Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Riscv lane #328

Merged
merged 1 commit into from
Feb 3, 2025
Merged

Riscv lane #328

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 8 additions & 2 deletions src/jit/ByteCodeParser.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -257,8 +257,6 @@ static bool isFloatGlobal(uint32_t globalIndex, Module* module)
OL2(OTV128ToI32, /* SD */ V128 | TMP, I32) \
OL4(OTOp3V128, /* SSSD */ V128 | TMP, V128 | TMP, V128 | NOTMP, V128 | TMP | S2) \
OL2(OTExtractLaneI64, /* SD */ V128 | TMP, I64) \
OL2(OTExtractLaneF32, /* SD */ V128 | TMP, F32 | S0) \
OL2(OTExtractLaneF64, /* SD */ V128 | TMP, F64 | S0) \
OL3(OTReplaceLaneI32, /* SSD */ V128 | NOTMP, I32, V128 | TMP | S0) \
OL3(OTReplaceLaneI64, /* SSD */ V128 | NOTMP, I64, V128 | TMP | S0) \
OL3(OTReplaceLaneF32, /* SSD */ V128 | NOTMP, F32 | NOTMP, V128 | TMP | S0) \
Expand All @@ -273,6 +271,8 @@ static bool isFloatGlobal(uint32_t globalIndex, Module* module)
OL4(OTOp2V128Tmp, /* SSDT */ V128 | NOTMP, V128 | TMP, V128 | TMP | S0, V128) \
OL3(OTOp2V128Rev, /* SSD */ V128 | TMP, V128 | NOTMP, V128 | TMP | S1) \
OL5(OTOp3DotAddV128, /* SSSDT */ V128 | TMP, V128 | TMP, V128 | NOTMP, V128 | TMP | S2, V128) \
OL2(OTExtractLaneF32, /* SD */ V128 | TMP, F32 | S0) \
OL2(OTExtractLaneF64, /* SD */ V128 | TMP, F64 | S0) \
OL3(OTShuffleV128, /* SSD */ V128 | NOTMP, V128 | NOTMP, V128 | TMP | S0) \
OL3(OTPopcntV128, /* SDT */ V128 | NOTMP, V128 | TMP | S0, V128) \
OL3(OTShiftV128, /* SSD */ V128 | NOTMP, I32, V128 | TMP | S0) \
Expand All @@ -289,6 +289,8 @@ static bool isFloatGlobal(uint32_t globalIndex, Module* module)
#define OPERAND_TYPE_LIST_SIMD_ARCH \
OL3(OTOp2V128, /* SSD */ V128 | TMP, V128 | TMP, V128 | TMP | S0 | S1) \
OL3(OTPMinMaxV128, /* SSD */ V128 | TMP, V128 | TMP, V128 | TMP) \
OL2(OTExtractLaneF32, /* SD */ V128 | TMP, F32 | S0) \
OL2(OTExtractLaneF64, /* SD */ V128 | TMP, F64 | S0) \
OL3(OTShuffleV128, /* SSD */ V128 | NOTMP, V128 | NOTMP, V128 | TMP | S0 | S1) \
OL3(OTShiftV128, /* SSD */ V128 | NOTMP, I32, V128 | TMP | S0)

Expand All @@ -309,6 +311,8 @@ static bool isFloatGlobal(uint32_t globalIndex, Module* module)
OL2(OTOp1V128CB, /* SD */ V128 | NOTMP, V128 | NOTMP) \
OL3(OTOp2V128, /* SSD */ V128 | TMP, V128 | TMP, V128 | TMP | S0 | S1) \
OL3(OTMinMaxV128, /* SSD */ V128 | NOTMP, V128 | NOTMP, V128 | NOTMP) \
OL2(OTExtractLaneF32, /* SD */ V128 | TMP, F32 | S0) \
OL2(OTExtractLaneF64, /* SD */ V128 | TMP, F64 | S0) \
OL3(OTSwizzleV128, /* SSD */ V128 | TMP, V128 | NOTMP, V128 | TMP | S1) \
OL3(OTShuffleV128, /* SSD */ V128 | TMP, V128 | TMP, V128 | TMP) \
OL3(OTShiftV128, /* SSD */ V128 | NOTMP, I32, V128 | TMP | S0)
Expand All @@ -328,6 +332,8 @@ static bool isFloatGlobal(uint32_t globalIndex, Module* module)
OL2(OTOp1V128CB, /* SD */ V128 | NOTMP, V128 | NOTMP) \
OL3(OTOp2V128, /* SSD */ V128 | TMP, V128 | TMP, V128 | TMP | S0 | S1) \
OL3(OTOp1V128Tmp, /* SDT */ V128 | NOTMP, V128 | TMP | S0, V128) \
OL2(OTExtractLaneF32, /* SD */ V128 | TMP, F32) \
OL2(OTExtractLaneF64, /* SD */ V128 | TMP, F64) \
OL3(OTSwizzleV128, /* SSD */ V128 | TMP, V128 | NOTMP, V128 | TMP | S1) \
OL3(OTShuffleV128, /* SSD */ V128 | TMP, V128 | TMP, V128 | TMP) \
OL3(OTShiftV128, /* SSD */ V128 | NOTMP, I32, V128 | TMP | S0)
Expand Down
41 changes: 34 additions & 7 deletions src/jit/SimdRiscvInl.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -77,6 +77,7 @@ enum TypeOpcode : uint32_t {
vminu_vv = InstructionType::opivv | OPCODE(0x4),
vmseq_vi = InstructionType::opivi | OPCODE(0x18),
vmseq_vv = InstructionType::opivv | OPCODE(0x18),
vmsgt_vi = InstructionType::opivi | OPCODE(0x1f),
vmsle_vv = InstructionType::opivv | OPCODE(0x1d),
vmsleu_vv = InstructionType::opivv | OPCODE(0x1c),
vmslt_vv = InstructionType::opivv | OPCODE(0x1b),
Expand Down Expand Up @@ -112,6 +113,7 @@ enum TypeOpcode : uint32_t {
vssub_vv = InstructionType::opivv | OPCODE(0x23),
vssubu_vv = InstructionType::opivv | OPCODE(0x22),
vsub_vv = InstructionType::opivv | OPCODE(0x2),
vsub_vx = InstructionType::opivx | OPCODE(0x2),
vwmul_vv = InstructionType::opmvv | OPCODE(0x3b),
vwmulu_vv = InstructionType::opmvv | OPCODE(0x38),
vxor_vi = InstructionType::opivi | OPCODE(0xb),
Expand Down Expand Up @@ -418,13 +420,12 @@ static void simdEmitPopcnt(sljit_compiler* compiler, sljit_s32 type, sljit_s32 r

