Firestorm: FCMLA (vector, 2D) measurements

Code:

  fcmla v0.2d, v1.2d, v2.2d, #90

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 cache writeback (a8)	ac	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	0	0	73	2	16	1	2	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	12	61	3407	25	1000	1000	1000	531908	1	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	0	0	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	0	0	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	84	3407	25	1000	1000	1000	531908	0	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	0	0	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	0	0	73	2	16	1	2	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	84	3407	25	1000	1000	1000	531908	1	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	0	0	73	1	16	1	1	3473	1000	4038	4038	4038	4085	4038
1004	4037	31	0	61	3398	25	1000	1000	1000	531908	1	4018	4037	4037	3258	3	3895	1000	1000	3000	4037	4037	1	1	1001	1000	0	0	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	0	0	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	0	0	73	1	16	1	1	3549	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	0	9	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038

Test 2: Latency 1->1

Code:

  fcmla v0.2d, v1.2d, v2.2d, #90

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)	09	l2 tlb miss data (0b)	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	6b	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 miss (a1)	ld unit uop (a6)	st unit uop (a7)	l1d cache writeback (a8)	a9	ac	cf	d0	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	0	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	9	0	6	710	1	3	16	2	2	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	7	0	9	710	1	2	16	2	2	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	7	0	9	710	1	2	16	2	2	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	0	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	1	0	9	710	1	2	16	2	2	39479	10000	100	40038	40038	40038	40038	40038
10204	40086	300	0	0	0	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	8	0	12	710	1	2	16	2	2	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	1	0	90	710	1	2	16	2	2	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	0	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	12	0	9	710	1	2	16	2	2	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	30	0	76	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	14	0	6	710	1	2	16	2	2	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	0	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	9	0	3	710	1	2	16	3	2	39479	10000	100	40038	40038	40038	40085	40038
10204	40037	300	0	0	0	0	0	372	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	10	0	9	712	1	2	16	2	2	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	09	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 tlb access (a0)	l1d tlb miss (a1)	st unit uop (a7)	l1d cache writeback (a8)	ac	cd	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	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	2	0	0	640	2	16	2	2	39473	0	10000	10	40085	40086	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	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	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	0	0	640	3	16	2	2	39473	0	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	3	126	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	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	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	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	1	27	0	0	640	2	16	2	2	39473	0	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	22	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40054	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	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	1156	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	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10006	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	17	9	0	640	3	16	2	2	39474	0	10000	10	40038	40038	40038	40038	40038

Test 3: Latency 1->2

Code:

  fcmla v0.2d, v0.2d, v1.2d, #90

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	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)	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	branch mispred nonspec (cb)	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	299	0	61	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40229	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	76	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	39526	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	39479	2	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40229	40081	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	0	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	39479	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	09	l2 tlb miss data (0b)	18	19	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 tlb miss (a1)	ld unit uop (a6)	l1d cache writeback (a8)	a9	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
10024	40037	300	0	0	0	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	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	0	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	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	0	0	0	61	39407	25	10010	10	10000	10	10153	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	0	640	2	16	2	4	39512	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	0	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	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	0	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	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	299	0	0	0	0	0	726	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	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	0	0	0	359	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	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	0	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	5	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	299	0	0	0	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	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	0	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	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038

Test 4: Latency 1->3

Code:

  fcmla v0.2d, v1.2d, v0.2d, #90

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	18	19	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)	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	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	eb	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	40037	300	3	1	1104	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	1	0	710	3	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	12	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	0	710	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	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	0	710	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	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	3	710	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	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	0	710	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	552	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	0	710	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	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	0	710	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	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	0	710	2	16	2	2	39538	0	10000	100	40038	40038	40038	40038	40038
10204	40084	300	0	0	654	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	0	710	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	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	0	710	2	16	2	2	39479	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	09	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	branch cond mispred nonspec (c5)	branch mispred nonspec (cb)	cd	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	40037	300	0	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	3	38767	10159	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	299	0	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	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	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	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	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10025	40037	300	0	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	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	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	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	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	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	0	726	39407	25	10010	10	10000	10	10000	50	5706908	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	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	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	3	0	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	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	0	0	640	2	16	2	2	39549	0	10000	10	40038	40038	40038	40038	40038
10024	40037	299	0	0	0	66	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	0	0	640	2	16	2	2	39473	0	10000	10	40038	40038	40038	40038	40038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  fcmla v0.2d, v8.2d, v9.2d, #90
  movi v1.16b, 0
  fcmla v1.2d, v8.2d, v9.2d, #90
  movi v2.16b, 0
  fcmla v2.2d, v8.2d, v9.2d, #90
  movi v3.16b, 0
  fcmla v3.2d, v8.2d, v9.2d, #90
  movi v4.16b, 0
  fcmla v4.2d, v8.2d, v9.2d, #90
  movi v5.16b, 0
  fcmla v5.2d, v8.2d, v9.2d, #90
  movi v6.16b, 0
  fcmla v6.2d, v8.2d, v9.2d, #90
  movi v7.16b, 0
  fcmla v7.2d, v8.2d, v9.2d, #90

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)	18	1e	1f	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	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)	branch mispred nonspec (cb)	cd	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160204	20089	150	0	0	24	0	29	25	80116	100	80016	100	80028	500	640196	0	20046	20065	20065	6	12	80128	200	80028	200	240084	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10119	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	0	15	0	29	25	80116	100	80016	100	80028	500	640196	1	20046	20065	20065	6	12	80128	200	80028	200	240084	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10119	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	1	0	0	0	29	25	80116	100	80016	100	80028	500	640196	1	20046	20065	20065	6	12	80128	200	80028	200	240084	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10119	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	0	3	0	29	25	80116	100	80016	100	80028	500	640196	1	20046	20065	20065	6	12	80128	200	80028	200	240084	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10119	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	151	0	0	0	0	40	25	80100	100	80000	100	80000	500	640000	1	20046	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10112	2	16	2	2	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	0	0	0	40	25	80100	100	80000	100	80000	500	640000	1	20046	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10112	2	16	2	2	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	0	0	0	40	25	80100	100	80000	100	80000	500	640000	1	20046	20065	20065	3	23	80100	200	80000	202	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10112	2	16	2	2	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	0	0	0	40	25	80100	100	80000	100	80000	500	640000	1	20046	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10112	2	16	2	2	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	0	0	0	40	25	80100	100	80000	100	80000	500	640000	1	20046	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10112	2	16	2	2	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	0	0	0	40	25	80100	100	80000	100	80000	500	640000	1	20046	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10112	2	16	2	2	20062	160000	100	20066	20066	20066	20066	20066

1000 unrolls and 10 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	l1i tlb fill (04)	mmu table walk instruction (07)	mmu table walk data (08)	09	l2 tlb miss instruction (0a)	18	1e	3a	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)	branch mispred nonspec (cb)	cf	d0	d2	l1i cache miss demand (d3)	l1i tlb miss demand (d4)	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	20113	150	0	2	0	1	1	0	0	0	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20044	20063	20063	3	112	80012	20	80000	20	240000	20063	20063	1	1	160021	10	9	10	10	160000	10	7	0	10045	16	6	2	1	33	36	3	2	2	28	28	20060	2	41	2	160000	10	20064	20064	20064	20064	20064
160024	20063	150	1	2	0	0	2	0	0	1	76	29	80012	12	80000	12	80000	62	640000	0	1	10	20044	20063	20063	3	23	80012	20	80000	20	240000	20063	20063	1	1	160021	10	9	10	10	160000	10	0	0	10045	16	7	2	1	18	36	3	2	2	27	23	20060	2	41	2	160000	10	20064	20064	20064	20064	20064
160024	20063	150	1	3	0	0	2	0	0	1	58	29	80012	12	80000	12	80000	62	640000	0	1	10	20044	20063	20063	3	23	80012	20	80000	20	240000	20063	20063	1	1	160021	10	9	10	10	160000	10	0	0	10057	16	7	2	1	30	36	3	2	2	19	31	20060	2	41	2	160000	10	20064	20064	20064	20064	20064
160024	20063	150	1	2	0	0	2	0	0	0	729	29	80012	12	80000	12	80000	62	640000	0	1	10	20110	20063	20063	3	23	80012	20	80000	20	240000	20063	20063	1	1	160021	10	9	10	10	160000	10	0	0	10067	16	6	2	1	28	36	3	2	2	29	30	20060	2	41	2	160000	10	20064	20064	20064	20064	20064
160024	20063	150	1	3	0	0	2	0	0	1	64	29	80012	12	80000	12	80000	62	640000	0	1	10	20044	20063	20063	3	23	80012	20	80000	20	240000	20063	20063	1	1	160021	10	9	10	10	160000	10	0	0	10050	16	7	2	1	22	36	3	2	2	30	18	20060	2	41	2	160000	10	20064	20064	20064	20064	20064
160024	20063	150	1	3	0	0	3	0	0	0	70	29	80012	12	80000	12	80000	62	640000	0	1	10	20044	20063	20063	3	23	80012	20	80000	20	240000	20063	20063	1	1	160021	10	9	10	10	160000	10	0	0	10057	16	7	2	1	29	36	3	2	2	30	30	20060	2	41	2	160000	10	20064	20064	20064	20064	20064
160024	20063	150	1	2	0	0	3	0	0	1	64	29	80012	12	80000	12	80000	62	640000	0	1	10	20044	20063	20063	3	23	80012	20	80000	20	240000	20063	20063	1	1	160021	10	9	10	10	160000	10	0	0	10056	16	6	2	1	27	36	3	4	2	28	28	20060	2	41	2	160000	10	20064	20064	20064	20064	20064
160024	20063	150	1	2	0	0	2	0	0	1	76	29	80012	12	80000	12	80000	62	640000	0	1	10	20044	20063	20063	3	23	80012	20	80000	20	240000	20063	20063	1	1	160021	10	9	10	10	160000	10	0	0	10055	16	7	2	1	27	36	3	2	2	16	29	20060	2	41	2	160000	10	20064	20064	20064	20064	20064
160024	20063	150	1	3	0	0	3	0	21	1	741	29	80012	12	80000	12	80000	62	640000	0	1	10	20044	20063	20063	3	23	80012	20	80000	20	240000	20063	20063	1	1	160021	10	9	10	10	160000	10	0	0	10043	16	6	2	1	23	36	3	2	2	30	18	20060	2	41	2	160000	10	20064	20064	20064	20064	20064
160024	20063	150	1	2	0	0	2	1	18	0	76	29	80012	12	80000	12	80000	62	640000	0	1	10	20044	20063	20063	3	23	80012	20	80000	20	240000	20063	20063	1	1	160021	10	9	10	10	160000	10	0	0	10051	16	6	2	1	18	36	3	2	2	27	23	20060	2	41	2	160000	10	20064	20064	20064	20064	20064

Test 6: throughput

Count: 16

Code:

  fcmla v0.2d, v16.2d, v17.2d, #90
  fcmla v1.2d, v16.2d, v17.2d, #90
  fcmla v2.2d, v16.2d, v17.2d, #90
  fcmla v3.2d, v16.2d, v17.2d, #90
  fcmla v4.2d, v16.2d, v17.2d, #90
  fcmla v5.2d, v16.2d, v17.2d, #90
  fcmla v6.2d, v16.2d, v17.2d, #90
  fcmla v7.2d, v16.2d, v17.2d, #90
  fcmla v8.2d, v16.2d, v17.2d, #90
  fcmla v9.2d, v16.2d, v17.2d, #90
  fcmla v10.2d, v16.2d, v17.2d, #90
  fcmla v11.2d, v16.2d, v17.2d, #90
  fcmla v12.2d, v16.2d, v17.2d, #90
  fcmla v13.2d, v16.2d, v17.2d, #90
  fcmla v14.2d, v16.2d, v17.2d, #90
  fcmla v15.2d, v16.2d, v17.2d, #90

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	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)	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 miss (a1)	l1d cache writeback (a8)	ac	branch cond mispred nonspec (c5)	cd	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	40059	312	0	0	43	61	0	25	160100	100	160000	100	160000	500	1707603	40021	40040	40040	19973	0	3	19998	160100	200	160000	200	480000	40040	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10110	9	16	9	9	40037	0	160000	100	40041	40044	40041	40041	40041
160204	41190	317	0	0	0	42	4485	25	160100	100	160000	100	160000	500	1280000	40021	40040	40040	19973	0	3	19998	160100	200	160000	200	480000	41190	40043	1	1	160201	100	99	100	100	160000	100	0	0	0	1	0	10110	8	16	8	8	41202	0	160000	100	40041	40041	40041	41206	40041
160204	41190	309	0	0	43	726	0	25	160100	100	160000	100	160000	500	1280000	41186	40040	40040	19973	37	3	19998	160100	200	160000	200	480000	40040	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10110	9	16	9	9	40037	0	160000	100	40041	40041	40041	41191	40041
160204	40040	308	0	0	0	42	0	25	160100	100	160000	100	160000	500	1280000	41171	41190	41205	19973	0	3	19998	160100	200	160000	200	480000	40040	41190	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10110	11	16	9	4	40037	0	160000	100	40041	40041	40041	41191	41191
160204	40040	309	0	0	0	42	0	25	160100	100	160142	100	160000	500	1280000	42308	40040	40040	19973	0	3	19998	160100	200	160000	200	480000	40040	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10110	8	16	8	8	40037	0	160000	100	41206	40041	40041	40041	40041
160204	40040	309	0	0	0	42	0	25	160100	100	160000	100	160000	500	2646333	40022	40040	41190	19973	0	3	19998	160100	200	160000	200	480000	40040	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10110	9	16	8	9	41187	0	160000	100	40041	40041	40041	42328	41206
160204	40040	308	0	0	0	61	0	25	160100	100	160000	100	160000	500	5717939	40021	41190	41205	21111	0	3	19998	160100	200	160000	200	480000	40040	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10110	9	16	9	9	40037	0	160000	100	41191	41206	40041	40041	40041
160204	42281	300	0	0	0	42	0	25	160100	100	160000	100	160000	500	1280000	41171	40040	40040	21109	0	3	19998	160100	200	160000	200	480000	41190	41205	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10110	9	16	9	9	40040	0	160000	100	41191	41206	40041	40041	41191
160204	40040	299	0	0	0	42	0	25	160143	100	160000	100	160000	500	1280000	40021	41205	40040	19973	0	3	19998	160100	200	160000	200	480000	41204	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10110	8	16	9	9	41202	0	160000	100	40041	40041	42295	41205	40042
160204	40040	308	0	0	0	517	0	25	160100	100	160000	100	160000	500	1280000	41186	40040	40040	19973	0	3	21163	160100	200	160000	200	480000	40040	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10110	9	16	9	9	40037	0	160000	100	40041	40041	40041	40041	40044

