TROWEXPANDDIV

Tile Operation Diagram

Introduction

Row-wise broadcast divide: divide each row of src0 by a per-row scalar vector src1.

Math Interpretation

For each element (i, j) in the valid region:

$${\mathrm{d}\mathrm{s}\mathrm{t}}_{i,j}=\frac{{\mathrm{s}\mathrm{r}\mathrm{c}0}_{i,j}}{{\mathrm{s}\mathrm{r}\mathrm{c}1}_{0,i}}$$

Assembly Syntax

Synchronous form:

%dst = trowexpanddiv %src0, %src1 : !pto.tile<...>, !pto.tile<...> -> !pto.tile<...>

AS Level 1 (SSA)

%dst = pto.trowexpanddiv %src0, %src1 : !pto.tile<...>, !pto.tile<...> -> !pto.tile<...>

AS Level 2 (DPS)

pto.trowexpanddiv ins(%src0, %src1 : !pto.tile_buf<...>, !pto.tile_buf<...>) outs(%dst : !pto.tile_buf<...>)

C++ Intrinsic

Declared in include/pto/common/pto_instr.hpp:

template <typename TileDataDst, typename TileDataSrc0, typename TileDataSrc1, typename... WaitEvents>
PTO_INST RecordEvent TROWEXPANDDIV(TileDataDst &dst, TileDataSrc0 &src0, TileDataSrc1 &src1, WaitEvents &... events);

template <typename TileDataDst, typename TileDataSrc0, typename TileDataSrc1, typename TileDataTmp,
          typename... WaitEvents>
PTO_INST RecordEvent TROWEXPANDDIV(TileDataDst &dst, TileDataSrc0 &src0, TileDataSrc1 &src1, TileDataTmp &tmp, WaitEvents &... events);

Constraints

• Implementation checks:
  • TileDataDst::DType == TileDataSrc0::DType == TileDataSrc1::DType (compile-time).
  • TileDataDst::DType, TileDataSrc0::DType, TileDataSrc1::DType must be one of: half, float, int16, int32, uint16, uint32.
  • Tile shape/layout constraint (compile-time): TileDataDst::isRowMajor.
  • Mode 1: src1 is expected to provide one scalar per row (i.e., its valid shape must cover R values).
  • Mode 2: src1 is expected to provide 32 bytes data per row.

Examples

Auto

#include <pto/pto-inst.hpp>

using namespace pto;

void example_auto() {
  using TileT = Tile<TileType::Vec, half, 16, 16>;
  using RowVecT = Tile<TileType::Vec, half, 16, 1, BLayout::ColMajor, 1, DYNAMIC, SLayout::NoneBox>;

  TileT src0, dst;
  RowVecT src1(16);
  TROWEXPANDDIV(dst, src0, src1);
}

Manual

#include <pto/pto-inst.hpp>

using namespace pto;

void example_manual() {
  using TileT = Tile<TileType::Vec, half, 16, 16>;
  using RowVecT = Tile<TileType::Vec, half, 16, 1, BLayout::ColMajor, 1, DYNAMIC, SLayout::NoneBox>;

  TileT src0, dst;
  RowVecT src1(16);
  TASSIGN(src0, 0x1000);
  TASSIGN(dst,  0x2000);
  TASSIGN(src1, 0x3000);
  TROWEXPANDDIV(dst, src0, src1);
}

ASM Form Examples

Auto Mode

# Auto mode: compiler/runtime-managed placement and scheduling.
%dst = pto.trowexpanddiv %src0, %src1 : !pto.tile<...>, !pto.tile<...> -> !pto.tile<...>

Manual Mode

# Manual mode: resources must be bound explicitly before issuing the instruction.
# Optional for tile operands:
# pto.tassign %arg0, @tile(0x1000)
# pto.tassign %arg1, @tile(0x2000)
%dst = pto.trowexpanddiv %src0, %src1 : !pto.tile<...>, !pto.tile<...> -> !pto.tile<...>

PTO Assembly Form

%dst = trowexpanddiv %src0, %src1 : !pto.tile<...>, !pto.tile<...> -> !pto.tile<...>
# AS Level 2 (DPS)
pto.trowexpanddiv ins(%src0, %src1 : !pto.tile_buf<...>, !pto.tile_buf<...>) outs(%dst : !pto.tile_buf<...>)