TGEMV_MX

1// PTO-INSTRUCTION: {"assembly":["TGEMV_MX <bundle operands>"],"block":["BSTART.TGEMVMX 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, optional A scale, B CUBE_N8 primary, optional B scale","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":[100],"catalog_records":[{"arguments":[{"operand":"destination0"},{"operand":"source0"},{"operand":"source1"},{"operand":"source2"},{"operand":"source3"}],"command_mnemonic":"BSTART.TGEMVMX","contract_status":"reviewed-complete","datr_contract":{"allowed_nonzero_fields":["DataType","RMode","Sat"],"pad_union":"must-zero"},"disposition":"accepted-direct-operation","effect_contract":"TGEMV_MX","family":"CUBE","fault_contract":"ExecuteTileInstruction","function":20,"legality_handler":"TileOperandsLegal_TGEMV_MX","name":"TGEMV_MX","operands":[{"field":"destination0","role":"destination"},{"field":"source0","role":"left-vector"},{"field":"source1","role":"row-scale"},{"field":"source2","role":"right-matrix"},{"field":"source3","role":"column-scale"}],"restart_contract":"CompleteBundleAtWithAcceptedApplicabilityRules","semantic_handler":"TGEMV_MX","state_effects":["operand:destination0:destination","operand:source0:left-vector","operand:source1:row-scale","operand:source2:right-matrix","operand:source3:column-scale"]}],"classification":["matrix-and-matrix-vector","matrix-vector"],"contract":{"block_composition":["BSTART.TGEMVMX 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, optional A scale, B CUBE_N8 primary, optional B scale","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_MX <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. Required E8M0 scales remain ordinary row-major Tiles.","TransA=0 and TransB=0 select no logical transpose. TGEMV requires both controls to remain zero."],"encoding_class":"selector-encoded-block-operation","examples":["BSTART.TGEMVMX 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, optional A scale, B CUBE_N8 primary, optional B scale; 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 20 and TileOperation_TGEMV_MX.","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. Required E8M0 scales remain ordinary row-major Tiles.","TGEMV is Local-only: TransA and TransB are zero and every effective Shared binding rejects before effects.","Each matrix side independently requires an E8M0 scale exactly when its MX input type is not FP16 or BF16. 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":"row-scale"},{"field":"source2","role":"right-matrix"},{"field":"source3","role":"column-scale"}],"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 using row and column scale Tiles.","After complete preflight, execute TGEMV_MX 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-MX","mnemonic":"TGEMV_MX","summary":"Multiply the matrix by the vector using row and column scale Tiles.","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-MX-CONTRACT-0014// ndf: kind=contract level=L1 layer=tile status=accepted5// TGEMV_MX MUST select CUBE Function 20 and TileOperation_TGEMV_MX.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// Each matrix side independently requires an E8M0 scale exactly when its MX11// input type is not FP16 or BF16. M is fixed to one and every Shared binding12// 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-MX-CONTRACT-00120// DOC-BEGIN: decode21readonly func InstructionContractOperation_TGEMV_MX()22    => TileOperation23begin24    return TileOperation_TGEMV_MX;25end;26// DOC-END: decode27// DOC-BEGIN: operation28readonly func InstructionContractCubeFunction_TGEMV_MX()29    => integer {0..31}30begin31    return 20;32end;33
34readonly func InstructionContractSharedOperandsAllowed_TGEMV_MX()35    => boolean36begin37    return FALSE;38end;39
40readonly func InstructionContractOperandsLegal_TGEMV_MX(41    destination: TileIndex,42    left_vector: TileIndex,43    row_scale: TileIndex,44    right_matrix: TileIndex,45    column_scale: TileIndex) => boolean46begin47    return TileOperandsLegal_TGEMV_MX(48        destination,49        left_vector,50        row_scale,51        right_matrix,52        column_scale);53end;54
55readonly func InstructionContractHandler_TGEMV_MX()56    => TileSemanticHandler57begin58    return TileHandler_TGEMV_MX;59end;60// DOC-END: operation61