PDEP (R64, R64, M64)

With unroll_count=500 and no inner loop

Code:

   0:	c4 42 b3 f5 06       	pdep   r8,r9,QWORD PTR [r14]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL

Show nanoBench command
Results:
- MPERF: 54.2
- APERF: 62.16
- UOPS: 119.0
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With loop_count=1000 and unroll_count=10

Code:

   0:	c4 42 b3 f5 06       	pdep   r8,r9,QWORD PTR [r14]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL

Show nanoBench command
Results:
- MPERF: 54.38
- APERF: 62.38
- UOPS: 119.2
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With loop_count=100 and unroll_count=100

Code:

   0:	c4 42 b3 f5 06       	pdep   r8,r9,QWORD PTR [r14]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL

Show nanoBench command
Results:
- MPERF: 54.21
- APERF: 62.18
- UOPS: 119.02
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With 4 independent instructions

With unroll_count=200 and no inner loop

Code:

   0:	c4 42 b3 f5 06       	pdep   r8,r9,QWORD PTR [r14]
   5:	c4 42 b3 f5 56 08    	pdep   r10,r9,QWORD PTR [r14+0x8]
   b:	c4 42 b3 f5 5e 10    	pdep   r11,r9,QWORD PTR [r14+0x10]
  11:	c4 42 b3 f5 66 18    	pdep   r12,r9,QWORD PTR [r14+0x18]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL

Show nanoBench command
Results:
- MPERF: 186.21
- APERF: 213.59
- UOPS: 412.0
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With loop_count=1000 and unroll_count=2

Code:

   0:	c4 42 b3 f5 06       	pdep   r8,r9,QWORD PTR [r14]
   5:	c4 42 b3 f5 56 08    	pdep   r10,r9,QWORD PTR [r14+0x8]
   b:	c4 42 b3 f5 5e 10    	pdep   r11,r9,QWORD PTR [r14+0x10]
  11:	c4 42 b3 f5 66 18    	pdep   r12,r9,QWORD PTR [r14+0x18]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL

Show nanoBench command
Results:
- MPERF: 187.53
- APERF: 215.11
- UOPS: 413.0
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With loop_count=100 and unroll_count=20

Code:

   0:	c4 42 b3 f5 06       	pdep   r8,r9,QWORD PTR [r14]
   5:	c4 42 b3 f5 56 08    	pdep   r10,r9,QWORD PTR [r14+0x8]
   b:	c4 42 b3 f5 5e 10    	pdep   r11,r9,QWORD PTR [r14+0x10]
  11:	c4 42 b3 f5 66 18    	pdep   r12,r9,QWORD PTR [r14+0x18]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL

Show nanoBench command
Results:
- MPERF: 187.66
- APERF: 215.26
- UOPS: 412.1
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With 7 independent instructions

With unroll_count=100 and no inner loop

Code:

   0:	c4 42 b3 f5 06       	pdep   r8,r9,QWORD PTR [r14]
   5:	c4 42 b3 f5 56 08    	pdep   r10,r9,QWORD PTR [r14+0x8]
   b:	c4 42 b3 f5 5e 10    	pdep   r11,r9,QWORD PTR [r14+0x10]
  11:	c4 42 b3 f5 66 18    	pdep   r12,r9,QWORD PTR [r14+0x18]
  17:	c4 c2 b3 f5 5e 20    	pdep   rbx,r9,QWORD PTR [r14+0x20]
  1d:	c4 c2 b3 f5 4e 28    	pdep   rcx,r9,QWORD PTR [r14+0x28]
  23:	c4 c2 b3 f5 56 30    	pdep   rdx,r9,QWORD PTR [r14+0x30]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL

Show nanoBench command
Results:
- MPERF: 275.51
- APERF: 316.0
- UOPS: 577.0
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With loop_count=1000 and unroll_count=1

Code:

   0:	c4 42 b3 f5 06       	pdep   r8,r9,QWORD PTR [r14]
   5:	c4 42 b3 f5 56 08    	pdep   r10,r9,QWORD PTR [r14+0x8]
   b:	c4 42 b3 f5 5e 10    	pdep   r11,r9,QWORD PTR [r14+0x10]
  11:	c4 42 b3 f5 66 18    	pdep   r12,r9,QWORD PTR [r14+0x18]
  17:	c4 c2 b3 f5 5e 20    	pdep   rbx,r9,QWORD PTR [r14+0x20]
  1d:	c4 c2 b3 f5 4e 28    	pdep   rcx,r9,QWORD PTR [r14+0x28]
  23:	c4 c2 b3 f5 56 30    	pdep   rdx,r9,QWORD PTR [r14+0x30]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL

Show nanoBench command
Results:
- MPERF: 278.97
- APERF: 320.01
- UOPS: 579.0
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With loop_count=100 and unroll_count=10

Code:

   0:	c4 42 b3 f5 06       	pdep   r8,r9,QWORD PTR [r14]
   5:	c4 42 b3 f5 56 08    	pdep   r10,r9,QWORD PTR [r14+0x8]
   b:	c4 42 b3 f5 5e 10    	pdep   r11,r9,QWORD PTR [r14+0x10]
  11:	c4 42 b3 f5 66 18    	pdep   r12,r9,QWORD PTR [r14+0x18]
  17:	c4 c2 b3 f5 5e 20    	pdep   rbx,r9,QWORD PTR [r14+0x20]
  1d:	c4 c2 b3 f5 4e 28    	pdep   rcx,r9,QWORD PTR [r14+0x28]
  23:	c4 c2 b3 f5 56 30    	pdep   rdx,r9,QWORD PTR [r14+0x30]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL

Show nanoBench command
Results:
- MPERF: 282.3
- APERF: 323.81
- UOPS: 577.2
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With an indexed addressing mode

With 1 independent instruction

With unroll_count=500 and no inner loop

Code:

   0:	c4 02 b3 f5 04 2e    	pdep   r8,r9,QWORD PTR [r14+r13*1]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL;
XOR R13, R13

Show nanoBench command
Results:
- MPERF: 54.2
- APERF: 62.16
- UOPS: 119.0
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With loop_count=1000 and unroll_count=10

Code:

   0:	c4 02 b3 f5 04 2e    	pdep   r8,r9,QWORD PTR [r14+r13*1]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL;
XOR R13, R13

Show nanoBench command
Results:
- MPERF: 54.38
- APERF: 62.38
- UOPS: 119.2
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With loop_count=100 and unroll_count=100

Code:

   0:	c4 02 b3 f5 04 2e    	pdep   r8,r9,QWORD PTR [r14+r13*1]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL;
XOR R13, R13

Show nanoBench command
Results:
- MPERF: 54.17
- APERF: 62.13
- UOPS: 119.02
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With 4 independent instructions

With unroll_count=200 and no inner loop

Code:

   0:	c4 02 b3 f5 04 2e    	pdep   r8,r9,QWORD PTR [r14+r13*1]
   6:	c4 02 b3 f5 54 2e 08 	pdep   r10,r9,QWORD PTR [r14+r13*1+0x8]
   d:	c4 02 b3 f5 5c 2e 10 	pdep   r11,r9,QWORD PTR [r14+r13*1+0x10]
  14:	c4 02 b3 f5 64 2e 18 	pdep   r12,r9,QWORD PTR [r14+r13*1+0x18]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL;
XOR R13, R13

Show nanoBench command
Results:
- MPERF: 188.36
- APERF: 216.13
- UOPS: 412.0
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With loop_count=1000 and unroll_count=2

Code:

   0:	c4 02 b3 f5 04 2e    	pdep   r8,r9,QWORD PTR [r14+r13*1]
   6:	c4 02 b3 f5 54 2e 08 	pdep   r10,r9,QWORD PTR [r14+r13*1+0x8]
   d:	c4 02 b3 f5 5c 2e 10 	pdep   r11,r9,QWORD PTR [r14+r13*1+0x10]
  14:	c4 02 b3 f5 64 2e 18 	pdep   r12,r9,QWORD PTR [r14+r13*1+0x18]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL;
XOR R13, R13

Show nanoBench command
Results:
- MPERF: 189.06
- APERF: 216.87
- UOPS: 413.0
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With loop_count=100 and unroll_count=20

Code:

   0:	c4 02 b3 f5 04 2e    	pdep   r8,r9,QWORD PTR [r14+r13*1]
   6:	c4 02 b3 f5 54 2e 08 	pdep   r10,r9,QWORD PTR [r14+r13*1+0x8]
   d:	c4 02 b3 f5 5c 2e 10 	pdep   r11,r9,QWORD PTR [r14+r13*1+0x10]
  14:	c4 02 b3 f5 64 2e 18 	pdep   r12,r9,QWORD PTR [r14+r13*1+0x18]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL;
XOR R13, R13

Show nanoBench command
Results:
- MPERF: 188.97
- APERF: 216.76
- UOPS: 412.1
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With 7 independent instructions

With unroll_count=100 and no inner loop

Code:

   0:	c4 02 b3 f5 04 2e    	pdep   r8,r9,QWORD PTR [r14+r13*1]
   6:	c4 02 b3 f5 54 2e 08 	pdep   r10,r9,QWORD PTR [r14+r13*1+0x8]
   d:	c4 02 b3 f5 5c 2e 10 	pdep   r11,r9,QWORD PTR [r14+r13*1+0x10]
  14:	c4 02 b3 f5 64 2e 18 	pdep   r12,r9,QWORD PTR [r14+r13*1+0x18]
  1b:	c4 82 b3 f5 5c 2e 20 	pdep   rbx,r9,QWORD PTR [r14+r13*1+0x20]
  22:	c4 82 b3 f5 4c 2e 28 	pdep   rcx,r9,QWORD PTR [r14+r13*1+0x28]
  29:	c4 82 b3 f5 54 2e 30 	pdep   rdx,r9,QWORD PTR [r14+r13*1+0x30]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL;
XOR R13, R13

Show nanoBench command
Results:
- MPERF: 284.58
- APERF: 326.18
- UOPS: 577.0
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With loop_count=1000 and unroll_count=1

Code:

   0:	c4 02 b3 f5 04 2e    	pdep   r8,r9,QWORD PTR [r14+r13*1]
   6:	c4 02 b3 f5 54 2e 08 	pdep   r10,r9,QWORD PTR [r14+r13*1+0x8]
   d:	c4 02 b3 f5 5c 2e 10 	pdep   r11,r9,QWORD PTR [r14+r13*1+0x10]
  14:	c4 02 b3 f5 64 2e 18 	pdep   r12,r9,QWORD PTR [r14+r13*1+0x18]
  1b:	c4 82 b3 f5 5c 2e 20 	pdep   rbx,r9,QWORD PTR [r14+r13*1+0x20]
  22:	c4 82 b3 f5 4c 2e 28 	pdep   rcx,r9,QWORD PTR [r14+r13*1+0x28]
  29:	c4 82 b3 f5 54 2e 30 	pdep   rdx,r9,QWORD PTR [r14+r13*1+0x30]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL;
XOR R13, R13

Show nanoBench command
Results:
- MPERF: 296.42
- APERF: 340.01
- UOPS: 579.0
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

With loop_count=100 and unroll_count=10

Code:

   0:	c4 02 b3 f5 04 2e    	pdep   r8,r9,QWORD PTR [r14+r13*1]
   6:	c4 02 b3 f5 54 2e 08 	pdep   r10,r9,QWORD PTR [r14+r13*1+0x8]
   d:	c4 02 b3 f5 5c 2e 10 	pdep   r11,r9,QWORD PTR [r14+r13*1+0x10]
  14:	c4 02 b3 f5 64 2e 18 	pdep   r12,r9,QWORD PTR [r14+r13*1+0x18]
  1b:	c4 82 b3 f5 5c 2e 20 	pdep   rbx,r9,QWORD PTR [r14+r13*1+0x20]
  22:	c4 82 b3 f5 4c 2e 28 	pdep   rcx,r9,QWORD PTR [r14+r13*1+0x28]
  29:	c4 82 b3 f5 54 2e 30 	pdep   rdx,r9,QWORD PTR [r14+r13*1+0x30]

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL;
XOR R13, R13

Show nanoBench command
Results:
- MPERF: 286.99
- APERF: 329.21
- UOPS: 577.2
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 0.0
- FpuPipeAssignment.Total2: 0.0
- FpuPipeAssignment.Total3: 0.0
- DIV_CYCLES: 0.0

PDEP (R64, R64, M64) - Throughput and Uops

With a non-indexed addressing mode

With 1 independent instruction

With unroll_count=500 and no inner loop

With loop_count=1000 and unroll_count=10

With loop_count=100 and unroll_count=100

With 4 independent instructions

With unroll_count=200 and no inner loop

With loop_count=1000 and unroll_count=2

With loop_count=100 and unroll_count=20

With 7 independent instructions

With unroll_count=100 and no inner loop

With loop_count=1000 and unroll_count=1

With loop_count=100 and unroll_count=10

With an indexed addressing mode

With 1 independent instruction

With unroll_count=500 and no inner loop

With loop_count=1000 and unroll_count=10

With loop_count=100 and unroll_count=100

With 4 independent instructions

With unroll_count=200 and no inner loop

With loop_count=1000 and unroll_count=2

With loop_count=100 and unroll_count=20

With 7 independent instructions

With unroll_count=100 and no inner loop

With loop_count=1000 and unroll_count=1

With loop_count=100 and unroll_count=10