Firestorm: SMLAL2 (by element, 2D) measurements

Code:

  smlal2 v0.2d, v1.4s, v2.s[1]

Initialization

  movi v0.16b, 1
  movi v1.16b, 2
  movi v2.16b, 3

(no loop instructions)

1000 unrolls and 1 iteration

Retires: 1.000

Issues: 1.000

Integer unit issues: 0.000

Load/store unit issues: 0.000

SIMD/FP unit issues: 1.000

All non-zero counters (first 10 of 64 runs)

retire uop (01)	cycle (02)	03	1e	3f	4e	51	schedule uop (52)	schedule simd uop (54)	dispatch simd uop (57)	ldst uops in schedulers (5b)	60	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map simd uop (7e)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst simd alu (9a)	l1d tlb access (a0)	l1d cache writeback (a8)	ac	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	f5	f6	f7	f8	fd
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	1	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	2	2	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	1	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	12	61	2548	25	1000	1000	1000	398313	1	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3086
1004	3037	22	0	82	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	1	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	1	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	1	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038

Test 2: Latency 1->1

Code:

  smlal2 v0.2d, v1.4s, v2.s[1]

Initialization

  movi v0.16b, 1
  movi v1.16b, 2
  movi v2.16b, 3

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0037

All non-zero counters (first 10 of 64 runs)

retire uop (01)	cycle (02)	03	18	19	1e	3a	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6d	6e	map stall dispatch (70)	simd prf full (72)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d cache writeback (a8)	ac	c2	branch mispred nonspec (cb)	cf	d0	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	30037	225	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	29634	6	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	82	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	29670	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	7	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	3	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	232	0	0	0	0	6726	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	712	1	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	0	3	28745	10100	204	10000	200	30000	30084	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	3	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	492	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	527	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

All non-zero counters (first 10 of 64 runs)

retire uop (01)	cycle (02)	03	1e	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6b	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb miss (a1)	l1d cache writeback (a8)	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10024	30037	225	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	4	4	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	552	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	4	16	3	3	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	300	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	9	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10459	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10022	10	9	10	10	10000	10	0	0	640	3	16	3	3	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	29630	10000	10	30038	30038	30038	30038	30038

Test 3: Latency 1->2

Code:

  smlal2 v0.2d, v0.4s, v1.s[1]

Initialization

  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0037

All non-zero counters (first 10 of 64 runs)

retire uop (01)	cycle (02)	03	1e	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d cache writeback (a8)	ac	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	30037	225	87	690	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	225	30	726	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	225	30	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	225	369	61	29548	25	10100	100	10000	100	10000	500	4277735	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	225	27	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	225	21	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	225	33	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	225	27	156	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	225	39	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

All non-zero counters (first 10 of 64 runs)

retire uop (01)	cycle (02)	03	1e	1f	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb miss (a1)	l1d cache writeback (a8)	ac	cd	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10024	30037	225	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	6	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	224	12	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	510	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	3	29666	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30085	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	33	0	61	29548	25	10010	12	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	30	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	3	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29630	10000	10	30086	30038	30038	30038	30038
10024	30037	225	39	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038

Test 4: Latency 1->3

Code:

  smlal2 v0.2d, v1.4s, v0.s[1]

Initialization

  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0037

All non-zero counters (first 10 of 64 runs)

retire uop (01)	cycle (02)	03	18	19	1e	1f	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	ld unit uop (a6)	l1d cache writeback (a8)	ac	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	30037	225	0	0	33	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	24	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	15	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	0	710	2	16	2	2	29634	24	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	24	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30086	30037	28265	7	28762	10100	200	10166	202	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	27	0	726	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	24	0	251	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	2	0	0	710	2	16	2	2	29634	0	10000	100	30038	30085	30087	30038	30085
10204	30037	225	2	1	303	0	346	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	27	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	0	710	2	16	2	2	29634	0	10000	100	30038	30038	30038	30038	30038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

All non-zero counters (first 10 of 64 runs)

retire uop (01)	cycle (02)	03	09	l2 tlb miss data (0b)	1e	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6b	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d cache writeback (a8)	ac	branch cond mispred nonspec (c5)	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10024	30037	225	2	0	66	156	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	7	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	24	232	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	33	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	3	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	3	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	3	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	27	943	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	21	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	36	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  smlal2 v0.2d, v8.4s, v9.s[1]
  movi v1.16b, 0
  smlal2 v1.2d, v8.4s, v9.s[1]
  movi v2.16b, 0
  smlal2 v2.2d, v8.4s, v9.s[1]
  movi v3.16b, 0
  smlal2 v3.2d, v8.4s, v9.s[1]
  movi v4.16b, 0
  smlal2 v4.2d, v8.4s, v9.s[1]
  movi v5.16b, 0
  smlal2 v5.2d, v8.4s, v9.s[1]
  movi v6.16b, 0
  smlal2 v6.2d, v8.4s, v9.s[1]
  movi v7.16b, 0
  smlal2 v7.2d, v8.4s, v9.s[1]

