Re: [patch 07/10] selftests/rseq: Validate legacy behavior

Dmitry Vyukov <[email protected]>
Newsgroups org.kernel.vger.linux-man,dev.linux.lists.regressions,org.kernel.vger.linux-kernel
Message-ID <CACT4Y+YEyR9X0D6ap94ttUv3byM7gSn+zJUsfGQ69-eSREnBPw@mail.gmail.com>
On Wed, 29 Apr 2026 at 01:34, Thomas Gleixner <[email protected]> wrote:
>
> The RSEQ legacy mode behavior requires that the ID fields in the rseq
> region are unconditionally updated on every context switch and before
> signal delivery even if not required by the ABI specification.
>
> To ensure that this behavior is preserved for legacy users in the future,
> add a test which validates that with a sleep() and a signal sent to self.
>
> Provide a run script which prevents GLIBC from registering a RSEQ region,
> so that the test can register it's own legacy sized region.
>
> Fixes: 566d8015f7ee ("rseq: Avoid CPU/MM CID updates when no event pending")
> Signed-off-by: Thomas Gleixner <[email protected]>
> Cc: [email protected]

Reviewed-by: Dmitry Vyukov <[email protected]>

> ---
>  tools/testing/selftests/rseq/Makefile            |    4 -
>  tools/testing/selftests/rseq/legacy_check.c      |   65 +++++++++++++++++++++++
>  tools/testing/selftests/rseq/run_legacy_check.sh |    4 +
>  3 files changed, 71 insertions(+), 2 deletions(-)
>
> --- a/tools/testing/selftests/rseq/Makefile
> +++ b/tools/testing/selftests/rseq/Makefile
> @@ -17,11 +17,11 @@ OVERRIDE_TARGETS = 1
>  TEST_GEN_PROGS = basic_test basic_percpu_ops_test basic_percpu_ops_mm_cid_test param_test \
>                 param_test_benchmark param_test_compare_twice param_test_mm_cid \
>                 param_test_mm_cid_benchmark param_test_mm_cid_compare_twice \
> -               syscall_errors_test slice_test
> +               syscall_errors_test slice_test legacy_check
>
>  TEST_GEN_PROGS_EXTENDED = librseq.so
>
> -TEST_PROGS = run_param_test.sh run_syscall_errors_test.sh
> +TEST_PROGS = run_param_test.sh run_syscall_errors_test.sh run_legacy_check.sh
>
>  TEST_FILES := settings
>
> --- /dev/null
> +++ b/tools/testing/selftests/rseq/legacy_check.c
> @@ -0,0 +1,65 @@
> +// SPDX-License-Identifier: GPL-2.0
> +#ifndef _GNU_SOURCE
> +#define _GNU_SOURCE
> +#endif
> +
> +#include <errno.h>
> +#include <signal.h>
> +#include <stdint.h>
> +#include <unistd.h>
> +
> +#include "rseq.h"
> +
> +#include "../kselftest_harness.h"
> +
> +FIXTURE(legacy)
> +{
> +};
> +
> +static int cpu_id_in_sigfn = -1;
> +
> +static void sigfn(int sig)
> +{
> +       struct rseq_abi *rs = rseq_get_abi();
> +
> +       cpu_id_in_sigfn = rs->cpu_id_start;
> +}
> +
> +FIXTURE_SETUP(legacy)
> +{
> +       int res = __rseq_register_current_thread(true, true);
> +
> +       switch (res) {
> +       case -ENOSYS:
> +               SKIP(return, "RSEQ not enabled\n");
> +       case -EBUSY:
> +               SKIP(return, "GLIBC owns RSEQ. Disable GLIBC RSEQ registration\n");
> +       default:
> +               ASSERT_EQ(res, 0);
> +       }
> +
> +       ASSERT_NE(signal(SIGUSR1, sigfn), SIG_ERR);
> +}
> +
> +FIXTURE_TEARDOWN(legacy)
> +{
> +}
> +
> +TEST_F(legacy, legacy_test)
> +{
> +       struct rseq_abi *rs = rseq_get_abi();
> +
> +       ASSERT_NE(rs, NULL);
> +
> +       /* Overwrite rs::cpu_id_start */
> +       rs->cpu_id_start = -1;
> +       sleep(1);
> +       ASSERT_NE(rs->cpu_id_start, -1);
> +
> +       rs->cpu_id_start = -1;
> +       ASSERT_EQ(raise(SIGUSR1), 0);
> +       ASSERT_NE(rs->cpu_id_start, -1);
> +       ASSERT_NE(cpu_id_in_sigfn, -1);
> +}
> +
> +TEST_HARNESS_MAIN
> --- /dev/null
> +++ b/tools/testing/selftests/rseq/run_legacy_check.sh
> @@ -0,0 +1,4 @@
> +#!/bin/bash
> +# SPDX-License-Identifier: GPL-2.0
> +
> +GLIBC_TUNABLES="${GLIBC_TUNABLES:-}:glibc.pthread.rseq=0" ./legacy_check
>
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.