1                SECure COMPuting with filters
   2                =============================
   7A large number of system calls are exposed to every userland process
   8with many of them going unused for the entire lifetime of the process.
   9As system calls change and mature, bugs are found and eradicated.  A
  10certain subset of userland applications benefit by having a reduced set
  11of available system calls.  The resulting set reduces the total kernel
  12surface exposed to the application.  System call filtering is meant for
  13use with those applications.
  15Seccomp filtering provides a means for a process to specify a filter for
  16incoming system calls.  The filter is expressed as a Berkeley Packet
  17Filter (BPF) program, as with socket filters, except that the data
  18operated on is related to the system call being made: system call
  19number and the system call arguments.  This allows for expressive
  20filtering of system calls using a filter program language with a long
  21history of being exposed to userland and a straightforward data set.
  23Additionally, BPF makes it impossible for users of seccomp to fall prey
  24to time-of-check-time-of-use (TOCTOU) attacks that are common in system
  25call interposition frameworks.  BPF programs may not dereference
  26pointers which constrains all filters to solely evaluating the system
  27call arguments directly.
  29What it isn't
  32System call filtering isn't a sandbox.  It provides a clearly defined
  33mechanism for minimizing the exposed kernel surface.  It is meant to be
  34a tool for sandbox developers to use.  Beyond that, policy for logical
  35behavior and information flow should be managed with a combination of
  36other system hardening techniques and, potentially, an LSM of your
  37choosing.  Expressive, dynamic filters provide further options down this
  38path (avoiding pathological sizes or selecting which of the multiplexed
  39system calls in socketcall() is allowed, for instance) which could be
  40construed, incorrectly, as a more complete sandboxing solution.
  45An additional seccomp mode is added and is enabled using the same
  46prctl(2) call as the strict seccomp.  If the architecture has
  47CONFIG_HAVE_ARCH_SECCOMP_FILTER, then filters may be added as below:
  50        Now takes an additional argument which specifies a new filter
  51        using a BPF program.
  52        The BPF program will be executed over struct seccomp_data
  53        reflecting the system call number, arguments, and other
  54        metadata.  The BPF program must then return one of the
  55        acceptable values to inform the kernel which action should be
  56        taken.
  58        Usage:
  59                prctl(PR_SET_SECCOMP, SECCOMP_MODE_FILTER, prog);
  61        The 'prog' argument is a pointer to a struct sock_fprog which
  62        will contain the filter program.  If the program is invalid, the
  63        call will return -1 and set errno to EINVAL.
  65        If fork/clone and execve are allowed by @prog, any child
  66        processes will be constrained to the same filters and system
  67        call ABI as the parent.
  69        Prior to use, the task must call prctl(PR_SET_NO_NEW_PRIVS, 1) or
  70        run with CAP_SYS_ADMIN privileges in its namespace.  If these are not
  71        true, -EACCES will be returned.  This requirement ensures that filter
  72        programs cannot be applied to child processes with greater privileges
  73        than the task that installed them.
  75        Additionally, if prctl(2) is allowed by the attached filter,
  76        additional filters may be layered on which will increase evaluation
  77        time, but allow for further decreasing the attack surface during
  78        execution of a process.
  80The above call returns 0 on success and non-zero on error.
  82Return values
  84A seccomp filter may return any of the following values. If multiple
  85filters exist, the return value for the evaluation of a given system
  86call will always use the highest precedent value. (For example,
  87SECCOMP_RET_KILL will always take precedence.)
  89In precedence order, they are:
  92        Results in the task exiting immediately without executing the
  93        system call.  The exit status of the task (status & 0x7f) will
  94        be SIGSYS, not SIGKILL.
  97        Results in the kernel sending a SIGSYS signal to the triggering
  98        task without executing the system call.  siginfo->si_call_addr
  99        will show the address of the system call instruction, and
 100        siginfo->si_syscall and siginfo->si_arch will indicate which
 101        syscall was attempted.  The program counter will be as though
 102        the syscall happened (i.e. it will not point to the syscall
 103        instruction).  The return value register will contain an arch-
 104        dependent value -- if resuming execution, set it to something
 105        sensible.  (The architecture dependency is because replacing
 106        it with -ENOSYS could overwrite some useful information.)
 108        The SECCOMP_RET_DATA portion of the return value will be passed
 109        as si_errno.
 111        SIGSYS triggered by seccomp will have a si_code of SYS_SECCOMP.
 114        Results in the lower 16-bits of the return value being passed
 115        to userland as the errno without executing the system call.
