Firestorm: FMLS (by element, 2D) measurements

Code:

  fmls v0.2d, v1.2d, v2.d[1]

Initialization

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

(no loop instructions)

1000 unrolls and 1 iteration

Retires: 1.000

Issues: 1.000

Integer unit issues: 0.000

Load/store unit issues: 0.000

SIMD/FP unit issues: 1.000

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

retire uop (01)	cycle (02)	03	mmu table walk data (08)	l2 tlb miss data (0b)	18	19	1e	1f	3f	4e	51	schedule uop (52)	schedule simd uop (54)	dispatch simd uop (57)	ldst uops in schedulers (5b)	60	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map simd uop (7e)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst simd alu (9a)	l1d tlb access (a0)	st unit uop (a7)	l1d cache writeback (a8)	a9	ac	c2	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	f5	f6	f7	f8	fd
1004	4037	30	0	0	0	0	0	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	0	0	0	0	73	2	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	0	0	0	0	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	0	0	0	0	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	0	0	0	0	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	0	0	69	0	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	0	0	0	0	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	0	0	12	0	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	0	0	0	0	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	0	0	0	0	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	0	0	0	0	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	0	0	0	0	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	0	0	0	0	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	0	0	0	0	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	0	0	0	0	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	0	0	0	0	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	31	0	0	0	0	0	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	0	0	0	0	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038
1004	4037	30	0	0	0	0	0	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	0	0	0	0	73	1	16	1	1	3473	1000	4038	4038	4038	4038	4038

Test 2: Latency 1->1

Code:

  fmls v0.2d, v1.2d, v2.d[1]

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0037

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

retire uop (01)	cycle (02)	03	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	6b	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb miss (a1)	l1d cache writeback (a8)	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	251	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	0	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	1	0	710	0	3	16	3	3	39479	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	0	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	1	0	710	1	3	16	3	4	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	0	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	0	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	3	0	710	1	3	16	3	3	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	89	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	0	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	1	0	710	1	3	16	4	3	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	0	61	39407	25	10136	100	10000	100	10000	500	5706908	1	40018	0	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	3	16	3	3	39479	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	0	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	0	0	712	1	3	16	3	4	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	0	103	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	0	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	3	16	3	3	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	9	1	61	39407	25	10100	100	10000	100	10000	500	5706908	0	40018	3	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	3	16	3	3	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	1	61	39407	25	10100	100	10000	100	10000	500	5706908	1	40018	0	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	0	9	0	710	1	3	16	3	3	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	726	39407	87	10100	100	10000	100	10000	500	5706908	1	40018	0	40037	40037	38108	3	38745	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	2	0	0	710	1	4	16	3	3	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	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	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
10024	40037	300	61	39407	25	10010	11	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	2	2	2	0	0	640	2	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	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	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	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	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	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	3	3	39545	10000	10	40038	40038	40038	40038	40038
10024	40037	300	89	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	4	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	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	4	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	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	3	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	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	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	299	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	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	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	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038

Test 3: Latency 1->2

Code:

  fmls v0.2d, v0.2d, v1.d[1]

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 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)	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)	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	c2	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	40037	300	0	1	0	0	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	1	0	0	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	0	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	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	0	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	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	0	0	0	0	0	0	162	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	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	0	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	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	0	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	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	0	0	6	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	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	0	0	0	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	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	0	0	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	0	710	1	16	1	1	39479	0	10000	100	40038	40038	40038	40038	40038
10204	40037	299	0	0	0	0	0	0	0	0	943	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	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	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 access (a0)	l1d tlb miss (a1)	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	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	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	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	9	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	2	0	0	0	0	640	2	16	2	2	39473	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	10000	10	40038	40038	40038	40038	40038
10024	40037	300	0	0	0	536	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	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	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	10000	10	40038	40038	40038	40038	40038
10024	40037	299	0	0	0	61	39407	25	10010	10	10000	10	10000	50	5706908	1	40018	40037	40037	38130	6	38786	10308	20	10163	20	30486	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	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	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	10000	10	40038	40038	40038	40038	40038

Test 4: Latency 1->3

Code:

  fmls v0.2d, v1.2d, v0.d[1]

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 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	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	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	82	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	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	3	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	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	9	61	39407	44	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	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	82	39389	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	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	38764	10100	200	10000	200	30000	40037	40037	1	1	10201	100	99	100	100	10000	100	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	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40038
10204	40037	300	0	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	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	710	1	16	1	1	39479	10000	100	40038	40038	40038	40038	40084

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	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	6b	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb access (a0)	l1d tlb miss (a1)	ld unit uop (a6)	l1d cache writeback (a8)	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	311	0	0	15	0	0	1444	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	0	0	0	640	4	16	3	3	39473	10000	10	40038	40085	40038	40224	40038
10024	40037	321	0	1	162	132	1	156	39389	25	10010	10	10000	10	10000	50	5706908	0	40018	0	40037	40037	38130	11	38805	10456	20	10000	20	30000	40037	40037	1	1	10021	10	9	10	10	10000	10	4	2	0	1	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	311	0	0	0	0	0	1151	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	0	0	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	306	0	0	0	0	0	908	39407	25	10018	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	0	0	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	306	0	0	0	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	0	0	0	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	305	0	0	0	0	0	103	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	0	0	0	640	3	16	3	3	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	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	0	0	640	3	16	3	5	39473	10000	10	40038	40038	40038	40038	40038
10024	40037	300	2	0	0	0	0	1603	39407	25	10010	10	10006	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	0	3	640	3	16	3	3	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	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	0	0	640	3	16	3	3	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	0	40037	40037	38130	3	38767	10010	20	10486	20	30000	40085	40085	1	1	10021	10	9	10	10	10000	10	0	0	2	3	9	640	3	16	3	3	39473	10000	10	40038	40038	40038	40038	40038

Test 5: throughput

Count: 8

Code:

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

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

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

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

retire uop (01)	cycle (02)	03	1e	3f	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6d	6e	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d cache writeback (a8)	ac	branch cond mispred nonspec (c5)	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160204	20077	150	0	1274	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	20	3	0	10111	1	16	2	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	705	25	80100	100	80000	117	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	10112	2	16	2	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	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	10111	2	16	2	2	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	1034	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	20	0	0	10112	2	16	2	2	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	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	10111	1	16	1	2	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	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	10112	1	16	2	2	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	103	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	10112	2	16	2	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	151	0	473	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	10111	1	16	1	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	40	25	80100	100	80000	100	80000	500	640000	1	20046	20135	20065	3	23	80100	200	80000	200	240000	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	10112	1	16	2	1	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	151	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	10112	2	16	2	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	mmu table walk instruction (07)	l2 tlb miss instruction (0a)	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)	ac	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	20097	150	0	0	2	88	27	80012	12	80000	12	80000	62	640000	1	1	0	20033	20052	20052	3	23	80012	20	80000	20	240000	20052	20052	1	1	160021	10	9	10	10	160000	10	2	3	10042	8	1	1	20	25	2	1	1	19	18	20049	2	40	2	160000	10	20062	20062	20062	20062	20062
160024	20061	150	2	1	1	85	29	80012	12	80000	12	80000	62	640000	0	1	5	20042	20061	20061	3	23	80012	20	80000	20	240000	20061	20061	1	1	160021	10	9	10	10	160000	10	0	0	10038	11	5	2	19	34	4	2	2	20	20	20058	2	40	2	160000	10	20062	20062	20062	20062	20062
160024	20061	150	4	2	1	64	29	80012	12	80109	12	80000	62	640000	0	1	5	20042	20061	20061	3	23	80012	20	80000	20	240000	20061	20061	1	1	160021	10	9	10	10	160000	10	0	0	10045	11	5	2	19	34	4	2	2	19	18	20058	2	40	2	160000	10	20062	20062	20062	20062	20062
160024	20061	151	1	1	2	117	29	80012	12	80000	12	80000	62	640000	0	1	5	20042	20061	20061	3	23	80012	20	80000	20	240000	20061	20130	1	1	160021	10	9	10	10	160000	10	0	0	10043	11	5	2	13	34	4	2	2	19	17	20058	2	20	1	160000	10	20053	20053	20053	20053	20053
160024	20052	150	1	0	2	58	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	0	10041	8	4	1	19	25	2	1	1	19	18	20049	2	20	1	160000	10	20053	20053	20053	20053	20053
160024	20052	150	1	1	1	123	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	0	10043	8	4	1	14	25	2	1	1	20	20	20049	2	20	1	160000	10	20053	20053	20053	20053	20053
160024	20052	150	1	1	1	182	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	3	10042	8	5	1	17	25	2	1	1	18	19	20049	2	20	1	160000	10	20053	20053	20053	20053	20053
160024	20052	150	1	1	1	886	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	0	10042	8	5	1	18	25	2	1	1	20	20	20049	2	20	1	160000	10	20053	20053	20053	20053	20053
160024	20052	150	0	0	2	953	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	0	10040	8	5	1	17	25	2	1	1	19	17	20049	2	20	1	160000	10	20053	20053	20053	20053	20053
160024	20052	150	2	1	2	138	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	0	10043	8	5	1	20	25	2	1	1	20	21	20049	2	20	1	160000	10	20053	20053	20053	20053	20053

Test 6: throughput

Count: 12

Code:

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

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.3337

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

retire uop (01)	cycle (02)	03	l2 tlb miss data (0b)	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)	60	69	6b	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb miss (a1)	l1d cache writeback (a8)	a9	ac	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
120204	40039	300	0	0	0	0	105	37966	25	120101	100	120001	100	120000	500	5630640	0	40020	0	40039	41691	26577	3	24997	120100	200	120000	200	360000	40039	41686	1	1	120201	100	99	100	100	120000	100	0	0	0	0	7610	3	16	1	1	40030	120000	100	40040	41692	40040	40040	40040
120204	40039	312	0	0	0	0	339	9961	25	120100	100	120003	100	120000	500	5630640	0	41667	0	41686	40039	24932	3	26649	120100	200	120000	200	360000	41686	40039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	7610	1	16	1	1	40030	120000	100	41687	40040	41692	40040	41692
120204	41688	312	0	0	0	0	898	9961	25	120100	100	120000	100	120000	500	5851993	0	41672	0	41686	40039	24932	3	24997	120100	200	120000	200	360000	41691	40039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	7610	1	16	1	1	41677	120000	100	41687	40040	41687	40040	41687
120204	41686	299	0	0	0	0	145	9961	25	120100	100	120000	100	120000	500	5630640	0	40020	0	41686	40039	24932	3	24997	120100	200	120000	200	360000	41686	40039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	7610	1	16	1	1	41677	120000	100	40040	41687	40040	41687	40040
120204	40039	312	0	0	3	0	61	9961	25	120100	100	120000	100	120000	500	5851869	0	41667	0	41691	40039	24932	3	24997	120100	200	120000	200	360000	40039	41688	1	1	120201	100	99	100	100	120000	100	0	0	0	3	7610	1	16	1	1	40030	120000	100	41687	40040	41687	40040	41692
120204	41691	300	0	0	0	0	61	9961	25	120100	100	120003	100	120000	500	5630640	1	40020	0	40039	41691	24932	3	24997	120100	200	120000	200	360000	40039	41686	1	1	120201	100	99	100	100	120000	100	0	0	0	0	7610	1	16	1	1	40030	120000	100	41687	40040	41692	40040	41687
120204	41691	299	0	0	0	0	61	9961	25	120103	100	120000	100	120000	500	5630640	0	40020	0	40039	41691	26577	3	26644	120100	200	120000	200	360000	41686	40039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	7610	1	16	1	1	41677	120000	100	40040	41687	40040	41692	40040
120204	40039	312	0	0	3	0	356	9961	25	120100	100	120000	100	120000	500	5851993	0	41667	0	41686	40039	24932	3	26644	120100	200	120000	200	360000	40039	40039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	7610	1	16	1	1	40030	120000	100	40040	41687	40040	41687	40040
120204	40039	312	0	0	0	0	489	35689	25	120103	100	120001	100	120000	500	5630640	0	40020	0	40039	41686	26582	3	26644	120100	200	120000	200	360000	41686	40039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	7610	1	16	1	1	40030	120000	100	41687	40040	41687	40040	41687
120204	41686	299	0	0	3	0	124	9961	25	120100	100	120000	100	120000	500	5851993	1	41672	0	41686	40039	24932	3	24997	120100	200	120000	200	360000	41691	40039	1	1	120201	100	99	100	100	120000	100	0	0	0	0	7610	1	16	1	1	40030	120000	100	40040	41692	40040	41689	40040

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.3337

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

retire uop (01)	cycle (02)	03	mmu table walk data (08)	09	18	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)	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)	a9	ac	cd	cf	d0	d5	map dispatch bubble (d6)	db	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
120024	40039	300	0	0	0	0	0	0	61	9961	25	120010	10	120000	10	120000	50	5630640	0	40020	40039	40039	24955	0	27	25019	120010	20	120000	20	360000	40039	40039	1	1	120021	10	9	10	10	120000	10	0	0	0	0	7520	69	25	16	0	10	13	40030	120000	10	40040	40040	40040	40040	40040
120024	40039	300	0	0	0	0	0	0	943	9961	25	120010	10	120000	10	120000	50	5630640	0	40020	40039	40039	24955	0	3	25019	120010	20	120000	20	360000	40039	40039	1	1	120021	10	9	10	10	120000	10	0	0	0	1	7520	69	11	16	0	13	11	40030	120000	10	40040	40040	40040	40040	40040
120024	40039	300	0	0	0	0	0	0	61	9961	25	120010	10	120000	10	120000	50	5630640	0	40020	40039	40039	24955	7	3	25019	120010	20	120000	20	360000	40039	40039	1	1	120021	10	9	10	10	120000	10	0	0	0	0	7520	66	11	16	0	12	12	40030	120000	10	40040	40040	40040	40040	40040
120024	40039	300	0	0	0	0	0	0	61	9961	25	120010	10	120000	10	120000	50	5630640	0	40020	40039	40039	24955	0	3	25019	120010	20	120000	20	360000	40039	40039	1	1	120021	10	9	10	10	120000	10	0	0	0	0	7520	63	12	16	0	13	10	40030	120000	10	40040	40040	40040	40040	40040
120024	40039	299	0	0	0	0	0	0	61	9961	25	120010	10	120000	10	120000	50	5630640	0	40020	40039	40039	24955	0	3	25019	120010	20	120000	20	360000	40039	40039	1	1	120021	10	9	10	10	120000	10	0	0	0	0	7520	60	12	16	0	12	24	41679	120000	10	40040	40040	40040	41692	40040
120024	40039	300	0	0	0	0	0	0	61	9961	25	120010	10	120000	10	120000	50	5630640	0	40020	40039	41691	24955	0	3	25019	120010	20	120000	20	360000	40039	40039	1	1	120021	10	9	10	10	120000	10	0	0	0	0	7520	57	10	16	0	13	13	40030	120000	10	40040	40040	40040	40040	40040
120024	40039	300	0	0	0	0	0	0	61	9961	25	120010	10	120000	10	120000	50	5630640	0	40020	40039	40039	24955	0	3	25019	120010	20	120000	20	360000	40039	40039	1	1	120021	10	9	10	10	120000	10	0	0	0	0	7520	48	13	16	0	14	10	40030	120000	10	40040	40040	40040	40040	40040
120024	41691	300	0	0	0	0	0	0	726	9961	25	120010	10	120000	10	120000	50	5630640	0	40020	40039	40039	24955	0	3	25019	120010	20	120000	20	360000	40039	40039	1	1	120021	10	9	10	10	120000	10	0	0	0	0	7520	54	12	16	0	15	11	40030	120000	10	40040	40040	40040	40040	40040
120024	40039	299	0	0	0	0	0	0	61	9961	25	120010	10	120000	10	120000	50	5630640	0	40020	40039	40039	24955	0	3	26671	120010	20	120000	20	360000	40039	40039	1	1	120021	10	9	10	10	120000	10	0	0	3	0	7520	48	10	16	0	10	12	40030	120000	10	40040	40040	40040	40040	40040
120024	40039	300	0	0	0	0	0	0	61	9961	25	120010	10	120000	10	120000	50	5630640	0	40020	40039	40039	24955	0	3	25019	120010	20	120000	20	360000	40039	40039	1	1	120021	10	9	10	10	120000	10	1	0	0	0	7520	51	14	16	0	12	12	40030	120000	10	40040	40040	41692	40040	40040

FMLS (by element, 2D)

Test 1: uops

1000 unrolls and 1 iteration

Test 2: Latency 1->1

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 3: Latency 1->2

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 4: Latency 1->3

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 5: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 6: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations