From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============7686323094706785872==" MIME-Version: 1.0 From: kernel test robot Subject: fs/ext4/mballoc.c:929:9: sparse: sparse: context imbalance in 'ext4_mb_choose_next_group' - different lock contexts for basic block Date: Mon, 01 Feb 2021 09:22:56 +0800 Message-ID: <202102010954.uegPiSp8-lkp@intel.com> List-Id: To: kbuild@lists.01.org --===============7686323094706785872== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable CC: kbuild-all(a)lists.01.org TO: Harshad Shirwadkar CC: 0day robot tree: https://github.com/0day-ci/linux/commits/UPDATE-20210201-083359/Har= shad-Shirwadkar/Improve-group-scanning-in-CR-0-and-CR-1-passes/20210130-063= 423 head: f45b77fb0c895083616f2581319210b1aa2726c5 commit: bef684db96416bd93c78a411f74086af1bfd9622 ext4: improve cr 0 / cr 1 = group scanning date: 49 minutes ago :::::: branch date: 49 minutes ago :::::: commit date: 49 minutes ago config: arc-randconfig-s031-20210201 (attached as .config) compiler: arceb-elf-gcc (GCC) 9.3.0 reproduce: wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/= make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross # apt-get install sparse # sparse version: v0.6.3-215-g0fb77bb6-dirty # https://github.com/0day-ci/linux/commit/bef684db96416bd93c78a411f= 74086af1bfd9622 git remote add linux-review https://github.com/0day-ci/linux git fetch --no-tags linux-review UPDATE-20210201-083359/Harshad-Shi= rwadkar/Improve-group-scanning-in-CR-0-and-CR-1-passes/20210130-063423 git checkout bef684db96416bd93c78a411f74086af1bfd9622 # save the attached .config to linux build tree COMPILER_INSTALL_PATH=3D$HOME/0day COMPILER=3Dgcc-9.3.0 make.cross = C=3D1 CF=3D'-fdiagnostic-prefix -D__CHECK_ENDIAN__' ARCH=3Darc = If you fix the issue, kindly add following tag as appropriate Reported-by: kernel test robot "sparse warnings: (new ones prefixed by >>)" >> fs/ext4/mballoc.c:929:9: sparse: sparse: context imbalance in 'ext4_mb_c= hoose_next_group' - different lock contexts for basic block fs/ext4/mballoc.c:2382:9: sparse: sparse: context imbalance in 'ext4_mb_= good_group_nolock' - different lock contexts for basic block vim +/ext4_mb_choose_next_group +929 fs/ext4/mballoc.c bef684db96416b Harshad Shirwadkar 2021-01-29 809 = bef684db96416b Harshad Shirwadkar 2021-01-29 810 /* bef684db96416b Harshad Shirwadkar 2021-01-29 811 * ext4_mb_choose_next_g= roup: choose next group for allocation. bef684db96416b Harshad Shirwadkar 2021-01-29 812 * bef684db96416b Harshad Shirwadkar 2021-01-29 813 * @ac Allocation= Context bef684db96416b Harshad Shirwadkar 2021-01-29 814 * @new_cr This is an= output parameter. If the there is no good group available bef684db96416b Harshad Shirwadkar 2021-01-29 815 * at current= CR level, this field is updated to indicate the new cr bef684db96416b Harshad Shirwadkar 2021-01-29 816 * level that= should be used. bef684db96416b Harshad Shirwadkar 2021-01-29 817 * @group This is an= input / output parameter. As an input it indicates the last bef684db96416b Harshad Shirwadkar 2021-01-29 818 * group used= for allocation. As output, this field indicates the bef684db96416b Harshad Shirwadkar 2021-01-29 819 * next group= that should be used. bef684db96416b Harshad Shirwadkar 2021-01-29 820 * @ngroups Total numb= er of groups bef684db96416b Harshad Shirwadkar 2021-01-29 821 */ bef684db96416b Harshad Shirwadkar 2021-01-29 822 static void ext4_mb_choo= se_next_group(struct ext4_allocation_context *ac, bef684db96416b Harshad Shirwadkar 2021-01-29 823 int *new_cr, ext4_grou= p_t *group, ext4_group_t ngroups) bef684db96416b Harshad Shirwadkar 2021-01-29 824 { bef684db96416b Harshad Shirwadkar 2021-01-29 825 struct ext4_sb_info *sb= i =3D EXT4_SB(ac->ac_sb); bef684db96416b Harshad Shirwadkar 2021-01-29 826 int avg_fragment_size, = best_so_far, i; bef684db96416b Harshad Shirwadkar 2021-01-29 827 struct rb_node *node, *= found; bef684db96416b Harshad Shirwadkar 2021-01-29 828 struct ext4_group_info = *grp; bef684db96416b Harshad Shirwadkar 2021-01-29 829 = bef684db96416b Harshad Shirwadkar 2021-01-29 830 *new_cr =3D ac->ac_crit= eria; bef684db96416b Harshad Shirwadkar 2021-01-29 831 if (*new_cr >=3D 2 || bef684db96416b Harshad Shirwadkar 2021-01-29 832 !ext4_test_inode_fl= ag(ac->ac_inode, EXT4_INODE_EXTENTS)) bef684db96416b Harshad Shirwadkar 2021-01-29 833 goto inc_and_return; bef684db96416b Harshad Shirwadkar 2021-01-29 834 = bef684db96416b Harshad Shirwadkar 2021-01-29 835 /* bef684db96416b Harshad Shirwadkar 2021-01-29 836 * If there is contenti= on on the lock, instead of waiting for the lock bef684db96416b Harshad Shirwadkar 2021-01-29 837 * to become available,= just continue searching lineraly. bef684db96416b Harshad Shirwadkar 2021-01-29 838 */ bef684db96416b Harshad Shirwadkar 2021-01-29 839 if (!read_trylock(&sbi-= >s_mb_rb_lock)) bef684db96416b Harshad Shirwadkar 2021-01-29 840 goto inc_and_return; bef684db96416b Harshad Shirwadkar 2021-01-29 841 = bef684db96416b Harshad Shirwadkar 2021-01-29 842 if (*new_cr =3D=3D 0) { bef684db96416b Harshad Shirwadkar 2021-01-29 843 grp =3D NULL; bef684db96416b Harshad Shirwadkar 2021-01-29 844 = bef684db96416b Harshad Shirwadkar 2021-01-29 845 if (ac->ac_status =3D= =3D AC_STATUS_FOUND) bef684db96416b Harshad Shirwadkar 2021-01-29 846 goto inc_and_return; bef684db96416b Harshad Shirwadkar 2021-01-29 847 = bef684db96416b Harshad Shirwadkar 2021-01-29 848 for (i =3D ac->ac_2ord= er; i < MB_NUM_ORDERS(ac->ac_sb); i++) { bef684db96416b Harshad Shirwadkar 2021-01-29 849 if (list_empty(&sbi->= s_mb_largest_free_orders[i])) bef684db96416b Harshad Shirwadkar 2021-01-29 