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
Code:
ld1 { v0.16b, v1.16b, v2.16b }, [x6]
nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nop ; nopmov x0, 1 mov x1, 2 mov x8, 0
(no loop instructions)
Retires (minus 60 nops): 3.000
Issues: 3.000
Integer unit issues: 0.001
Load/store unit issues: 3.000
SIMD/FP unit issues: 0.000
| retire uop (01) | cycle (02) | schedule uop (52) | schedule int uop (53) | schedule ldst uop (55) | dispatch ldst uop (58) | simd uops in schedulers (5a) | dispatch uop (78) | map ldst uop (7d) | map ldst uop inputs (80) | ? int output thing (e9) | ? ldst retires (ed) |
| 63005 | 29665 | 3007 | 1 | 3006 | 3000 | 9000 | 3000 | 3000 | 3000 | 1 | 3000 |
| 63004 | 29406 | 3001 | 1 | 3000 | 3000 | 9000 | 3000 | 3000 | 3000 | 1 | 3000 |
| 63004 | 29345 | 3001 | 1 | 3000 | 3000 | 9000 | 3000 | 3000 | 3000 | 1 | 3000 |
| 63004 | 29352 | 3001 | 1 | 3000 | 3000 | 9000 | 3000 | 3000 | 3000 | 1 | 3000 |
| 63004 | 29341 | 3001 | 1 | 3000 | 3000 | 9000 | 3000 | 3000 | 3000 | 1 | 3000 |
| 63004 | 29345 | 3001 | 1 | 3000 | 3000 | 9000 | 3000 | 3000 | 3000 | 1 | 3000 |
| 63004 | 29345 | 3001 | 1 | 3000 | 3000 | 9000 | 3000 | 3000 | 3000 | 1 | 3000 |
| 63004 | 29349 | 3001 | 1 | 3000 | 3000 | 9000 | 3000 | 3000 | 3000 | 1 | 3000 |
| 63004 | 29345 | 3001 | 1 | 3000 | 3000 | 9000 | 3000 | 3000 | 3000 | 1 | 3000 |
| 63004 | 29347 | 3001 | 1 | 3000 | 3000 | 9000 | 3000 | 3000 | 3000 | 1 | 3000 |
Count: 8
Code:
ld1 { v0.16b, v1.16b, v2.16b }, [x6]
ld1 { v0.16b, v1.16b, v2.16b }, [x6]
ld1 { v0.16b, v1.16b, v2.16b }, [x6]
ld1 { v0.16b, v1.16b, v2.16b }, [x6]
ld1 { v0.16b, v1.16b, v2.16b }, [x6]
ld1 { v0.16b, v1.16b, v2.16b }, [x6]
ld1 { v0.16b, v1.16b, v2.16b }, [x6]
ld1 { v0.16b, v1.16b, v2.16b }, [x6]mov x7, x6 mov x8, x6 mov x9, x6 mov x10, x6 mov x11, x6 mov x12, x6 mov x13, x6
(fused SUBS/B.cc loop)
Result (median cycles for code divided by count): 1.5007
| retire uop (01) | cycle (02) | schedule uop (52) | schedule int uop (53) | schedule ldst uop (55) | dispatch int uop (56) | dispatch ldst uop (58) | int uops in schedulers (59) | simd uops in schedulers (5a) | dispatch uop (78) | map int uop (7c) | map ldst uop (7d) | map int uop inputs (7f) | map ldst uop inputs (80) | ? int output thing (e9) | ? ldst retires (ed) | ? int retires (ef) |
| 240206 | 120558 | 240161 | 101 | 240060 | 100 | 240008 | 300 | 768188 | 240108 | 200 | 240012 | 200 | 240012 | 1 | 240000 | 100 |
| 240205 | 120107 | 240138 | 101 | 240037 | 100 | 240057 | 300 | 1924571 | 240157 | 200 | 240068 | 200 | 240012 | 1 | 240000 | 100 |
| 240204 | 120053 | 240101 | 101 | 240000 | 100 | 240008 | 300 | 1920194 | 240108 | 200 | 240012 | 200 | 240012 | 1 | 240000 | 100 |
| 240204 | 120053 | 240101 | 101 | 240000 | 100 | 240008 | 300 | 1920194 | 240108 | 200 | 240012 | 200 | 240012 | 1 | 240000 | 100 |
| 240204 | 120053 | 240101 | 101 | 240000 | 100 | 240056 | 300 | 1285814 | 240156 | 200 | 240068 | 200 | 240072 | 1 | 240000 | 100 |
| 240204 | 120053 | 240101 | 101 | 240000 | 100 | 240057 | 300 | 1542521 | 240157 | 200 | 240068 | 200 | 240012 | 1 | 240000 | 100 |
| 240204 | 120053 | 240101 | 101 | 240000 | 100 | 240008 | 300 | 1920194 | 240108 | 200 | 240012 | 200 | 240012 | 1 | 240000 | 100 |
| 240205 | 120399 | 240131 | 101 | 240030 | 100 | 240008 | 300 | 1920194 | 240108 | 200 | 240012 | 200 | 240012 | 1 | 240000 | 100 |
| 240204 | 120057 | 240101 | 101 | 240000 | 100 | 240008 | 300 | 1920374 | 240108 | 200 | 240012 | 200 | 240012 | 1 | 240000 | 100 |
| 240204 | 120064 | 240101 | 101 | 240000 | 100 | 240008 | 300 | 1920410 | 240108 | 200 | 240012 | 200 | 240012 | 1 | 240000 | 100 |
Result (median cycles for code divided by count): 1.5007
| retire uop (01) | cycle (02) | schedule uop (52) | schedule int uop (53) | schedule ldst uop (55) | dispatch int uop (56) | dispatch ldst uop (58) | int uops in schedulers (59) | simd uops in schedulers (5a) | dispatch uop (78) | map int uop (7c) | map ldst uop (7d) | map int uop inputs (7f) | map ldst uop inputs (80) | ? int output thing (e9) | ? ldst retires (ed) | ? int retires (ef) |
| 240025 | 120190 | 240041 | 11 | 240030 | 10 | 240056 | 30 | 1200512 | 240066 | 20 | 240068 | 20 | 240012 | 1 | 240000 | 10 |
| 240024 | 120051 | 240011 | 11 | 240000 | 10 | 240000 | 30 | 1920164 | 240010 | 20 | 240000 | 20 | 240000 | 1 | 240000 | 10 |
| 240024 | 120051 | 240011 | 11 | 240000 | 10 | 240000 | 30 | 1920164 | 240010 | 20 | 240000 | 20 | 240069 | 1 | 240000 | 10 |
| 240024 | 120085 | 240011 | 11 | 240000 | 10 | 240056 | 30 | 942812 | 240066 | 20 | 240068 | 20 | 240000 | 1 | 240000 | 10 |
| 240024 | 120401 | 240011 | 11 | 240000 | 10 | 240000 | 30 | 1920326 | 240010 | 20 | 240000 | 20 | 240000 | 1 | 240000 | 10 |
| 240024 | 120051 | 240011 | 11 | 240000 | 10 | 240000 | 30 | 1920164 | 240010 | 20 | 240000 | 20 | 240000 | 1 | 240000 | 10 |
| 240024 | 120051 | 240011 | 11 | 240000 | 10 | 240000 | 30 | 1920164 | 240010 | 20 | 240000 | 20 | 240000 | 1 | 240000 | 10 |
| 240024 | 120051 | 240011 | 11 | 240000 | 10 | 240000 | 30 | 1920164 | 240010 | 20 | 240000 | 20 | 240000 | 1 | 240000 | 10 |
| 240024 | 120091 | 240011 | 11 | 240000 | 10 | 240000 | 30 | 1920434 | 240010 | 20 | 240000 | 20 | 240069 | 1 | 240000 | 10 |
| 240024 | 120066 | 240011 | 11 | 240000 | 10 | 240008 | 30 | 1920410 | 240018 | 20 | 240012 | 20 | 240000 | 1 | 240000 | 10 |