VMASKMOVPD (M256, YMM, YMM) - Port Usage

With blocking instructions for port '1':

Code:

   0:	41 0f c4 c0 02       	pinsrw mm0,r8d,0x2
   5:	41 0f c4 c8 02       	pinsrw mm1,r8d,0x2
   a:	41 0f c4 d0 02       	pinsrw mm2,r8d,0x2
   f:	41 0f c4 d8 02       	pinsrw mm3,r8d,0x2
  14:	41 0f c4 e0 02       	pinsrw mm4,r8d,0x2
  19:	41 0f c4 e8 02       	pinsrw mm5,r8d,0x2
  1e:	41 0f c4 f0 02       	pinsrw mm6,r8d,0x2
  23:	41 0f c4 f8 02       	pinsrw mm7,r8d,0x2
  28:	41 0f c4 c0 02       	pinsrw mm0,r8d,0x2
  2d:	41 0f c4 c8 02       	pinsrw mm1,r8d,0x2
  32:	41 0f c4 d0 02       	pinsrw mm2,r8d,0x2
  37:	41 0f c4 d8 02       	pinsrw mm3,r8d,0x2
  3c:	41 0f c4 e0 02       	pinsrw mm4,r8d,0x2
  41:	41 0f c4 e8 02       	pinsrw mm5,r8d,0x2
  46:	41 0f c4 f0 02       	pinsrw mm6,r8d,0x2
  4b:	41 0f c4 f8 02       	pinsrw mm7,r8d,0x2
  50:	41 0f c4 c0 02       	pinsrw mm0,r8d,0x2
  55:	41 0f c4 c8 02       	pinsrw mm1,r8d,0x2
  5a:	c4 c2 7d 2f 0e       	vmaskmovpd YMMWORD PTR [r14],ymm0,ymm1

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
PXOR MM0, MM0;
PXOR MM1, MM1;
PXOR MM2, MM2;
PXOR MM3, MM3;
PXOR MM4, MM4;
PXOR MM5, MM5;
PXOR MM6, MM6;
PXOR MM7, MM7;
VZEROALL;
VXORPS YMM0, YMM0, YMM0;
MOV RAX, 0x4000000040000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOV [R14+16], RAX;
MOV [R14+24], RAX;
VMOVUPD YMM1, [R14]

Show nanoBench command
Results:
- MPERF: 18.87
- APERF: 23.13
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 20.99
- FpuPipeAssignment.Total2: 0.01
- FpuPipeAssignment.Total3: 0.0
- FpuPipeAssignment.Total4: 5.0
- FpuPipeAssignment.Total5: 0.0

⇨ 3 μops that can only use port '1'

With blocking instructions for ports {'4', '5'}:

Code:

   0:	c4 c1 79 7e d0       	vmovd  r8d,xmm2
   5:	c4 c1 79 7e d1       	vmovd  r9d,xmm2
   a:	c4 c1 79 7e d2       	vmovd  r10d,xmm2
   f:	c4 c1 79 7e d3       	vmovd  r11d,xmm2
  14:	c4 c1 79 7e d4       	vmovd  r12d,xmm2
  19:	c5 f9 7e d3          	vmovd  ebx,xmm2
  1d:	c5 f9 7e d1          	vmovd  ecx,xmm2
  21:	c5 f9 7e d2          	vmovd  edx,xmm2
  25:	c4 c1 79 7e d0       	vmovd  r8d,xmm2
  2a:	c4 c1 79 7e d1       	vmovd  r9d,xmm2
  2f:	c4 c1 79 7e d2       	vmovd  r10d,xmm2
  34:	c4 c1 79 7e d3       	vmovd  r11d,xmm2
  39:	c4 c1 79 7e d4       	vmovd  r12d,xmm2
  3e:	c5 f9 7e d3          	vmovd  ebx,xmm2
  42:	c5 f9 7e d1          	vmovd  ecx,xmm2
  46:	c5 f9 7e d2          	vmovd  edx,xmm2
  4a:	c4 c1 79 7e d0       	vmovd  r8d,xmm2
  4f:	c4 c1 79 7e d1       	vmovd  r9d,xmm2
  54:	c4 c1 79 7e d2       	vmovd  r10d,xmm2
  59:	c4 c1 79 7e d3       	vmovd  r11d,xmm2
  5e:	c4 c1 79 7e d4       	vmovd  r12d,xmm2
  63:	c5 f9 7e d3          	vmovd  ebx,xmm2
  67:	c5 f9 7e d1          	vmovd  ecx,xmm2
  6b:	c5 f9 7e d2          	vmovd  edx,xmm2
  6f:	c4 c1 79 7e d0       	vmovd  r8d,xmm2
  74:	c4 c1 79 7e d1       	vmovd  r9d,xmm2
  79:	c4 c1 79 7e d2       	vmovd  r10d,xmm2
  7e:	c4 c1 79 7e d3       	vmovd  r11d,xmm2
  83:	c4 c1 79 7e d4       	vmovd  r12d,xmm2
  88:	c5 f9 7e d3          	vmovd  ebx,xmm2
  8c:	c5 f9 7e d1          	vmovd  ecx,xmm2
  90:	c5 f9 7e d2          	vmovd  edx,xmm2
  94:	c4 c1 79 7e d0       	vmovd  r8d,xmm2
  99:	c4 c1 79 7e d1       	vmovd  r9d,xmm2
  9e:	c4 c1 79 7e d2       	vmovd  r10d,xmm2
  a3:	c4 c1 79 7e d3       	vmovd  r11d,xmm2
  a8:	c4 c2 7d 2f 0e       	vmaskmovpd YMMWORD PTR [r14],ymm0,ymm1

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL;
VXORPS XMM2, XMM2, XMM2;
VXORPS YMM0, YMM0, YMM0;
MOV RAX, 0x4000000040000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOV [R14+16], RAX;
MOV [R14+24], RAX;
VMOVUPD YMM1, [R14]

Show nanoBench command
Results:
- MPERF: 32.75
- APERF: 40.15
- FpuPipeAssignment.Total0: 0.62
- FpuPipeAssignment.Total1: 1.67
- FpuPipeAssignment.Total2: 0.72
- FpuPipeAssignment.Total3: 0.0
- FpuPipeAssignment.Total4: 20.47
- FpuPipeAssignment.Total5: 20.54

⇨ 5 μops that can only use ports {'4', '5'}

With an indexed addressing mode

With blocking instructions for port '1':

Code:

   0:	41 0f c4 c0 02       	pinsrw mm0,r8d,0x2
   5:	41 0f c4 c8 02       	pinsrw mm1,r8d,0x2
   a:	41 0f c4 d0 02       	pinsrw mm2,r8d,0x2
   f:	41 0f c4 d8 02       	pinsrw mm3,r8d,0x2
  14:	41 0f c4 e0 02       	pinsrw mm4,r8d,0x2
  19:	41 0f c4 e8 02       	pinsrw mm5,r8d,0x2
  1e:	41 0f c4 f0 02       	pinsrw mm6,r8d,0x2
  23:	41 0f c4 f8 02       	pinsrw mm7,r8d,0x2
  28:	41 0f c4 c0 02       	pinsrw mm0,r8d,0x2
  2d:	41 0f c4 c8 02       	pinsrw mm1,r8d,0x2
  32:	41 0f c4 d0 02       	pinsrw mm2,r8d,0x2
  37:	41 0f c4 d8 02       	pinsrw mm3,r8d,0x2
  3c:	41 0f c4 e0 02       	pinsrw mm4,r8d,0x2
  41:	41 0f c4 e8 02       	pinsrw mm5,r8d,0x2
  46:	41 0f c4 f0 02       	pinsrw mm6,r8d,0x2
  4b:	41 0f c4 f8 02       	pinsrw mm7,r8d,0x2
  50:	41 0f c4 c0 02       	pinsrw mm0,r8d,0x2
  55:	41 0f c4 c8 02       	pinsrw mm1,r8d,0x2
  5a:	c4 82 7d 2f 0c 2e    	vmaskmovpd YMMWORD PTR [r14+r13*1],ymm0,ymm1

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
PXOR MM0, MM0;
PXOR MM1, MM1;
PXOR MM2, MM2;
PXOR MM3, MM3;
PXOR MM4, MM4;
PXOR MM5, MM5;
PXOR MM6, MM6;
PXOR MM7, MM7;
VZEROALL;
VXORPS YMM0, YMM0, YMM0;
MOV RAX, 0x4000000040000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOV [R14+16], RAX;
MOV [R14+24], RAX;
VMOVUPD YMM1, [R14];
XOR R13, R13

