TCOLEXPAND¶

指令示意图¶

简介¶

将每个源列的第一个元素广播到目标列中。

数学语义¶

Let R = dst.GetValidRow() and C = dst.GetValidCol(). 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：

%dst = tcolexpand %src : !pto.tile<...> -> !pto.tile<...>

AS Level 1 (SSA)¶

%dst = pto.tcolexpand %src : !pto.tile<...> -> !pto.tile<...>

AS Level 2 (DPS)¶

pto.tcolexpand ins(%src : !pto.tile_buf<...>) outs(%dst : !pto.tile_buf<...>)

AS Level 1（SSA）¶

%dst = pto.tcolexpand %src : !pto.tile<...> -> !pto.tile<...>

AS Level 2（DPS）¶

pto.tcolexpand ins(%src : !pto.tile_buf<...>) outs(%dst : !pto.tile_buf<...>)

C++ 内建接口¶

声明于 include/pto/common/pto_instr.hpp:

template <typename TileDataDst, typename TileDataSrc, typename... WaitEvents>
PTO_INST RecordEvent TCOLEXPAND(TileDataDst &dst, TileDataSrc &src, WaitEvents &... events);

约束¶

• The op iterates over dst.GetValidRow() / dst.GetValidCol().

示例¶

#include <pto/pto-inst.hpp>

using namespace pto;

void example() {
  using TileT = Tile<TileType::Vec, float, 16, 16>;
  TileT src, dst;
  TCOLEXPAND(dst, src);
}