Firestorm: FRINTX (vector, 4H) measurements

Code:

  frintx v0.4h, v0.4h

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	mmu table walk data (08)	09	l2 tlb miss data (0b)	18	1e	1f	3a	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 miss (a1)	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	3037	22	0	0	0	0	0	0	0	82	2547	25	1000	1000	1000	398160	1	3018	3037	3037	2414	3	2895	1000	1000	1000	3037	3037	1	1	1001	1000	0	0	0	0	0	73	1	16	1	1	2629	1000	3038	3038	3038	3038	3038
1004	3037	22	0	0	0	0	0	0	0	61	2547	25	1000	1000	1000	398160	1	3018	3037	3037	2414	3	2895	1000	1000	1000	3037	3037	1	1	1001	1000	0	0	0	0	0	73	1	16	1	1	2629	1000	3038	3038	3038	3038	3038
1004	3037	22	0	0	0	0	0	0	0	61	2547	25	1000	1000	1000	398160	1	3018	3037	3037	2414	3	2895	1000	1000	1000	3037	3037	1	1	1001	1000	0	0	0	0	0	73	1	16	1	1	2629	1000	3038	3038	3038	3038	3038
1004	3037	22	0	0	0	0	0	0	0	103	2547	25	1000	1000	1000	398160	1	3018	3037	3037	2414	3	2895	1000	1000	1000	3037	3037	1	1	1001	1000	0	0	0	0	0	73	1	16	1	1	2629	1000	3038	3038	3038	3038	3038
1004	3037	22	0	0	0	0	0	0	0	61	2547	25	1000	1000	1000	398160	1	3018	3037	3037	2414	3	2895	1000	1000	1000	3037	3037	1	1	1001	1000	0	1	0	2638	0	94	1	24	1	1	2645	1000	3038	3085	3086	3038	3085
1004	3084	23	0	1	0	1	132	0	0	177	2538	49	1000	1008	1150	398160	1	3054	3037	3037	2418	8	2895	1000	1162	1163	3084	3037	2	1	1001	1000	0	0	0	2618	2	73	1	16	1	1	2675	1000	3085	3038	3085	3074	3085
1004	3085	23	1	1	1	1	141	88	1	103	2538	42	1008	1000	1000	399512	1	3018	3037	3085	2415	3	2895	1000	1162	1162	3084	3085	2	1	1001	1000	4	0	4	2815	0	73	1	16	1	1	2629	1000	3038	3038	3038	3038	3038
1004	3037	23	0	0	0	0	0	0	0	84	2547	25	1000	1000	1000	398160	1	3018	3037	3037	2414	3	2895	1000	1000	1000	3037	3037	1	1	1001	1000	0	0	0	0	0	73	1	16	1	1	2629	1000	3038	3038	3038	3038	3038
1004	3037	23	0	0	0	0	0	0	0	126	2547	25	1000	1000	1000	398160	1	3018	3037	3037	2414	3	2895	1000	1000	1000	3037	3037	1	1	1001	1000	0	0	0	0	0	73	1	16	1	1	2629	1000	3038	3038	3038	3038	3038
1004	3037	22	0	0	0	0	12	0	0	251	2547	25	1000	1000	1000	398160	1	3018	3037	3037	2414	3	2895	1000	1000	1000	3037	3037	1	1	1001	1000	0	0	0	0	0	73	1	16	1	1	2629	1000	3038	3038	3038	3038	3038

Test 2: Latency 1->2

Code:

  frintx v0.4h, v0.4h

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	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 miss (a1)	ld unit uop (a6)	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
10204	30037	224	0	255	0	61	29547	25	10100	100	10000	100	10000	500	4277160	0	30018	30037	30037	28271	6	28740	10100	200	10008	200	10008	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	1	1	1	718	0	16	0	0	29646	10000	100	30038	30038	30038	30038	30038
10204	30037	224	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	1	30018	30037	30037	28271	6	28745	10100	200	10000	200	10000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	0	710	1	16	1	1	29633	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	1	30018	30037	30037	28264	3	28745	10100	200	10000	200	10000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	0	710	1	16	1	1	29633	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	15	0	61	29547	25	10100	100	10000	100	10000	500	4277160	1	30018	30037	30037	28264	3	28745	10100	200	10000	200	10000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	2	0	0	0	710	1	16	1	1	29633	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	0	30018	30037	30037	28264	3	28745	10100	200	10000	200	10000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	0	710	1	16	1	1	29695	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	0	30018	30037	30037	28264	3	28745	10100	200	10000	200	10000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	0	710	1	16	1	1	29633	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	0	30018	30037	30037	28264	3	28745	10100	200	10000	200	10000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	0	710	1	16	1	1	29633	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	0	30018	30037	30037	28264	3	28745	10100	200	10000	200	10000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	0	710	1	16	1	1	29633	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	0	30018	30037	30037	28264	3	28745	10100	200	10000	200	10000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	1	0	0	0	0	0	710	1	16	1	1	29633	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	3	0	726	29547	25	10100	100	10000	100	10000	500	4277160	0	30018	30037	30037	28264	3	28745	10100	200	10000	200	10000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	0	710	1	16	1	1	29633	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	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)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb access (a0)	l1d tlb miss (a1)	st unit uop (a7)	l1d cache writeback (a8)	a9	ac	c2	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
10024	30037	225	0	0	0	45	0	0	103	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	3	28767	10010	20	10000	20	10000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	1	0	0	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	224	0	0	0	12	0	0	103	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	3	28767	10010	20	10000	20	10000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	1	0	3	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	6	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	0	30018	30037	30037	28286	3	28767	10010	20	10000	20	10000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	1	0	0	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	36	0	0	103	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	3	28767	10010	20	10000	20	10000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	1	0	3	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	0	0	61	29547	25	10010	10	10000	10	10000	55	4277160	1	30018	30037	30037	28286	3	28767	10010	20	10000	20	10000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	2	0	1	0	0	0	0	640	2	16	2	2	29629	1	10000	10	30038	30038	30038	30085	30084
10024	30037	224	0	0	0	48	0	0	103	29538	45	10019	10	10000	10	10150	50	4279864	1	30054	30037	30037	28286	3	28785	10010	20	10000	20	10000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	1	0	3	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	3	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	3	28767	10010	20	10000	20	10000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	1	0	3	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	36	0	0	818	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30083	30037	28286	3	28767	10010	20	10000	20	10000	30037	30037	4	1	10021	10	9	10	10	10000	10	0	0	0	1	2	0	0	0	640	2	16	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	51	0	0	3894	29520	154	10037	18	10040	13	10450	60	4281216	1	30018	30037	30037	28286	3	28767	10010	20	10000	20	10000	30037	30037	2	1	10021	10	9	10	10	10000	10	2	4	1	1	0	19308	2	0	682	4	101	2	2	29629	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	1	7	1170	88	1	1335	29482	117	10028	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	3	28786	11217	20	10000	24	10324	30037	30037	1	1	10021	10	9	10	10	10000	10	2	4	1	1	2	25201	2	0	748	2	16	2	2	29629	0	10000	10	30180	30038	30085	30179	30038

Test 3: throughput

Count: 8

Code:

  frintx v0.4h, v8.4h
  frintx v1.4h, v8.4h
  frintx v2.4h, v8.4h
  frintx v3.4h, v8.4h
  frintx v4.4h, v8.4h
  frintx v5.4h, v8.4h
  frintx v6.4h, v8.4h
  frintx v7.4h, v8.4h

Initialization

  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.2505

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 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)	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
80204	20039	150	90	30	25	80108	100	80008	100	80020	500	640132	1	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	100	80000	100	0	0	1	1	1	5118	4	16	3	4	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	150	0	410	25	80108	100	80008	100	80020	500	640132	1	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	100	80000	100	0	0	1	1	1	5118	4	16	3	3	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	150	0	30	25	80108	100	80008	100	80020	500	640132	1	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	100	80000	100	0	0	1	1	1	5118	3	16	3	3	20086	80000	100	20040	20040	20040	20040	20040
80204	20039	150	6	30	25	80108	100	80008	100	80020	500	640132	1	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	100	80000	100	0	0	1	1	1	5118	4	16	4	3	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	150	0	30	25	80108	100	80008	100	80020	500	640132	1	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	100	80000	100	0	0	1	1	1	5118	3	16	4	4	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	150	0	30	25	80108	100	80008	100	80020	500	640132	1	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	100	80000	100	0	0	1	1	1	5118	4	16	4	3	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	150	30	30	25	80108	100	80008	100	80020	500	640132	1	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	100	80000	100	0	0	1	1	1	5118	3	16	3	4	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	150	453	30	25	80108	100	80008	100	80020	500	640132	1	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	100	80000	100	0	0	1	1	1	5118	3	16	4	3	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	150	0	30	25	80108	100	80008	100	80020	500	640132	1	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	100	80000	100	0	0	1	1	1	5118	3	16	3	3	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	150	0	30	25	80108	100	80008	100	80020	500	640132	1	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	100	80000	100	0	0	1	1	1	5118	4	16	3	4	20036	80000	100	20040	20040	20040	20040	20040

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2505

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

retire uop (01)	cycle (02)	03	18	1e	1f	3f	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6d	6e	map 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	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
80024	20039	150	0	12	0	82	25	80010	10	80000	10	80000	50	640000	1	20020	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	1	3	5020	1	16	1	1	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	150	0	36	0	40	25	80010	10	80000	10	80000	50	640000	0	20020	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	0	21	5020	1	16	1	1	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	150	0	27	0	82	25	80010	10	80000	10	80000	50	640000	0	20020	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	1	0	5020	1	16	1	1	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	149	0	30	0	40	25	80010	10	80000	10	80000	50	640000	0	20020	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	1	3	5020	1	16	1	1	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	150	0	0	0	82	25	80010	10	80000	10	80000	50	640000	1	20020	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	1	3	5020	1	16	1	1	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	150	0	186	0	82	25	80010	10	80000	10	80000	50	640000	1	20020	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	1	3	5020	1	16	1	1	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	150	0	30	0	40	25	80010	10	80000	10	80000	50	640000	1	20020	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	1	3	5020	1	16	1	1	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	150	0	39	0	82	25	80010	10	80000	10	80000	50	640000	0	20020	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	0	3	5020	1	16	1	1	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	150	0	309	0	82	25	80010	10	80000	10	80000	50	640000	1	20020	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	0	0	5020	1	16	1	1	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	150	0	9	0	40	25	80010	10	80000	10	80000	50	640000	1	20020	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	1	3	5020	1	16	1	1	20036	80000	10	20040	20040	20040	20040	20040

FRINTX (vector, 4H)

Test 1: uops

1000 unrolls and 1 iteration

Test 2: Latency 1->2

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 3: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations