Firestorm: FMLA (vector, 2S) measurements

Code:

  fmla v0.2s, v1.2s, v2.2s

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)	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	f5	f6	f7	f8	fd
1004	4037	30	0	61	3407	25	1000	1000	1000	531908	1	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	61	3407	25	1000	1000	1000	531908	1	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	73	1	16	1	1	3473	1000	4038	4038	4085	4111	4038
1004	4037	30	0	61	3407	25	1000	1000	1000	531908	1	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	61	3407	25	1000	1000	1000	531908	1	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	61	3407	25	1000	1000	1000	531908	1	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	9	61	3407	25	1000	1000	1000	531908	1	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	31	0	61	3407	25	1000	1000	1000	531908	1	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	61	3407	25	1000	1000	1000	531908	1	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	61	3407	25	1004	1000	1000	531908	1	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	3	61	3407	25	1000	1000	1000	531908	1	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038

Test 2: Latency 1->1

Code:

  fmla v0.2s, v1.2s, v2.2s

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): 4.0037

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

retire uop (01)	cycle (02)	03	mmu table walk data (08)	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)	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	40037	300	0	0	0	920	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	27	0	0	710	1	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	147	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	1	304	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10251	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	214	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	170	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	804	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	317	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	12	0	208	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	3	2	0	712	1	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40133
10204	40037	300	0	0	0	145	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	1	0	0	0	710	1	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	12	1	726	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	2	16	2	2	40034	0	10000	100	40038	40038	40038	40038	40038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

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

retire uop (01)	cycle (02)	03	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)	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 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	40037	299	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	3	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	0	40018	40084	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5708307	0	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	3	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	2	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40079
10024	40037	299	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	3	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	832	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	0	1	640	2	16	2	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	299	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	251	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	3	2	39473	10000	10	40038	40038	40038	40038	40038

Test 3: Latency 1->2

Code:

  fmla v0.2s, v0.2s, v1.2s

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0037

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)	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	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	40037	300	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	710	2	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40083
10204	40037	300	87	0	61	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	710	2	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	61	39407	25	10100	100	10000	125	10000	500	5706908	1	40018	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	3	710	2	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	61	39407	25	10100	125	10000	100	10000	500	5706908	1	40018	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	710	2	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	40037	40037	38108	0	3	38745	10251	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	710	2	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	61	39407	25	10100	100	10000	100	10000	626	5706908	1	40018	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	710	2	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	103	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	710	2	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	710	2	16	1	1	39479	25	10000	100	40038	40038	40038	40038	40038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

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 access (a0)	ac	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10024	40037	300	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	0	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40084	300	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	0	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	299	0	61	39407	25	10010	10	10000	10	10000	50	5706908	0	40018	0	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	61	39407	25	10010	10	10000	10	10000	50	5706908	0	40018	0	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	61	39407	25	10010	10	10000	10	10000	50	5706908	0	40018	0	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	82	39407	25	10010	10	10000	10	10000	50	5706908	0	40018	0	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	61	39407	25	10010	10	10000	10	10000	50	5706908	0	40018	0	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	299	15	726	39407	25	10010	10	10000	10	10000	50	5706908	0	40018	0	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	299	0	61	39407	25	10010	12	10000	10	10000	50	5706908	0	40018	0	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	61	39407	25	10010	10	10000	10	10000	50	5706908	0	40018	0	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	49	3	3	39473	10000	10	40038	40038	40038	40038	40038

Test 4: Latency 1->3

Code:

  fmla v0.2s, v1.2s, v0.2s

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0037

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

retire uop (01)	cycle (02)	03	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)	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	branch cond mispred nonspec (c5)	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	40037	300	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40083
10204	40037	300	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40084	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	1	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	346	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

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 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
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	103	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	82	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	661	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10022	10	9	10	10	10000	10	0	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	299	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10022	10	9	10	10	10000	10	0	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  fmla v0.2s, v8.2s, v9.2s
  movi v1.16b, 0
  fmla v1.2s, v8.2s, v9.2s
  movi v2.16b, 0
  fmla v2.2s, v8.2s, v9.2s
  movi v3.16b, 0
  fmla v3.2s, v8.2s, v9.2s
  movi v4.16b, 0
  fmla v4.2s, v8.2s, v9.2s
  movi v5.16b, 0
  fmla v5.2s, v8.2s, v9.2s
  movi v6.16b, 0
  fmla v6.2s, v8.2s, v9.2s
  movi v7.16b, 0
  fmla v7.2s, v8.2s, v9.2s

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	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	6b	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	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
160204	20065	150	0	40	25	80100	100	80000	100	80000	500	640000	1	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	10111	1	16	1	1	20062	0	160000	100	20066	20066	20066	20066	20066
160204	20065	150	348	40	25	80100	100	80000	100	80000	500	640000	1	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	10111	1	16	1	1	20062	0	160000	100	20066	20066	20066	20066	20066
160204	20065	150	336	40	25	80100	100	80000	100	80000	500	640000	1	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	10111	1	16	1	1	20062	0	160000	100	20066	20066	20066	20066	20066
160204	20065	151	0	40	25	80100	100	80000	100	80000	500	640000	1	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	10111	1	16	1	1	33072	0	160000	100	20066	20066	20066	20066	20066
160204	20065	151	0	40	25	80100	100	80000	100	80000	500	640000	1	20046	0	20065	20065	3	23	80100	200	80109	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	10111	1	16	1	1	20062	0	160000	100	20066	20066	20066	20066	20066
160204	20065	150	6	40	25	80100	100	80000	100	80000	500	640000	1	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	10111	1	16	1	1	20062	0	160000	100	20066	20066	20066	20066	20066
160204	20065	150	222	61	25	80100	100	80000	100	80000	500	640000	0	20046	0	20065	20065	3	23	80134	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	10111	1	16	1	1	20062	0	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	40	25	80100	100	80000	100	80000	500	640000	1	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	10111	1	16	1	1	20062	0	160000	100	20066	20066	20066	20066	20066
160204	20065	150	381	40	25	80100	100	80000	100	80000	500	640000	0	20046	3	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	9	10111	1	16	1	1	20062	9	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	40	25	80100	100	80000	100	80000	500	640000	1	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	10111	1	16	1	1	20062	0	160000	100	20066	20066	20066	20066	20066

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)	l2 tlb miss instruction (0a)	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	61	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)	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	ec	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160024	20086	150	0	0	0	236	27	80012	12	80000	12	80000	62	640000	0	1	5	20033	20052	20052	3	23	80012	20	80000	20	240000	20061	20061	1	1	160021	10	9	10	10	160000	10	0	10031	8	2	2	25	25	2	1	1	14	9	20049	2	20	1	160000	10	20053	20053	20053	20053	20053
160024	20052	150	0	0	372	58	27	80012	12	80000	12	80000	62	640000	0	1	5	20033	20052	20052	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	0	10029	8	4	1	8	25	2	1	1	14	9	20049	2	20	1	160000	10	20053	20053	20053	20053	20053
160024	20052	150	0	0	0	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	0	10028	8	4	1	12	34	2	1	1	7	14	20058	2	20	1	160000	10	20053	20053	20053	20062	20062
160024	20052	150	0	0	0	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20042	20052	20052	3	23	80012	20	80000	20	240000	20061	20052	1	1	160021	10	9	10	10	160000	10	0	10028	8	4	1	14	25	2	1	1	7	6	20049	2	20	1	160000	10	20053	20053	20053	20053	20053
160024	20052	150	0	0	351	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	0	10114	11	4	2	6	25	2	2	1	12	6	20049	2	20	2	160000	10	20053	20053	20053	20053	20053
160024	20052	150	0	0	0	46	27	80012	12	80000	12	80000	68	640000	1	1	5	20033	20052	20052	3	23	80012	20	80000	20	240000	20061	20052	1	1	160021	10	9	10	10	160000	10	0	10035	8	4	1	6	25	2	1	1	5	6	20049	2	20	1	160000	10	20053	20053	20053	20053	20053
160024	20052	150	0	0	0	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	0	10033	8	4	1	7	25	4	1	1	7	6	20049	2	20	2	160000	10	20053	20053	20062	20053	20053
160024	20052	150	0	0	372	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	0	10029	8	4	1	13	25	2	1	1	7	13	20049	2	20	1	160000	10	20053	20053	20053	20053	20053
160024	20052	150	0	0	15	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20042	20061	20052	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	0	10029	8	4	1	4	25	2	1	1	6	6	20049	2	20	1	160000	10	20053	20053	20053	20053	20053
160024	20052	150	0	0	0	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	3	51	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	0	10032	8	4	2	6	25	2	1	1	6	6	20049	2	40	2	160000	10	20053	20053	20053	20053	20053

Test 6: throughput

Count: 16

Code:

  fmla v0.2s, v16.2s, v17.2s
  fmla v1.2s, v16.2s, v17.2s
  fmla v2.2s, v16.2s, v17.2s
  fmla v3.2s, v16.2s, v17.2s
  fmla v4.2s, v16.2s, v17.2s
  fmla v5.2s, v16.2s, v17.2s
  fmla v6.2s, v16.2s, v17.2s
  fmla v7.2s, v16.2s, v17.2s
  fmla v8.2s, v16.2s, v17.2s
  fmla v9.2s, v16.2s, v17.2s
  fmla v10.2s, v16.2s, v17.2s
  fmla v11.2s, v16.2s, v17.2s
  fmla v12.2s, v16.2s, v17.2s
  fmla v13.2s, v16.2s, v17.2s
  fmla v14.2s, v16.2s, v17.2s
  fmla v15.2s, v16.2s, v17.2s

Initialization

  movi v16.16b, 17
  movi v17.16b, 18

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.2502

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

retire uop (01)	cycle (02)	03	18	19	1e	1f	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)	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	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160204	40130	300	0	0	0	0	0	42	0	25	160117	100	160001	100	160000	500	1280000	1	40021	41190	41205	19973	3	21148	160100	200	160000	200	480000	40040	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	1	16	1	1	41187	160000	100	41191	41206	40041	40041	40041
160204	40040	300	0	0	0	0	22	45	0	25	160100	100	160000	100	160000	500	1280000	1	40021	40040	40040	19973	3	19998	160100	200	160000	200	480000	40040	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	1	16	0	2	41202	160000	100	40041	40041	40041	40041	40041
160204	41190	308	0	0	0	0	0	42	0	25	160100	100	160000	100	160000	500	1280000	0	40021	40041	40040	19973	3	19998	160100	200	160000	200	480000	40040	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	1	16	1	1	40037	160000	100	41191	40044	40041	40041	40041
160204	40040	300	0	0	0	0	0	42	4485	25	160100	100	160000	100	160000	500	1280000	1	40021	42280	40040	19973	3	21148	160100	200	160000	200	480000	40040	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	1	16	1	1	41187	160000	100	40041	41191	42286	40041	42286
160204	40040	300	0	0	0	0	0	42	0	25	160100	100	160000	100	160000	500	1280000	1	40021	40040	40040	19973	3	19998	160100	200	160000	200	480000	40040	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	1	16	1	1	40037	160000	100	40041	40041	40041	40041	40041
160204	40040	300	0	0	0	0	0	61	0	25	160117	100	160000	100	160000	500	1280000	0	40021	40040	40040	19973	3	21148	160100	200	160000	200	480000	40040	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	1	16	1	1	40037	160000	100	41191	41206	40041	40041	40041
160204	40040	300	0	0	0	0	0	42	0	25	160100	100	160000	100	160000	500	1280000	0	40021	40040	41190	19973	3	21148	160100	200	160000	200	480000	41291	41205	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	1	16	1	1	40037	160000	100	40041	40041	40041	40041	40041
160204	40040	299	0	0	0	0	22	42	4485	25	160122	100	160022	100	160000	500	1280000	0	42266	40040	41190	21095	3	19998	160100	200	160000	200	480000	41190	41205	1	1	160201	100	99	100	100	160000	100	0	0	165	10110	1	16	1	1	41187	160000	100	40041	40041	40041	40041	40041
160204	40040	299	0	0	0	0	22	707	0	25	160100	100	160000	100	160000	500	1280000	1	41171	40040	40040	19973	3	19998	160100	200	160000	200	480000	40040	40043	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	1	16	1	1	41202	160000	100	40041	40041	40041	40041	41191
160204	40040	300	0	0	0	0	43	61	4485	25	160122	100	160000	100	160000	500	1280000	1	40021	40040	40040	21095	3	19998	160100	200	160000	200	480000	40043	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	1	16	1	1	40037	160000	100	40044	41191	41206	40041	40041

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2502

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

retire uop (01)	cycle (02)	03	mmu table walk data (08)	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	61	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 tlb access (a0)	ld unit uop (a6)	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	ea	eb	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160024	40049	306	0	0	0	53	0	25	160010	10	160000	10	160000	50	1280000	1	1	0	40021	40040	40040	21173	0	3	20020	160010	20	160000	20	480000	40040	41219	1	1	160021	10	9	10	10	160000	10	0	0	1	0	0	0	10022	3	1	1	3	16	2	1	1	4	4	40037	15	6	160000	10	40041	40041	40140	40041	42319
160024	40040	300	0	0	0	47	0	25	160011	10	160000	10	160000	50	1320000	1	1	0	40021	41245	40040	19996	0	3	20021	160010	20	160000	20	480000	42327	40040	1	1	160021	10	9	10	10	160000	10	0	0	3	0	78	0	10022	8	2	1	3	16	2	1	1	3	2	40037	15	6	160000	10	40991	42377	41743	40041	40042
160024	40040	300	0	0	0	47	8863	25	160010	10	160000	10	160000	50	1280000	0	1	5	40021	40040	40040	21173	0	3	20020	160010	20	160000	20	480000	42285	40040	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	0	10024	11	3	2	3	16	4	2	2	3	4	40037	30	10	160000	10	40041	40041	40062	40041	40041
160024	40040	300	0	0	0	48	8863	25	160045	10	160000	10	160000	50	1280000	1	1	5	40021	40040	40040	19996	0	3	20020	160010	20	160000	20	480000	42318	40040	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	0	10022	8	2	1	5	16	2	1	1	4	3	40037	15	6	160000	10	40041	40041	40066	40041	40041
160024	40040	317	0	0	0	237	0	25	160010	10	160053	10	160000	50	5868333	0	1	5	40021	40040	40040	19996	0	3	20020	160010	20	160000	20	480000	40040	40040	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	0	10024	8	2	2	3	17	4	2	2	2	4	40037	15	6	160000	10	40041	40041	40134	42297	40041
160024	41229	300	0	0	35	47	8913	25	160010	10	160000	10	160000	50	5721073	1	1	5	42266	40040	40040	19996	0	3	22274	160010	20	160000	20	480000	40040	40040	1	1	160021	10	9	10	10	160000	10	0	0	4	0	0	0	10022	8	2	1	3	16	4	1	1	3	2	40037	15	5	160000	10	42319	40041	40180	40042	40041
160024	40040	300	0	0	0	47	0	25	160010	10	160000	10	160000	50	1280000	1	1	5	40021	40040	40041	19996	0	3	20020	160010	20	160000	20	480000	40040	41245	1	1	160021	10	9	10	10	160000	10	0	3	0	0	6	0	10022	8	3	1	3	16	2	1	1	4	5	40037	15	5	160000	10	40106	40041	42111	42288	40041
160024	40040	300	0	0	0	47	0	25	160010	10	160000	10	160000	50	1280000	0	1	5	40021	40040	40040	19996	0	3	20020	160010	20	160000	20	480000	40040	40040	1	1	160021	10	9	10	10	160000	10	0	0	2	0	0	0	10022	8	3	2	2	16	2	1	2	3	3	40037	15	5	160000	10	40041	42286	40066	40041	40041
160024	40040	300	0	0	0	47	0	25	160010	10	160000	10	160000	50	1280000	1	1	5	40021	40040	40040	19996	0	3	20020	160010	20	160000	20	480000	40040	40040	1	1	160021	10	9	10	10	160000	10	0	0	1	0	0	0	10024	8	2	1	4	16	2	1	1	3	3	40037	15	5	160000	10	42296	42286	40148	40041	40042
160024	40040	300	0	0	0	67	8970	25	160010	10	160000	10	160000	50	1280000	1	1	5	40021	40040	40040	19996	0	3	20020	160010	20	160000	20	480000	40040	40040	1	1	160021	10	9	10	10	160000	10	0	0	6	0	6	0	10024	8	2	1	3	16	2	1	1	4	5	40038	15	5	160000	10	40041	40041	40240	40041	40041

FMLA (vector, 2S)

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