Skip to content

L
linux
Switch branch/tag
  1. 31 Mar, 2014 8 commits
    • Alexei Starovoitov's avatar
      net: filter: rework/optimize internal BPF interpreter's instruction set · bd4cf0ed
      Alexei Starovoitov authored 
      This patch replaces/reworks the kernel-internal BPF interpreter with
an optimized BPF instruction set format that is modelled closer to
mimic native instruction sets and is designed to be JITed with one to
one mapping. Thus, the new interpreter is noticeably faster than the
current implementation of sk_run_filter(); mainly for two reasons:

1. Fall-through jumps:

  BPF jump instructions are forced to go either 'true' or 'false'
  branch which causes branch-miss penalty. The new BPF jump
  instructions have only one branch and fall-through otherwise,
  which fits the CPU branch predictor logic better. `perf stat`
  shows drastic difference for branch-misses between the old and
  new code.

2. Jump-threaded implementation of interpreter vs switch
   statement:

  Instead of single table-jump at the top of 'switch' statement,
  gcc will now generate multiple table-jump instructions, which
  helps CPU branch predictor logic.

Note that the verification of filters is still being done through
sk_chk_filter() in classical BPF format, so filters from user- or
kernel space are verified in the same way as we do now, and same
restrictions/constraints hold as well.

We reuse current BPF JIT compilers in a way that this upgrade would
even be fine as is, but nevertheless allows for a successive upgrade
of BPF JIT compilers to the new format.

The internal instruction set migration is being done after the
probing for JIT compilation, so in case JIT compilers are able to
create a native opcode image, we're going to use that, and in all
other cases we're doing a follow-up migration of the BPF program's
instruction set, so that it can be transparently run in the new
interpreter.

In short, the *internal* format extends BPF in the following way (more
details can be taken from the appended documentation):

  - Number of registers increase from 2 to 10
  - Register width increases from 32-bit to 64-bit
  - Conditional jt/jf targets replaced with jt/fall-through
  - Adds signed > and >= insns
  - 16 4-byte stack slots for register spill-fill replaced
    with up to 512 bytes of multi-use stack space
  - Introduction of bpf_call insn and register passing convention
    for zero overhead calls from/to other kernel functions
  - Adds arithmetic right shift and endianness conversion insns
  - Adds atomic_add insn
  - Old tax/txa insns are replaced with 'mov dst,src' insn

Performance of two BPF filters generated by libpcap resp. bpf_asm
was measured on x86_64, i386 and arm32 (other libpcap programs
have similar performance differences):

fprog #1 is taken from Documentation/networking/filter.txt:
tcpdump -i eth0 port 22 -dd

fprog #2 is taken from 'man tcpdump':
tcpdump -i eth0 'tcp port 22 and (((ip[2:2] - ((ip[0]&0xf)<<2)) -
   ((tcp[12]&0xf0)>>2)) != 0)' -dd

Raw performance data from BPF micro-benchmark: SK_RUN_FILTER on the
same SKB (cache-hit) or 10k SKBs (cache-miss); time in ns per call,
smaller is better:

--x86_64--
         fprog #1  fprog #1   fprog #2  fprog #2
         cache-hit cache-miss cache-hit cache-miss
old BPF      90       101        192       202
new BPF      31        71         47        97
old BPF jit  12        34         17        44
new BPF jit TBD

--i386--
         fprog #1  fprog #1   fprog #2  fprog #2
         cache-hit cache-miss cache-hit cache-miss
old BPF     107       136        227       252
new BPF      40       119         69       172

--arm32--
         fprog #1  fprog #1   fprog #2  fprog #2
         cache-hit cache-miss cache-hit cache-miss
old BPF     202       300        475       540
new BPF     180       270        330       470
old BPF jit  26       182         37       202
new BPF jit TBD

Thus, without changing any userland BPF filters, applications on
top of AF_PACKET (or other families) such as libpcap/tcpdump, cls_bpf
classifier, netfilter's xt_bpf, team driver's load-balancing mode,
and many more will have better interpreter filtering performance.

