VSCATTERQPD (VSIB_ZMM, K, ZMM) - Port Usage

With blocking instructions for port '0':

Code:

   0:	41 0f 7e c0          	movd   r8d,mm0
   4:	41 0f 7e c1          	movd   r9d,mm0
   8:	41 0f 7e c2          	movd   r10d,mm0
   c:	41 0f 7e c3          	movd   r11d,mm0
  10:	41 0f 7e c4          	movd   r12d,mm0
  14:	0f 7e c3             	movd   ebx,mm0
  17:	0f 7e c1             	movd   ecx,mm0
  1a:	0f 7e c2             	movd   edx,mm0
  1d:	41 0f 7e c0          	movd   r8d,mm0
  21:	41 0f 7e c1          	movd   r9d,mm0
  25:	41 0f 7e c2          	movd   r10d,mm0
  29:	41 0f 7e c3          	movd   r11d,mm0
  2d:	41 0f 7e c4          	movd   r12d,mm0
  31:	0f 7e c3             	movd   ebx,mm0
  34:	0f 7e c1             	movd   ecx,mm0
  37:	0f 7e c2             	movd   edx,mm0
  3a:	41 0f 7e c0          	movd   r8d,mm0
  3e:	41 0f 7e c1          	movd   r9d,mm0
  42:	41 0f 7e c2          	movd   r10d,mm0
  46:	62 92 fd 49 a3 04 36 	vscatterqpd QWORD PTR [r14+zmm14*1]{k1},zmm0

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
PXOR MM0, MM0;
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 ZMM14, ZMM14, ZMM14

Show nanoBench command
Results:
- Instructions retired: 20.0
- Core cycles: 21.38
- Reference cycles: 16.83
- UOPS_PORT_0: 21.01

⇨ 2 μops that can only use port '0'

With blocking instructions for port '4':

Code:

   0:	4d 89 46 40          	mov    QWORD PTR [r14+0x40],r8
   4:	4d 89 46 48          	mov    QWORD PTR [r14+0x48],r8
   8:	4d 89 46 50          	mov    QWORD PTR [r14+0x50],r8
   c:	4d 89 46 58          	mov    QWORD PTR [r14+0x58],r8
  10:	4d 89 46 60          	mov    QWORD PTR [r14+0x60],r8
  14:	4d 89 46 68          	mov    QWORD PTR [r14+0x68],r8
  18:	4d 89 46 70          	mov    QWORD PTR [r14+0x70],r8
  1c:	4d 89 46 78          	mov    QWORD PTR [r14+0x78],r8
  20:	4d 89 86 80 00 00 00 	mov    QWORD PTR [r14+0x80],r8
  27:	4d 89 86 88 00 00 00 	mov    QWORD PTR [r14+0x88],r8
  2e:	4d 89 86 90 00 00 00 	mov    QWORD PTR [r14+0x90],r8
  35:	4d 89 86 98 00 00 00 	mov    QWORD PTR [r14+0x98],r8
  3c:	4d 89 86 a0 00 00 00 	mov    QWORD PTR [r14+0xa0],r8
  43:	4d 89 86 a8 00 00 00 	mov    QWORD PTR [r14+0xa8],r8
  4a:	4d 89 86 b0 00 00 00 	mov    QWORD PTR [r14+0xb0],r8
  51:	4d 89 86 b8 00 00 00 	mov    QWORD PTR [r14+0xb8],r8
  58:	4d 89 46 40          	mov    QWORD PTR [r14+0x40],r8
  5c:	4d 89 46 48          	mov    QWORD PTR [r14+0x48],r8
  60:	4d 89 46 50          	mov    QWORD PTR [r14+0x50],r8
  64:	62 92 fd 49 a3 04 36 	vscatterqpd QWORD PTR [r14+zmm14*1]{k1},zmm0

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 ZMM14, ZMM14, ZMM14

Show nanoBench command
Results:
- Instructions retired: 20.0
- Core cycles: 28.1
- Reference cycles: 23.64
- UOPS_PORT_4: 27.0

⇨ 8 μops that can only use port '4'

With blocking instructions for ports {'0', '6'}:

Code:

