rfc: https://lore.kernel.org/lkml/20200616074934.1600036-1-keescook@xxxxxxxxxxxx/ alternative: https://lore.kernel.org/containers/cover.1600661418.git.yifeifz2@xxxxxxxxxxxx/ v1: - rebase to for-next/seccomp - finish X86_X32 support for both pinning and bitmaps - replace TLB magic with Jann's emulator - add JSET insn TODO: - add ALU|AND insn - significantly more testing Hi, This is a refresh of my earlier constant action bitmap series. It looks like the RFC was missed on the container list, so I've CCed it now. :) I'd like to work from this series, as it handles the multi-architecture stuff. Repeating the commit log from patch 3: seccomp: Implement constant action bitmaps One of the most common pain points with seccomp filters has been dealing with the overhead of processing the filters, especially for "always allow" or "always reject" cases. While BPF is extremely fast[1], it will always have overhead associated with it. Additionally, due to seccomp's design, filters are layered, which means processing time goes up as the number of filters attached goes up. In the past, efforts have been focused on making filter execution complete in a shorter amount of time. For example, filters were rewritten from using linear if/then/else syscall search to using balanced binary trees, or moving tests for syscalls common to the process's workload to the front of the filter. However, there are limits to this, especially when some processes are dealing with tens of filters[2], or when some architectures have a less efficient BPF engine[3]. The most common use of seccomp, constructing syscall block/allow-lists, where syscalls that are always allowed or always rejected (without regard to any arguments), also tends to produce the most pathological runtime problems, in that a large number of syscall checks in the filter need to be performed to come to a determination. In order to optimize these cases from O(n) to O(1), seccomp can use bitmaps to immediately determine the desired action. A critical observation in the prior paragraph bears repeating: the common case for syscall tests do not check arguments. For any given filter, there is a constant mapping from the combination of architecture and syscall to the seccomp action result. (For kernels/architectures without CONFIG_COMPAT, there is a single architecture.). As such, it is possible to construct a mapping of arch/syscall to action, which can be updated as new filters are attached to a process. In order to build this mapping at filter attach time, each filter is executed for every syscall (under each possible architecture), and checked for any accesses of struct seccomp_data that are not the "arch" nor "nr" (syscall) members. If only "arch" and "nr" are examined, then there is a constant mapping for that syscall, and bitmaps can be updated accordingly. If any accesses happen outside of those struct members, seccomp must not bypass filter execution for that syscall, since program state will be used to determine filter action result. (This logic comes in the next patch.) [1] https://lore.kernel.org/bpf/20200531171915.wsxvdjeetmhpsdv2@xxxxxxxxxxxxxxxxxxxxxxxxxxxx/ [2] https://lore.kernel.org/bpf/20200601101137.GA121847@gardel-login/ [3] https://lore.kernel.org/bpf/717a06e7f35740ccb4c70470ec70fb2f@xxxxxxxxxx/ Thanks! -Kees Kees Cook (6): seccomp: Introduce SECCOMP_PIN_ARCHITECTURE x86: Enable seccomp architecture tracking seccomp: Implement constant action bitmaps seccomp: Emulate basic filters for constant action results selftests/seccomp: Compare bitmap vs filter overhead [DEBUG] seccomp: Report bitmap coverage ranges arch/x86/include/asm/seccomp.h | 14 + include/linux/seccomp.h | 27 + include/uapi/linux/seccomp.h | 1 + kernel/seccomp.c | 473 +++++++++++++++++- net/core/filter.c | 3 +- .../selftests/seccomp/seccomp_benchmark.c | 151 +++++- tools/testing/selftests/seccomp/seccomp_bpf.c | 33 ++ tools/testing/selftests/seccomp/settings | 2 +- 8 files changed, 674 insertions(+), 30 deletions(-) -- 2.25.1