850 continue; bef684db96416b Harshad Shirwadkar 2021-01-29 851 grp =3D list_first_en= try(&sbi->s_mb_largest_free_orders[i], bef684db96416b Harshad Shirwadkar 2021-01-29 852 struct ext4_= group_info, bef684db96416b Harshad Shirwadkar 2021-01-29 853 bb_largest_f= ree_order_node); bef684db96416b Harshad Shirwadkar 2021-01-29 854 break; bef684db96416b Harshad Shirwadkar 2021-01-29 855 } bef684db96416b Harshad Shirwadkar 2021-01-29 856 = bef684db96416b Harshad Shirwadkar 2021-01-29 857 if (grp) { bef684db96416b Harshad Shirwadkar 2021-01-29 858 *group =3D grp->bb_gr= oup; bef684db96416b Harshad Shirwadkar 2021-01-29 859 goto done; bef684db96416b Harshad Shirwadkar 2021-01-29 860 } bef684db96416b Harshad Shirwadkar 2021-01-29 861 /* Increment cr and se= arch again */ bef684db96416b Harshad Shirwadkar 2021-01-29 862 *new_cr =3D 1; bef684db96416b Harshad Shirwadkar 2021-01-29 863 } bef684db96416b Harshad Shirwadkar 2021-01-29 864 = bef684db96416b Harshad Shirwadkar 2021-01-29 865 /* bef684db96416b Harshad Shirwadkar 2021-01-29 866 * At CR 1, if enough g= roups are not loaded, we just fallback to bef684db96416b Harshad Shirwadkar 2021-01-29 867 * linear search bef684db96416b Harshad Shirwadkar 2021-01-29 868 */ bef684db96416b Harshad Shirwadkar 2021-01-29 869 if (atomic_read(&sbi->s= _mb_buddies_generated) < bef684db96416b Harshad Shirwadkar 2021-01-29 870 ext4_get_groups_cou= nt(ac->ac_sb)) { bef684db96416b Harshad Shirwadkar 2021-01-29 871 read_unlock(&sbi->s_mb= _rb_lock); bef684db96416b Harshad Shirwadkar 2021-01-29 872 goto inc_and_return; bef684db96416b Harshad Shirwadkar 2021-01-29 873 } bef684db96416b Harshad Shirwadkar 2021-01-29 874 = bef684db96416b Harshad Shirwadkar 2021-01-29 875 if (*new_cr =3D=3D 1) { bef684db96416b Harshad Shirwadkar 2021-01-29 876 if (ac->ac_f_ex.fe_len= > 0) { bef684db96416b Harshad Shirwadkar 2021-01-29 877 /* We have found some= thing at CR 1 in the past */ bef684db96416b Harshad Shirwadkar 2021-01-29 878 grp =3D ext4_get_grou= p_info(ac->ac_sb, ac->ac_last_optimal_group); bef684db96416b Harshad Shirwadkar 2021-01-29 879 found =3D rb_next(&gr= p->bb_avg_fragment_size_rb); bef684db96416b Harshad Shirwadkar 2021-01-29 880 if (found) { bef684db96416b Harshad Shirwadkar 2021-01-29 881 grp =3D rb_entry(fou= nd, struct ext4_group_info, bef684db96416b Harshad Shirwadkar 2021-01-29 882 bb_avg_fragm= ent_size_rb); bef684db96416b Harshad Shirwadkar 2021-01-29 883 *group =3D grp->bb_g= roup; bef684db96416b Harshad Shirwadkar 2021-01-29 884 } else { bef684db96416b Harshad Shirwadkar 2021-01-29 885 *new_cr =3D 2; bef684db96416b Harshad Shirwadkar 2021-01-29 886 } bef684db96416b Harshad Shirwadkar 2021-01-29 887 goto done; bef684db96416b Harshad Shirwadkar 2021-01-29 888 } bef684db96416b Harshad Shirwadkar 2021-01-29 889 = bef684db96416b Harshad Shirwadkar 2021-01-29 890 /* This is the first t= ime we are searching in the tree */ bef684db96416b Harshad Shirwadkar 2021-01-29 891 node =3D sbi->s_mb_avg= _fragment_size_root.rb_node; bef684db96416b Harshad Shirwadkar 2021-01-29 892 best_so_far =3D 0; bef684db96416b Harshad Shirwadkar 2021-01-29 893 found =3D NULL; bef684db96416b Harshad Shirwadkar 2021-01-29 894 = bef684db96416b Harshad Shirwadkar 2021-01-29 895 while (node) { bef684db96416b Harshad Shirwadkar 2021-01-29 896 grp =3D rb_entry(node= , struct ext4_group_info, bef684db96416b Harshad Shirwadkar 2021-01-29 897 bb_avg_fragment_size= _rb); bef684db96416b Harshad Shirwadkar 2021-01-29 898 avg_fragment_size =3D= grp->bb_fragments ? bef684db96416b Harshad Shirwadkar 2021-01-29 899 grp->bb_free / grp->= bb_fragments : 0; bef684db96416b Harshad Shirwadkar 2021-01-29 900 if (avg_fragment_size= > ac->ac_g_ex.fe_len) { bef684db96416b Harshad Shirwadkar 2021-01-29 901 if (!best_so_far || = avg_fragment_size < best_so_far) { bef684db96416b Harshad Shirwadkar 2021-01-29 902 best_so_far =3D avg= _fragment_size; bef684db96416b Harshad Shirwadkar 2021-01-29 903 found =3D node; bef684db96416b Harshad Shirwadkar 2021-01-29 904 } bef684db96416b Harshad Shirwadkar 2021-01-29 905 } bef684db96416b Harshad Shirwadkar 2021-01-29 906 if (avg_fragment_size= > ac->ac_g_ex.fe_len) bef684db96416b Harshad Shirwadkar 2021-01-29 907 node =3D node->rb_ri= ght; bef684db96416b Harshad Shirwadkar 2021-01-29 908 else bef684db96416b Harshad Shirwadkar 2021-01-29 909 node =3D node->rb_le= ft; bef684db96416b Harshad Shirwadkar 2021-01-29 910 } bef684db96416b Harshad Shirwadkar 2021-01-29 911 if (found) { bef684db96416b Harshad Shirwadkar 2021-01-29 912 grp =3D rb_entry(foun= d, struct ext4_group_info, bef684db96416b Harshad Shirwadkar 2021-01-29 913 bb_avg_fragment_size= _rb); bef684db96416b Harshad Shirwadkar 2021-01-29 914 *group =3D grp->bb_gr= oup; bef684db96416b Harshad Shirwadkar 2021-01-29 915 } else { bef684db96416b Harshad Shirwadkar 2021-01-29 916 *new_cr =3D 2; bef684db96416b Harshad Shirwadkar 2021-01-29 917 } bef684db96416b Harshad Shirwadkar 2021-01-29 918 } bef684db96416b Harshad Shirwadkar 2021-01-29 919 done: bef684db96416b Harshad Shirwadkar 2021-01-29 920 read_unlock(&sbi->s_mb_= rb_lock); bef684db96416b Harshad Shirwadkar 2021-01-29 921 ac->ac_last_optimal_gro= up =3D *group; bef684db96416b Harshad Shirwadkar 2021-01-29 922 return; bef684db96416b Harshad Shirwadkar 2021-01-29 923 = bef684db96416b Harshad Shirwadkar 2021-01-29 924 inc_and_return: bef684db96416b Harshad Shirwadkar 2021-01-29 925 /* bef684db96416b Harshad Shirwadkar 2021-01-29 926 * Artificially restric= ted ngroups for non-extent bef684db96416b Harshad Shirwadkar 2021-01-29 927 * files makes group > = ngroups possible on first loop. bef684db96416b Harshad Shirwadkar 