Firestorm: SQRSHRN2 (4S) measurements

Code:

  sqrshrn2 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)	60	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 cache writeback (a8)	a9	ac	cf	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? simd retires (ee)	f5	f6	f7	f8	fd
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	1	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	16	0	0	73	3	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	1	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	43	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3133	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	1	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2646	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	23	0	61	2548	25	1000	1000	1000	398313	1	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2672	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	0	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038
1004	3037	22	0	61	2548	25	1000	1000	1000	398313	1	3018	3037	3037	2415	3	2895	1000	1000	2000	3037	3037	1	1	1001	1000	0	0	0	73	2	16	2	2	2630	1000	3038	3038	3038	3038	3038

Test 2: Latency 1->1

Code:

  sqrshrn2 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)	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)	ld unit uop (a6)	st unit uop (a7)	l1d cache writeback (a8)	a9	ac	c2	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	232	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	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	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30086	30100	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	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	0	710	1	16	1	1	29634	0	10000	100	30038	30038	30038	30038	30038
10204	30037	224	0	1	0	0	0	3	0	103	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28265	25	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	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	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	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	141	0	61	29548	25	10100	100	10000	116	10000	500	4277313	1	30018	30037	30037	28265	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	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	3	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	0	0	710	1	16	1	1	29634	15	10000	100	30038	30038	30038	30038	30227
10204	30037	224	0	0	0	0	0	0	0	82	29548	25	10100	100	10000	100	10000	500	4277313	1	30018	30037	30037	28272	12	28745	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	3	1	1	4	2783	0	733	1	16	1	1	29850	28	10000	100	30740	30994	30894	30954	30276
10204	30517	232	1	1	0	8	9	1059	704	5774	29494	282	10330	186	10144	146	11192	745	4301774	1	30270	30416	30457	28295	37	28902	11185	222	11330	234	22716	30359	30414	8	1	10201	100	99	100	100	10000	100	4	4	3	1	30	2	14578	2	851	4	105	1	2	29942	31	10000	100	30368	30370	30420	30418	30421
10204	30464	228	0	1	1	9	8	1191	704	5265	29466	200	10208	149	10072	157	11267	718	4288169	1	30306	30514	30466	28302	31	28878	11335	226	11324	224	22664	30232	30324	10	1	10201	100	99	100	100	10000	100	0	0	2	1	0	4	25098	2	889	4	106	3	1	29820	31	10000	100	30275	30552	30562	30470	30407

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)	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)	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 miss (a1)	ld unit uop (a6)	l1d cache writeback (a8)	a9	ac	c2	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
10024	30037	225	0	0	0	0	0	9	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	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30086	30132	30038	30038	30038
10024	30037	225	0	0	0	0	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	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	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	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	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	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	233	0	0	0	0	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	43	0	9	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	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	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	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	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	224	0	0	0	0	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	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	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	0	0	6	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	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	0	0	0	0	640	2	16	2	2	29630	0	10000	10	30038	30038	30038	30038	30038

Test 3: Latency 1->2

Code:

  sqrshrn2 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	mmu table walk data (08)	l2 tlb miss data (0b)	18	19	1e	1f	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	l1d tlb miss (a1)	ld unit uop (a6)	l1d cache writeback (a8)	a9	ac	c2	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
10204	30037	225	0	0	0	0	30	0	0	82	29547	25	10100	100	10000	104	10450	500	4277160	30018	30037	30037	28271	6	28741	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	1	1	1	718	0	16	0	0	29646	0	10000	100	30232	30038	30038	30038	30038
10204	30037	224	0	0	0	0	0	0	0	804	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28252	6	28733	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	4	1	1	1	722	2	24	2	2	29629	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	0	0	97	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28252	6	28733	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	1	1	1	722	2	24	2	2	29629	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	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	0	0	0	0	0	1	1	1	717	0	16	0	0	29646	0	10000	100	30038	30038	30038	30038	30038
10204	30037	224	0	0	0	0	0	0	0	536	29547	25	10100	100	10000	104	10000	522	4277160	30018	30037	30037	28271	6	28741	10100	200	10008	200	20016	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	1	0	6	0	1	1	1	722	2	24	2	2	29629	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	0	0	97	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28252	6	28733	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	3	0	1	1	1	722	2	24	2	2	29629	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	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	0	0	0	0	0	1	1	1	718	0	16	0	0	29646	0	10000	100	30038	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	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	0	0	0	0	0	1	1	1	722	2	31	2	2	29629	0	10000	100	30038	30038	30038	30087	30038
10204	30037	224	0	0	0	0	408	0	0	97	29547	25	10100	100	10000	100	10000	500	4277160	30018	30037	30037	28252	6	28733	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	0	0	1	1	1	722	2	24	2	2	29629	0	10000	100	30230	30038	30038	30038	30038
10204	30037	225	0	0	0	0	0	0	0	97	29547	25	10100	100	10008	100	10000	500	4277160	30018	30037	30037	28252	6	28733	10100	200	10000	200	20000	30037	30037	1	1	10201	100	99	100	100	10000	100	0	0	0	0	3	0	1	1	1	722	2	24	2	2	29629	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)	18	19	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	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	a9	ac	cd	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	0	61	29547	25	10010	10	10000	10	10000	50	4277160	30018	0	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	224	0	0	0	0	61	29547	25	10010	10	10000	10	10000	66	4277160	30018	0	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	30018	0	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	30018	0	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	30018	0	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	30018	0	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	30018	0	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	30018	0	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	30018	0	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	0	0	0	640	2	16	2	2	29629	10000	10	30038	30038	30038	30038	30038
10024	30037	225	0	0	0	0	61	29547	25	10010	10	10000	10	10000	50	4277160	30018	0	30037	30037	28286	3	28767	10010	20	10000	20	20000	30037	30037	1	1	10021	10	9	10	10	10000	10	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
  sqrshrn2 v0.8h, v8.4s, #3
  movi v1.16b, 0
  sqrshrn2 v1.8h, v8.4s, #3
  movi v2.16b, 0
  sqrshrn2 v2.8h, v8.4s, #3
  movi v3.16b, 0
  sqrshrn2 v3.8h, v8.4s, #3
  movi v4.16b, 0
  sqrshrn2 v4.8h, v8.4s, #3
  movi v5.16b, 0
  sqrshrn2 v5.8h, v8.4s, #3
  movi v6.16b, 0
  sqrshrn2 v6.8h, v8.4s, #3
  movi v7.16b, 0
  sqrshrn2 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	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	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	l1d cache writeback (a8)	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
160204	20090	150	0	29	25	80116	100	80016	100	80028	500	640196	0	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	1	1	1	10126	8	16	8	9	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	151	0	29	49	80222	100	80016	100	80028	500	640196	0	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	1	1	1	10122	8	16	8	9	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	29	25	80116	100	80016	100	80028	500	640196	0	1	20045	20065	20065	0	6	12	80128	200	80136	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10122	3	16	8	3	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	29	25	80116	100	80016	100	80028	500	640196	0	1	20045	20065	20065	0	6	12	80128	200	80028	200	160056	20065	20065	1	1	160201	100	99	100	100	160000	100	1	0	1	1	1	10128	8	16	2	8	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	29	25	80116	100	80016	100	80028	500	640196	0	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	1	1	1	10127	8	16	8	8	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	151	0	29	25	80116	100	80016	100	80028	500	640196	0	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	1	1	1	10125	6	16	8	10	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	9	29	25	80116	100	80016	100	80028	500	640196	1	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	1	1	1	10127	7	16	7	7	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	29	25	80116	100	80016	100	80028	500	640196	0	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	1	1	1	10127	3	16	6	6	20062	160000	100	20066	20066	20066	20066	20066
160204	20065	150	0	29	25	80116	100	80016	100	80028	500	640196	0	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	1	1	1	10126	8	16	8	8	20062	160000	100	20147	20150	20066	20066	20146
160204	20065	151	0	29	25	80116	100	80016	100	80028	500	640196	0	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	1	1	1	10127	3	16	8	7	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	cf	d0	d2	l1i cache miss demand (d3)	d5	map dispatch bubble (d6)	d9	da	db	dd	fetch restart (de)	e0	ea	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160024	20088	150	0	283	25	80010	10	80000	10	80000	50	640000	1	1	5	20027	20046	20046	3	22	80010	20	80000	20	160000	20050	20046	1	1	160021	10	9	10	10	160000	10	0	0	10026	13	3	1	6	20	4	1	1	4	4	20043	32	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	233	25	80010	10	80000	10	80000	50	640000	1	1	10	20027	20046	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10027	13	5	1	6	20	2	1	2	4	4	20043	64	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	223	25	80010	10	80000	10	80000	50	640000	0	1	10	20027	20046	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	3	10027	13	5	1	4	24	2	1	1	7	6	20043	32	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	1766	25	80010	10	80000	10	80000	50	640000	1	1	10	20027	20046	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10027	13	6	1	4	20	2	1	1	6	4	20043	32	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	220	25	80010	10	80000	10	80000	50	640000	1	1	10	20027	20046	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10029	13	6	1	7	24	4	2	1	4	6	20043	76	160000	10	20051	20047	20051	20051	20051
160024	20050	150	0	224	25	80115	10	80000	10	80000	50	640000	0	1	10	20027	20046	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	78	10029	13	6	1	3	20	2	1	1	6	7	20043	64	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	218	25	80010	10	80000	10	80000	50	640000	1	1	10	20027	20046	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10027	13	6	1	4	20	2	1	1	4	7	20043	32	160000	10	20047	20047	20047	20047	20047
160024	20050	151	0	184	25	80010	10	80000	10	80000	50	640000	1	1	10	20027	20050	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10026	13	6	1	3	20	2	2	1	4	4	20043	32	160000	10	20047	20047	20047	20047	20047
160024	20046	151	0	183	25	80010	10	80000	10	80000	50	640000	1	1	10	20027	20046	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10027	13	6	1	4	20	2	1	1	4	4	20043	32	160000	10	20047	20047	20047	20047	20047
160024	20046	150	0	197	25	80010	10	80000	10	80000	50	640000	1	1	10	20027	20046	20046	3	22	80010	20	80000	20	160000	20046	20046	1	1	160021	10	9	10	10	160000	10	0	0	10030	13	6	2	7	20	2	1	1	6	6	20043	32	160000	10	20047	20047	20047	20051	20047

Test 5: throughput

Count: 16

Code:

  sqrshrn2 v0.8h, v16.4s, #3
  sqrshrn2 v1.8h, v16.4s, #3
  sqrshrn2 v2.8h, v16.4s, #3
  sqrshrn2 v3.8h, v16.4s, #3
  sqrshrn2 v4.8h, v16.4s, #3
  sqrshrn2 v5.8h, v16.4s, #3
  sqrshrn2 v6.8h, v16.4s, #3
  sqrshrn2 v7.8h, v16.4s, #3
  sqrshrn2 v8.8h, v16.4s, #3
  sqrshrn2 v9.8h, v16.4s, #3
  sqrshrn2 v10.8h, v16.4s, #3
  sqrshrn2 v11.8h, v16.4s, #3
  sqrshrn2 v12.8h, v16.4s, #3
  sqrshrn2 v13.8h, v16.4s, #3
  sqrshrn2 v14.8h, v16.4s, #3
  sqrshrn2 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): 0.2503

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

retire uop (01)	cycle (02)	03	19	1f	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	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	branch cond mispred nonspec (c5)	branch mispred nonspec (cb)	cd	cf	d2	d5	map dispatch bubble (d6)	dd	fetch restart (de)	e0	? int output thing (e9)	eb	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160204	40061	299	0	0	51	25	160108	100	160008	100	160020	500	1280132	0	0	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10118	0	1	16	0	0	40036	0	0	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	0	695	25	160108	100	160008	100	160020	500	1280132	0	1	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10118	0	2	16	0	0	40036	0	0	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	0	30	25	160108	100	160008	100	160020	500	1280132	0	1	40020	40039	40039	19977	6	19990	160120	200	160144	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	3	1	1	1	10118	0	1	16	0	0	40036	0	0	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	0	30	25	160108	100	160008	100	160020	500	1280132	0	0	40020	40039	40039	19977	6	19990	160120	200	160032	200	320064	40039	40039	1	1	160201	100	99	100	100	160000	100	0	0	1	1	1	10118	0	2	16	0	0	40036	0	0	160000	100	40040	40040	40040	40040	40040
160204	40039	300	0	0	30	25	160108	100	160008	100	160020	500	1280132	0	1	40020	40039	40039	19977	6	19990	160120	200	160038	200	320076	40049	40048	1	1	160201	100	99	100	100	160000	100	0	0	2	2	2	10128	0	2	23	1	1	40046	0	0	160000	100	40050	40050	40049	40050	40049
160204	40049	300	0	0	64	26	160116	100	160016	100	160028	500	1280196	0	1	40029	40048	40048	19976	9	19986	160128	200	160038	200	320076	40048	40048	1	1	160201	100	99	100	100	160000	100	0	0	2	2	2	10129	0	3	23	1	1	40046	0	0	160000	100	40050	40049	40049	40050	40049
160204	40049	300	0	0	64	26	160116	100	160016	100	160028	500	1280196	1	1	40029	40049	40048	19976	9	19986	160128	200	160038	200	320076	40049	40048	1	1	160201	100	99	100	100	160000	100	0	0	2	2	2	10128	0	3	23	1	1	40046	0	0	160000	100	40049	40049	40050	40049	40049
160204	40048	300	0	0	64	26	160116	100	160016	100	160028	500	1280196	0	0	40029	40048	40049	19976	9	19986	160128	200	160038	200	320076	40048	40049	1	1	160201	100	99	100	100	160000	100	0	0	2	2	2	10128	0	2	23	2	1	40045	0	0	160000	100	40049	40049	40049	40050	40049
160204	40048	299	0	0	64	26	160116	100	160016	100	160028	500	1280196	0	0	40029	40048	40048	19976	9	19986	160128	200	160038	200	320076	40048	40048	1	1	160201	100	99	100	100	160000	100	0	0	2	2	2	10128	0	3	23	1	1	40046	0	0	160000	100	40049	40049	40049	40049	40049
160204	40049	300	0	0	64	26	160116	100	160016	100	160028	500	1280196	0	1	40029	40049	40048	19976	9	19986	160128	200	160038	200	320076	40049	40048	1	1	160201	100	99	100	100	160000	100	0	0	2	2	2	10128	0	2	23	1	1	40045	0	0	160000	100	40049	40049	40050	40049	40049

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	19	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	branch cond mispred nonspec (c5)	cd	cf	d0	d2	l1i cache miss demand (d3)	d5	map dispatch bubble (d6)	d9	da	db	dd	fetch restart (de)	e0	? int output thing (e9)	ea	eb	ec	? simd retires (ee)	? int retires (ef)	f5	f6	f7	f8	fd
160024	40050	300	0	0	46	25	160010	10	160000	10	160000	50	1280000	1	1	10	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40041	40039	1	1	160021	10	9	10	10	160000	10	0	3	0	0	10022	13	2	1	9	16	4	2	2	13	6	40036	0	40	12	0	160000	10	40143	40040	40040	40040	40040
160024	40039	300	1	171	648	25	160010	10	160099	10	160000	50	1280000	1	1	10	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40041	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	10022	13	7	1	12	16	2	1	2	5	6	40036	0	20	6	0	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	46	25	160010	10	160000	10	160000	50	1280000	1	1	10	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40041	40039	1	1	160021	10	9	10	10	160000	10	0	3	0	0	10022	13	6	1	10	16	2	1	1	11	10	40036	0	20	6	0	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	12	46	25	160010	10	160000	10	160000	50	1280000	1	1	10	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40041	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	10022	13	6	1	4	16	2	1	1	5	10	40036	0	40	6	0	160000	10	40040	40040	40040	40040	40040
160024	40039	299	0	0	46	25	160010	10	160000	10	160210	50	1282464	1	1	10	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320814	40041	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	10022	13	6	1	6	16	2	1	1	6	10	40036	1	20	6	0	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	46	25	160010	10	160000	10	160000	50	1280000	1	1	10	40020	0	40039	40039	19996	3	20019	160010	20	160000	20	320000	40041	40039	1	1	160021	10	9	10	10	160000	10	0	0	0	0	10022	13	6	1	4	16	2	1	1	7	4	40036	0	20	6	0	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	6	88	25	160010	10	160000	10	160000	50	1280788	1	1	10	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	0	0	10022	13	6	1	5	16	4	1	2	4	6	40036	0	20	12	0	160000	10	40040	40040	40040	40040	40040
160024	40039	299	0	0	46	25	160010	10	160000	10	160000	50	1280000	1	1	10	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	0	0	10024	13	6	1	7	16	4	1	2	4	6	40036	0	20	12	0	160000	10	40040	40040	40040	40040	40040
160024	40039	300	0	0	52	25	160010	10	160000	10	160000	50	1280000	1	1	10	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	0	0	10022	13	6	1	5	16	2	1	1	4	6	40036	0	20	6	0	160000	10	40040	40040	40040	40040	40040
160024	40039	299	0	0	46	25	160010	10	160000	10	160000	50	1280000	1	1	10	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	0	0	10022	13	6	1	12	16	2	1	2	4	12	40036	0	20	6	0	160000	10	40040	40040	40040	40040	40040

SQRSHRN2 (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