   0:	66 41 c1 c0 00       	rol    r8w,0x0
   5:	66 41 c1 c1 00       	rol    r9w,0x0
   a:	66 41 c1 c2 00       	rol    r10w,0x0
   f:	66 41 c1 c3 00       	rol    r11w,0x0
  14:	66 41 c1 c4 00       	rol    r12w,0x0
  19:	66 c1 c3 00          	rol    bx,0x0
  1d:	66 c1 c1 00          	rol    cx,0x0
  21:	66 c1 c2 00          	rol    dx,0x0
  25:	66 41 c1 c0 00       	rol    r8w,0x0
  2a:	66 41 c1 c1 00       	rol    r9w,0x0
  2f:	66 41 c1 c2 00       	rol    r10w,0x0
  34:	66 41 c1 c3 00       	rol    r11w,0x0
  39:	66 41 c1 c4 00       	rol    r12w,0x0
  3e:	66 c1 c3 00          	rol    bx,0x0
  42:	66 c1 c1 00          	rol    cx,0x0
  46:	66 c1 c2 00          	rol    dx,0x0
  4a:	66 41 c1 c0 00       	rol    r8w,0x0
  4f:	66 41 c1 c1 00       	rol    r9w,0x0
  54:	66 41 c1 c2 00       	rol    r10w,0x0
  59:	66 41 c1 c3 00       	rol    r11w,0x0
  5e:	66 41 c1 c4 00       	rol    r12w,0x0
  63:	66 c1 c3 00          	rol    bx,0x0
  67:	66 c1 c1 00          	rol    cx,0x0
  6b:	66 c1 c2 00          	rol    dx,0x0
  6f:	66 41 c1 c0 00       	rol    r8w,0x0
  74:	66 41 c1 c1 00       	rol    r9w,0x0
  79:	66 41 c1 c2 00       	rol    r10w,0x0
  7e:	66 41 c1 c3 00       	rol    r11w,0x0
  83:	66 41 c1 c4 00       	rol    r12w,0x0
  88:	66 c1 c3 00          	rol    bx,0x0
  8c:	66 c1 c1 00          	rol    cx,0x0
  90:	66 c1 c2 00          	rol    dx,0x0
  94:	66 41 c1 c0 00       	rol    r8w,0x0
  99:	66 41 c1 c1 00       	rol    r9w,0x0
  9e:	66 41 c1 c2 00       	rol    r10w,0x0
  a3:	66 41 c1 c3 00       	rol    r11w,0x0
  a8:	66 41 c1 c4 00       	rol    r12w,0x0
  ad:	66 c1 c3 00          	rol    bx,0x0
  b1:	62 92 fd 49 a3 04 36 	vscatterqpd QWORD PTR [r14+zmm14*1]{k1},zmm0

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 ZMM14, ZMM14, ZMM14

Show nanoBench command
Results:
- Instructions retired: 39.0
- Core cycles: 20.88
- Reference cycles: 16.83
- UOPS_PORT_0: 20.04
- UOPS_PORT_6: 19.96

With blocking instructions for ports {'2', '3'}:

Code:

