OV - attention-aocc-znver5-512 - Loop 38 - attention-aocc-znver5-512

	Loop Id: 38	Module: attention-aocc-znver5-512	Source: attention_v2.cpp:52-53	Coverage: 0.25%

0x6ad0 VMOVUPS	-0xc0(%R12,%RAX,4),%ZMM0    [2]

0x6ad8 VMOVAPS	0x4c0(%RSP),%ZMM4    [1]

0x6ae0 VMOVUPS	(%R12,%RAX,4),%ZMM3    [2]

0x6ae7 VMOVAPS	%ZMM1,0x500(%RSP)    [1]

0x6aef VMOVUPS	-0x80(%R12,%RAX,4),%ZMM1    [2]

0x6af7 VMOVUPS	-0x40(%R12,%RAX,4),%ZMM2    [2]

0x6aff MOV	%RAX,0x480(%RSP)    [1]

0x6b07 VSUBPS	%ZMM4,%ZMM0,%ZMM5

0x6b0d VSUBPS	%ZMM4,%ZMM3,%ZMM0

0x6b13 VSUBPS	%ZMM4,%ZMM1,%ZMM6

0x6b19 VSUBPS	%ZMM4,%ZMM2,%ZMM1

0x6b1f VMOVAPS	%ZMM0,0x240(%RSP)    [1]

0x6b27 VEXTRACTF32X4	$0x3,%ZMM5,%XMM0

0x6b2e VMOVAPS	%ZMM5,0xc0(%RSP)    [1]

0x6b36 VMOVAPS	%ZMM6,0x180(%RSP)    [1]

0x6b3e VMOVAPS	%ZMM1,0x1c0(%RSP)    [1]

0x6b46 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x6b4c VZEROUPPER

0x6b4f CALL	8830 <@plt_start@+0x20>

0x6b54 VMOVAPS	%XMM0,0x140(%RSP)    [1]

0x6b5d VMOVSHDUP	0x20(%RSP),%XMM0    [1]

0x6b63 CALL	8830 <@plt_start@+0x20>

0x6b68 VMOVAPS	0x140(%RSP),%XMM1    [1]

0x6b71 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x6b77 VMOVAPS	%XMM0,0x140(%RSP)    [1]

0x6b80 VPERMILPD	$0x1,0x20(%RSP),%XMM0    [1]

0x6b88 CALL	8830 <@plt_start@+0x20>

0x6b8d VMOVAPS	0x140(%RSP),%XMM1    [1]

0x6b96 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x6b9c VMOVAPS	%XMM0,0x140(%RSP)    [1]

0x6ba5 VPERMILPS	$-0x1,0x20(%RSP),%XMM0    [1]

0x6bad CALL	8830 <@plt_start@+0x20>

0x6bb2 VMOVAPS	0x140(%RSP),%XMM1    [1]

0x6bbb VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x6bc1 VMOVAPS	%XMM0,0x140(%RSP)    [1]

0x6bca VMOVAPS	0xc0(%RSP),%ZMM0    [1]

0x6bd2 VEXTRACTF32X4	$0x2,%ZMM0,%XMM0

0x6bd9 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x6bdf VZEROUPPER

0x6be2 CALL	8830 <@plt_start@+0x20>

0x6be7 VMOVAPS	%XMM0,0x90(%RSP)    [1]

0x6bf0 VMOVSHDUP	0x20(%RSP),%XMM0    [1]

0x6bf6 CALL	8830 <@plt_start@+0x20>

0x6bfb VMOVAPS	0x90(%RSP),%XMM1    [1]

0x6c04 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x6c0a VMOVAPS	%XMM0,0x90(%RSP)    [1]

0x6c13 VPERMILPD	$0x1,0x20(%RSP),%XMM0    [1]

0x6c1b CALL	8830 <@plt_start@+0x20>

0x6c20 VMOVAPS	0x90(%RSP),%XMM1    [1]

0x6c29 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x6c2f VMOVAPS	%XMM0,0x90(%RSP)    [1]

0x6c38 VPERMILPS	$-0x1,0x20(%RSP),%XMM0    [1]

0x6c40 CALL	8830 <@plt_start@+0x20>

0x6c45 VMOVAPS	0x90(%RSP),%XMM1    [1]

0x6c4e VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x6c54 VINSERTF128	$0x1,0x140(%RSP),%YMM0,%YMM0    [1]

0x6c5f VMOVAPS	%YMM0,0x140(%RSP)    [1]

0x6c68 VMOVAPS	0xc0(%RSP),%ZMM0    [1]