Initialization

  movi v8.16b, 9
  movi v9.16b, 10

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.2508

All non-zero counters (first 10 of 64 runs)

retire uop (01)	cycle (02)	03	l2 tlb miss data (0b)	1e	3f	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6d	6e	rob full (74)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d cache writeback (a8)	ac	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	ec	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160204	20065	151	0	12	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	0	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	10111	1	16	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	150	0	0	39	25	80100	100	80000	100	80000	500	640000	1	20045	20064	20064	0	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	10111	1	16	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	150	0	0	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	0	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	10111	1	16	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	151	0	9	39	25	80100	100	80000	100	80000	500	640000	1	20045	20064	20064	0	3	22	80100	200	80000	200	240000	20064	20307	1	1	160201	100	99	100	100	160000	100	0	0	10111	1	16	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	150	0	0	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	0	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	3	10111	1	16	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	151	0	87	39	25	80100	100	80000	100	80000	500	640000	1	20045	20064	20064	0	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	10111	1	16	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	150	0	0	39	25	80100	100	80000	100	80000	500	640000	1	20045	20064	20064	0	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	10111	1	16	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	150	0	0	60	25	80100	100	80000	100	80000	500	640000	1	20045	20064	20064	0	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	10111	1	16	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	150	0	0	704	25	80100	100	80000	100	80000	500	640000	1	20045	20064	20064	0	3	22	80100	200	80000	200	240627	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	10111	1	16	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	150	0	0	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	0	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	10111	1	16	1	1	20061	0	160000	100	20065	20065	20065	20065	20065

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2506

All non-zero counters (first 10 of 64 runs)

retire uop (01)	cycle (02)	03	mmu table walk instruction (07)	1e	3f	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	5f	60	69	6d	6e	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d cache writeback (a8)	c2	cf	d0	d2	l1i cache miss demand (d3)	d5	map dispatch bubble (d6)	d9	da	db	dd	fetch restart (de)	e0	? int output thing (e9)	ea	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160024	20078	150	0	12	425	25	80012	12	80000	12	80000	62	640000	1	1	20027	20046	20046	3	22	80012	20	80000	20	240000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10028	3	1	1	9	20	2	1	1	4	7	20043	2	15	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	24	560	25	80012	12	80000	12	80000	62	640000	1	1	20027	20046	20046	3	22	80012	20	80000	20	240000	20050	20046	1	1	160021	10	9	10	10	160000	10	0	0	10030	3	1	1	4	20	2	1	1	4	7	20043	2	15	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	0	45	25	80012	12	80000	12	80000	62	640000	1	1	20027	20046	20046	3	22	80012	20	80000	20	240000	20046	20291	1	1	160021	10	9	10	10	160000	10	0	0	10029	6	1	1	5	20	2	1	1	3	7	20043	2	15	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	0	45	25	80012	12	80000	12	80000	62	640000	1	1	20027	20046	20046	3	22	80012	20	80000	20	240000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10030	3	1	1	9	20	2	1	1	4	9	20043	2	15	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	0	45	25	80012	12	80000	12	80000	62	640000	1	1	20027	20046	20046	3	22	80012	20	80000	20	240000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10029	3	1	1	7	20	2	1	1	6	4	20047	2	15	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	0	45	25	80012	12	80000	12	80000	62	640000	0	1	20027	20050	20046	3	22	80012	20	80000	20	240000	20050	20046	1	1	160021	10	9	10	10	160000	10	0	0	10028	3	1	1	7	20	2	1	1	6	4	20043	2	15	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	0	45	25	80012	12	80000	12	80000	62	640000	1	1	20027	20046	20046	3	22	80012	20	80000	20	240000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10027	3	1	1	5	20	2	2	1	6	9	20043	2	15	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	75	45	25	80012	12	80000	12	80000	62	640000	1	1	20027	20050	20046	3	22	80012	20	80000	20	240000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10027	3	1	1	6	20	2	1	1	6	7	20043	2	15	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	3	45	25	80012	12	80000	12	80000	62	640000	1	1	20027	20046	20046	3	22	80012	20	80000	20	240000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10029	3	1	2	5	20	2	1	1	5	4	20043	2	15	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	0	45	25	80012	12	80000	12	80000	62	640000	1	1	20027	20046	20046	3	22	80012	20	80000	20	240000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10028	3	1	1	3	20	2	1	1	8	9	20043	2	15	160000	10	20047	20047	20047	20047	20047

