Firestorm: SQSHRUN2 (4S) measurements

Code:

  sqshrun2 v0.8h, v1.4s, #3

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: 0.000

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	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 tlb access (a0)	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	22	0	82	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	1	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	18	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	1	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	156	2548	25	1000	1000	1000	398313	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	1	0	73	1	16	1	1	2630	1000	3038	3038	3038	3038	3038

Test 2: Latency 1->1

Code:

  sqshrun2 v0.8h, v1.4s, #3

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	mmu table walk data (08)	09	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)	simd prf full (72)	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)	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	30037	225	0	0	0	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	20000	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	30086	30086	30133	30038
10204	30037	225	0	0	0	0	0	0	0	323	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	20000	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	224	0	0	0	0	0	0	0	328	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	20000	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	225	0	0	0	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	20000	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	225	0	0	0	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	1	0	3	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	20000	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	225	0	0	0	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	20000	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	225	0	0	0	0	0	0	0	166	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	20000	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	225	0	0	0	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	20000	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	225	0	0	0	0	0	0	0	61	29548	25	10100	100	10000	100	10000	500	4277313	0	30018	30037	30037	28265	0	3	28745	10100	200	10000	200	20000	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)	l2 tlb miss data (0b)	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	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10024	30037	225	0	0	0	168	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	3	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	224	0	0	0	631	29548	25	10010	10	10000	10	10000	50	4277313	1	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30086	30085	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	105	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	61	29548	25	10010	10	10000	10	10000	50	4277313	0	30018	30037	30037	28287	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	640	2	16	2	2	29630	10000	10	30038	30038	30038	30038	30038

Test 3: Latency 1->2

Code:

  sqshrun2 v0.8h, v0.4s, #3

Initialization

  movi v0.16b, 1

(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)	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	ac	branch cond mispred nonspec (c5)	branch mispred nonspec (cb)	cd	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
10204	30037	225	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28271	7	28741	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	1	1	717	0	16	0	0	29645	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28271	7	28741	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	1	1	717	0	16	0	0	29646	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28271	7	28740	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	1	1	722	2	24	2	2	29629	10000	100	30089	30038	30038	30038	30038
10204	30037	225	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28271	6	28740	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	1	1	718	0	16	0	0	29645	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28271	6	28740	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	1	1	717	0	16	0	0	29645	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28271	7	28741	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	1	1	718	0	16	0	0	29645	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	3845	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28271	7	28741	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	1	1	718	0	16	0	0	29645	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28271	7	28740	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	1	1	717	0	16	0	0	29646	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28271	7	28741	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	1	1	718	0	16	0	0	29646	10000	100	30038	30038	30038	30038	30038
10204	30037	224	0	0	61	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28271	6	28740	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	1	1	1	717	0	16	0	0	29646	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	l2 tlb miss data (0b)	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)	simd prf full (72)	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)	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	225	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	0	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	0	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	0	30018	30037	30037	28286	0	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	0	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	224	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	0	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	0	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	0	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	1	30018	30037	30037	28286	0	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	0	30018	30037	30037	28286	0	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	0	30018	30037	30037	28286	0	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038

Test 4: throughput

Count: 8

Code:

  movi v0.16b, 0
  sqshrun2 v0.8h, v8.4s, #3
  movi v1.16b, 0
  sqshrun2 v1.8h, v8.4s, #3
  movi v2.16b, 0
  sqshrun2 v2.8h, v8.4s, #3
  movi v3.16b, 0
  sqshrun2 v3.8h, v8.4s, #3
  movi v4.16b, 0
  sqshrun2 v4.8h, v8.4s, #3
  movi v5.16b, 0
  sqshrun2 v5.8h, v8.4s, #3
  movi v6.16b, 0
  sqshrun2 v6.8h, v8.4s, #3
  movi v7.16b, 0
  sqshrun2 v7.8h, v8.4s, #3

Initialization

  movi v8.16b, 9

(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	l1i tlb fill (04)	mmu table walk data (08)	09	19	1e	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	rob full (74)	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)	ac	c2	branch cond mispred nonspec (c5)	branch mispred nonspec (cb)	cd	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160204	20078	150	1	0	0	0	12	3	36	25	80116	100	80016	100	80028	500	640196	1	20045	20065	20065	0	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	1	1	1	10133	7	16	11	13	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	151	0	0	0	0	0	5	36	25	80116	100	80016	100	80028	500	640196	1	20045	20065	20065	0	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	1	1	1	10135	13	16	13	14	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	1	0	0	0	0	5	67	25	80116	100	80016	100	80028	500	640196	1	20045	20065	20065	0	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	1	1	1	10127	14	16	12	11	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	1	0	0	0	12	4	156	25	80116	100	80016	100	80028	500	640196	1	20045	20065	20065	0	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	3	2	1	1	1	10135	13	16	6	12	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	1	0	0	0	0	4	36	25	80116	100	80016	100	80028	500	640196	1	20045	20065	20065	0	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	1	1	1	10134	11	16	9	12	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	1	0	0	0	9	4	61	25	80116	100	80016	100	80028	500	640196	1	20045	20065	20065	0	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	1	1	1	10133	12	16	6	12	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	151	0	0	0	0	21	4	61	104	80116	100	80016	100	80028	500	640196	1	20045	20065	20065	0	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	1	1	1	10135	13	16	14	13	20130	160000	100	20066	20066	20066	20066	20066
160204	20065	151	0	0	1	1	15	4	67	25	80116	100	80016	100	80028	500	640196	1	20045	20065	20065	0	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	3	0	1	1	1	10135	13	16	11	13	20132	160000	100	20155	20150	20157	20149	20066
160204	20065	151	1	0	0	3	0	6	107	25	80116	100	80016	100	80028	500	640196	1	20045	20065	20065	0	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	0	1	1	1	10133	11	16	5	15	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	1	0	0	0	0	3	61	25	80116	100	80016	100	80028	500	640196	1	20045	20065	20065	0	28	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	1	0	0	1	1	1	10134	11	16	13	11	20062	160000	100	20066	20066	20066	20066	20066

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	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)	ac	c2	cf	d0	d2	l1i cache miss demand (d3)	d5	map dispatch bubble (d6)	d9	da	db	dd	fetch restart (de)	e0	ea	eb	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160024	20081	156	360	45	25	80010	10	80000	10	80000	50	640000	0	1	0	20027	20046	20046	3	22	80010	20	80000	20	160000	20050	20046	1	1	160021	10	9	10	10	160000	10	0	0	0	10029	3	2	1	6	20	2	1	1	5	8	20043	15	0	160000	10	20047	20051	20047	20047	20047
160024	20046	151	0	45	25	80010	10	80000	10	80000	50	640000	0	1	0	20027	20046	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	0	10031	3	1	1	4	20	2	1	1	5	6	20043	15	0	160000	10	20047	20047	20047	20047	20051
160024	20046	150	27	324	25	80010	10	80000	10	80000	50	640000	1	1	0	20027	20046	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	0	10029	3	1	1	7	20	2	1	1	6	8	20043	15	0	160000	10	20047	20047	20047	20047	20047
160024	20046	150	39	45	36	80010	10	80000	10	80000	50	640000	1	1	0	20027	20050	20046	3	22	80114	20	80000	20	160000	20052	20046	1	1	160021	10	9	10	10	160000	10	0	0	0	10028	3	2	1	6	20	2	1	1	6	6	20043	15	0	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	45	25	80010	10	80000	10	80000	50	640000	1	1	0	20027	20046	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	0	10029	3	2	2	6	20	2	1	1	6	6	20049	15	0	160000	10	20051	20047	20047	20047	20047
160024	20046	150	48	45	25	80010	10	80000	10	80000	50	640000	1	1	0	20027	20046	20046	3	22	80010	20	80000	20	160000	20046	20052	1	1	160021	10	9	10	10	160000	10	0	0	0	10029	6	2	2	6	20	4	2	1	7	6	20043	15	0	160000	10	20047	20051	20051	20051	20051
160024	20050	150	0	45	25	80010	10	80000	10	80000	50	640000	0	1	0	20033	20046	20046	3	22	80010	20	80000	20	160000	20050	20046	1	1	160021	10	9	10	10	160000	10	0	0	0	10028	6	1	1	8	20	2	1	1	6	5	20043	15	0	160000	10	20047	20047	20053	20051	20047
160024	20046	150	0	45	25	80010	10	80000	10	80000	50	640000	1	1	0	20027	20046	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	0	10030	3	1	2	7	20	3	1	2	8	6	20047	15	0	160000	10	20051	20047	20051	20047	20051
160024	20050	150	0	45	25	80010	10	80000	10	80000	50	640000	1	1	0	20027	20050	20046	3	22	80429	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	0	10029	6	1	1	5	24	2	1	2	5	7	20049	15	0	160000	10	20051	20047	20047	20053	20047
160024	20050	150	423	45	25	80010	10	80000	10	80000	50	640000	1	1	0	20027	20050	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	0	10029	3	1	1	6	20	2	1	1	5	6	20043	15	0	160000	10	20047	20047	20047	20047	20047

Test 5: throughput

Count: 16

Code:

  sqshrun2 v0.8h, v16.4s, #3
  sqshrun2 v1.8h, v16.4s, #3
  sqshrun2 v2.8h, v16.4s, #3
  sqshrun2 v3.8h, v16.4s, #3
  sqshrun2 v4.8h, v16.4s, #3
  sqshrun2 v5.8h, v16.4s, #3
  sqshrun2 v6.8h, v16.4s, #3
  sqshrun2 v7.8h, v16.4s, #3
  sqshrun2 v8.8h, v16.4s, #3
  sqshrun2 v9.8h, v16.4s, #3
  sqshrun2 v10.8h, v16.4s, #3
  sqshrun2 v11.8h, v16.4s, #3
  sqshrun2 v12.8h, v16.4s, #3
  sqshrun2 v13.8h, v16.4s, #3
  sqshrun2 v14.8h, v16.4s, #3
  sqshrun2 v15.8h, v16.4s, #3

Initialization

  movi v16.16b, 17

(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	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)	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)	91	inst branch cond (94)	inst int alu (97)	inst simd alu (9a)	9f	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
160204	40039	300	39	30	25	160108	100	160008	100	160020	500	1280132	1	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	0	100	100	160000	100	0	1	1	1	10118	0	16	0	0	40036	0	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	30	25	160108	100	160008	100	160020	500	1280132	1	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	0	100	100	160000	100	0	1	1	1	10118	0	16	0	0	40036	0	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	30	25	160108	100	160008	100	160020	500	1280132	1	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	0	100	100	160000	100	0	1	1	1	10118	0	16	0	0	40036	0	160000	100	40040	40040	40040	40040	40040
160204	40039	299	0	30	25	160108	100	160008	100	160020	500	1280132	1	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	0	100	100	160000	100	0	1	1	1	10118	0	16	0	0	40036	0	160000	100	40040	40040	40040	40040	40040
160204	40039	299	0	30	25	160108	100	160008	100	160020	500	1280132	0	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	0	100	100	160000	100	0	1	1	1	10118	1	16	1	0	40036	0	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	30	25	160108	100	160008	100	160020	500	1280132	1	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	0	100	100	160000	100	0	1	1	1	10118	0	16	0	0	40036	0	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	30	25	160108	100	160008	100	160020	500	1280132	0	40020	40039	40189	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	0	100	100	160000	100	0	1	1	1	10118	0	16	0	0	40036	3	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	30	25	160108	100	160008	100	160020	500	1280132	1	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	0	100	100	160000	100	0	1	1	1	10118	0	16	1	1	40036	0	160000	100	40040	40040	40040	40040	40040
160204	40039	300	3	30	25	160108	100	160008	100	160020	500	1280132	1	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	0	100	100	160000	100	0	1	1	1	10118	0	16	0	0	40036	0	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	30	25	160108	100	160008	100	160020	500	1280132	0	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	0	100	100	160000	100	0	1	1	1	10118	0	16	1	0	40036	0	160000	100	40040	40040	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	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	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 cache writeback (a8)	ac	cf	d0	d2	l1i cache miss demand (d3)	d5	map dispatch bubble (d6)	d9	da	db	dd	fetch restart (de)	e0	ea	eb	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160024	40051	300	96	46	25	160010	10	160000	10	160000	50	1280000	1	1	0	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	10022	8	5	1	24	16	4	1	1	24	12	40036	20	6	160000	10	40040	40040	40040	40040	40040
160024	40039	300	27	46	25	160010	10	160000	10	160000	50	1280000	1	1	5	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	10022	8	3	1	12	16	2	1	1	24	12	40036	20	6	160000	10	40040	40040	40040	40040	40040
160024	40039	300	39	46	25	160010	10	160000	10	160000	50	1280000	1	1	5	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	10022	8	3	1	24	16	2	2	1	24	24	40036	20	6	160000	10	40040	40040	40040	40040	40040
160024	40039	300	264	46	25	160010	10	160000	10	160000	50	1280000	1	1	5	40020	3	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	10022	8	5	1	12	16	2	1	1	12	24	40036	20	6	160000	10	40040	40040	40040	40040	40040
160024	40039	300	36	46	25	160010	10	160000	10	160000	50	1280000	1	1	5	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	10022	8	5	1	10	16	2	1	1	24	12	40036	20	6	160000	10	40040	40040	40040	40040	40040
160024	40039	300	54	67	62	160306	10	160292	10	160000	50	1280000	1	1	5	40020	0	40039	40039	19996	20	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	10022	8	5	1	24	16	2	1	1	10	24	40036	20	6	160000	10	40040	40040	40040	40040	40040
160024	40039	299	36	46	25	160010	10	160000	10	160000	50	1280000	1	1	5	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	10022	8	5	1	12	16	2	1	1	24	12	40036	20	6	160000	10	40040	40040	40040	40040	40040
160024	40039	300	48	46	25	160010	10	160000	10	160000	50	1280000	1	1	5	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	10022	8	5	1	24	16	2	1	1	24	12	40036	20	6	160000	10	40040	40040	40040	40040	40040
160024	40039	300	30	46	25	160010	10	160000	10	160000	50	1280000	1	1	5	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	0	10022	8	5	1	24	16	2	1	1	19	24	40036	20	12	160000	10	40040	40040	40040	40040	40040
160024	40039	299	27	46	25	160010	10	160000	10	160000	50	1280000	1	1	5	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40039	40039	1	1	160021	10	9	10	10	160000	10	0	75	10022	8	5	1	24	16	2	1	1	24	24	40036	20	6	160000	10	40040	40040	40040	40040	40040

SQSHRUN2 (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