a_block_desc_m0_m1_m2_k | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
A_K0 | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
A_K1 | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
a_thread_copy_ | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | protected |
a_thread_desc_ | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | protectedstatic |
AThreadCopy typedef | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | protected |
b_block_desc_n0_n1_n2_k | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
B_K0 | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
B_K1 | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
b_thread_copy_ | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | protected |
b_thread_desc_ | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | protectedstatic |
BlockwiseGemmXdlops_v2(Tuple4 a_origin=CalculateAThreadOriginDataIndex(), Tuple4 b_origin=CalculateBThreadOriginDataIndex()) | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inline |
BThreadCopy typedef | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | protected |
c_thread_buf_ | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | |
c_thread_desc_ | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | protectedstatic |
CalculateAThreadOriginDataIndex() | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inlinestatic |
CalculateBThreadOriginDataIndex() | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inlinestatic |
CalculateCThreadOriginDataIndex(Number< m0 >, Number< n0 >, Number< xdlops_i >, Number< blk_i >) | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inlinestatic |
CalculateCThreadOriginDataIndex8D(Number< m0 >, Number< n0 >, Number< xdlops_i >, Number< blk_i >) | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inlinestatic |
GetCBlockDescriptor_G_M0_N0_M1_N1_M2_M3_M4_N2() | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inlinestatic |
GetCBlockDescriptor_M0_N0_M1_N1_M2_M3_M4_N2() | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inlinestatic |
GetCBlockDescriptor_M0_N0_M1_N1_M2_N2_N3_N4() | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inlinestatic |
GetCThreadBuffer() | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inline |
GetCThreadDescriptor_G_M0_N0_M1_N1_M2_M3_M4_N2() | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inlinestatic |
GetCThreadDescriptor_M0_N0_M1_N1_M2_M3_M4_N2() | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inlinestatic |
GetCThreadDescriptor_M0_N0_M1_N1_M2_N2_N3_N4() | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inlinestatic |
GetWaveIdx() | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inlinestatic |
I0 | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
I1 | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
I2 | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
I3 | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
KPerThread | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
MakeCGridDescriptor_G_M0_N0_M1_N1_M2_M3_M4_N2(const CGridDesc_G_M_N &c_grid_desc_g_m_n) | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inlinestatic |
MakeCGridDescriptor_M0_N0_M1_N1_M2_M3_M4_N2(const CGridDesc_M_N &c_grid_desc_m_n) | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inlinestatic |
MWaves | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
NWaves | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
Run(const ABlockBuffer &a_block_buf, const BBlockBuffer &b_block_buf, CThreadBuffer &c_thread_buf) const | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | inline |
ThisThreadBlock typedef | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | |
Tuple4 typedef | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | |
WaveSize | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |
xdlops_gemm | ck::BlockwiseGemmXdlops_v2< BlockSize, FloatAB, FloatAcc, ATileDesc, BTileDesc, AMmaTileDesc, BMmaTileDesc, MPerBlock, NPerBlock, KPerBlock, MPerXDL, NPerXDL, MRepeat, NRepeat, KPack, TransposeC, AMmaKStride, BMmaKStride > | static |