 118        When returned, this value will cause the kernel to attempt to
 119        notify a ptrace()-based tracer prior to executing the system
 120        call.  If there is no tracer present, -ENOSYS is returned to
 121        userland and the system call is not executed.
 123        A tracer will be notified if it requests PTRACE_O_TRACESECCOMP
 124        using ptrace(PTRACE_SETOPTIONS).  The tracer will be notified
 125        of a PTRACE_EVENT_SECCOMP and the SECCOMP_RET_DATA portion of
 126        the BPF program return value will be available to the tracer
 127        via PTRACE_GETEVENTMSG.
 129        The tracer can skip the system call by changing the syscall number
 130        to -1.  Alternatively, the tracer can change the system call
 131        requested by changing the system call to a valid syscall number.  If
 132        the tracer asks to skip the system call, then the system call will
 133        appear to return the value that the tracer puts in the return value
 134        register.
 136        The seccomp check will not be run again after the tracer is
 137        notified.  (This means that seccomp-based sandboxes MUST NOT
 138        allow use of ptrace, even of other sandboxed processes, without
 139        extreme care; ptracers can use this mechanism to escape.)
 142        Results in the system call being executed.
 144If multiple filters exist, the return value for the evaluation of a
 145given system call will always use the highest precedent value.
 147Precedence is only determined using the SECCOMP_RET_ACTION mask.  When
 148multiple filters return values of the same precedence, only the
 149SECCOMP_RET_DATA from the most recently installed filter will be
 155The biggest pitfall to avoid during use is filtering on system call
 156number without checking the architecture value.  Why?  On any
 157architecture that supports multiple system call invocation conventions,
 158the system call numbers may vary based on the specific invocation.  If
 159the numbers in the different calling conventions overlap, then checks in
 160the filters may be abused.  Always check the arch value!
 165The samples/seccomp/ directory contains both an x86-specific example
 166and a more generic example of a higher level macro interface for BPF
 167program generation.
 171Adding architecture support
 174See arch/Kconfig for the authoritative requirements.  In general, if an
 175architecture supports both ptrace_event and seccomp, it will be able to
 176support seccomp filter with minor fixup: SIGSYS support and seccomp return
 177value checking.  Then it must just add CONFIG_HAVE_ARCH_SECCOMP_FILTER
 178to its arch-specific Kconfig.
 185The vDSO can cause some system calls to run entirely in userspace,
 186leading to surprises when you run programs on different machines that
 187fall back to real syscalls.  To minimize these surprises on x86, make
 188sure you test with
 189/sys/devices/system/clocksource/clocksource0/current_clocksource set to
 190something like acpi_pm.
 192On x86-64, vsyscall emulation is enabled by default.  (vsyscalls are
 193legacy variants on vDSO calls.)  Currently, emulated vsyscalls will honor seccomp, with a few oddities:
 195- A return value of SECCOMP_RET_TRAP will set a si_call_addr pointing to
 196  the vsyscall entry for the given call and not the address after the
 197  'syscall' instruction.  Any code which wants to restart the call
 198  should be aware that (a) a ret instruction has been emulated and (b)
 199  trying to resume the syscall will again trigger the standard vsyscall
 200  emulation security checks, making resuming the syscall mostly
 201  pointless.
 203- A return value of SECCOMP_RET_TRACE will signal the tracer as usual,
 204  but the syscall may not be changed to another system call using the
 205  orig_rax register. It may only be changed to -1 order to skip the
 206  currently emulated call. Any other change MAY terminate the process.
 207  The rip value seen by the tracer will be the syscall entry address;
 208  this is different from normal behavior.  The tracer MUST NOT modify
 209  rip or rsp.  (Do not rely on other changes terminating the process.
 210  They might work.  For example, on some kernels, choosing a syscall
 211  that only exists in future kernels will be correctly emulated (by
 212  returning -ENOSYS).
 214To detect this quirky behavior, check for addr & ~0x0C00 ==
 2150xFFFFFFFFFF600000.  (For SECCOMP_RET_TRACE, use rip.  For
 216SECCOMP_RET_TRAP, use siginfo->si_call_addr.)  Do not check any other
 217condition: future kernels may improve vsyscall emulation and current
 218kernels in vsyscall=native mode will behave differently, but the
 219instructions at 0xF...F600{0,4,8,C}00 will not be system calls in these
 222Note that modern systems are unlikely to use vsyscalls at all -- they
 223are a legacy feature and they are considerably slower than standard
 224syscalls.  New code will use the vDSO, and vDSO-issued system calls
 225are indistinguishable from normal system calls.
 226 kindly hosted by Redpill Linpro AS, provider of Linux consulting and operations services since 1995.