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

FRSQRTS (vector, 4S)

Test 1: uops

Code:

  frsqrts v0.4s, v0.4s, v1.4s
  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)dispatch ldst uop (58)int uops in schedulers (59)simd uops in schedulers (5a)ldst uops in schedulers (5b)dispatch uop (78)map int uop (7c)map ldst uop (7d)map simd uop (7e)map simd uop inputs (81)? int output thing (e9)? simd retires (ee)
1004403310011100010000001015571000001000200011000
1004403310011100010000001015571000001000200011000
1004403310011100010000001015571000001000200011000
1004403310011100010000001015571000001000200011000
1004403310011100010000001015571000001000200011000
1004403310011100010000001015571000001000200011000
1004403310011100010000001015571000001000200011000
1004403310011100010000001015571000001000200011000
1004403310011100010000001015571000001000200011000
1004403310011100010000001015571000001000200011000

Test 2: Latency 1->2

Code:

  frsqrts v0.4s, v0.4s, v1.4s
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0033

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)
10204400331010110110000100100003001028557101002001000620020012110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0033

retire uop (01)cycle (02)schedule uop (52)schedule int uop (53)schedule simd uop (54)dispatch int uop (56)dispatch simd uop (57)dispatch ldst uop (58)int uops in schedulers (59)simd uops in schedulers (5a)ldst uops in schedulers (5b)dispatch uop (78)map int uop (7c)map ldst uop (7d)map simd uop (7e)map int uop inputs (7f)map simd uop inputs (81)? int output thing (e9)? simd retires (ee)? int retires (ef)
100244003310021211000020100000700102855710020200100002020000111000010
100244003310021211000020100000700102855710020200100002020000111000010
100244003310021211000020100000700102855710020200100002020000111000010
100244003310021211000020100000700102855710020200100002020000111000010
100244003310021211000020100000700102855710020200100002020000111000010
100244003310021211000020100000700102855710020200100002020000111000010
100244003310021211000020100000700102855710020200100002020000111000010
100244003310021211000020100000700102855710020200100002020000111000010
100244003310021211000020100000700102855710020200100002020000111000010
100244003310021211000020100000700102855710020200100002020000111000010

Test 3: Latency 1->3

Code:

  frsqrts v0.4s, v1.4s, v0.4s
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0033

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)
10204400331010110110000100100003001028557101002001000620020012110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028904101302001004420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100
10204400331010110110000100100003001028557101002001000420020008110000100

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0033

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)
100244003310021211000020100007010285571002020100002020000111000010
100244003310021211000020100007010285571002020100002020000111000010
100244003310021211000020100007010285571002020100002020000111000010
100244003310021211000020100007010285571002020100002020000111000010
100244003310021211000020100007010285571002020100002020000111000010
100244003310021211000020100007010285571002020100002020000111000010
100244003310021211000020100007010285571002020100002020000111000010
100244003310021211000020100007010285571002020100002020000111000010
100244003310021211000020100007010285571002020100002020000111000010
100244003310021211000020100007010285571002020100002020000111000010

Test 4: throughput

Count: 8

Code:

  frsqrts v0.4s, v8.4s, v9.4s
  frsqrts v1.4s, v8.4s, v9.4s
  frsqrts v2.4s, v8.4s, v9.4s
  frsqrts v3.4s, v8.4s, v9.4s
  frsqrts v4.4s, v8.4s, v9.4s
  frsqrts v5.4s, v8.4s, v9.4s
  frsqrts v6.4s, v8.4s, v9.4s
  frsqrts v7.4s, v8.4s, v9.4s
  movi v8.16b, 9
  movi v9.16b, 10

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.5004

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)
80204400888010710180006100800103003200448011020080014200160028180000100
80205400738014510180044100800563003200368010820080012200160024180000100
80204400368010510180004100800083003200368010820080012200160024180000100
80204400368010510180004100800083003200368010820080012200160024180000100
80204400368010510180004100800083003200368010820080012200160024180000100
80204400368010510180004100800083003200368010820080012200160024180000100
80204400368010510180004100800083003200368010820080012200160024180000100
80204400368010510180004100800083003200368010820080012200160024180000100
80204400368010510180004100800083003200368010820080012200160024180000100
80204400368010510180004100800083003200368010820080012200160024180000100

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.5004

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)dispatch ldst uop (58)int uops in schedulers (59)simd uops in schedulers (5a)ldst uops in schedulers (5b)dispatch uop (78)map int uop (7c)map ldst uop (7d)map simd uop (7e)map int uop inputs (7f)map simd uop inputs (81)? int output thing (e9)? simd retires (ee)? int retires (ef)
8002440137800272180006020800100680320052800322008001820160036118000010
8002440042800212180000020800001332684981752332395183419264316368010420160000118000010
8002440036800212180000020800000650320000800202008000020160000118000010
8002440036800212180000020800000650320000800202008000020160000118000010
8002440036800212180000020800000650320000800202008000020160000118000010
8002440036800212180000020800000650320000800202008000020160000118000010
8002440036800212180000020800000650320000800202008000020160000118000010
8002440036800212180000020800000650320000800202008000020160000118000010
8002440036800212180000020800000650320000800202008000020160000118000010
8002440036800212180000020800000650320000800202008000020160000118000010