Firestorm: UDOT (by element, 16B) measurements

Code:

  udot v0.4s, v1.16b, v2.4b[1]

Initialization

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

(no loop instructions)

1000 unrolls and 1 iteration

Retires: 1.000

Issues: 1.000

Integer unit issues: 0.000

Load/store unit issues: 0.000

SIMD/FP unit issues: 1.000

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

retire uop (01)	cycle (02)	03	1e	1f	3f	4e	51	schedule uop (52)	schedule simd uop (54)	dispatch simd uop (57)	ldst uops in schedulers (5b)	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	24	0	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	24	0	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	24	0	252	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	6	3	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	1	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	3000	3037	3037	1	1	1001	1000	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038

Test 2: Latency 1->1

Code:

  udot v0.4s, v1.16b, v2.4b[1]

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0037

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

retire uop (01)	cycle (02)	03	1e	3a	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb miss (a1)	l1d cache writeback (a8)	a9	ac	branch cond mispred nonspec (c5)	cf	d0	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	30037	233	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	0	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	0	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	84	29548	25	10100	100	10016	100	10000	500	4277313	0	30018	30037	30084	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	4	0	0	710	0	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	103	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	0	2	16	2	2	29634	10000	100	30038	30038	30085	30038	30038
10204	30131	233	12	0	82	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30084	30085	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	0	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	0	2	16	2	3	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	232	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	0	0	0	710	0	2	16	2	3	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	0	2	16	3	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	241	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	2	0	6	0	710	1	2	16	2	2	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	1	2	16	3	2	29634	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)	eb	ec	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10024	30037	233	0	0	0	0	0	0	111	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	6	640	3	16	3	3	29630	0	0	0	10000	10	30038	30038	30038	30038	30038
10024	30037	232	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	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	3	16	3	3	29630	0	0	0	10000	10	30038	30038	30038	30038	30038
10024	30037	232	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	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	3	16	3	3	29630	0	0	0	10000	10	30038	30038	30038	30038	30038
10024	30037	233	0	0	0	0	0	0	82	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	3	16	3	3	29630	0	1	0	10000	10	30038	30038	30038	30038	30038
10024	30037	233	0	0	0	0	0	0	84	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	1	0	0	640	3	16	3	3	29630	0	0	0	10000	10	30038	30038	30038	30038	30038
10024	30037	232	0	0	0	0	36	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	3	640	3	16	4	3	29630	0	0	0	10000	10	30038	30038	30038	30038	30038
10024	30037	233	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	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	3	16	3	3	29630	0	0	0	10000	10	30038	30038	30038	30038	30038
10024	30037	233	0	0	0	0	0	0	84	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	3	16	3	3	29630	0	0	0	10000	10	30038	30038	30038	30038	30038
10024	30037	232	0	0	0	0	0	0	2007	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	3	16	3	4	29630	0	0	0	10000	10	30038	30038	30038	30038	30038
10024	30037	233	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	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	640	3	16	3	3	29630	0	0	0	10000	10	30038	30038	30038	30038	30038

Test 3: Latency 1->2

Code:

  udot v0.4s, v0.16b, v1.4b[1]

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	19	1e	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d cache writeback (a8)	ac	branch cond mispred nonspec (c5)	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	233	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28272	6	28740	10100	200	10008	200	30024	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	1	1	1	718	0	16	0	0	29646	0	10000	100	30038	30038	30038	30038	30038
10204	30037	232	0	12	103	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28272	7	28740	10100	200	10008	200	30024	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	1	1	1	718	0	16	0	0	29646	0	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	726	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	82	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	0	103	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	3	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	232	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	3	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	233	2	162	103	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

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

retire uop (01)	cycle (02)	03	mmu table walk data (08)	09	l2 tlb miss data (0b)	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	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	233	0	0	0	0	0	0	84	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10609	20	10168	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	233	0	0	0	0	12	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	232	0	0	0	0	12	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	233	0	0	0	0	159	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	232	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	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	232	0	0	0	0	12	0	126	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	232	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	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	232	0	0	0	0	0	0	202	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	233	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	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	3	0	640	2	16	2	2	29668	10000	10	30227	30038	30038	30038	30038
10024	30037	233	0	0	1	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	3	16	2	2	29630	10000	10	30038	30038	30038	30038	30038

Test 4: Latency 1->3

Code:

  udot v0.4s, v1.16b, v0.4b[1]

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0037

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

retire uop (01)	cycle (02)	03	1e	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	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	30037	233	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	1	16	1	1	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	1	16	1	1	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	1	16	1	1	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	241	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	1	16	1	1	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	241	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	1	16	1	1	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	88	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	1	16	1	1	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	1	16	1	1	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	232	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	1	16	1	1	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	710	1	16	1	1	29634	10000	100	30038	30038	30038	30038	30038
10204	30037	233	0	61	29548	25	10100	100	10000	100	10000	500	4277313	30018	30037	30037	28265	3	28745	10100	200	10000	200	30000	30037	30037	1	1	10201	100	99	100	100	10000	100	1	3	710	1	16	1	1	29634	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	3a	3f	4e	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	60	69	6b	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb access (a0)	l1d tlb miss (a1)	ld unit uop (a6)	l1d cache writeback (a8)	a9	ac	c2	cf	l1i cache miss demand (d3)	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	233	0	0	0	0	12	0	0	61	29530	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	0	640	0	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	233	0	0	0	0	0	0	0	103	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	0	640	0	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	232	0	0	0	0	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	0	640	0	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	233	0	0	0	0	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	0	640	0	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	233	0	0	0	0	0	0	0	631	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	0	640	0	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	232	0	0	0	0	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	1	0	0	0	640	0	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	233	0	0	0	0	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	0	640	0	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	233	0	0	0	0	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	0	640	0	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	233	0	0	0	0	0	0	0	103	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	0	640	0	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	233	0	0	0	0	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	0	30037	30037	28287	3	28767	10010	20	10000	20	30000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	0	0	0	640	0	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  udot v0.4s, v8.16b, v9.4b[1]
  movi v1.16b, 0
  udot v1.4s, v8.16b, v9.4b[1]
  movi v2.16b, 0
  udot v2.4s, v8.16b, v9.4b[1]
  movi v3.16b, 0
  udot v3.4s, v8.16b, v9.4b[1]
  movi v4.16b, 0
  udot v4.4s, v8.16b, v9.4b[1]
  movi v5.16b, 0
  udot v5.4s, v8.16b, v9.4b[1]
  movi v6.16b, 0
  udot v6.4s, v8.16b, v9.4b[1]
  movi v7.16b, 0
  udot v7.4s, v8.16b, v9.4b[1]

Initialization

  movi v8.16b, 9
  movi v9.16b, 10

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.2508

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

retire uop (01)	cycle (02)	03	mmu table walk instruction (07)	mmu table walk data (08)	09	l2 tlb miss instruction (0a)	l2 tlb miss data (0b)	18	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 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)	a9	ac	branch mispred nonspec (cb)	cf	d5	map dispatch bubble (d6)	da	dd	fetch restart (de)	e0	? int output thing (e9)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160204	20089	156	0	0	0	0	0	0	0	3	0	0	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	0	10111	2	16	0	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	155	0	0	0	0	0	0	0	0	0	0	419	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	0	10111	1	16	0	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	155	0	0	0	0	0	0	0	0	0	0	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	0	10111	1	16	0	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	156	0	0	0	0	0	0	0	0	0	0	39	25	80100	100	80000	100	80000	500	640000	1	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	0	10111	1	16	0	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	155	0	0	0	0	0	0	0	0	0	0	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	0	10111	1	16	0	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	156	0	0	0	0	0	0	0	27	0	0	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	0	10111	1	16	0	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	155	0	0	0	0	0	0	0	0	0	0	81	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	0	10111	1	16	0	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	155	0	0	0	0	0	0	0	0	0	0	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	0	10111	1	16	0	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	155	0	0	0	0	0	0	0	0	0	0	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	0	10111	1	16	0	1	1	20061	0	160000	100	20065	20065	20065	20065	20065
160204	20064	155	0	0	0	0	0	0	0	0	0	0	39	25	80100	100	80000	100	80000	500	640000	0	20045	20064	20064	3	22	80100	200	80000	200	240000	20064	20064	1	1	160201	100	99	100	100	160000	100	0	0	0	0	0	0	0	10111	1	16	0	1	1	20061	0	160000	100	20065	20065	20065	20065	20065

