Icestorm: CMP (sxtb, 32-bit) measurements

Code:

  cmp w0, w1, sxtb

Initialization

  mov x0, 1
  mov x1, 2

(no loop instructions)

1000 unrolls and 1 iteration

Retires: 1.000

Issues: 2.000

Integer unit issues: 2.001

Load/store unit issues: 0.000

SIMD/FP unit issues: 0.000

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

retire uop (01)	cycle (02)	schedule uop (52)	schedule int uop (53)	dispatch int uop (56)	int uops in schedulers (59)	dispatch uop (78)	map int uop (7c)	map int uop inputs (7f)	? int output thing (e9)
1004	698	2001	2001	1000	14678	1000	1000	2000	2001
1004	698	2001	2001	1000	14678	1000	1000	2000	2001
1004	698	2001	2001	1000	14678	1000	1000	2000	2001
1004	698	2001	2001	1000	14678	1000	1000	2000	2001
1004	698	2001	2001	1000	14678	1000	1000	2000	2001
1004	698	2001	2001	1000	14678	1000	1000	2000	2001
1004	698	2001	2001	1000	14678	1000	1000	2000	2001
1004	698	2001	2001	1000	14678	1000	1000	2000	2001
1004	698	2001	2001	1000	14678	1000	1000	2000	2001
1004	698	2001	2001	1000	14678	1000	1000	2000	2001

Test 2: Latency 3->1

Chain cycles: 1

Code:

  cmp w0, w1, sxtb
  cset x0, cc

Initialization

  mov x0, 1
  mov x1, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code, minus 1 chain cycle): 2.0030

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

retire uop (01)	cycle (02)	schedule uop (52)	schedule int uop (53)	dispatch int uop (56)	int uops in schedulers (59)	dispatch uop (78)	map int uop (7c)	map int uop inputs (7f)	? int output thing (e9)	? int retires (ef)
20204	30030	30101	30101	20106	789231	20105	20210	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100

1000 unrolls and 10 iterations

Result (median cycles for code, minus 1 chain cycle): 2.0030

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

retire uop (01)	cycle (02)	schedule uop (52)	schedule int uop (53)	dispatch int uop (56)	int uops in schedulers (59)	dispatch uop (78)	map int uop (7c)	map int uop inputs (7f)	? int output thing (e9)	? int retires (ef)
20024	30224	30095	30095	20168	789380	20010	20020	30020	30001	10010
20024	30030	30011	30011	20010	789438	20010	20020	30020	30001	10010
20024	30030	30011	30011	20010	789438	20010	20020	30020	30001	10010
20024	30030	30011	30011	20010	789438	20010	20020	30020	30001	10010
20024	30030	30011	30011	20010	789438	20010	20020	30020	30001	10010
20024	30030	30011	30011	20010	789438	20010	20020	30020	30001	10010
20024	30030	30011	30011	20010	789438	20010	20020	30020	30001	10010
20024	30030	30011	30011	20010	789438	20010	20020	30020	30001	10010
20024	30030	30011	30011	20010	789438	20010	20020	30020	30001	10010
20024	30030	30011	30011	20010	789438	20010	20020	30020	30001	10010

Test 3: Latency 3->2

Chain cycles: 1

Code:

  cmp w0, w1, sxtb
  cset x1, cc

Initialization

  mov x0, 1
  mov x1, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code, minus 1 chain cycle): 2.0030

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

retire uop (01)	cycle (02)	schedule uop (52)	schedule int uop (53)	dispatch int uop (56)	int uops in schedulers (59)	dispatch uop (78)	map int uop (7c)	map int uop inputs (7f)	? int output thing (e9)	? int retires (ef)
20204	30030	30101	30101	20106	789369	20105	20212	30221	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100
20204	30030	30101	30101	20105	789369	20105	20212	30218	30001	10100

1000 unrolls and 10 iterations

Result (median cycles for code, minus 1 chain cycle): 2.0030

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

retire uop (01)	cycle (02)	schedule uop (52)	schedule int uop (53)	dispatch int uop (56)	int uops in schedulers (59)	dispatch uop (78)	map int uop (7c)	map int uop inputs (7f)	map ldst uop inputs (80)	? int output thing (e9)	? ldst retires (ed)	? int retires (ef)
20024	30030	30011	30011	20015	789381	20015	20030	30020	0	30001	0	10010
20025	30060	30025	30025	20050	789379	20010	20020	30020	0	30001	0	10010
20024	30030	30011	30011	20010	789438	20010	20020	30020	0	30001	0	10010
20024	30030	30011	30011	20010	789197	20015	20030	30035	0	30001	0	10010
20024	30030	30011	30011	20015	789459	20015	20032	30038	0	30001	0	10010
20024	30030	30011	30011	20015	789459	20015	20032	30038	0	30001	0	10010
20024	30030	30011	30011	20015	789459	20015	20032	30038	0	30001	0	10010
20024	30030	30011	30011	20015	789459	20015	20032	30038	0	30001	0	10010
20025	30060	30025	30025	20050	789459	20015	20032	30098	0	30015	0	10010
20024	30030	30011	30011	20010	789438	20010	20020	30020	0	30001	0	10010

Test 4: throughput

Count: 8

Code:

  cmp w0, w1, sxtb
  cmp w0, w1, sxtb
  cmp w0, w1, sxtb
  cmp w0, w1, sxtb
  cmp w0, w1, sxtb
  cmp w0, w1, sxtb
  cmp w0, w1, sxtb
  cmp w0, w1, sxtb

Initialization

  mov x0, 1
  mov x1, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.6675

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

retire uop (01)	cycle (02)	schedule uop (52)	schedule int uop (53)	dispatch int uop (56)	int uops in schedulers (59)	dispatch uop (78)	map int uop (7c)	map int uop inputs (7f)	map ldst uop inputs (80)	? int output thing (e9)	? ldst retires (ed)	? simd retires (ee)	? int retires (ef)
80204	53427	160118	160118	80126	1177761	80130	80230	160252	0	160016	0	0	100
80205	53435	160162	160162	80163	1177635	80123	80224	160248	0	160014	0	0	100
80204	53402	160114	160114	80123	1177607	80123	80224	160248	0	160014	0	0	100
80204	53402	160114	160114	80123	1177627	80123	80224	160248	0	160014	0	0	100
80204	53402	160114	160114	80123	1177603	80123	80224	160248	0	160014	0	0	100
80204	53402	160114	160114	80123	1177631	80123	80224	160248	0	160014	0	0	100
80204	53402	160114	160114	80123	1177597	80123	80224	160248	0	160014	0	0	100
80204	53402	160114	160114	80123	1177627	80123	80224	160248	0	160014	0	0	100
80204	53402	160114	160114	80123	1177844	80166	80268	160248	0	160014	0	0	100
80204	53402	160114	160114	80123	1177599	80123	80224	160248	0	160014	0	0	100

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.6671

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

retire uop (01)	cycle (02)	schedule uop (52)	schedule int uop (53)	dispatch int uop (56)	int uops in schedulers (59)	dispatch uop (78)	map int uop (7c)	map int uop inputs (7f)	? int output thing (e9)	? int retires (ef)
80024	53403	160039	160039	80047	1170032	80020	80020	160020	160011	10
80024	53371	160021	160021	80020	1170032	80020	80020	160020	160011	10
80024	53371	160021	160021	80020	1170026	80020	80020	160020	160011	10
80024	53371	160021	160021	80020	1170032	80020	80020	160020	160011	10
80024	53371	160021	160021	80020	1170032	80020	80020	160020	160011	10
80024	53371	160021	160021	80020	1170032	80020	80020	160020	160011	10
80024	53371	160021	160021	80020	1170032	80020	80020	160148	160074	10
80024	53371	160021	160021	80020	1170032	80020	80020	160020	160011	10
80024	53371	160021	160021	80020	1170032	80020	80020	160020	160011	10
80024	53371	160021	160021	80020	1170032	80020	80020	160020	160011	10

CMP (sxtb, 32-bit)

Test 1: uops

1000 unrolls and 1 iteration

Test 2: Latency 3->1

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 3: Latency 3->2

100 unrolls and 100 iterations

1000 unrolls and 10 iterations

Test 4: throughput

100 unrolls and 100 iterations

1000 unrolls and 10 iterations