From: Will Drewry <wad@chromium.org>
To: linux-kernel@vger.kernel.org
Cc: keescook@chromium.org, john.johansen@canonical.com,
serge.hallyn@canonical.com, coreyb@linux.vnet.ibm.com,
pmoore@redhat.com, eparis@redhat.com, djm@mindrot.org,
torvalds@linux-foundation.org, segoon@openwall.com,
rostedt@goodmis.org, jmorris@namei.org, scarybeasts@gmail.com,
avi@redhat.com, penberg@cs.helsinki.fi, viro@zeniv.linux.org.uk,
wad@chromium.org, luto@mit.edu, mingo@elte.hu,
akpm@linux-foundation.org, khilman@ti.com,
borislav.petkov@amd.com, amwang@redhat.com, oleg@redhat.com,
ak@linux.intel.com, eric.dumazet@gmail.com, gregkh@suse.de,
dhowells@redhat.com, daniel.lezcano@free.fr,
linux-fsdevel@vger.kernel.org,
linux-security-module@vger.kernel.org, olofj@chromium.org,
mhalcrow@google.com, dlaor@redhat.com
Subject: [RFC,PATCH 2/2] Documentation: prctl/seccomp_filter
Date: Wed, 11 Jan 2012 11:25:10 -0600 [thread overview]
Message-ID: <1326302710-9427-3-git-send-email-wad@chromium.org> (raw)
In-Reply-To: <1326302710-9427-1-git-send-email-wad@chromium.org>
Document how system call filtering with BPF works
and can be used.
Signed-off-by: Will Drewry <wad@chromium.org>
---
Documentation/prctl/seccomp_filter.txt | 159 ++++++++++++++++++++++++++++++++
1 files changed, 159 insertions(+), 0 deletions(-)
create mode 100644 Documentation/prctl/seccomp_filter.txt
diff --git a/Documentation/prctl/seccomp_filter.txt b/Documentation/prctl/seccomp_filter.txt
new file mode 100644
index 0000000..5fb3f44
--- /dev/null
+++ b/Documentation/prctl/seccomp_filter.txt
@@ -0,0 +1,159 @@
+ Seccomp filtering
+ =================
+
+Introduction
+------------
+
+A large number of system calls are exposed to every userland process
+with many of them going unused for the entire lifetime of the process.
+As system calls change and mature, bugs are found and eradicated. A
+certain subset of userland applications benefit by having a reduced set
+of available system calls. The resulting set reduces the total kernel
+surface exposed to the application. System call filtering is meant for
+use with those applications.
+
+Seccomp filtering provides a means for a process to specify a filter
+for incoming system calls. The filter is expressed as a Berkeley Packet
+Filter program, as with socket filters, except that the data operated on
+is the current user_regs_struct. This allows for expressive filtering
+of system calls using the pre-existing system call ABI and using a filter
+program language with a long history of being exposed to userland.
+Additionally, BPF makes it impossible for users of seccomp to fall prey to
+time-of-check-time-of-use (TOCTOU) attacks that are common in system call
+interposition frameworks because the evaluated data is solely register state
+just after system call entry.
+
+What it isn't
+-------------
+
+System call filtering isn't a sandbox. It provides a clearly defined
+mechanism for minimizing the exposed kernel surface. Beyond that,
+policy for logical behavior and information flow should be managed with
+a combinations of other system hardening techniques and, potentially, a
+LSM of your choosing. Expressive, dynamic filters provide further options down
+this path (avoiding pathological sizes or selecting which of the multiplexed
+system calls in socketcall() is allowed, for instance) which could be
+construed, incorrectly, as a more complete sandboxing solution.
+
+Usage
+-----
+
+An additional seccomp mode is added, but they are not directly set by the
+consuming process. The new mode, '2', is only available if
+CONFIG_SECCOMP_FILTER is set and enabled using prctl with the
+PR_ATTACH_SECCOMP_FILTER argument.
+
+Interacting with seccomp filters is done using one prctl(2) call.
+
+PR_ATTACH_SECCOMP_FILTER:
+ Allows the specification of a new filter using a BPF program.
+ The BPF program will be executed over a user_regs_struct data
+ reflecting system call time except with the system call number
+ resident in orig_[register]. To allow a system call, the size
+ of the data must be returned. At present, all other return values
+ result in the system call being blocked, but it is recommended to
+ return 0 in those cases. This will allow for future custom return
+ values to be introduced, if ever desired.
+
+ Usage:
+ prctl(PR_ATTACH_SECCOMP_FILTER, prog);
+
+ The 'prog' argument is a pointer to a struct sock_fprog which will
+ contain the filter program. If the program is invalid, the call
+ will return -1 and set errno to -EINVAL.