While we are replacing the internal BPF interpreter, we also need
to convert seccomp BPF in the same step to make use of the new
internal structure since it makes use of lower-level API details
without being further decoupled through higher-level calls like
sk_unattached_filter_{create,destroy}(), for example.

Just as for normal socket filtering, also seccomp BPF experiences
a time-to-verdict speedup:

05-sim-long_jumps.c of libseccomp was used as micro-benchmark:

  seccomp_rule_add_exact(ctx,...
  seccomp_rule_add_exact(ctx,...

  rc = seccomp_load(ctx);

  for (i = 0; i < 10000000; i++)
     syscall(199, 100);

'short filter' has 2 rules
'large filter' has 200 rules

'short filter' performance is slightly better on x86_64/i386/arm32
'large filter' is much faster on x86_64 and i386 and shows no
               difference on arm32

--x86_64-- short filter
old BPF: 2.7 sec
 39.12%  bench  libc-2.15.so       [.] syscall
  8.10%  bench  [kernel.kallsyms]  [k] sk_run_filter
  6.31%  bench  [kernel.kallsyms]  [k] system_call
  5.59%  bench  [kernel.kallsyms]  [k] trace_hardirqs_on_caller
  4.37%  bench  [kernel.kallsyms]  [k] trace_hardirqs_off_caller
  3.70%  bench  [kernel.kallsyms]  [k] __secure_computing
  3.67%  bench  [kernel.kallsyms]  [k] lock_is_held
  3.03%  bench  [kernel.kallsyms]  [k] seccomp_bpf_load
new BPF: 2.58 sec
 42.05%  bench  libc-2.15.so       [.] syscall
  6.91%  bench  [kernel.kallsyms]  [k] system_call
  6.25%  bench  [kernel.kallsyms]  [k] trace_hardirqs_on_caller
  6.07%  bench  [kernel.kallsyms]  [k] __secure_computing
  5.08%  bench  [kernel.kallsyms]  [k] sk_run_filter_int_seccomp

--arm32-- short filter
old BPF: 4.0 sec
 39.92%  bench  [kernel.kallsyms]  [k] vector_swi
 16.60%  bench  [kernel.kallsyms]  [k] sk_run_filter
 14.66%  bench  libc-2.17.so       [.] syscall
  5.42%  bench  [kernel.kallsyms]  [k] seccomp_bpf_load
  5.10%  bench  [kernel.kallsyms]  [k] __secure_computing
new BPF: 3.7 sec
 35.93%  bench  [kernel.kallsyms]  [k] vector_swi
 21.89%  bench  libc-2.17.so       [.] syscall
 13.45%  bench  [kernel.kallsyms]  [k] sk_run_filter_int_seccomp
  6.25%  bench  [kernel.kallsyms]  [k] __secure_computing
  3.96%  bench  [kernel.kallsyms]  [k] syscall_trace_exit

--x86_64-- large filter
old BPF: 8.6 seconds
    73.38%    bench  [kernel.kallsyms]  [k] sk_run_filter
    10.70%    bench  libc-2.15.so       [.] syscall
     5.09%    bench  [kernel.kallsyms]  [k] seccomp_bpf_load
     1.97%    bench  [kernel.kallsyms]  [k] system_call
new BPF: 5.7 seconds
    66.20%    bench  [kernel.kallsyms]  [k] sk_run_filter_int_seccomp
    16.75%    bench  libc-2.15.so       [.] syscall
     3.31%    bench  [kernel.kallsyms]  [k] system_call
     2.88%    bench  [kernel.kallsyms]  [k] __secure_computing

