TGEMV_BIAS

1// PTO-INSTRUCTION: {"assembly":["TGEMV_BIAS <bundle operands>"],"block":["BSTART.TGEMV.BIAS AType","B.DATR BType, RMode, Sat (optional; BType defaults to AType)","B.FPATR PreQuantMode, ReluMode, GroupNCode, RowMaxEn, GroupMaxEn, RowMaxInit, MaxAbsEn, TransA, TransB, CScaleEn (exactly one)","B.DIM LB0 M (optional, default 1; TGEMV permits only M=1)","B.DIM LB1 N (optional, default 1)","B.DIM LB2 K (optional, default 1)","B.IOT ordered Local mathematical sources: A CUBE_M16/M32 primary, B CUBE_N8 primary, 1xN Bias","B.IOT D matching A's CUBE_M16/M32 layout, optional RowMaxOut, optional GroupMaxOut destinations","B.IOT/B.IOR postprocess operands selected by B.FPATR","BSTOP or the next BSTART completion boundary"],"catalog_indices":[98],"catalog_records":[{"arguments":[{"operand":"destination0"},{"operand":"source0"},{"operand":"source1"},{"operand":"source2"}],"command_mnemonic":"BSTART.TGEMV.BIAS","contract_status":"reviewed-complete","datr_contract":{"allowed_nonzero_fields":["DataType","RMode","Sat"],"pad_union":"must-zero"},"disposition":"accepted-direct-operation","effect_contract":"TGEMV_BIAS","family":"CUBE","fault_contract":"ExecuteTileInstruction","function":17,"legality_handler":"TileOperandsLegal_TGEMV_BIAS","name":"TGEMV_BIAS","operands":[{"field":"destination0","role":"destination"},{"field":"source0","role":"left-vector"},{"field":"source1","role":"right-matrix"},{"field":"source2","role":"bias"}],"restart_contract":"CompleteBundleAtWithAcceptedApplicabilityRules","semantic_handler":"TGEMV_BIAS","state_effects":["operand:destination0:destination","operand:source0:left-vector","operand:source1:right-matrix","operand:source2:bias"]}],"classification":["matrix-and-matrix-vector","matrix-vector"],"contract":{"block_composition":["BSTART.TGEMV.BIAS AType","B.DATR BType, RMode, Sat (optional; BType defaults to AType)","B.FPATR PreQuantMode, ReluMode, GroupNCode, RowMaxEn, GroupMaxEn, RowMaxInit, MaxAbsEn, TransA, TransB, CScaleEn (exactly one)","B.DIM LB0 M (optional, default 1; TGEMV permits only M=1)","B.DIM LB1 N (optional, default 1)","B.DIM LB2 K (optional, default 1)","B.IOT ordered Local mathematical sources: A CUBE_M16/M32 primary, B CUBE_N8 primary, 1xN Bias","B.IOT D matching A's CUBE_M16/M32 layout, optional RowMaxOut, optional GroupMaxOut destinations","B.IOT/B.IOR postprocess operands selected by B.FPATR","BSTOP or the next BSTART completion boundary"],"canonical_assembly":["TGEMV_BIAS <bundle operands>"],"defaults":["Encoded DataType is always AType. Omitted B.DATR preserves AType as BType, selects RNE, and disables saturation.","Omitted LB0, LB1, and LB2 default M, N, and K independently to one; TGEMV fixes M to one.","Exactly one all-zero B.FPATR selects no conversion, activation, or reduction; B.IOR and auxiliary B.IOT operands exist only when a selected postprocess mode requires them.","Local A uses persistent CUBE_M16 or CUBE_M32, Local B uses persistent CUBE_N8, and D is newly allocated in A's M layout. M is fixed to one; N and K are arbitrary positive values independent of per-PE TSize. Bias remains an ordinary row-major 1xN accumulator-type Tile.","TransA=0 and TransB=0 select no logical transpose. TGEMV requires both controls to remain zero."],"encoding_class":"selector-encoded-block-operation","examples":["BSTART.TGEMV.BIAS AType; B.DATR BType, RMode, Sat (optional; BType defaults to AType); B.FPATR PreQuantMode, ReluMode, GroupNCode, RowMaxEn, GroupMaxEn, RowMaxInit, MaxAbsEn, TransA, TransB, CScaleEn (exactly one); B.DIM LB0 M (optional, default 1; TGEMV permits only M=1); B.DIM LB1 N (optional, default 1); B.DIM LB2 K (optional, default 1); B.IOT ordered Local mathematical sources: A CUBE_M16/M32 primary, B CUBE_N8 primary, 1xN Bias; B.IOT D matching A's CUBE_M16/M32 layout, optional RowMaxOut, optional GroupMaxOut destinations; B.IOT/B.IOR postprocess operands selected by B.FPATR; BSTOP or the next BSTART completion boundary"],"exceptions":["A reserved DataType or fixed-bit mismatch raises Fault_IllegalInstruction before block state changes.","Missing, duplicate, or non-Matrix B.FPATR use raises Fault_BundleControl before allocation or payload effects.","Illegal types, dimensions, masks, binding streams, descriptors, shapes, capacities, aliases, readiness, or postprocess values raise Fault_TileLegality before source snapshots and effects."],"field_contracts":{},"field_zero_meanings":{},"legality":["The