linux-kernel - Re: Linux 5.0-rc2 seccomp_bpf user_notification

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAGXu5jLGqXstRCmM1_fxUxn4djs1YLUun2Fc8ExcaN4-CKyELQ@mail.gmail.com>
Date:   Wed, 16 Jan 2019 16:30:26 -0800
From:   Kees Cook <keescook@...omium.org>
To:     shuah <shuah@...nel.org>, Tycho Andersen <tycho@...ho.ws>
Cc:     James Morris <jmorris@...ei.org>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        "open list:KERNEL SELFTEST FRAMEWORK" 
        <linux-kselftest@...r.kernel.org>
Subject: Re: Linux 5.0-rc2 seccomp_bpf user_notification_basic test hangs

On Wed, Jan 16, 2019 at 4:01 PM shuah <shuah@...nel.org> wrote:
>
> Hi Kees and James,
>
> seccomp_bpf test hangs right after the following test passes
> with EBUSY. Please see log at the end.
>
> /* Installing a second listener in the chain should EBUSY */
>          EXPECT_EQ(user_trap_syscall(__NR_getpid,
>                                      SECCOMP_FILTER_FLAG_NEW_LISTENER),
>                    -1);
>          EXPECT_EQ(errno, EBUSY);
>
>
> The user_notification_basic test starts running I assume and then
> the hang.
>
> The only commit I see that could be suspect is the following as
> it talks about adding SECCOMP_RET_USER_NOTIF
>
> commit d9a7fa67b4bfe6ce93ee9aab23ae2e7ca0763e84
> Merge: f218a29c25ad 55b8cbe470d1
> Author: Linus Torvalds <torvalds@...ux-foundation.org>
> Date:   Wed Jan 2 09:48:13 2019 -0800
>
>      Merge branch 'next-seccomp' of
> git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security
>
>      Pull seccomp updates from James Morris:
>
>       - Add SECCOMP_RET_USER_NOTIF
>
>       - seccomp fixes for sparse warnings and s390 build (Tycho)
>
>      * 'next-seccomp' of
> git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security:
>        seccomp, s390: fix build for syscall type change
>        seccomp: fix poor type promotion
>        samples: add an example of seccomp user trap
>        seccomp: add a return code to trap to userspace
>        seccomp: switch system call argument type to void *
>        seccomp: hoist struct seccomp_data recalculation higher
>
>
> Any ideas on how to proceed? Here is the log. The following
> reproduces the problem.
>
> make -C tools/testing/selftests/seccomp/ run_tests
>
>
> seccomp_bpf.c:2947:global.get_metadata:Expected 0 (0) ==
> seccomp(SECCOMP_SET_MODE_FILTER, SECCOMP_FILTER_FLAG_LOG, &prog)
> (18446744073709551615)
> seccomp_bpf.c:2959:global.get_metadata:Expected 1 (1) == read(pipefd[0],
> &buf, 1) (0)
> global.get_metadata: Test terminated by assertion
> [     FAIL ] global.get_metadata
> [ RUN      ] global.user_notification_basic
> seccomp_bpf.c:3036:global.user_notification_basic:Expected 0 (0) ==
> WEXITSTATUS(status) (1)
> seccomp_bpf.c:3039:global.user_notification_basic:Expected
> seccomp(SECCOMP_SET_MODE_FILTER, 0, &prog) (18446744073709551615) == 0 (0)
> seccomp_bpf.c:3040:global.user_notification_basic:Expected
> seccomp(SECCOMP_SET_MODE_FILTER, 0, &prog) (18446744073709551615) == 0 (0)
> seccomp_bpf.c:3041:global.user_notification_basic:Expected
> seccomp(SECCOMP_SET_MODE_FILTER, 0, &prog) (18446744073709551615) == 0 (0)
> seccomp_bpf.c:3042:global.user_notification_basic:Expected
> seccomp(SECCOMP_SET_MODE_FILTER, 0, &prog) (18446744073709551615) == 0 (0)
> seccomp_bpf.c:3047:global.user_notification_basic:Expected listener
> (18446744073709551615) >= 0 (0)
> seccomp_bpf.c:3053:global.user_notification_basic:Expected errno (13) ==
> EBUSY (16)

Looks like the test is unfriendly when running the current selftest on
an old kernel version. A quick look seems like it's missing some
ASSERT_* cases where EXPECT_* is used. I'll send a patch.

-- 
Kees Cook