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

FRSQRTE (scalar, H)

Test 1: uops

Code:

  frsqrte h0, h0
  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)
1004303310011100010003769410001000100011000
1004303310011100010003769410001000100011000
1004303310011100010003769410001000100011000
1004303310011100010003769410001000100011000
1004303310011100010003769410001000100011000
1004303310011100010003769410001000100011000
1004303310011100010003769410001000100011000
1004303310011100010003769410001000100011000
1004303310011100010003769410001000100011000
1004303310011100010003769410001000100011000

Test 2: Latency 1->2

Code:

  frsqrte h0, h0
  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 ldst uop inputs (80)map simd uop inputs (81)? int output thing (e9)? ldst retires (ed)? simd retires (ee)? int retires (ef)
102043003310101101100001001000030037969410100200100042000100061010000100
102043003310101101100001001000030037969410100200100042000100041010000100
102043003310101101100001001000030037969410100200100042000100041010000100
102043003310101101100001001000030037969410100200100042000100041010000100
102043003310101101100001001000030037969410100200100042000100041010000100
102043003310101101100001001000030037969410100200100042000100061010000100
102043003310101101100001001000030037969410100200100042000100041010000100
102043003310101101100001001000030037969410100200100042000100041010000100
102043003310101101100001001000030037969410100200100042000100041010000100
102043003310101101100001001000030037969410100200100042000100041010000100

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)
10024300331002121100002010000703796941002020100002010000111000010
10024300331002121100002010000703796941002020100002010000111000010
10024300331002121100002010000703796941002020100002010000111000010
10024300331002121100002010000693798501003820100322010000111000010
10024300331002121100002010000703796941002020100002010000111000010
10024300331002121100002010000703796941002020100002010000111000010
10024300331002121100002010000703796941002020100002010000111000010
10024300331002121100002010000703796941002020100002010000111000010
10024300331002121100002010000703796941002020100002010000111000010
10024300331002121100002010000703796941002020100002010000111000010

Test 3: throughput

Count: 8

Code:

  frsqrte h0, h8
  frsqrte h1, h8
  frsqrte h2, h8
  frsqrte h3, h8
  frsqrte h4, h8
  frsqrte h5, h8
  frsqrte h6, h8
  frsqrte h7, h8
  movi v8.16b, 9

(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)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)
8020480049801011018000010080001300320005801012008000820080008180000100
8020480035801011018000010080001300320005801012008000820080008180000100
8020580070801191018001810080024300320005801012008000820080008180000100
8020480035801011018000010080001300320005801012008000820080008180000100
8020480035801011018000010080001300320005801012008000820080008180000100
8020480035801011018000010080001300320005801012008000820080036180000100
8020480035801011018000010080001300320005801012008000820080008180000100
8020480035801011018000010080001300320005801012008000820080008180000100
8020480035801011018000010080001300320005801012008000820080008180000100
8020580070801211018002010080026300320005801012008000820080008180000100

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)schedule simd uop (54)schedule ldst uop (55)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)
800248003580023218000202080004703200108002220800082080000118000010
800248003580021218000002080000703200008002020800002080000118000010
800248003580021218000002080000703200008002020800002080000118000010
800248003580021218000002080000703200008002020800002080000118000010
800248003580021218000002080000703200008002020800002080000118000010
800248003580021218000002080000703200928004420800362080000118000010
800248003580021218000002080000703200008002020800002080000118000010
800248003580021218000002080000703200008002020800002080000118000010
800248003580021218000002080000703200008002020800002080000118000010
800248003580021218000002080000703200008002020800002080000118000010