carrier selects exactly CUBE Function 17 and TileOperation_TGEMV_BIAS.","Local A uses persistent CUBE_M16 or CUBE_M32, Local B uses persistent CUBE_N8, and D is newly allocated in A's M layout. M is fixed to one; N and K are arbitrary positive values independent of per-PE TSize. Bias remains an ordinary row-major 1xN accumulator-type Tile.","TGEMV is Local-only: TransA and TransB are zero and every effective Shared binding rejects before effects.","AType and BType must be supported ordinary Matrix types from one numeric class. Bias is one Local row-major 1xN accumulator-type source. M is fixed to one and every Shared binding is illegal.","Every common nonzero four-bit PE_MASK is legal; all four PEs complete cooperative Shared readiness while only selected PEs allocate and publish. Mask zero is a strict no-op before descriptor reads, faults, allocation, readiness checks, or lifetime effects.","B.DATR permits only BType, RMode, and Sat. Exactly one B.FPATR is mandatory and closes the conditional postprocess schema."],"memory_effects":["none"],"operands":[{"field":"destination0","role":"destination"},{"field":"source0","role":"left-vector"},{"field":"source1","role":"right-matrix"},{"field":"source2","role":"bias"}],"ordering":["Complete schema, field, type, dimension, descriptor, shape, capacity, readiness, alias, and allocation preflight precedes every source snapshot and destination effect.","D and every enabled reduction output publish as one atomic group; rejection publishes none and successful sources persist."],"standalone_opcode":false,"state_effects":["Multiply the matrix by the vector and add the bias Tile.","After complete preflight, execute TGEMV_BIAS with the operand bindings listed above; destination definedness changes only as specified by that handler.","For Local execution, publish D with A's CUBE_M16 or CUBE_M32 layout and final output dtype; ordinary Bias, MX scales, and enabled reduction auxiliaries keep their operation-owned layouts."]},"depends_on":["PTO-BLOCK-MODEL-SCHEMA-PROFILE-ENCODING"],"engine":"CUBE","id":"PTO-TILE-TGEMV-BIAS","mnemonic":"TGEMV_BIAS","summary":"Multiply the matrix by the vector and add the bias Tile.","surface":"tile"}2// PTO-REVIEW: {"review_method":"formal-definition-read","outcome":"FORMAL-COMPLETE","reviewed_fields":["assembly","encoding","defaults","operation","state","memory","ordering","faults","reserved"]}3// NDF-BEGIN: PTO-TGEMV-BIAS-CONTRACT-0014// ndf: kind=contract level=L1 layer=tile status=accepted5// TGEMV_BIAS MUST select CUBE Function 17 and TileOperation_TGEMV_BIAS.6// Its encoded DataType is AType; omitted B.DATR MUST preserve AType as7// BType, omitted LB0/LB1/LB2 MUST default M/N/K independently to one,8// Local N and K MUST be positive and independent of per-PE TSize; M9// MUST equal one.10// AType and BType must be supported ordinary Matrix types from one numeric11// class. Bias is one Local row-major 1xN accumulator-type source. M is fixed12// to one and every Shared binding is illegal.13// TransA and TransB MUST remain zero and every Shared binding MUST reject.14// Every common nonzero Local PE mask is legal; mask zero is a strict no-op.15// Exactly one B.FPATR MUST close postprocess defaults and operands.16// Complete schema, type, shape, capacity, definedness, readiness, alias,17// and allocation preflight MUST precede source snapshots and effects;18// successful outputs MUST publish atomically and sources MUST persist.19// NDF-END: PTO-TGEMV-BIAS-CONTRACT-00120// DOC-BEGIN: decode21readonly func InstructionContractOperation_TGEMV_BIAS()22    => TileOperation23begin24    return TileOperation_TGEMV_BIAS;25end;26// DOC-END: decode27// DOC-BEGIN: operation28readonly func InstructionContractCubeFunction_TGEMV_BIAS()29    => integer {0..31}30begin31    return 17;32end;33
34readonly func InstructionContractSharedOperandsAllowed_TGEMV_BIAS()35    => boolean36begin37    return FALSE;38end;39
40readonly func InstructionContractOperandsLegal_TGEMV_BIAS(41    destination: TileIndex,42    left_vector: TileIndex,43    right_matrix: TileIndex,44    bias: TileIndex) => boolean45begin46    return TileOperandsLegal_TGEMV_BIAS(47        destination,48        left_vector,49        right_matrix,50        bias);51end;52
53readonly func InstructionContractHandler_TGEMV_BIAS()54    => TileSemanticHandler55begin56    return TileHandler_TGEMV_BIAS;57end;58// DOC-END: operation59