TGEMV_ACC¶
指令示意图¶
简介¶
带显式累加器输入 Tile(cInMatrix)和输出 Tile(cOutMatrix)的 GEMV。
另请参见¶
- 基础 GEMV 指令:
docs/isa/TGEMV.md。 - 偏置变体:
docs/isa/TGEMV_BIAS.md。
C++ 内建接口¶
声明于 include/pto/common/pto_instr.hpp:
template <typename TileRes, typename TileLeft, typename TileRight, typename... WaitEvents>
PTO_INST RecordEvent TGEMV_ACC(TileRes &cOutMatrix, TileRes &cInMatrix, TileLeft &aMatrix, TileRight &bMatrix, WaitEvents &... events);
template <AccPhase Phase, typename TileRes, typename TileLeft, typename TileRight, typename... WaitEvents>
PTO_INST RecordEvent TGEMV_ACC(TileRes &cOutMatrix, TileRes &cInMatrix, TileLeft &aMatrix, TileRight &bMatrix, WaitEvents &... events);
数学语义¶
设:
M = 1K = bMatrix.GetValidRow()N = bMatrix.GetValidCol()
对于 0 <= j < N(累加到已有输出 Tile):
\[ \mathrm{C}_{0,j} \gets \mathrm{C}_{0,j} + \sum_{k=0}^{K-1} \mathrm{A}_{0,k} \cdot \mathrm{B}_{k,j} \]
注意: 精确的累加器行为和数据类型提升由目标/实现定义。
汇编语法¶
PTO-AS 形式:参见 PTO-AS 规范。
同步形式:
%acc1 = tgemv.acc %acc0, %a, %b : (!pto.tile<...>, !pto.tile<...>, !pto.tile<...>) -> !pto.tile<...>
AS Level 1(SSA)¶
%c_out = pto.tgemv.acc %c_in, %a, %b : (!pto.tile<...>, !pto.tile<...>, !pto.tile<...>) -> !pto.tile<...>
AS Level 2(DPS)¶
pto.tgemv.acc ins(%c_in, %a, %b : !pto.tile_buf<...>, !pto.tile_buf<...>, !pto.tile_buf<...>) outs(%c_out : !pto.tile_buf<...>)
约束¶
通用形状与位置约束¶
- 静态形状约束:
TileLeft::Rows == TileRes::RowsTileLeft::Cols == TileRight::RowsTileRight::Cols == TileRes::Cols
- Tile 位置约束:
TileLeft::Loc == LeftTileRight::Loc == RightTileRes::Loc == Acc
- 运行时有效尺寸约束:
m必须为1k和n(取自bMatrix.GetValidRow()与bMatrix.GetValidCol())必须位于[1, 4095]
数据类型约束¶
- 实现检查 (A2A3):
- 支持的
(CType, AType, BType)三元组:(int32_t, int8_t, int8_t)(float, half, half)(float, float, float)(float, bfloat16_t, bfloat16_t)
- 支持的
- 实现检查 (A5):
- 累加器类型必须是
int32_t或float。 - 如果为
int32_t:AType == int8_t且BType == int8_t。 - 如果为
float:支持half、bfloat16_t、float以及选定的 fp8 组合(目标定义)。 - 会强制执行以下分形/布局约束:
- Left:
Loc == Left、!isRowMajor、SFractal == RowMajor - Right:
Loc == Right、isRowMajor、SFractal == ColMajor - Acc:
Loc == Acc、!isRowMajor、SFractal == RowMajor
- Left:
- 除上述 GEMV 约定外,底层 A5 matmul 实现不会再单独补充一组显式的
m/k/n运行时断言。
- 累加器类型必须是
示例¶
自动(Auto)¶
#include <pto/pto-inst.hpp>
using namespace pto;
void example_auto() {
using A = TileLeft<half, 1, 16>;
using B = TileRight<half, 16, 16>;
using C = TileAcc<float, 1, 16>;
A a;
B b;
C c0, c1;
TGEMV_ACC(c1, c0, a, b);
}
手动(Manual)¶
#include <pto/pto-inst.hpp>
using namespace pto;
void example_manual() {
using A = TileLeft<half, 1, 16>;
using B = TileRight<half, 16, 16>;
using C = TileAcc<float, 1, 16>;
A a;
B b;
C c0, c1;
TASSIGN(a, 0x1000);
TASSIGN(b, 0x2000);
TASSIGN(c0, 0x3000);
TASSIGN(c1, 0x4000);
TGEMV_ACC(c1, c0, a, b);
}
汇编示例(ASM)¶
自动模式¶
# 自动模式:由编译器/运行时负责资源放置与调度。
%c_out = pto.tgemv.acc %c_in, %a, %b : (!pto.tile<...>, !pto.tile<...>, !pto.tile<...>) -> !pto.tile<...>
手动模式¶
# 手动模式:先显式绑定资源,再发射指令。
# 可选(当该指令包含 tile 操作数时):
# pto.tassign %arg0, @tile(0x1000)
# pto.tassign %arg1, @tile(0x2000)
%c_out = pto.tgemv.acc %c_in, %a, %b : (!pto.tile<...>, !pto.tile<...>, !pto.tile<...>) -> !pto.tile<...>
PTO 汇编形式¶
%acc1 = tgemv.acc %acc0, %a, %b : (!pto.tile<...>, !pto.tile<...>, !pto.tile<...>) -> !pto.tile<...>
# AS Level 2 (DPS)
pto.tgemv.acc ins(%c_in, %a, %b : !pto.tile_buf<...>, !pto.tile_buf<...>, !pto.tile_buf<...>) outs(%c_out : !pto.tile_buf<...>)