static void simdEmitSwizzle(sljit_compiler* compiler, sljit_s32 type, sljit_s32 rd, sljit_s32 rn, sljit_s32 rm)
{
sljit_s32 tmp = SLJIT_TMP_DEST_VREG;
if (rd == rn) {
simdEmitTypedOp(compiler, type, SimdOp::vrgather_vv, tmp, rn, rm);
simdEmitOp(compiler, SimdOp::vmv_vv, rd, 0, tmp);
} else {
simdEmitTypedOp(compiler, type, SimdOp::vrgather_vv, rd, rn, rm);
}
sljit_s32 tmp1 = SLJIT_TMP_DEST_VREG;
sljit_s32 tmp2 = SLJIT_VR0;

simdEmitTypedOp(compiler, type, SimdOp::vrgather_vv, tmp1, rn, rm);
simdEmitOp(compiler, SimdOp::vmsgt_vi, tmp2, rm, 0xf, SimdOp::rmIsImm);
simdEmitOp(compiler, SimdOp::vmerge_vi, rd, tmp1, 0, SimdOp::rmIsImm);
}

static void emitUnarySIMD(sljit_compiler* compiler, Instruction* instr)
Expand Down Expand Up @@ -1309,6 +1310,32 @@ static void emitSelectSIMD(sljit_compiler* compiler, Instruction* instr)

static void emitShuffleSIMD(sljit_compiler* compiler, Instruction* instr)
{
Operand* operands = instr->operands();
JITArg args[3];

simdOperandToArg(compiler, operands, args[0], SLJIT_SIMD_ELEM_128, SLJIT_TMP_FR0);
simdOperandToArg(compiler, operands + 1, args[1], SLJIT_SIMD_ELEM_128, SLJIT_TMP_FR1);
args[2].set(operands + 2);
sljit_s32 dst = GET_TARGET_REG(args[2].arg, instr->requiredReg(0));

I8X16Shuffle* shuffle = reinterpret_cast<I8X16Shuffle*>(instr->byteCode());
sljit_s32 tmp1 = SLJIT_TMP_DEST_VREG;
sljit_s32 tmp2 = SLJIT_VR0;

sljit_emit_simd_mov(compiler, SLJIT_SIMD_LOAD | SLJIT_SIMD_REG_128 | SLJIT_SIMD_ELEM_8, tmp2,
SLJIT_MEM0(), reinterpret_cast<sljit_sw>(shuffle->value()));
sljit_emit_op1(compiler, SLJIT_MOV, SLJIT_TMP_DEST_REG, 0, SLJIT_IMM, 16);

simdEmitTypedOp(compiler, SLJIT_SIMD_ELEM_8, SimdOp::vrgather_vv, tmp1, args[0].arg, tmp2);
simdEmitOp(compiler, SimdOp::vsub_vx, tmp2, tmp2, SLJIT_TMP_DEST_REG, SimdOp::rmIsGpr);
simdEmitOp(compiler, SimdOp::vrgather_vv, dst, args[1].arg, tmp2);

simdEmitOp(compiler, SimdOp::vmsgt_vi, tmp2, tmp2, 0x1f, SimdOp::rmIsImm);
simdEmitOp(compiler, SimdOp::vmerge_vv, dst, tmp1, dst);

if (args[2].arg != dst) {
sljit_emit_simd_mov(compiler, SLJIT_SIMD_STORE | SLJIT_SIMD_REG_128 | SLJIT_SIMD_ELEM_128, dst, args[2].arg, args[2].argw);
}
}

static void emitShiftSIMD(sljit_compiler* compiler, Instruction* instr)
Expand Down
Loading