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

SMNEGL

Test 1: uops

Code:

  smnegl x0, w0, w1
  mov x0, 1
  mov x1, 2

(no loop instructions)

1000 unrolls and 1 iteration

Retires: 1.000

Issues: 1.000

Integer unit issues: 1.001

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)
100430301001100110002585610001000200010011000
100430301001100110002585610001000200010011000
100430301001100110002585610001000200010011000
100430301001100110002585610001000200010011000
100430301001100110002585610001000200010011000
100430301001100110002585610001000200010011000
100430301001100110002585610001000200010011000
100430301001100110002585610001000200010011000
100430301001100110002585610001000200010011000
100430301001100110002585610001000200010011000

Test 2: Latency 1->2

Code:

  smnegl x0, w0, w1
  mov x0, 1
  mov x1, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0030

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)
10204300301010110101101002601501010010206202121000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0030

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)map ldst uop inputs (80)? int output thing (e9)? ldst retires (ed)? simd retires (ee)? int retires (ef)
10024300301002110021100202599161002010020200200100110010010
10024300301002110021100202599161002010020200200100110010010
10024300301002110021100202599161002010020200200100110010010
10024300301002110021100202599161002010020200200100110010010
10024300301002110021100202599161002010020200200100110010010
10024300301002110021100202599161002010020200200100110010010
10025300601002610026100342599161002010020200200100110010010
10024300301002110021100202599161002010020200200100110010010
10024300301002110021100202599161002010020200200100110010010
10024300301002110021100202599161002010020200200100110010010

Test 3: Latency 1->3

Code:

  smnegl x0, w1, w0
  mov x0, 1
  mov x1, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0030

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)
10204300301010110101101002601441010010206202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100
10204300301010110101101002601561010010208202161000110100

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0030

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)
10024300301002110021100202599071002010020200841001610010
10024300301002110021100202599161002010020200201001110010
10024300301002110021100202599161002010020200201001110010
10024300301002110021100202599161002010020200201001110010
10024300301002110021100202599161002010020200201001110010
10024300301002110021100202599161002010020200201001110010
10024300301002110021100202599161002010020200201001110010
10024300301002110021100202599161002010020201451003210010
10024300301002110021100202599161002010020200201001110010
10024300301002110021100202599161002010020200201001110010

Test 4: throughput

Count: 8

Code:

  smnegl x0, w8, w9
  smnegl x1, w8, w9
  smnegl x2, w8, w9
  smnegl x3, w8, w9
  smnegl x4, w8, w9
  smnegl x5, w8, w9
  smnegl x6, w8, w9
  smnegl x7, w8, w9
  mov x8, 9
  mov x9, 10

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 1.0004

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)
802048005480104801048010524031580105802101602208000480100
802048003480104801048010524031580105802121602248000480100
802048003480104801048010524031580105802121602248000480100
802048003480104801048010524031580105802121602808002080100
802048003480104801048010524031580105802121602248000480100
802048003480104801048010524031580105802121602248000480100
802048003480104801048010524031580105802121602248000480100
802048003480104801048010524031580105802121602248000480100
802048003480104801048010524031580105802121602248000480100
802048003480104801048010524031580105802121602248000480100

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 1.0004

retire uop (01)cycle (02)schedule uop (52)schedule int uop (53)dispatch int uop (56)dispatch ldst uop (58)int uops in schedulers (59)simd uops in schedulers (5a)dispatch uop (78)map int uop (7c)map ldst uop (7d)map int uop inputs (7f)? int output thing (e9)? int retires (ef)
800248005780025800258002602400780800268003401600208001180010
800248003280021800218002002400600800208002001600208001180010
800248003280021800218002002400600800208002001600208001180010
800248003280021800218002002400600800208002001600208001180010
800248003280021800218002002400600800208002001600208001180010
800248003280021800218002002400600800208002001600208001180010
800248003280021800218002002400600800208002001600208001180010
800248003280021800218002002400600800208002001600208001180010
800248003280021800218002002400600800208002001600208001180010
800248003280021800218002002400600800208002001600208001180010