OV - attention-avx512 - Loop 37

	Loop Id: 37	Module: attention-avx512	Source: attention.cpp:52-53	Coverage: 0.10%

0x6f90 VMOVUPS	-0xc0(%RDI,%R13,4),%ZMM0    [2]

0x6f98 VMOVAPS	0x480(%RSP),%ZMM4    [1]

0x6fa0 VMOVUPS	(%RDI,%R13,4),%ZMM3    [2]

0x6fa7 VMOVAPS	%ZMM1,0x4c0(%RSP)    [1]

0x6faf VMOVUPS	-0x80(%RDI,%R13,4),%ZMM1    [2]

0x6fb7 VMOVUPS	-0x40(%RDI,%R13,4),%ZMM2    [2]

0x6fbf VSUBPS	%ZMM4,%ZMM0,%ZMM5

0x6fc5 VSUBPS	%ZMM4,%ZMM3,%ZMM0

0x6fcb VSUBPS	%ZMM4,%ZMM1,%ZMM6

0x6fd1 VSUBPS	%ZMM4,%ZMM2,%ZMM1

0x6fd7 VMOVAPS	%ZMM0,0x240(%RSP)    [1]

0x6fdf VEXTRACTF32X4	$0x3,%ZMM5,%XMM0

0x6fe6 VMOVAPS	%ZMM5,0x40(%RSP)    [1]

0x6fee VMOVAPS	%ZMM6,0x100(%RSP)    [1]

0x6ff6 VMOVAPS	%ZMM1,0x1c0(%RSP)    [1]

0x6ffe VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x7007 VZEROUPPER

0x700a CALL	89b0 <@plt_start@+0x20>

0x700f VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x7018 VMOVSHDUP	0xc0(%RSP),%XMM0    [1]

0x7021 CALL	89b0 <@plt_start@+0x20>

0x7026 VMOVAPS	0x180(%RSP),%XMM1    [1]

0x702f VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x7035 VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x703e VPERMILPD	$0x1,0xc0(%RSP),%XMM0    [1]

0x7049 CALL	89b0 <@plt_start@+0x20>

0x704e VMOVAPS	0x180(%RSP),%XMM1    [1]

0x7057 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x705d VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x7066 VPERMILPS	$-0x1,0xc0(%RSP),%XMM0    [1]

0x7071 CALL	89b0 <@plt_start@+0x20>

0x7076 VMOVAPS	0x180(%RSP),%XMM1    [1]

0x707f VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x7085 VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x708e VMOVAPS	0x40(%RSP),%ZMM0    [1]

0x7096 VEXTRACTF32X4	$0x2,%ZMM0,%XMM0

0x709d VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x70a6 VZEROUPPER

0x70a9 CALL	89b0 <@plt_start@+0x20>

0x70ae VMOVAPS	%XMM0,0x160(%RSP)    [1]

0x70b7 VMOVSHDUP	0xc0(%RSP),%XMM0    [1]

0x70c0 CALL	89b0 <@plt_start@+0x20>

0x70c5 VMOVAPS	0x160(%RSP),%XMM1    [1]

0x70ce VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x70d4 VMOVAPS	%XMM0,0x160(%RSP)    [1]

0x70dd VPERMILPD	$0x1,0xc0(%RSP),%XMM0    [1]

0x70e8 CALL	89b0 <@plt_start@+0x20>

0x70ed VMOVAPS	0x160(%RSP),%XMM1    [1]

0x70f6 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x70fc VMOVAPS	%XMM0,0x160(%RSP)    [1]

0x7105 VPERMILPS	$-0x1,0xc0(%RSP),%XMM0    [1]

0x7110 CALL	89b0 <@plt_start@+0x20>

0x7115 VMOVAPS	0x160(%RSP),%XMM1    [1]

0x711e VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x7124 VINSERTF128	$0x1,0x180(%RSP),%YMM0,%YMM0    [1]

0x712f VMOVAPS	%YMM0,0x180(%RSP)    [1]

0x7138 VMOVAPS	0x40(%RSP),%ZMM0    [1]

0x7140 VEXTRACTF128	$0x1,%YMM0,%XMM0

0x7146 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x714f VZEROUPPER

0x7152 CALL	89b0 <@plt_start@+0x20>

0x7157 VMOVAPS	%XMM0,0x160(%RSP)    [1]

0x7160 VMOVSHDUP	0xc0(%RSP),%XMM0    [1]

0x7169 CALL	89b0 <@plt_start@+0x20>

0x716e VMOVAPS	0x160(%RSP),%XMM1    [1]

0x7177 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x717d VMOVAPS	%XMM0,0x160(%RSP)    [1]

0x7186 VPERMILPD	$0x1,0xc0(%RSP),%XMM0    [1]

0x7191 CALL	89b0 <@plt_start@+0x20>

0x7196 VMOVAPS	0x160(%RSP),%XMM1    [1]

0x719f VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x71a5 VMOVAPS	%XMM0,0x160(%RSP)    [1]

0x71ae VPERMILPS	$-0x1,0xc0(%RSP),%XMM0    [1]

0x71b9 CALL	89b0 <@plt_start@+0x20>

0x71be VMOVAPS	0x160(%RSP),%XMM1    [1]

0x71c7 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x71cd VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x71d6 VMOVAPS	0x40(%RSP),%ZMM0    [1]

0x71de VZEROUPPER

0x71e1 CALL	89b0 <@plt_start@+0x20>

0x71e6 VMOVAPS	%XMM0,0x160(%RSP)    [1]