+
+ The struct user_regs_struct the @prog will see is based on the
+ personality of the task at the time of this prctl call. Additionally,
+ is_compat_task is also tracked for the @prog. This means that once set
+ the calling task will have all of its system calls blocked if it
+ switches its system call ABI (via personality or other means).
+
+ If the @prog is installed while the task has CAP_SYS_ADMIN in its user
+ namespace, the @prog will be marked as inheritable across execve. Any
+ inherited filters are still subject to the system call ABI constraints
+ above and any ABI mismatched system calls will result in process death.
+
+All of the above calls return 0 on success and non-zero on error.
+
+
+Example
+-------
+
+Assume a process would like to cleanly read and write to stdin/out/err and exit
+cleanly. Without using a BPF compiler, it may be done as follows on x86 32-bit:
+
+#include <asm/unistd.h>
+#include <linux/filter.h>
+#include <stdio.h>
+#include <stddef.h>
+#include <sys/user.h>
+#include <unistd.h>
+
+#define regoffset(_reg) (offsetof(struct user_regs_struct, _reg))
+int install_filter(void)
+{
+ struct sock_filter filter[] = {
+ /* Grab the system call number */
+ BPF_STMT(BPF_LD+BPF_W+BPF_IND, regoffset(orig_eax)),
+ /* Jump table for the allowed syscalls */
+ BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, __NR_rt_sigreturn, 10, 0),
+ BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, __NR_sigreturn, 9, 0),
+ BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, __NR_exit_group, 8, 0),
+ BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, __NR_exit, 7, 0),
+ BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, __NR_read, 1, 0),
+ BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, __NR_write, 2, 6),
+
+ /* Check that read is only using stdin. */
+ BPF_STMT(BPF_LD+BPF_W+BPF_IND, regoffset(ebx)),
+ BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, STDIN_FILENO, 3, 4),
+
+ /* Check that write is only using stdout/stderr */
+ BPF_STMT(BPF_LD+BPF_W+BPF_IND, regoffset(ebx)),
+ BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, STDOUT_FILENO, 1, 0),
+ BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, STDERR_FILENO, 0, 1),
+
+ /* Put the "accept" value in A */
+ BPF_STMT(BPF_LD+BPF_W+BPF_LEN, 0),
+
+ BPF_STMT(BPF_RET+BPF_A,0),
+ };
+ struct sock_fprog prog = {
+ .len = (unsigned short)(sizeof(filter)/sizeof(filter[0])),
+ .filter = filter,
+ };
+ if (prctl(36, &prog)) {
+ perror("prctl");
+ return 1;
+ }
+ return 0;
+}
+
+#define payload(_c) _c, sizeof(_c)
+int main(int argc, char **argv) {
+ char buf[4096];
+ ssize_t bytes = 0;
+ if (install_filter())
+ return 1;
+ syscall(__NR_write, STDOUT_FILENO, payload("OHAI! WHAT IS YOUR NAME? "));
+ bytes = syscall(__NR_read, STDIN_FILENO, buf, sizeof(buf));
+ syscall(__NR_write, STDOUT_FILENO, payload("HELLO, "));
+ syscall(__NR_write, STDOUT_FILENO, buf, bytes);
+ return 0;
+}
+
+Additionally, if prctl(2) is allowed by the installed filter, additional
+filters may be layered on which will increase evaluation time, but allow for
+further decreasing the attack surface during execution of a process.
+
+
+Caveats
+-------
+
+- execve will fail unless the most recently attached filter was installed by
+ a process with CAP_SYS_ADMIN (in its namespace).
+
+Adding architecture support
+-----------------------
+
+Any platform with seccomp support will support seccomp filters
+as long as CONFIG_SECCOMP_FILTER is enabled.
