Firestorm: FRINT64X (vector, 2D) measurements

Code:

  frint64x v0.2d, v0.2d

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)	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 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)	f5	f6	f7	f8	fd
1004	3037	25	0	0	0	0	89	2547	25	1000	1000	1150	398160	0	3018	3037	3037	2417	3	2895	1000	1161	1000	3037	3085	2	1	1001	1000	11	0	0	0	0	73	1	16	1	1	2629	1000	3085	3038	3038	3038	3085
1004	3073	32	1	1	0	0	686	2547	44	1000	1008	1000	399512	0	3054	3084	3037	2415	7	2925	1150	1168	1161	3133	3120	3	1	1001	1000	0	0	2	0	3320	111	1	45	1	1	2727	1000	3136	3086	3085	3085	3085
1004	3084	25	0	0	0	0	61	2547	43	1000	1000	1150	398160	0	3054	3133	3084	2414	10	2919	1150	1168	1121	3085	3132	2	1	1001	1000	0	2	0	0	3640	73	1	16	1	1	2629	1000	3038	3038	3038	3038	3038
1004	3037	24	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	23	0	0	0	0	61	2547	25	1000	1000	1000	398160	0	3018	3037	3037	2414	3	2895	1000	1000	1000	3037	3037	1	1	1001	1000	0	0	0	1	0	73	1	16	1	1	2629	1000	3038	3038	3038	3038	3038
1004	3037	24	0	0	0	0	61	2547	25	1000	1000	1000	398160	0	3018	3037	3037	2414	3	2895	1000	1000	1000	3037	3037	1	1	1001	1000	0	0	0	0	30	73	1	16	1	1	2629	1000	3038	3038	3038	3038	3038
1004	3037	24	0	0	0	0	82	2547	25	1000	1000	1000	398160	0	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	24	0	0	0	0	61	2547	25	1000	1000	1000	398160	0	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	24	0	0	0	0	61	2547	25	1000	1000	1000	398160	0	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	61	2547	25	1000	1000	1000	398160	0	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:

  frint64x v0.2d, v0.2d

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	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)	ld unit uop (a6)	st unit uop (a7)	l1d cache writeback (a8)	a9	ac	branch cond mispred nonspec (c5)	branch mispred nonspec (cb)	cd	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	30037	232	0	0	0	0	3	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	0	0	710	1	16	1	1	29633	0	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	0	0	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	0	0	710	1	16	1	1	29633	0	10000	100	30038	30038	30038	30038	30038
10204	30037	232	0	0	0	0	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	3	0	0	0	710	1	16	1	1	29633	0	10000	100	30038	30038	30038	30038	30038
10204	30037	242	0	0	0	0	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	0	0	710	1	16	1	1	29633	0	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	0	0	18	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	0	0	710	1	16	1	1	29652	0	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	0	0	0	0	0	108	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	0	0	710	1	16	1	1	29633	0	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	0	0	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	30134	2	1	10201	100	99	100	100	10000	100	2	4	2	0	0	4	3060	0	0	0	732	1	33	1	1	29702	2	10000	100	30085	30085	30086	30133	30133
10204	30132	233	1	0	2	2	264	176	0	61	29538	44	10122	114	10008	104	10150	644	4279864	1	30054	30131	30132	28269	13	28763	10428	206	10331	202	10333	30132	30132	3	1	10201	100	99	100	100	10000	100	2	4	2	1	0	2	19450	0	0	0	850	1	91	3	1	29930	28	10000	100	30419	30407	30420	30408	30371
10204	30367	235	0	0	8	7	924	616	0	6082	29454	212	10246	149	10088	149	11050	737	4290527	1	30378	30368	30502	28300	48	28925	11802	218	11664	236	11813	30419	30514	9	1	10201	100	99	100	100	10000	100	2	2	2	0	0	0	32695	0	0	0	942	1	97	2	1	30029	31	10000	100	30518	30562	30325	30561	30597
10204	30519	237	1	1	11	10	1344	994	0	7128	29502	233	10243	152	10088	153	11650	763	4292032	1	30414	30563	30563	28310	57	28934	11032	236	11821	220	11652	30567	30565	12	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	27893	0	0	0	944	3	127	1	2	29958	37	10000	100	30468	30326	30472	30454	30508

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	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	30037	224	210	89	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	640	2	16	2	2	29629	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	10000	30037	30037	1	1	10021	10	9	10	10	10000	10	7	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	12	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	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	15	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	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	21	61	29547	25	10010	10	10000	10	10150	50	4277160	0	30018	30037	30037	28286	3	28842	10010	20	10000	20	10000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	49	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	33	61	29547	25	10010	10	10008	10	10000	50	4277160	0	30018	30037	30037	28286	3	28767	10010	20	10329	20	10000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	5	9	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	27	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	640	2	17	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	36	341	29547	25	10010	10	10000	10	10000	50	4277160	0	30018	30037	30037	28286	6	28767	10010	20	10000	20	10000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	18	103	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	1	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	3	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	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038

Test 3: throughput

Count: 8

Code:

  frint64x v0.2d, v8.2d
  frint64x v1.2d, v8.2d
  frint64x v2.2d, v8.2d
  frint64x v3.2d, v8.2d
  frint64x v4.2d, v8.2d
  frint64x v5.2d, v8.2d
  frint64x v6.2d, v8.2d
  frint64x v7.2d, v8.2d

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	19	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 load (95)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb miss (a1)	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	20059	155	0	0	51	25	80108	100	80008	100	80020	500	640132	0	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	0	100	80000	100	0	0	3	1	1	1	5118	0	16	0	0	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	156	0	0	64	26	80116	100	80016	100	80028	500	640196	1	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	0	100	80000	100	0	16	0	1	1	1	5118	1	16	0	0	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	155	0	0	30	25	80108	100	80008	100	80020	500	640132	0	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	0	100	80000	100	0	4	3	1	1	1	5118	0	16	0	0	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	155	0	0	30	25	80108	100	80008	100	80020	500	640132	0	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	0	100	80000	100	0	0	0	1	1	1	5118	0	16	0	0	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	156	0	0	30	25	80108	100	80008	100	80020	500	640132	0	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	0	100	80000	100	0	0	0	1	1	1	5118	0	16	0	0	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	155	0	0	30	25	80108	100	80008	100	80020	500	640132	0	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	0	100	80000	100	0	0	0	1	1	1	5118	0	16	0	0	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	155	0	0	30	25	80108	100	80008	100	80020	500	640132	0	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	0	100	80000	100	0	1	0	1	1	1	5118	0	16	0	0	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	155	0	0	72	25	80108	100	80008	100	80020	500	640132	0	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	0	100	80000	100	0	0	0	1	1	1	5118	0	16	0	0	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	166	0	306	30	25	80108	100	80008	100	80020	500	640132	0	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	0	100	80000	100	0	0	0	1	1	1	5118	0	16	0	0	20036	80000	100	20040	20040	20040	20040	20040
80204	20039	156	0	0	30	25	80108	100	80008	100	80020	500	640132	0	20020	20039	20039	9977	6	9990	80120	200	80032	200	80032	20039	20039	1	1	80201	100	99	100	0	100	80000	100	0	0	0	1	1	1	5118	0	16	0	0	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	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	61	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	b5	c2	cf	d0	d2	d5	map dispatch bubble (d6)	da	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
80024	20050	155	0	40	25	80010	10	80000	10	80000	50	640000	0	0	20020	0	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	0	0	0	0	5020	0	0	16	16	0	8	13	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	155	0	230	25	80010	10	80000	10	80000	50	640812	0	0	20020	0	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	15	0	0	0	5020	0	0	13	16	0	13	8	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	156	0	40	25	80010	10	80000	10	80000	50	640000	0	0	20020	0	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	1	0	0	0	5020	0	0	13	16	0	12	12	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	155	0	40	25	80010	10	80000	10	80000	50	640000	0	0	20020	0	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	33	0	0	0	5020	0	0	10	16	0	12	11	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	155	0	40	25	80010	10	80000	10	80000	50	640000	0	0	20020	0	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	0	0	0	0	5020	0	0	8	16	0	8	6	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	161	0	40	25	80010	10	80000	10	80000	50	640000	0	0	20020	0	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	0	0	0	0	5020	0	0	8	16	0	7	9	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	155	0	40	25	80010	10	80000	10	80000	50	640000	0	0	20020	0	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	0	0	0	0	5020	0	0	10	16	0	8	11	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	155	0	40	25	80010	10	80000	10	80000	50	640000	0	0	20020	0	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	0	0	0	0	5020	0	0	7	16	0	5	7	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	155	0	40	25	80010	10	80000	10	80000	50	640000	0	0	20020	0	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	0	0	0	0	5020	0	0	11	16	0	8	11	20036	80000	10	20040	20040	20040	20040	20040
80024	20039	155	0	40	25	80010	10	80000	10	80000	50	640000	0	0	20020	0	20039	20039	9996	3	10019	80010	20	80000	20	80000	20039	20039	1	1	80021	10	9	10	10	80000	10	0	2	0	0	0	5020	0	0	11	16	0	6	9	20036	80000	10	20040	20040	20040	20040	20040

FRINT64X (vector, 2D)

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