0x6c70 VEXTRACTF128	$0x1,%YMM0,%XMM0

0x6c76 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x6c7c VZEROUPPER

0x6c7f CALL	8830 <@plt_start@+0x20>

0x6c84 VMOVAPS	%XMM0,0x90(%RSP)    [1]

0x6c8d VMOVSHDUP	0x20(%RSP),%XMM0    [1]

0x6c93 CALL	8830 <@plt_start@+0x20>

0x6c98 VMOVAPS	0x90(%RSP),%XMM1    [1]

0x6ca1 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x6ca7 VMOVAPS	%XMM0,0x90(%RSP)    [1]

0x6cb0 VPERMILPD	$0x1,0x20(%RSP),%XMM0    [1]

0x6cb8 CALL	8830 <@plt_start@+0x20>

0x6cbd VMOVAPS	0x90(%RSP),%XMM1    [1]

0x6cc6 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x6ccc VMOVAPS	%XMM0,0x90(%RSP)    [1]

0x6cd5 VPERMILPS	$-0x1,0x20(%RSP),%XMM0    [1]

0x6cdd CALL	8830 <@plt_start@+0x20>

0x6ce2 VMOVAPS	0x90(%RSP),%XMM1    [1]

0x6ceb VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x6cf1 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x6cf7 VMOVAPS	0xc0(%RSP),%ZMM0    [1]

0x6cff VZEROUPPER

0x6d02 CALL	8830 <@plt_start@+0x20>

0x6d07 VMOVAPS	%XMM0,0x90(%RSP)    [1]

0x6d10 VMOVSHDUP	0xc0(%RSP),%XMM0    [1]

0x6d19 CALL	8830 <@plt_start@+0x20>

0x6d1e VMOVAPS	0x90(%RSP),%XMM1    [1]

0x6d27 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x6d2d VMOVAPS	%XMM0,0x90(%RSP)    [1]

0x6d36 VPERMILPD	$0x1,0xc0(%RSP),%XMM0    [1]

0x6d41 CALL	8830 <@plt_start@+0x20>

0x6d46 VMOVAPS	0x90(%RSP),%XMM1    [1]

0x6d4f VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x6d55 VMOVAPS	%XMM0,0x90(%RSP)    [1]

0x6d5e VPERMILPS	$-0x1,0xc0(%RSP),%XMM0    [1]

0x6d69 CALL	8830 <@plt_start@+0x20>

0x6d6e VMOVAPS	0x90(%RSP),%XMM1    [1]

0x6d77 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x6d7d VINSERTF128	$0x1,0x20(%RSP),%YMM0,%YMM0    [1]

0x6d85 VMOVAPS	0x300(%RSP),%ZMM1    [1]

0x6d8d VINSERTF64X4	$0x1,0x140(%RSP),%ZMM0,%ZMM0    [1]

0x6d96 VADDPS	%ZMM1,%ZMM0,%ZMM1

0x6d9c VMOVAPS	0x180(%RSP),%ZMM0    [1]

0x6da4 VMOVAPS	%ZMM1,0x300(%RSP)    [1]

0x6dac VEXTRACTF32X4	$0x3,%ZMM0,%XMM0

0x6db3 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x6dbc VZEROUPPER

0x6dbf CALL	8830 <@plt_start@+0x20>

0x6dc4 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x6dca VMOVSHDUP	0xc0(%RSP),%XMM0    [1]

0x6dd3 CALL	8830 <@plt_start@+0x20>

0x6dd8 VMOVAPS	0x20(%RSP),%XMM1    [1]

0x6dde VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x6de4 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x6dea VPERMILPD	$0x1,0xc0(%RSP),%XMM0    [1]

0x6df5 CALL	8830 <@plt_start@+0x20>

0x6dfa VMOVAPS	0x20(%RSP),%XMM1    [1]

0x6e00 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x6e06 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x6e0c VPERMILPS	$-0x1,0xc0(%RSP),%XMM0    [1]

0x6e17 CALL	8830 <@plt_start@+0x20>

0x6e1c VMOVAPS	0x20(%RSP),%XMM1    [1]

0x6e22 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x6e28 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x6e2e VMOVAPS	0x180(%RSP),%ZMM0    [1]

0x6e36 VEXTRACTF32X4	$0x2,%ZMM0,%XMM0

0x6e3d VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x6e46 VZEROUPPER

0x6e49 CALL	8830 <@plt_start@+0x20>

0x6e4e VMOVAPS	%XMM0,0x140(%RSP)    [1]

0x6e57 VMOVSHDUP	0xc0(%RSP),%XMM0    [1]

0x6e60 CALL	8830 <@plt_start@+0x20>

0x6e65 VMOVAPS	0x140(%RSP),%XMM1    [1]

0x6e6e VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x6e74 VMOVAPS	%XMM0,0x140(%RSP)    [1]

0x6e7d VPERMILPD	$0x1,0xc0(%RSP),%XMM0    [1]

0x6e88 CALL	8830 <@plt_start@+0x20>

0x6e8d VMOVAPS	0x140(%RSP),%XMM1    [1]

0x6e96 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x6e9c VMOVAPS	%XMM0,0x140(%RSP)    [1]

0x6ea5 VPERMILPS	$-0x1,0xc0(%RSP),%XMM0    [1]

0x6eb0 CALL	8830 <@plt_start@+0x20>

0x6eb5 VMOVAPS	0x140(%RSP),%XMM1    [1]

0x6ebe VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x6ec4 VINSERTF128	$0x1,0x20(%RSP),%YMM0,%YMM0    [1]

0x6ecc VMOVAPS	%YMM0,0x20(%RSP)    [1]

0x6ed2 VMOVAPS	0x180(%RSP),%ZMM0    [1]

0x6eda VEXTRACTF128	$0x1,%YMM0,%XMM0

0x6ee0 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x6ee9 VZEROUPPER

0x6eec CALL	8830 <@plt_start@+0x20>

0x6ef1 VMOVAPS	%XMM0,0x140(%RSP)    [1]

0x6efa VMOVSHDUP	0xc0(%RSP),%XMM0    [1]

0x6f03 CALL	8830 <@plt_start@+0x20>

0x6f08 VMOVAPS	0x140(%RSP),%XMM1    [1]

0x6f11 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x6f17 VMOVAPS	%XMM0,0x140(%RSP)    [1]

0x6f20 VPERMILPD	$0x1,0xc0(%RSP),%XMM0    [1]

0x6f2b CALL	8830 <@plt_start@+0x20>

0x6f30 VMOVAPS	0x140(%RSP),%XMM1    [1]

0x6f39 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x6f3f VMOVAPS	%XMM0,0x140(%RSP)    [1]

0x6f48 VPERMILPS	$-0x1,0xc0(%RSP),%XMM0    [1]

0x6f53 CALL	8830 <@plt_start@+0x20>

0x6f58 VMOVAPS	0x140(%RSP),%XMM1    [1]

0x6f61 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x6f67 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x6f70 VMOVAPS	0x180(%RSP),%ZMM0    [1]

0x6f78 VZEROUPPER

0x6f7b CALL	8830 <@plt_start@+0x20>

0x6f80 VMOVAPS	%XMM0,0x140(%RSP)    [1]

0x6f89 VMOVSHDUP	0x180(%RSP),%XMM0    [1]

0x6f92 CALL	8830 <@plt_start@+0x20>

0x6f97 VMOVAPS	0x140(%RSP),%XMM1    [1]

0x6fa0 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x6fa6 VMOVAPS	%XMM0,0x140(%RSP)    [1]

0x6faf VPERMILPD	$0x1,0x180(%RSP),%XMM0    [1]

0x6fba CALL	8830 <@plt_start@+0x20>

0x6fbf VMOVAPS	0x140(%RSP),%XMM1    [1]

0x6fc8 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x6fce VMOVAPS	%XMM0,0x140(%RSP)    [1]

0x6fd7 VPERMILPS	$-0x1,0x180(%RSP),%XMM0    [1]

0x6fe2 CALL	8830 <@plt_start@+0x20>

0x6fe7 VMOVAPS	0x140(%RSP),%XMM1    [1]

0x6ff0 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x6ff6 VINSERTF128	$0x1,0xc0(%RSP),%YMM0,%YMM0    [1]

0x7001 VMOVAPS	0x400(%RSP),%ZMM1    [1]

0x7009 VINSERTF64X4	$0x1,0x20(%RSP),%ZMM0,%ZMM0    [1]

0x7012 VADDPS	%ZMM1,%ZMM0,%ZMM1

0x7018 VMOVAPS	0x1c0(%RSP),%ZMM0    [1]

0x7020 VMOVAPS	%ZMM1,0x400(%RSP)    [1]

0x7028 VEXTRACTF32X4	$0x3,%ZMM0,%XMM0

0x702f VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x7038 VZEROUPPER

0x703b CALL	8830 <@plt_start@+0x20>

0x7040 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x7049 VMOVSHDUP	0x180(%RSP),%XMM0    [1]