0x71ef VMOVSHDUP	0x40(%RSP),%XMM0    [1]

0x71f5 CALL	89b0 <@plt_start@+0x20>

0x71fa VMOVAPS	0x160(%RSP),%XMM1    [1]

0x7203 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x7209 VMOVAPS	%XMM0,0x160(%RSP)    [1]

0x7212 VPERMILPD	$0x1,0x40(%RSP),%XMM0    [1]

0x721a CALL	89b0 <@plt_start@+0x20>

0x721f VMOVAPS	0x160(%RSP),%XMM1    [1]

0x7228 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x722e VMOVAPS	%XMM0,0x160(%RSP)    [1]

0x7237 VPERMILPS	$-0x1,0x40(%RSP),%XMM0    [1]

0x723f CALL	89b0 <@plt_start@+0x20>

0x7244 VMOVAPS	0x160(%RSP),%XMM1    [1]

0x724d VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x7253 VINSERTF128	$0x1,0xc0(%RSP),%YMM0,%YMM0    [1]

0x725e VMOVAPS	0x340(%RSP),%ZMM1    [1]

0x7266 VINSERTF64X4	$0x1,0x180(%RSP),%ZMM0,%ZMM0    [1]

0x726f VADDPS	%ZMM1,%ZMM0,%ZMM1

0x7275 VMOVAPS	0x100(%RSP),%ZMM0    [1]

0x727d VMOVAPS	%ZMM1,0x340(%RSP)    [1]

0x7285 VEXTRACTF32X4	$0x3,%ZMM0,%XMM0

0x728c VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x7292 VZEROUPPER

0x7295 CALL	89b0 <@plt_start@+0x20>

0x729a VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x72a3 VMOVSHDUP	0x40(%RSP),%XMM0    [1]

0x72a9 CALL	89b0 <@plt_start@+0x20>

0x72ae VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x72b7 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x72bd VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x72c6 VPERMILPD	$0x1,0x40(%RSP),%XMM0    [1]

0x72ce CALL	89b0 <@plt_start@+0x20>

0x72d3 VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x72dc VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x72e2 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x72eb VPERMILPS	$-0x1,0x40(%RSP),%XMM0    [1]

0x72f3 CALL	89b0 <@plt_start@+0x20>

0x72f8 VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x7301 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x7307 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x7310 VMOVAPS	0x100(%RSP),%ZMM0    [1]

0x7318 VEXTRACTF32X4	$0x2,%ZMM0,%XMM0

0x731f VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x7325 VZEROUPPER

0x7328 CALL	89b0 <@plt_start@+0x20>

0x732d VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x7336 VMOVSHDUP	0x40(%RSP),%XMM0    [1]

0x733c CALL	89b0 <@plt_start@+0x20>

0x7341 VMOVAPS	0x180(%RSP),%XMM1    [1]

0x734a VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x7350 VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x7359 VPERMILPD	$0x1,0x40(%RSP),%XMM0    [1]

0x7361 CALL	89b0 <@plt_start@+0x20>

0x7366 VMOVAPS	0x180(%RSP),%XMM1    [1]

0x736f VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x7375 VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x737e VPERMILPS	$-0x1,0x40(%RSP),%XMM0    [1]

0x7386 CALL	89b0 <@plt_start@+0x20>

0x738b VMOVAPS	0x180(%RSP),%XMM1    [1]

0x7394 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x739a VINSERTF128	$0x1,0xc0(%RSP),%YMM0,%YMM0    [1]

0x73a5 VMOVAPS	%YMM0,0xc0(%RSP)    [1]

0x73ae VMOVAPS	0x100(%RSP),%ZMM0    [1]

0x73b6 VEXTRACTF128	$0x1,%YMM0,%XMM0

0x73bc VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x73c2 VZEROUPPER

0x73c5 CALL	89b0 <@plt_start@+0x20>

0x73ca VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x73d3 VMOVSHDUP	0x40(%RSP),%XMM0    [1]

0x73d9 CALL	89b0 <@plt_start@+0x20>

0x73de VMOVAPS	0x180(%RSP),%XMM1    [1]

0x73e7 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x73ed VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x73f6 VPERMILPD	$0x1,0x40(%RSP),%XMM0    [1]

0x73fe CALL	89b0 <@plt_start@+0x20>

0x7403 VMOVAPS	0x180(%RSP),%XMM1    [1]

0x740c VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x7412 VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x741b VPERMILPS	$-0x1,0x40(%RSP),%XMM0    [1]

0x7423 CALL	89b0 <@plt_start@+0x20>

0x7428 VMOVAPS	0x180(%RSP),%XMM1    [1]

0x7431 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x7437 VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x743d VMOVAPS	0x100(%RSP),%ZMM0    [1]

0x7445 VZEROUPPER

0x7448 CALL	89b0 <@plt_start@+0x20>

0x744d VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x7456 VMOVSHDUP	0x100(%RSP),%XMM0    [1]

0x745f CALL	89b0 <@plt_start@+0x20>

0x7464 VMOVAPS	0x180(%RSP),%XMM1    [1]

0x746d VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x7473 VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x747c VPERMILPD	$0x1,0x100(%RSP),%XMM0    [1]

0x7487 CALL	89b0 <@plt_start@+0x20>

0x748c VMOVAPS	0x180(%RSP),%XMM1    [1]

0x7495 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x749b VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x74a4 VPERMILPS	$-0x1,0x100(%RSP),%XMM0    [1]

0x74af CALL	89b0 <@plt_start@+0x20>

0x74b4 VMOVAPS	0x180(%RSP),%XMM1    [1]

