llvm-project/llvm/lib/Target/BPF/BPFInstrFormats.td
Peilin Ye 17bfc00f7c
[BPF] Add load-acquire and store-release instructions under -mcpu=v4 (#108636)
As discussed in [1], introduce BPF instructions with load-acquire and
store-release semantics under -mcpu=v4.  Define 2 new flags:

  BPF_LOAD_ACQ    0x100
  BPF_STORE_REL   0x110

A "load-acquire" is a BPF_STX | BPF_ATOMIC instruction with the 'imm'
field set to BPF_LOAD_ACQ (0x100).

Similarly, a "store-release" is a BPF_STX | BPF_ATOMIC instruction with
the 'imm' field set to BPF_STORE_REL (0x110).

Unlike existing atomic read-modify-write operations that only support
BPF_W (32-bit) and BPF_DW (64-bit) size modifiers, load-acquires and
store-releases also support BPF_B (8-bit) and BPF_H (16-bit).  An 8- or
16-bit load-acquire zero-extends the value before writing it to a 32-bit
register, just like ARM64 instruction LDAPRH and friends.

As an example (assuming little-endian):

  long foo(long *ptr) {
      return __atomic_load_n(ptr, __ATOMIC_ACQUIRE);
  }

foo() can be compiled to:

  db 10 00 00 00 01 00 00  r0 = load_acquire((u64 *)(r1 + 0x0))
  95 00 00 00 00 00 00 00  exit

  opcode (0xdb): BPF_ATOMIC | BPF_DW | BPF_STX
  imm (0x00000100): BPF_LOAD_ACQ

Similarly:

  void bar(short *ptr, short val) {
      __atomic_store_n(ptr, val, __ATOMIC_RELEASE);
  }

bar() can be compiled to:

  cb 21 00 00 10 01 00 00  store_release((u16 *)(r1 + 0x0), w2)
  95 00 00 00 00 00 00 00  exit

  opcode (0xcb): BPF_ATOMIC | BPF_H | BPF_STX
  imm (0x00000110): BPF_STORE_REL

Inline assembly is also supported.

Add a pre-defined macro, __BPF_FEATURE_LOAD_ACQ_STORE_REL, to let
developers detect this new feature.  It can also be disabled using a new
llc option, -disable-load-acq-store-rel.

Using __ATOMIC_RELAXED for __atomic_store{,_n}() will generate a "plain"
store (BPF_MEM | BPF_STX) instruction:

  void foo(short *ptr, short val) {
      __atomic_store_n(ptr, val, __ATOMIC_RELAXED);
  }

  6b 21 00 00 00 00 00 00  *(u16 *)(r1 + 0x0) = w2
  95 00 00 00 00 00 00 00  exit

Similarly, using __ATOMIC_RELAXED for __atomic_load{,_n}() will generate
a zero-extending, "plain" load (BPF_MEM | BPF_LDX) instruction:

  int foo(char *ptr) {
      return __atomic_load_n(ptr, __ATOMIC_RELAXED);
  }

  71 11 00 00 00 00 00 00  w1 = *(u8 *)(r1 + 0x0)
  bc 10 08 00 00 00 00 00  w0 = (s8)w1
  95 00 00 00 00 00 00 00  exit

Currently __ATOMIC_CONSUME is an alias for __ATOMIC_ACQUIRE.  Using
__ATOMIC_SEQ_CST ("sequentially consistent") is not supported yet and
will cause an error:

  $ clang --target=bpf -mcpu=v4 -c bar.c > /dev/null
bar.c:1:5: error: sequentially consistent (seq_cst) atomic load/store is
not supported
1 | int foo(int *ptr) { return __atomic_load_n(ptr, __ATOMIC_SEQ_CST); }
      |     ^
  ...

Finally, rename those isST*() and isLD*() helper functions in
BPFMISimplifyPatchable.cpp based on what the instructions actually do,
rather than their instruction class.

[1]
https://lore.kernel.org/all/20240729183246.4110549-1-yepeilin@google.com/
2025-03-04 09:19:39 -08:00

135 lines
3.2 KiB
TableGen

//===-- BPFInstrFormats.td - BPF Instruction Formats -------*- tablegen -*-===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
class BPFOpClass<bits<3> val> {
bits<3> Value = val;
}
def BPF_LD : BPFOpClass<0x0>;
def BPF_LDX : BPFOpClass<0x1>;
def BPF_ST : BPFOpClass<0x2>;
def BPF_STX : BPFOpClass<0x3>;
def BPF_ALU : BPFOpClass<0x4>;
def BPF_JMP : BPFOpClass<0x5>;
def BPF_JMP32 : BPFOpClass<0x6>;
def BPF_ALU64 : BPFOpClass<0x7>;
class BPFSrcType<bits<1> val> {
bits<1> Value = val;
}
def BPF_K : BPFSrcType<0x0>;
def BPF_X : BPFSrcType<0x1>;
class BPFArithOp<bits<4> val> {
bits<4> Value = val;
}
def BPF_ADD : BPFArithOp<0x0>;
def BPF_SUB : BPFArithOp<0x1>;
def BPF_MUL : BPFArithOp<0x2>;
def BPF_DIV : BPFArithOp<0x3>;
def BPF_OR : BPFArithOp<0x4>;
def BPF_AND : BPFArithOp<0x5>;
def BPF_LSH : BPFArithOp<0x6>;
def BPF_RSH : BPFArithOp<0x7>;
def BPF_NEG : BPFArithOp<0x8>;
def BPF_MOD : BPFArithOp<0x9>;
def BPF_XOR : BPFArithOp<0xa>;
def BPF_MOV : BPFArithOp<0xb>;
def BPF_ARSH : BPFArithOp<0xc>;
def BPF_END : BPFArithOp<0xd>;
def BPF_XCHG : BPFArithOp<0xe>;
def BPF_CMPXCHG : BPFArithOp<0xf>;
class BPFAtomicOp<bits<5> val> {
bits<5> Value = val;
}
def BPF_LOAD_ACQ : BPFAtomicOp<0x10>;
def BPF_STORE_REL : BPFAtomicOp<0x11>;
class BPFEndDir<bits<1> val> {
bits<1> Value = val;
}
def BPF_TO_LE : BPFSrcType<0x0>;
def BPF_TO_BE : BPFSrcType<0x1>;
class BPFJumpOp<bits<4> val> {
bits<4> Value = val;
}
def BPF_JA : BPFJumpOp<0x0>;
def BPF_JEQ : BPFJumpOp<0x1>;
def BPF_JGT : BPFJumpOp<0x2>;
def BPF_JGE : BPFJumpOp<0x3>;
def BPF_JSET : BPFJumpOp<0x4>;
def BPF_JNE : BPFJumpOp<0x5>;
def BPF_JSGT : BPFJumpOp<0x6>;
def BPF_JSGE : BPFJumpOp<0x7>;
def BPF_CALL : BPFJumpOp<0x8>;
def BPF_EXIT : BPFJumpOp<0x9>;
def BPF_JLT : BPFJumpOp<0xa>;
def BPF_JLE : BPFJumpOp<0xb>;
def BPF_JSLT : BPFJumpOp<0xc>;
def BPF_JSLE : BPFJumpOp<0xd>;
def BPF_JCOND : BPFJumpOp<0xe>;
class BPFWidthModifer<bits<2> val> {
bits<2> Value = val;
}
def BPF_W : BPFWidthModifer<0x0>;
def BPF_H : BPFWidthModifer<0x1>;
def BPF_B : BPFWidthModifer<0x2>;
def BPF_DW : BPFWidthModifer<0x3>;
class BPFModeModifer<bits<3> val> {
bits<3> Value = val;
}
def BPF_IMM : BPFModeModifer<0x0>;
def BPF_ABS : BPFModeModifer<0x1>;
def BPF_IND : BPFModeModifer<0x2>;
def BPF_MEM : BPFModeModifer<0x3>;
def BPF_MEMSX : BPFModeModifer<0x4>;
def BPF_ATOMIC : BPFModeModifer<0x6>;
class BPFAtomicFlag<bits<4> val> {
bits<4> Value = val;
}
def BPF_FETCH : BPFAtomicFlag<0x1>;
class InstBPF<dag outs, dag ins, string asmstr, list<dag> pattern>
: Instruction {
field bits<64> Inst;
field bits<64> SoftFail = 0;
let Size = 8;
let Namespace = "BPF";
let DecoderNamespace = "BPF";
BPFOpClass BPFClass;
let Inst{58-56} = BPFClass.Value;
dag OutOperandList = outs;
dag InOperandList = ins;
let AsmString = asmstr;
let Pattern = pattern;
}
// Pseudo instructions
class Pseudo<dag outs, dag ins, string asmstr, list<dag> pattern>
: InstBPF<outs, ins, asmstr, pattern> {
let Inst{63-0} = 0;
let isPseudo = 1;
}