   0:	45 8b 46 40          	mov    r8d,DWORD PTR [r14+0x40]
   4:	45 8b 4e 44          	mov    r9d,DWORD PTR [r14+0x44]
   8:	45 8b 56 48          	mov    r10d,DWORD PTR [r14+0x48]
   c:	45 8b 5e 4c          	mov    r11d,DWORD PTR [r14+0x4c]
  10:	45 8b 66 50          	mov    r12d,DWORD PTR [r14+0x50]
  14:	41 8b 5e 54          	mov    ebx,DWORD PTR [r14+0x54]
  18:	41 8b 4e 58          	mov    ecx,DWORD PTR [r14+0x58]
  1c:	41 8b 56 5c          	mov    edx,DWORD PTR [r14+0x5c]
  20:	45 8b 46 40          	mov    r8d,DWORD PTR [r14+0x40]
  24:	45 8b 4e 44          	mov    r9d,DWORD PTR [r14+0x44]
  28:	45 8b 56 48          	mov    r10d,DWORD PTR [r14+0x48]
  2c:	45 8b 5e 4c          	mov    r11d,DWORD PTR [r14+0x4c]
  30:	45 8b 66 50          	mov    r12d,DWORD PTR [r14+0x50]
  34:	41 8b 5e 54          	mov    ebx,DWORD PTR [r14+0x54]
  38:	41 8b 4e 58          	mov    ecx,DWORD PTR [r14+0x58]
  3c:	41 8b 56 5c          	mov    edx,DWORD PTR [r14+0x5c]
  40:	45 8b 46 40          	mov    r8d,DWORD PTR [r14+0x40]
  44:	45 8b 4e 44          	mov    r9d,DWORD PTR [r14+0x44]
  48:	45 8b 56 48          	mov    r10d,DWORD PTR [r14+0x48]
  4c:	45 8b 5e 4c          	mov    r11d,DWORD PTR [r14+0x4c]
  50:	45 8b 66 50          	mov    r12d,DWORD PTR [r14+0x50]
  54:	41 8b 5e 54          	mov    ebx,DWORD PTR [r14+0x54]
  58:	41 8b 4e 58          	mov    ecx,DWORD PTR [r14+0x58]
  5c:	41 8b 56 5c          	mov    edx,DWORD PTR [r14+0x5c]
  60:	45 8b 46 40          	mov    r8d,DWORD PTR [r14+0x40]
  64:	45 8b 4e 44          	mov    r9d,DWORD PTR [r14+0x44]
  68:	45 8b 56 48          	mov    r10d,DWORD PTR [r14+0x48]
  6c:	45 8b 5e 4c          	mov    r11d,DWORD PTR [r14+0x4c]
  70:	45 8b 66 50          	mov    r12d,DWORD PTR [r14+0x50]
  74:	41 8b 5e 54          	mov    ebx,DWORD PTR [r14+0x54]
  78:	41 8b 4e 58          	mov    ecx,DWORD PTR [r14+0x58]
  7c:	41 8b 56 5c          	mov    edx,DWORD PTR [r14+0x5c]
  80:	45 8b 46 40          	mov    r8d,DWORD PTR [r14+0x40]
  84:	45 8b 4e 44          	mov    r9d,DWORD PTR [r14+0x44]
  88:	45 8b 56 48          	mov    r10d,DWORD PTR [r14+0x48]
  8c:	45 8b 5e 4c          	mov    r11d,DWORD PTR [r14+0x4c]
  90:	45 8b 66 50          	mov    r12d,DWORD PTR [r14+0x50]
  94:	41 8b 5e 54          	mov    ebx,DWORD PTR [r14+0x54]
  98:	62 92 fd 49 a3 04 36 	vscatterqpd QWORD PTR [r14+zmm14*1]{k1},zmm0

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 ZMM14, ZMM14, ZMM14

Show nanoBench command
Results:
- Instructions retired: 39.0
- Core cycles: 23.46
- Reference cycles: 19.23
- UOPS_PORT_3: 23.02
- UOPS_PORT_2: 22.98

⇨ 8 μops that can only use ports {'2', '3'}

With blocking instructions for ports {'0', '1', '5', '6'}:

Code:

   0:	45 0f bf c1          	movsx  r8d,r9w
   4:	45 0f bf d1          	movsx  r10d,r9w
   8:	45 0f bf d9          	movsx  r11d,r9w
   c:	45 0f bf e1          	movsx  r12d,r9w
  10:	41 0f bf d9          	movsx  ebx,r9w
  14:	41 0f bf c9          	movsx  ecx,r9w
  18:	41 0f bf d1          	movsx  edx,r9w
  1c:	45 0f bf c1          	movsx  r8d,r9w
  20:	45 0f bf d1          	movsx  r10d,r9w
  24:	45 0f bf d9          	movsx  r11d,r9w
  28:	45 0f bf e1          	movsx  r12d,r9w
  2c:	41 0f bf d9          	movsx  ebx,r9w
  30:	41 0f bf c9          	movsx  ecx,r9w
  34:	41 0f bf d1          	movsx  edx,r9w
  38:	45 0f bf c1          	movsx  r8d,r9w
  3c:	45 0f bf d1          	movsx  r10d,r9w
  40:	45 0f bf d9          	movsx  r11d,r9w
  44:	45 0f bf e1          	movsx  r12d,r9w
  48:	41 0f bf d9          	movsx  ebx,r9w
  4c:	41 0f bf c9          	movsx  ecx,r9w
  50:	41 0f bf d1          	movsx  edx,r9w
  54:	45 0f bf c1          	movsx  r8d,r9w
  58:	45 0f bf d1          	movsx  r10d,r9w
  5c:	45 0f bf d9          	movsx  r11d,r9w
  60:	45 0f bf e1          	movsx  r12d,r9w
  64:	41 0f bf d9          	movsx  ebx,r9w
  68:	41 0f bf c9          	movsx  ecx,r9w
  6c:	41 0f bf d1          	movsx  edx,r9w
  70:	45 0f bf c1          	movsx  r8d,r9w
  74:	45 0f bf d1          	movsx  r10d,r9w
  78:	45 0f bf d9          	movsx  r11d,r9w
  7c:	45 0f bf e1          	movsx  r12d,r9w
  80:	41 0f bf d9          	movsx  ebx,r9w
  84:	41 0f bf c9          	movsx  ecx,r9w
  88:	41 0f bf d1          	movsx  edx,r9w
  8c:	45 0f bf c1          	movsx  r8d,r9w
  90:	45 0f bf d1          	movsx  r10d,r9w
  94:	45 0f bf d9          	movsx  r11d,r9w
  98:	45 0f bf e1          	movsx  r12d,r9w
  9c:	41 0f bf d9          	movsx  ebx,r9w
  a0:	41 0f bf c9          	movsx  ecx,r9w
  a4:	41 0f bf d1          	movsx  edx,r9w
  a8:	45 0f bf c1          	movsx  r8d,r9w
  ac:	45 0f bf d1          	movsx  r10d,r9w
  b0:	45 0f bf d9          	movsx  r11d,r9w
  b4:	45 0f bf e1          	movsx  r12d,r9w
  b8:	41 0f bf d9          	movsx  ebx,r9w
  bc:	41 0f bf c9          	movsx  ecx,r9w
  c0:	41 0f bf d1          	movsx  edx,r9w
  c4:	45 0f bf c1          	movsx  r8d,r9w
  c8:	45 0f bf d1          	movsx  r10d,r9w
  cc:	45 0f bf d9          	movsx  r11d,r9w
  d0:	45 0f bf e1          	movsx  r12d,r9w
  d4:	41 0f bf d9          	movsx  ebx,r9w
  d8:	41 0f bf c9          	movsx  ecx,r9w
  dc:	41 0f bf d1          	movsx  edx,r9w
  e0:	45 0f bf c1          	movsx  r8d,r9w
  e4:	45 0f bf d1          	movsx  r10d,r9w
  e8:	45 0f bf d9          	movsx  r11d,r9w
  ec:	45 0f bf e1          	movsx  r12d,r9w
  f0:	41 0f bf d9          	movsx  ebx,r9w
  f4:	41 0f bf c9          	movsx  ecx,r9w
  f8:	41 0f bf d1          	movsx  edx,r9w
  fc:	45 0f bf c1          	movsx  r8d,r9w
 100:	45 0f bf d1          	movsx  r10d,r9w
 104:	45 0f bf d9          	movsx  r11d,r9w
 108:	45 0f bf e1          	movsx  r12d,r9w
 10c:	41 0f bf d9          	movsx  ebx,r9w
 110:	41 0f bf c9          	movsx  ecx,r9w
 114:	41 0f bf d1          	movsx  edx,r9w
 118:	45 0f bf c1          	movsx  r8d,r9w
 11c:	45 0f bf d1          	movsx  r10d,r9w
 120:	45 0f bf d9          	movsx  r11d,r9w
 124:	45 0f bf e1          	movsx  r12d,r9w
 128:	41 0f bf d9          	movsx  ebx,r9w
 12c:	41 0f bf c9          	movsx  ecx,r9w
 130:	62 92 fd 49 a3 04 36 	vscatterqpd QWORD PTR [r14+zmm14*1]{k1},zmm0

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 ZMM14, ZMM14, ZMM14

Show nanoBench command
Results:
- Instructions retired: 77.0
- Core cycles: 27.46
- Reference cycles: 19.23
- UOPS_PORT_6: 20.31
- UOPS_PORT_0: 19.92
- UOPS_PORT_5: 20.31
- UOPS_PORT_1: 18.23