--i386-- large filter
old BPF: 5.4 sec
new BPF: 3.8 sec

--arm32-- large filter
old BPF: 13.5 sec
 73.88%  bench  [kernel.kallsyms]  [k] sk_run_filter
 10.29%  bench  [kernel.kallsyms]  [k] vector_swi
  6.46%  bench  libc-2.17.so       [.] syscall
  2.94%  bench  [kernel.kallsyms]  [k] seccomp_bpf_load
  1.19%  bench  [kernel.kallsyms]  [k] __secure_computing
  0.87%  bench  [kernel.kallsyms]  [k] sys_getuid
new BPF: 13.5 sec
 76.08%  bench  [kernel.kallsyms]  [k] sk_run_filter_int_seccomp
 10.98%  bench  [kernel.kallsyms]  [k] vector_swi
  5.87%  bench  libc-2.17.so       [.] syscall
  1.77%  bench  [kernel.kallsyms]  [k] __secure_computing
  0.93%  bench  [kernel.kallsyms]  [k] sys_getuid

BPF filters generated by seccomp are very branchy, so the new
internal BPF performance is better than the old one. Performance
gains will be even higher when BPF JIT is committed for the
new structure, which is planned in future work (as successive
JIT migrations).

BPF has also been stress-tested with trinity's BPF fuzzer.

Joint work with Daniel Borkmann.
Signed-off-by: default avatarAlexei Starovoitov <ast@plumgrid.com>
Signed-off-by: default avatarDaniel Borkmann <dborkman@redhat.com>
Cc: Hagen Paul Pfeifer <hagen@jauu.net>
Cc: Kees Cook <keescook@chromium.org>
Cc: Paul Moore <pmoore@redhat.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: H. Peter Anvin <hpa@linux.intel.com>
Cc: linux-kernel@vger.kernel.org
Acked-by: default avatarKees Cook <keescook@chromium.org>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      bd4cf0ed
    • Daniel Borkmann's avatar
      net: isdn: use sk_unattached_filter api · 77e0114a
      Daniel Borkmann authored 
      Similarly as in ppp, we need to migrate the ISDN/PPP code to make use
of the sk_unattached_filter api in order to decouple having direct
filter structure access. By using sk_unattached_filter_{create,destroy},
we can allow for the possibility to jit compile filters for faster
filter verdicts as well.

Joint work with Alexei Starovoitov.
Signed-off-by: default avatarDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: default avatarAlexei Starovoitov <ast@plumgrid.com>
Cc: Karsten Keil <isdn@linux-pingi.de>
Cc: isdn4linux@listserv.isdn4linux.de
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      77e0114a
    • Daniel Borkmann's avatar
      net: ppp: use sk_unattached_filter api · 568f194e
      Daniel Borkmann authored 
      For the ppp driver, there are currently two open-coded BPF filters in use,
that is, pass_filter and active_filter. Migrate both to make proper use
of sk_unattached_filter_{create,destroy} API so that the actual BPF code
is decoupled from direct access, and filters can be jited as a side-effect
by the internal filter compiler.

Joint work with Alexei Starovoitov.
Signed-off-by: default avatarDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: default avatarAlexei Starovoitov <ast@plumgrid.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: linux-ppp@vger.kernel.org
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      568f194e
    • Daniel Borkmann's avatar
      net: ptp: do not reimplement PTP/BPF classifier · 164d8c66
      Daniel Borkmann authored 
      There are currently pch_gbe, cpts, and ixp4xx_eth drivers that open-code
and reimplement a BPF classifier for the PTP protocol. Since all of them
effectively do the very same thing and load the very same PTP/BPF filter,
we can just consolidate that code by introducing ptp_classify_raw() in
the time-stamping core framework which can be used in drivers.

As drivers get initialized after bootstrapping the core networking
subsystem, they can make use of ptp_insns wrapped through
ptp_classify_raw(), which allows to simplify and remove PTP classifier
setup code in drivers.

Joint work with Alexei Starovoitov.
Signed-off-by: default avatarDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: default avatarAlexei Starovoitov <ast@plumgrid.com>
Cc: Richard Cochran <richard.cochran@omicron.at>
Cc: Jiri Benc <jbenc@redhat.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      164d8c66
    • Daniel Borkmann's avatar
      net: ptp: use sk_unattached_filter_create() for BPF · e62d2df0
      Daniel Borkmann authored 
      This patch migrates an open-coded sk_run_filter() implementation with
proper use of the BPF API, that is, sk_unattached_filter_create(). This
migration is needed, as we will be internally transforming the filter
to a different representation, and therefore needs to be decoupled.