Show nanoBench command
Results:
- MPERF: 18.87
- APERF: 23.13
- FpuPipeAssignment.Total0: 0.0
- FpuPipeAssignment.Total1: 20.99
- FpuPipeAssignment.Total2: 0.01
- FpuPipeAssignment.Total3: 0.0
- FpuPipeAssignment.Total4: 5.0
- FpuPipeAssignment.Total5: 0.0

⇨ 3 μops that can only use port '1'

With blocking instructions for ports {'4', '5'}:

Code:

   0:	c4 c1 79 7e d0       	vmovd  r8d,xmm2
   5:	c4 c1 79 7e d1       	vmovd  r9d,xmm2
   a:	c4 c1 79 7e d2       	vmovd  r10d,xmm2
   f:	c4 c1 79 7e d3       	vmovd  r11d,xmm2
  14:	c4 c1 79 7e d4       	vmovd  r12d,xmm2
  19:	c5 f9 7e d3          	vmovd  ebx,xmm2
  1d:	c5 f9 7e d1          	vmovd  ecx,xmm2
  21:	c5 f9 7e d2          	vmovd  edx,xmm2
  25:	c4 c1 79 7e d0       	vmovd  r8d,xmm2
  2a:	c4 c1 79 7e d1       	vmovd  r9d,xmm2
  2f:	c4 c1 79 7e d2       	vmovd  r10d,xmm2
  34:	c4 c1 79 7e d3       	vmovd  r11d,xmm2
  39:	c4 c1 79 7e d4       	vmovd  r12d,xmm2
  3e:	c5 f9 7e d3          	vmovd  ebx,xmm2
  42:	c5 f9 7e d1          	vmovd  ecx,xmm2
  46:	c5 f9 7e d2          	vmovd  edx,xmm2
  4a:	c4 c1 79 7e d0       	vmovd  r8d,xmm2
  4f:	c4 c1 79 7e d1       	vmovd  r9d,xmm2
  54:	c4 c1 79 7e d2       	vmovd  r10d,xmm2
  59:	c4 c1 79 7e d3       	vmovd  r11d,xmm2
  5e:	c4 c1 79 7e d4       	vmovd  r12d,xmm2
  63:	c5 f9 7e d3          	vmovd  ebx,xmm2
  67:	c5 f9 7e d1          	vmovd  ecx,xmm2
  6b:	c5 f9 7e d2          	vmovd  edx,xmm2
  6f:	c4 c1 79 7e d0       	vmovd  r8d,xmm2
  74:	c4 c1 79 7e d1       	vmovd  r9d,xmm2
  79:	c4 c1 79 7e d2       	vmovd  r10d,xmm2
  7e:	c4 c1 79 7e d3       	vmovd  r11d,xmm2
  83:	c4 c1 79 7e d4       	vmovd  r12d,xmm2
  88:	c5 f9 7e d3          	vmovd  ebx,xmm2
  8c:	c5 f9 7e d1          	vmovd  ecx,xmm2
  90:	c5 f9 7e d2          	vmovd  edx,xmm2
  94:	c4 c1 79 7e d0       	vmovd  r8d,xmm2
  99:	c4 c1 79 7e d1       	vmovd  r9d,xmm2
  9e:	c4 c1 79 7e d2       	vmovd  r10d,xmm2
  a3:	c4 c1 79 7e d3       	vmovd  r11d,xmm2
  a8:	c4 82 7d 2f 0c 2e    	vmaskmovpd YMMWORD PTR [r14+r13*1],ymm0,ymm1

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
VZEROALL;
VXORPS XMM2, XMM2, XMM2;
VXORPS YMM0, YMM0, YMM0;
MOV RAX, 0x4000000040000000;
MOV [R14+0], RAX;
MOV [R14+8], RAX;
MOV [R14+16], RAX;
MOV [R14+24], RAX;
VMOVUPD YMM1, [R14];
XOR R13, R13

Show nanoBench command
Results:
- MPERF: 32.87
- APERF: 40.16
- FpuPipeAssignment.Total0: 0.62
- FpuPipeAssignment.Total1: 1.67
- FpuPipeAssignment.Total2: 0.72
- FpuPipeAssignment.Total3: 0.0
- FpuPipeAssignment.Total4: 20.47
- FpuPipeAssignment.Total5: 20.54

⇨ 5 μops that can only use ports {'4', '5'}