TSTORE_FP¶

指令示意图¶

简介¶

使用缩放 (fp) Tile 作为向量量化参数，将累加器 Tile 存储到全局内存。

数学语义¶

Let R = src.GetValidRow() and C = src.GetValidCol(). Conceptually (2D view, with a base offset), for 0 <= i < R and 0 <= j < C:

汇编语法¶

PTO-AS 形式：参见 PTO-AS Specification.

同步形式：

tstore.fp %src, %fp, %sv_out[%c0, %c0]

AS Level 1 (SSA)¶

pto.tstore.fp %src, %fp, %mem : (!pto.tile<...>, !pto.tile<...>, !pto.partition_tensor_view<MxNxdtype>) -> ()

AS Level 2 (DPS)¶

pto.tstore.fp ins(%src, %fp : !pto.tile_buf<...>, !pto.tile_buf<...>) outs(%mem : !pto.partition_tensor_view<MxNxdtype>)

AS Level 1（SSA）¶

pto.tstore.fp %src, %fp, %mem : (!pto.tile<...>, !pto.tile<...>, !pto.partition_tensor_view<MxNxdtype>) -> ()

AS Level 2（DPS）¶

pto.tstore.fp ins(%src, %fp : !pto.tile_buf<...>, !pto.tile_buf<...>) outs(%mem : !pto.partition_tensor_view<MxNxdtype>)

C++ 内建接口¶

声明于 include/pto/common/pto_instr.hpp and include/pto/common/constants.hpp:

template <typename TileData, typename GlobalData, typename FpTileData, AtomicType atomicType = AtomicType::AtomicNone,
          ReluPreMode reluPreMode = ReluPreMode::NoRelu, typename... WaitEvents>
PTO_INST RecordEvent TSTORE_FP(GlobalData &dst, TileData &src, FpTileData &fp, WaitEvents &... events);

约束¶

• 实现检查 (A2A3):
  • The fp store path is implemented via TSTORE_IMPL(dst, src, fp) and uses the same accumulator-to-GM legality checks as quantized accumulator stores:
  • Destination layout must be ND or NZ.
  • Source dtype must be int32_t or float.
  • Static shape constraints: 1 <= TileData::Cols <= 4095; if ND then 1 <= TileData::Rows <= 8192; if NZ then 1 <= TileData::Rows <= 65535 and TileData::Cols % 16 == 0.
  • Runtime: 1 <= src.GetValidCol() <= 4095.
  • No explicit static_assert is enforced on FpTileData (the implementation uses fp to set FPC state).
• 实现检查 (A5):
  • Implemented via TSTORE_IMPL(dst, src, fp) and validated by CheckStaticAcc<..., true>() for the accumulator path (ND/NZ only, int32_t/float source dtype, rows/cols ranges).
  • No explicit static_assert is enforced on FpTileData (the implementation uses fp to set FPC state).

示例¶

自动（Auto）¶

#include <pto/pto-inst.hpp>

using namespace pto;

void example_auto(__gm__ int8_t* out) {
  using AccT = TileAcc<float, 16, 16>;
  using FpT = Tile<TileType::Scaling, uint64_t, 1, 16, BLayout::RowMajor, 1, DYNAMIC, SLayout::NoneBox>;
  using GShape = Shape<1, 1, 1, 16, 16>;
  using GStride = BaseShape2D<int8_t, 16, 16, Layout::ND>;
  using GT = GlobalTensor<int8_t, GShape, GStride, Layout::ND>;

  GT gout(out);
  AccT acc;
  FpT fp(16);
  TSTORE_FP(gout, acc, fp);
}

手动（Manual）¶

#include <pto/pto-inst.hpp>

using namespace pto;

void example_manual(__gm__ int8_t* out) {
  using AccT = TileAcc<float, 16, 16>;
  using FpT = Tile<TileType::Scaling, uint64_t, 1, 16, BLayout::RowMajor, 1, DYNAMIC, SLayout::NoneBox>;
  using GShape = Shape<1, 1, 1, 16, 16>;
  using GStride = BaseShape2D<int8_t, 16, 16, Layout::ND>;
  using GT = GlobalTensor<int8_t, GShape, GStride, Layout::ND>;

  GT gout(out);
  AccT acc;
  FpT fp(16);
  TASSIGN(acc, 0x1000);
  TASSIGN(fp,  0x2000);
  TSTORE_FP(gout, acc, fp);
}