--
1.7.5.4
next prev parent reply other threads:[~2012-01-11 17:26 UTC|newest]
Thread overview: 222+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-01-11 17:25 [RFC,PATCH 0/2] dynamic seccomp policies (using BPF filters) Will Drewry
2012-01-11 17:25 ` [RFC,PATCH 1/2] seccomp_filters: system call filtering using BPF Will Drewry
2012-01-12 8:53 ` Serge Hallyn
2012-01-12 16:54 ` Will Drewry
2012-01-12 14:50 ` Oleg Nesterov
2012-01-12 16:55 ` Will Drewry
2012-01-12 15:43 ` Steven Rostedt
2012-01-12 16:14 ` Oleg Nesterov
2012-01-12 16:38 ` Steven Rostedt
2012-01-12 16:47 ` Oleg Nesterov
2012-01-12 17:08 ` Will Drewry
2012-01-12 17:30 ` Jamie Lokier
2012-01-12 17:40 ` Steven Rostedt
2012-01-12 17:44 ` Jamie Lokier
2012-01-12 17:56 ` Steven Rostedt
2012-01-12 23:27 ` Alan Cox
2012-01-12 23:38 ` Linus Torvalds
2012-01-12 22:18 ` Will Drewry
2012-01-12 23:00 ` Andrew Lutomirski
2012-01-12 16:14 ` Andrew Lutomirski
2012-01-12 16:27 ` Steven Rostedt
2012-01-12 16:51 ` Andrew Lutomirski
2012-01-12 17:09 ` Linus Torvalds
2012-01-12 17:17 ` Steven Rostedt
2012-01-12 18:18 ` Andrew Lutomirski
2012-01-12 18:32 ` Linus Torvalds
2012-01-12 18:44 ` Andrew Lutomirski
2012-01-12 19:08 ` Kyle Moffett
2012-01-12 23:05 ` Eric Paris
2012-01-12 23:33 ` Andrew Lutomirski
2012-01-12 19:40 ` Will Drewry
2012-01-12 19:42 ` Will Drewry
2012-01-12 19:46 ` Andrew Lutomirski
2012-01-12 20:00 ` Linus Torvalds
2012-01-12 16:59 ` Will Drewry
2012-01-12 17:22 ` Jamie Lokier
2012-01-12 17:35 ` Will Drewry
2012-01-12 17:57 ` Jamie Lokier
2012-01-12 18:03 ` Will Drewry
2012-01-13 1:34 ` Jamie Lokier
2012-01-13 6:33 ` Chris Evans
2012-01-12 17:36 ` Jamie Lokier
2012-01-12 16:18 ` Alan Cox
2012-01-12 17:03 ` Will Drewry
2012-01-12 17:11 ` Alan Cox
2012-01-12 17:52 ` Will Drewry
2012-01-13 1:31 ` James Morris
2012-01-12 16:22 ` Oleg Nesterov
2012-01-12 17:10 ` Will Drewry
2012-01-12 17:23 ` Oleg Nesterov
2012-01-12 17:51 ` Will Drewry
2012-01-13 17:31 ` Oleg Nesterov
2012-01-13 19:01 ` Will Drewry
2012-01-13 23:10 ` Will Drewry
2012-01-13 23:12 ` Will Drewry
2012-01-13 23:30 ` Eric Paris
2012-01-16 18:37 ` Oleg Nesterov
2012-01-16 20:15 ` Will Drewry
2012-01-17 16:45 ` Oleg Nesterov
2012-01-17 16:56 ` Will Drewry
2012-01-17 17:01 ` Andrew Lutomirski
2012-01-17 17:05 ` Oleg Nesterov
2012-01-17 17:45 ` Andrew Lutomirski
2012-01-18 0:56 ` Compat 32-bit syscall entry from 64-bit task!? [was: Re: [RFC,PATCH 1/2] seccomp_filters: system call filtering using BPF] Indan Zupancic
2012-01-18 1:01 ` Andrew Lutomirski
2012-01-19 1:06 ` Indan Zupancic
2012-01-19 1:19 ` Andrew Lutomirski
2012-01-19 1:47 ` Indan Zupancic
2012-01-18 1:07 ` Roland McGrath
2012-01-18 1:47 ` Indan Zupancic
2012-01-18 1:48 ` Jamie Lokier
2012-01-18 1:50 ` Andi Kleen
2012-01-18 2:00 ` Steven Rostedt
2012-01-18 2:04 ` Jamie Lokier
2012-01-18 2:22 ` Andi Kleen
2012-01-18 2:25 ` Andrew Lutomirski
2012-01-18 4:22 ` Indan Zupancic
2012-01-18 5:23 ` Linus Torvalds
2012-01-18 6:25 ` Linus Torvalds
2012-01-18 13:12 ` Compat 32-bit syscall entry from 64-bit task!? Indan Zupancic
