Icestorm: ORR (vector, immediate, 8H) measurements

Code:

  orr v0.8h, #1

Initialization

  movi v0.16b, 1

(no loop instructions)

1000 unrolls and 1 iteration

Retires: 1.000

Issues: 1.000

Integer unit issues: 0.001

Load/store unit issues: 0.000

SIMD/FP unit issues: 1.000

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

retire uop (01)	cycle (02)	schedule uop (52)	schedule int uop (53)	schedule simd uop (54)	dispatch simd uop (57)	ldst uops in schedulers (5b)	dispatch uop (78)	map simd uop (7e)	map simd uop inputs (81)	? int output thing (e9)	? simd retires (ee)
1004	2033	1001	1	1000	1000	50248	1000	1000	1000	1	1000
1004	2033	1001	1	1000	1000	50248	1000	1000	1000	1	1000
1004	2033	1001	1	1000	1000	50248	1000	1000	1000	1	1000
1004	2033	1001	1	1000	1000	50248	1000	1000	1000	1	1000
1004	2033	1001	1	1000	1000	50248	1000	1000	1000	1	1000
1004	2033	1001	1	1000	1000	50248	1000	1000	1000	1	1000
1004	2033	1001	1	1000	1000	50248	1000	1000	1000	1	1000
1004	2033	1001	1	1000	1000	50248	1000	1000	1000	1	1000
1004	2033	1001	1	1000	1000	50248	1000	1000	1000	1	1000
1004	2033	1001	1	1000	1000	50248	1000	1000	1000	1	1000

Test 2: Latency 1->1

Code:

  orr v0.8h, #1

Initialization

  movi v0.16b, 1

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 2.0033

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

retire uop (01)	cycle (02)	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)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	? int output thing (e9)	? simd retires (ee)	? int retires (ef)
10204	20033	10101	101	10000	100	10000	300	509248	10100	200	10004	200	10004	1	10000	100
10204	20033	10101	101	10000	100	10000	300	509248	10100	200	10004	200	10004	1	10000	100
10204	20033	10101	101	10000	100	10000	300	509248	10100	200	10004	200	10004	1	10000	100
10204	20033	10101	101	10000	100	10000	300	509248	10100	200	10004	200	10004	1	10000	100
10204	20033	10101	101	10000	100	10000	300	511376	10244	200	10168	200	10004	1	10000	100
10204	20033	10101	101	10000	100	10000	300	509248	10100	200	10004	200	10004	1	10000	100
10204	20033	10101	101	10000	100	10000	300	509248	10100	200	10004	200	10004	1	10000	100
10204	20033	10101	101	10000	100	10000	300	509248	10100	200	10004	200	10004	1	10000	100
10204	20033	10101	101	10000	100	10000	300	509248	10100	200	10004	200	10004	1	10000	100
10204	20033	10101	101	10000	100	10000	300	509248	10100	200	10004	200	10004	1	10000	100

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0033

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

retire uop (01)	cycle (02)	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)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	? int output thing (e9)	? simd retires (ee)	? int retires (ef)
10024	20033	10021	21	10000	20	10000	70	509248	10020	20	10000	20	10000	11	10000	10
10024	20033	10021	21	10000	20	10000	70	509248	10020	20	10000	20	10000	11	10000	10
10024	20033	10021	21	10000	20	10000	70	509248	10020	20	10000	20	10000	11	10000	10
10024	20033	10021	21	10000	20	10000	70	509248	10020	20	10000	20	10000	11	10000	10
10024	20033	10021	21	10000	20	10000	70	509248	10020	20	10000	20	10000	11	10000	10
10024	20033	10021	21	10000	20	10000	70	509248	10020	20	10000	20	10000	11	10000	10
10024	20033	10021	21	10000	20	10000	70	509248	10020	20	10000	20	10000	11	10000	10
10024	20033	10021	21	10000	20	10000	70	509248	10020	20	10000	20	10000	11	10000	10
10024	20033	10021	21	10000	20	10000	70	509248	10020	20	10000	20	10000	11	10000	10
10024	20033	10021	21	10000	20	10000	70	509248	10020	20	10000	20	10000	11	10000	10

Test 3: throughput

Count: 8

Code:

  movi v0.16b, 0
  orr v0.8h, #1
  movi v1.16b, 0
  orr v1.8h, #1
  movi v2.16b, 0
  orr v2.8h, #1
  movi v3.16b, 0
  orr v3.8h, #1
  movi v4.16b, 0
  orr v4.8h, #1
  movi v5.16b, 0
  orr v5.8h, #1
  movi v6.16b, 0
  orr v6.8h, #1
  movi v7.16b, 0
  orr v7.8h, #1

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.5011

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

retire uop (01)	cycle (02)	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)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	? int output thing (e9)	? simd retires (ee)	? int retires (ef)
160204	40476	80109	101	80008	100	80012	300	320056	80113	200	80013	200	80013	1	160000	100
160204	40110	80110	101	80009	100	80013	300	320052	80112	200	80012	200	80012	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	80012	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	80012	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	80012	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	80012	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	80012	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320056	80113	200	80013	200	80013	1	160000	100
160204	40103	80110	101	80009	100	80013	300	320052	80112	200	80012	200	80012	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	80012	1	160000	100

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.5055

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

retire uop (01)	cycle (02)	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)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	? int output thing (e9)	? simd retires (ee)	? int retires (ef)
160024	44113	80019	11	80008	10	80012	30	320000	80010	20	80000	20	80000	1	160000	10
160024	41380	80011	11	80000	10	80000	30	320000	80010	20	80000	20	80000	1	160000	10
160024	40419	80011	11	80000	10	80000	30	320000	80010	20	80000	20	80000	1	160000	10
160024	42039	80011	11	80000	10	80000	30	320000	80010	20	80000	20	80000	1	160000	10
160024	40400	80011	11	80000	10	80000	30	320000	80010	20	80000	20	80012	1	160000	10
160024	40518	80020	11	80009	10	80013	30	320056	80023	20	80013	20	80000	1	160000	10
160024	40476	80011	11	80000	10	80000	30	320000	80010	20	80000	20	80000	1	160000	10
160024	40441	80011	11	80000	10	80000	30	320000	80010	20	80000	20	80000	1	160000	10
160024	40434	80011	11	80000	10	80000	30	320000	80010	20	80000	20	80000	1	160000	10
160024	40449	80011	11	80000	10	80000	30	320000	80010	20	80000	20	80000	1	160000	10

Test 4: throughput

Count: 16

Code:

  orr v0.8h, #1
  orr v1.8h, #1
  orr v2.8h, #1
  orr v3.8h, #1
  orr v4.8h, #1
  orr v5.8h, #1
  orr v6.8h, #1
  orr v7.8h, #1
  orr v8.8h, #1
  orr v9.8h, #1
  orr v10.8h, #1
  orr v11.8h, #1
  orr v12.8h, #1
  orr v13.8h, #1
  orr v14.8h, #1
  orr v15.8h, #1

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.5002

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

retire uop (01)	cycle (02)	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)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map ldst uop inputs (80)	map simd uop inputs (81)	? int output thing (e9)	? ldst retires (ed)	? simd retires (ee)	? int retires (ef)
160204	80038	160107	101	160006	100	160010	300	640432	160207	200	160111	200	0	160062	1	0	160000	100
160204	80043	160107	101	160006	100	160010	300	640036	160108	200	160012	200	0	160012	1	0	160000	100
160204	80034	160105	101	160004	100	160008	300	640036	160108	200	160012	200	0	160012	1	0	160000	100
160204	80034	160105	101	160004	100	160008	300	640036	160108	200	160012	200	0	160012	1	0	160000	100
160204	80034	160105	101	160004	100	160008	300	640036	160108	200	160012	200	0	160012	1	0	160000	100
160204	80034	160105	101	160004	100	160008	300	640036	160108	200	160012	200	0	160012	1	0	160000	100
160204	80034	160105	101	160004	100	160008	300	640044	160110	200	160014	200	0	160012	1	0	160000	100
160204	80034	160105	101	160004	100	160008	300	640036	160108	200	160012	200	0	160060	1	0	160000	100
160204	80034	160105	101	160004	100	160008	300	640036	160108	200	160012	200	0	160012	1	0	160000	100
160204	80034	160105	101	160004	100	160008	300	640036	160108	200	160012	200	0	160012	1	0	160000	100

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.5002

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

retire uop (01)	cycle (02)	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)	dispatch uop (78)	map int uop (7c)	map simd uop (7e)	map int uop inputs (7f)	map simd uop inputs (81)	? int output thing (e9)	? simd retires (ee)	? int retires (ef)
160024	80179	160017	11	160006	10	160010	30	640048	160021	20	160014	20	160000	1	160000	10
160025	80098	160052	11	160041	10	160053	30	640000	160010	20	160000	20	160000	1	160000	10
160024	80034	160011	11	160000	10	160000	30	640044	160020	20	160013	20	160000	1	160000	10
160024	80034	160011	11	160000	10	160000	30	640000	160010	20	160000	20	160000	1	160000	10
160024	80036	160011	11	160000	10	160000	30	640000	160010	20	160000	20	160000	1	160000	10
160024	80034	160011	11	160000	10	160000	30	640000	160010	20	160000	20	160000	1	160000	10
160024	80034	160011	11	160000	10	160000	30	640000	160010	20	160000	20	160000	1	160000	10
160024	80034	160011	11	160000	10	160000	30	640000	160010	20	160000	20	160000	1	160000	10
160024	80034	160011	11	160000	10	160000	30	640208	160064	20	160060	20	160000	1	160000	10
160024	80034	160011	11	160000	10	160000	30	640000	160010	20	160000	20	160000	1	160000	10

ORR (vector, immediate, 8H)

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: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 4: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations