RDMSR

Operands

• Operand 1 (w, suppressed): Register (EAX)
• Operand 2 (w, suppressed): Register (EDX)
• Operand 3 (r, suppressed): Register (ECX)
• Operand 4 (r, suppressed): Register (MSRS)

Available performance data

• Alder Lake-P
• Alder Lake-E
• Rocket Lake
• Tiger Lake
• Ice Lake
• Cascade Lake
• Cannon Lake
• Skylake-X
• Coffee Lake
• Kaby Lake
• Skylake
• Broadwell
• Haswell
• Ivy Bridge
• Sandy Bridge
• Westmere
• Nehalem
• Wolfdale
• Conroe
• Tremont
• Goldmont Plus
• Goldmont
• Airmont
• Bonnell
• AMD Zen 4
• AMD Zen 3
• AMD Zen 2
• AMD Zen+

Alder Lake-P

• Measurements
• Throughput
  • Computed from the port usage: 13.25
  • Measured (loop): 51.00
  • Measured (unrolled): 51.00
• Number of μops
  • Executed: 54
  • Retire slots: 62
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 62
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 1*p01+6*p015+1*p05+10*p056+20*p06+8*p1+5*p15+1*p15B+2*p5

Alder Lake-E

• Measurements
• Throughput
  • Measured (loop): 36.80
  • Measured (unrolled): 39.00
• Number of μops
  • Executed: 28
  • Microcode Sequencer (MS): 27
  • Requires the complex decoder

Rocket Lake

• Measurements
• Throughput
  • Computed from the port usage: 15.50
  • Measured (loop): 91.08
  • Measured (unrolled): 91.00
• Number of μops
  • Executed: 62
  • Retire slots: 71
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 71
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 1*p0+3*p01+10*p015+4*p0156+8*p05+22*p06+5*p1+5*p15+4*p5

Tiger Lake

• Measurements
• Throughput
  • Computed from the port usage: 11.50
  • Measured (loop): 44.00
  • Measured (unrolled): 44.00
• Number of μops
  • Executed: 46
  • Retire slots: 53
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 53
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 1*p01+14*p015+4*p0156+1*p05+18*p06+2*p1+4*p15+2*p5

Ice Lake

• Measurements
• Throughput
  • Computed from the port usage: 15.25
  • Measured (loop): 91.07
  • Measured (unrolled): 91.08
• Number of μops
  • Executed: 61
  • Retire slots: 70
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 70
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 3*p01+9*p015+3*p0156+10*p05+22*p06+5*p1+5*p15+4*p5

Cascade Lake

• Measurements
• Throughput
  • Computed from the port usage: 21.25
  • Measured (loop): 57.00
  • Measured (unrolled): 57.00
• Number of μops
  • Executed: 85
  • Retire slots: 87
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 87
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 1*p0+10*p01+15*p05+39*p06+5*p1+11*p15+4*p5

Cannon Lake

• Measurements
• Throughput
  • Computed from the port usage: 17.00
  • Measured (loop): 94.08
  • Measured (unrolled): 94.25
• Number of μops
  • Executed: 63
  • Retire slots: 74
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 74
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 1*p0+7*p01+13*p05+25*p06+5*p1+13*p15+4*p5

Skylake-X

• Measurements
• Throughput
  • Computed from the port usage: 21.50
  • Measured (loop): 89.08
  • Measured (unrolled): 89.29
• Number of μops
  • Executed: 84
  • Retire slots: 86
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 86
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 1*p0+9*p01+16*p05+37*p06+6*p1+12*p15+5*p5

Coffee Lake

• Measurements
• Throughput
  • Computed from the port usage: 21.50
  • Measured (loop): 92.08
  • Measured (unrolled): 92.28
• Number of μops
  • Executed: 80
  • Retire slots: 83
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 83
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 1*p0+10*p01+15*p05+37*p06+6*p1+12*p15+5*p5

Kaby Lake

• Measurements
• Throughput
  • Computed from the port usage: 20.50
  • Measured (loop): 91.08
  • Measured (unrolled): 91.05
• Number of μops
  • Executed: 80
  • Retire slots: 83
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 83
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 1*p0+9*p01+15*p05+35*p06+6*p1+11*p15+5*p5

Skylake

• Measurements
• Throughput
  • Computed from the port usage: 20.00
  • Measured (loop): 87.08
  • Measured (unrolled): 87.29
• Number of μops
  • Executed: 79
  • Retire slots: 81
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 81
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 1*p0+7*p01+6*p015+13*p05+33*p06+6*p1+9*p15+5*p5

Broadwell

• Measurements
• Throughput
  • Computed from the port usage: 19.50
  • Measured (loop): 91.30
  • Measured (unrolled): 91.14
• Number of μops
  • Executed: 72
  • Retire slots: 68
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 74
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 1*p0+4*p01+16*p015+38*p06+5*p1+5*p15+4*p5

Haswell

• Measurements
• Throughput
  • Computed from the port usage: 19.00
  • Measured (loop): 92.30
  • Measured (unrolled): 92.29
• Number of μops
  • Executed: 70
  • Retire slots: 65
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 71
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 1*p0+6*p01+14*p015+37*p06+5*p1+6*p15+3*p5

Ivy Bridge

• Measurements
• Throughput
  • Computed from the port usage: 39.00
  • Measured (loop): 108.63
  • Measured (unrolled): 107.50
• Number of μops
  • Executed: 81
  • Retire slots: 76
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 82
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 2*p0+20*p01+11*p05+5*p1+4*p15+39*p5

Sandy Bridge

• Measurements
• Throughput
  • Computed from the port usage: 36.00
  • Measured (loop): 105.33
  • Measured (unrolled): 104.50
• Number of μops
  • Executed: 77
  • Retire slots: 77
  • Decoded (MITE): 0
  • Microcode Sequencer (MS): 83
  • Requires the complex decoder (no other instruction can be decoded with simple decoders in the same cycle)
• Port usage: 2*p0+23*p01+12*p05+5*p1+6*p15+36*p5

Westmere

• Measurements
• Throughput
  • Computed from the port usage: 21.67
  • Measured (loop): 76.80
  • Measured (unrolled): 77.00
• Number of μops
  • Executed: 66
  • Retire slots: 66
  • Microcode Sequencer (MS): 210
  • Requires the complex decoder
• Port usage: 3*p0+21*p015+14*p05+8*p1+1*p2+19*p5

Nehalem

• Measurements
• Throughput
  • Computed from the port usage: 23.67
  • Measured (loop): 80.00
  • Measured (unrolled): 80.00
• Number of μops
  • Executed: 72
  • Retire slots: 72
  • Microcode Sequencer (MS): 240
  • Requires the complex decoder
• Port usage: 4*p0+26*p015+14*p05+6*p1+1*p2+21*p5

Wolfdale

• Measurements
• Throughput
  • Computed from the port usage: 39.00
  • Measured (loop): 166.60
  • Measured (unrolled): 168.00
• Number of μops
  • Executed: 76
• Port usage: 4*p0+9*p015+5*p05+18*p1+1*p2+39*p5

Conroe

• Measurements
• Throughput
  • Computed from the port usage: 37.00
  • Measured (loop): 155.90
  • Measured (unrolled): 154.00
• Number of μops
  • Executed: 83
• Port usage: 11*p0+6*p015+5*p05+23*p1+1*p2+37*p5

Tremont

• Measurements
• Throughput
  • Measured (loop): 72.30
  • Measured (unrolled): 72.50
• Number of μops
  • Executed: 61
  • Microcode Sequencer (MS): 61
  • Requires the complex decoder

Goldmont Plus

• Measurements
• Throughput
  • Measured (loop): 50.00
  • Measured (unrolled): 50.00
• Number of μops
  • Executed: 58
  • Microcode Sequencer (MS): 58
  • Requires the complex decoder

Goldmont

• Measurements
• Throughput
  • Measured (loop): 41.63
  • Measured (unrolled): 41.67
• Number of μops
  • Executed: 56
  • Microcode Sequencer (MS): 56
  • Requires the complex decoder

Airmont

• Measurements
• Throughput
  • Measured (loop): 69.00
  • Measured (unrolled): 69.00
• Number of μops
  • Executed: 57
  • Microcode Sequencer (MS): 57
  • Requires the complex decoder

Bonnell

• Measurements
• Throughput
  • Measured (loop): 90.00
  • Measured (unrolled): 90.00
• Number of μops
  • Executed: 55
  • Microcode Sequencer (MS): 55
  • Requires the complex decoder

AMD Zen 4

• Measurements
• Throughput
  • Measured (loop): 152.10
  • Measured (unrolled): 151.63
• Number of μops
  • Executed: 114

AMD Zen 3

• Measurements
• Throughput
  • Measured (loop): 172.70
  • Measured (unrolled): 172.00
• Number of μops
  • Executed: 120
• Documentation
• Number of μops: ucode

AMD Zen 2

• Measurements
• Throughput
  • Measured (loop): 149.60
  • Measured (unrolled): 149.00
• Number of μops
  • Executed: 98
• Documentation
• Number of μops: ucode

AMD Zen+

• Measurements
• Throughput
  • Measured (loop): 148.90
  • Measured (unrolled): 148.00
• Number of μops
  • Executed: 99
• Documentation
• Number of μops: ucode