Firestorm: SQDMLAL2 (vector, 2D) measurements

Code:

  sqdmlal2 v0.2d, v1.4s, v2.4s

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)	st unit uop (a7)	l1d cache writeback (a8)	ac	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	f5	f6	f7	f8	fd
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	9	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	94	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038

Test 2: Latency 1->1

Code:

  sqdmlal2 v0.2d, v1.4s, v2.4s

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0037

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

retire uop (01)	cycle (02)	03	mmu table walk instruction (07)	mmu table walk data (08)	09	l2 tlb miss instruction (0a)	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)	ld unit uop (a6)	l1d cache writeback (a8)	ac	branch cond mispred nonspec (c5)	branch mispred nonspec (cb)	cd	cf	d0	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	30037	225	1	0	0	1	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	0	30037	30037	28272	7	28741	10100	200	10008	200	30024	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	1	1	1	720	0	1	16	1	1	29650	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	1	0	0	1	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	0	30037	30037	28272	7	28740	10100	200	10008	200	30024	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	1	1	1	718	0	1	16	1	1	29651	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	1	0	0	1	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	0	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	710	1	3	16	3	3	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	0	0	61	29548	25	10100	100	10000	100	10000	522	4277313	1	30018	0	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	710	1	3	16	3	3	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	0	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	710	1	3	16	3	3	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	9	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	0	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	710	1	3	16	3	3	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	224	0	0	0	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	0	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	710	1	3	16	3	3	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	224	0	0	0	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	0	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	710	1	3	16	3	3	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	0	0	470	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	0	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	712	1	3	16	3	3	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	0	30037	30037	28265	27	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	710	1	3	16	3	3	29634	0	10000	100	30038	30038	30038	30038	30038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

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

retire uop (01)	cycle (02)	03	mmu table walk data (08)	3a	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb miss (a1)	l1d cache writeback (a8)	ac	c2	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
10024	30037	224	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29548	25	10019	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	1	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30083
10024	30037	225	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	1	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	3	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	3	2	29630	10000	10	30038	30038	30038	30038	30038

Test 3: Latency 1->2

Code:

  sqdmlal2 v0.2d, v0.4s, v1.4s

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0037

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

retire uop (01)	cycle (02)	03	mmu table walk data (08)	l2 tlb miss data (0b)	18	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)	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d cache writeback (a8)	ac	c2	cf	d2	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	30037	225	0	0	0	0	740	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	195	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	224	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	6	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	533	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	435	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	631	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	484	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	185	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	684	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	451	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

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

retire uop (01)	cycle (02)	03	19	1e	1f	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	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)	ac	c2	cd	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10024	30037	225	0	0	0	1183	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	640	6	16	3	4	29630	10000	10	30038	30132	30038	30038	30038
10024	30037	225	0	57	0	61	29548	25	10010	10	10000	12	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	640	4	16	3	5	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	1	640	4	16	4	3	29630	10000	10	30038	30038	30038	30038	30038
10024	30084	225	0	0	0	61	29548	83	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	2	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	640	4	16	4	3	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30180	1	1	10021	10	9	10	10	10000	10	0	2	0	0	0	0	640	3	16	3	4	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	82	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	640	4	16	6	4	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	39	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	640	4	16	3	4	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	82	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	640	3	16	3	4	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	640	3	16	4	3	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	2	2	1	0	0	0	640	3	16	3	4	29630	10000	10	30038	30038	30038	30038	30038

Test 4: Latency 1->3

Code:

  sqdmlal2 v0.2d, v1.4s, v0.4s

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0037

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

retire uop (01)	cycle (02)	03	1e	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d cache writeback (a8)	ac	branch cond mispred nonspec (c5)	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	30037	225	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30226	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	2	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30133	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	441	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

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

retire uop (01)	cycle (02)	03	mmu table walk data (08)	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)	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	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30225	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28305	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	13	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	94	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10160	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	0	10000	10	30085	30038	30038	30038	30038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  sqdmlal2 v0.2d, v8.4s, v9.4s
  movi v1.16b, 0
  sqdmlal2 v1.2d, v8.4s, v9.4s
  movi v2.16b, 0
  sqdmlal2 v2.2d, v8.4s, v9.4s
  movi v3.16b, 0
  sqdmlal2 v3.2d, v8.4s, v9.4s
  movi v4.16b, 0
  sqdmlal2 v4.2d, v8.4s, v9.4s
  movi v5.16b, 0
  sqdmlal2 v5.2d, v8.4s, v9.4s
  movi v6.16b, 0
  sqdmlal2 v6.2d, v8.4s, v9.4s
  movi v7.16b, 0
  sqdmlal2 v7.2d, v8.4s, v9.4s

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 tlb access (a0)	ld unit uop (a6)	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
160204	20091	150	699	989	25	80100	100	80000	100	80000	500	640000	1	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10114	3	16	4	4	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	151	0	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10114	4	16	4	5	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	150	0	83	25	80100	100	80000	100	80000	500	640000	1	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	1	3	0	10114	3	16	5	4	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	150	12	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10113	4	16	3	4	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	150	0	39	25	80100	100	80000	100	80000	500	640000	1	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10115	5	16	5	4	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	151	0	39	25	80100	100	80000	100	80000	500	640000	0	20045	20076	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10115	4	16	5	5	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	150	39	39	25	80100	100	80000	100	80000	500	640000	1	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	1	10113	4	16	4	4	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	150	138	39	25	80100	100	80000	100	80000	500	640000	1	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10113	5	16	3	5	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	150	0	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10114	5	16	5	5	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	151	6	39	25	80100	100	80000	100	80000	500	640000	1	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	10115	3	16	3	4	20061	0	160000	100	20065	20065	20065	20065	20065

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2507

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

retire uop (01)	cycle (02)	03	mmu table walk instruction (07)	1e	3f	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	5f	60	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)	91	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	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	1	0	45	27	80012	12	80000	12	80000	62	640000	1	1	0	20032	20051	20051	3	22	80012	20	80000	20	240000	20051	20051	1	1	160021	10	9	0	10	10	160000	10	0	10043	3	1	1	36	25	2	1	1	14	14	20048	2	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	150	0	0	45	27	80012	12	80000	12	80000	62	640000	1	1	5	20032	20051	20051	3	22	80012	20	80000	20	240000	20060	20060	1	1	160021	10	9	0	10	10	160000	10	0	10041	11	4	2	11	34	4	2	2	14	16	20057	2	40	2	160000	10	20061	20061	20061	20061	20061
160024	20060	150	0	0	51	29	80012	12	80000	12	80000	62	640000	0	1	5	20041	20060	20060	3	22	80012	20	80000	20	240000	20060	20060	1	1	160021	10	9	0	10	10	160000	10	0	10038	13	4	1	14	25	2	1	1	16	11	20048	2	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	150	0	0	45	27	80012	12	80000	12	80000	62	640000	1	1	10	20032	20051	20051	3	22	80012	20	80000	20	240000	20371	20051	2	1	160021	10	9	0	10	10	160000	10	0	10042	13	4	1	20	25	2	1	1	14	16	20048	2	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	150	0	0	140	27	80012	12	80000	12	80000	62	640000	1	1	10	20032	20051	20051	3	22	80012	20	80000	20	240000	20051	20051	1	1	160021	10	9	0	10	10	160000	10	0	10037	13	5	1	13	25	2	1	1	13	10	20048	2	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	151	0	0	66	27	80012	12	80000	12	80000	62	640000	1	1	10	20032	20051	20051	3	22	80012	20	80000	20	240000	20051	20051	1	1	160021	10	9	0	10	10	160000	10	0	10037	13	5	1	15	25	2	1	1	13	11	20048	2	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	150	0	0	45	27	80012	12	80000	12	80000	62	640000	1	1	10	20032	20051	20051	3	22	80012	20	80000	20	240000	20051	20051	1	1	160021	10	9	0	10	10	160000	10	0	10034	13	5	1	20	25	2	1	1	13	11	20048	2	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	150	0	0	45	27	80012	12	80000	12	80000	62	640000	1	1	10	20032	20051	20051	3	22	80012	20	80000	20	240000	20051	20051	1	1	160021	10	9	0	10	10	160000	10	0	10060	13	5	1	17	25	2	1	1	13	10	20048	2	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	150	0	0	45	27	80012	12	80000	12	80000	62	640000	1	1	10	20032	20051	20051	3	22	80012	20	80000	20	240000	20051	20051	1	1	160021	10	9	0	10	10	160000	10	0	10041	13	5	1	13	25	2	2	1	18	11	20048	2	20	1	160000	10	20052	20052	20052	20052	20052
160024	20060	150	0	0	45	27	80012	12	80000	12	80000	62	640000	1	1	10	20032	20051	20051	3	22	80012	20	80000	20	240000	20051	20060	1	1	160021	10	9	0	10	10	160000	10	0	10036	13	5	1	17	25	2	1	1	11	18	20048	2	20	1	160000	10	20052	20052	20052	20052	20052

Test 6: throughput

Count: 16

Code:

  sqdmlal2 v0.2d, v16.4s, v17.4s
  sqdmlal2 v1.2d, v16.4s, v17.4s
  sqdmlal2 v2.2d, v16.4s, v17.4s
  sqdmlal2 v3.2d, v16.4s, v17.4s
  sqdmlal2 v4.2d, v16.4s, v17.4s
  sqdmlal2 v5.2d, v16.4s, v17.4s
  sqdmlal2 v6.2d, v16.4s, v17.4s
  sqdmlal2 v7.2d, v16.4s, v17.4s
  sqdmlal2 v8.2d, v16.4s, v17.4s
  sqdmlal2 v9.2d, v16.4s, v17.4s
  sqdmlal2 v10.2d, v16.4s, v17.4s
  sqdmlal2 v11.2d, v16.4s, v17.4s
  sqdmlal2 v12.2d, v16.4s, v17.4s
  sqdmlal2 v13.2d, v16.4s, v17.4s
  sqdmlal2 v14.2d, v16.4s, v17.4s
  sqdmlal2 v15.2d, v16.4s, v17.4s

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): 0.2503

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

retire uop (01)	cycle (02)	03	mmu table walk data (08)	l2 tlb miss data (0b)	18	19	1e	1f	37	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)	l1d cache writeback (a8)	ac	c2	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	40049	300	0	0	0	0	42	0	0	715	0	25	160100	100	160017	100	160000	500	1280000	1	40029	40039	40039	19973	0	3	20007	160100	200	160000	200	480000	40048	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	10110	1	16	1	1	40036	0	160000	100	40040	40041	40040	40040	40049
160204	40039	300	0	0	0	0	0	0	1	51	0	25	160100	100	160000	100	160000	500	1280000	1	40029	40276	40091	19980	0	3	20006	160100	200	160000	200	480000	40039	40049	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	10110	1	16	1	1	40068	0	160000	100	40040	40050	40049	40040	40049
160204	40049	300	0	0	0	0	0	0	17	50	0	25	160100	100	160017	100	160000	500	1280000	1	40030	40039	40039	19973	0	3	19997	160100	200	160000	200	480000	40039	40040	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	10110	1	16	1	1	40046	0	160000	100	40049	40072	40040	40040	40049
160204	40039	300	0	0	0	0	0	0	17	61	0	25	160117	100	160000	100	160000	500	1320000	1	40020	40039	40039	19973	0	3	19997	160100	200	160000	200	480000	40039	40049	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	10110	1	16	1	1	40036	0	160000	100	40040	40040	40050	40049	40050
160204	40049	300	0	0	0	0	0	0	0	61	0	25	160117	100	160000	100	160000	500	1280000	1	40020	40039	40039	19973	0	3	19997	160100	200	160000	200	480000	40039	40048	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	10110	1	16	1	1	40036	0	160000	100	40049	40040	40050	40049	40050
160204	40048	300	0	0	0	0	0	0	0	41	0	25	160100	100	160017	100	160000	500	1320000	1	40029	40049	40048	19973	0	3	19997	160100	200	160000	200	480000	40039	40049	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	10110	1	16	1	1	40036	0	160000	100	40050	40049	40040	40041	40040
160204	40048	300	0	0	0	0	0	0	0	51	25	25	160100	100	160000	100	160000	500	5387188	1	40020	40148	40039	19973	0	3	19997	160100	200	160000	200	480000	40048	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	10110	1	16	1	1	40036	0	160000	100	40040	40050	40040	40040	40072
160204	40039	300	0	0	0	0	0	0	0	51	0	25	160117	100	160000	100	160000	500	1280000	1	40020	40039	40049	19973	7	3	19997	160100	200	160000	200	480000	40039	40048	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	10110	1	16	1	1	40068	0	160000	100	40041	40040	40041	40040	40049
160204	40039	300	0	0	0	0	0	0	1	41	25	25	160117	100	160001	100	160000	500	1280000	1	40020	40039	40039	19973	0	3	19997	160100	200	160000	200	480000	40039	40049	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	10110	1	16	1	1	40045	0	160000	100	40072	40040	40040	40041	40041
160204	40040	300	0	0	0	0	0	0	0	41	0	25	160100	100	160000	100	160000	500	1280000	1	40020	40039	40039	19973	0	3	20007	160100	200	160000	200	480000	40049	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	10110	1	16	1	1	40068	0	160000	100	40040	40040	40050	40040	40040

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	l2 tlb miss data (0b)	1e	37	3f	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	5f	60	69	6d	6e	map 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	st unit uop (a7)	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	eb	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160024	40039	300	0	24	0	56	25	160010	10	160000	10	160000	50	1280000	1	1	40020	40040	40039	19996	3	20028	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	10024	6	2	2	30	16	2	1	1	23	23	40046	0	15	5	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	39	0	52	25	160010	10	160000	10	160000	50	1280000	0	1	40020	40040	40039	19996	3	20019	160010	20	160000	20	480000	40039	40048	1	1	160021	10	9	10	10	160000	10	0	0	0	10024	6	2	2	19	16	4	2	2	26	23	40036	0	30	10	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	0	52	25	160010	10	160000	10	160000	50	1280000	0	1	40020	40040	40049	19996	3	20019	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	10024	6	2	2	22	16	4	2	2	21	21	40036	0	30	10	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	0	52	25	160010	10	160000	10	160000	50	1280000	0	1	40020	40045	40048	19996	3	20053	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	10024	6	2	2	22	16	4	2	2	21	22	40036	0	30	10	160000	10	40040	40053	40040	40050	40040
160024	40039	300	0	0	0	52	25	160010	10	160000	10	160000	50	1280000	0	1	40020	40040	40039	19996	3	20019	160010	20	160000	20	480000	40039	40049	1	1	160021	10	9	10	10	160000	10	0	0	0	10022	3	1	1	25	16	2	1	1	22	22	40036	0	15	5	160000	10	40040	40040	40040	40040	40040
160024	40039	299	0	0	0	46	25	160010	10	160000	10	160000	50	1280000	1	1	40020	40040	40049	19996	3	20019	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	10022	3	1	1	22	16	2	1	1	21	22	40036	2	15	5	160000	10	40040	40040	40040	40050	40040
160024	40039	299	0	60	0	46	25	160010	10	160000	10	160000	50	1280000	1	1	40020	40051	40039	19996	3	20019	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	10022	3	1	1	23	16	2	1	1	21	22	40036	0	15	6	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	0	52	25	160010	10	160000	10	160000	50	2438865	0	1	40020	40040	40048	19996	3	20019	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	10024	6	2	2	22	16	4	2	2	23	22	40036	0	30	10	160000	10	40050	40040	40040	40040	40040
160024	40039	300	0	0	0	46	25	160010	10	160000	10	160000	50	1280000	1	1	40020	40052	40039	19996	3	20019	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	10022	3	1	1	21	16	2	1	1	16	24	40036	0	15	5	160000	10	40040	40040	40040	40040	40050
160024	40039	300	0	0	0	46	25	160010	10	160000	10	160000	50	1280000	1	1	40020	40040	40039	19996	3	20019	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	10024	6	2	2	22	16	4	2	2	24	24	40037	0	30	10	160000	10	40040	40040	40040	40040	40040

SQDMLAL2 (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