[PATCH 0/7] Atomics for eBPF

[Brendan Jackman <jackmanb@google.com>]

Status of the patches
=====================

This patchset is currently incomplete: it oughtn't be released as-is,
because that would necessitate incrementing architecture
version number for a partial feature-set. However I've implemented
the parts that I think are most interesting/controversial, so I'd
like to get the review started.

The missing elements that I'm aware of are:

* Atomic subtract
* Atomic & and |
* Documentation updates

The prog_test that's added depends on Clang/LLVM features added by
Yonghong in https://reviews.llvm.org/D72184

This only includes a JIT implementation for x86_64 - I don't plan to
implement JIT support myself for other architectures.

Operations
==========

This patchset adds atomic operations to the eBPF instruction set. The
use-case that motivated this work was a trivial and efficient way to
generate globally-unique cookies in BPF progs, but I think it's
obvious that these features are pretty widely applicable.  The
instructions that are added here can be summarised with this list of
kernel operations:

* atomic[64]_[fetch_]add
* atomic[64]_[fetch_]sub (TODO)
* atomic[64]_[fetch_]and (TODO)
* atomic[64]_[fetch_]or  (TODO)
* atomic[64]_xchg
* atomic[64]_cmpxchg

The following are left out of scope for this effort:

* 16 and 8 bit operations
* Explicit memory barriers

Encoding
========

I originally planned to add new values for bpf_insn.opcode. This was
rather unpleasant: the opcode space has holes in it but no entire
instruction classes[1]. Yonghong Song had a better idea: use the
immediate field of the existing STX XADD instruction to encode the
operation. This works nicely, without breaking existing programs,
because the immediate field is currently reserved-must-be-zero, and
extra-nicely because BPF_ADD happens to be zero.

Note that this of course makes immediate-source atomic operations
impossible. It's hard to imagine a measurable speedup from such
instructions, and if it existed it would certainly not benefit x86,
which has no support for them.

The BPF_OP opcode fields are re-used in the immediate, and an
additional flag BPF_FETCH is used to mark instructions that should
fetch a pre-modification value from memory.

So, BPF_XADD is now called BPF_ATOMIC (the old name is kept to avoid
breaking userspace builds), and where we previously had .imm = 0, we
now have .imm = BPF_ADD (which is 0).

Operands
========

Reg-source eBPF instructions only have two operands, while these
atomic operations have up to four. To avoid needing to encode
additional operands, then:

- One of the input registers is re-used as an output register
  (e.g. atomic_fetch_add both reads from and writes to the source
  register).

- Where necessary (i.e. for cmpxchg) , R0 is "hard-coded" as one of
  the operands.

This approach also allows the new eBPF instructions to map directly
to single x86 instructions.

[1] Visualisation of eBPF opcode space:
    https://gist.github.com/bjackman/00fdad2d5dfff601c1918bc29b16e778

Brendan Jackman (7):
  bpf: Factor out emission of ModR/M for *(reg + off)
  bpf: x86: Factor out emission of REX byte
  bpf: Rename BPF_XADD and prepare to encode other atomics in .imm
  bpf: Move BPF_STX reserved field check into BPF_STX verifier code
  bpf: Add BPF_FETCH field / create atomic_fetch_add instruction
  bpf: Add instructions for atomic_cmpxchg and friends
  bpf: Add tests for new BPF atomic operations

 Documentation/networking/filter.rst           |  27 ++--
 arch/arm/net/bpf_jit_32.c                     |   7 +-
 arch/arm64/net/bpf_jit_comp.c                 |  16 +-
 arch/mips/net/ebpf_jit.c                      |  11 +-
 arch/powerpc/net/bpf_jit_comp64.c             |  25 ++-
 arch/riscv/net/bpf_jit_comp32.c               |  20 ++-
 arch/riscv/net/bpf_jit_comp64.c               |  16 +-
 arch/s390/net/bpf_jit_comp.c                  |  26 ++--
 arch/sparc/net/bpf_jit_comp_64.c              |  14 +-
 arch/x86/net/bpf_jit_comp.c                   | 131 ++++++++++------
 arch/x86/net/bpf_jit_comp32.c                 |   6 +-
 drivers/net/ethernet/netronome/nfp/bpf/jit.c  |  14 +-
 drivers/net/ethernet/netronome/nfp/bpf/main.h |   4 +-
 .../net/ethernet/netronome/nfp/bpf/verifier.c |  13 +-
 include/linux/filter.h                        |  47 +++++-
 include/uapi/linux/bpf.h                      |   9 +-
 kernel/bpf/core.c                             |  64 ++++++--
 kernel/bpf/disasm.c                           |  25 ++-
 kernel/bpf/verifier.c                         |  72 ++++++---
 lib/test_bpf.c                                |   2 +-
 samples/bpf/bpf_insn.h                        |   4 +-
 samples/bpf/sock_example.c                    |   3 +-
 samples/bpf/test_cgrp2_attach.c               |   6 +-
 tools/include/linux/filter.h                  |  47 +++++-
 tools/include/uapi/linux/bpf.h                |   9 +-
 tools/testing/selftests/bpf/Makefile          |   2 +-
 .../selftests/bpf/prog_tests/atomics_test.c   | 145 ++++++++++++++++++
 .../bpf/prog_tests/cgroup_attach_multi.c      |   6 +-
 .../selftests/bpf/progs/atomics_test.c        |  61 ++++++++
 .../selftests/bpf/verifier/atomic_cmpxchg.c   |  96 ++++++++++++
 .../selftests/bpf/verifier/atomic_fetch_add.c | 106 +++++++++++++
 .../selftests/bpf/verifier/atomic_xchg.c      | 113 ++++++++++++++
 tools/testing/selftests/bpf/verifier/ctx.c    |   6 +-
 .../testing/selftests/bpf/verifier/leak_ptr.c |   4 +-
 tools/testing/selftests/bpf/verifier/unpriv.c |   3 +-
 tools/testing/selftests/bpf/verifier/xadd.c   |   2 +-
 36 files changed, 1002 insertions(+), 160 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/atomics_test.c
 create mode 100644 tools/testing/selftests/bpf/progs/atomics_test.c
 create mode 100644 tools/testing/selftests/bpf/verifier/atomic_cmpxchg.c
 create mode 100644 tools/testing/selftests/bpf/verifier/atomic_fetch_add.c
 create mode 100644 tools/testing/selftests/bpf/verifier/atomic_xchg.c

--
2.29.2.454.gaff20da3a2-goog