VCVTPS2PD (ZMM, YMM) - Latency

Experiment 1

• Instruction: VCVTPS2PD ZMM1, YMM0
• Chain instructions: VSHUFPD YMM0, YMM1, YMM1, 0;VSHUFPD YMM0, YMM0, YMM0, 0;VSHUFPD YMM0, YMM0, YMM0, 0;VSHUFPD YMM0, YMM0, YMM0, 0;VSHUFPD YMM0, YMM0, YMM0, 0
• Chain latency: 5
• Code:

     0:	62 f1 7c 48 5a c8    	vcvtps2pd zmm1,ymm0
     6:	c5 f5 c6 c1 00       	vshufpd ymm0,ymm1,ymm1,0x0
     b:	c5 fd c6 c0 00       	vshufpd ymm0,ymm0,ymm0,0x0
    10:	c5 fd c6 c0 00       	vshufpd ymm0,ymm0,ymm0,0x0
    15:	c5 fd c6 c0 00       	vshufpd ymm0,ymm0,ymm0,0x0
    1a:	c5 fd c6 c0 00       	vshufpd ymm0,ymm0,ymm0,0x0
  

• 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;
  VMOVUPD YMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 6.0
  • Core cycles: 12.0
  • Reference cycles: 4.08
  • UOPS_EXECUTED.THREAD: 7.0

Experiment 2

• Instruction: VCVTPS2PD ZMM1, YMM0
• Chain instructions: VPAND YMM0, YMM1, YMM1;VPAND YMM0, YMM0, YMM0;VPAND YMM0, YMM0, YMM0;VPAND YMM0, YMM0, YMM0;VPAND YMM0, YMM0, YMM0
• Chain latency: 5
• Code:

     0:	62 f1 7c 48 5a c8    	vcvtps2pd zmm1,ymm0
     6:	c5 f5 db c1          	vpand  ymm0,ymm1,ymm1
     a:	c5 fd db c0          	vpand  ymm0,ymm0,ymm0
     e:	c5 fd db c0          	vpand  ymm0,ymm0,ymm0
    12:	c5 fd db c0          	vpand  ymm0,ymm0,ymm0
    16:	c5 fd db c0          	vpand  ymm0,ymm0,ymm0
  

• 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;
  VMOVUPD YMM0, [R14]

• Show nanoBench command
• Results:
  • Instructions retired: 6.0
  • Core cycles: 12.52
  • Reference cycles: 4.08
  • UOPS_EXECUTED.THREAD: 7.0