2012-01-18 19:31 ` Linus Torvalds
2012-01-18 19:36 ` Andi Kleen
2012-01-18 19:39 ` Linus Torvalds
2012-01-18 19:44 ` Andi Kleen
2012-01-18 19:47 ` Linus Torvalds
2012-01-18 19:52 ` Will Drewry
2012-01-18 19:58 ` Will Drewry
2012-01-18 19:41 ` Martin Mares
2012-01-18 19:38 ` Andrew Lutomirski
2012-01-19 16:01 ` Jamie Lokier
2012-01-19 16:13 ` Andrew Lutomirski
2012-01-19 19:21 ` Linus Torvalds
2012-01-19 19:30 ` Andrew Lutomirski
2012-01-19 19:37 ` Linus Torvalds
2012-01-19 19:41 ` Linus Torvalds
2012-01-19 23:54 ` Jamie Lokier
2012-01-20 0:05 ` Linus Torvalds
2012-01-20 15:35 ` Will Drewry
2012-01-20 17:56 ` Roland McGrath
2012-01-20 19:45 ` Will Drewry
2012-01-18 20:26 ` Linus Torvalds
2012-01-18 20:55 ` H. Peter Anvin
2012-01-18 21:01 ` Linus Torvalds
2012-01-18 21:04 ` H. Peter Anvin
2012-01-18 21:21 ` H. Peter Anvin
2012-01-18 21:51 ` Roland McGrath
2012-01-18 21:53 ` H. Peter Anvin
2012-01-18 23:28 ` Linus Torvalds
2012-01-19 0:38 ` H. Peter Anvin
2012-01-20 21:51 ` Denys Vlasenko
2012-01-20 22:40 ` Roland McGrath
2012-01-20 22:41 ` H. Peter Anvin
2012-01-20 23:49 ` Indan Zupancic
2012-01-20 23:55 ` Roland McGrath
2012-01-20 23:58 ` hpanvin@gmail.com
2012-01-23 2:14 ` Indan Zupancic
2012-01-21 0:07 ` Denys Vlasenko
2012-01-21 0:10 ` Roland McGrath
2012-01-21 1:23 ` Jamie Lokier
2012-01-23 2:37 ` Indan Zupancic
2012-01-23 16:48 ` Oleg Nesterov
2012-01-24 8:19 ` Indan Zupancic
2012-02-06 20:30 ` H. Peter Anvin
2012-02-06 20:39 ` Roland McGrath
2012-02-06 20:42 ` H. Peter Anvin
2012-01-18 21:26 ` Linus Torvalds
2012-01-18 21:30 ` H. Peter Anvin
2012-01-18 21:42 ` Linus Torvalds
2012-01-18 21:47 ` H. Peter Anvin
2012-01-19 1:45 ` Indan Zupancic
2012-01-19 2:16 ` H. Peter Anvin
2012-02-06 8:32 ` Indan Zupancic
2012-02-06 17:02 ` H. Peter Anvin
2012-02-07 1:52 ` Indan Zupancic
2012-02-09 0:19 ` H. Peter Anvin
2012-02-09 4:20 ` Indan Zupancic
2012-02-09 4:29 ` H. Peter Anvin
2012-02-09 6:03 ` Indan Zupancic
2012-02-09 14:47 ` H. Peter Anvin
2012-02-09 16:00 ` H.J. Lu
2012-02-10 1:09 ` Indan Zupancic
2012-02-10 1:15 ` H. Peter Anvin
2012-02-10 2:29 ` Indan Zupancic
2012-02-10 2:47 ` H. Peter Anvin
[not found] ` <cc95fcf4b1c28ee6f73e373d04593634.squirrel@webmail.greenhost.nl>
2012-02-10 15:53 ` H. Peter Anvin
2012-02-10 22:42 ` Indan Zupancic
2012-02-10 22:56 ` H. Peter Anvin
2012-02-12 12:07 ` Indan Zupancic
2012-01-25 19:36 ` Oleg Nesterov
2012-01-25 20:20 ` Pedro Alves
2012-01-25 23:36 ` Denys Vlasenko
2012-01-25 23:32 ` Denys Vlasenko
2012-01-26 0:40 ` Indan Zupancic
2012-01-26 1:08 ` Jamie Lokier
2012-01-26 1:22 ` Denys Vlasenko
2012-01-26 6:34 ` Indan Zupancic
2012-01-26 10:31 ` Jamie Lokier
2012-01-26 10:40 ` Denys Vlasenko
2012-01-26 11:01 ` Jamie Lokier
2012-01-26 14:02 ` Denys Vlasenko
2012-01-26 11:19 ` Indan Zupancic
2012-01-26 11:20 ` Indan Zupancic
2012-01-26 11:47 ` Jamie Lokier
2012-01-26 14:05 ` Denys Vlasenko
2012-01-27 7:23 ` Indan Zupancic
2012-02-10 2:02 ` Jamie Lokier
2012-02-10 3:37 ` Indan Zupancic
2012-02-10 21:19 ` Denys Vlasenko
2012-01-26 1:09 ` Denys Vlasenko
2012-01-26 3:47 ` Linus Torvalds
2012-01-26 18:03 ` Denys Vlasenko
2017-03-08 23:41 ` Dmitry V. Levin
2017-03-09 4:39 ` Andrew Lutomirski
2017-03-14 2:57 ` Dmitry V. Levin
2012-01-26 5:57 ` Indan Zupancic
2012-01-26 0:59 ` Jamie Lokier
2012-01-26 1:21 ` Denys Vlasenko
2012-01-26 8:23 ` Pedro Alves
2012-01-26 8:53 ` Denys Vlasenko
2012-01-26 9:51 ` Pedro Alves
2012-01-26 18:44 ` Oleg Nesterov
2012-02-10 2:51 ` Jamie Lokier
2012-01-18 15:04 ` Compat 32-bit syscall entry from 64-bit task!? [was: Re: [RFC,PATCH 1/2] seccomp_filters: system call filtering using BPF] Eric Paris
2012-01-18 17:51 ` Linus Torvalds
2012-01-18 5:43 ` Chris Evans
2012-01-18 12:12 ` Indan Zupancic
2012-01-18 21:13 ` Chris Evans
2012-01-19 0:14 ` Indan Zupancic
2012-01-19 8:16 ` Chris Evans
2012-01-19 11:34 ` Indan Zupancic
2012-01-19 16:11 ` Jamie Lokier
2012-01-19 15:40 ` Jamie Lokier
2012-01-18 17:00 ` Oleg Nesterov
2012-01-18 17:12 ` Oleg Nesterov
2012-01-18 21:09 ` Chris Evans
2012-01-23 16:56 ` Oleg Nesterov
2012-01-23 22:23 ` Chris Evans
2012-02-07 11:45 ` Indan Zupancic
2012-01-19 0:29 ` Indan Zupancic
2012-01-18 2:27 ` Linus Torvalds
2012-01-18 2:31 ` Andi Kleen
2012-01-18 2:46 ` Linus Torvalds
2012-01-18 14:06 ` Martin Mares
2012-01-18 18:24 ` Andi Kleen
2012-01-19 16:04 ` Jamie Lokier
2012-01-20 0:21 ` Indan Zupancic
2012-01-20 0:53 ` Linus Torvalds
2012-01-20 2:02 ` Indan Zupancic
2012-01-17 17:06 ` [RFC,PATCH 1/2] seccomp_filters: system call filtering using BPF Will Drewry
2012-01-17 19:35 ` Will Drewry
2012-01-12 17:02 ` Andrew Lutomirski
2012-01-16 20:28 ` Will Drewry
2012-01-11 17:25 ` Will Drewry [this message]
2012-01-11 20:03 ` [RFC,PATCH 2/2] Documentation: prctl/seccomp_filter Jonathan Corbet
2012-01-11 20:10 ` Will Drewry
2012-01-11 23:19 ` [PATCH v2 " Will Drewry
2012-01-12 0:29 ` Will Drewry
2012-01-12 18:16 ` Randy Dunlap
2012-01-12 17:23 ` Will Drewry
2012-01-12 17:34 ` Steven Rostedt
2012-01-12 13:13 ` [RFC,PATCH " Łukasz Sowa
2012-01-12 17:25 ` Will Drewry
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1326302710-9427-3-git-send-email-wad@chromium.org \
--to=wad@chromium.org \
--cc=ak@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=amwang@redhat.com \
--cc=avi@redhat.com \
--cc=borislav.petkov@amd.com \
--cc=coreyb@linux.vnet.ibm.com \
--cc=daniel.lezcano@free.fr \
--cc=dhowells@redhat.com \
--cc=djm@mindrot.org \
--cc=dlaor@redhat.com \
--cc=eparis@redhat.com \
--cc=eric.dumazet@gmail.com \
--cc=gregkh@suse.de \
--cc=jmorris@namei.org \
--cc=john.johansen@canonical.com \
--cc=keescook@chromium.org \
--cc=khilman@ti.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-security-module@vger.kernel.org \
--cc=luto@mit.edu \
--cc=mhalcrow@google.com \
--cc=mingo@elte.hu \
--cc=oleg@redhat.com \
--cc=olofj@chromium.org \
--cc=penberg@cs.helsinki.fi \
--cc=pmoore@redhat.com \
--cc=rostedt@goodmis.org \
--cc=scarybeasts@gmail.com \
--cc=segoon@openwall.com \
--cc=serge.hallyn@canonical.com \
--cc=torvalds@linux-foundation.org \
--cc=viro@zeniv.linux.org.uk \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).