0x7052 CALL	8830 <@plt_start@+0x20>

0x7057 VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x7060 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x7066 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x706f VPERMILPD	$0x1,0x180(%RSP),%XMM0    [1]

0x707a CALL	8830 <@plt_start@+0x20>

0x707f VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x7088 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x708e VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x7097 VPERMILPS	$-0x1,0x180(%RSP),%XMM0    [1]

0x70a2 CALL	8830 <@plt_start@+0x20>

0x70a7 VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x70b0 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x70b6 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x70bf VMOVAPS	0x1c0(%RSP),%ZMM0    [1]

0x70c7 VEXTRACTF32X4	$0x2,%ZMM0,%XMM0

0x70ce VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x70d7 VZEROUPPER

0x70da CALL	8830 <@plt_start@+0x20>

0x70df VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x70e5 VMOVSHDUP	0x180(%RSP),%XMM0    [1]

0x70ee CALL	8830 <@plt_start@+0x20>

0x70f3 VMOVAPS	0x20(%RSP),%XMM1    [1]

0x70f9 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x70ff VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x7105 VPERMILPD	$0x1,0x180(%RSP),%XMM0    [1]

0x7110 CALL	8830 <@plt_start@+0x20>

0x7115 VMOVAPS	0x20(%RSP),%XMM1    [1]

0x711b VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x7121 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x7127 VPERMILPS	$-0x1,0x180(%RSP),%XMM0    [1]

0x7132 CALL	8830 <@plt_start@+0x20>

0x7137 VMOVAPS	0x20(%RSP),%XMM1    [1]

0x713d VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x7143 VINSERTF128	$0x1,0xc0(%RSP),%YMM0,%YMM0    [1]

0x714e VMOVAPS	%YMM0,0xc0(%RSP)    [1]

0x7157 VMOVAPS	0x1c0(%RSP),%ZMM0    [1]

0x715f VEXTRACTF128	$0x1,%YMM0,%XMM0

0x7165 VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x716e VZEROUPPER

0x7171 CALL	8830 <@plt_start@+0x20>

0x7176 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x717c VMOVSHDUP	0x180(%RSP),%XMM0    [1]

0x7185 CALL	8830 <@plt_start@+0x20>

0x718a VMOVAPS	0x20(%RSP),%XMM1    [1]

0x7190 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x7196 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x719c VPERMILPD	$0x1,0x180(%RSP),%XMM0    [1]

0x71a7 CALL	8830 <@plt_start@+0x20>

0x71ac VMOVAPS	0x20(%RSP),%XMM1    [1]

0x71b2 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x71b8 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x71be VPERMILPS	$-0x1,0x180(%RSP),%XMM0    [1]

0x71c9 CALL	8830 <@plt_start@+0x20>

0x71ce VMOVAPS	0x20(%RSP),%XMM1    [1]

0x71d4 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x71da VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x71e3 VMOVAPS	0x1c0(%RSP),%ZMM0    [1]

0x71eb VZEROUPPER

0x71ee CALL	8830 <@plt_start@+0x20>

0x71f3 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x71f9 VMOVSHDUP	0x1c0(%RSP),%XMM0    [1]

0x7202 CALL	8830 <@plt_start@+0x20>

0x7207 VMOVAPS	0x20(%RSP),%XMM1    [1]

0x720d VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x7213 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x7219 VPERMILPD	$0x1,0x1c0(%RSP),%XMM0    [1]

0x7224 CALL	8830 <@plt_start@+0x20>

0x7229 VMOVAPS	0x20(%RSP),%XMM1    [1]

0x722f VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x7235 VMOVAPS	%XMM0,0x20(%RSP)    [1]

0x723b VPERMILPS	$-0x1,0x1c0(%RSP),%XMM0    [1]

0x7246 CALL	8830 <@plt_start@+0x20>

0x724b VMOVAPS	0x20(%RSP),%XMM1    [1]

0x7251 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x7257 VINSERTF128	$0x1,0x180(%RSP),%YMM0,%YMM0    [1]

0x7262 VMOVAPS	0x440(%RSP),%ZMM1    [1]

0x726a VINSERTF64X4	$0x1,0xc0(%RSP),%ZMM0,%ZMM0    [1]

0x7273 VADDPS	%ZMM1,%ZMM0,%ZMM1

0x7279 VMOVAPS	0x240(%RSP),%ZMM0    [1]

0x7281 VMOVAPS	%ZMM1,0x440(%RSP)    [1]

0x7289 VEXTRACTF32X4	$0x3,%ZMM0,%XMM0

0x7290 VMOVAPS	%XMM0,0x1c0(%RSP)    [1]

0x7299 VZEROUPPER

0x729c CALL	8830 <@plt_start@+0x20>

0x72a1 VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x72aa VMOVSHDUP	0x1c0(%RSP),%XMM0    [1]

0x72b3 CALL	8830 <@plt_start@+0x20>

0x72b8 VMOVAPS	0x180(%RSP),%XMM1    [1]

0x72c1 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x72c7 VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x72d0 VPERMILPD	$0x1,0x1c0(%RSP),%XMM0    [1]

0x72db CALL	8830 <@plt_start@+0x20>

0x72e0 VMOVAPS	0x180(%RSP),%XMM1    [1]

0x72e9 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x72ef VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x72f8 VPERMILPS	$-0x1,0x1c0(%RSP),%XMM0    [1]

0x7303 CALL	8830 <@plt_start@+0x20>

0x7308 VMOVAPS	0x180(%RSP),%XMM1    [1]

0x7311 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x7317 VMOVAPS	%XMM0,0x180(%RSP)    [1]

0x7320 VMOVAPS	0x240(%RSP),%ZMM0    [1]

0x7328 VEXTRACTF32X4	$0x2,%ZMM0,%XMM0

0x732f VMOVAPS	%XMM0,0x1c0(%RSP)    [1]

0x7338 VZEROUPPER

0x733b CALL	8830 <@plt_start@+0x20>

0x7340 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x7349 VMOVSHDUP	0x1c0(%RSP),%XMM0    [1]

0x7352 CALL	8830 <@plt_start@+0x20>

0x7357 VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x7360 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x7366 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x736f VPERMILPD	$0x1,0x1c0(%RSP),%XMM0    [1]

0x737a CALL	8830 <@plt_start@+0x20>

0x737f VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x7388 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x738e VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x7397 VPERMILPS	$-0x1,0x1c0(%RSP),%XMM0    [1]

0x73a2 CALL	8830 <@plt_start@+0x20>

0x73a7 VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x73b0 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x73b6 VINSERTF128	$0x1,0x180(%RSP),%YMM0,%YMM0    [1]

0x73c1 VMOVAPS	%YMM0,0x180(%RSP)    [1]

0x73ca VMOVAPS	0x240(%RSP),%ZMM0    [1]

0x73d2 VEXTRACTF128	$0x1,%YMM0,%XMM0

0x73d8 VMOVAPS	%XMM0,0x1c0(%RSP)    [1]

0x73e1 VZEROUPPER

0x73e4 CALL	8830 <@plt_start@+0x20>

0x73e9 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x73f2 VMOVSHDUP	0x1c0(%RSP),%XMM0    [1]

0x73fb CALL	8830 <@plt_start@+0x20>

0x7400 VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x7409 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x740f VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x7418 VPERMILPD	$0x1,0x1c0(%RSP),%XMM0    [1]

0x7423 CALL	8830 <@plt_start@+0x20>

0x7428 VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x7431 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x7437 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x7440 VPERMILPS	$-0x1,0x1c0(%RSP),%XMM0    [1]

0x744b CALL	8830 <@plt_start@+0x20>

0x7450 VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x7459 VINSERTPS	$0x30,%XMM0,%XMM1,%XMM0

0x745f VMOVAPS	%XMM0,0x1c0(%RSP)    [1]

0x7468 VMOVAPS	0x240(%RSP),%ZMM0    [1]

0x7470 VZEROUPPER

0x7473 CALL	8830 <@plt_start@+0x20>

0x7478 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x7481 VMOVSHDUP	0x240(%RSP),%XMM0    [1]

0x748a CALL	8830 <@plt_start@+0x20>

0x748f VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x7498 VINSERTPS	$0x10,%XMM0,%XMM1,%XMM0

0x749e VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x74a7 VPERMILPD	$0x1,0x240(%RSP),%XMM0    [1]

0x74b2 CALL	8830 <@plt_start@+0x20>

0x74b7 VMOVAPS	0xc0(%RSP),%XMM1    [1]

0x74c0 VINSERTPS	$0x20,%XMM0,%XMM1,%XMM0

0x74c6 VMOVAPS	%XMM0,0xc0(%RSP)    [1]

0x74cf VPERMILPS	$-0x1,0x240(%RSP),%XMM0    [1]

0x74da CALL	8830 <@plt_start@+0x20>

0x74df VMOVAPS	0xc0(%RSP),%XMM2    [1]