0x74bd VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x74c3 VINSERTF128	$0x1,0x40(%RSP),%YMM0,%YMM0    [1]

0x74cb VMOVAPS	0x400(%RSP),%ZMM1    [1]

0x74d3 VINSERTF64X4	$0x1,0xc0(%RSP),%ZMM0,%ZMM0    [1]

0x74dc VADDPS	%ZMM1,%ZMM0,%ZMM1

0x74e2 VMOVAPS	0x1c0(%RSP),%ZMM0    [1]

0x74ea VMOVAPS	%ZMM1,0x400(%RSP)    [1]

0x74f2 VEXTRACTF32X4	$0x3,%ZMM0,%XMM0

0x74f9 VMOVAPS	%XMM0,0x100(%RSP)    [1]

0x7502 VZEROUPPER

0x7505 CALL	89b0 <@plt_start@+0x20>

0x750a VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x7510 VMOVSHDUP	0x100(%RSP),%XMM0    [1]

0x7519 CALL	89b0 <@plt_start@+0x20>

0x751e VMOVAPS	0x40(%RSP),%XMM1    [1]

0x7524 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x752a VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x7530 VPERMILPD	$0x1,0x100(%RSP),%XMM0    [1]

0x753b CALL	89b0 <@plt_start@+0x20>

0x7540 VMOVAPS	0x40(%RSP),%XMM1    [1]

0x7546 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x754c VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x7552 VPERMILPS	$-0x1,0x100(%RSP),%XMM0    [1]

0x755d CALL	89b0 <@plt_start@+0x20>

0x7562 VMOVAPS	0x40(%RSP),%XMM1    [1]

0x7568 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x756e VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x7574 VMOVAPS	0x1c0(%RSP),%ZMM0    [1]

0x757c VEXTRACTF32X4	$0x2,%ZMM0,%XMM0

0x7583 VMOVAPS	%XMM0,0x100(%RSP)    [1]

0x758c VZEROUPPER

0x758f CALL	89b0 <@plt_start@+0x20>

0x7594 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x759d VMOVSHDUP	0x100(%RSP),%XMM0    [1]

0x75a6 CALL	89b0 <@plt_start@+0x20>

0x75ab VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x75b4 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x75ba VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x75c3 VPERMILPD	$0x1,0x100(%RSP),%XMM0    [1]

0x75ce CALL	89b0 <@plt_start@+0x20>

0x75d3 VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x75dc VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x75e2 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x75eb VPERMILPS	$-0x1,0x100(%RSP),%XMM0    [1]

0x75f6 CALL	89b0 <@plt_start@+0x20>

0x75fb VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x7604 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x760a VINSERTF128	$0x1,0x40(%RSP),%YMM0,%YMM0    [1]

0x7612 VMOVAPS	%YMM0,0x40(%RSP)    [1]

0x7618 VMOVAPS	0x1c0(%RSP),%ZMM0    [1]

0x7620 VEXTRACTF128	$0x1,%YMM0,%XMM0

0x7626 VMOVAPS	%XMM0,0x100(%RSP)    [1]

0x762f VZEROUPPER

0x7632 CALL	89b0 <@plt_start@+0x20>

0x7637 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x7640 VMOVSHDUP	0x100(%RSP),%XMM0    [1]

0x7649 CALL	89b0 <@plt_start@+0x20>

0x764e VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x7657 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x765d VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x7666 VPERMILPD	$0x1,0x100(%RSP),%XMM0    [1]

0x7671 CALL	89b0 <@plt_start@+0x20>

0x7676 VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x767f VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x7685 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x768e VPERMILPS	$-0x1,0x100(%RSP),%XMM0    [1]

0x7699 CALL	89b0 <@plt_start@+0x20>

0x769e VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x76a7 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x76ad VMOVAPS	%XMM0,0x100(%RSP)    [1]

0x76b6 VMOVAPS	0x1c0(%RSP),%ZMM0    [1]

0x76be VZEROUPPER

0x76c1 CALL	89b0 <@plt_start@+0x20>

0x76c6 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x76cf VMOVSHDUP	0x1c0(%RSP),%XMM0    [1]

0x76d8 CALL	89b0 <@plt_start@+0x20>

0x76dd VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x76e6 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x76ec VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x76f5 VPERMILPD	$0x1,0x1c0(%RSP),%XMM0    [1]

0x7700 CALL	89b0 <@plt_start@+0x20>

0x7705 VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x770e VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x7714 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x771d VPERMILPS	$-0x1,0x1c0(%RSP),%XMM0    [1]

0x7728 CALL	89b0 <@plt_start@+0x20>

0x772d VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x7736 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x773c VINSERTF128	$0x1,0x100(%RSP),%YMM0,%YMM0    [1]

0x7747 VMOVAPS	0x440(%RSP),%ZMM1    [1]

0x774f VINSERTF64X4	$0x1,0x40(%RSP),%ZMM0,%ZMM0    [1]

0x7758 VADDPS	%ZMM1,%ZMM0,%ZMM1

0x775e VMOVAPS	0x240(%RSP),%ZMM0    [1]

0x7766 VMOVAPS	%ZMM1,0x440(%RSP)    [1]

0x776e VEXTRACTF32X4	$0x3,%ZMM0,%XMM0

0x7775 VMOVAPS	%XMM0,0x1c0(%RSP)    [1]

0x777e VZEROUPPER

0x7781 CALL	89b0 <@plt_start@+0x20>

0x7786 VMOVAPS	%XMM0,0x100(%RSP)    [1]

0x778f VMOVSHDUP	0x1c0(%RSP),%XMM0    [1]

0x7798 CALL	89b0 <@plt_start@+0x20>

0x779d VMOVAPS	0x100(%RSP),%XMM1    [1]

0x77a6 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x77ac VMOVAPS	%XMM0,0x100(%RSP)    [1]

0x77b5 VPERMILPD	$0x1,0x1c0(%RSP),%XMM0    [1]

0x77c0 CALL	89b0 <@plt_start@+0x20>

0x77c5 VMOVAPS	0x100(%RSP),%XMM1    [1]

0x77ce VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x77d4 VMOVAPS	%XMM0,0x100(%RSP)    [1]

0x77dd VPERMILPS	$-0x1,0x1c0(%RSP),%XMM0    [1]

0x77e8 CALL	89b0 <@plt_start@+0x20>

0x77ed VMOVAPS	0x100(%RSP),%XMM1    [1]

0x77f6 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x77fc VMOVAPS	%XMM0,0x100(%RSP)    [1]

0x7805 VMOVAPS	0x240(%RSP),%ZMM0    [1]

0x780d VEXTRACTF32X4	$0x2,%ZMM0,%XMM0

0x7814 VMOVAPS	%XMM0,0x1c0(%RSP)    [1]

0x781d VZEROUPPER

0x7820 CALL	89b0 <@plt_start@+0x20>

0x7825 VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x782b VMOVSHDUP	0x1c0(%RSP),%XMM0    [1]

0x7834 CALL	89b0 <@plt_start@+0x20>

0x7839 VMOVAPS	0x40(%RSP),%XMM1    [1]

0x783f VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x7845 VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x784b VPERMILPD	$0x1,0x1c0(%RSP),%XMM0    [1]

0x7856 CALL	89b0 <@plt_start@+0x20>

0x785b VMOVAPS	0x40(%RSP),%XMM1    [1]

0x7861 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x7867 VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x786d VPERMILPS	$-0x1,0x1c0(%RSP),%XMM0    [1]

0x7878 CALL	89b0 <@plt_start@+0x20>

0x787d VMOVAPS	0x40(%RSP),%XMM1    [1]

0x7883 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x7889 VINSERTF128	$0x1,0x100(%RSP),%YMM0,%YMM0    [1]

0x7894 VMOVAPS	%YMM0,0x100(%RSP)    [1]

0x789d VMOVAPS	0x240(%RSP),%ZMM0    [1]

0x78a5 VEXTRACTF128	$0x1,%YMM0,%XMM0

0x78ab VMOVAPS	%XMM0,0x1c0(%RSP)    [1]

0x78b4 VZEROUPPER

0x78b7 CALL	89b0 <@plt_start@+0x20>

0x78bc VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x78c2 VMOVSHDUP	0x1c0(%RSP),%XMM0    [1]

0x78cb CALL	89b0 <@plt_start@+0x20>

0x78d0 VMOVAPS	0x40(%RSP),%XMM1    [1]

0x78d6 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x78dc VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x78e2 VPERMILPD	$0x1,0x1c0(%RSP),%XMM0    [1]

0x78ed CALL	89b0 <@plt_start@+0x20>

0x78f2 VMOVAPS	0x40(%RSP),%XMM1    [1]

0x78f8 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x78fe VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x7904 VPERMILPS	$-0x1,0x1c0(%RSP),%XMM0    [1]

0x790f CALL	89b0 <@plt_start@+0x20>

0x7914 VMOVAPS	0x40(%RSP),%XMM1    [1]

0x791a VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x7920 VMOVAPS	%XMM0,0x1c0(%RSP)    [1]

0x7929 VMOVAPS	0x240(%RSP),%ZMM0    [1]

0x7931 VZEROUPPER

0x7934 CALL	89b0 <@plt_start@+0x20>

0x7939 VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x793f VMOVSHDUP	0x240(%RSP),%XMM0    [1]

0x7948 CALL	89b0 <@plt_start@+0x20>

0x794d VMOVAPS	0x40(%RSP),%XMM1    [1]

0x7953 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x7959 VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x795f VPERMILPD	$0x1,0x240(%RSP),%XMM0    [1]

0x796a CALL	89b0 <@plt_start@+0x20>

0x796f VMOVAPS	0x40(%RSP),%XMM1    [1]

0x7975 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x797b VMOVAPS	%XMM0,0x40(%RSP)    [1]

0x7981 VPERMILPS	$-0x1,0x240(%RSP),%XMM0    [1]

0x798c CALL	89b0 <@plt_start@+0x20>

0x7991 VMOVAPS	0x40(%RSP),%XMM2    [1]

0x7997 VMOVAPS	0x4c0(%RSP),%ZMM1    [1]

0x799f MOV	0x3e8(%RSP),%R10    [1]

0x79a7 MOV	0x330(%RSP),%RDI    [1]

0x79af ADD	$0x40,%R13

0x79b3 VINSERTPS	$0x30,%XMM0,%XMM2,%XMM0

0x79b9 VINSERTF128	$0x1,0x1c0(%RSP),%YMM0,%YMM0    [1]

0x79c4 VINSERTF64X4	$0x1,0x100(%RSP),%ZMM0,%ZMM0    [1]

0x79cd VADDPS	%ZMM1,%ZMM0,%ZMM1

0x79d3 CMP	%R13,%R10

0x79d6 JNE	6f90

/home/eoseret/llm-attention/attention.cpp: 52 - 53

--------------------------------------------------------------------------------

52:       for (int idx = 0; idx <= row; ++idx) // vectorised

53:          sum += expf(S_row[idx] - max_val);

Coverage (%)	Name	Source Location	Module

min	med	avg	max

Percentile Index	10	20	30	40	50	60	70	80	90	100
Value

min	med	avg	max

Percentile Index	10	20	30	40	50	60	70	80	90	100
Value

Path /

Metric	Value
CQA speedup if no scalar integer	1.31
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	1.94
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.28
Bottlenecks	P6, P7, P8, P9,
Function	main
Source	attention.cpp:52-53
Source loop unroll info	unrolled by 64
Source loop unroll confidence level	max
Unroll/vectorization loop type	main
Unroll factor	64
CQA cycles	69.50
CQA cycles if no scalar integer	53.25
CQA cycles if FP arith vectorized	69.50
CQA cycles if fully vectorized	35.75
Front-end cycles	54.50
P0 cycles	0.33
P1 cycles	0.33
P2 cycles	0.33
P3 cycles	21.67
P4 cycles	21.67
P5 cycles	21.67
P6 cycles	69.50
P7 cycles	69.50
P8 cycles	69.50
P9 cycles	69.50
P10 cycles	28.00
P11 cycles	28.00
P12 cycles	28.00
P13 cycles	28.00
P14 cycles	44.00
P15 cycles	44.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	365.00
Nb uops	436.00
Nb loads	134.00
Nb stores	80.00
Nb stack references	14.00
FLOP/cycle	1.84
Nb FLOP add-sub	128.00
Nb FLOP mul	0.00
Nb FLOP fma	0.00
Nb FLOP div	0.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	72.98
Bytes prefetched	0.00
Bytes loaded	3344.00
Bytes stored	1728.00
Stride 0	1.00
Stride 1	1.00
Stride n	0.00
Stride unknown	0.00
Stride indirect	0.00
Vectorization ratio all	83.78
Vectorization ratio load	100.00
Vectorization ratio store	100.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	100.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	60.00
Vector-efficiency ratio all	32.09
Vector-efficiency ratio load	37.88
Vector-efficiency ratio store	33.75
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	100.00
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	18.33

Metric	Value
CQA speedup if no scalar integer	1.31
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	1.94
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.28
Bottlenecks	P6, P7, P8, P9,
Function	main
Source	attention.cpp:52-53
Source loop unroll info	unrolled by 64
Source loop unroll confidence level	max
Unroll/vectorization loop type	main
Unroll factor	64
CQA cycles	69.50
CQA cycles if no scalar integer	53.25
CQA cycles if FP arith vectorized	69.50
CQA cycles if fully vectorized	35.75
Front-end cycles	54.50
P0 cycles	0.33
P1 cycles	0.33
P2 cycles	0.33
P3 cycles	21.67
P4 cycles	21.67
P5 cycles	21.67
P6 cycles	69.50
P7 cycles	69.50
P8 cycles	69.50
P9 cycles	69.50
P10 cycles	28.00
P11 cycles	28.00
P12 cycles	28.00
P13 cycles	28.00
P14 cycles	44.00
P15 cycles	44.00
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	365.00
Nb uops	436.00
Nb loads	134.00
Nb stores	80.00
Nb stack references	14.00
FLOP/cycle	1.84
Nb FLOP add-sub	128.00
Nb FLOP mul	0.00
Nb FLOP fma	0.00
Nb FLOP div	0.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	72.98
Bytes prefetched	0.00
Bytes loaded	3344.00
Bytes stored	1728.00
Stride 0	1.00
Stride 1	1.00
Stride n	0.00
Stride unknown	0.00
Stride indirect	0.00
Vectorization ratio all	83.78
Vectorization ratio load	100.00
Vectorization ratio store	100.00
Vectorization ratio mul	NA
Vectorization ratio add_sub	100.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	60.00
Vector-efficiency ratio all	32.09
Vector-efficiency ratio load	37.88
Vector-efficiency ratio store	33.75
Vector-efficiency ratio mul	NA
Vector-efficiency ratio add_sub	100.00
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	18.33

Path /

Average path: Display a virtual path defined by average values of all real paths

Function	main
Source file and lines	attention.cpp:52-53
Module	attention-avx512

The loop is defined in /home/eoseret/llm-attention/attention.cpp:52-53.

It is main loop of related source loop which is unrolled by 64 (including vectorization).

gain
potential
hint
expert

Code clean check

Detected a slowdown caused by scalar integer instructions (typically used for address computation). By removing them, you can lower the cost of an iteration from 69.50 to 53.25 cycles (1.31x speedup).

Workaround

Try to reorganize arrays of structures to structures of arrays
Consider to permute loops (see vectorization gain report)

Vectorization

Your loop is partially vectorized. Only 32% of vector register length is used (average across all SSE/AVX instructions). By fully vectorizing your loop, you can lower the cost of an iteration from 69.50 to 35.75 cycles (1.94x speedup).

Details

83% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

60% of SSE/AVX instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

Since your execution units are vector units, only a fully vectorized loop can use their full power.

Workaround

Pass to your compiler a micro-architecture specialization option:
- use march=native
Use vector aligned instructions:
1. align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
2. inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.

Execution units bottlenecks

Performance is limited by:

reading data from caches/RAM (load units are a bottleneck)
writing data to caches/RAM (the store unit is a bottleneck)

By removing all these bottlenecks, you can lower the cost of an iteration from 69.50 to 54.50 cycles (1.28x speedup).

Workaround

Read less array elements
Write less array elements
Provide more information to your compiler:
- hardcode the bounds of the corresponding 'for' loop

No data for this section

CALL instructions

Detected function call instructions.

Details

Calling (and then returning from) a function prevents many compiler optimizations (like vectorization), breaks control flow (which reduces pipeline performance) and executes extra instructions to save/restore the registers used inside it, which is very expensive (dozens of cycles). Consider to inline small functions.

unknown: 64 occurrences►
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53

Complex instructions

Detected COMPLEX INSTRUCTIONS.

Details

These instructions generate more than one micro-operation and only one of them can be decoded during a cycle and the extra micro-operations increase pressure on execution units.

VMOVAPS: 8 occurrences►
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53

Vector unaligned load/store instructions

Detected 24 suboptimal vector unaligned load/store instructions.

Details

VEXTRACTF128: 4 occurrences►
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
VEXTRACTF32X4: 8 occurrences►
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
VINSERTF128: 8 occurrences►
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
VINSERTF64X4: 4 occurrences►
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53

Workaround

Pass to your compiler a micro-architecture specialization option:
- use march=native
Use vector aligned instructions:
1. align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
2. inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.

Type of elements and instruction set

8 AVX-512 instructions are processing arithmetic or math operations on single precision FP elements in vector mode (sixteen at a time).

Matching between your loop (in the source code) and the binary loop

The binary loop is composed of 128 FP arithmetical operations:

128: addition or subtraction

The binary loop is loading 3344 bytes (836 single precision FP elements). The binary loop is storing 1728 bytes (432 single precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.03 FP operations per loaded or stored byte.

General properties

nb instructions	365
nb uops	436
loop length	2636
used x86 registers	4
used mmx registers	0
used xmm registers	3
used ymm registers	1
used zmm registers	7
nb stack references	14

Front-end

ASSUMED MACRO FUSION FIT IN UOP CACHE

micro-operation queue	54.50 cycles
front end	54.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	0.33	0.33	0.33	21.67	21.67	21.67	69.50	69.50	69.50	69.50	28.00	28.00	28.00	28.00	44.00	44.00
cycles	0.33	0.33	0.33	21.67	21.67	21.67	69.50	69.50	69.50	69.50	28.00	28.00	28.00	28.00	44.00	44.00

Execution ports to units layout:

P0: ALU
P1: ALU
P2: ALU
P3: ALU, BRU
P4: ALU, BRU
P5: ALU, BRU
P6 (512 bits): store address, load
P7 (512 bits): store address, load
P8 (512 bits): store address, load
P9 (512 bits): store address, load
P10 (512 bits): VPU, DIV/SQRT
P11 (512 bits): VPU, DIV/SQRT
P12 (512 bits): VPU
P13 (512 bits): VPU
P14 (512 bits): FP store data
P15 (512 bits): FP store data

Cycles executing div or sqrt instructions	NA
Longest recurrence chain latency (RecMII)	1.00

Cycles summary

Front-end	54.50
Dispatch	69.50
Data deps.	1.00
Overall L1	69.50

Vectorization ratios

INT

all	100%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	100%

all	82%
load	100%
store	100%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	100%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	53%

INT+FP

all	83%
load	100%
store	100%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	100%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	60%

Vector efficiency ratios

INT

all	25%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	25%

all	32%
load	37%
store	33%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	100%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	17%

INT+FP

all	32%
load	37%
store	33%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	100%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	18%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 69.50 cycles. At this rate:

18% of peak load performance is reached (48.12 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
19% of peak store performance is reached (24.86 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the loop is: 6f90

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	Latency	Recip. throughput	Vectorization
VMOVUPS -0xc0(%RDI,%R13,4),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVAPS 0x480(%RSP),%ZMM4	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVUPS (%RDI,%R13,4),%ZMM3	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVAPS %ZMM1,0x4c0(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VMOVUPS -0x80(%RDI,%R13,4),%ZMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVUPS -0x40(%RDI,%R13,4),%ZMM2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VSUBPS %ZMM4,%ZMM0,%ZMM5	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VSUBPS %ZMM4,%ZMM3,%ZMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VSUBPS %ZMM4,%ZMM1,%ZMM6	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VSUBPS %ZMM4,%ZMM2,%ZMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VMOVAPS %ZMM0,0x240(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VEXTRACTF32X4 $0x3,%ZMM5,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %ZMM5,0x40(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VMOVAPS %ZMM6,0x100(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VMOVAPS %ZMM1,0x1c0(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x40(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF32X4 $0x2,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0x180(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS %YMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (50.0%)
VMOVAPS 0x40(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF128 $0x1,%YMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	3	0.50	vect (25.0%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x40(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0xc0(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS 0x340(%RSP),%ZMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VINSERTF64X4 $0x1,0x180(%RSP),%ZMM0,%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (50.0%)
VADDPS %ZMM1,%ZMM0,%ZMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VMOVAPS 0x100(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVAPS %ZMM1,0x340(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VEXTRACTF32X4 $0x3,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x100(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF32X4 $0x2,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0xc0(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS %YMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (50.0%)
VMOVAPS 0x100(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF128 $0x1,%YMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	3	0.50	vect (25.0%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x100(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0x40(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS 0x400(%RSP),%ZMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VINSERTF64X4 $0x1,0xc0(%RSP),%ZMM0,%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (50.0%)
VADDPS %ZMM1,%ZMM0,%ZMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VMOVAPS 0x1c0(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVAPS %ZMM1,0x400(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VEXTRACTF32X4 $0x3,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x1c0(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF32X4 $0x2,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0x40(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS %YMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (50.0%)
VMOVAPS 0x1c0(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF128 $0x1,%YMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	3	0.50	vect (25.0%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x1c0(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0x100(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS 0x440(%RSP),%ZMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VINSERTF64X4 $0x1,0x40(%RSP),%ZMM0,%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (50.0%)
VADDPS %ZMM1,%ZMM0,%ZMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VMOVAPS 0x240(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVAPS %ZMM1,0x440(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VEXTRACTF32X4 $0x3,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0x1c0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x100(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x100(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x100(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x240(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF32X4 $0x2,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0x1c0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0x100(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS %YMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (50.0%)
VMOVAPS 0x240(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF128 $0x1,%YMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	3	0.50	vect (25.0%)
VMOVAPS %XMM0,0x1c0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x1c0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x240(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x240(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x240(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x240(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VMOVAPS 0x4c0(%RSP),%ZMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
MOV 0x3e8(%RSP),%R10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.25	N/A
MOV 0x330(%RSP),%RDI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.25	N/A
ADD $0x40,%R13	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VINSERTPS $0x30,%XMM0,%XMM2,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0x1c0(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VINSERTF64X4 $0x1,0x100(%RSP),%ZMM0,%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (50.0%)
VADDPS %ZMM1,%ZMM0,%ZMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
CMP %R13,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
JNE 6f90 <main+0x2540>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	0.33-1	N/A

Function	main
Source file and lines	attention.cpp:52-53
Module	attention-avx512

The loop is defined in /home/eoseret/llm-attention/attention.cpp:52-53.

It is main loop of related source loop which is unrolled by 64 (including vectorization).

gain
potential
hint
expert

Code clean check

Workaround

Try to reorganize arrays of structures to structures of arrays
Consider to permute loops (see vectorization gain report)

Vectorization

Details

83% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):

60% of SSE/AVX instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.

Since your execution units are vector units, only a fully vectorized loop can use their full power.

Workaround

Pass to your compiler a micro-architecture specialization option:
- use march=native
Use vector aligned instructions:
1. align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
2. inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.

Execution units bottlenecks

Performance is limited by:

reading data from caches/RAM (load units are a bottleneck)
writing data to caches/RAM (the store unit is a bottleneck)

By removing all these bottlenecks, you can lower the cost of an iteration from 69.50 to 54.50 cycles (1.28x speedup).

Workaround

Read less array elements
Write less array elements
Provide more information to your compiler:
- hardcode the bounds of the corresponding 'for' loop

No data for this section

CALL instructions

Detected function call instructions.

Details

unknown: 64 occurrences►
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53

Complex instructions

Detected COMPLEX INSTRUCTIONS.

Details

These instructions generate more than one micro-operation and only one of them can be decoded during a cycle and the extra micro-operations increase pressure on execution units.

VMOVAPS: 8 occurrences►
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53

Vector unaligned load/store instructions

Detected 24 suboptimal vector unaligned load/store instructions.

Details

VEXTRACTF128: 4 occurrences►
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
VEXTRACTF32X4: 8 occurrences►
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
VINSERTF128: 8 occurrences►
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
VINSERTF64X4: 4 occurrences►
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53
- /home/eoseret/llm-attention/attention.cpp:53

Workaround

Pass to your compiler a micro-architecture specialization option:
- use march=native
Use vector aligned instructions:
1. align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
2. inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.

Type of elements and instruction set

8 AVX-512 instructions are processing arithmetic or math operations on single precision FP elements in vector mode (sixteen at a time).

Matching between your loop (in the source code) and the binary loop

The binary loop is composed of 128 FP arithmetical operations:

128: addition or subtraction

The binary loop is loading 3344 bytes (836 single precision FP elements). The binary loop is storing 1728 bytes (432 single precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.03 FP operations per loaded or stored byte.

General properties

nb instructions	365
nb uops	436
loop length	2636
used x86 registers	4
used mmx registers	0
used xmm registers	3
used ymm registers	1
used zmm registers	7
nb stack references	14

Front-end

ASSUMED MACRO FUSION FIT IN UOP CACHE

micro-operation queue	54.50 cycles
front end	54.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	0.33	0.33	0.33	21.67	21.67	21.67	69.50	69.50	69.50	69.50	28.00	28.00	28.00	28.00	44.00	44.00
cycles	0.33	0.33	0.33	21.67	21.67	21.67	69.50	69.50	69.50	69.50	28.00	28.00	28.00	28.00	44.00	44.00

Execution ports to units layout:

P0: ALU
P1: ALU
P2: ALU
P3: ALU, BRU
P4: ALU, BRU
P5: ALU, BRU
P6 (512 bits): store address, load
P7 (512 bits): store address, load
P8 (512 bits): store address, load
P9 (512 bits): store address, load
P10 (512 bits): VPU, DIV/SQRT
P11 (512 bits): VPU, DIV/SQRT
P12 (512 bits): VPU
P13 (512 bits): VPU
P14 (512 bits): FP store data
P15 (512 bits): FP store data

Cycles executing div or sqrt instructions	NA
Longest recurrence chain latency (RecMII)	1.00

Cycles summary

Front-end	54.50
Dispatch	69.50
Data deps.	1.00
Overall L1	69.50

Vectorization ratios

INT

all	100%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	100%

all	82%
load	100%
store	100%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	100%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	53%

INT+FP

all	83%
load	100%
store	100%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	100%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	60%

Vector efficiency ratios

INT

all	25%
load	NA (no load vectorizable/vectorized instructions)
store	NA (no store vectorizable/vectorized instructions)
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
other	25%

all	32%
load	37%
store	33%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	100%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	17%

INT+FP

all	32%
load	37%
store	33%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	100%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	18%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 69.50 cycles. At this rate:

18% of peak load performance is reached (48.12 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
19% of peak store performance is reached (24.86 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the loop is: 6f90

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15	Latency	Recip. throughput	Vectorization
VMOVUPS -0xc0(%RDI,%R13,4),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVAPS 0x480(%RSP),%ZMM4	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVUPS (%RDI,%R13,4),%ZMM3	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVAPS %ZMM1,0x4c0(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VMOVUPS -0x80(%RDI,%R13,4),%ZMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVUPS -0x40(%RDI,%R13,4),%ZMM2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VSUBPS %ZMM4,%ZMM0,%ZMM5	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VSUBPS %ZMM4,%ZMM3,%ZMM0	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VSUBPS %ZMM4,%ZMM1,%ZMM6	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VSUBPS %ZMM4,%ZMM2,%ZMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VMOVAPS %ZMM0,0x240(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VEXTRACTF32X4 $0x3,%ZMM5,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %ZMM5,0x40(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VMOVAPS %ZMM6,0x100(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VMOVAPS %ZMM1,0x1c0(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x40(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF32X4 $0x2,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0x180(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS %YMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (50.0%)
VMOVAPS 0x40(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF128 $0x1,%YMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	3	0.50	vect (25.0%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0xc0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x40(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x160(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x160(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0xc0(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS 0x340(%RSP),%ZMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VINSERTF64X4 $0x1,0x180(%RSP),%ZMM0,%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (50.0%)
VADDPS %ZMM1,%ZMM0,%ZMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VMOVAPS 0x100(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVAPS %ZMM1,0x340(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VEXTRACTF32X4 $0x3,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x100(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF32X4 $0x2,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0xc0(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS %YMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (50.0%)
VMOVAPS 0x100(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF128 $0x1,%YMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	3	0.50	vect (25.0%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x40(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x100(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x180(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x180(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0x40(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS 0x400(%RSP),%ZMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VINSERTF64X4 $0x1,0xc0(%RSP),%ZMM0,%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (50.0%)
VADDPS %ZMM1,%ZMM0,%ZMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VMOVAPS 0x1c0(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVAPS %ZMM1,0x400(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VEXTRACTF32X4 $0x3,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x1c0(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF32X4 $0x2,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0x40(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS %YMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (50.0%)
VMOVAPS 0x1c0(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF128 $0x1,%YMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	3	0.50	vect (25.0%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x100(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x1c0(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0xc0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0xc0(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0x100(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS 0x440(%RSP),%ZMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VINSERTF64X4 $0x1,0x40(%RSP),%ZMM0,%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (50.0%)
VADDPS %ZMM1,%ZMM0,%ZMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
VMOVAPS 0x240(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VMOVAPS %ZMM1,0x440(%RSP)	2	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	1	1	4	1	vect (100.0%)
VEXTRACTF32X4 $0x3,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0x1c0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x100(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x100(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x100(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x240(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF32X4 $0x2,%ZMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	5	0.50	vect (25.0%)
VMOVAPS %XMM0,0x1c0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0x100(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VMOVAPS %YMM0,0x100(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (50.0%)
VMOVAPS 0x240(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VEXTRACTF128 $0x1,%YMM0,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	0	3	0.50	vect (25.0%)
VMOVAPS %XMM0,0x1c0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x1c0(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x30,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x1c0(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVAPS 0x240(%RSP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VMOVSHDUP 0x240(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	1	0.50	vect (12.5%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x10,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPD $0x1,0x240(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VINSERTPS $0x20,%XMM0,%XMM1,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VMOVAPS %XMM0,0x40(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0.50	0.50	4	0.50	vect (25.0%)
VPERMILPS $-0x1,0x240(%RSP),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0	0	1	0.50	vect (25.0%)
CALL 89b0 <@plt_start@+0x20>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0.50	N/A
VMOVAPS 0x40(%RSP),%XMM2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (25.0%)
VMOVAPS 0x4c0(%RSP),%ZMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.50	vect (100.0%)
MOV 0x3e8(%RSP),%R10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.25	N/A
MOV 0x330(%RSP),%RDI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	3	0.25	N/A
ADD $0x40,%R13	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VINSERTPS $0x30,%XMM0,%XMM2,%XMM0	1	0	0	0	0	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.25	scal (6.3%)
VINSERTF128 $0x1,0x1c0(%RSP),%YMM0,%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (25.0%)
VINSERTF64X4 $0x1,0x100(%RSP),%ZMM0,%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0.50	0.50	0	0	0	1	0.50	vect (50.0%)
VADDPS %ZMM1,%ZMM0,%ZMM1	1	0	0	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0	2	0.50	vect (100.0%)
CMP %R13,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
JNE 6f90 <main+0x2540>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	0.33-1	N/A

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

Workaround

CALL instructions

Details

Complex instructions

Details

Vector unaligned load/store instructions

Details

Workaround

Type of elements and instruction set

Matching between your loop (in the source code) and the binary loop

Arithmetic intensity

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

Workaround

CALL instructions

Details

Complex instructions

Details

Vector unaligned load/store instructions

Details

Workaround

Type of elements and instruction set

Matching between your loop (in the source code) and the binary loop

Arithmetic intensity

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code