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2576

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	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)	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 cache writeback (a8)	ac	branch cond mispred nonspec (c5)	cd	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	40050	300	1	0	0	0	1	53	0	35	160011	10	160000	10	160000	50	5710654	0	1	10	40021	42318	40040	19996	0	3	21206	160010	20	160000	20	480000	42318	40040	1	1	160021	10	9	10	10	160000	10	3	0	0	0	10023	16	4	1	23	16	3	2	2	29	16	40039	31	7	160000	10	40041	40041	42319	40041	40041
160024	41229	316	0	1	0	53	0	73	8970	25	160070	10	160053	10	160000	50	1280000	1	1	10	41207	41226	41226	21141	25	3	21206	160010	20	160000	20	480000	41226	41226	1	1	160021	10	9	10	10	160000	10	9	6	0	0	10025	16	6	2	29	16	3	1	2	17	28	40037	16	11	160000	10	40041	40041	42319	40041	42295
160024	42318	299	1	1	0	1	0	59	4576	25	160045	10	160102	10	160000	50	5728047	1	1	10	41210	41226	40040	21141	0	3	20020	160010	20	160000	20	480000	40040	40431	3	1	160021	10	9	10	10	160000	10	4	0	0	0	10024	16	6	2	25	16	3	2	2	21	25	42315	16	11	160000	10	41227	41227	41227	41227	40041
160024	42281	300	1	1	0	0	0	53	0	25	160070	10	160053	10	160000	50	1280000	0	1	10	40021	41226	41226	21141	0	3	21199	160010	20	160000	20	480000	41219	41226	1	1	160021	10	9	10	10	160000	10	8	9	0	0	10023	16	6	1	20	16	3	2	2	26	17	40039	31	11	160000	10	40041	40041	40041	40041	40041
160024	40040	317	1	1	0	1	0	53	0	25	160010	10	160000	10	160000	50	1280000	0	1	10	40021	40040	42318	22244	0	3	22298	160010	20	160000	20	480000	40040	41226	1	1	160021	10	9	10	10	160000	10	9	9	0	0	10025	16	7	2	26	16	1	2	2	15	26	41223	31	14	160000	10	40041	42319	40041	42319	40041
160024	40040	317	0	0	0	0	0	65	4576	25	160010	10	160000	10	160000	50	1280000	1	1	10	41207	41219	40040	21136	0	3	21183	160010	20	160000	20	480000	40040	40040	1	1	160021	10	9	10	10	160000	10	4	3	0	0	10023	16	7	1	27	16	1	2	2	16	26	42315	31	11	160000	10	40041	41227	41227	41227	40041
160024	40040	300	1	0	12	0	1	85	0	25	160028	10	160000	10	160000	50	5868333	0	1	10	40021	40040	40040	19996	0	3	20020	160010	20	160000	20	480000	40040	40040	1	1	160021	10	9	10	10	160000	10	20	12	0	0	10025	13	7	2	25	16	3	2	2	26	26	42315	31	8	160000	10	42319	40041	42319	40041	40041
160024	40040	308	1	1	0	53	1	67	4576	25	160070	10	160001	10	160000	50	1280000	0	1	10	40021	41226	40040	21141	25	3	20020	160010	20	160000	20	480000	41226	41226	1	1	160021	10	9	10	10	160000	10	8	0	0	0	10024	16	7	1	21	16	3	1	2	26	17	40037	16	12	160000	10	42319	40041	40041	40041	40041
160024	40040	317	1	1	0	0	1	65	8970	25	160053	10	160000	10	160000	50	5868333	1	1	10	42299	42318	40040	19996	0	3	22298	160010	20	160000	20	480000	42318	40040	1	1	160021	10	9	10	10	160000	10	8	0	1	0	10022	16	7	1	26	16	1	2	1	20	25	40037	31	8	160000	10	41227	41227	41227	41227	40041
160024	40040	309	0	1	0	0	1	65	4589	25	160011	10	160000	10	160000	50	1280000	0	1	10	40021	40040	40040	22244	0	3	22298	160010	20	160000	20	480000	42318	40040	1	1	160021	10	9	10	10	160000	10	6	0	0	0	10025	16	7	2	26	16	3	2	2	15	25	40037	31	11	160000	10	40041	42319	40041	42319	40041

FCMLA (vector, 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