Firestorm: SUQADD (vector, 4S) measurements

Code:

  suqadd v0.4s, v1.4s

Initialization

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

(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	3037	22	0	84	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	73	2	16	3	3	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	1163	2000	3037	3037	1	1	1001	1000	3	0	73	3	16	3	3	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	1	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	73	3	16	3	3	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	2000	3037	3037	1	1	1001	1000	0	0	73	3	16	3	3	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	6	61	2548	25	1000	1000	1000	398313	1	3018	3037	3037	2415	3	2895	1148	1000	2000	3037	3037	1	1	1001	1000	0	0	73	3	16	3	3	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	93	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	73	3	16	3	3	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	2000	3037	3037	1	1	1001	1000	0	0	73	3	16	3	3	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	2000	3037	3037	1	1	1001	1000	0	0	73	3	16	3	3	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	387	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	73	3	16	3	3	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	2000	3037	3037	1	1	1001	1000	0	0	73	3	16	3	3	2630	1000	3038	3038	3038	3038	3038

Test 2: Latency 1->1

Code:

  suqadd 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)	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	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	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10205	30037	225	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	0	1	16	1	1	29690	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30230	30037	28265	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	0	2	25	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	5	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30054	30037	30037	28265	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	0	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	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	mmu table walk data (08)	l2 tlb miss data (0b)	18	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)	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	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	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	4	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30084	225	0	0	0	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	224	0	0	0	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30162	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30085	30038	30038	30038	30038
10024	30037	224	0	0	0	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	9	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30180	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	3	0	0	640	2	16	2	2	29630	0	10000	10	30321	30180	30180	30179	30228
10024	30142	227	1	1	3	3	414	264	2123	29503	106	10035	15	10040	16	10447	60	4282741	1	30126	30226	30131	28295	28	28878	10460	20	10490	22	20982	30226	30309	2	1	10021	10	9	10	10	10000	10	0	0	0	0	2730	640	3	16	2	2	29648	0	10000	10	30038	30038	30038	30038	30038

Test 3: Latency 1->2

Code:

  suqadd v0.4s, v0.4s

Initialization

  movi v0.16b, 1

(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	l2 tlb miss data (0b)	18	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)	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)	ac	branch cond mispred nonspec (c5)	branch mispred nonspec (cb)	cd	cf	map dispatch bubble (d6)	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	30037	224	0	0	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	1	30162	30037	30085	28271	6	28741	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	4	0	0	1	1	1	761	16	0	29645	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	12	0	61	29547	25	10100	100	10000	100	10000	500	4277160	1	30018	30084	30037	28271	10	28741	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	1	1	1	717	16	0	29645	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	1	30018	30037	30037	28271	7	28740	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	1	1	1	717	16	0	29645	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	0	30018	30037	30037	28271	6	28740	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	1	1	1	718	16	0	29646	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	0	30018	30037	30037	28271	7	28740	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	1	1	1	718	16	0	29645	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	1	30018	30084	30037	28271	7	28741	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	3	1	1	1	718	16	0	29646	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	2	0	0	293	29547	25	10100	102	10016	100	10000	511	4277160	0	30018	30037	30179	28271	6	28740	10100	206	10008	208	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	1	1	1	717	16	0	29645	0	10000	100	30038	30038	30038	30038	30038
10204	30037	224	0	0	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	1	30018	30037	30037	28271	6	28741	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	3	1	1	1	717	16	0	29646	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	3	0	61	29547	25	10100	100	10000	100	10000	500	4277160	1	30018	30037	30037	28271	6	28740	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	1	1	1	718	16	0	29646	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	1	30018	30037	30037	28271	7	28740	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	1	1	1	717	16	0	29645	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	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	ec	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10024	30037	225	0	61	29547	25	10010	10	10000	10	10000	50	4277160	0	30018	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	224	0	61	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	61	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	61	29547	25	10010	10	10000	10	10000	50	4277160	0	30018	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	61	29547	25	10010	10	10000	10	10000	50	4277160	0	30162	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30084	1	1	10021	10	9	10	10	10000	10	0	0	640	3	24	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	61	29547	25	10010	10	10000	10	10000	50	4277160	0	30018	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	61	29547	25	10010	10	10000	10	10000	50	4277160	0	30018	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29629	3	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	61	29547	25	10010	10	10000	10	10000	50	4277160	0	30018	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29629	0	10000	10	30085	30038	30038	30038	30038
10024	30037	225	0	61	29547	25	10010	10	10000	10	10000	50	4277160	0	30018	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	224	0	61	29547	25	10010	10	10000	10	10000	50	4277160	0	30018	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038

Test 4: throughput

Count: 8

Code:

  movi v0.16b, 0
  suqadd v0.4s, v8.4s
  movi v1.16b, 0
  suqadd v1.4s, v8.4s
  movi v2.16b, 0
  suqadd v2.4s, v8.4s
  movi v3.16b, 0
  suqadd v3.4s, v8.4s
  movi v4.16b, 0
  suqadd v4.4s, v8.4s
  movi v5.16b, 0
  suqadd v5.4s, v8.4s
  movi v6.16b, 0
  suqadd v6.4s, v8.4s
  movi v7.16b, 0
  suqadd v7.4s, v8.4s

Initialization

  movi v8.16b, 9

(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	mmu table walk instruction (07)	l2 tlb miss instruction (0a)	18	1e	3f	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6b	6d	6e	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d cache writeback (a8)	ac	branch cond mispred nonspec (c5)	branch mispred nonspec (cb)	cd	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160204	20091	151	1	1	0	0	29	25	80116	100	80016	100	80028	500	640196	1	20045	0	20065	20065	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10123	3	16	4	4	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	1	1	0	33	29	25	80116	100	80016	100	80028	500	640196	0	20045	0	20065	20065	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10123	4	16	4	5	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	151	1	1	0	0	29	25	80116	100	80016	100	80028	500	640196	0	20045	0	20065	20065	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10123	5	16	5	4	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	1	1	0	399	29	25	80116	100	80016	100	80028	500	640196	0	20045	0	20065	20065	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10123	5	16	5	5	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	1	1	0	0	29	25	80116	100	80016	100	80028	500	640196	0	20045	0	20065	20065	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10123	4	16	4	5	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	1	1	0	312	29	25	80116	100	80016	100	80028	500	640196	0	20045	0	20065	20065	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10122	5	16	5	5	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	151	1	1	0	456	29	25	80116	100	80016	100	80028	500	640196	0	20045	0	20065	20065	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10123	4	16	5	5	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	1	1	0	0	219	25	80116	100	80016	100	80028	500	640196	0	20045	0	20065	20065	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10123	5	16	5	4	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	1	1	0	0	29	25	80116	100	80016	100	80028	500	640196	0	20045	0	20065	20065	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10122	4	16	4	4	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	1	1	0	0	29	25	80116	100	80016	100	80028	500	640196	0	20045	0	20065	20065	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10123	4	16	3	5	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	18	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	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 miss (a1)	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	ea	ec	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160024	20082	150	0	0	45	27	80010	10	80000	10	80000	50	640000	2	1	20032	20051	20051	3	22	80010	20	80000	20	160000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	0	0	10036	6	2	2	11	25	2	1	1	10	7	20048	20	1	160000	10	20052	20052	20052	20061	20052
160024	20051	150	0	24	68	27	80010	10	80000	10	80000	50	640000	0	1	20032	20051	20051	3	22	80010	20	80000	20	160000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	0	0	10033	6	2	2	10	34	4	2	2	10	10	20057	40	2	160000	10	20061	20061	20163	20061	20061
160024	20060	150	0	0	51	29	80010	10	80000	10	80000	50	640000	3	1	20041	20060	20060	3	22	80010	20	80000	20	160000	20060	20051	1	1	160021	10	9	10	10	160000	10	0	0	0	10030	3	1	1	10	25	2	1	1	7	10	20048	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	151	0	0	45	29	80010	10	80000	10	80000	50	640000	0	0	20032	20051	20051	3	22	80010	20	80000	20	160000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	0	0	10030	3	1	1	10	25	4	1	1	10	7	20048	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	150	0	372	51	27	80010	10	80000	10	80000	50	640000	3	1	20032	20051	20051	3	22	80010	20	80000	20	160000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	0	0	10034	3	1	1	7	25	2	1	1	7	10	20048	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	151	0	0	45	27	80010	10	80000	10	80000	50	640000	3	0	20032	20051	20051	3	22	80010	20	80000	20	160000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	0	0	10033	6	1	1	10	25	2	1	1	7	10	20048	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	150	0	0	45	27	80010	10	80000	10	80000	50	640000	3	1	20032	20051	20051	3	22	80010	20	80000	20	160000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	0	0	10033	3	1	1	10	25	2	1	1	10	10	20048	20	1	160000	10	20052	20052	20052	20061	20061
160024	20051	150	1	24	45	27	80010	10	80000	10	80000	50	640000	0	1	20032	20051	20051	3	22	80010	20	80000	20	160000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	0	0	10030	3	1	1	7	25	2	1	1	7	10	20048	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	150	0	0	45	27	80010	10	80000	10	80000	50	640000	0	1	20032	20051	20051	3	22	80010	20	80000	20	160000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	0	0	10033	3	1	1	10	25	2	1	1	10	7	20048	20	2	160000	10	20061	20052	20052	20052	20052
160024	20051	155	0	0	45	27	80010	10	80000	10	80000	50	640000	0	1	20032	20051	20051	3	22	80010	20	80000	20	160000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	0	0	10031	3	1	1	10	25	2	1	1	10	7	20048	20	1	160000	10	20052	20052	20052	20052	20052

Test 5: throughput

Count: 16

Code:

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

Initialization

  movi v16.16b, 17

(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)	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)	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	ld unit uop (a6)	l1d cache writeback (a8)	a9	ac	c2	branch cond mispred nonspec (c5)	branch mispred nonspec (cb)	cd	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160204	40059	300	0	0	30	25	160108	100	160008	100	160020	500	1280132	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	1	1	1	10118	1	16	1	1	40036	160000	100	40040	40040	40040	40040	40040
160204	40090	299	0	0	30	25	160108	100	160008	100	160020	500	1280132	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	1	1	1	10118	1	16	1	0	40036	160000	100	40040	40040	40040	40040	40040
160204	40039	299	0	0	35	25	160108	100	160008	100	160020	500	1280132	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	1	1	1	10118	1	16	1	1	40036	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	0	30	25	160108	100	160008	100	160020	500	1280132	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	1	1	1	10118	1	16	1	1	40036	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	0	30	25	160108	100	160008	100	160020	500	1280132	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	583	0	1	1	1	10118	1	16	1	1	40036	160000	100	40040	40040	40040	40040	40040
160204	40039	299	0	0	30	25	160108	100	160008	100	160020	500	1280132	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	1	1	1	10118	1	16	1	1	40036	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	0	30	25	160108	100	160008	100	160020	500	1280132	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	1	1	1	10118	1	16	1	1	40036	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	0	30	25	160108	100	160008	101	160020	500	1280132	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	1	1	1	10118	1	16	1	1	40036	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	0	30	25	160108	100	160008	100	160020	500	1280132	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	1	1	1	10118	1	16	1	1	40036	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	0	30	25	160184	100	160008	100	160020	500	1280132	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	1	1	1	10118	1	16	1	1	40036	160000	100	40040	40040	40040	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	mmu table walk instruction (07)	l2 tlb miss instruction (0a)	18	1e	1f	3f	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	5f	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 cache writeback (a8)	a9	ac	branch mispred nonspec (cb)	cd	cf	d0	d2	l1i cache miss demand (d3)	d5	map dispatch bubble (d6)	d9	da	db	dd	fetch restart (de)	e0	ea	eb	ec	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160024	40050	300	0	0	0	0	0	46	25	160010	10	160000	10	160000	50	1280000	1	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	1	0	0	0	0	10022	3	1	1	6	16	2	1	1	6	9	40127	15	6	1	160000	10	40812	40860	41220	41219	40861
160024	40915	319	2	2	4	0	0	127	25	160010	10	160000	10	160000	50	1280000	1	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	3	0	0	10022	3	1	1	9	16	2	1	2	6	9	40036	15	5	0	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	0	36	88	115	25	160010	10	160000	10	160000	50	1280000	1	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40128	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	10022	3	1	1	9	16	2	1	1	6	9	40036	15	5	0	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	0	0	0	46	25	160010	10	160000	10	160000	50	1280000	1	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	10022	3	1	1	7	16	2	1	1	9	9	40036	15	5	0	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	0	0	0	46	25	160010	10	160000	10	160000	50	1280000	1	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	10022	3	1	1	9	16	2	1	1	9	6	40036	15	5	0	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	0	0	0	46	25	160010	10	160000	10	160000	50	1280000	3	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	10024	3	1	1	10	16	2	1	1	9	6	40036	15	5	0	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	0	0	0	236	25	160010	10	160000	10	160000	50	1280000	1	1	40020	0	40039	40039	19996	3	20048	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	10022	3	1	1	9	16	2	1	1	10	9	40036	15	5	0	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	0	0	0	46	25	160010	10	160000	10	160000	50	1280000	1	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	10022	3	1	1	10	16	2	1	1	9	7	40036	15	5	0	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	0	0	0	46	25	160010	10	160000	10	160000	50	1280000	0	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	10022	3	1	1	9	16	2	1	1	9	6	40036	15	5	0	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	0	0	0	258	25	160010	10	160000	10	160000	50	1280000	2	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40088	11	1	160021	10	9	10	10	160000	10	34	0	3	0	0	10022	6	2	2	9	16	2	1	1	6	9	40036	15	10	0	160000	10	40040	40040	40040	40040	40040

SUQADD (vector, 4S)

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: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 5: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations