VMASKMOVPD (M256, YMM, YMM)

Experiment 1

Instruction: VMASKMOVPD ymmword ptr [R14], YMM0, YMM1
Chain instructions: MOV R12, [R14];MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;XOR R14, R12; XOR R14, R12
Chain latency: ≥10

Code:

   0:	c4 c2 7d 2f 0e       	vmaskmovpd YMMWORD PTR [r14],ymm0,ymm1
   5:	4d 8b 26             	mov    r12,QWORD PTR [r14]
   8:	4d 63 e4             	movsxd r12,r12d
   b:	4d 63 e4             	movsxd r12,r12d
   e:	4d 63 e4             	movsxd r12,r12d
  11:	4d 63 e4             	movsxd r12,r12d
  14:	4d 63 e4             	movsxd r12,r12d
  17:	4d 63 e4             	movsxd r12,r12d
  1a:	4d 63 e4             	movsxd r12,r12d
  1d:	4d 63 e4             	movsxd r12,r12d
  20:	4d 31 e6             	xor    r14,r12
  23:	4d 31 e6             	xor    r14,r12

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
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: 12.22
- APERF: 14.08
- UOPS: 29.0

Experiment 2 (with additional nop)

Instruction: VMASKMOVPD ymmword ptr [R14], YMM0, YMM1
Chain instructions: MOV R12, [R14];MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;XOR R14, R12; XOR R14, R12;nop
Chain latency: ≥10

Code:

   0:	c4 c2 7d 2f 0e       	vmaskmovpd YMMWORD PTR [r14],ymm0,ymm1
   5:	4d 8b 26             	mov    r12,QWORD PTR [r14]
   8:	4d 63 e4             	movsxd r12,r12d
   b:	4d 63 e4             	movsxd r12,r12d
   e:	4d 63 e4             	movsxd r12,r12d
  11:	4d 63 e4             	movsxd r12,r12d
  14:	4d 63 e4             	movsxd r12,r12d
  17:	4d 63 e4             	movsxd r12,r12d
  1a:	4d 63 e4             	movsxd r12,r12d
  1d:	4d 63 e4             	movsxd r12,r12d
  20:	4d 31 e6             	xor    r14,r12
  23:	4d 31 e6             	xor    r14,r12
  26:	90                   	nop

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
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: 12.22
- APERF: 14.08
- UOPS: 29.0

Latency operand 1 → 1 (address, index register): ≤8

Experiment 1

Instruction: VMASKMOVPD ymmword ptr [R14+R13], YMM0, YMM1
Chain instructions: MOV R12, [R14+R13];MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;XOR R13, R12; XOR R13, R12
Chain latency: ≥10

Code:

   0:	c4 82 7d 2f 0c 2e    	vmaskmovpd YMMWORD PTR [r14+r13*1],ymm0,ymm1
   6:	4f 8b 24 2e          	mov    r12,QWORD PTR [r14+r13*1]
   a:	4d 63 e4             	movsxd r12,r12d
   d:	4d 63 e4             	movsxd r12,r12d
  10:	4d 63 e4             	movsxd r12,r12d
  13:	4d 63 e4             	movsxd r12,r12d
  16:	4d 63 e4             	movsxd r12,r12d
  19:	4d 63 e4             	movsxd r12,r12d
  1c:	4d 63 e4             	movsxd r12,r12d
  1f:	4d 63 e4             	movsxd r12,r12d
  22:	4d 31 e5             	xor    r13,r12
  25:	4d 31 e5             	xor    r13,r12

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
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: 15.51
- APERF: 18.07
- UOPS: 29.0

Experiment 2 (with additional nop)

Instruction: VMASKMOVPD ymmword ptr [R14+R13], YMM0, YMM1
Chain instructions: MOV R12, [R14+R13];MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;MOVSX R12, R12D;XOR R13, R12; XOR R13, R12;nop
Chain latency: ≥10

Code:

   0:	c4 82 7d 2f 0c 2e    	vmaskmovpd YMMWORD PTR [r14+r13*1],ymm0,ymm1
   6:	4f 8b 24 2e          	mov    r12,QWORD PTR [r14+r13*1]
   a:	4d 63 e4             	movsxd r12,r12d
   d:	4d 63 e4             	movsxd r12,r12d
  10:	4d 63 e4             	movsxd r12,r12d
  13:	4d 63 e4             	movsxd r12,r12d
  16:	4d 63 e4             	movsxd r12,r12d
  19:	4d 63 e4             	movsxd r12,r12d
  1c:	4d 63 e4             	movsxd r12,r12d
  1f:	4d 63 e4             	movsxd r12,r12d
  22:	4d 31 e5             	xor    r13,r12
  25:	4d 31 e5             	xor    r13,r12
  28:	90                   	nop

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
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: 15.51
- APERF: 18.07
- UOPS: 29.0

Latency operand 2 → 1: ≤0

Experiment 1

Instruction: VMASKMOVPD ymmword ptr [R14], YMM0, YMM1
Chain instructions: VMOVUPD XMM13, [R14];VANDPD XMM0, XMM13, XMM13;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0
Chain latency: ≥10

Code:

   0:	c4 c2 7d 2f 0e       	vmaskmovpd YMMWORD PTR [r14],ymm0,ymm1
   5:	c4 41 79 10 2e       	vmovupd xmm13,XMMWORD PTR [r14]
   a:	c4 c1 11 54 c5       	vandpd xmm0,xmm13,xmm13
   f:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  13:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  17:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  1b:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  1f:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  23:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  27:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  2b:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0

Init:

MOV R15, 10000;
L: VADDPS YMM0, YMM1, YMM1;
VADDPS YMM0, YMM1, YMM1;
DEC R15;
JNZ L;
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: 7.83
- APERF: 9.1
- UOPS: 28.0

Latency operand 3 → 1: ≤0

Experiment 1

Instruction: VMASKMOVPD ymmword ptr [R14], YMM1, YMM0
Chain instructions: VMOVUPD XMM13, [R14];VANDPD XMM0, XMM13, XMM13;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0;VANDPD XMM0, XMM0, XMM0
Chain latency: ≥10

Code:

   0:	c4 c2 75 2f 06       	vmaskmovpd YMMWORD PTR [r14],ymm1,ymm0
   5:	c4 41 79 10 2e       	vmovupd xmm13,XMMWORD PTR [r14]
   a:	c4 c1 11 54 c5       	vandpd xmm0,xmm13,xmm13
   f:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  13:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  17:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  1b:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  1f:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  23:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  27:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0
  2b:	c5 f9 54 c0          	vandpd xmm0,xmm0,xmm0

Init:

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

Show nanoBench command
Results:
- MPERF: 6.03
- APERF: 6.99
- UOPS: 28.0

VMASKMOVPD (M256, YMM, YMM) - Latency

Operands

Latency operand 1 → 1 (address, base register): ≤4

Latency operand 1 → 1 (address, index register): ≤8

Latency operand 2 → 1: ≤0

Latency operand 3 → 1: ≤0

Latency operand 1 → 1 (address, base register): ≤4

Experiment 1

Experiment 2 (with additional nop)

Latency operand 1 → 1 (address, index register): ≤8

Experiment 1

Experiment 2 (with additional nop)

Latency operand 2 → 1: ≤0

Experiment 1

Latency operand 3 → 1: ≤0

Experiment 1