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

CSNEG (64-bit)

Test 1: uops

Code:

  csneg x0, x0, x1, hi
  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)
100410301001100110002560310001000300010011000
100410301001100110002560310001000300010011000
100410301001100110002560310001000300010011000
100410301001100110002560310001000300010011000
100410301001100110002560310001000300010011000
100410301001100110002560310001000300010011000
100410301001100110002560310001000300010011000
100410301001100110002560310001000300010011000
100410301001100110002560310001000300010011000
100410301001100110002560310001000300010011000

Test 2: Latency 1->2

Code:

  csneg x0, x0, x1, hi
  mov x0, 1
  mov x1, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 1.0030

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)
10204100301010110101101072593381010810214302361000110100
10204100301010110101101072595391010710212302361000110100
10204100301010110101101072595391010710212303881001610100
10204100301010110101101072595391010710212302361000110100
10204100301010110101101072595391010710212302361000110100
10204100301010110101101072595391010710212302361000110100
10204101801016810168102432595391010710212302361000110100
10204100301010110101101072595391010710212302361000110100
10204100301010110101101072595391010710212302361000110100
10204100301010110101101072595391010710212302361000110100

1000 unrolls and 10 iterations

Result (median cycles for code): 1.0030

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)
10024100301002110021100292594741002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010

Test 3: Latency 1->3

Code:

  csneg x0, x1, x0, hi
  mov x0, 1
  mov x1, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 1.0030

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)
10204100301010110101101082592931010710214302361000110100
10204100301010110101101072595391010710212302361000110100
10204100301010110101101072595391010710212302361000110100
10204100301010110101101072595391010710212302361000110100
10204100301010110101101072595391010710212302361000110100
10204100301010110101101072595391010710212302361000110100
10204100301010110101101072595391010710212302361000110100
10204100301010110101101072595391010710212302361000110100
10204100301010110101101072595391010710212302361000110100
10204100301010110101101072595391010710212302361000110100

1000 unrolls and 10 iterations

Result (median cycles for code): 1.0030

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)
10024100301002110021100282595281003010038300681001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010
10024100301002110021100202595911002010020300201001110010

Test 4: Latency 1->4

Chain cycles: 1

Code:

  csneg x0, x1, x2, hi
  tst x0, 1
  mov x0, 1
  mov x1, 2
  mov x2, 3

(non-fused SUB/CBNZ loop)

100 unrolls and 100 iterations

Result (median cycles for code, minus 1 chain cycle): 1.0030

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)
20204200302020120201202095193292020820216402322010110100
20204200302020120201202085194482020820216402322010110100
20204200302020120201202085194482020820216402322010110100
20204200302020120201202085194482020820216402322010110100
20204200302020120201202085194482020820216402322010110100
20204200302020120201202085194482020820216402322010110100
20204200302020120201202085194482020820216402322010110100
20204200302020120201202085194482020820216402322010110100
20204200302020120201202085194482020820216402322010110100
20204200302020120201202085194482020820216402322010110100

1000 unrolls and 10 iterations

Result (median cycles for code, minus 1 chain cycle): 1.0030

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)
20024200302002120021200285194722002020020400202001110010
20024200302002120021200205195912002020020400202001110010
20024200302002120021200205195912002020020400202001110010
20024200302002120021200205195912002020020400202001110010
20024200302002120021200205195912002020020400202001110010
20024200302002120021200205195912002020020400202001110010
20024200302002120021200205195912002020020400202001110010
20024200302002120021200205195912002020020400202001110010
20024200302002120021200205195912002020020400202001110010
20024200302002120021200205195912002020020400202001110010

Test 5: throughput

Count: 8

Code:

  csneg x0, x8, x9, hi
  csneg x1, x8, x9, hi
  csneg x2, x8, x9, hi
  csneg x3, x8, x9, hi
  csneg x4, x8, x9, hi
  csneg x5, x8, x9, hi
  csneg x6, x8, x9, hi
  csneg x7, x8, x9, hi
  mov x8, 9
  mov x9, 10

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.3342

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)
802042689680115801158012029136780120802222404348005780100
802042675280115801158012029270280120802222402638001580100
802042673780115801158012029323680120802242402728001580100
802042673780115801158012029323680120802242402728001580100
802042673780115801158012029323680120802242402728001580100
802042673780115801158012029323680120802242402728001580100
802042673780115801158012029323680120802242402728001580100
802042673780115801158012029323680120802242402728001580100
802042673780115801158012029323680120802242402728001580100
802042673780115801158012029323680120802242402728001580100

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.3340

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)
800252813580057800578007330941680020800202400208001180010
800242677480021800218002038943780020800202400208001180010
800242671980021800218002042678080020800202400208001180010
800242672480021800218002042678080020800202400208001180010
800242671980021800218002042678080020800202400208001180010
800242671980021800218002042678080020800202400208001180010
800242671980021800218002042678080020800202400208001180010
800242674380021800218002042678080020800202400208001180010
800242671980021800218002038943780020800202400208001180010
800242671980021800218002042678080020800202400208001180010