Firestorm: FCSEL (scalar, S) measurements

Code:

  fcsel s0, s0, s1, lt

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 2.0037

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

retire uop (01)	cycle (02)	03	l1i tlb fill (04)	1e	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb miss (a1)	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
10204	20037	150	1	0	441	19822	25	10100	100	10000	100	10000	500	1423880	1	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	1	0	710	1	16	1	1	19907	0	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	0	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	0	9	710	1	16	1	1	19907	0	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	0	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	0	0	710	1	16	1	1	19907	0	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	0	232	19822	25	10100	100	10000	100	10000	500	1423880	0	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	0	0	710	1	16	1	1	19907	0	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	21	61	19822	25	10100	100	10000	100	10000	500	1423880	0	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	4	44	15	710	1	16	1	1	19907	0	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	0	536	19822	25	10100	100	10000	100	10000	500	1423880	1	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	0	0	710	1	16	1	1	19907	0	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	1	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	0	147	710	1	16	1	1	19907	0	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	0	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	2	3	710	1	16	1	1	19907	0	10000	100	20038	20038	20038	20038	20038
10204	20037	149	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	1	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	6	93	710	1	16	1	1	19907	0	10000	100	20038	20038	20038	20038	20038
10204	20037	149	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	0	20018	20037	20037	18566	3	18760	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	5	0	710	1	16	1	1	19907	0	10000	100	20038	20038	20038	20038	20038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

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)	ld unit uop (a6)	st unit uop (a7)	l1d cache writeback (a8)	a9	ac	c2	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10024	20037	150	0	0	0	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	0	0	640	2	16	2	2	19907	1	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	0	20018	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	0	0	640	2	16	2	2	19907	0	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	0	20018	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	1	0	3	0	640	2	16	2	2	19907	0	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	0	0	640	2	16	2	2	19907	0	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	0	0	0	89	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	0	0	640	2	16	2	2	19907	0	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	1	0	0	0	640	2	16	2	2	19907	0	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	0	0	640	2	16	2	2	19907	0	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	0	0	640	2	16	2	2	19907	0	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	0	0	640	2	16	2	2	19907	0	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	0	0	640	2	16	2	2	19907	0	10000	10	20038	20038	20038	20038	20038

Test 2: uops

Code:

  fcsel s0, s1, s0, lt

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)	9f	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	2037	16	0	61	1822	25	1000	1000	1000	136880	1	2018	2037	2037	1716	3	1895	1000	1000	3000	2037	2037	1	1	1001	1000	1000	0	0	73	1	16	1	1	1907	1000	2038	2038	2038	2038	2038
1004	2037	15	0	61	1822	25	1000	1000	1000	136880	1	2018	2037	2037	1716	3	1895	1000	1000	3000	2037	2037	1	1	1001	1000	1000	0	0	73	1	16	1	1	1907	1000	2038	2038	2038	2038	2038
1004	2037	15	0	61	1822	25	1000	1000	1000	136880	1	2018	2037	2037	1716	3	1895	1000	1000	3000	2037	2037	1	1	1001	1000	1000	0	0	73	1	16	1	1	1907	1000	2038	2038	2038	2038	2038
1004	2037	15	0	61	1822	25	1000	1000	1000	136880	1	2018	2037	2037	1716	3	1895	1000	1000	3000	2037	2037	1	1	1001	1000	1000	7	0	73	1	16	1	1	1907	1000	2038	2038	2038	2038	2038
1004	2037	15	0	61	1822	25	1000	1000	1000	136880	1	2018	2037	2037	1716	3	1895	1000	1000	3000	2037	2037	1	1	1001	1000	1000	0	0	73	1	16	1	1	1907	1000	2038	2038	2038	2038	2038
1004	2037	16	12	61	1822	25	1000	1000	1000	136880	1	2018	2037	2037	1716	3	1895	1000	1000	3000	2037	2037	1	1	1001	1000	1000	0	0	73	1	16	1	1	1907	1000	2038	2038	2038	2038	2038
1004	2037	15	0	61	1822	25	1000	1000	1000	136880	1	2018	2037	2037	1716	3	1895	1000	1000	3000	2037	2037	1	1	1001	1000	1000	0	0	73	1	16	1	1	1907	1000	2038	2038	2038	2038	2038
1004	2037	15	0	61	1822	25	1000	1000	1000	136880	1	2018	2037	2037	1716	3	1895	1000	1000	3000	2037	2037	1	1	1001	1000	1000	0	0	73	1	16	1	1	1907	1000	2038	2038	2038	2038	2038
1004	2037	15	0	61	1822	25	1000	1000	1000	136880	1	2018	2037	2037	1716	3	1895	1000	1000	3000	2037	2037	1	1	1001	1000	1000	0	0	73	1	16	1	1	1907	1000	2038	2038	2038	2038	2038
1004	2037	16	0	61	1822	25	1000	1000	1000	136880	0	2018	2037	2037	1716	3	1895	1000	1000	3000	2037	2037	1	1	1001	1000	1000	0	9	73	1	16	1	1	1907	1000	2038	2038	2038	2038	2038

Test 3: Latency 1->3

Code:

  fcsel s0, s1, s0, lt

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 2.0037

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)	60	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d cache writeback (a8)	ac	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	20037	150	0	0	726	19822	25	10100	100	10000	100	10000	500	1423880	1	20018	20037	20037	18566	3	18745	10198	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	2	0	710	1	16	1	1	19907	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	0	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	177	710	1	16	1	1	19907	10000	100	20038	20038	20038	20038	20038
10204	20037	149	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	0	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	111	710	1	16	1	1	19907	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	0	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	2	0	710	1	16	1	1	19907	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	1	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	0	710	1	16	1	1	19907	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	0	536	19822	25	10100	100	10000	100	10000	500	1423880	1	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	42	0	710	1	16	1	1	19907	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	1	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	0	710	1	16	1	1	19907	10000	100	20038	20038	20038	20038	20038
10204	20037	149	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	1	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	0	710	1	16	1	1	19907	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	1	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	0	710	1	16	0	1	19907	10000	100	20038	20038	20038	20038	20038
10204	20037	150	0	0	61	19822	25	10100	100	10000	100	10000	500	1423880	1	20018	20037	20037	18566	3	18745	10100	200	10000	200	30000	20037	20037	1	1	10201	100	99	100	100	10000	10100	0	0	710	1	16	1	1	19907	10000	100	20038	20038	20038	20038	20038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

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

retire uop (01)	cycle (02)	03	mmu table walk data (08)	09	l2 tlb miss data (0b)	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	6b	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb access (a0)	l1d cache writeback (a8)	a9	ac	cd	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10024	20037	150	0	0	0	0	103	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	0	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	1	0	3	0	640	2	16	2	2	19907	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	166	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	0	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	640	2	16	2	2	19907	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	0	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	640	2	16	2	2	19907	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	540	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	0	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	640	2	16	2	2	19907	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	0	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	1	0	0	0	640	2	16	2	2	19907	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	0	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	640	2	16	2	2	19907	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	0	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	640	2	16	2	2	19907	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	0	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	1	640	2	16	2	2	19907	10000	10	20038	20227	20038	20038	20038
10024	20037	150	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	0	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	640	2	16	2	2	19907	10000	10	20038	20038	20038	20038	20038
10024	20037	150	0	0	0	0	61	19822	25	10010	10	10000	10	10000	50	1423880	1	20018	0	20037	20037	18588	3	18767	10010	20	10000	20	30000	20037	20037	1	1	10021	10	9	10	10	10000	10010	0	0	0	0	0	640	2	16	2	2	19907	10000	10	20038	20038	20038	20038	20038

Test 4: Latency 1->4

Chain cycles: 2

Code:

  fcsel s0, s1, s2, lt
  fcmp d0, d3

Initialization

  movi v0.16b, 1
  movi v1.16b, 2
  movi v2.16b, 3
  movi v3.16b, 4

(non-fused SUB/CBNZ loop)

