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

FRSQRTE (vector, 4S)

Test 1: uops

Code:

  frsqrte v0.4s, v0.4s
  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

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)
1004403310011100010005060810001000100011000
1004403310011100010005060810001000100011000
1004403310011100010005060810001000100011000
1004403310011100010005060810001000100011000
1004403310011100010005060810001000100011000
1004403310011100010005060810001000100011000
1004403310011100010005060810001000100011000
1004403310011100010005060810001000100011000
1004403310011100010005060810001000100011000
1004403310011100010005060810001000100011000

Test 2: Latency 1->2

Code:

  frsqrte v0.4s, v0.4s
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0033

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)
1020440033101011011000010010000300509608101002001000620010006110000100
1020540066101051011000410010017300509608101002001000420010004110000100
1020440033101011011000010010000300509608101002001000420010004110000100
1020440033101011011000010010000300509608101002001000420210034110000100
1020440033101011011000010010000300509608101002001000420010004110000100
1020440033101011011000010010000300509608101002001000420010004110000100
1020440033101011011000010010000300509608101002001000420010004110000100
1020440033101011011000010010000300509608101002001000420010004110000100
1020440033101011011000010010000300509608101002001000420010004110000100
1020440033101011011000010010000300509608101002001000420010004110000100

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0033

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)
10024400331002121100002010000705096081002020100062010000111000010
10024400331002121100002010000705096081002020100002010000111000010
10024400331002121100002010000705096081002020100002010000111000010
10024400331002121100002010000705096081002020100002010000111000010
10025400661002521100042010017705096081002020100002010000111000010
10024400331002121100002010000705096081002020100002010000111000010
10024400331002121100002010000705096081002020100002010000111000010
10024400331002121100002010000705096081002020100002010000111000010
10024400331002121100002010000705096081002020100002010000111000010
10024400331002121100002010000705096081002020100002010000111000010

Test 3: throughput

Count: 8

Code:

  frsqrte v0.4s, v8.4s
  frsqrte v1.4s, v8.4s
  frsqrte v2.4s, v8.4s
  frsqrte v3.4s, v8.4s
  frsqrte v4.4s, v8.4s
  frsqrte v5.4s, v8.4s
  frsqrte v6.4s, v8.4s
  frsqrte v7.4s, v8.4s
  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 2.0004

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)
802041600358010110180000100800003001999769801002008000420080035180000100
802041600358010110180000100800003001999769801002008000420080004180000100
802041600358010110180000100800003001999769801002008000420080004180000100
802041600358010110180000100800003001999932801222008003620080004180000100
802051600708011010180009100800223001999769801002008000420080004180000100
802041600858011410180013100800243001999769801002008000420080004180000100
802041600358010110180000100800003002000287801482008006420080004180000100
802041600358010110180000100800003002000805801962008012620080036180000100
802041600358010110180000100800003002000287801482008006420080004180000100
802041600358010110180000100800003001999769801002008000420080064180000100

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 2.0004

retire uop (01)cycle (02)schedule uop (52)schedule int uop (53)schedule simd uop (54)dispatch int uop (56)dispatch simd uop (57)dispatch ldst uop (58)int uops in schedulers (59)simd uops in schedulers (5a)ldst uops in schedulers (5b)dispatch uop (78)map int uop (7c)map ldst uop (7d)map simd uop (7e)map int uop inputs (7f)map simd uop inputs (81)? int output thing (e9)? simd retires (ee)? int retires (ef)
8002416003580021218000020800000700199976980020200800042080000118000010
8002416003580021218000020800000700199976980020200800002080000118000010
8002416003580021218000020800000680199993280042200800362080000118000010
8002416003580021218000020800000700199976980020200800002080000118000010
8002416003580021218000020800000700199976980020200800002080000118000010
8002416003580021218000020800000700199976980020200800002080000118000010
8002416003580021218000020800000700199976980020200800002080029118000010
8002416003580021218000020800000700199976980020200800002080000118000010
8002416003580021218000020800000700199976980020200800002080000118000010
8002416003580021218000020800000700199976980020200800002080000118000010