Icestorm: SMLSL2 (by element, 4S) measurements

Code:

  smlsl2 v0.4s, v1.8h, v2.h[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: 0.001

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	3000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	3000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	3000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	3000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	3000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	3000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	3000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	3000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	3000	1	1000
1004	3033	1001	1	1000	1000	75905	1000	1000	3000	1	1000

Test 2: Latency 1->1

Code:

  smlsl2 v0.4s, v1.8h, v2.h[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): 3.0033

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	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	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	30012	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30144	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10

Test 3: Latency 1->2

Code:

  smlsl2 v0.4s, v0.8h, v1.h[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): 3.0033

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	10006	200	30018	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	30012	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	67	769247	10051	20	10046	20	30012	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	30000	11	10000	10

Test 4: Latency 1->3

Code:

  smlsl2 v0.4s, v1.8h, v0.h[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): 3.0033

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)	schedule ldst uop (55)	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	0	100	10000	300	768905	10100	200	10006	200	30012	1	10000	100
10204	30033	10101	101	10000	0	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	0	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	0	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	0	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	0	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	0	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	0	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	0	100	10000	300	768905	10100	200	10004	200	30012	1	10000	100
10204	30033	10101	101	10000	0	100	10000	300	768905	10100	200	10004	200	30012	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)	schedule ldst uop (55)	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)
10024	30033	10021	21	10000	0	20	10000	70	768905	10020	20	10000	20	0	30000	11	0	10000	10
10024	30033	10021	21	10000	0	20	10000	70	768905	10020	20	10000	20	0	30000	11	0	10000	10
10024	30033	10021	21	10000	0	20	10000	70	768905	10020	20	10000	20	0	30000	11	0	10000	10
10024	30033	10021	21	10000	0	20	10000	70	768905	10020	20	10000	20	0	30000	11	0	10000	10
10024	30033	10021	21	10000	0	20	10000	70	768905	10020	20	10000	20	0	30129	11	0	10000	10
10024	30033	10021	21	10000	0	20	10000	70	768905	10020	20	10000	20	0	30000	11	0	10000	10
10024	30033	10021	21	10000	0	20	10000	70	768905	10020	20	10000	20	0	30000	11	0	10000	10
10024	30033	10021	21	10000	0	20	10000	70	768905	10020	20	10000	20	0	30000	11	0	10000	10
10024	30033	10021	21	10000	0	20	10000	70	768905	10020	20	10000	20	0	30000	11	0	10000	10
10024	30033	10021	21	10000	0	20	10000	70	768905	10020	20	10000	20	0	30000	11	0	10000	10

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  smlsl2 v0.4s, v8.8h, v9.h[1]
  movi v1.16b, 0
  smlsl2 v1.4s, v8.8h, v9.h[1]
  movi v2.16b, 0
  smlsl2 v2.4s, v8.8h, v9.h[1]
  movi v3.16b, 0
  smlsl2 v3.4s, v8.8h, v9.h[1]
  movi v4.16b, 0
  smlsl2 v4.4s, v8.8h, v9.h[1]
  movi v5.16b, 0
  smlsl2 v5.4s, v8.8h, v9.h[1]
  movi v6.16b, 0
  smlsl2 v6.4s, v8.8h, v9.h[1]
  movi v7.16b, 0
  smlsl2 v7.4s, v8.8h, v9.h[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): 0.5011

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	40455	80110	101	80009	100	80013	300	320056	80113	200	80013	200	240039	1	160000	100
160204	40108	80110	101	80009	100	80013	300	320052	80112	200	80012	200	240036	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320204	80150	200	80050	200	240036	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	240036	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	240036	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	240036	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	240036	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	240036	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	240036	1	160000	100
160204	40086	80109	101	80008	100	80012	300	320052	80112	200	80012	200	240036	1	160000	100

