VDIVPS (ZMM, ZMM, ZMM)

Experiment 1

Instruction: VDIVPS ZMM0, ZMM2, ZMM4
Chain instructions: VORPS ZMM2, ZMM0, ZMM0; VORPS ZMM2, ZMM2, ZMM1; VANDPS ZMM2, ZMM2, ZMM1; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2;
Chain latency: ≥11

Code:

   0:	62 f1 6c 48 5e c4    	vdivps zmm0,zmm2,zmm4
   6:	62 f1 7c 48 56 d0    	vorps  zmm2,zmm0,zmm0
   c:	62 f1 6c 48 56 d1    	vorps  zmm2,zmm2,zmm1
  12:	62 f1 6c 48 54 d1    	vandps zmm2,zmm2,zmm1
  18:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  1e:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  24:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  2a:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  30:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  36:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  3c:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  42:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x54ed392654ed3926;
MOV RBX, 0x3f99f4c03f99f4c0;
MOV [R14+0], RBX;
MOV [R14+8], RBX;
MOV [R14+16], RBX;
MOV [R14+24], RBX;
MOV [R14+32], RBX;
MOV [R14+40], RBX;
MOV [R14+48], RBX;
MOV [R14+56], RBX;
MOV [R14+64], RAX;
MOV [R14+72], RAX;
MOV [R14+80], RAX;
MOV [R14+88], RAX;
MOV [R14+96], RAX;
MOV [R14+104], RAX;
MOV [R14+112], RAX;
MOV [R14+120], RAX;
VMOVUPS ZMM1, [R14+64];
VMOVUPS ZMM2, [R14+64];
VMOVUPS ZMM3, [R14];
VMOVUPS ZMM4, [R14]

Show nanoBench command
Results:
- MPERF: 18.72
- APERF: 22.0
- UOPS: 12.0

Experiment 2

Instruction: VDIVPS ZMM2, ZMM2, ZMM4

Code:

   0:	62 f1 6c 48 5e d4    	vdivps zmm2,zmm2,zmm4

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x3f8000003f800000;
MOV RBX, 0x3f8000003f800000;
MOV [R14+0], RBX;
MOV [R14+8], RBX;
MOV [R14+16], RBX;
MOV [R14+24], RBX;
MOV [R14+32], RBX;
MOV [R14+40], RBX;
MOV [R14+48], RBX;
MOV [R14+56], RBX;
MOV [R14+64], RAX;
MOV [R14+72], RAX;
MOV [R14+80], RAX;
MOV [R14+88], RAX;
MOV [R14+96], RAX;
MOV [R14+104], RAX;
MOV [R14+112], RAX;
MOV [R14+120], RAX;
VMOVUPS ZMM1, [R14+64];
VMOVUPS ZMM2, [R14+64];
VMOVUPS ZMM3, [R14];
VMOVUPS ZMM4, [R14]

Show nanoBench command
Results:
- MPERF: 9.4
- APERF: 11.0
- UOPS: 1.0

Experiment 3

Instruction: VDIVPS ZMM0, ZMM2, ZMM4
Chain instructions: VORPS ZMM2, ZMM0, ZMM0; VORPS ZMM2, ZMM2, ZMM1; VANDPS ZMM2, ZMM2, ZMM1; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2; VORPS ZMM2, ZMM2, ZMM2;
Chain latency: ≥11

Code:

   0:	62 f1 6c 48 5e c4    	vdivps zmm0,zmm2,zmm4
   6:	62 f1 7c 48 56 d0    	vorps  zmm2,zmm0,zmm0
   c:	62 f1 6c 48 56 d1    	vorps  zmm2,zmm2,zmm1
  12:	62 f1 6c 48 54 d1    	vandps zmm2,zmm2,zmm1
  18:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  1e:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  24:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  2a:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  30:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  36:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  3c:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2
  42:	62 f1 6c 48 56 d2    	vorps  zmm2,zmm2,zmm2

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x3f8000003f800000;
MOV RBX, 0x3f8000003f800000;
MOV [R14+0], RBX;
MOV [R14+8], RBX;
MOV [R14+16], RBX;
MOV [R14+24], RBX;
MOV [R14+32], RBX;
MOV [R14+40], RBX;
MOV [R14+48], RBX;
MOV [R14+56], RBX;
MOV [R14+64], RAX;
MOV [R14+72], RAX;
MOV [R14+80], RAX;
MOV [R14+88], RAX;
MOV [R14+96], RAX;
MOV [R14+104], RAX;
MOV [R14+112], RAX;
MOV [R14+120], RAX;
VMOVUPS ZMM1, [R14+64];
VMOVUPS ZMM2, [R14+64];
VMOVUPS ZMM3, [R14];
VMOVUPS ZMM4, [R14]

Show nanoBench command
Results:
- MPERF: 19.11
- APERF: 22.0
- UOPS: 12.0

Experiment 1

Instruction: VDIVPS ZMM0, ZMM2, ZMM4
Chain instructions: VORPS ZMM4, ZMM0, ZMM0; VORPS ZMM4, ZMM4, ZMM3; VANDPS ZMM4, ZMM4, ZMM3; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4;
Chain latency: ≥11