It is okay to do so as skb_timestamping_init() is called during
initialization of the network stack in core initcall via sock_init().
This would effectively also allow for PTP filters to be jit compiled if
bpf_jit_enable is set.

For better readability, there are also some newlines introduced, also
ptp_classify.h is only in kernel space.

Joint work with Alexei Starovoitov.
Signed-off-by: default avatarDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: default avatarAlexei Starovoitov <ast@plumgrid.com>
Cc: Richard Cochran <richard.cochran@omicron.at>
Cc: Jiri Benc <jbenc@redhat.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      e62d2df0
    • Daniel Borkmann's avatar
      net: filter: move filter accounting to filter core · fbc907f0
      Daniel Borkmann authored 
      This patch basically does two things, i) removes the extern keyword
from the include/linux/filter.h file to be more consistent with the
rest of Joe's changes, and ii) moves filter accounting into the filter
core framework.

Filter accounting mainly done through sk_filter_{un,}charge() take
care of the case when sockets are being cloned through sk_clone_lock()
so that removal of the filter on one socket won't result in eviction
as it's still referenced by the other.

These functions actually belong to net/core/filter.c and not
include/net/sock.h as we want to keep all that in a central place.
It's also not in fast-path so uninlining them is fine and even allows
us to get rd of sk_filter_release_rcu()'s EXPORT_SYMBOL and a forward
declaration.

Joint work with Alexei Starovoitov.
Signed-off-by: default avatarDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: default avatarAlexei Starovoitov <ast@plumgrid.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      fbc907f0
    • Daniel Borkmann's avatar
      net: filter: keep original BPF program around · a3ea269b
      Daniel Borkmann authored 
      In order to open up the possibility to internally transform a BPF program
into an alternative and possibly non-trivial reversible representation, we
need to keep the original BPF program around, so that it can be passed back
to user space w/o the need of a complex decoder.

The reason for that use case resides in commit a8fc9277 ("sk-filter:
Add ability to get socket filter program (v2)"), that is, the ability
to retrieve the currently attached BPF filter from a given socket used
mainly by the checkpoint-restore project, for example.

Therefore, we add two helpers sk_{store,release}_orig_filter for taking
care of that. In the sk_unattached_filter_create() case, there's no such
possibility/requirement to retrieve a loaded BPF program. Therefore, we
can spare us the work in that case.

This approach will simplify and slightly speed up both, sk_get_filter()
and sock_diag_put_filterinfo() handlers as we won't need to successively
decode filters anymore through sk_decode_filter(). As we still need
sk_decode_filter() later on, we're keeping it around.

Joint work with Alexei Starovoitov.
Signed-off-by: default avatarAlexei Starovoitov <ast@plumgrid.com>
Signed-off-by: default avatarDaniel Borkmann <dborkman@redhat.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      a3ea269b
    • Daniel Borkmann's avatar
      net: filter: add jited flag to indicate jit compiled filters · f8bbbfc3
      Daniel Borkmann authored 
      This patch adds a jited flag into sk_filter struct in order to indicate
whether a filter is currently jited or not. The size of sk_filter is
not being expanded as the 32 bit 'len' member allows upper bits to be
reused since a filter can currently only grow as large as BPF_MAXINSNS.

Therefore, there's enough room also for other in future needed flags to
reuse 'len' field if necessary. The jited flag also allows for having
alternative interpreter functions running as currently, we can only
detect jit compiled filters by testing fp->bpf_func to not equal the
address of sk_run_filter().

Joint work with Alexei Starovoitov.
Signed-off-by: default avatarAlexei Starovoitov <ast@plumgrid.com>
Signed-off-by: default avatarDaniel Borkmann <dborkman@redhat.com>
Cc: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      f8bbbfc3
  3. 29 Mar, 2014 22 commits
  5. 28 Mar, 2014 10 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7