Message-ID: <20190829064502.2750303-1-yhs@fb.com>
Date:   Wed, 28 Aug 2019 23:45:02 -0700
From:   Yonghong Song <yhs@...com>
To:     <bpf@...r.kernel.org>, <netdev@...r.kernel.org>
CC:     Alexei Starovoitov <ast@...com>,
        Brian Vazquez <brianvv@...gle.com>,
        Daniel Borkmann <daniel@...earbox.net>, <kernel-team@...com>,
        Yonghong Song <yhs@...com>
Subject: [PATCH bpf-next 00/13] bpf: adding map batch processing support

Brian Vazquez has proposed BPF_MAP_DUMP command to look up more than one
map entries per syscall.
  https://lore.kernel.org/bpf/CABCgpaU3xxX6CMMxD+1knApivtc2jLBHysDXw-0E9bQEL0qC3A@mail.gmail.com/T/#t

During discussion, we found more use cases can be supported in a similar
map operation batching framework. For example, batched map lookup and delete,
which can be really helpful for bcc.
  https://github.com/iovisor/bcc/blob/master/tools/tcptop.py#L233-L243
  https://github.com/iovisor/bcc/blob/master/tools/slabratetop.py#L129-L138
    
Also, in bcc, we have API to delete all entries in a map.
  https://github.com/iovisor/bcc/blob/master/src/cc/api/BPFTable.h#L257-L264

For map update, batched operations also useful as sometimes applications need
to populate initial maps with more than one entry. For example, the below
example is from kernel/samples/bpf/xdp_redirect_cpu_user.c:
  https://github.com/torvalds/linux/blob/master/samples/bpf/xdp_redirect_cpu_user.c#L543-L550

This patch addresses all the above use cases. To make uapi stable, it also
covers other potential use cases. Four bpf syscall subcommands are introduced:
    BPF_MAP_LOOKUP_BATCH
    BPF_MAP_LOOKUP_AND_DELETE_BATCH
    BPF_MAP_UPDATE_BATCH
    BPF_MAP_DELETE_BATCH

In userspace, application can iterate through the whole map one batch
as a time, e.g., bpf_map_lookup_batch() in the below:
    p_key = NULL;
    p_next_key = &key;
    while (true) {
       err = bpf_map_lookup_batch(fd, p_key, &p_next_key, keys, values,
                                  &batch_size, elem_flags, flags);
       if (err) ...
       if (p_next_key) break; // done
       if (!p_key) p_key = p_next_key;
    }
Please look at individual patches for details of new syscall subcommands
and examples of user codes.

The testing is also done in a qemu VM environment:
      measure_lookup: max_entries 1000000, batch 10, time 342ms
      measure_lookup: max_entries 1000000, batch 1000, time 295ms
      measure_lookup: max_entries 1000000, batch 1000000, time 270ms
      measure_lookup: max_entries 1000000, no batching, time 1346ms
      measure_lookup_delete: max_entries 1000000, batch 10, time 433ms
      measure_lookup_delete: max_entries 1000000, batch 1000, time 363ms
      measure_lookup_delete: max_entries 1000000, batch 1000000, time 357ms
      measure_lookup_delete: max_entries 1000000, not batch, time 1894ms
      measure_delete: max_entries 1000000, batch, time 220ms
      measure_delete: max_entries 1000000, not batch, time 1289ms
For a 1M entry hash table, batch size of 10 can reduce cpu time
by 70%. Please see patch "tools/bpf: measure map batching perf"
for details of test codes.

Brian Vazquez (1):
  bpf: add bpf_map_value_size and bp_map_copy_value helper functions

Yonghong Song (12):
  bpf: refactor map_update_elem()
  bpf: refactor map_delete_elem()
  bpf: refactor map_get_next_key()
  bpf: adding map batch processing support
  tools/bpf: sync uapi header bpf.h
  tools/bpf: implement libbpf API functions for map batch operations
  tools/bpf: add test for bpf_map_update_batch()
  tools/bpf: add test for bpf_map_lookup_batch()
  tools/bpf: add test for bpf_map_lookup_and_delete_batch()
  tools/bpf: add test for bpf_map_delete_batch()
  tools/bpf: add a multithreaded test for map batch operations
  tools/bpf: measure map batching perf

 include/uapi/linux/bpf.h                      |  27 +
 kernel/bpf/syscall.c                          | 752 ++++++++++++++----
 tools/include/uapi/linux/bpf.h                |  27 +
 tools/lib/bpf/bpf.c                           |  67 ++
 tools/lib/bpf/bpf.h                           |  17 +
 tools/lib/bpf/libbpf.map                      |   4 +
 .../selftests/bpf/map_tests/map_batch_mt.c    | 126 +++
 .../selftests/bpf/map_tests/map_batch_perf.c  | 242 ++++++
 .../bpf/map_tests/map_delete_batch.c          | 139 ++++
 .../map_tests/map_lookup_and_delete_batch.c   | 164 ++++
 .../bpf/map_tests/map_lookup_batch.c          | 166 ++++
 .../bpf/map_tests/map_update_batch.c          | 115 +++
 12 files changed, 1707 insertions(+), 139 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/map_tests/map_batch_mt.c
 create mode 100644 tools/testing/selftests/bpf/map_tests/map_batch_perf.c
 create mode 100644 tools/testing/selftests/bpf/map_tests/map_delete_batch.c
 create mode 100644 tools/testing/selftests/bpf/map_tests/map_lookup_and_delete_batch.c
 create mode 100644 tools/testing/selftests/bpf/map_tests/map_lookup_batch.c
 create mode 100644 tools/testing/selftests/bpf/map_tests/map_update_batch.c

-- 
2.17.1

Powered by blists - more mailing lists