Skip to content

K
Kernel

openeuler / Kernel
1 年多 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 27 9月, 2017 1 次提交
  3. 29 8月, 2017 1 次提交
  5. 20 8月, 2017 1 次提交
  7. 17 8月, 2017 1 次提交
    • J
      bpf: sockmap sample program · 69e8cc13
      由 John Fastabend 提交于 
      This program binds a program to a cgroup and then matches hard
coded IP addresses and adds these to a sockmap.

This will receive messages from the backend and send them to
the client.

     client:X <---> frontend:10000 client:X <---> backend:10001

To keep things simple this is only designed for 1:1 connections
using hard coded values. A more complete example would allow many
backends and clients.

To run,

 # sockmap <cgroup2_dir>
Signed-off-by: NJohn Fastabend <john.fastabend@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      69e8cc13
  9. 10 8月, 2017 1 次提交
    • D
      bpf: add BPF_J{LT,LE,SLT,SLE} instructions · 92b31a9a
      由 Daniel Borkmann 提交于 
      Currently, eBPF only understands BPF_JGT (>), BPF_JGE (>=),
BPF_JSGT (s>), BPF_JSGE (s>=) instructions, this means that
particularly *JLT/*JLE counterparts involving immediates need
to be rewritten from e.g. X < [IMM] by swapping arguments into
[IMM] > X, meaning the immediate first is required to be loaded
into a register Y := [IMM], such that then we can compare with
Y > X. Note that the destination operand is always required to
be a register.

This has the downside of having unnecessarily increased register
pressure, meaning complex program would need to spill other
registers temporarily to stack in order to obtain an unused
register for the [IMM]. Loading to registers will thus also
affect state pruning since we need to account for that register
use and potentially those registers that had to be spilled/filled
again. As a consequence slightly more stack space might have
been used due to spilling, and BPF programs are a bit longer
due to extra code involving the register load and potentially
required spill/fills.

Thus, add BPF_JLT (<), BPF_JLE (<=), BPF_JSLT (s<), BPF_JSLE (s<=)
counterparts to the eBPF instruction set. Modifying LLVM to
remove the NegateCC() workaround in a PoC patch at [1] and
allowing it to also emit the new instructions resulted in
cilium's BPF programs that are injected into the fast-path to
have a reduced program length in the range of 2-3% (e.g.
accumulated main and tail call sections from one of the object
file reduced from 4864 to 4729 insns), reduced complexity in
the range of 10-30% (e.g. accumulated sections reduced in one
of the cases from 116432 to 88428 insns), and reduced stack
usage in the range of 1-5% (e.g. accumulated sections from one
of the object files reduced from 824 to 784b).

The modification for LLVM will be incorporated in a backwards
compatible way. Plan is for LLVM to have i) a target specific
option to offer a possibility to explicitly enable the extension
by the user (as we have with -m target specific extensions today
for various CPU insns), and ii) have the kernel checked for
presence of the extensions and enable them transparently when
the user is selecting more aggressive options such as -march=native
in a bpf target context. (Other frontends generating BPF byte
code, e.g. ply can probe the kernel directly for its code
generation.)

  [1] https://github.com/borkmann/llvm/tree/bpf-insnsSigned-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      92b31a9a
  11. 09 8月, 2017 1 次提交
  13. 01 8月, 2017 1 次提交
    • A
      tools headers: Fixup tools/include/uapi/linux/bpf.h copy of kernel ABI header · f1d6cb2d
      由 Arnaldo Carvalho de Melo 提交于 
      In 04df41e3 ("bpf: update tools/include/uapi/linux/bpf.h") the files
added in 40304b2a ("bpf: BPF support for sock_ops") were added to
tools/include, but not in a verbatim way, missing the comments, which
ends up triggering this warning when build tools/perf/:

  make: Entering directory '/home/acme/git/linux/tools/perf'
    BUILD:   Doing 'make -j4' parallel build
  Warning: Kernel ABI header at 'tools/include/uapi/linux/bpf.h' differs from latest version at 'include/uapi/linux/bpf.h'

Make sure the the lines are equal, to fix the simple header copy
drift detector in tools/perf/.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: David Ahern <dsahern@gmail.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Lawrence Brakmo <brakmo@fb.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Fixes: 04df41e3 ("bpf: update tools/include/uapi/linux/bpf.h")
Link: http://lkml.kernel.org/n/tip-z9qyyqht9qq3yyxu76sfy0dh@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      f1d6cb2d
  15. 31 7月, 2017 1 次提交
  17. 30 7月, 2017 1 次提交
  19. 03 7月, 2017 1 次提交
  21. 02 7月, 2017 1 次提交
  23. 11 6月, 2017 1 次提交
  25. 07 6月, 2017 1 次提交
  27. 05 6月, 2017 1 次提交
  29. 12 5月, 2017 1 次提交
  31. 22 4月, 2017 1 次提交
  33. 10 4月, 2017 1 次提交
  35. 02 4月, 2017 1 次提交
  37. 24 3月, 2017 2 次提交
  39. 23 3月, 2017 1 次提交
  41. 15 2月, 2017 1 次提交
  43. 11 2月, 2017 1 次提交
    • M
      tools: Sync {,tools/}include/uapi/linux/bpf.h · 9a738266
      由 Mickaël Salaün 提交于 
      The tools version of this header is out of date; update it to the latest
version from kernel header.