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	l2 tlb miss instruction (0a)	1e	3f	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	5f	60	61	69	6d	6e	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d cache writeback (a8)	cf	d0	d2	l1i cache miss demand (d3)	d5	map dispatch bubble (d6)	d9	da	db	dd	fetch restart (de)	e0	? int output thing (e9)	ea	ec	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160024	20074	155	0	0	45	27	80012	12	80000	12	80000	62	640000	2	1	10	20032	20053	20051	3	22	80012	20	80000	20	240000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	10025	13	2	1	11	34	2	1	1	7	4	20048	2	20	1	160000	10	20061	20061	20075	20059	20052
160024	20051	155	0	0	80	29	80012	12	80000	12	80000	62	640000	3	1	10	20032	20051	20051	3	22	80012	20	80000	20	240000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	10028	13	5	1	2	34	2	1	1	2	4	20048	2	40	1	160000	10	20052	20061	20052	20052	20052
160024	20051	155	0	0	45	27	80012	12	80000	12	80000	62	640000	2	1	10	20032	20121	20131	3	22	80012	20	80000	20	240000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	10030	13	6	1	4	25	2	1	1	4	6	20057	2	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	156	0	0	45	27	80012	12	80000	12	80000	62	640000	3	1	10	20032	20051	20051	3	22	80012	20	80000	20	240000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	10025	13	5	1	4	25	2	2	1	4	2	20048	2	20	1	160000	10	20052	20052	20061	20052	20052
160024	20060	155	0	0	520	27	80012	12	80000	12	80000	62	640000	3	1	10	20032	20051	20051	3	22	80012	20	80000	20	240000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	10027	13	5	1	4	25	2	1	1	4	2	20048	2	20	1	160000	10	20052	20052	20052	20052	20052
160024	20051	155	0	0	45	27	80012	12	80000	12	80000	62	640000	3	1	10	20039	20051	20051	3	22	80012	20	80000	20	240000	20051	20051	1	1	160021	10	9	10	10	160000	10	0	10027	25	8	5	6	48	8	3	5	6	4	20048	2	20	1	160000	10	20052	20052	20052	20053	20052
160024	20051	156	0	0	45	27	80012	12	80000	12	80000	62	640000	3	1	10	20041	20051	20060	3	22	80012	20	80000	20	240000	20051	20060	1	1	160021	10	9	10	10	160000	10	0	10030	16	5	2	2	34	2	1	1	2	4	20048	2	40	2	160000	10	20061	20052	20061	20052	20052
160024	20051	155	0	0	93	29	80012	12	80000	12	80000	62	640000	3	1	10	20041	20060	20060	3	22	80012	20	80000	20	240000	20060	20060	1	1	160021	10	9	10	10	160000	10	0	10028	16	5	2	4	34	4	2	2	2	4	20057	2	40	2	160000	10	20061	20061	20052	20061	20061
160024	20060	155	0	0	51	29	80012	12	80000	12	80000	62	640000	2	1	10	20032	20060	20051	3	22	80012	20	80000	20	240000	20060	20051	1	1	160021	10	9	10	10	160000	10	0	10028	16	3	10	4	34	4	2	2	2	4	20057	2	20	2	160000	10	20053	20061	20061	20061	20061
160024	20060	155	0	0	51	27	80012	12	80000	12	80000	62	640000	3	1	10	20032	20060	20060	3	22	80012	20	80000	20	240000	20060	20060	1	1	160021	10	9	10	10	160000	10	0	10028	16	6	2	4	34	4	2	2	4	2	20062	2	20	1	160000	10	20061	20052	20052	20052	20061

Test 6: throughput

Count: 16

Code:

  udot v0.4s, v16.16b, v17.4b[1]
  udot v1.4s, v16.16b, v17.4b[1]
  udot v2.4s, v16.16b, v17.4b[1]
  udot v3.4s, v16.16b, v17.4b[1]
  udot v4.4s, v16.16b, v17.4b[1]
  udot v5.4s, v16.16b, v17.4b[1]
  udot v6.4s, v16.16b, v17.4b[1]
  udot v7.4s, v16.16b, v17.4b[1]
  udot v8.4s, v16.16b, v17.4b[1]
  udot v9.4s, v16.16b, v17.4b[1]
  udot v10.4s, v16.16b, v17.4b[1]
  udot v11.4s, v16.16b, v17.4b[1]
  udot v12.4s, v16.16b, v17.4b[1]
  udot v13.4s, v16.16b, v17.4b[1]
  udot v14.4s, v16.16b, v17.4b[1]
  udot v15.4s, v16.16b, v17.4b[1]

Initialization

  movi v16.16b, 17
  movi v17.16b, 18

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.2502

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

retire uop (01)	cycle (02)	03	09	19	1e	37	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	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)	ac	branch mispred nonspec (cb)	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	eb	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160204	40059	310	0	0	0	1	0	41	25	160100	100	160017	100	160000	500	1280000	0	40020	0	40039	40048	19973	3	19998	160100	200	160000	200	480000	40039	40039	1	1	160201	100	99	100	100	160000	100	0	5	0	0	10110	2	16	2	2	40036	0	160000	100	40049	40040	40040	40049	40040
160204	40039	311	0	0	0	0	0	246	25	160100	100	160000	100	160000	500	1280000	0	40029	0	40039	40039	19973	3	19997	160100	200	160000	200	480000	40039	40048	1	1	160201	100	99	100	100	160000	100	0	1	0	0	10110	2	16	2	2	40036	0	160000	100	40049	40049	40049	40049	40284
160204	40039	311	0	0	0	0	0	50	25	160100	100	160000	100	160000	500	1280000	0	40020	0	40048	40048	19973	3	20006	160100	200	160000	200	480000	40039	40048	1	1	160201	100	99	100	100	160000	100	0	1	0	0	10110	2	16	2	2	40045	0	160000	100	40040	40049	40049	40090	40040
160204	40048	310	0	0	0	0	0	41	25	160100	100	160000	100	160000	500	2398999	1	40029	0	40039	40039	19973	3	20006	160100	200	160000	200	480000	40039	40048	1	1	160201	100	99	100	100	160000	100	0	0	0	0	10110	2	16	2	2	40045	0	160000	100	40040	40040	40049	40040	40049
160204	40039	310	0	0	0	17	0	41	25	160100	100	160000	100	160000	500	1280000	1	40029	0	40039	40048	19973	3	19997	160100	200	160000	200	480000	40048	40039	1	1	160201	100	99	100	100	160000	100	0	0	3	0	10110	2	16	2	2	40036	0	160000	100	40040	40049	40040	40040	40040
160204	40039	311	0	0	0	0	0	50	25	160100	100	160017	100	160000	500	1280000	1	40020	0	40048	40048	19973	3	20006	160100	200	160000	200	480000	40039	40040	1	1	160201	100	99	100	100	160000	100	0	2	0	0	10110	2	16	2	2	40036	0	160000	100	40040	40049	40040	40040	40040
160204	40039	310	0	0	0	0	0	50	25	160117	100	160000	100	160000	500	1280000	1	40029	0	40039	40048	19973	3	19997	160100	200	160000	200	480000	40039	40048	1	1	160201	100	99	100	100	160000	100	0	0	0	0	10110	2	16	2	2	40045	0	160000	100	40040	40040	40049	40040	40049
160204	40048	310	0	0	0	0	0	41	25	160101	100	160017	100	160000	500	1280000	1	40020	0	40039	40048	19973	3	19997	160100	200	160000	200	480000	40048	40039	1	1	160201	100	99	100	100	160000	100	0	4	0	0	10110	2	16	2	2	40045	0	160000	100	40040	40049	40040	40040	40040
160204	40039	310	0	0	9	17	0	50	25	160117	100	160000	100	160000	500	1280000	1	40020	0	40048	40048	19973	3	19997	160100	200	160000	200	480000	40048	40039	1	1	160201	100	99	100	100	160000	100	0	0	0	0	10110	2	16	2	2	40045	0	160000	100	40040	40040	40040	40053	40040
160204	40048	310	0	0	0	17	0	41	25	160117	100	160000	100	160000	500	1280000	1	40029	0	40039	40039	19973	3	19997	160100	200	160000	200	480000	40040	40071	1	1	160201	100	99	100	100	160000	100	0	0	0	0	10110	2	16	2	2	40045	0	160000	100	40049	40049	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 data (08)	l2 tlb miss data (0b)	1e	1f	37	3f	51	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch int uop (56)	dispatch simd uop (57)	int uops in schedulers (59)	ldst uops in schedulers (5b)	5f	60	69	6b	6d	6e	map stall dispatch (70)	map rewind (75)	map stall (76)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	82	83	flush restart other nonspec (84)	85	inst all (8c)	inst branch (8d)	inst branch taken (90)	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	l1d tlb access (a0)	ld unit uop (a6)	l1d cache writeback (a8)	a9	ac	branch cond mispred nonspec (c5)	cf	d0	d2	l1i cache miss demand (d3)	d5	map dispatch bubble (d6)	d9	da	db	dd	fetch restart (de)	e0	? int output thing (e9)	ea	eb	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160024	40048	310	0	0	0	0	0	46	25	160010	10	160017	10	160000	50	1319999	1	1	40029	0	40039	40048	19996	3	20028	160010	20	160000	20	480000	40048	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	30	0	10022	3	1	1	11	16	2	1	1	11	11	40036	0	20	6	160000	10	40049	40040	40049	40040	40049
160024	40049	310	0	0	0	0	0	55	25	160138	12	160097	10	160000	50	2398999	1	1	40029	0	40039	40048	19996	3	20028	160010	20	160000	20	480000	40048	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	27	0	10022	3	1	1	10	16	3	399	2	11	12	40045	0	15	6	160000	10	40040	40041	40049	40049	40040
160024	40039	310	0	0	0	0	17	46	25	160010	10	160003	10	160000	50	2438865	1	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	480000	40040	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	6	0	10022	3	1	1	10	16	4	1	2	11	9	40036	0	15	5	160000	10	40050	40040	40049	40040	40040
160024	40039	311	0	0	0	0	0	52	25	160027	10	160002	10	160000	50	2398999	1	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	168	0	10022	3	1	1	9	16	2	1	1	9	8	40036	0	15	6	160000	10	40040	40040	40040	40040	40040
160024	40048	310	0	0	0	0	0	46	25	160010	10	160034	10	160000	50	1280000	1	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	171	0	10022	3	1	1	10	16	2	1	1	10	10	40036	0	15	5	160000	10	40041	40050	40040	40040	40040
160024	40039	310	0	0	0	0	17	55	25	160010	10	160002	10	160000	50	2398999	1	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	102	0	10022	3	1	1	9	16	2	1	1	9	8	40036	0	15	5	160000	10	40049	40040	40040	40040	40041
160024	40039	310	0	0	0	0	0	46	25	160027	10	160015	10	160000	50	2398999	1	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	0	0	10022	3	1	1	8	16	2	829	1	10	11	40045	0	15	5	160000	10	40050	40041	40040	40040	40040
160024	40039	310	0	0	0	0	1	53	25	160010	10	160057	10	160000	50	2398999	1	1	40030	0	40039	40039	19996	3	20019	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	1	0	6	0	10022	6	2	2	9	16	2	1	1	11	9	40036	0	30	10	160000	10	40050	40041	40049	40050	40040
160024	40048	311	0	0	0	0	0	52	25	160010	10	160035	10	160000	50	2398999	1	1	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	480000	40049	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	180	0	10022	3	1	1	9	16	2	1	1	11	10	40036	0	15	7	160000	10	40041	40049	40050	40040	40040
160024	40039	310	0	0	0	0	0	46	25	160011	10	160002	10	160000	50	1280000	1	1	40020	0	40039	40039	19996	3	20029	160010	20	160000	20	480000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	129	0	10024	6	1	2	8	16	2	1	1	11	11	40036	0	30	5	160000	10	40041	40049	40040	40040	40040

UDOT (by element, 16B)

Test 1: uops

1000 unrolls and 1 iteration

Test 2: Latency 1->1

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 3: Latency 1->2

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 4: Latency 1->3

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 5: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 6: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations