Icestorm: UADALP (vector, 4S) measurements

Code:

  uadalp v0.4s, v1.8h

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: 0.001

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	3033	1001	1	1000	1000	75905	1000	1000	2000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	2000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	2000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	2000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	2000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	2000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	2000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	2000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	2000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	2000	1	1000

Test 2: Latency 1->1

Code:

  uadalp v0.4s, v1.8h

Initialization

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

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0033

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 ldst uop inputs (80)	map simd uop inputs (81)	? int output thing (e9)	? ldst retires (ed)	? simd retires (ee)	? int retires (ef)
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	0	20012	1	0	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	0	20008	1	0	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	0	20008	1	0	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	0	20008	1	0	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	0	20008	1	0	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	0	20008	1	0	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	0	20008	1	0	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	0	20008	1	0	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	0	20008	1	0	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	0	20008	1	0	10000	100

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0033

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	30033	10021	21	10000	20	10000	70	768905	10020	20	10006	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	68	769247	10051	20	10046	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20090	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10
10025	30066	10027	21	10006	20	10031	70	768905	10020	20	10000	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10

Test 3: Latency 1->2

Code:

  uadalp v0.4s, v0.8h

Initialization

  movi v0.16b, 1

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0033

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	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	20008	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	20008	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	20008	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	20008	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	20008	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	20008	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	20008	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	202	20090	2	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	20008	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	20008	1	10000	100

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0033

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	30033	10021	21	10000	20	10000	70	768905	10020	20	10004	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	20000	11	10000	10

Test 4: throughput

Count: 8

Code:

  movi v0.16b, 0
  uadalp v0.4s, v8.8h
  movi v1.16b, 0
  uadalp v1.4s, v8.8h
  movi v2.16b, 0
  uadalp v2.4s, v8.8h
  movi v3.16b, 0
  uadalp v3.4s, v8.8h
  movi v4.16b, 0
  uadalp v4.4s, v8.8h
  movi v5.16b, 0
  uadalp v5.4s, v8.8h
  movi v6.16b, 0
  uadalp v6.4s, v8.8h
  movi v7.16b, 0
  uadalp v7.4s, v8.8h

Initialization

  movi v8.16b, 9

(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	40525	80110	101	80009	100	80013	300	320052	80112	200	80012	200	160026	1	160000	100
160204	40122	80109	101	80008	100	80012	300	320056	80113	200	80013	200	160024	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	160094	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320056	80113	200	80013	200	160024	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320056	80113	200	80013	200	160026	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320056	80113	200	80013	200	160024	1	160000	100
160204	40096	80110	101	80009	100	80013	300	320056	80113	200	80013	200	160024	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320056	80113	200	80013	200	160024	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320056	80113	200	80013	200	160024	1	160000	100
160205	40133	80146	101	80045	100	80049	300	320052	80112	200	80012	200	160024	1	160000	100

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.5054

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	43906	80020	11	80009	10	80013	30	320052	80022	20	80012	20	160000	1	160000	10
160024	41180	80011	11	80000	10	80000	30	320000	80010	20	80000	20	160000	1	160000	10
160024	40387	80011	11	80000	10	80000	30	320000	80010	20	80000	20	160000	1	160000	10
160024	40428	80011	11	80000	10	80000	30	320000	80010	20	80000	20	160000	1	160000	10
160024	40419	80011	11	80000	10	80000	30	320000	80010	20	80000	20	160000	1	160000	10
160024	40397	80011	11	80000	10	80000	30	320000	80010	20	80000	20	160000	1	160000	10
160024	40435	80011	11	80000	10	80000	30	320000	80010	20	80000	20	160000	1	160000	10
160024	40444	80011	11	80000	10	80000	30	320000	80010	20	80000	20	160000	1	160000	10
160024	40408	80011	11	80000	10	80000	30	320000	80010	20	80000	20	160000	1	160000	10
160024	40393	80011	11	80000	10	80000	30	320000	80010	20	80000	20	160000	1	160000	10

Test 5: throughput

Count: 16

Code:

  uadalp v0.4s, v16.8h
  uadalp v1.4s, v16.8h
  uadalp v2.4s, v16.8h
  uadalp v3.4s, v16.8h
  uadalp v4.4s, v16.8h
  uadalp v5.4s, v16.8h
  uadalp v6.4s, v16.8h
  uadalp v7.4s, v16.8h
  uadalp v8.4s, v16.8h
  uadalp v9.4s, v16.8h
  uadalp v10.4s, v16.8h
  uadalp v11.4s, v16.8h
  uadalp v12.4s, v16.8h
  uadalp v13.4s, v16.8h
  uadalp v14.4s, v16.8h
  uadalp v15.4s, v16.8h

Initialization

  movi v16.16b, 17

(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 simd uop inputs (81)	? int output thing (e9)	? simd retires (ee)	? int retires (ef)
160204	80065	160107	101	160006	100	160010	300	640036	160108	200	160012	200	320028	1	160000	100
160204	80035	160107	101	160006	100	160010	300	640036	160108	200	160012	200	320024	1	160000	100
160204	80035	160105	101	160004	100	160008	300	640044	160110	200	160014	200	320326	1	160000	100
160204	80037	160107	101	160006	100	160010	300	640036	160108	200	160012	200	320024	1	160000	100
160204	80035	160105	101	160004	100	160008	300	640036	160108	200	160012	200	320024	1	160000	100
160204	80035	160105	101	160004	100	160008	300	640036	160108	200	160012	200	320024	1	160000	100
160204	80035	160105	101	160004	100	160008	300	640036	160108	200	160012	200	320024	1	160000	100
160204	80035	160105	101	160004	100	160008	300	640036	160108	200	160012	200	320024	1	160000	100
160204	80035	160105	101	160004	100	160008	300	640036	160108	200	160012	200	320024	1	160000	100
160204	80334	160301	101	160200	100	160204	300	640036	160108	200	160012	200	320024	1	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	80409	160017	11	160006	10	160010	30	640000	160010	20	160000	20	320000	1	160000	10
160024	80084	160011	11	160000	10	160000	30	640000	160010	20	160000	20	320000	1	160000	10
160024	80038	160011	11	160000	10	160000	30	640000	160010	20	160000	20	320000	1	160000	10
160025	80075	160052	11	160041	10	160053	30	640000	160010	20	160000	20	320000	1	160000	10
160024	80038	160011	11	160000	10	160000	30	640000	160010	20	160000	20	320000	1	160000	10
160024	80038	160011	11	160000	10	160000	30	640000	160010	20	160000	20	320000	1	160000	10
160024	80038	160011	11	160000	10	160000	30	640000	160010	20	160000	20	320000	1	160000	10
160024	80038	160011	11	160000	10	160000	30	640000	160010	20	160000	20	320000	1	160000	10
160024	80038	160011	11	160000	10	160000	30	640000	160010	20	160000	20	320000	1	160000	10
160024	80038	160011	11	160000	10	160000	30	640000	160010	20	160000	20	320000	1	160000	10

UADALP (vector, 4S)

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: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 5: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations