Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1221561 > unrolled thread

Re: [PATCH] selftests/seccomp: add support for s390

Started byKees Cook <keescook@chromium.org>
First post2015-09-09 18:50 +0200
Last post2015-09-10 11:20 +0200
Articles 7 — 3 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH] selftests/seccomp: add support for s390 Kees Cook <keescook@chromium.org> - 2015-09-09 18:50 +0200
    Re: [PATCH] selftests/seccomp: add support for s390 Shuah Khan <shuahkh@osg.samsung.com> - 2015-09-09 18:50 +0200
      Re: [PATCH] selftests/seccomp: add support for s390 Shuah Khan <shuahkh@osg.samsung.com> - 2015-09-09 19:00 +0200
        Re: [PATCH] selftests/seccomp: add support for s390 Kees Cook <keescook@chromium.org> - 2015-09-09 19:30 +0200
          Re: [PATCH] selftests/seccomp: add support for s390 Shuah Khan <shuahkh@osg.samsung.com> - 2015-09-09 20:50 +0200
            Re: [PATCH] selftests/seccomp: add support for s390 Shuah Khan <shuahkh@osg.samsung.com> - 2015-09-15 01:00 +0200
          Re: [PATCH] selftests/seccomp: add support for s390 Michael Ellerman <mpe@ellerman.id.au> - 2015-09-10 11:20 +0200

#1221561 — Re: [PATCH] selftests/seccomp: add support for s390

FromKees Cook <keescook@chromium.org>
Date2015-09-09 18:50 +0200
SubjectRe: [PATCH] selftests/seccomp: add support for s390
Message-ID<q6QJk-ho-9@gated-at.bofh.it>
Re-ping. Can someone pull this into their tree?

-Kees

On Fri, Aug 21, 2015 at 11:22 AM, Kees Cook <keescook@chromium.org> wrote:
> This adds support for s390 to the seccomp selftests. Some improvements
> were made to enhance the accuracy of failure reporting, and additional
> tests were added to validate assumptions about the currently traced
> syscall. Also adds early asserts for running on older kernels to avoid
> noise when the seccomp syscall is not implemented.
>
> Signed-off-by: Kees Cook <keescook@chromium.org>
> ---
> This applies on top of -next, following the addition of the powerpc tests.
> ---
>  tools/testing/selftests/seccomp/seccomp_bpf.c  | 37 +++++++++++++++++++++++++-
>  tools/testing/selftests/seccomp/test_harness.h |  7 ++---
>  2 files changed, 38 insertions(+), 6 deletions(-)
>
> diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
> index a004b4cce99e..770f47adf295 100644
> --- a/tools/testing/selftests/seccomp/seccomp_bpf.c
> +++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
> @@ -1210,6 +1210,10 @@ TEST_F(TRACE_poke, getpid_runs_normally)
>  # define ARCH_REGS     struct pt_regs
>  # define SYSCALL_NUM   gpr[0]
>  # define SYSCALL_RET   gpr[3]
> +#elif defined(__s390__)
> +# define ARCH_REGS     s390_regs
> +# define SYSCALL_NUM   gprs[2]
> +# define SYSCALL_RET   gprs[2]
>  #else
>  # error "Do not know how to find your architecture's registers and syscalls"
>  #endif
> @@ -1243,7 +1247,8 @@ void change_syscall(struct __test_metadata *_metadata,
>         ret = ptrace(PTRACE_GETREGSET, tracee, NT_PRSTATUS, &iov);
>         EXPECT_EQ(0, ret);
>
> -#if defined(__x86_64__) || defined(__i386__) || defined(__aarch64__) || defined(__powerpc__)
> +#if defined(__x86_64__) || defined(__i386__) || defined(__aarch64__) || \
> +    defined(__powerpc__) || defined(__s390__)
>         {
>                 regs.SYSCALL_NUM = syscall;
>         }
> @@ -1281,17 +1286,21 @@ void tracer_syscall(struct __test_metadata *_metadata, pid_t tracee,
>         ret = ptrace(PTRACE_GETEVENTMSG, tracee, NULL, &msg);
>         EXPECT_EQ(0, ret);
>
> +       /* Validate and take action on expected syscalls. */
>         switch (msg) {
>         case 0x1002:
>                 /* change getpid to getppid. */
> +               EXPECT_EQ(__NR_getpid, get_syscall(_metadata, tracee));
>                 change_syscall(_metadata, tracee, __NR_getppid);
>                 break;
>         case 0x1003:
>                 /* skip gettid. */
> +               EXPECT_EQ(__NR_gettid, get_syscall(_metadata, tracee));
>                 change_syscall(_metadata, tracee, -1);
>                 break;
>         case 0x1004:
>                 /* do nothing (allow getppid) */
> +               EXPECT_EQ(__NR_getppid, get_syscall(_metadata, tracee));
>                 break;
>         default:
>                 EXPECT_EQ(0, msg) {
> @@ -1409,6 +1418,8 @@ TEST_F(TRACE_syscall, syscall_dropped)
>  #  define __NR_seccomp 277
>  # elif defined(__powerpc__)
>  #  define __NR_seccomp 358
> +# elif defined(__s390__)
> +#  define __NR_seccomp 348
>  # else
>  #  warning "seccomp syscall number unknown for this architecture"
>  #  define __NR_seccomp 0xffff
> @@ -1453,6 +1464,9 @@ TEST(seccomp_syscall)
>
>         /* Reject insane operation. */
>         ret = seccomp(-1, 0, &prog);
> +       ASSERT_NE(ENOSYS, errno) {
> +               TH_LOG("Kernel does not support seccomp syscall!");
> +       }
>         EXPECT_EQ(EINVAL, errno) {
>                 TH_LOG("Did not reject crazy op value!");
>         }
> @@ -1501,6 +1515,9 @@ TEST(seccomp_syscall_mode_lock)
>         }
>
>         ret = seccomp(SECCOMP_SET_MODE_FILTER, 0, &prog);
> +       ASSERT_NE(ENOSYS, errno) {
> +               TH_LOG("Kernel does not support seccomp syscall!");
> +       }
>         EXPECT_EQ(0, ret) {
>                 TH_LOG("Could not install filter!");
>         }
> @@ -1535,6 +1552,9 @@ TEST(TSYNC_first)
>
>         ret = seccomp(SECCOMP_SET_MODE_FILTER, SECCOMP_FLAG_FILTER_TSYNC,
>                       &prog);
> +       ASSERT_NE(ENOSYS, errno) {
> +               TH_LOG("Kernel does not support seccomp syscall!");
> +       }
>         EXPECT_EQ(0, ret) {
>                 TH_LOG("Could not install initial filter with TSYNC!");
>         }
> @@ -1694,6 +1714,9 @@ TEST_F(TSYNC, siblings_fail_prctl)
>
>         /* Check prctl failure detection by requesting sib 0 diverge. */
>         ret = seccomp(SECCOMP_SET_MODE_FILTER, 0, &prog);
> +       ASSERT_NE(ENOSYS, errno) {
> +               TH_LOG("Kernel does not support seccomp syscall!");
> +       }
>         ASSERT_EQ(0, ret) {
>                 TH_LOG("setting filter failed");
>         }
> @@ -1731,6 +1754,9 @@ TEST_F(TSYNC, two_siblings_with_ancestor)
>         }
>
>         ret = seccomp(SECCOMP_SET_MODE_FILTER, 0, &self->root_prog);
> +       ASSERT_NE(ENOSYS, errno) {
> +               TH_LOG("Kernel does not support seccomp syscall!");
> +       }
>         ASSERT_EQ(0, ret) {
>                 TH_LOG("Kernel does not support SECCOMP_SET_MODE_FILTER!");
>         }
> @@ -1805,6 +1831,9 @@ TEST_F(TSYNC, two_siblings_with_no_filter)
>
>         ret = seccomp(SECCOMP_SET_MODE_FILTER, SECCOMP_FLAG_FILTER_TSYNC,
>                       &self->apply_prog);
> +       ASSERT_NE(ENOSYS, errno) {
> +               TH_LOG("Kernel does not support seccomp syscall!");
> +       }
>         ASSERT_EQ(0, ret) {
>                 TH_LOG("Could install filter on all threads!");
>         }
> @@ -1833,6 +1862,9 @@ TEST_F(TSYNC, two_siblings_with_one_divergence)
>         }
>
>         ret = seccomp(SECCOMP_SET_MODE_FILTER, 0, &self->root_prog);
> +       ASSERT_NE(ENOSYS, errno) {
> +               TH_LOG("Kernel does not support seccomp syscall!");
> +       }
>         ASSERT_EQ(0, ret) {
>                 TH_LOG("Kernel does not support SECCOMP_SET_MODE_FILTER!");
>         }
> @@ -1890,6 +1922,9 @@ TEST_F(TSYNC, two_siblings_not_under_filter)
>         }
>
>         ret = seccomp(SECCOMP_SET_MODE_FILTER, 0, &self->root_prog);
> +       ASSERT_NE(ENOSYS, errno) {
> +               TH_LOG("Kernel does not support seccomp syscall!");
> +       }
>         ASSERT_EQ(0, ret) {
>                 TH_LOG("Kernel does not support SECCOMP_SET_MODE_FILTER!");
>         }
> diff --git a/tools/testing/selftests/seccomp/test_harness.h b/tools/testing/selftests/seccomp/test_harness.h
> index 977a6afc4489..fb2841601f2f 100644
> --- a/tools/testing/selftests/seccomp/test_harness.h
> +++ b/tools/testing/selftests/seccomp/test_harness.h
> @@ -370,11 +370,8 @@
>         __typeof__(_expected) __exp = (_expected); \
>         __typeof__(_seen) __seen = (_seen); \
>         if (!(__exp _t __seen)) { \
> -               unsigned long long __exp_print = 0; \
> -               unsigned long long __seen_print = 0; \
> -               /* Avoid casting complaints the scariest way we can. */ \
> -               memcpy(&__exp_print, &__exp, sizeof(__exp)); \
> -               memcpy(&__seen_print, &__seen, sizeof(__seen)); \
> +               unsigned long long __exp_print = (unsigned long long)__exp; \
> +               unsigned long long __seen_print = (unsigned long long)__seen; \
>                 __TH_LOG("Expected %s (%llu) %s %s (%llu)", \
>                          #_expected, __exp_print, #_t, \
>                          #_seen, __seen_print); \
> --
> 1.9.1
>
>
> --
> Kees Cook
> Chrome OS Security



-- 
Kees Cook
Chrome OS Security
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1221562

FromShuah Khan <shuahkh@osg.samsung.com>
Date2015-09-09 18:50 +0200
Message-ID<q6QJk-ho-15@gated-at.bofh.it>
In reply to#1221561
On 09/09/2015 10:40 AM, Kees Cook wrote:
> Re-ping. Can someone pull this into their tree?
> 
> -Kees
> 
> On Fri, Aug 21, 2015 at 11:22 AM, Kees Cook <keescook@chromium.org> wrote:
>> This adds support for s390 to the seccomp selftests. Some improvements
>> were made to enhance the accuracy of failure reporting, and additional
>> tests were added to validate assumptions about the currently traced
>> syscall. Also adds early asserts for running on older kernels to avoid
>> noise when the seccomp syscall is not implemented.
>>
>> Signed-off-by: Kees Cook <keescook@chromium.org>
>> ---


Sorry looks like it got lost in my LinuxCon backlog. I will queue it
up for 4.3-rc1.

thanks,
-- Shuah


-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1221567

FromShuah Khan <shuahkh@osg.samsung.com>
Date2015-09-09 19:00 +0200
Message-ID<q6QT0-sK-5@gated-at.bofh.it>
In reply to#1221562
On 09/09/2015 10:45 AM, Shuah Khan wrote:
> On 09/09/2015 10:40 AM, Kees Cook wrote:
>> Re-ping. Can someone pull this into their tree?
>>
>> -Kees
>>
>> On Fri, Aug 21, 2015 at 11:22 AM, Kees Cook <keescook@chromium.org> wrote:
>>> This adds support for s390 to the seccomp selftests. Some improvements
>>> were made to enhance the accuracy of failure reporting, and additional
>>> tests were added to validate assumptions about the currently traced
>>> syscall. Also adds early asserts for running on older kernels to avoid
>>> noise when the seccomp syscall is not implemented.
>>>
>>> Signed-off-by: Kees Cook <keescook@chromium.org>
>>> ---
> 
> 
> Sorry looks like it got lost in my LinuxCon backlog. I will queue it
> up for 4.3-rc1.
> 

Hmm. It doesn't apply to linux-kselftest next. Based on your comment

"This applies on top of -next, following the addition of the powerpc
tests."

This one probably should go through PowerPc tree.

Adding Michael Ellerman to the thread,. Michael! Something
you can take in for 4.3-rc1??

thanks,
-- Shuah



-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1221583

FromKees Cook <keescook@chromium.org>
Date2015-09-09 19:30 +0200
Message-ID<q6Rm2-1gb-19@gated-at.bofh.it>
In reply to#1221567
On Wed, Sep 9, 2015 at 9:56 AM, Shuah Khan <shuahkh@osg.samsung.com> wrote:
> On 09/09/2015 10:45 AM, Shuah Khan wrote:
>> On 09/09/2015 10:40 AM, Kees Cook wrote:
>>> Re-ping. Can someone pull this into their tree?
>>>
>>> -Kees
>>>
>>> On Fri, Aug 21, 2015 at 11:22 AM, Kees Cook <keescook@chromium.org> wrote:
>>>> This adds support for s390 to the seccomp selftests. Some improvements
>>>> were made to enhance the accuracy of failure reporting, and additional
>>>> tests were added to validate assumptions about the currently traced
>>>> syscall. Also adds early asserts for running on older kernels to avoid
>>>> noise when the seccomp syscall is not implemented.
>>>>
>>>> Signed-off-by: Kees Cook <keescook@chromium.org>
>>>> ---
>>
>>
>> Sorry looks like it got lost in my LinuxCon backlog. I will queue it
>> up for 4.3-rc1.
>>
>
> Hmm. It doesn't apply to linux-kselftest next. Based on your comment
>
> "This applies on top of -next, following the addition of the powerpc
> tests."

Well, it was -next when I sent it. :) The powerpc changes are now in
Linus's tree, so there should be no problem. Is your linux-kselftest
merged with 4.3-rc yet?

-Kees

>
> This one probably should go through PowerPc tree.
>
> Adding Michael Ellerman to the thread,. Michael! Something
> you can take in for 4.3-rc1??
>
> thanks,
> -- Shuah
>
>
>
> --
> Shuah Khan
> Sr. Linux Kernel Developer
> Open Source Innovation Group
> Samsung Research America (Silicon Valley)
> shuahkh@osg.samsung.com | (970) 217-8978



-- 
Kees Cook
Chrome OS Security
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1221632

FromShuah Khan <shuahkh@osg.samsung.com>
Date2015-09-09 20:50 +0200
Message-ID<q6SBr-2Y3-3@gated-at.bofh.it>
In reply to#1221583
On 09/09/2015 11:23 AM, Kees Cook wrote:
> On Wed, Sep 9, 2015 at 9:56 AM, Shuah Khan <shuahkh@osg.samsung.com> wrote:
>> On 09/09/2015 10:45 AM, Shuah Khan wrote:
>>> On 09/09/2015 10:40 AM, Kees Cook wrote:
>>>> Re-ping. Can someone pull this into their tree?
>>>>
>>>> -Kees
>>>>
>>>> On Fri, Aug 21, 2015 at 11:22 AM, Kees Cook <keescook@chromium.org> wrote:
>>>>> This adds support for s390 to the seccomp selftests. Some improvements
>>>>> were made to enhance the accuracy of failure reporting, and additional
>>>>> tests were added to validate assumptions about the currently traced
>>>>> syscall. Also adds early asserts for running on older kernels to avoid
>>>>> noise when the seccomp syscall is not implemented.
>>>>>
>>>>> Signed-off-by: Kees Cook <keescook@chromium.org>
>>>>> ---
>>>
>>>
>>> Sorry looks like it got lost in my LinuxCon backlog. I will queue it
>>> up for 4.3-rc1.
>>>
>>
>> Hmm. It doesn't apply to linux-kselftest next. Based on your comment
>>
>> "This applies on top of -next, following the addition of the powerpc
>> tests."
> 
> Well, it was -next when I sent it. :) The powerpc changes are now in
> Linus's tree, so there should be no problem. Is your linux-kselftest
> merged with 4.3-rc yet?
> 

Kees,

I will get this into 4.3-rc2 as this patch fixes failure paths with
other fixes I am planning to send in rc2. Thanks for the ping on this.

thanks,
-- Shuah


-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1224525

FromShuah Khan <shuahkh@osg.samsung.com>
Date2015-09-15 01:00 +0200
Message-ID<q8KT8-3i1-19@gated-at.bofh.it>
In reply to#1221632
On 09/09/2015 12:46 PM, Shuah Khan wrote:
> On 09/09/2015 11:23 AM, Kees Cook wrote:
>> On Wed, Sep 9, 2015 at 9:56 AM, Shuah Khan <shuahkh@osg.samsung.com> wrote:
>>> On 09/09/2015 10:45 AM, Shuah Khan wrote:
>>>> On 09/09/2015 10:40 AM, Kees Cook wrote:
>>>>> Re-ping. Can someone pull this into their tree?
>>>>>
>>>>> -Kees
>>>>>
>>>>> On Fri, Aug 21, 2015 at 11:22 AM, Kees Cook <keescook@chromium.org> wrote:
>>>>>> This adds support for s390 to the seccomp selftests. Some improvements
>>>>>> were made to enhance the accuracy of failure reporting, and additional
>>>>>> tests were added to validate assumptions about the currently traced
>>>>>> syscall. Also adds early asserts for running on older kernels to avoid
>>>>>> noise when the seccomp syscall is not implemented.
>>>>>>
>>>>>> Signed-off-by: Kees Cook <keescook@chromium.org>
>>>>>> ---
>>>>
>>>>
>>>> Sorry looks like it got lost in my LinuxCon backlog. I will queue it
>>>> up for 4.3-rc1.
>>>>
>>>
>>> Hmm. It doesn't apply to linux-kselftest next. Based on your comment
>>>
>>> "This applies on top of -next, following the addition of the powerpc
>>> tests."
>>
>> Well, it was -next when I sent it. :) The powerpc changes are now in
>> Linus's tree, so there should be no problem. Is your linux-kselftest
>> merged with 4.3-rc yet?
>>
> 
> Kees,
> 
> I will get this into 4.3-rc2 as this patch fixes failure paths with
> other fixes I am planning to send in rc2. Thanks for the ping on this.
> 

Applied to linux-kselftest fixes for 4.3-rc2

thanks,
-- Shuah


-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1222054

FromMichael Ellerman <mpe@ellerman.id.au>
Date2015-09-10 11:20 +0200
Message-ID<q76bo-5y3-11@gated-at.bofh.it>
In reply to#1221583
On Wed, 2015-09-09 at 10:23 -0700, Kees Cook wrote:
> On Wed, Sep 9, 2015 at 9:56 AM, Shuah Khan <shuahkh@osg.samsung.com> wrote:
> > On 09/09/2015 10:45 AM, Shuah Khan wrote:
> >> On 09/09/2015 10:40 AM, Kees Cook wrote:
> >>> On Fri, Aug 21, 2015 at 11:22 AM, Kees Cook <keescook@chromium.org> wrote:
> >>>> This adds support for s390 to the seccomp selftests. Some improvements
> >>>> were made to enhance the accuracy of failure reporting, and additional
> >>>> tests were added to validate assumptions about the currently traced
> >>>> syscall. Also adds early asserts for running on older kernels to avoid
> >>>> noise when the seccomp syscall is not implemented.
> >>
> >> Sorry looks like it got lost in my LinuxCon backlog. I will queue it
> >> up for 4.3-rc1.
> >
> > Hmm. It doesn't apply to linux-kselftest next. Based on your comment
> >
> > "This applies on top of -next, following the addition of the powerpc
> > tests."
> 
> Well, it was -next when I sent it. :) The powerpc changes are now in
> Linus's tree, so there should be no problem. Is your linux-kselftest
> merged with 4.3-rc yet?

Right, it applies cleanly to Linus' tree as of ~now (b8889c4fc6ba).

So it should be fine to go via the selftest tree as a fix for rc1.

cheers


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web