0x74e8 VMOVAPS	0x500(%RSP),%ZMM1    [1]

0x74f0 MOV	0x480(%RSP),%RAX    [1]

0x74f8 MOV	0x3f0(%RSP),%R11    [1]

0x7500 ADD	$0x40,%RAX

0x7504 VINSERTPS	$0x30,%XMM0,%XMM2,%XMM0

0x750a VINSERTF128	$0x1,0x1c0(%RSP),%YMM0,%YMM0    [1]

0x7515 VINSERTF64X4	$0x1,0x180(%RSP),%ZMM0,%ZMM0    [1]

0x751e VADDPS	%ZMM1,%ZMM0,%ZMM1

0x7524 CMP	%RAX,%R11

0x7527 JNE	6ad0

/home/eoseret/llm-attention/attention_v2.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	1.02
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	2.23
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.02
Bottlenecks	micro-operation queue,
Function	main
Source	attention_v2.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	54.63
CQA cycles if no scalar integer	53.50
CQA cycles if FP arith vectorized	54.63
CQA cycles if fully vectorized	24.50
Front-end cycles	54.63
P0 cycles	0.33
P1 cycles	0.33
P2 cycles	0.33
P3 cycles	0.33
P4 cycles	0.33
P5 cycles	0.33
P6 cycles	53.75
P7 cycles	53.75
P8 cycles	53.75
P9 cycles	53.75
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	0
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	366.00
Nb uops	437.00
Nb loads	134.00
Nb stores	81.00
Nb stack references	14.00
FLOP/cycle	2.34
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	93.00
Bytes prefetched	0.00
Bytes loaded	3344.00
Bytes stored	1736.00
Stride 0	1.00
Stride 1	0.00
Stride n	0.00
Stride unknown	0.00
Stride indirect	1.00
Vectorization ratio all	83.50
Vectorization ratio load	100.00
Vectorization ratio store	98.77
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	32.03
Vector-efficiency ratio load	37.88
Vector-efficiency ratio store	33.49
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	1.02
CQA speedup if FP arith vectorized	1.00
CQA speedup if fully vectorized	2.23
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.02
Bottlenecks	micro-operation queue,
Function	main
Source	attention_v2.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	54.63
CQA cycles if no scalar integer	53.50
CQA cycles if FP arith vectorized	54.63
CQA cycles if fully vectorized	24.50
Front-end cycles	54.63
P0 cycles	0.33
P1 cycles	0.33
P2 cycles	0.33
P3 cycles	0.33
P4 cycles	0.33
P5 cycles	0.33
P6 cycles	53.75
P7 cycles	53.75
P8 cycles	53.75
P9 cycles	53.75
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	0
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	366.00
Nb uops	437.00
Nb loads	134.00
Nb stores	81.00
Nb stack references	14.00
FLOP/cycle	2.34
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	93.00
Bytes prefetched	0.00
Bytes loaded	3344.00
Bytes stored	1736.00
Stride 0	1.00
Stride 1	0.00
Stride n	0.00
Stride unknown	0.00
Stride indirect	1.00
Vectorization ratio all	83.50
Vectorization ratio load	100.00
Vectorization ratio store	98.77
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	32.03
Vector-efficiency ratio load	37.88
Vector-efficiency ratio store	33.49
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_v2.cpp:52-53
Module	attention-aocc-znver5-512

The loop is defined in /home/eoseret/llm-attention/attention_v2.cpp:52-53.

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

gain
potential
hint
expert

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 54.63 to 24.50 cycles (2.23x speedup).

Details

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

98% of SSE/AVX stores are used in vector version.
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

Found no such bottlenecks but see expert reports for more complex bottlenecks.

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_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.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_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53

Slow data structures access

Detected data structures (typically arrays) that cannot be efficiently read/written

Details

Irregular (variable stride) or indirect: 1 occurrence(s)

Non-unit stride (uncontiguous) accesses are not efficiently using data caches

Workaround

Try to remove indirect accesses. If applicable, precompute elements out of the innermost loop.

Vector unaligned load/store instructions

Detected 24 suboptimal vector unaligned load/store instructions.

Details

VEXTRACTF128: 4 occurrences►
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
VEXTRACTF32X4: 8 occurrences►
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
VINSERTF128: 8 occurrences►
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
VINSERTF64X4: 4 occurrences►
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.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 1736 bytes (434 single precision FP elements).

Arithmetic intensity

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

General properties

nb instructions	366
nb uops	437
loop length	2653
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	54.63 cycles
front end	54.63 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	0.33	0.33	0.33	53.75	53.75	53.75	53.75	28.00	28.00	28.00	28.00	44.00	44.00
cycles	0.33	0.33	0.33	0.33	0.33	0.33	53.75	53.75	53.75	53.75	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)	0.00

Cycles summary

Front-end	54.63
Dispatch	53.75
Data deps.	0.00
Overall L1	54.63

Vectorization ratios

INT

all	94%
load	NA (no load vectorizable/vectorized instructions)
store	0%
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	98%
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	24%
load	NA (no load vectorizable/vectorized instructions)
store	12%
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 54.63 cycles. At this rate:

23% of peak load performance is reached (61.22 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
24% of peak store performance is reached (31.78 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Performance is limited by instruction throughput (loading/decoding program instructions to execution core) (front-end is a bottleneck).

ASM code

In the binary file, the address of the loop is: 6ad0

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(%R12,%RAX,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 0x4c0(%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 (%R12,%RAX,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 %ZMM1,0x500(%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(%R12,%RAX,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(%R12,%RAX,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%)
MOV %RAX,0x480(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	4	0.50	scal (12.5%)
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 %ZMM5,0xc0(%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 %ZMM6,0x180(%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 %XMM0,0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x140(%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 0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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 $0x1,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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 $-0x1,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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 0xc0(%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 %XMM0,0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x90(%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 0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x90(%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 $0x1,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x90(%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 $-0x1,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VINSERTF128 $0x1,0x140(%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	2	0.50	vect (25.0%)
VMOVAPS %YMM0,0x140(%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 0xc0(%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	0.25	0.25	0.25	0.25	0	0	2	0.25	vect (25.0%)
VMOVAPS %XMM0,0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x90(%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 0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x90(%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 $0x1,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x90(%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 $-0x1,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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 0xc0(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x90(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x90(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x90(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VINSERTF128 $0x1,0x20(%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	2	0.50	vect (25.0%)
VMOVAPS 0x300(%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 $0x1,0x140(%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	1-2	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 0x180(%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 %ZMM1,0x300(%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 %XMM0,0xc0(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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 0x180(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x140(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VINSERTF128 $0x1,0x20(%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	2	0.50	vect (25.0%)
VMOVAPS %YMM0,0x20(%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 0x180(%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	0.25	0.25	0.25	0.25	0	0	2	0.25	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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x140(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	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 0x180(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x140(%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 0x180(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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 $0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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 $-0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	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	2	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 $0x1,0x20(%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	1-2	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 %XMM0,0x180(%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%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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 0x180(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 $0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 $-0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 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 %XMM0,0x180(%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%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x20(%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 0x180(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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 $0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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 $-0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	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	2	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 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	0.25	0.25	0.25	0.25	0	0	2	0.25	vect (25.0%)
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%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x20(%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 0x180(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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 $0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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 $-0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	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 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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	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	2	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 $0x1,0xc0(%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	1-2	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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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 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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 $0x1,0x1c0(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 $-0x1,0x1c0(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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	2	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 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	0.25	0.25	0.25	0.25	0	0	2	0.25	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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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 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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 $0x1,0x240(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 $-0x1,0x240(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0xc0(%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 0x500(%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 0x480(%RSP),%RAX	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 0x3f0(%RSP),%R11	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,%RAX	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	2	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	2	0.50	vect (25.0%)
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	1-2	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 %RAX,%R11	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 6ad0 <main+0x2170>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	0.33-0.50	N/A

Function	main
Source file and lines	attention_v2.cpp:52-53
Module	attention-aocc-znver5-512

The loop is defined in /home/eoseret/llm-attention/attention_v2.cpp:52-53.

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

gain
potential
hint
expert

Vectorization

Details

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

98% of SSE/AVX stores are used in vector version.
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

Found no such bottlenecks but see expert reports for more complex bottlenecks.

No data for this section

CALL instructions

Detected function call instructions.

Details

unknown: 64 occurrences►
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.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_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53

Slow data structures access

Detected data structures (typically arrays) that cannot be efficiently read/written

Details

Irregular (variable stride) or indirect: 1 occurrence(s)

Non-unit stride (uncontiguous) accesses are not efficiently using data caches

Workaround

Try to remove indirect accesses. If applicable, precompute elements out of the innermost loop.

Vector unaligned load/store instructions

Detected 24 suboptimal vector unaligned load/store instructions.

Details

VEXTRACTF128: 4 occurrences►
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
VEXTRACTF32X4: 8 occurrences►
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
VINSERTF128: 8 occurrences►
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
VINSERTF64X4: 4 occurrences►
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.cpp:53
- /home/eoseret/llm-attention/attention_v2.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 1736 bytes (434 single precision FP elements).

Arithmetic intensity

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

General properties

nb instructions	366
nb uops	437
loop length	2653
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	54.63 cycles
front end	54.63 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	0.33	0.33	0.33	53.75	53.75	53.75	53.75	28.00	28.00	28.00	28.00	44.00	44.00
cycles	0.33	0.33	0.33	0.33	0.33	0.33	53.75	53.75	53.75	53.75	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)	0.00

Cycles summary

Front-end	54.63
Dispatch	53.75
Data deps.	0.00
Overall L1	54.63

Vectorization ratios

INT

all	94%
load	NA (no load vectorizable/vectorized instructions)
store	0%
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	98%
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	24%
load	NA (no load vectorizable/vectorized instructions)
store	12%
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 54.63 cycles. At this rate:

23% of peak load performance is reached (61.22 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
24% of peak store performance is reached (31.78 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Performance is limited by instruction throughput (loading/decoding program instructions to execution core) (front-end is a bottleneck).

ASM code

In the binary file, the address of the loop is: 6ad0

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(%R12,%RAX,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 0x4c0(%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 (%R12,%RAX,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 %ZMM1,0x500(%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(%R12,%RAX,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(%R12,%RAX,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%)
MOV %RAX,0x480(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	4	0.50	scal (12.5%)
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 %ZMM5,0xc0(%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 %ZMM6,0x180(%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 %XMM0,0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x140(%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 0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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 $0x1,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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 $-0x1,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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 0xc0(%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 %XMM0,0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x90(%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 0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x90(%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 $0x1,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x90(%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 $-0x1,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VINSERTF128 $0x1,0x140(%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	2	0.50	vect (25.0%)
VMOVAPS %YMM0,0x140(%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 0xc0(%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	0.25	0.25	0.25	0.25	0	0	2	0.25	vect (25.0%)
VMOVAPS %XMM0,0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x90(%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 0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x90(%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 $0x1,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x90(%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 $-0x1,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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 0xc0(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x90(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x90(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x90(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x90(%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	2	0.25	scal (6.3%)
VINSERTF128 $0x1,0x20(%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	2	0.50	vect (25.0%)
VMOVAPS 0x300(%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 $0x1,0x140(%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	1-2	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 0x180(%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 %ZMM1,0x300(%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 %XMM0,0xc0(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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 0x180(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x140(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VINSERTF128 $0x1,0x20(%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	2	0.50	vect (25.0%)
VMOVAPS %YMM0,0x20(%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 0x180(%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	0.25	0.25	0.25	0.25	0	0	2	0.25	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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x140(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	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 0x180(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x140(%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 0x180(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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 $0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x140(%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 $-0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x140(%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	2	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	2	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 $0x1,0x20(%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	1-2	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 %XMM0,0x180(%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%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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 0x180(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 $0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 $-0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 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 %XMM0,0x180(%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%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x20(%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 0x180(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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 $0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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 $-0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	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	2	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 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	0.25	0.25	0.25	0.25	0	0	2	0.25	vect (25.0%)
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%)
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x20(%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 0x180(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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 $0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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 $-0x1,0x180(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	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 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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS %XMM0,0x20(%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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	0.25	scal (6.3%)
VMOVAPS %XMM0,0x20(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0x20(%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	2	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	2	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 $0x1,0xc0(%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	1-2	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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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 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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 $0x1,0x1c0(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 $-0x1,0x1c0(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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	2	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 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	0.25	0.25	0.25	0.25	0	0	2	0.25	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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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 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 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 $0x1,0x240(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	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	2	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 $-0x1,0x240(%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	1-2	0.50	vect (25.0%)
CALL 8830 <@plt_start@+0x20>	2	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	N/A
VMOVAPS 0xc0(%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 0x500(%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 0x480(%RSP),%RAX	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 0x3f0(%RSP),%R11	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,%RAX	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	2	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	2	0.50	vect (25.0%)
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	1-2	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 %RAX,%R11	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 6ad0 <main+0x2170>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	0	0	0.33-0.50	N/A

Report Configuration

Vectorization

Details

Workaround

Execution units bottlenecks

CALL instructions

Details

Complex instructions

Details

Slow data structures access

Details

Workaround

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

Vectorization

Details

Workaround

Execution units bottlenecks

CALL instructions

Details

Complex instructions

Details

Slow data structures access

Details

Workaround

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