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

MVN (register, ror, 64-bit)

Test 1: uops

Code:

  mvn x0, x0, ror #17
  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

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)
100420302001200110005226510001000100020011000
100420302001200110005226510001000100020011000
100420302001200110005226510001000100020011000
100420302001200110005226510001000100020011000
100420302001200110005226510001000100020011000
100420302001200110005226510001000100020011000
100420302001200110005226510001000100020011000
100420302001200110005226510001000100020011000
100420302001200110005226510001000100020011000
100420302001200110005226510001000100020011000

Test 2: Latency 1->2

Code:

  mvn x0, x0, ror #17
  mov x0, 1
  mov x1, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 2.0030

retire uop (01)cycle (02)schedule uop (52)schedule int uop (53)schedule simd uop (54)schedule ldst uop (55)dispatch int uop (56)dispatch simd uop (57)int uops in schedulers (59)dispatch uop (78)map int uop (7c)map int uop inputs (7f)? int output thing (e9)? int retires (ef)
10204200302010120101001010405291431010410210102102000110100
10204200302010120101001010405291861010410212102122000110100
10204200302010120101001010405291861010410212102122000110100
10204200302010120101001010405291861010410212102122000110100
10204200302010120101001010405291861010410212102122000110100
10204200302010120101001010405291861010410212102122000110100
10204200302010120101001010405291861010410212102122000110100
10204200302010120101001010405291861010410212102122000110100
10204200302010120101001010405291861010410212102122000110100
10204200302010120101001010405291861010410212102122000110100

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0030

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)
10024200302002120021100255292531002010020100202001110010
10024200302002120021100205292531002010020100202001110010
10024200302002120021100205292531002010020100202001110010
10024200302002120021100205292531002010020100202001110010
10024200302002120021100205292531002010020100202001110010
10024200302002120021100205292531002010020100202001110010
10024200302002120021100205292531002010020100202001110010
10024200302002120021100205292531002010020100202001110010
10024200302002120021100205292531002010020100202001110010
10024200302002120021100205292531002010020100732002510010

Test 3: throughput

Count: 8

Code:

  mvn x0, x8, ror #17
  mvn x1, x8, ror #17
  mvn x2, x8, ror #17
  mvn x3, x8, ror #17
  mvn x4, x8, ror #17
  mvn x5, x8, ror #17
  mvn x6, x8, ror #17
  mvn x7, x8, ror #17
  mov x8, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

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

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)
802045340316011716011780130136030180130802368023616001780100
802045340416011716011780130136083880130802368028816005980100
802045340416011716011780130136083880130802368023616001780100
802045340416011716011780130136083880130802368023616001780100
802045340416011716011780130136083880130802368023616001780100
802045340416011716011780130136083880130802368023616001780100
802045340416011716011780130136083880130802368023616001780100
802045340416011716011780130136083880130802368023616001780100
802045340416011716011780130136083880130802368023616001780100
802045340416011716011780130136083880130802368023616001780100

1000 unrolls and 10 iterations

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

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)
800245340216003916003980051135997580051800568002016001180010
800245337116002116002180020135990380020800208002016001180010
800245337116002116002180020135990380020800208002016001180010
800245337116002116002180020135990380020800208002016001180010
800245337116002116002180020135990380020800208002016001180010
800245337116002116002180020135990380020800208002016001180010
800245337116002116002180020135990380020800208002016001180010
800255341016008516008580099135990380020800208002016001180010
800245337116002116002180020135990380020800208002016001180010
800245337116002116002180020135990380020800208002016001180010