Firestorm: FCMLA (vector, 4H) measurements

Code:

  fcmla v0.4h, v1.4h, v2.4h, #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	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	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	18	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	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	61	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	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	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	0	18	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	0	0	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038

Test 2: Latency 1->1

Code:

  fcmla v0.4h, v1.4h, v2.4h, #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	l2 tlb miss data (0b)	18	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	ld unit uop (a6)	l1d cache writeback (a8)	a9	ac	branch cond mispred nonspec (c5)	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	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	0	0	0	0	710	0	3	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	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	0	0	0	0	710	0	2	16	2	2	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	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	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	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	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	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	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	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	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	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	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	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	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	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	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	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	0	0	0	0	710	1	2	16	2	3	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	mmu table walk data (08)	09	l2 tlb miss data (0b)	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 access (a0)	ld unit uop (a6)	l1d cache writeback (a8)	a9	ac	c2	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	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	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	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	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	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	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	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	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	0	0	0	82	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	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	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	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	0	40018	40037	40037	38130	3	38767	10158	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	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	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	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	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	10000	10	40085	40086	40038	40038	40038
10024	40037	300	0	0	1	159	0	61	39407	25	10016	10	10000	10	10000	61	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	0	0	0	640	3	16	2	2	39473	10000	10	40038	40038	40038	40038	40038

Test 3: Latency 1->2

Code:

  fcmla v0.4h, v0.4h, v1.4h, #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	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 mispred nonspec (cb)	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	61	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	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	61	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	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	2742	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	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	61	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	40037	40037	38108	3	38745	10264	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	61	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	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	61	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	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	61	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	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	61	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	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	61	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	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	61	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	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	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)	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	72	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	301	0	0	372	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	516	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	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	4	0	474	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	1	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	114	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	299	0	0	696	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	582	726	39407	25	10010	10	10000	10	10000	50	5708304	1	40018	40037	40037	38130	17	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	468	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	1	0	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	225	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	299	0	0	453	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

Test 4: Latency 1->3

Code:

  fcmla v0.4h, v1.4h, v0.4h, #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	1f	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	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
10204	40037	300	0	0	69	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	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	600	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	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	528	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	1	0	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	24	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	141	0	710	1	16	1	1	39553	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	39	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	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	36	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	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	33	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	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	279	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	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	513	0	0	82	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	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	33	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	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	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	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 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	0	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	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	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	159	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	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	162	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	25	10010	10	10000	12	10148	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	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	10148	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	1	27	640	2	16	2	2	39473	10000	10	40038	40038	40133	40038	40038
10024	40037	300	9	0	61	39407	44	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	45	3	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	0	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	54	6	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	0	40037	40037	38130	3	38767	10010	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	0	50	3	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	61	39407	44	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	6	9	640	2	16	2	2	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	12	0	103	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	0	40037	40037	38130	3	38767	10010	20	10000	20	30501	40085	40037	1	1	10021	10	9	10	10	10000	10	2	3	3	640	2	16	2	2	39473	10000	10	40038	40038	40085	40038	40038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  fcmla v0.4h, v8.4h, v9.4h, #90
  movi v1.16b, 0
  fcmla v1.4h, v8.4h, v9.4h, #90
  movi v2.16b, 0
  fcmla v2.4h, v8.4h, v9.4h, #90
  movi v3.16b, 0
  fcmla v3.4h, v8.4h, v9.4h, #90
  movi v4.16b, 0
  fcmla v4.4h, v8.4h, v9.4h, #90
  movi v5.16b, 0
  fcmla v5.4h, v8.4h, v9.4h, #90
  movi v6.16b, 0
  fcmla v6.4h, v8.4h, v9.4h, #90
  movi v7.16b, 0
  fcmla v7.4h, v8.4h, v9.4h, #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	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	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	l1d tlb miss (a1)	l1d cache writeback (a8)	ac	branch mispred nonspec (cb)	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160204	20091	151	0	0	40	25	80100	100	80000	100	80000	500	640000	0	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	1	0	3	0	10111	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	0	40	25	80100	100	80000	100	80000	500	640000	0	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	0	10111	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	0	40	25	80100	100	80000	100	80000	500	640000	0	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	0	10111	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	155	0	0	40	25	80100	100	80000	100	80000	500	640000	0	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	1	93	0	10111	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	0	40	25	80100	100	80000	100	80000	500	640000	0	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	15	0	10111	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	0	40	25	80100	100	80000	100	80000	500	640000	0	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	30	0	10111	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	0	40	25	80100	100	80000	100	80000	500	640000	0	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	0	10111	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	158	0	0	40	25	80100	100	80000	100	80000	500	640000	0	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	3	0	10111	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	151	0	0	40	25	80100	100	80000	100	80000	500	640000	0	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	34	0	0	10111	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	151	0	0	40	25	80100	100	80000	100	80000	500	640000	0	20046	0	20065	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	3	0	10111	1	16	1	1	20062	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	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	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	branch mispred nonspec (cb)	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	20100	150	0	46	27	80012	12	80000	12	80000	62	640000	1	1	0	20033	20052	20052	0	3	23	80012	20	80000	20	240000	20052	20061	1	1	160021	10	9	10	10	160000	10	4	3	0	10035	8	1	1	19	25	2	1	1	11	5	20049	2	20	1	160000	10	20053	20241	20088	20053	20053
160024	20052	150	0	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	0	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	1	0	0	10039	8	4	1	16	25	2	1	1	5	11	20049	2	20	1	160000	10	20053	20217	20090	20053	20053
160024	20052	150	0	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	0	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	0	0	0	10034	8	4	1	10	25	2	1	1	6	16	20049	2	20	1	160000	10	20053	20242	20066	20053	20053
160024	20052	150	0	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	0	3	23	80012	20	80000	20	240000	20052	20052	1	1	160022	10	9	10	10	160000	10	2	3	0	10029	8	4	1	17	25	2	1	1	16	12	20049	2	20	1	160000	10	20053	20243	20088	20053	20053
160024	20052	150	0	74	29	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	0	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	1	0	0	10035	8	4	1	12	25	2	1	1	11	6	20049	2	20	1	160000	10	20053	20227	20088	20053	20053
160024	20052	151	0	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	0	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	6	0	0	10028	8	4	1	12	25	2	1	1	6	16	20049	2	20	1	160000	10	20053	20218	20066	20053	20053
160024	20052	150	0	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	0	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	0	0	0	10035	8	4	1	6	25	2	1	1	12	6	20049	2	20	1	160000	10	20053	20235	20066	20053	20053
160024	20052	150	0	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	0	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	0	0	0	10034	8	4	1	12	25	2	1	1	12	5	20049	2	20	1	160000	10	20053	20203	20088	20053	20053
160024	20057	150	0	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	0	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	0	0	0	10028	8	4	1	18	25	2	1	1	16	12	20049	2	20	1	160000	10	20053	20219	20066	20055	20053
160024	20052	150	0	46	27	80012	12	80000	12	80000	62	640000	1	1	5	20033	20052	20052	0	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	56	9	0	10038	8	4	1	6	25	2	1	1	4	10	20049	2	20	1	160000	10	20053	20213	20088	20053	20053

Test 6: throughput

Count: 16

Code:

  fcmla v0.4h, v16.4h, v17.4h, #90
  fcmla v1.4h, v16.4h, v17.4h, #90
  fcmla v2.4h, v16.4h, v17.4h, #90
  fcmla v3.4h, v16.4h, v17.4h, #90
  fcmla v4.4h, v16.4h, v17.4h, #90
  fcmla v5.4h, v16.4h, v17.4h, #90
  fcmla v6.4h, v16.4h, v17.4h, #90
  fcmla v7.4h, v16.4h, v17.4h, #90
  fcmla v8.4h, v16.4h, v17.4h, #90
  fcmla v9.4h, v16.4h, v17.4h, #90
  fcmla v10.4h, v16.4h, v17.4h, #90
  fcmla v11.4h, v16.4h, v17.4h, #90
  fcmla v12.4h, v16.4h, v17.4h, #90
  fcmla v13.4h, v16.4h, v17.4h, #90
  fcmla v14.4h, v16.4h, v17.4h, #90
  fcmla v15.4h, v16.4h, v17.4h, #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	l2 tlb miss data (0b)	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)	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)	l1d cache writeback (a8)	ac	cf	d0	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160204	42042	304	0	0	0	0	34	42	0	25	160118	100	160000	100	160000	500	5863038	0	40021	40040	40040	19973	0	3	19998	160100	200	160000	200	480000	40040	40040	1	1	160201	100	99	100	100	160000	100	0	1	0	10110	0	1	16	1	1	42315	160000	100	40041	40044	42282	42306	40041
160204	40040	299	0	0	0	0	53	42	0	25	160153	100	160017	100	160000	500	1280000	0	40021	40040	40042	19973	0	3	19998	160100	200	160210	200	480000	40040	42305	1	1	160201	100	99	100	100	160000	100	0	1	220	10110	0	3	16	1	1	40037	160000	100	42319	40041	40041	40041	42282
160204	42318	300	0	0	0	0	0	42	8970	25	160153	100	160053	100	160000	500	5868333	0	40021	40040	40040	19973	0	3	19998	160100	200	160000	200	480000	40040	42318	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	0	1	16	1	1	40037	160000	100	42319	40041	42319	40041	40041
160204	40040	300	0	0	0	0	22	42	0	25	160100	100	160000	100	160000	500	1280000	0	40021	40040	40040	21111	0	3	22276	160100	200	160000	200	480000	42318	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	0	1	16	1	1	40037	160000	100	40041	40041	42319	40041	42319
160204	41190	309	0	0	0	0	0	42	0	25	160153	100	160000	100	160000	500	1280000	0	42299	42318	40040	22222	0	3	19998	160100	200	160000	200	480000	42305	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	0	1	16	1	1	40037	160000	100	42319	40041	42319	40041	40041
160204	40040	300	0	0	0	0	1	42	0	25	160153	100	160000	100	160000	500	1280000	0	42299	42318	40040	19973	0	3	19998	160100	200	160000	200	480000	40040	42318	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	0	1	16	1	1	40037	160000	100	40041	40041	40041	42306	40041
160204	42309	317	0	0	0	0	0	61	0	25	160100	100	160000	100	160000	500	1280000	0	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	10110	0	1	16	1	1	42315	160000	100	42306	42315	40041	42319	40041
160204	40040	316	0	0	0	0	0	61	0	25	160151	100	160000	100	160000	500	1280000	0	42299	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	10110	0	1	16	1	1	40037	160000	100	40041	42310	40041	41191	40041
160204	40040	317	0	0	0	0	51	309	0	25	160100	100	160000	100	160000	500	1280000	0	40021	40040	40040	19973	0	3	19998	160100	200	160000	200	480000	42318	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	0	1	16	1	1	42315	160000	100	40041	40041	40041	40041	40041
160204	40040	299	0	0	0	0	0	61	0	25	160100	100	160000	100	160000	500	1280000	0	42299	40040	42318	22222	0	3	22276	160100	200	160000	200	480000	42318	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	10110	0	1	16	2	1	42315	160000	100	42319	40041	40042	40042	41220

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	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)	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	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	40040	306	0	0	53	0	67	0	25	160010	10	160053	10	160000	50	1280000	1	1	0	40021	40040	40040	19996	3	21225	160010	20	160000	20	480000	40040	40040	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	0	10024	11	4	1	11	16	4	1	2	10	10	40037	40	16	160000	10	42286	40041	42286	40041	40041
160024	40090	317	0	39	0	0	53	8970	25	160010	10	160000	10	160000	50	1280000	0	1	0	40021	42318	40040	19996	3	20021	160010	20	160000	20	480000	40040	40040	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	0	10022	11	6	2	12	16	4	2	2	10	11	40037	40	16	160000	10	40041	40041	40041	40041	42295
160024	42318	300	0	0	0	0	53	0	25	160010	10	160000	10	160000	50	1280000	0	1	5	40021	40040	40040	19996	3	20020	160010	20	160000	20	480000	40040	42318	1	1	160021	10	9	10	10	160000	10	2	0	0	0	0	0	10049	11	6	2	10	16	4	2	2	7	10	40037	40	16	160000	10	40041	40041	40041	40041	40041
160024	40040	300	0	0	17	0	53	4695	25	160045	10	160035	10	160000	50	1280000	0	1	5	40021	40040	40040	19996	3	20020	160010	20	160000	20	480000	40040	42318	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	0	10024	11	6	2	10	16	4	2	2	10	11	41242	40	16	160000	10	40041	41246	40041	41246	40041
160024	40040	300	0	0	0	0	53	0	25	160063	10	160177	10	160000	50	1280000	0	1	5	40021	40040	40040	19996	3	20020	160010	20	160000	20	480000	40040	42318	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	0	10024	11	6	2	7	16	4	5	2	11	10	40037	40	16	160000	10	40041	42286	40041	42286	40041
160024	40040	316	0	0	0	0	73	8970	25	160063	10	160000	10	160000	50	1280000	0	1	0	40021	40043	40040	19996	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	11	2	1	8	16	4	1	2	12	7	40037	40	16	160000	10	42295	40041	42319	40041	40041
160024	40040	300	0	0	0	0	528	8863	25	160027	10	160017	10	160000	50	5867562	1	1	5	42262	40040	40040	19996	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	6	2	11	16	4	2	1	8	13	40037	20	16	160000	10	42319	40041	42319	40041	40041
160024	40040	317	0	0	0	0	53	0	25	160010	10	160000	10	160000	50	5721073	0	0	0	40021	40040	40040	19996	3	20020	160010	20	160000	20	480000	40040	42285	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	0	10024	11	6	2	10	16	4	2	2	10	7	40037	40	8	160000	10	40041	42319	40041	40041	42319
160024	42318	300	0	0	0	0	73	0	25	160046	10	160000	10	160000	50	1280000	0	1	5	40021	40040	41205	21173	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	11	2	2	10	16	2	1	1	8	10	40037	40	16	160000	10	40041	40041	40041	40041	40041
160024	40040	317	0	0	17	0	73	0	25	160027	10	160017	10	160000	50	1280000	1	1	5	42299	40040	40040	19996	3	20020	160010	20	160000	20	480000	42281	40040	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	0	10022	11	5	2	12	16	2	1	1	13	13	40037	40	8	160000	10	40041	40041	40041	41246	40041

FCMLA (vector, 4H)

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