Test 6: throughput

Count: 12

Code:

  smlal2 v0.2d, v12.4s, v13.s[1]
  smlal2 v1.2d, v12.4s, v13.s[1]
  smlal2 v2.2d, v12.4s, v13.s[1]
  smlal2 v3.2d, v12.4s, v13.s[1]
  smlal2 v4.2d, v12.4s, v13.s[1]
  smlal2 v5.2d, v12.4s, v13.s[1]
  smlal2 v6.2d, v12.4s, v13.s[1]
  smlal2 v7.2d, v12.4s, v13.s[1]
  smlal2 v8.2d, v12.4s, v13.s[1]
  smlal2 v9.2d, v12.4s, v13.s[1]
  smlal2 v10.2d, v12.4s, v13.s[1]
  smlal2 v11.2d, v12.4s, v13.s[1]

Initialization

  movi v12.16b, 13
  movi v13.16b, 14

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.2503

All non-zero counters (first 10 of 64 runs)

retire uop (01)	cycle (02)	03	mmu table walk data (08)	l2 tlb miss data (0b)	18	19	1e	1f	37	3a	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6b	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb access (a0)	l1d tlb miss (a1)	ld unit uop (a6)	l1d cache writeback (a8)	a9	ac	c2	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
120204	30296	225	0	0	0	0	0	0	0	0	125	0	25	120100	100	120000	100	120000	500	4286184	0	30020	0	30039	30039	15128	3	14997	120100	200	120000	200	360000	30951	30039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	0	0	0	7610	1	16	1	6	30036	0	120000	100	30040	30040	30040	30040	30040
120204	30039	225	0	1	0	0	57	0	0	0	41	0	25	120100	100	120000	100	120104	500	960000	1	30020	0	30039	30039	14973	3	14997	120100	200	120000	200	360000	30039	30039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	0	0	0	7610	1	16	1	4	30036	0	120000	100	30040	30040	30040	30040	30040
120204	30039	232	0	0	0	0	0	0	52	0	105	0	25	120100	100	120000	100	120000	500	4286184	0	30020	0	30039	30039	14973	3	14997	120100	200	120000	200	360000	30039	30039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	0	0	0	7610	1	16	1	2	30036	0	120000	100	30040	30040	30040	30040	30040
120204	30039	231	0	0	0	0	21	0	0	0	146	0	25	120100	100	120000	100	120000	500	960000	1	30932	0	30039	30039	15009	3	14997	120100	200	120000	200	360000	30039	30039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	4	9	0	7610	1	16	1	3	30036	0	120000	100	30040	30040	30040	30040	30040
120204	30039	225	0	0	0	0	0	0	2	0	104	0	25	120100	100	120000	100	120000	500	4286184	0	30020	0	30039	30039	14973	3	14997	120100	200	120000	200	360000	30039	30039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	0	0	0	7610	1	16	1	4	30036	0	120000	100	30040	30952	30040	30040	30040
120204	30039	231	0	0	2	0	0	0	1	0	41	0	25	120100	100	120000	100	120000	500	960000	0	30020	0	30039	30039	14973	3	15636	120100	200	120000	200	360000	30039	30039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	0	0	0	7610	1	16	1	4	30036	0	120000	100	30040	30040	30040	30040	30040
120204	30039	225	0	0	0	0	0	0	0	0	41	0	25	120100	100	120000	100	120000	500	960000	1	30020	0	30039	30039	14973	3	14997	120100	200	120000	200	360000	30039	30039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	0	0	0	7610	1	16	1	2	30948	0	120000	100	30040	30040	30952	30040	30040
120204	30039	225	0	0	0	0	3	0	0	0	83	3523	25	120100	100	120000	100	120000	500	960000	0	30020	0	30039	30039	14973	3	15909	120100	200	120000	200	360000	30039	30039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	0	0	0	7610	1	16	1	4	30036	0	120000	100	30040	30040	30040	30040	30040
120204	30039	225	0	0	0	0	0	0	0	0	171	0	25	120100	100	120000	100	120000	500	4286184	0	30932	0	30039	30039	14973	3	14997	120100	200	120000	200	360000	30039	30039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	0	0	0	7610	1	16	1	10	30036	0	120000	100	30040	30040	30040	30040	30040
120204	30039	225	0	0	0	0	0	0	0	0	41	0	25	120100	100	120000	100	120000	500	960000	1	30020	3	30039	30039	14973	7	14997	120100	200	120000	200	360000	30039	30039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	0	0	0	7610	1	16	1	2	30036	0	120000	100	30040	30040	30040	30040	30040

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2503

All non-zero counters (first 10 of 64 runs)

retire uop (01)	cycle (02)	03	mmu table walk data (08)	l2 tlb miss instruction (0a)	l2 tlb miss data (0b)	18	19	1e	37	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	5f	60	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d cache writeback (a8)	a9	ac	c2	cf	d0	d2	l1i cache miss demand (d3)	d5	map dispatch bubble (d6)	d9	da	db	dd	fetch restart (de)	e0	? int output thing (e9)	ea	eb	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
120024	30039	225	0	0	0	0	0	0	0	46	0	25	120010	10	120000	10	120000	50	960000	1	1	30020	30039	30039	14989	3	15012	120010	20	120000	20	360000	30039	30039	1	1	120021	10	9	10	10	120000	10	0	0	18	0	7522	3	1	1	5	16	2	1	1	5	5	30036	0	20	9	120000	10	30040	30040	30040	30040	30040
120024	30922	225	0	0	0	0	0	0	0	88	0	25	120046	10	120036	10	120000	50	4283915	1	1	30020	30039	30039	14989	3	15012	120010	20	120000	20	360000	30039	30039	1	1	120021	10	9	10	10	120000	10	0	0	24	0	7522	3	1	1	6	16	2	1	1	6	6	30036	0	20	9	120000	10	30923	30040	30040	30040	30040
120024	31010	225	0	0	0	0	0	0	0	52	0	25	120010	10	120000	10	120000	50	960000	0	1	30020	30039	30039	14989	3	15012	120010	20	120000	20	360000	30039	30039	1	1	120021	10	9	10	10	120000	10	0	0	30	0	7524	3	2	1	4	16	4	2	2	5	4	30036	0	20	15	120000	10	30040	30040	30040	30040	30040
120024	30039	224	0	0	0	0	0	0	18	73	6713	25	120010	10	120000	10	120000	50	960000	0	1	30020	30039	30039	14989	3	15012	120010	20	120000	20	360000	30039	30039	1	1	120021	10	9	10	10	120000	10	0	0	27	0	7524	6	1	2	5	16	4	1	2	4	5	30036	0	40	9	120000	10	30040	30040	30040	30040	30040
120024	30039	225	0	0	0	0	0	0	0	46	0	25	120028	10	120000	10	120000	50	960000	1	1	30020	30039	30039	14989	3	15012	120010	20	120000	20	360000	30039	30039	1	1	120021	10	9	10	10	120000	10	0	0	0	0	7524	3	2	2	6	16	4	2	2	6	6	30036	0	40	15	120000	10	30040	30040	30040	31749	30040
120024	30039	225	0	0	0	0	0	0	0	52	0	25	120010	10	120000	10	120000	50	960000	0	1	30020	30039	30039	14989	3	15012	120010	20	120000	20	360000	30039	30039	1	1	120021	10	9	10	10	120000	10	0	0	90	0	7524	6	2	2	5	16	2	2	2	6	6	30036	0	40	15	120000	10	30040	30040	30040	30040	30040
120024	30041	225	0	0	0	0	0	0	0	52	0	25	120010	10	120000	10	120000	50	960000	0	1	30020	31010	31010	14989	3	15012	120010	20	120000	20	360000	30039	30922	1	1	120021	10	9	10	10	120000	10	0	0	75	0	7524	6	1	1	4	16	2	1	2	4	5	30036	0	40	9	120000	10	30040	30040	30040	30040	30040
120024	30039	225	0	0	0	0	0	0	0	52	0	25	120011	10	120000	10	120000	50	960000	0	1	30020	30039	30039	14989	3	15012	120010	20	120000	20	360000	30039	30039	1	1	120021	10	9	10	10	120000	10	0	0	105	0	7522	6	2	1	4	16	4	1	2	5	5	30036	0	40	15	120000	10	30040	30923	30040	30040	30040
120024	30039	225	0	0	0	0	0	0	0	52	0	25	120010	10	120000	10	120000	50	960000	1	1	30903	30039	31748	14989	3	15012	120010	20	120000	20	360000	30039	30039	1	1	120021	10	9	10	10	120000	10	0	0	96	0	7524	6	2	2	5	16	2	2	2	5	4	30036	0	40	9	120000	10	30040	30040	30040	30040	30040
120024	30039	225	0	0	0	0	0	0	0	52	0	25	120010	10	120000	10	120000	50	4283622	0	1	30020	30039	30039	14989	3	15012	120010	20	120000	20	360000	30039	30039	1	1	120021	10	9	10	10	120000	10	0	0	0	0	7522	3	1	1	5	16	2	1	1	5	5	30036	0	20	9	120000	10	30040	30040	30040	31749	30040

SMLAL2 (by element, 2D)

Test 1: uops

1000 unrolls and 1 iteration

Test 2: Latency 1->1

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 3: Latency 1->2

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 4: Latency 1->3

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 5: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 6: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations