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

ANDS (register, lsl, 64-bit)

Test 1: uops

Code:

  ands x0, x0, x1, lsl #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)
100420302001200110005226510001000200020011000
100420302001200110005226510001000200020011000
100420302001200110005226510001000200020011000
100420302001200110005226510001000200020011000
100420302001200110005226510001000200020011000
100420302001200110005226510001000200020011000
100420302001200110005226510001000200020011000
100420302001200110005226510001000200020011000
100420302001200110005226510001000200020011000
100420302001200110005226510001000200020011000

Test 2: Latency 1->2

Code:

  ands x0, x0, x1, lsl #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)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)
10205200602011720117101395290441010410206202122000110100
10204200302010120101101045290871010410208202162000110100
10204200302010120101101045290871010410208202162000110100
10204200302010120101101045290871010410208202162000110100
10204200302010120101101045290871010410208202162000110100
10204200302010120101101045290871010410208202162000110100
10204200302010120101101045290871010410208202162000110100
10204200302010120101101045290871010410208202862001610100
10204200302010120101101045290871010410208202162000110100
10204200302010120101101045290871010410208202162000110100

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)
10024200302002120021100255292151002510030200202001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010
10025200602003520035100585292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020203262009510010

Test 3: Latency 1->3

Code:

  ands x0, x1, x0, lsl #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)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)
10204200302010120101101045290441010410206202160200010010100
10204200302010120101101045290441010410206202120200010010100
10204200302010120101101045290871010410208202160200010010100
10204200302010120101101045290871010410208202160200010010100
10204200302010120101101045290871010410208202160200010010100
1020420030201012010110104529087101041020821751109020655596311030
10204200302010120101101045290871010410208202160200010010100
10204200302010120101101045290871010410208202160200010010100
10204200302010120101101045290871010410208202160200010010100
10204200302010120101101045290871010410208202160200010010100

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)
10024200302002120021100255292151002510030200402001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010
10024200302002120021100205292491002010020200202001110010

Test 4: Latency 4->2

Chain cycles: 1

Code:

  ands x0, x1, x2, lsl #17
  cset x1, cc
  mov x0, 1
  mov x1, 2
  mov x2, 3
  mov x3, 4
  mov x4, 5

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code, minus 1 chain cycle): 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)
20204300303010130101201057892332010520212302153000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057893692010520212302183000120100

1000 unrolls and 10 iterations

Result (median cycles for code, minus 1 chain cycle): 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)
20024300303001130011200157894012001520032300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010

Test 5: Latency 4->3

Chain cycles: 1

Code:

  ands x0, x1, x2, lsl #17
  cset x2, cc
  mov x0, 1
  mov x1, 2
  mov x2, 3
  mov x3, 4
  mov x4, 5

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code, minus 1 chain cycle): 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)
20204300303010130101201067893112010520212302213000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057893692010520212302183000120100
20204300303010130101201057896102014020254302183000120100
20204300303010130101201057896102014020254302183000120100
20204300303010130101201057893692010520212302183000120100

1000 unrolls and 10 iterations

Result (median cycles for code, minus 1 chain cycle): 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)
20024300303001130011200157893782001520032300383000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010
20024300303001130011200107894382001020020300203000120010

Test 6: throughput

Count: 8

Code:

  ands x0, x8, x9, lsl #17
  ands x1, x8, x9, lsl #17
  ands x2, x8, x9, lsl #17
  ands x3, x8, x9, lsl #17
  ands x4, x8, x9, lsl #17
  ands x5, x8, x9, lsl #17
  ands x6, x8, x9, lsl #17
  ands x7, x8, x9, lsl #17
  mov x8, 9
  mov x9, 10
  mov x10, 11

(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)
8020453404160114160114801231099833801278022816024816001480100
8020453417160114160114801231100076801238022416024816001480100
8020453404160114160114801231100076801238022416024816001480100
8020453404160114160114801231100076801238022416024816001480100
8020453404160114160114801231100076801238022416024816001480100
8020553436160162160162801641100076801238022416024816001480100
8020453404160114160114801231100076801238022416024816001480100
8020453404160114160114801231100076801238022416024816001480100
8020453404160114160114801231100076801238022416024816001480100
8020453404160114160114801231100076801238022416024816001480100

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)
8002453384160040160040800461107564800208002016002016001180010
8002453371160021160021800201107732800208002016002016001180010
8002553410160084160084800881107732800208002016002016001180010
8002453371160021160021800201107732800208002016002016001180010
8002453371160021160021800201107732800208002016002016001180010
8002453371160021160021800201114275800848008416002016001180010
8002453371160021160021800201107732800208002016002016001180010
8002453371160021160021800201107732800208002016002016001180010
8002453371160021160021800201107732800208002016002016001180010
8002453371160021160021800201107732800208002016002016001180010