100 unrolls and 100 iterations

Result (median cycles for code, minus 2 chain cycles): 2.0037

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)	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 int alu (97)	inst simd alu (9a)	9f	l1d tlb miss (a1)	ld unit uop (a6)	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
20204	40037	300	0	0	0	0	0	0	536	39822	25	20200	200	20000	200	20000	1100	2854371	40018	40037	40037	37316	3	37495	20200	200	20000	200	50000	40037	40037	1	1	20201	100	99	100	20000	10000	0	0	0	0	1310	4	16	2	2	39909	100	10000	100	40038	40038	40038	40038	40038
20204	40037	300	0	1	0	0	0	0	61	39822	100	20200	200	20000	200	20000	1105	2853880	40018	40037	40037	37316	3	37495	20200	200	20000	204	50000	40037	40037	1	1	20201	100	99	100	20000	10000	2	0	0	0	1310	2	16	2	2	39909	100	10000	100	40038	40038	40038	40038	40038
20204	40037	300	0	0	0	0	0	0	61	39822	25	20200	200	20000	200	20000	1100	2853880	40018	40037	40037	37316	3	37495	20200	200	20000	200	50000	40037	40225	1	1	20201	100	99	100	20000	10000	0	3	0	0	1310	2	16	2	2	39909	100	10000	100	40038	40038	40038	40038	40038
20204	40037	300	0	0	0	0	0	0	61	39822	25	20200	200	20000	200	20000	1100	2853880	40018	40037	40037	37316	3	37495	20200	200	20000	200	50000	40037	40037	1	1	20201	100	99	100	20000	10000	0	2	3	0	1310	2	16	2	2	39909	100	10000	100	40038	40038	40038	40038	40038
20204	40037	300	0	1	0	0	0	0	103	39822	25	20200	200	20000	200	20000	1100	2853880	40018	40037	40037	37316	18	37495	20200	200	20000	200	50000	40037	40037	1	1	20201	100	99	100	20000	10000	0	0	0	0	1310	2	16	2	2	39909	100	10000	100	40038	40038	40038	40038	40038
20204	40037	299	0	0	0	0	0	0	61	39822	25	20200	200	20000	200	20000	1100	2853880	40018	40037	40037	37316	3	37495	20200	200	20000	200	50000	40037	40037	1	1	20201	100	99	100	20000	10000	0	0	0	0	1310	2	16	2	2	39909	100	10000	100	40038	40038	40038	40038	40038
20204	40037	299	0	0	0	0	0	0	61	39822	25	20200	200	20000	200	20000	1100	2853880	40018	40037	40037	37316	3	37495	20200	200	20000	200	50000	40037	40037	1	1	20201	100	99	100	20000	10000	0	0	0	0	1310	2	16	2	2	39909	100	10000	100	40038	40038	40038	40038	40038
20204	40037	300	0	0	0	0	0	0	441	39822	25	20200	200	20000	200	20000	1100	2853880	40018	40037	40037	37316	3	37495	20200	200	20000	200	50000	40037	40037	1	1	20201	100	99	100	20000	10000	0	0	0	0	1310	2	16	2	2	39909	100	10000	100	40038	40038	40038	40038	40038
20204	40037	300	0	0	0	0	0	0	61	39822	25	20200	200	20000	200	20000	1100	2853880	40018	40037	40037	37316	3	37495	20200	200	20000	200	50000	40037	40037	1	1	20201	100	99	100	20000	10000	0	0	0	0	1310	2	16	2	2	39909	100	10000	100	40038	40038	40038	40038	40038
20204	40037	300	0	0	0	0	0	0	94	39822	25	20200	200	20000	200	20000	1100	2853880	40018	40037	40037	37316	3	37495	20200	200	20000	200	50000	40037	40037	1	1	20201	100	99	100	20000	10000	0	0	0	0	1310	2	16	2	2	39909	100	10000	100	40038	40038	40038	40038	40038

1000 unrolls and 10 iterations

Result (median cycles for code, minus 2 chain cycles): 2.0037

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)	l2 tlb miss instruction (0a)	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 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	c2	cd	cf	d5	map dispatch bubble (d6)	d8	dd	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
20024	40037	299	0	0	1	0	0	0	9	0	0	89	39822	25	20020	20	20012	20	20000	110	2853880	1	40018	40037	40037	37338	3	37517	20020	20	20000	20	50000	40037	40037	1	1	20021	10	9	10	20000	10000	0	0	0	0	0	0	3	0	0	1270	9	16	0	10	9	39909	10	10000	10	40038	40038	40038	40038	40038
20024	40037	300	0	0	0	0	0	0	12	0	0	61	39822	25	20020	20	20000	20	20000	109	2854371	1	40018	40037	40037	37338	3	37517	20020	20	20000	20	50000	40037	40037	1	1	20021	10	9	10	20000	10000	0	0	0	0	0	0	0	0	0	1270	11	41	208	10	11	39909	10	10000	10	40323	40038	40323	40038	40132
20024	40037	300	0	0	0	0	0	0	144	264	0	61	39822	25	20020	20	20000	20	20000	110	2853880	1	40018	40274	40037	37338	30	37517	20020	20	20000	20	51641	40037	40368	1	1	20021	10	9	10	20000	10000	0	0	0	0	0	0	0	0	0	1386	14	16	0	9	10	39909	11	10000	10	40038	40278	40038	40275	40038
20024	40037	302	0	0	0	0	0	0	0	0	0	61	39822	25	20020	20	20000	20	20000	110	2853880	1	40018	40226	40037	37338	25	37517	20020	20	20000	20	50000	40037	40037	1	1	20021	10	9	10	20000	10000	2	0	0	0	0	4	5393	4	0	1322	9	16	0	11	9	39909	10	10000	10	40038	40038	40038	40038	40038
20024	40037	300	0	0	0	0	0	0	0	0	0	61	39822	25	20096	20	20000	20	20240	110	2854371	1	40018	40132	40037	37338	7	37517	20020	20	20000	20	50000	40037	40037	1	1	20021	10	9	10	20000	10000	6	0	0	0	0	0	0	0	0	1270	8	16	0	11	11	40052	10	10000	10	40322	40038	40038	40369	40038
20024	40037	300	0	0	0	0	0	0	0	0	0	1790	39822	25	20104	20	20036	20	20558	110	2853880	1	40162	40037	40037	37338	3	37517	20020	20	20000	20	50227	40037	40037	1	1	20021	10	9	10	20000	10000	0	0	0	0	0	0	0	0	0	1270	10	16	0	11	10	39909	10	10000	10	40038	40038	40038	40038	40038
20024	40037	300	0	0	0	1	3	1	405	0	0	61	39822	25	20020	20	20000	20	20000	110	2853880	0	40018	40318	40037	37413	30	37517	20580	20	20000	20	50000	40180	40132	1	1	20021	10	9	10	20000	10000	0	0	2	0	0	0	0	0	0	1270	12	16	0	8	10	39945	10	10000	10	40038	40038	40038	40038	40038
20024	40037	300	0	1	4	0	0	0	132	0	0	61	39714	62	20020	20	20000	20	20000	110	2857317	0	40018	40323	40037	37420	20	37545	20420	20	20000	20	50000	40037	40037	1	1	20021	10	9	10	20000	10000	0	0	0	0	0	0	0	0	0	1270	12	16	0	10	8	39909	10	10000	10	40038	40038	40085	40038	40038
20024	40037	300	0	1	0	0	0	2	0	352	0	1425	39822	25	20020	20	20072	20	20560	110	2853880	1	40270	40037	40321	37338	3	37517	20020	20	20000	20	50708	40131	40037	1	1	20021	10	9	10	20000	10000	0	0	0	0	0	2	0	0	0	1270	8	16	0	10	11	39909	10	10000	10	40321	40038	40038	40084	40038
20024	40037	300	0	0	1	0	0	0	0	0	0	61	39822	25	20020	20	20000	20	20000	108	2853880	1	40018	40037	40037	37350	3	37517	20180	20	20000	20	50000	40037	40037	1	1	20021	10	9	10	20000	10000	0	0	0	0	1	0	0	0	0	1270	9	16	0	8	10	39909	10	10000	10	40038	40038	40038	40038	40038

Test 5: throughput

Count: 8

Code:

  fcsel s0, s8, s9, lt
  fcsel s1, s8, s9, lt
  fcsel s2, s8, s9, lt
  fcsel s3, s8, s9, lt
  fcsel s4, s8, s9, lt
  fcsel s5, s8, s9, lt
  fcsel s6, s8, s9, lt
  fcsel s7, s8, s9, lt

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): 0.5005

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 tlb access (a0)	l1d tlb miss (a1)	ld unit uop (a6)	l1d cache writeback (a8)	ac	cd	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
80204	40050	300	0	43	25	80100	100	80000	100	80000	500	640000	1	40019	40038	40038	29970	3	29996	80100	200	80000	200	240000	40038	40038	1	1	80201	100	99	100	100	80000	80100	0	0	0	0	0	0	5110	2	16	1	1	40035	80000	100	40039	40039	40039	40039	40039
80204	40038	299	0	43	25	80100	100	80000	100	80000	500	640000	0	40019	40038	40038	29970	3	29996	80100	200	80000	200	240000	40038	40038	1	1	80201	100	99	100	100	80000	80100	0	0	0	0	0	0	5110	1	16	1	1	40035	80000	100	40039	40039	40039	40039	40039
80204	40038	300	0	43	25	80100	100	80000	100	80000	500	640000	0	40169	40038	40038	29970	3	29996	80100	200	80000	200	240000	40038	40038	1	1	80201	100	99	100	100	80000	80100	0	0	0	0	0	0	5110	1	16	1	1	40035	80000	100	40039	40039	40039	40039	40039
80204	40038	299	0	43	25	80100	100	80000	100	80000	500	640000	0	40019	40038	40038	29970	3	29996	80100	200	80000	200	240000	40038	40038	1	1	80201	100	99	100	100	80000	80100	0	0	0	0	0	0	5110	1	16	1	1	40035	80000	100	40039	40039	40039	40039	40039
80204	40038	300	0	138	25	80100	100	80000	100	80000	500	640000	1	40019	40038	40038	29970	3	29996	80100	200	80000	200	240000	40038	40038	1	1	80201	100	99	100	100	80000	80100	0	0	0	0	0	0	5110	1	16	1	1	40035	80000	100	40039	40039	40039	40039	40039
80204	40038	300	0	43	25	80100	100	80000	100	80000	500	640000	0	40019	40038	40038	29970	3	29996	80100	200	80000	200	240000	40038	40038	1	1	80201	100	99	100	100	80000	80100	0	0	0	0	0	0	5110	1	16	1	1	40035	80000	100	40039	40039	40039	40039	40039
80204	40038	300	0	43	25	80100	100	80000	100	80000	500	640000	0	40019	40038	40038	29970	3	29996	80100	200	80000	200	240000	40038	40038	1	1	80201	100	99	100	100	80000	80100	0	0	0	0	0	0	5110	1	16	1	1	40035	80000	100	40039	40039	40039	40039	40039
80204	40038	300	0	43	25	80100	100	80000	100	80000	500	640000	0	40019	40038	40038	29970	3	29996	80100	200	80000	200	240000	40038	40038	1	1	80201	100	99	100	100	80000	80100	0	0	0	0	0	0	5110	1	16	1	1	40035	80000	100	40039	40039	40039	40039	40039
80204	40038	300	0	43	25	80100	100	80000	100	80000	500	640000	0	40019	40038	40038	29970	3	29996	80100	200	80000	200	240000	40038	40038	1	1	80201	100	99	100	100	80000	80100	0	0	0	0	0	0	5110	1	16	1	1	40035	80000	100	40039	40039	40039	40039	40039
80204	40038	300	0	43	25	80100	100	80000	100	80000	500	640000	0	40019	40038	40038	29970	3	29996	80100	200	80000	200	240000	40038	40038	2	1	80201	100	99	100	100	80000	80100	0	0	0	0	0	0	5110	1	16	1	1	40035	80000	100	40039	40039	40039	40039	40039

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.5005

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)	19	1e	1f	3a	3f	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	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	d2	l1i cache miss demand (d3)	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
80024	40044	300	0	0	0	0	0	0	43	25	80010	10	80000	10	80000	50	640000	1	40019	40038	40038	29992	3	30018	80010	20	80000	20	240000	40038	40038	1	1	80021	10	9	10	10	80000	80010	0	0	5020	0	0	18	16	12	9	40035	80000	10	40039	40039	40039	40039	40039
80024	40038	300	0	0	0	0	0	0	43	25	80010	10	80000	10	80000	50	640873	1	40019	40038	40038	29992	3	30018	80010	20	80000	20	240000	40038	40038	1	1	80021	10	9	10	10	80000	80010	6	0	5020	0	0	9	16	13	14	40035	80000	10	40039	40039	40039	40039	40039
80024	40038	299	0	0	0	0	0	0	43	25	80010	10	80000	10	80000	50	640000	1	40019	40038	40038	29992	3	30018	80010	20	80000	20	240000	40038	40038	1	1	80021	10	9	10	10	80000	80010	0	0	5020	0	0	13	16	9	12	40035	80000	10	40039	40039	40039	40039	40039
80024	40038	300	0	0	0	0	0	0	43	25	80010	10	80000	10	80000	50	640000	1	40019	40038	40038	29992	3	30018	80010	20	80000	20	240000	40038	40038	1	1	80021	10	9	10	10	80000	80010	0	0	5020	0	0	8	16	12	9	40035	80000	10	40039	40039	40039	40039	40039
80024	40038	300	0	0	0	0	0	0	43	25	80010	10	80000	10	80000	50	640000	1	40019	40038	40038	29992	3	30018	80010	20	80000	20	240000	40038	40038	1	1	80021	10	9	10	10	80000	80010	0	0	5020	0	0	13	16	13	13	40035	80000	10	40039	40039	40039	40039	40039
80024	40038	299	0	0	0	0	0	0	43	25	80010	10	80000	10	80000	50	640000	1	40019	40038	40038	29992	3	30018	80010	20	80000	20	240000	40038	40038	1	1	80021	10	9	10	10	80000	80010	0	0	5020	0	0	13	16	12	8	40035	80000	10	40039	40039	40039	40039	40039
80024	40038	300	0	0	0	0	0	0	43	25	80010	10	80000	10	80000	50	640000	1	40019	40038	40038	29992	3	30018	80010	20	80000	20	240000	40038	40038	1	1	80021	10	9	10	10	80000	80010	0	0	5020	0	2	13	16	12	9	40035	80000	10	40039	40039	40039	40039	40138
80024	40135	300	0	1	0	312	192	0	43	25	80010	10	80000	10	80000	56	640000	1	40094	40133	40136	30026	3	30018	80010	20	80000	20	240000	40038	40038	1	1	80021	10	9	10	10	80000	80010	0	0	5020	0	0	9	16	18	8	40035	80000	10	40039	40039	40039	40039	40039
80024	40038	300	0	0	1	0	0	0	708	25	80010	10	80000	10	80000	60	640390	1	40019	40038	40038	29992	3	30018	80010	20	80000	20	240000	40038	40038	1	1	80021	10	9	10	10	80000	80010	0	0	5020	0	0	13	16	13	14	40035	80000	10	40039	40039	40039	40039	40039
80024	40038	300	0	0	0	0	0	0	43	25	80010	10	80000	10	80000	50	640000	1	40019	40038	40038	29992	3	30018	80010	20	80000	20	240000	40038	40038	1	1	80021	10	9	10	10	80000	80010	0	0	5020	0	0	13	16	13	9	40035	80000	10	40039	40039	40039	40039	40039

FCSEL (scalar, S)

Test 1: Latency 1->2

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 2: uops

1000 unrolls and 1 iteration

Test 3: Latency 1->3

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 4: Latency 1->4

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 5: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations