* presentation-inlined.c:13:3: === analyzing loop === [pass=vect] [count(precise)=76800000] ** presentation-inlined.c:13:3: === analyze_loop_nest === [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_analyze_loop_form === [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: === get_loop_niters === [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: symbolic number of iterations is (unsigned int) n_8(D) [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_analyze_data_refs === [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: got vectype for stmt: _4 = *_3;vector(4) int [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_analyze_scalar_cycles === [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: Analyze phi: i_17 = PHI <0(5), i_10(6)> [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: Access function of PHI: {0, +, 1}_1 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: step: 1, init: 0 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: Detected induction [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: Analyze phi: sum_20 = PHI <0(5), _12(6)> [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: Access function of PHI: {0, +, _4}_1 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: step: _4, init: 0 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: step unknown [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: Analyze phi: sum_20 = PHI <0(5), _12(6)> [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: detected reduction: _12 = _4 + sum_20; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: Detected reduction [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_pattern_recog === [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand _1 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: def_stmt: _1 = (long unsigned int) i_17; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: type of def: internal [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand i_17 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: def_stmt: i_17 = PHI <0(5), i_10(6)> [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: type of def: induction [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand 4 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand _4 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: def_stmt: _4 = *_3; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: type of def: internal [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand _4 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: def_stmt: _4 = *_3; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: type of def: internal [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand _4 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: def_stmt: _4 = *_3; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: type of def: internal [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_analyze_data_ref_accesses === [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_mark_stmts_to_be_vectorized === [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand _4 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: def_stmt: _4 = *_3; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: type of def: internal [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand _4 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: def_stmt: _4 = *_3; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: type of def: internal [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand sum_20 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: def_stmt: sum_20 = PHI <0(5), _12(6)> [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: type of def: reduction [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand 0 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand _12 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: def_stmt: _12 = _4 + sum_20; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: type of def: reduction [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_analyze_data_ref_dependences === [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_determine_vectorization_factor === [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining phi: i_17 = PHI <0(5), i_10(6)> [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining phi: sum_20 = PHI <0(5), _12(6)> [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: get vectype for scalar type: int [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vectype: vector(4) int [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: nunits = 4 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: # DEBUG sum => sum_20 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: # DEBUG i => i_17 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: # DEBUG BEGIN_STMT [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: _1 = (long unsigned int) i_17; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: _2 = _1 * 4; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: _3 = arr_9(D) + _2; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: _4 = *_3; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: get vectype for scalar type: int [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vectype: vector(4) int [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: nunits = 4 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: # DEBUG x => _4 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: # DEBUG a => &sum [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: # DEBUG BEGIN_STMT [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: _12 = _4 + sum_20; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: get vectype for scalar type: int [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vectype: vector(4) int [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: get vectype for scalar type: int [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vectype: vector(4) int [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: nunits = 4 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: # DEBUG sum => _12 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: # DEBUG x => NULL [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: # DEBUG a => NULL [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: i_10 = i_17 + 1; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: # DEBUG i => i_10 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: # DEBUG sum => _12 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: # DEBUG i => i_10 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: ==> examining statement: if (n_8(D) > i_10) [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vectorization factor = 4 [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_analyze_slp === [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_make_slp_decision === [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_analyze_data_refs_alignment === [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: recording new base alignment for arr_9(D) alignment: 4 misalignment: 0 based on: _4 = *_3; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_compute_data_ref_alignment: [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: can't force alignment of ref: *_3 [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_prune_runtime_alias_test_list === [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_enhance_data_refs_alignment === [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: Unknown misalignment, naturally aligned [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: Analyze phi: i_17 = PHI <0(5), i_10(6)> [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: Analyze phi: sum_20 = PHI <0(5), _12(6)> [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: reduc or virtual phi. skip [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_get_data_access_cost: inside_cost = 12, outside_cost = 0. [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: cost model: epilogue peel iters set to vf/2 because loop iterations are unknown [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_get_data_access_cost: inside_cost = 12, outside_cost = 0. [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: cost model: epilogue peel iters set to vf/2 because loop iterations are unknown [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: Vectorizing an unaligned access [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: === vect_analyze_loop_operations === [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: examining phi: i_17 = PHI <0(5), i_10(6)> [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: examining phi: sum_20 = PHI <0(5), _12(6)> [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand *_3 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: not ssa-name [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: use not simple [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand *_3 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: not ssa-name [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: use not simple [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: can't use a fully-masked loop because the target doesn't have the appropriate masked load or store. [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand _4 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: def_stmt: _4 = *_3; [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: type of def: internal [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: vect_is_simple_use: operand sum_20 [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: def_stmt: sum_20 = PHI <0(5), _12(6)> [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: type of def: reduction [pass=vect] [count(precise)=76800000] **** presentation-inlined.c:13:3: reduc op not supported by target [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: not using a fully-masked loop [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: cost model: epilogue peel iters set to vf/2 because loop iterations are unknown [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: Cost model analysis: Vector inside of loop cost: 16 Vector prologue cost: 36 Vector epilogue cost: 52 Scalar iteration cost: 16 Scalar outside cost: 32 Vector outside cost: 88 prologue iterations: 0 epilogue iterations: 2 [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: Runtime profitability threshold = 5 [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: Static estimate profitability threshold = 9 [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: epilog loop required [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: Analyze phi: i_17 = PHI <0(5), i_10(6)> [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: Analyze phi: sum_20 = PHI <0(5), _12(6)> [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: reduc or virtual phi. skip [pass=vect] [count(precise)=76800000] ** presentation-inlined.c:13:3: loop vectorized [pass=vect] [count(precise)=76800000] ** presentation-inlined.c:13:3: === vec_transform_loop === [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: Profitability threshold is 5 loop iterations [pass=vect] [count(precise)=76800000] *** presentation-inlined.c:13:3: Analyze phi: i_17 = PHI [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: Analyze phi: sum_20 = PHI <_12(6), 0(9)> [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: reduc or virtual phi. skip [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: vect_update_ivs_after_vectorizer: phi: i_17 = PHI [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: vect_update_ivs_after_vectorizer: phi: sum_20 = PHI <_12(6), 0(9)> [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: reduc or virtual phi. skip. [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing phi: i_17 = PHI [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing phi: sum_20 = PHI <_12(6), 0(16)> [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: transform phi [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing phi: vect__12.4_33 = PHI <(6), (16)> [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: # DEBUG sum => sum_20 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: # DEBUG i => i_17 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: # DEBUG BEGIN_STMT [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: _1 = (long unsigned int) i_17; [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: _2 = _1 * 4; [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: _3 = arr_9(D) + _2; [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: _4 = *_3; [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: transform statement [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: create vector_type-pointer variable to type: vector(4) int vectorizing a pointer ref: *arr_9(D) [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: created arr_9(D) [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: # DEBUG x => _4 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: # DEBUG a => &sum [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: # DEBUG BEGIN_STMT [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: _12 = _4 + sum_20; [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: transform statement [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: vect_is_simple_use: operand _4 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: def_stmt: _4 = *_3; [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: type of def: internal [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: vect_is_simple_use: operand sum_20 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: def_stmt: sum_20 = PHI <_12(6), 0(16)> [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: type of def: reduction [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: reduc op not supported by target [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: transform reduction [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: vect_get_vec_def_for_operand: _4 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: vect_is_simple_use: operand _4 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: def_stmt: _4 = *_3; [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: type of def: internal [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: def_stmt = _4 = *_3; [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: vect_get_vec_def_for_operand: sum_20 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: vect_is_simple_use: operand sum_20 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: def_stmt: sum_20 = PHI <_12(6), 0(16)> [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: type of def: reduction [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: def_stmt = sum_20 = PHI <_12(6), 0(16)> [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: vect_is_simple_use: operand 0 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: transform reduction: created def-use cycle: vect__12.4_33 = PHI vect__12.8_37 = vect__4.7_36 + vect__12.4_33; [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: Reduce using vector shifts [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: extract scalar result [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: # DEBUG sum => _12 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: # DEBUG x => NULL [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: # DEBUG a => NULL [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: i_10 = i_17 + 1; [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: # DEBUG i => i_10 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: # DEBUG sum => _12 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: # DEBUG i => i_10 [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: vectp_arr.5_35 = vectp_arr.5_34 + 16; [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: ------>vectorizing statement: if (n_8(D) > i_10) [pass=vect] [count(guessed)=69120000] *** presentation-inlined.c:13:3: New loop exit condition: if (ivtmp_46 < bnd.1_29) [pass=vect] *** presentation-inlined.c:13:3: LOOP VECTORIZED [pass=vect] * presentation-not-inlined.c:9:5: not inlinable: compute_sum_without_inlining/0 -> accumulate/1: function body not available [pass=inline] [count(precise)=76800000] * presentation-not-inlined.c:8:3: === analyzing loop === [pass=vect] [count(precise)=76800000] ** presentation-not-inlined.c:8:3: === analyze_loop_nest === [pass=vect] [count(precise)=76800000] *** presentation-not-inlined.c:8:3: === vect_analyze_loop_form === [pass=vect] [count(precise)=76800000] **** presentation-not-inlined.c:8:3: === get_loop_niters === [pass=vect] [count(precise)=76800000] *** presentation-not-inlined.c:8:3: symbolic number of iterations is (unsigned int) n_9(D) [pass=vect] [count(precise)=76800000] *** presentation-not-inlined.c:8:3: not vectorized: loop contains function calls or data references that cannot be analyzed [pass=vect] [count(precise)=76800000] * presentation-not-inlined.c:8:3: vectorized 0 loops in function [pass=vect] [count(precise)=76800000] * loop-interchange.c:36:3: === analyzing loop === [pass=vect] [count(precise)=5346680] ** loop-interchange.c:36:3: === analyze_loop_nest === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_analyze_loop_form === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: === get_loop_niters === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_analyze_data_refs === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: got vectype for stmt: u[i_13] = 2.0e+0;vector(2) double [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_analyze_scalar_cycles === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: Analyze phi: i_13 = PHI <0(2), i_6(7)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: Access function of PHI: {0, +, 1}_1 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: step: 1, init: 0 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: Detected induction [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: Analyze phi: .MEM_14 = PHI <.MEM_4(D)(2), .MEM_5(7)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: Analyze phi: ivtmp_2 = PHI <1782225(2), ivtmp_11(7)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: Access function of PHI: {1782225, +, 4294967295}_1 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: step: 4294967295, init: 1782225 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: Detected induction [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_pattern_recog === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_analyze_data_ref_accesses === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_mark_stmts_to_be_vectorized === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_analyze_data_ref_dependences === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_determine_vectorization_factor === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: ==> examining phi: i_13 = PHI <0(2), i_6(7)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: ==> examining phi: .MEM_14 = PHI <.MEM_4(D)(2), .MEM_5(7)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: ==> examining phi: ivtmp_2 = PHI <1782225(2), ivtmp_11(7)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: ==> examining statement: # DEBUG i => i_13 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: ==> examining statement: # DEBUG BEGIN_STMT [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: ==> examining statement: u[i_13] = 2.0e+0; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: get vectype for scalar type: double [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: vectype: vector(2) double [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: nunits = 2 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: ==> examining statement: i_6 = i_13 + 1; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: ==> examining statement: # DEBUG i => i_6 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: ==> examining statement: # DEBUG i => i_6 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: ==> examining statement: ivtmp_11 = ivtmp_2 - 1; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: ==> examining statement: if (ivtmp_11 != 0) [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: vectorization factor = 2 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_analyze_slp === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_make_slp_decision === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: vectorization_factor = 2, niters = 1782225 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_analyze_data_refs_alignment === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: recording new base alignment for &u alignment: 16 misalignment: 0 based on: u[i_13] = 2.0e+0; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: vect_compute_data_ref_alignment: [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: misalign = 0 bytes of ref u[i_13] [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_prune_runtime_alias_test_list === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_enhance_data_refs_alignment === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: Analyze phi: i_13 = PHI <0(2), i_6(7)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: Analyze phi: .MEM_14 = PHI <.MEM_4(D)(2), .MEM_5(7)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: reduc or virtual phi. skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: Analyze phi: ivtmp_2 = PHI <1782225(2), ivtmp_11(7)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: vect_get_data_access_cost: inside_cost = 12, outside_cost = 0. [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: === vect_analyze_loop_operations === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: examining phi: i_13 = PHI <0(2), i_6(7)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: examining phi: .MEM_14 = PHI <.MEM_4(D)(2), .MEM_5(7)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: examining phi: ivtmp_2 = PHI <1782225(2), ivtmp_11(7)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: vect_is_simple_use: operand 2.0e+0 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:36:3: can't use a fully-masked loop because the target doesn't have the appropriate masked load or store. [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: not using a fully-masked loop [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: Cost model analysis: Vector inside of loop cost: 12 Vector prologue cost: 4 Vector epilogue cost: 12 Scalar iteration cost: 12 Scalar outside cost: 0 Vector outside cost: 16 prologue iterations: 0 epilogue iterations: 1 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: Runtime profitability threshold = 2 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: Static estimate profitability threshold = 2 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: epilog loop required [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: Analyze phi: i_13 = PHI <0(2), i_6(7)> [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: Analyze phi: .MEM_14 = PHI <.MEM_4(D)(2), .MEM_5(7)> [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: reduc or virtual phi. skip [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: Analyze phi: ivtmp_2 = PHI <1782225(2), ivtmp_11(7)> [pass=vect] [count(precise)=5346680] ** loop-interchange.c:36:3: loop vectorized [pass=vect] [count(precise)=5346680] ** loop-interchange.c:36:3: === vec_transform_loop === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: Analyze phi: i_13 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: Analyze phi: .MEM_14 = PHI <.MEM_5(7), .MEM_4(D)(2)> [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: reduc or virtual phi. skip [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: Analyze phi: ivtmp_2 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: vect_update_ivs_after_vectorizer: phi: i_13 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: vect_update_ivs_after_vectorizer: phi: .MEM_14 = PHI <.MEM_5(7), .MEM_4(D)(2)> [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: reduc or virtual phi. skip. [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: vect_update_ivs_after_vectorizer: phi: ivtmp_2 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: ------>vectorizing phi: i_13 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: ------>vectorizing phi: .MEM_14 = PHI <.MEM_5(7), .MEM_4(D)(12)> [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: ------>vectorizing phi: ivtmp_2 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: ------>vectorizing statement: # DEBUG i => i_13 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: ------>vectorizing statement: # DEBUG BEGIN_STMT [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: ------>vectorizing statement: u[i_13] = 2.0e+0; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: transform statement [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: vect_is_simple_use: operand 2.0e+0 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: vect_get_vec_def_for_operand: 2.0e+0 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: vect_is_simple_use: operand 2.0e+0 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: create vector_type-pointer variable to type: vector(2) double vectorizing an array ref: u [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: created &u [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: ------>vectorizing statement: i_6 = i_13 + 1; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: ------>vectorizing statement: # DEBUG i => i_6 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: ------>vectorizing statement: # DEBUG i => i_6 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: ------>vectorizing statement: ivtmp_11 = ivtmp_2 - 1; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: ------>vectorizing statement: vectp_u.3_22 = vectp_u.3_21 + 16; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: ------>vectorizing statement: if (ivtmp_11 != 0) [pass=vect] [count(precise)=5346680] *** loop-interchange.c:36:3: New loop exit condition: if (ivtmp_25 < 891112) [pass=vect] *** loop-interchange.c:36:3: LOOP VECTORIZED [pass=vect] * loop-interchange.c:20:5: === considering loop for interchange === [pass=linterchange] [count(precise)=5346680] ** loop-interchange.c:20:5: === prepare_perfect_loop_nest === [pass=linterchange] [count(precise)=5346680] *** interchange gives better data locality behavior: iloop_strides: 10680 > (oloop_strides: 8 * ratio: 3) [pass=linterchange] *** loop-interchange.c:19:3: consider loop interchange for loop_nest<2 - 3> [pass=linterchange] [count(precise)=4005] ** loop-interchange.c:19:3: === tree_loop_interchange::interchange === [pass=linterchange] [count(precise)=4005] *** interchange gives better data locality behavior: iloop_strides: 10680 > (oloop_strides: 8 * ratio: 3) [pass=linterchange] *** loop-interchange.c:19:3: Loop_pair is interchanged [pass=linterchange] [count(precise)=4005] *** loop-interchange.c:20:25: removing dead stmt: j_8 = j_17 + 1; [pass=linterchange] [count(precise)=5346680] *** loop-interchange.c:19:23: removing dead stmt: i_9 = i_26 + 1; [pass=linterchange] [count(precise)=5346680] *** removing dead stmt: ivtmp_30 = ivtmp_24 - 1; [pass=linterchange] [count(precise)=5346680] *** removing dead stmt: ivtmp_32 = ivtmp_6 - 1; [pass=linterchange] [count(precise)=5346680] *** loop-interchange.c:19:3: loops interchanged in loop nest [pass=linterchange] [count(precise)=4005] * loop-interchange.c:20:5: === analyzing loop === [pass=vect] [count(precise)=5346680] ** loop-interchange.c:20:5: === analyze_loop_nest === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_analyze_loop_form === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: === get_loop_niters === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_analyze_data_refs === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: got vectype for stmt: _5 = u[_4];vector(2) double [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_analyze_scalar_cycles === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Analyze phi: sum_40 = PHI [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Access function of PHI: {sum_42, +, _5}_3 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: step: _5, init: sum_42 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: step unknown [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Analyze phi: i_26 = PHI <0(5), i_25(11)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Access function of PHI: {0, +, 1}_3 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: step: 1, init: 0 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Detected induction [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Analyze phi: ivtmp_24 = PHI <1335(5), ivtmp_19(11)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Access function of PHI: {1335, +, 4294967295}_3 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: step: 4294967295, init: 1335 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Detected induction [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Analyze phi: ivtmp_45 = PHI <1335(5), ivtmp_46(11)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Access function of PHI: {1335, +, 4294967295}_3 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: step: 4294967295, init: 1335 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Detected induction [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Analyze phi: sum_40 = PHI [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: detected reduction: sum_7 = _5 + sum_40; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Detected reduction [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_pattern_recog === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_is_simple_use: operand j_17 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: def_stmt: j_17 = PHI <0(2), j_10(10)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: type of def: external [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_is_simple_use: operand _5 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: def_stmt: _5 = u[_4]; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: type of def: internal [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_is_simple_use: operand _5 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: def_stmt: _5 = u[_4]; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: type of def: internal [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_is_simple_use: operand _5 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: def_stmt: _5 = u[_4]; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: type of def: internal [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_analyze_data_ref_accesses === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_mark_stmts_to_be_vectorized === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_is_simple_use: operand _5 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: def_stmt: _5 = u[_4]; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: type of def: internal [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_is_simple_use: operand _5 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: def_stmt: _5 = u[_4]; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: type of def: internal [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_is_simple_use: operand sum_40 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: def_stmt: sum_40 = PHI [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: type of def: reduction [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_is_simple_use: operand sum_42 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: def_stmt: sum_42 = PHI <0.0(2), sum_37(10)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: type of def: external [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_is_simple_use: operand sum_7 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: def_stmt: sum_7 = _5 + sum_40; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: type of def: reduction [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_analyze_data_ref_dependences === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_determine_vectorization_factor === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining phi: sum_40 = PHI [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: get vectype for scalar type: double [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vectype: vector(2) double [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: nunits = 2 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining phi: i_26 = PHI <0(5), i_25(11)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining phi: ivtmp_24 = PHI <1335(5), ivtmp_19(11)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining phi: ivtmp_45 = PHI <1335(5), ivtmp_46(11)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG sum => NULL [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG i => i_26 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG sum => NULL [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG j => 0 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG sum => sum_40 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG j => j_17 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG BEGIN_STMT [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: _2 = j_17 * 1335; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: _4 = _2 + i_26; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: _5 = u[_4]; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: get vectype for scalar type: double [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vectype: vector(2) double [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: nunits = 2 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: sum_7 = _5 + sum_40; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: get vectype for scalar type: double [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vectype: vector(2) double [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: get vectype for scalar type: double [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vectype: vector(2) double [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: nunits = 2 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG sum => sum_7 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG D#2 => j_17 + 1 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG j => D#2 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG sum => sum_7 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG j => D#2 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG i => NULL [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG sum => NULL [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: # DEBUG i => NULL [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: ivtmp_19 = ivtmp_24 - 1; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: i_25 = i_26 + 1; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: ivtmp_46 = ivtmp_45 - 1; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: ==> examining statement: if (ivtmp_46 != 0) [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vectorization factor = 2 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_analyze_slp === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_make_slp_decision === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: vectorization_factor = 2, niters = 1335 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_analyze_data_refs_alignment === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: recording new base alignment for &u alignment: 16 misalignment: 0 based on: _5 = u[_4]; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_compute_data_ref_alignment: [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Unknown alignment for access: u[_4] [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_prune_runtime_alias_test_list === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_enhance_data_refs_alignment === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Unknown misalignment, naturally aligned [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Analyze phi: sum_40 = PHI [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: reduc or virtual phi. skip [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Analyze phi: i_26 = PHI <0(5), i_25(11)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Analyze phi: ivtmp_24 = PHI <1335(5), ivtmp_19(11)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Analyze phi: ivtmp_45 = PHI <1335(5), ivtmp_46(11)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_get_data_access_cost: inside_cost = 12, outside_cost = 0. [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_get_data_access_cost: inside_cost = 12, outside_cost = 0. [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: Vectorizing an unaligned access [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: === vect_analyze_loop_operations === [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: examining phi: sum_40 = PHI [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: examining phi: i_26 = PHI <0(5), i_25(11)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: examining phi: ivtmp_24 = PHI <1335(5), ivtmp_19(11)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: examining phi: ivtmp_45 = PHI <1335(5), ivtmp_46(11)> [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: num. args = 4 (not unary/binary/ternary op). [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_is_simple_use: operand u[_4] [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: not ssa-name [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: use not simple [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: can't use a fully-masked loop because the target doesn't have the appropriate masked load or store. [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_is_simple_use: operand _5 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: def_stmt: _5 = u[_4]; [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: type of def: internal [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: vect_is_simple_use: operand sum_40 [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: def_stmt: sum_40 = PHI [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: type of def: reduction [pass=vect] [count(precise)=5346680] **** loop-interchange.c:20:5: reduc op not supported by target [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: not using a fully-masked loop [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: Cost model analysis: Vector inside of loop cost: 24 Vector prologue cost: 4 Vector epilogue cost: 44 Scalar iteration cost: 16 Scalar outside cost: 0 Vector outside cost: 48 prologue iterations: 0 epilogue iterations: 1 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: Runtime profitability threshold = 10 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: Static estimate profitability threshold = 10 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: epilog loop required [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: Analyze phi: sum_40 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: reduc or virtual phi. skip [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: Analyze phi: i_26 = PHI <0(5), i_25(11)> [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: Analyze phi: ivtmp_24 = PHI <1335(5), ivtmp_19(11)> [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: Analyze phi: ivtmp_45 = PHI <1335(5), ivtmp_46(11)> [pass=vect] [count(precise)=5346680] ** loop-interchange.c:20:5: loop vectorized [pass=vect] [count(precise)=5346680] ** loop-interchange.c:20:5: === vec_transform_loop === [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: Analyze phi: sum_40 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: reduc or virtual phi. skip [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: Analyze phi: i_26 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: Analyze phi: ivtmp_24 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: Analyze phi: ivtmp_45 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: vect_update_ivs_after_vectorizer: phi: sum_40 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: reduc or virtual phi. skip. [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: vect_update_ivs_after_vectorizer: phi: i_26 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: vect_update_ivs_after_vectorizer: phi: ivtmp_24 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: vect_update_ivs_after_vectorizer: phi: ivtmp_45 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing phi: sum_40 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: transform phi [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing phi: i_26 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing phi: ivtmp_24 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing phi: ivtmp_45 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing phi: vect_sum_7.15_51 = PHI <(11), (16)> [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG sum => NULL [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG i => i_26 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG sum => NULL [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG j => 0 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG sum => sum_40 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG j => j_17 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG BEGIN_STMT [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: _2 = j_17 * 1335; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: _4 = _2 + i_26; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: _5 = u[_4]; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: transform statement [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: create vector_type-pointer variable to type: vector(2) double vectorizing an array ref: u [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: created vectp_u.17_52 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: sum_7 = _5 + sum_40; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: transform statement [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: vect_is_simple_use: operand _5 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: def_stmt: _5 = u[_4]; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: type of def: internal [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: vect_is_simple_use: operand sum_40 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: def_stmt: sum_40 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: type of def: reduction [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: reduc op not supported by target [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: transform reduction [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: vect_get_vec_def_for_operand: _5 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: vect_is_simple_use: operand _5 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: def_stmt: _5 = u[_4]; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: type of def: internal [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: def_stmt = _5 = u[_4]; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: vect_get_vec_def_for_operand: sum_40 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: vect_is_simple_use: operand sum_40 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: def_stmt: sum_40 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: type of def: reduction [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: def_stmt = sum_40 = PHI [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: vect_is_simple_use: operand sum_42 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: def_stmt: sum_42 = PHI <0.0(2), sum_37(10)> [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: type of def: external [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: transform reduction: created def-use cycle: vect_sum_7.15_51 = PHI vect_sum_7.19_59 = vect__5.18_58 + vect_sum_7.15_51; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: Reduce using vector shifts [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: extract scalar result [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG sum => sum_7 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG D#2 => j_17 + 1 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG j => D#2 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG sum => sum_7 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG j => D#2 [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG i => NULL [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG sum => NULL [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: # DEBUG i => NULL [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: ivtmp_19 = ivtmp_24 - 1; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: i_25 = i_26 + 1; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: ivtmp_46 = ivtmp_45 - 1; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: vectp_u.16_57 = vectp_u.16_56 + 16; [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: ------>vectorizing statement: if (ivtmp_46 != 0) [pass=vect] [count(precise)=5346680] *** loop-interchange.c:20:5: New loop exit condition: if (ivtmp_66 < 667) [pass=vect] *** loop-interchange.c:20:5: LOOP VECTORIZED [pass=vect] * replaced redundant expr _50 = (void *) ivtmp.43_26; with vectp_u.17_52 [pass=dom] [count(precise)=5346680] * loop-opts.c:7:7: not inlinable: loop_opts_c_body/0 -> not_inlinable/1: function body not available [pass=inline] [count(precise)=300000] * main.c:19:7: not inlinable: main/12 -> compute_sum_without_inlining/18: function body not available [pass=inline] [count(precise)=300000] * main.c:18:7: not inlinable: main/12 -> compute_sum_with_inlining/17: function body not available [pass=inline] [count(precise)=300000] * main.c:17:7: not inlinable: main/12 -> workload_c_body/16: function body not available [pass=inline] [count(precise)=300000] * main.c:16:7: not inlinable: main/12 -> loop_opts_c_body/15: function body not available [pass=inline] [count(precise)=300000] * main.c:15:7: not inlinable: main/12 -> devirt_cc_body/14: function body not available [pass=inline] [count(precise)=300000] * main.c:13:3: === analyzing loop === [pass=vect] [count(precise)=300000] ** main.c:13:3: === analyze_loop_nest === [pass=vect] [count(precise)=300000] *** main.c:13:3: === vect_analyze_loop_form === [pass=vect] [count(precise)=300000] **** main.c:13:3: === get_loop_niters === [pass=vect] [count(precise)=300000] *** main.c:13:3: not vectorized: loop contains function calls or data references that cannot be analyzed [pass=vect] [count(precise)=300000] * main.c:13:3: vectorized 0 loops in function [pass=vect] [count(precise)=300000] * replaced redundant expr _25 = (unsigned int) n_8(D); with niters.0_19 [pass=dom] [count(precise)=298833] * replaced redundant expr _24 = (unsigned long) arr_10(D); with ivtmp.3_16 [pass=dom] [count(precise)=298833] * loop-interchange.c:23:3: === considering loop for interchange === [pass=linterchange] [count(precise)=4005] ** loop-interchange.c:23:3: === prepare_perfect_loop_nest === [pass=linterchange] [count(precise)=4005] * loop-interchange.c:19:3: === analyzing loop === [pass=vect] [count(precise)=4005] ** loop-interchange.c:19:3: === analyze_loop_nest === [pass=vect] [count(precise)=4005] *** loop-interchange.c:19:3: === vect_analyze_loop_form === [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: === vect_analyze_loop_form === [pass=vect] [count(precise)=4005] ***** loop-interchange.c:19:3: === get_loop_niters === [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Considering outer-loop vectorization [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: === get_loop_niters === [pass=vect] [count(precise)=4005] *** loop-interchange.c:19:3: === vect_analyze_data_refs === [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: analyze in outer loop: *(&u + (sizetype) (j_17 * 1335) * 8) [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: outer base_address: &u outer offset from base address: 0 outer constant offset from base address: 0 outer step: 10680 outer base alignment: 16 outer base misalignment: 0 outer offset alignment: 128 outer step alignment: 8 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: got vectype for stmt: _5 = u[_4];vector(2) double [pass=vect] [count(precise)=4005] *** loop-interchange.c:19:3: === vect_analyze_scalar_cycles === [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Analyze phi: sum_42 = PHI <0.0(2), sum_37(10)> [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Access function of PHI: sum_42 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Analyze phi: j_17 = PHI <0(2), j_10(10)> [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Access function of PHI: {0, +, 1}_2 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: step: 1, init: 0 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Detected induction [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Analyze phi: ivtmp_6 = PHI <1335(2), ivtmp_3(10)> [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Access function of PHI: {1335, +, 4294967295}_2 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: step: 4294967295, init: 1335 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Detected induction [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Analyze phi: ivtmp_1 = PHI <1335(2), ivtmp_44(10)> [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Access function of PHI: {1335, +, 4294967295}_2 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: step: 4294967295, init: 1335 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Detected induction [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Analyze phi: sum_42 = PHI <0.0(2), sum_37(10)> [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: detected double reduction: sum_37 = PHI [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Detected double reduction [pass=vect] [count(precise)=4005] *** loop-interchange.c:19:3: === vect_analyze_scalar_cycles === [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Analyze phi: sum_40 = PHI [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Access function of PHI: {sum_42, +, _5}_3 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: step: _5, init: sum_42 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: step unknown [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Analyze phi: i_26 = PHI <0(5), i_25(11)> [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Access function of PHI: {0, +, 1}_3 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: step: 1, init: 0 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Detected induction [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Analyze phi: ivtmp_24 = PHI <1335(5), ivtmp_19(11)> [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Access function of PHI: {1335, +, 4294967295}_3 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: step: 4294967295, init: 1335 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Detected induction [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Analyze phi: ivtmp_45 = PHI <1335(5), ivtmp_46(11)> [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Access function of PHI: {1335, +, 4294967295}_3 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: step: 4294967295, init: 1335 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Detected induction [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Analyze phi: sum_40 = PHI [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: detected reduction: sum_7 = _5 + sum_40; [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: Detected vectorizable nested cycle [pass=vect] [count(precise)=4005] *** loop-interchange.c:19:3: === vect_pattern_recog === [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: vect_is_simple_use: operand j_17 [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: def_stmt: j_17 = PHI <0(2), j_10(10)> [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: type of def: induction [pass=vect] [count(precise)=4005] *** loop-interchange.c:19:3: === vect_analyze_data_ref_accesses === [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: grouped access in outer loop. [pass=vect] [count(precise)=4005] **** loop-interchange.c:19:3: not vectorized: complicated access pattern. [pass=vect] [count(precise)=4005] *** loop-interchange.c:19:3: bad data access [pass=vect] [count(precise)=4005] * replaced redundant expr _8 = (void *) ivtmp.29_37; with _48 [pass=dom] [count(precise)=4005] * loop-interchange.c:39:3: not inlinable: loop_interchange_c_main/2 -> foo.constprop/4: function not inlinable [pass=inline] [count(precise)=3] * loop-interchange.c:39:3: not inlinable: loop_interchange_c_main/2 -> foo.constprop/4: function not inlinable [pass=inline] [count(precise)=3] * loop-interchange.c:39:3: not inlinable: loop_interchange_c_main/2 -> foo.constprop/4: function not inlinable [pass=inline] [count(precise)=3] * loop-interchange.c:48:10: not attempting to optimize loop nest [pass=vect] [count(precise)=3] * replaced redundant expr _2 = (unsigned long) &u; with ivtmp.43_9 [pass=dom] [count(precise)=3] * replaced redundant expr ivtmp.29_49 = (unsigned long) &u; with ivtmp.43_9 [pass=dom] [count(adjusted)=3] * replaced redundant expr _30 = (unsigned long) &u; with ivtmp.43_9 [pass=dom] [count(adjusted)=3] * main.c:24:3: not inlinable: main/12 -> loop_interchange_c_main/21: function body not available [pass=inline] [count(precise)=3] * main.c:22:3: not inlinable: main/12 -> printf/20: function body not available [pass=inline] [count(precise)=3] * main.c:11:3: not inlinable: main/12 -> __builtin_puts/13: function body not available [pass=inline] [count(precise)=3] * presentation-inlined.c:14:5: inlining accumulate/0 into compute_sum_with_inlining/1 [pass=einline] * presentation-inlined.c:14:5: Inlining accumulate/2 to compute_sum_with_inlining/1 with frequency 1.00 [pass=einline] * presentation-inlined.c:13:3: vectorized 1 loops in function [pass=vect] * vectorized 0 loops in function [pass=vect] * loop-interchange.c:46:5: not inlinable: loop_interchange_c_main/2 -> abort/3: function body not available [pass=inline] [count(precise)=0] * loop-interchange.c:36:3: vectorized 1 loops in function [pass=vect] * loop-interchange.c:20:5: vectorized 1 loops in function [pass=vect] * devirt.cc:26:8: inlining void test(base*)/1 into void devirt_cc_body()/8 [pass=einline] * devirt.cc:26:8: Inlining void test(base*)/1 to void devirt_cc_body()/8 with frequency 1.00 [pass=einline] * devirt.cc:19:15: folding virtual function call to virtual void subclass::meth() [pass=einline] * vectorized 0 loops in function [pass=vect]