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

UQRSHRN2 (2D)

Test 1: uops

Code:

  uqrshrn2 v0.4s, v1.2d, #3
  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)
1004303310011100010007590510001000200011000
1004303310011100010007590510001000200011000
1004303310011100010007590510001000200011000
1004303310011100010007590510001000200011000
1004303310011100010007590510001000200011000
1004303310011100010007590510001000200011000
1004303310011100010007590510001000200011000
1004303310011100010007590510001000200011000
1004303310011100010007590510001000200011000
1004303310011100010007590510001000200011000

Test 2: Latency 1->1

Code:

  uqrshrn2 v0.4s, v1.2d, #3
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0033

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)
1020430033101011011000010010000300768905101002001000620020012110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430136101191031001610210066300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0033

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)
10024300331002121100002010000707689051002020100002020000111000010
10024300331002121100002010000707689051002020100002020000111000010
10024300331002121100002010000707689051002020100002020000111000010
10024300331002121100002010000707689051002020100002020000111000010
10024300331002121100002010000707689051002020100002020000111000010
10024300331002121100002010000707689051002020100002020000111000010
10024300331002121100002010000707689051002020100002020000111000010
10024300331002121100002010000707689051002020100002020000111000010
10024300331002121100002010000707689051002020100002020000111000010
10024300331002121100002010000707689051002020100002020000111000010

Test 3: Latency 1->2

Code:

  uqrshrn2 v0.4s, v0.2d, #3
  movi v0.16b, 1

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0033

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)
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100
1020430033101011011000010010000300768905101002001000420020008110000100

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0033

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 ldst uop inputs (80)map simd uop inputs (81)? int output thing (e9)? ldst retires (ed)? simd retires (ee)? int retires (ef)
1002430033100212110000201000070768905100202010000200200001101000010
1002430033100212110000201000070768905100202010000200200001101000010
1002430033100212110000201000070768905100202010000200200001101000010
1002430033100212110000201000070768905100202010000200200001101000010
1002430033100212110000201000070768905100202010000200200001101000010
1002430033100212110000201000070768905100202010000200200001101000010
1002430033100212110000201000070768905100202010000200200001101000010
1002430033100212110000201000070768905100202010000200200001101000010
1002430033100212110000201000070768905100202010000200200001101000010
1002430033100212110000201000070768905100202010000200200001101000010

Test 4: throughput

Count: 8

Code:

  movi v0.16b, 0
  uqrshrn2 v0.4s, v8.2d, #3
  movi v1.16b, 0
  uqrshrn2 v1.4s, v8.2d, #3
  movi v2.16b, 0
  uqrshrn2 v2.4s, v8.2d, #3
  movi v3.16b, 0
  uqrshrn2 v3.4s, v8.2d, #3
  movi v4.16b, 0
  uqrshrn2 v4.4s, v8.2d, #3
  movi v5.16b, 0
  uqrshrn2 v5.4s, v8.2d, #3
  movi v6.16b, 0
  uqrshrn2 v6.4s, v8.2d, #3
  movi v7.16b, 0
  uqrshrn2 v7.4s, v8.2d, #3
  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.5011

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)
1602044046780109101800081008001230032005680113200800132001600241160000100
1602044010880110101800091008001330032005280112200800122001600241160000100
1602044008680109101800081008001230032005680113200800132001600261160000100
1602044009680110101800091008001330032005280112200800122001600241160000100
1602044008680109101800081008001230032005280112200800122001600241160000100
1602044008680109101800081008001230032005280112200800122001600241160000100
1602044008680109101800081008001230032005280112200800122001600241160000100
1602044008680109101800081008001230032005280112200800122001600241160000100
1602044008680109101800081008001230032005280112200800122001600241160000100
1602044008680109101800081008001230032005280112200800122001600241160000100

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.5052

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)
1600244390280019118000810800123032005280022208001220160000116000010
1600254216480046118003510800353032000080010208000020160000116000010
1600244041880011118000010800003032000080010208000020160000116000010
1600244039280011118000010800003032000080010208000020160000116000010
1600244041680011118000010800003032000080010208000020160000116000010
1600244045480011118000010800003032000080010208000020160000116000010
1600244040080011118000010800003032000080010208000020160000116000010
1600244041680011118000010800003032000080010208000020160000116000010
1600244041680011118000010800003032000080010208000020160000116000010
1600244041680011118000010800003032000080010208000020160000116000010

Test 5: throughput

Count: 16

Code:

  uqrshrn2 v0.4s, v16.2d, #3
  uqrshrn2 v1.4s, v16.2d, #3
  uqrshrn2 v2.4s, v16.2d, #3
  uqrshrn2 v3.4s, v16.2d, #3
  uqrshrn2 v4.4s, v16.2d, #3
  uqrshrn2 v5.4s, v16.2d, #3
  uqrshrn2 v6.4s, v16.2d, #3
  uqrshrn2 v7.4s, v16.2d, #3
  uqrshrn2 v8.4s, v16.2d, #3
  uqrshrn2 v9.4s, v16.2d, #3
  uqrshrn2 v10.4s, v16.2d, #3
  uqrshrn2 v11.4s, v16.2d, #3
  uqrshrn2 v12.4s, v16.2d, #3
  uqrshrn2 v13.4s, v16.2d, #3
  uqrshrn2 v14.4s, v16.2d, #3
  uqrshrn2 v15.4s, v16.2d, #3
  movi v16.16b, 17

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.5002

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 ldst uop inputs (80)map simd uop inputs (81)? int output thing (e9)? ldst retires (ed)? simd retires (ee)? int retires (ef)
16020480086160105101160004100160008300640036160108200160012200032002410160000100
16020480035160107101160006100160010300640036160108200160012200032002410160000100
16020480035160105101160004100160008300640036160108200160012200032002410160000100
16020480035160105101160004100160008300640036160108200160012200032002410160000100
16020480035160105101160004100160008300640036160108200160012200032002410160000100
16020480035160105101160004100160008300640036160108200160012200032002410160000100
16020480035160105101160004100160008300640036160108200160012200032002410160000100
16020480035160105101160004100160008300640036160108200160012200032002410160000100
16020480035160105101160004100160008300640036160108200160012200032002410160000100
16020480035160105101160004100160008300640036160108200160012200032002410160000100

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.5002

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)
160024801161600171116000610160010306400361600182016001220320000116000010
160024800521600111116000010160000306400001600102016000020320000116000010
160024800351600111116000010160000306400001600102016000020320000116000010
160024800351600111116000010160000306400001600102016000020320000116000010
160024800351600111116000010160000306400001600102016000020320000116000010
160024800351600111116000010160000306400001600102016000020320000116000010
160024800351600111116000010160000306400001600102016000020320000116000010
160024800351600111116000010160000306400001600102016000020320000116000010
160024800351600111116000010160000306400001600102016000020320000116000010
160024800351600111116000010160000306400001600102016000020320000116000010