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.5053

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	43863	80019	11	80008	10	80012	30	320056	80023	20	80013	20	240000	1	160000	10
160024	41252	80011	11	80000	10	80000	30	320000	80010	20	80000	20	240000	1	160000	10
160024	40397	80011	11	80000	10	80000	30	320000	80010	20	80000	20	240000	1	160000	10
160024	41642	80011	11	80000	10	80000	30	320000	80010	20	80000	20	240000	1	160000	10
160024	40434	80011	11	80000	10	80000	30	320000	80010	20	80000	20	240000	1	160000	10
160024	40435	80011	11	80000	10	80000	30	320000	80010	20	80000	20	240000	1	160000	10
160024	40424	80011	11	80000	10	80000	30	320196	80058	20	80048	20	240000	1	160000	10
160024	40497	80011	11	80000	10	80000	30	320000	80010	20	80000	20	240000	1	160000	10
160024	40437	80011	11	80000	10	80000	30	320000	80010	20	80000	20	240000	1	160000	10
160024	40436	80011	11	80000	10	80000	30	320000	80010	20	80000	20	240000	1	160000	10

Test 6: throughput

Count: 12

Code:

  smlsl2 v0.4s, v12.8h, v13.h[1]
  smlsl2 v1.4s, v12.8h, v13.h[1]
  smlsl2 v2.4s, v12.8h, v13.h[1]
  smlsl2 v3.4s, v12.8h, v13.h[1]
  smlsl2 v4.4s, v12.8h, v13.h[1]
  smlsl2 v5.4s, v12.8h, v13.h[1]
  smlsl2 v6.4s, v12.8h, v13.h[1]
  smlsl2 v7.4s, v12.8h, v13.h[1]
  smlsl2 v8.4s, v12.8h, v13.h[1]
  smlsl2 v9.4s, v12.8h, v13.h[1]
  smlsl2 v10.4s, v12.8h, v13.h[1]
  smlsl2 v11.4s, v12.8h, v13.h[1]

Initialization

  movi v12.16b, 13
  movi v13.16b, 14

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.5003

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)
120204	60075	120107	101	120006	100	120010	300	480044	120110	200	120014	200	360042	1	120000	100
120204	60035	120107	101	120006	100	120010	300	480044	120110	200	120013	200	360036	1	120000	100
120204	60035	120105	101	120004	100	120008	300	480036	120108	200	120012	200	360036	1	120000	100
120204	60035	120105	101	120004	100	120008	300	480036	120108	200	120012	200	360036	1	120000	100
120204	60035	120105	101	120004	100	120008	300	480036	120108	200	120012	200	360036	1	120000	100
120204	60035	120105	101	120004	100	120008	300	480036	120108	200	120012	200	360036	1	120000	100
120204	60035	120105	101	120004	100	120008	300	480036	120108	200	120012	200	360036	1	120000	100
120204	60035	120105	101	120004	100	120008	300	480036	120108	200	120012	200	360036	1	120000	100
120204	60035	120105	101	120004	100	120008	300	480036	120108	200	120012	200	360036	1	120000	100
120204	60035	120105	101	120004	100	120008	300	480380	120200	200	120113	200	360036	1	120000	100

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.5003

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)
120024	60181	120015	11	120004	10	120008	30	480044	120020	20	120014	20	360000	1	120000	10
120024	60035	120011	11	120000	10	120000	30	480000	120010	20	120000	20	360000	1	120000	10
120024	60035	120011	11	120000	10	120000	30	480000	120010	20	120000	20	360000	1	120000	10
120024	60035	120011	11	120000	10	120000	30	480000	120010	20	120000	20	360000	1	120000	10
120024	60035	120011	11	120000	10	120000	30	480000	120010	20	120000	20	360000	1	120000	10
120024	60035	120011	11	120000	10	120000	30	480000	120010	20	120000	20	360000	1	120000	10
120024	60035	120011	11	120000	10	120000	30	480000	120010	20	120000	20	360000	1	120000	10
120024	60035	120011	11	120000	10	120000	30	480000	120010	20	120000	20	360000	1	120000	10
120024	60035	120011	11	120000	10	120000	30	480000	120010	20	120000	20	360000	1	120000	10
120024	60035	120011	11	120000	10	120000	30	480000	120010	20	120000	20	360000	1	120000	10

SMLSL2 (by element, 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: 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