Synchronize with the following commits:
* b95a5c4d ("bpf: add a longest prefix match trie map implementation")
* a5e8c070 ("bpf: add bpf_probe_read_str helper")
* d1b662ad ("bpf: allow option for setting bpf_l4_csum_replace from scratch")
Signed-off-by: NMickaël Salaün <mic@digikod.net>
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Daniel Mack <daniel@zonque.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: Gianluca Borello <g.borello@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      9a738266
  45. 16 12月, 2016 1 次提交
  47. 06 10月, 2016 1 次提交
  49. 09 8月, 2016 1 次提交
    • A
      toops: Sync tools/include/uapi/linux/bpf.h with the kernel · 791cceb8
      由 Arnaldo Carvalho de Melo 提交于 
      The way we're using kernel headers in tools/ now, with a copy that is
made to the same path prefixed by "tools/" plus checking if that copy
got stale, i.e. if the kernel counterpart changed, helps in keeping
track with new features that may be useful for tools to exploit.

For instance, looking at all the changes to bpf.h since it was last
copied to tools/include brings this to toolers' attention:

Need to investigate this one to check how to run a program via perf, setting up
a BPF event, that will take advantage of the way perf already calls clang/LLVM,
sets up the event and runs the workload in a single command line, helping in
debugging such semi cooperative programs:

  96ae5227 ("bpf: Add bpf_probe_write_user BPF helper to be called in tracers")

This one needs further investigation about using the feature it improves
in 'perf trace' to do some tcpdumpin' mixed with syscalls, tracepoints,
probe points, callgraphs, etc:

  555c8a86 ("bpf: avoid stack copy and use skb ctx for event output")

Add tracing just packets that are related to some container to that mix:

  4a482f34 ("cgroup: bpf: Add bpf_skb_in_cgroup_proto")
  4ed8ec52 ("cgroup: bpf: Add BPF_MAP_TYPE_CGROUP_ARRAY")

Definetely needs to have example programs accessing task_struct from a bpf proggie
started from 'perf trace':

  606274c5 ("bpf: introduce bpf_get_current_task() helper")

Core networking related, XDP:

  6ce96ca3 ("bpf: add XDP_TX xdp_action for direct forwarding")
  6a773a15 ("bpf: add XDP prog type for early driver filter")
  13c5c240 ("bpf: add bpf_get_hash_recalc helper")
  d2485c42 ("bpf: add bpf_skb_change_type helper")
  6578171a ("bpf: add bpf_skb_change_proto helper")

Changes detected by the tools build system:

  $ make -C tools/perf O=/tmp/build/perf install-bin
  make: Entering directory '/home/acme/git/linux/tools/perf'
    BUILD:   Doing 'make -j4' parallel build
  Warning: tools/include/uapi/linux/bpf.h differs from kernel
    INSTALL  GTK UI
    CC       /tmp/build/perf/bench/mem-memcpy-x86-64-asm.o
  <SNIP>
  $

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Brenden Blanco <bblanco@plumgrid.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: David Ahern <dsahern@gmail.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Sargun Dhillon <sargun@sargun.me>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-difq4ts1xvww6eyfs9e7zlft@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
      791cceb8
  51. 13 7月, 2016 1 次提交
  53. 09 7月, 2016 1 次提交
  55. 05 7月, 2016 1 次提交
  57. 02 7月, 2016 2 次提交
  59. 30 6月, 2016 3 次提交
    • D
      bpf: add bpf_skb_change_type helper · d2485c42
      由 Daniel Borkmann 提交于 
      This work adds a helper for changing skb->pkt_type in a controlled way.
We only allow a subset of possible values and can extend that in future
should other use cases come up. Doing this as a helper has the advantage
that errors can be handeled gracefully and thus helper kept extensible.

It's a write counterpart to pkt_type member we can already read from
struct __sk_buff context. Major use case is to change incoming skbs to
PACKET_HOST in a programmatic way instead of having to recirculate via
redirect(..., BPF_F_INGRESS), for example.
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      d2485c42
    • D
      bpf: add bpf_skb_change_proto helper · 6578171a
      由 Daniel Borkmann 提交于 
      This patch adds a minimal helper for doing the groundwork of changing
the skb->protocol in a controlled way. Currently supported is v4 to
v6 and vice versa transitions, which allows f.e. for a minimal, static
nat64 implementation where applications in containers that still
require IPv4 can be transparently operated in an IPv6-only environment.
For example, host facing veth of the container can transparently do
the transitions in a programmatic way with the help of clsact qdisc
and cls_bpf.

Idea is to separate concerns for keeping complexity of the helper
lower, which means that the programs utilize bpf_skb_change_proto(),
bpf_skb_store_bytes() and bpf_lX_csum_replace() to get the job done,
instead of doing everything in a single helper (and thus partially
duplicating helper functionality). Also, bpf_skb_change_proto()
shouldn't need to deal with raw packet data as this is done by other
helpers.

bpf_skb_proto_6_to_4() and bpf_skb_proto_4_to_6() unclone the skb to
operate on a private one, push or pop additionally required header
space and migrate the gso/gro meta data from the shared info. We do
mark the gso type as dodgy so that headers are checked and segs
recalculated by the gso/gro engine. The gso_size target is adapted
as well. The flags argument added is currently reserved and can be
used for future extensions.
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      6578171a
    • D
      bpf, trace: add BPF_F_CURRENT_CPU flag for bpf_perf_event_read · 6816a7ff
      由 Daniel Borkmann 提交于 
      Follow-up commit to 1e33759c ("bpf, trace: add BPF_F_CURRENT_CPU
flag for bpf_perf_event_output") to add the same functionality into
bpf_perf_event_read() helper. The split of index into flags and index
component is also safe here, since such large maps are rejected during
map allocation time.
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      6816a7ff
  61. 07 5月, 2016 1 次提交
    • A
      bpf: direct packet access · 969bf05e
      由 Alexei Starovoitov 提交于 
      Extended BPF carried over two instructions from classic to access
packet data: LD_ABS and LD_IND. They're highly optimized in JITs,
but due to their design they have to do length check for every access.
When BPF is processing 20M packets per second single LD_ABS after JIT
is consuming 3% cpu. Hence the need to optimize it further by amortizing
the cost of 'off < skb_headlen' over multiple packet accesses.
One option is to introduce two new eBPF instructions LD_ABS_DW and LD_IND_DW
with similar usage as skb_header_pointer().
The kernel part for interpreter and x64 JIT was implemented in [1], but such
new insns behave like old ld_abs and abort the program with 'return 0' if
access is beyond linear data. Such hidden control flow is hard to workaround
plus changing JITs and rolling out new llvm is incovenient.

Therefore allow cls_bpf/act_bpf program access skb->data directly:
int bpf_prog(struct __sk_buff *skb)
{
  struct iphdr *ip;

  if (skb->data + sizeof(struct iphdr) + ETH_HLEN > skb->data_end)
      /* packet too small */
      return 0;

  ip = skb->data + ETH_HLEN;

  /* access IP header fields with direct loads */
  if (ip->version != 4 || ip->saddr == 0x7f000001)
      return 1;
  [...]
}

This solution avoids introduction of new instructions. llvm stays
the same and all JITs stay the same, but verifier has to work extra hard
to prove safety of the above program.

For XDP the direct store instructions can be allowed as well.

The skb->data is NET_IP_ALIGNED, so for common cases the verifier can check
the alignment. The complex packet parsers where packet pointer is adjusted
incrementally cannot be tracked for alignment, so allow byte access in such cases
and misaligned access on architectures that define efficient_unaligned_access

[1] https://git.kernel.org/cgit/linux/kernel/git/ast/bpf.git/?h=ld_abs_dwSigned-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      969bf05e
  63. 20 4月, 2016 1 次提交
  65. 08 4月, 2016 1 次提交
    • A
      perf, bpf: allow bpf programs attach to tracepoints · 98b5c2c6
      由 Alexei Starovoitov 提交于 
      introduce BPF_PROG_TYPE_TRACEPOINT program type and allow it to be attached
to the perf tracepoint handler, which will copy the arguments into
the per-cpu buffer and pass it to the bpf program as its first argument.
The layout of the fields can be discovered by doing
'cat /sys/kernel/debug/tracing/events/sched/sched_switch/format'
prior to the compilation of the program with exception that first 8 bytes
are reserved and not accessible to the program. This area is used to store
the pointer to 'struct pt_regs' which some of the bpf helpers will use:
+---------+
| 8 bytes | hidden 'struct pt_regs *' (inaccessible to bpf program)
+---------+
| N bytes | static tracepoint fields defined in tracepoint/format (bpf readonly)
+---------+
| dynamic | __dynamic_array bytes of tracepoint (inaccessible to bpf yet)
+---------+

Not that all of the fields are already dumped to user space via perf ring buffer
and broken application access it directly without consulting tracepoint/format.
Same rule applies here: static tracepoint fields should only be accessed
in a format defined in tracepoint/format. The order of fields and
field sizes are not an ABI.
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      98b5c2c6
  67. 31 3月, 2016 1 次提交
    • D
      bpf: make padding in bpf_tunnel_key explicit · c0e760c9
      由 Daniel Borkmann 提交于 
      Make the 2 byte padding in struct bpf_tunnel_key between tunnel_ttl
and tunnel_label members explicit. No issue has been observed, and
gcc/llvm does padding for the old struct already, where tunnel_label
was not yet present, so the current code works, but since it's part
of uapi, make sure we don't introduce holes in structs.

Therefore, add tunnel_ext that we can use generically in future
(f.e. to flag OAM messages for backends, etc). Also add the offset
to the compat tests to be sure should some compilers not padd the
tail of the old version of bpf_tunnel_key.

Fixes: 4018ab18 ("bpf: support flow label for bpf_skb_{set, get}_tunnel_key")
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      c0e760c9
  69. 12 3月, 2016 1 次提交
  71. 09 3月, 2016 1 次提交
    • A
      bpf: pre-allocate hash map elements · 6c905981
      由 Alexei Starovoitov 提交于 
      If kprobe is placed on spin_unlock then calling kmalloc/kfree from
bpf programs is not safe, since the following dead lock is possible:
kfree->spin_lock(kmem_cache_node->lock)...spin_unlock->kprobe->
bpf_prog->map_update->kmalloc->spin_lock(of the same kmem_cache_node->lock)
and deadlocks.

The following solutions were considered and some implemented, but
eventually discarded
- kmem_cache_create for every map
- add recursion check to slow-path of slub
- use reserved memory in bpf_map_update for in_irq or in preempt_disabled
- kmalloc via irq_work

At the end pre-allocation of all map elements turned out to be the simplest
solution and since the user is charged upfront for all the memory, such
pre-allocation doesn't affect the user space visible behavior.

Since it's impossible to tell whether kprobe is triggered in a safe
location from kmalloc point of view, use pre-allocation by default
and introduce new BPF_F_NO_PREALLOC flag.

While testing of per-cpu hash maps it was discovered
that alloc_percpu(GFP_ATOMIC) has odd corner cases and often
fails to allocate memory even when 90% of it is free.
The pre-allocation of per-cpu hash elements solves this problem as well.

Turned out that bpf_map_update() quickly followed by
bpf_map_lookup()+bpf_map_delete() is very common pattern used
in many of iovisor/bcc/tools, so there is additional benefit of
pre-allocation, since such use cases are must faster.

Since all hash map elements are now pre-allocated we can remove
atomic increment of htab->count and save few more cycles.

Also add bpf_map_precharge_memlock() to check rlimit_memlock early to avoid
large malloc/free done by users who don't have sufficient limits.

Pre-allocation is done with vmalloc and alloc/free is done
via percpu_freelist. Here are performance numbers for different
pre-allocation algorithms that were implemented, but discarded
in favor of percpu_freelist:

1 cpu:
pcpu_ida	2.1M
pcpu_ida nolock	2.3M
bt		2.4M
kmalloc		1.8M
hlist+spinlock	2.3M
pcpu_freelist	2.6M

4 cpu:
pcpu_ida	1.5M
pcpu_ida nolock	1.8M
bt w/smp_align	1.7M
bt no/smp_align	1.1M
kmalloc		0.7M
hlist+spinlock	0.2M
pcpu_freelist	2.0M

8 cpu:
pcpu_ida	0.7M
bt w/smp_align	0.8M
kmalloc		0.4M
pcpu_freelist	1.5M

32 cpu:
kmalloc		0.13M
pcpu_freelist	0.49M

pcpu_ida nolock is a modified percpu_ida algorithm without
percpu_ida_cpu locks and without cross-cpu tag stealing.
It's faster than existing percpu_ida, but not as fast as pcpu_freelist.

bt is a variant of block/blk-mq-tag.c simlified and customized
for bpf use case. bt w/smp_align is using cache line for every 'long'
(similar to blk-mq-tag). bt no/smp_align allocates 'long'
bitmasks continuously to save memory. It's comparable to percpu_ida
and in some cases faster, but slower than percpu_freelist

hlist+spinlock is the simplest free list with single spinlock.
As expeceted it has very bad scaling in SMP.

kmalloc is existing implementation which is still available via
BPF_F_NO_PREALLOC flag. It's significantly slower in single cpu and
in 8 cpu setup it's 3 times slower than pre-allocation with pcpu_freelist,
but saves memory, so in cases where map->max_entries can be large
and number of map update/delete per second is low, it may make
sense to use it.
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      6c905981