Code:

   0:	62 f1 6c 48 5e c4    	vdivps zmm0,zmm2,zmm4
   6:	62 f1 7c 48 56 e0    	vorps  zmm4,zmm0,zmm0
   c:	62 f1 5c 48 56 e3    	vorps  zmm4,zmm4,zmm3
  12:	62 f1 5c 48 54 e3    	vandps zmm4,zmm4,zmm3
  18:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  1e:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  24:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  2a:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  30:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  36:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  3c:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  42:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x54ed392654ed3926;
MOV RBX, 0x3f99f4c03f99f4c0;
MOV [R14+0], RBX;
MOV [R14+8], RBX;
MOV [R14+16], RBX;
MOV [R14+24], RBX;
MOV [R14+32], RBX;
MOV [R14+40], RBX;
MOV [R14+48], RBX;
MOV [R14+56], RBX;
MOV [R14+64], RAX;
MOV [R14+72], RAX;
MOV [R14+80], RAX;
MOV [R14+88], RAX;
MOV [R14+96], RAX;
MOV [R14+104], RAX;
MOV [R14+112], RAX;
MOV [R14+120], RAX;
VMOVUPS ZMM1, [R14+64];
VMOVUPS ZMM2, [R14+64];
VMOVUPS ZMM3, [R14];
VMOVUPS ZMM4, [R14]

Show nanoBench command
Results:
- MPERF: 18.8
- APERF: 22.0
- UOPS: 12.0

Experiment 2

Instruction: VDIVPS ZMM4, ZMM2, ZMM4

Code:

   0:	62 f1 6c 48 5e e4    	vdivps zmm4,zmm2,zmm4

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x3f8000003f800000;
MOV RBX, 0x3f8000003f800000;
MOV [R14+0], RBX;
MOV [R14+8], RBX;
MOV [R14+16], RBX;
MOV [R14+24], RBX;
MOV [R14+32], RBX;
MOV [R14+40], RBX;
MOV [R14+48], RBX;
MOV [R14+56], RBX;
MOV [R14+64], RAX;
MOV [R14+72], RAX;
MOV [R14+80], RAX;
MOV [R14+88], RAX;
MOV [R14+96], RAX;
MOV [R14+104], RAX;
MOV [R14+112], RAX;
MOV [R14+120], RAX;
VMOVUPS ZMM1, [R14+64];
VMOVUPS ZMM2, [R14+64];
VMOVUPS ZMM3, [R14];
VMOVUPS ZMM4, [R14]

Show nanoBench command
Results:
- MPERF: 9.4
- APERF: 11.0
- UOPS: 1.0

Experiment 3

Instruction: VDIVPS ZMM0, ZMM2, ZMM4
Chain instructions: VORPS ZMM4, ZMM0, ZMM0; VORPS ZMM4, ZMM4, ZMM3; VANDPS ZMM4, ZMM4, ZMM3; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4; VORPS ZMM4, ZMM4, ZMM4;
Chain latency: ≥11

Code:

   0:	62 f1 6c 48 5e c4    	vdivps zmm0,zmm2,zmm4
   6:	62 f1 7c 48 56 e0    	vorps  zmm4,zmm0,zmm0
   c:	62 f1 5c 48 56 e3    	vorps  zmm4,zmm4,zmm3
  12:	62 f1 5c 48 54 e3    	vandps zmm4,zmm4,zmm3
  18:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  1e:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  24:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  2a:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  30:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  36:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  3c:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4
  42:	62 f1 5c 48 56 e4    	vorps  zmm4,zmm4,zmm4

Init:

MOV R15, 10000;
L: VADDPS ZMM0, ZMM1, ZMM1;
VADDPS ZMM0, ZMM1, ZMM1;
DEC R15;
JNZ L;
VZEROALL;
MOV RAX, 0x3f8000003f800000;
MOV RBX, 0x3f8000003f800000;
MOV [R14+0], RBX;
MOV [R14+8], RBX;
MOV [R14+16], RBX;
MOV [R14+24], RBX;
MOV [R14+32], RBX;
MOV [R14+40], RBX;
MOV [R14+48], RBX;
MOV [R14+56], RBX;
MOV [R14+64], RAX;
MOV [R14+72], RAX;
MOV [R14+80], RAX;
MOV [R14+88], RAX;
MOV [R14+96], RAX;
MOV [R14+104], RAX;
MOV [R14+112], RAX;
MOV [R14+120], RAX;
VMOVUPS ZMM1, [R14+64];
VMOVUPS ZMM2, [R14+64];
VMOVUPS ZMM3, [R14];
VMOVUPS ZMM4, [R14]

Show nanoBench command
Results:
- MPERF: 19.19
- APERF: 22.0
- UOPS: 12.0

VDIVPS (ZMM, ZMM, ZMM) - Latency

Operands

Latency operand 2 → 1: ≤11

Latency operand 3 → 1: ≤11

Latency operand 2 → 1: ≤11

Experiment 1

Experiment 2

Experiment 3

Latency operand 3 → 1: ≤11

Experiment 1

Experiment 2

Experiment 3