VGATHERDPD (ZMM, K, VSIB

With unroll_count=500 and no inner loop

Code:

   0:	62 92 fd 49 92 04 36 	vgatherdpd zmm0{k1},QWORD PTR [r14+ymm14*1]

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x4000000040000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOV [R14+16], RAX;
MOV [R14+24], RAX;
MOV [R14+32], RAX;
MOV [R14+40], RAX;
MOV [R14+48], RAX;
MOV [R14+56], RAX;
VMOVUPD ZMM0, [R14];
VXORPS YMM14, YMM14, YMM14

Show nanoBench command
Results:
- MPERF: 9.6
- APERF: 11.15
- UOPS: 46.0
- FpuPipeAssignment.Total0: 3.65
- FpuPipeAssignment.Total1: 2.86
- FpuPipeAssignment.Total2: 3.27
- FpuPipeAssignment.Total3: 3.22
- FpuPipeAssignment.Total4: 5.05
- FpuPipeAssignment.Total5: 4.95
- DIV_CYCLES: 0.0

With loop_count=1000 and unroll_count=10

Code:

   0:	62 92 fd 49 92 04 36 	vgatherdpd zmm0{k1},QWORD PTR [r14+ymm14*1]

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x4000000040000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOV [R14+16], RAX;
MOV [R14+24], RAX;
MOV [R14+32], RAX;
MOV [R14+40], RAX;
MOV [R14+48], RAX;
MOV [R14+56], RAX;
VMOVUPD ZMM0, [R14];
VXORPS YMM14, YMM14, YMM14

Show nanoBench command
Results:
- MPERF: 9.59
- APERF: 11.12
- UOPS: 46.1
- FpuPipeAssignment.Total0: 3.64
- FpuPipeAssignment.Total1: 2.85
- FpuPipeAssignment.Total2: 3.28
- FpuPipeAssignment.Total3: 3.22
- FpuPipeAssignment.Total4: 5.03
- FpuPipeAssignment.Total5: 4.97
- DIV_CYCLES: 0.0

With loop_count=100 and unroll_count=100

Code:

   0:	62 92 fd 49 92 04 36 	vgatherdpd zmm0{k1},QWORD PTR [r14+ymm14*1]

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x4000000040000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOV [R14+16], RAX;
MOV [R14+24], RAX;
MOV [R14+32], RAX;
MOV [R14+40], RAX;
MOV [R14+48], RAX;
MOV [R14+56], RAX;
VMOVUPD ZMM0, [R14];
VXORPS YMM14, YMM14, YMM14

Show nanoBench command
Results:
- MPERF: 9.58
- APERF: 11.11
- UOPS: 46.01
- FpuPipeAssignment.Total0: 3.64
- FpuPipeAssignment.Total1: 2.86
- FpuPipeAssignment.Total2: 3.28
- FpuPipeAssignment.Total3: 3.22
- FpuPipeAssignment.Total4: 5.03
- FpuPipeAssignment.Total5: 4.97
- DIV_CYCLES: 0.0

With 4 independent instructions

With unroll_count=200 and no inner loop

Code:

   0:	62 92 fd 49 92 04 36 	vgatherdpd zmm0{k1},QWORD PTR [r14+ymm14*1]
   7:	62 92 fd 4a 92 4c 36 	vgatherdpd zmm1{k2},QWORD PTR [r14+ymm14*1+0x8]
   e:	01 
   f:	62 92 fd 4b 92 54 36 	vgatherdpd zmm2{k3},QWORD PTR [r14+ymm14*1+0x10]
  16:	02 
  17:	62 92 fd 4c 92 5c 36 	vgatherdpd zmm3{k4},QWORD PTR [r14+ymm14*1+0x18]
  1e:	03

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x4000000040000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOV [R14+16], RAX;
MOV [R14+24], RAX;
MOV [R14+32], RAX;
MOV [R14+40], RAX;
MOV [R14+48], RAX;
MOV [R14+56], RAX;
VMOVUPD ZMM0, [R14];
VXORPS YMM14, YMM14, YMM14;
VMOVUPD ZMM1, [R14];
VMOVUPD ZMM2, [R14];
VMOVUPD ZMM3, [R14]

Show nanoBench command
Results:
- MPERF: 27.53
- APERF: 32.0
- UOPS: 184.0
- FpuPipeAssignment.Total0: 16.0
- FpuPipeAssignment.Total1: 12.0
- FpuPipeAssignment.Total2: 16.0
- FpuPipeAssignment.Total3: 8.0
- FpuPipeAssignment.Total4: 16.0
- FpuPipeAssignment.Total5: 24.0
- DIV_CYCLES: 0.0

With loop_count=1000 and unroll_count=2

Code:

   0:	62 92 fd 49 92 04 36 	vgatherdpd zmm0{k1},QWORD PTR [r14+ymm14*1]
   7:	62 92 fd 4a 92 4c 36 	vgatherdpd zmm1{k2},QWORD PTR [r14+ymm14*1+0x8]
   e:	01 
   f:	62 92 fd 4b 92 54 36 	vgatherdpd zmm2{k3},QWORD PTR [r14+ymm14*1+0x10]
  16:	02 
  17:	62 92 fd 4c 92 5c 36 	vgatherdpd zmm3{k4},QWORD PTR [r14+ymm14*1+0x18]
  1e:	03

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x4000000040000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOV [R14+16], RAX;
MOV [R14+24], RAX;
MOV [R14+32], RAX;
MOV [R14+40], RAX;
MOV [R14+48], RAX;
MOV [R14+56], RAX;
VMOVUPD ZMM0, [R14];
VXORPS YMM14, YMM14, YMM14;
VMOVUPD ZMM1, [R14];
VMOVUPD ZMM2, [R14];
VMOVUPD ZMM3, [R14]

Show nanoBench command
Results:
- MPERF: 28.53
- APERF: 33.08
- UOPS: 184.5
- FpuPipeAssignment.Total0: 15.98
- FpuPipeAssignment.Total1: 12.0
- FpuPipeAssignment.Total2: 16.0
- FpuPipeAssignment.Total3: 8.02
- FpuPipeAssignment.Total4: 16.01
- FpuPipeAssignment.Total5: 23.99
- DIV_CYCLES: 0.0

With loop_count=100 and unroll_count=20

Code:

   0:	62 92 fd 49 92 04 36 	vgatherdpd zmm0{k1},QWORD PTR [r14+ymm14*1]
   7:	62 92 fd 4a 92 4c 36 	vgatherdpd zmm1{k2},QWORD PTR [r14+ymm14*1+0x8]
   e:	01 
   f:	62 92 fd 4b 92 54 36 	vgatherdpd zmm2{k3},QWORD PTR [r14+ymm14*1+0x10]
  16:	02 
  17:	62 92 fd 4c 92 5c 36 	vgatherdpd zmm3{k4},QWORD PTR [r14+ymm14*1+0x18]
  1e:	03

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x4000000040000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOV [R14+16], RAX;
MOV [R14+24], RAX;
MOV [R14+32], RAX;
MOV [R14+40], RAX;
MOV [R14+48], RAX;
MOV [R14+56], RAX;
VMOVUPD ZMM0, [R14];
VXORPS YMM14, YMM14, YMM14;
VMOVUPD ZMM1, [R14];
VMOVUPD ZMM2, [R14];
VMOVUPD ZMM3, [R14]

Show nanoBench command
Results:
- MPERF: 27.73
- APERF: 32.16
- UOPS: 184.05
- FpuPipeAssignment.Total0: 11.36
- FpuPipeAssignment.Total1: 13.54
- FpuPipeAssignment.Total2: 15.99
- FpuPipeAssignment.Total3: 11.12
- FpuPipeAssignment.Total4: 19.11
- FpuPipeAssignment.Total5: 20.94
- DIV_CYCLES: 0.0

With 7 independent instructions

With unroll_count=100 and no inner loop

Code:

   0:	62 92 fd 49 92 04 36 	vgatherdpd zmm0{k1},QWORD PTR [r14+ymm14*1]
   7:	62 92 fd 4a 92 4c 36 	vgatherdpd zmm1{k2},QWORD PTR [r14+ymm14*1+0x8]
   e:	01 
   f:	62 92 fd 4b 92 54 36 	vgatherdpd zmm2{k3},QWORD PTR [r14+ymm14*1+0x10]
  16:	02 
  17:	62 92 fd 4c 92 5c 36 	vgatherdpd zmm3{k4},QWORD PTR [r14+ymm14*1+0x18]
  1e:	03 
  1f:	62 92 fd 4d 92 64 36 	vgatherdpd zmm4{k5},QWORD PTR [r14+ymm14*1+0x20]
  26:	04 
  27:	62 92 fd 4e 92 6c 36 	vgatherdpd zmm5{k6},QWORD PTR [r14+ymm14*1+0x28]
  2e:	05 
  2f:	62 92 fd 4f 92 74 36 	vgatherdpd zmm6{k7},QWORD PTR [r14+ymm14*1+0x30]
  36:	06

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x4000000040000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOV [R14+16], RAX;
MOV [R14+24], RAX;
MOV [R14+32], RAX;
MOV [R14+40], RAX;
MOV [R14+48], RAX;
MOV [R14+56], RAX;
VMOVUPD ZMM0, [R14];
VXORPS YMM14, YMM14, YMM14;
VMOVUPD ZMM1, [R14];
VMOVUPD ZMM2, [R14];
VMOVUPD ZMM3, [R14];
VMOVUPD ZMM4, [R14];
VMOVUPD ZMM5, [R14];
VMOVUPD ZMM6, [R14]

Show nanoBench command
Results:
- MPERF: 52.72
- APERF: 61.25
- UOPS: 322.0
- FpuPipeAssignment.Total0: 17.5
- FpuPipeAssignment.Total1: 24.5
- FpuPipeAssignment.Total2: 28.0
- FpuPipeAssignment.Total3: 21.0
- FpuPipeAssignment.Total4: 35.0
- FpuPipeAssignment.Total5: 35.0
- DIV_CYCLES: 0.0

With loop_count=1000 and unroll_count=1

Code:

   0:	62 92 fd 49 92 04 36 	vgatherdpd zmm0{k1},QWORD PTR [r14+ymm14*1]
   7:	62 92 fd 4a 92 4c 36 	vgatherdpd zmm1{k2},QWORD PTR [r14+ymm14*1+0x8]
   e:	01 
   f:	62 92 fd 4b 92 54 36 	vgatherdpd zmm2{k3},QWORD PTR [r14+ymm14*1+0x10]
  16:	02 
  17:	62 92 fd 4c 92 5c 36 	vgatherdpd zmm3{k4},QWORD PTR [r14+ymm14*1+0x18]
  1e:	03 
  1f:	62 92 fd 4d 92 64 36 	vgatherdpd zmm4{k5},QWORD PTR [r14+ymm14*1+0x20]
  26:	04 
  27:	62 92 fd 4e 92 6c 36 	vgatherdpd zmm5{k6},QWORD PTR [r14+ymm14*1+0x28]
  2e:	05 
  2f:	62 92 fd 4f 92 74 36 	vgatherdpd zmm6{k7},QWORD PTR [r14+ymm14*1+0x30]
  36:	06

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x4000000040000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOV [R14+16], RAX;
MOV [R14+24], RAX;
MOV [R14+32], RAX;
MOV [R14+40], RAX;
MOV [R14+48], RAX;
MOV [R14+56], RAX;
VMOVUPD ZMM0, [R14];
VXORPS YMM14, YMM14, YMM14;
VMOVUPD ZMM1, [R14];
VMOVUPD ZMM2, [R14];
VMOVUPD ZMM3, [R14];
VMOVUPD ZMM4, [R14];
VMOVUPD ZMM5, [R14];
VMOVUPD ZMM6, [R14]

Show nanoBench command
Results:
- MPERF: 49.8
- APERF: 57.75
- UOPS: 323.0
- FpuPipeAssignment.Total0: 28.0
- FpuPipeAssignment.Total1: 21.0
- FpuPipeAssignment.Total2: 28.0
- FpuPipeAssignment.Total3: 14.0
- FpuPipeAssignment.Total4: 28.0
- FpuPipeAssignment.Total5: 42.0
- DIV_CYCLES: 0.0

With loop_count=100 and unroll_count=10

Code:

   0:	62 92 fd 49 92 04 36 	vgatherdpd zmm0{k1},QWORD PTR [r14+ymm14*1]
   7:	62 92 fd 4a 92 4c 36 	vgatherdpd zmm1{k2},QWORD PTR [r14+ymm14*1+0x8]
   e:	01 
   f:	62 92 fd 4b 92 54 36 	vgatherdpd zmm2{k3},QWORD PTR [r14+ymm14*1+0x10]
  16:	02 
  17:	62 92 fd 4c 92 5c 36 	vgatherdpd zmm3{k4},QWORD PTR [r14+ymm14*1+0x18]
  1e:	03 
  1f:	62 92 fd 4d 92 64 36 	vgatherdpd zmm4{k5},QWORD PTR [r14+ymm14*1+0x20]
  26:	04 
  27:	62 92 fd 4e 92 6c 36 	vgatherdpd zmm5{k6},QWORD PTR [r14+ymm14*1+0x28]
  2e:	05 
  2f:	62 92 fd 4f 92 74 36 	vgatherdpd zmm6{k7},QWORD PTR [r14+ymm14*1+0x30]
  36:	06

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x4000000040000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOV [R14+16], RAX;
MOV [R14+24], RAX;
MOV [R14+32], RAX;
MOV [R14+40], RAX;
MOV [R14+48], RAX;
MOV [R14+56], RAX;
VMOVUPD ZMM0, [R14];
VXORPS YMM14, YMM14, YMM14;
VMOVUPD ZMM1, [R14];
VMOVUPD ZMM2, [R14];
VMOVUPD ZMM3, [R14];
VMOVUPD ZMM4, [R14];
VMOVUPD ZMM5, [R14];
VMOVUPD ZMM6, [R14]

Show nanoBench command
Results:
- MPERF: 53.03
- APERF: 61.49
- UOPS: 322.1
- FpuPipeAssignment.Total0: 20.01
- FpuPipeAssignment.Total1: 23.67
- FpuPipeAssignment.Total2: 28.0
- FpuPipeAssignment.Total3: 19.33
- FpuPipeAssignment.Total4: 33.33
- FpuPipeAssignment.Total5: 36.67
- DIV_CYCLES: 0.0

VGATHERDPD (ZMM, K, VSIB_YMM) - Throughput and Uops

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