Apple Microarchitecture Research by Dougall Johnson

M1/A14 P-core (Firestorm): Overview | Base Instructions | SIMD and FP Instructions
M1/A14 E-core (Icestorm):  Overview | Base Instructions | SIMD and FP Instructions

SCVTF (scalar, integer, H from H)

Test 1: uops

Code:

  scvtf h0, h0

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

Test 2: Latency 1->2

Code:

  scvtf h0, h0

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	10006	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	10004	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	10004	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	10004	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	10004	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	10004	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	10004	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	10004	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	10004	1	10000	100
10204	30033	10101	101	10000	100	10000	300	768905	10100	200	10004	200	10004	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	10000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	10000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	10000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	10000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	10039	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	10000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	10000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	10000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	10000	11	10000	10
10024	30033	10021	21	10000	20	10000	70	768905	10020	20	10000	20	10044	11	10000	10

Test 3: throughput

Count: 8

Code:

  scvtf h0, h8
  scvtf h1, h8
  scvtf h2, h8
  scvtf h3, h8
  scvtf h4, h8
  scvtf h5, h8
  scvtf h6, h8
  scvtf h7, h8

Initialization

  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.5004

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)
80204	40552	80415	102	80313	101	80321	314	328128	82135	204	82035	204	81828	3	80000	100
80205	40070	80141	101	80040	100	80052	300	320044	80110	200	80014	200	80012	1	80000	100
80204	40035	80105	101	80004	100	80008	300	320036	80108	200	80012	200	80012	1	80000	100
80204	40035	80105	101	80004	100	80008	300	320036	80108	200	80012	200	80012	1	80000	100
80204	40035	80105	101	80004	100	80008	300	320036	80108	200	80012	200	80012	1	80000	100
80204	40035	80105	101	80004	100	80008	300	320036	80108	200	80012	202	80064	1	80000	100
80204	40035	80105	101	80004	100	80008	300	320036	80108	200	80012	200	80012	1	80000	100
80204	40035	80105	101	80004	100	80008	300	320036	80108	200	80012	200	80060	1	80000	100
80204	40035	80105	101	80004	100	80008	300	320036	80108	200	80012	200	80012	1	80000	100
80204	40035	80105	101	80004	100	80008	300	320036	80108	200	80012	200	80062	1	80000	100

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.5004

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)
80024	40157	80027	21	80006	20	80010	70	320044	80030	20	80016	20	0	80064	11	0	80000	10
80024	40035	80021	21	80000	20	80000	70	320000	80020	20	80000	20	0	80000	11	0	80000	10
80024	40035	80021	21	80000	20	80000	70	320000	80020	20	80000	20	0	80000	11	0	80000	10
80024	40035	80021	21	80000	20	80000	70	320000	80020	20	80000	20	0	80000	11	0	80000	10
80025	40080	80065	21	80044	20	80056	70	320000	80020	20	80000	20	0	80000	11	0	80000	10
80024	40035	80021	21	80000	20	80000	70	320000	80020	20	80000	20	0	80000	11	0	80000	10
80024	40038	80021	21	80000	20	80000	70	320000	80020	20	80000	164	199	80065	91	88	80000	140
80024	40035	80021	21	80000	20	80000	70	320000	80020	20	80000	20	0	80000	11	0	80000	10
80024	40035	80021	21	80000	20	80000	70	320000	80020	20	80000	20	0	80000	11	0	80000	10
80024	40035	80021	21	80000	20	80000	70	320000	80020	20	80000	20	0	80000	11	0	80000	10