Hello Kees,
On Fri, Sep 21, 2012 at 1:37 AM, Kees Cook <[email protected]> wrote:
> This adds a short summary of the arguments used for "mode 2" (BPF)
> seccomp.
Thanks very much. Applied.
Cheers,
Michael
> Signed-off-by: Kees Cook <[email protected]>
> ---
> man2/prctl.2 | 40 ++++++++++++++++++++++++++++++----------
> 1 files changed, 30 insertions(+), 10 deletions(-)
>
> diff --git a/man2/prctl.2 b/man2/prctl.2
> index eb53aa8..a44aa32 100644
> --- a/man2/prctl.2
> +++ b/man2/prctl.2
> @@ -41,11 +41,12 @@
> .\" 2012-04 Cyrill Gorcunov, Document PR_SET_MM
> .\" 2012-04-25 Michael Kerrisk, Document PR_TASK_PERF_EVENTS_DISABLE and
> .\" PR_TASK_PERF_EVENTS_ENABLE
> +.\" 2012-09-20 Kees Cook, update PR_SET_SECCOMP for mode 2
> .\"
> .\" FIXME: Document PR_SET_TIMERSLACK and PR_GET_TIMERSLACK (new in 2.6.28)
> .\" commit 6976675d94042fbd446231d1bd8b7de71a980ada
> .\"
> -.TH PRCTL 2 2012-04-23 "Linux" "Linux Programmer's Manual"
> +.TH PRCTL 2 2012-09-20 "Linux" "Linux Programmer's Manual"
> .SH NAME
> prctl \- operations on a process
> .SH SYNOPSIS
> @@ -235,11 +236,17 @@ in the location pointed to by
> .\" See http://thread.gmane.org/gmane.linux.kernel/542632
> .\" [PATCH 0 of 2] seccomp updates
> .\" [email protected]
> -Set the secure computing mode for the calling thread.
> -In the current implementation,
> +Set the secure computing mode for the calling thread, to limit
> +the available syscalls.
> +The seccomp mode is selected via
> +.IR arg2 .
> +(The seccomp constants are defined in
> +.IR <linux/seccomp.h> .)
> +
> +With
> .IR arg2
> -must be 1.
> -After the secure computing mode has been set to 1,
> +set to
> +.BR SECCOMP_MODE_STRICT
> the only system calls that the thread is permitted to make are
> .BR read (2),
> .BR write (2),
> @@ -249,22 +256,35 @@ and
> Other system calls result in the delivery of a
> .BR SIGKILL
> signal.
> -Secure computing mode is useful for number-crunching applications
> +Strict secure computing mode is useful for number-crunching applications
> that may need to execute untrusted byte code,
> perhaps obtained by reading from a pipe or socket.
> This operation is only available
> if the kernel is configured with CONFIG_SECCOMP enabled.
> +
> +With
> +.IR arg2
> +set to
> +.BR SECCOMP_MODE_FILTER " (since Linux 3.5)"
> +the system calls allowed are
> +defined by a pointer passed in
> +.IR arg3
> +to a Berkeley Packet Filter, which can be designed to filter
> +arbitrary syscalls and syscall arguments.
> +This mode is only available if the kernel is configured with
> +CONFIG_SECCOMP_FILTER enabled.
> +
> .TP
> .BR PR_GET_SECCOMP " (since Linux 2.6.23)"
> Return the secure computing mode of the calling thread.
> -Not very useful for the current implementation (mode equals 1),
> -but may be useful for other possible future modes:
> -if the caller is not in secure computing mode, this operation returns 0;
> -if the caller is in secure computing mode, then the
> +If the caller is not in secure computing mode, this operation returns 0;
> +if the caller is in strict secure computing mode, then the
> .BR prctl ()
> call will cause a
> .B SIGKILL
> signal to be sent to the process.
> +If the caller is in filter mode, and this syscall is allowed by the
> +seccomp filters, it returns 2.
> This operation is only available
> if the kernel is configured with CONFIG_SECCOMP enabled.
> .TP
> --
> 1.7.0.4
>
> --
> Kees Cook
> Chrome OS Security
--
Michael Kerrisk
Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
Author of "The Linux Programming Interface"; http://man7.org/tlpi/
Kees,
A couple of questions about SECCOMP_MODE_FILTER.
I added some words that the arg3 is a pointer to 'struct fprog'. Can
you confirmn that's correct?
If the CONFIG_SECCOMP_FILTER permits fork(), is the seccomp setting
inherited across fork()? Similar question for execve().
Thanks,
Michael
On Mon, Oct 22, 2012 at 6:56 AM, Michael Kerrisk (man-pages)
<[email protected]> wrote:
> Hello Kees,
>
> On Fri, Sep 21, 2012 at 1:37 AM, Kees Cook <[email protected]> wrote:
>> This adds a short summary of the arguments used for "mode 2" (BPF)
>> seccomp.
>
> Thanks very much. Applied.
>
> Cheers,
>
> Michael
>
>
>> Signed-off-by: Kees Cook <[email protected]>
>> ---
>> man2/prctl.2 | 40 ++++++++++++++++++++++++++++++----------
>> 1 files changed, 30 insertions(+), 10 deletions(-)
>>
>> diff --git a/man2/prctl.2 b/man2/prctl.2
>> index eb53aa8..a44aa32 100644
>> --- a/man2/prctl.2
>> +++ b/man2/prctl.2
>> @@ -41,11 +41,12 @@
>> .\" 2012-04 Cyrill Gorcunov, Document PR_SET_MM
>> .\" 2012-04-25 Michael Kerrisk, Document PR_TASK_PERF_EVENTS_DISABLE and
>> .\" PR_TASK_PERF_EVENTS_ENABLE
>> +.\" 2012-09-20 Kees Cook, update PR_SET_SECCOMP for mode 2
>> .\"
>> .\" FIXME: Document PR_SET_TIMERSLACK and PR_GET_TIMERSLACK (new in 2.6.28)
>> .\" commit 6976675d94042fbd446231d1bd8b7de71a980ada
>> .\"
>> -.TH PRCTL 2 2012-04-23 "Linux" "Linux Programmer's Manual"
>> +.TH PRCTL 2 2012-09-20 "Linux" "Linux Programmer's Manual"
>> .SH NAME
>> prctl \- operations on a process
>> .SH SYNOPSIS
>> @@ -235,11 +236,17 @@ in the location pointed to by
>> .\" See http://thread.gmane.org/gmane.linux.kernel/542632
>> .\" [PATCH 0 of 2] seccomp updates
>> .\" [email protected]
>> -Set the secure computing mode for the calling thread.
>> -In the current implementation,
>> +Set the secure computing mode for the calling thread, to limit
>> +the available syscalls.
>> +The seccomp mode is selected via
>> +.IR arg2 .
>> +(The seccomp constants are defined in
>> +.IR <linux/seccomp.h> .)
>> +
>> +With
>> .IR arg2
>> -must be 1.
>> -After the secure computing mode has been set to 1,
>> +set to
>> +.BR SECCOMP_MODE_STRICT
>> the only system calls that the thread is permitted to make are
>> .BR read (2),
>> .BR write (2),
>> @@ -249,22 +256,35 @@ and
>> Other system calls result in the delivery of a
>> .BR SIGKILL
>> signal.
>> -Secure computing mode is useful for number-crunching applications
>> +Strict secure computing mode is useful for number-crunching applications
>> that may need to execute untrusted byte code,
>> perhaps obtained by reading from a pipe or socket.
>> This operation is only available
>> if the kernel is configured with CONFIG_SECCOMP enabled.
>> +
>> +With
>> +.IR arg2
>> +set to
>> +.BR SECCOMP_MODE_FILTER " (since Linux 3.5)"
>> +the system calls allowed are
>> +defined by a pointer passed in
>> +.IR arg3
>> +to a Berkeley Packet Filter, which can be designed to filter
>> +arbitrary syscalls and syscall arguments.
>> +This mode is only available if the kernel is configured with
>> +CONFIG_SECCOMP_FILTER enabled.
>> +
>> .TP
>> .BR PR_GET_SECCOMP " (since Linux 2.6.23)"
>> Return the secure computing mode of the calling thread.
>> -Not very useful for the current implementation (mode equals 1),
>> -but may be useful for other possible future modes:
>> -if the caller is not in secure computing mode, this operation returns 0;
>> -if the caller is in secure computing mode, then the
>> +If the caller is not in secure computing mode, this operation returns 0;
>> +if the caller is in strict secure computing mode, then the
>> .BR prctl ()
>> call will cause a
>> .B SIGKILL
>> signal to be sent to the process.
>> +If the caller is in filter mode, and this syscall is allowed by the
>> +seccomp filters, it returns 2.
>> This operation is only available
>> if the kernel is configured with CONFIG_SECCOMP enabled.
>> .TP
>> --
>> 1.7.0.4
>>
>> --
>> Kees Cook
>> Chrome OS Security
>
>
>
> --
> Michael Kerrisk
> Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
> Author of "The Linux Programming Interface"; http://man7.org/tlpi/
--
Michael Kerrisk
Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
Author of "The Linux Programming Interface"; http://man7.org/tlpi/
On Mon, Oct 22, 2012 at 12:34 AM, Michael Kerrisk (man-pages)
<[email protected]> wrote:
> Kees,
>
> A couple of questions about SECCOMP_MODE_FILTER.
>
> I added some words that the arg3 is a pointer to 'struct fprog'. Can
> you confirmn that's correct?
Correct. Good idea to add this detail.
> If the CONFIG_SECCOMP_FILTER permits fork(), is the seccomp setting
> inherited across fork()? Similar question for execve().
Yes for both. Additionally, the filters are cumulative. (If the
filters allows prctl, additional filters can be appended; they are run
in order until the first non-allow result is seen.)
-Kees
--
Kees Cook
Chrome OS Security
>> If the CONFIG_SECCOMP_FILTER permits fork(), is the seccomp setting
>> inherited across fork()? Similar question for execve().
>
> Yes for both. Additionally, the filters are cumulative. (If the
> filters allows prctl, additional filters can be appended; they are run
> in order until the first non-allow result is seen.)
Thanks. I'll add some words to the man page.
Cheers,
Michael
--
Michael Kerrisk
Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
Author of "The Linux Programming Interface"; http://man7.org/tlpi/