WIFSIGNALED(status), "child killed by signal"))
> +		goto out;
> +	ASSERT_NEQ(WTERMSIG(status), SIGSEGV, "not killed by SIGSEGV");
> +out:
> +	if (cgroup_fd >= 0)
> +		close(cgroup_fd);
> +	cleanup_cgroup_environment();
> +}
> diff --git a/tools/testing/selftests/bpf/progs/arena_memcg.c b/tools/testing/selftests/bpf/progs/arena_memcg.c
> new file mode 100644
> index 000000000000..adecd9e8463e
> --- /dev/null
> +++ b/tools/testing/selftests/bpf/progs/arena_memcg.c
> @@ -0,0 +1,24 @@
> +// SPDX-License-Identifier: GPL-2.0
> +
> +#include <vmlinux.h>
> +#include <bpf/bpf_helpers.h>
> +#include "bpf_arena_common.h"
> +
> +struct {
> +	__uint(type, BPF_MAP_TYPE_ARENA);
> +	__uint(map_flags, BPF_F_MMAPABLE);
> +	__uint(max_entries, 100000); /* number of pages */
> +#ifdef __TARGET_ARCH_arm64
> +	__ulong(map_extra, 0x1ull << 32); /* start of mmap() region */
> +#else
> +	__ulong(map_extra, 0x1ull << 44); /* start of mmap() region */
> +#endif
> +} arena SEC(".maps");
> +
> +SEC("syscall")
> +int noop(void *ctx)
> +{
> +	return 0;
> +}
> +
> +char _license[] SEC("license") = "GPL";