2021-01-29 928 */ bef684db96416b Harshad Shirwadkar 2021-01-29 @929 *group =3D *group + 1; bef684db96416b Harshad Shirwadkar 2021-01-29 930 if (*group >=3D ngroups) bef684db96416b Harshad Shirwadkar 2021-01-29 931 *group =3D 0; bef684db96416b Harshad Shirwadkar 2021-01-29 932 } bef684db96416b Harshad Shirwadkar 2021-01-29 933 = --- 0-DAY CI Kernel Test Service, Intel Corporation https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org --===============7686323094706785872== Content-Type: application/gzip MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="config.gz" H4sICFZWF2AAAy5jb25maWcAjFxJk9w2sr7Pr6iQLzMHy71IsvRe9AEEwSJcJMEGwFr6wii1SnKH e1F0lzzWv59MgAsAgiU5wnYzM7EnMr9MAPXLv35ZkG/Hp4f98e52f3//ffHl8Hh43h8Pnxaf7+4P /79IxaISesFSrl+DcHH3+O2f3/bPt4u3r8/PXp/9+nz7drE6PD8e7hf06fHz3ZdvUPru6fFfv/yL iirjy5bSds2k4qJqNdvqq1dQ+vDx18P951+/3N4u/r2k9D+LD68vX5+9copw1QLj6ntPWo7VXH04 uzw76xlFOtAvLt+cmX+GegpSLQf2mVN9TlRLVNkuhRZjIw6DVwWv2Mji8rrdCLkaKUnDi1TzkrWa JAVrlZAauDDwXxZLM4v3i5fD8dvXcSoSKVasamEmVFk7dVdct6xat0TCeHjJ9dXlBdTS90qUNYcG NFN6cfeyeHw6YsXDBAhKin6Mr17FyC1p3GGanreKFNqRz8matSsmK1a0yxvudM/lFDclGTm++NBh RzbS35RlpCm0GbXTfk/OhdIVKdnVq38/Pj0e/jMIqA1xOqV2as1rOiHg/6kuRnotFN+25XXDGhan TopsiKZ5G5SgUijVlqwUctcSrQnN3TE3ihU8iQyXNLB1er0ALVq8fPv48v3leHgY9WLJKiY5NUqm crHx1S4VJeGVT1O8jAm1OWeSSJrv3OWrUtCeTgBk3W67LacsaZaZcsfwy+Lw+Gnx9DnoeNhvCsq2 YmtWadWPVN89HJ5fYoPNb9oaSomUU7cnlUAOh65GJtEwXemcL/NWMtXiBpTxPk+6MBavJWNlraHe ivllA4G1KJpKE7mLdKqTcZSqK0QFlJmQcXt2k0Pr5je9f/lrcYQuLvbQ3Zfj/viy2N/ePn17PN49 fhmnS3O6aqFAS6ipl1dLdyISlUIDgjJQTpCIGQhN1EppYtbGIcF6F2RnCgWMbUcbGjFULpwuxCdN 8ehC/MRwzbRI2izUVGE0zF8LvOlEe0T4aNkWVMsZjvIkTEUBCefGFO10OcKakJqUxehaEsqmfYKp Lwo04qWofE7FGJhhtqRJwZX2eRmpRGP8wITYFoxkV+fvxolHXiKE7x48biVoglo0px5j52FXkbQt E7P+3QL6CzOYnZX94+ohpBiNdOzTKoc6YZuOkoVAN5SBteOZvjr/fVxZXukV+KaMhTKXodVRNIf5 M7an31jq9s/Dp2/3h+fF58P++O358GLI3TAi3MG6L6VoauUqPZh6GtfzpFh1BSLTaRm2c+MUZITL NsqhmWoTMNEbnmrPo0jtFoh2pGur5qk6xZdp1BF33Ax2xQ2TbtMdJ2VrTuP2sZMAjZoxOp1AUmfO UPtqwc84e1TQ1cAi2oEXCANUDXrprUujVVupSJPo/SvHyIFLlh4B5sn7rpj2vmGe6aoWoIHoWrSQ jvu3yoY4yvTUwx2wgikDm0SJNis7dDXkteuL+A5FSzyrarAMBizJuBLAxkeXg3/H1oG2AnxPyW9Y mwmJrhf+V5KKMm/F58VacRlXgaCIgj8iHQjBldWIoRZrsSPlSgCDHJcwnGqsECyqs7cswgnBnUUG rtNA0+KiYEcJWZHBHLoLnhAF42q8hhoIYIJPUKoAKVsyLestzd0WauHWpfiyIkXmGALTX5dgAJVL UDnYpPGTcAfTg3dupMUGPTtdc8X66XImAipJiJTcndoViuxKNaW03lwPVDM9qLear5m3utMFwlUs BTjNVIKw9KVhbxSCeNsG5U0wkMU1HvrP0tS3ie4KoPK2PhztQtT68Pz56flh/3h7WLC/D48AQgj4 BoowBLCiBWeduoyVREHNT9bYd2xd2sosOPQUUxVNMhhFL+ojGkLGVXQOVEFi0QbW5VkgEIMFl0vW B1iztRlHgECklbCfRPkTgjmRKTjj2EKovMkyCDtqAk2DNkA0CgbV28ualcbgY4DOMw4C3EVIgCEy XngqbUCKMdDKhSd+tD3ov3QgCMZEJranELcBtIDY3hisoG6Mg7KCLMHKNHUtXCiJMAMs+ZRhPYOA qB0mBjxWazroqvkQJqmmDLoEjWnYjxD9Yw7B2aOlgyIh+uECG21LNwQeqiUQeUrwLxbRTwXyDYNw ye0yRK8rO+BxOEb5oVML8nz7593xcIsIaZLUGaTq+/0RNf839UR/S572z5/c3QMSbQ1ja3VyfraN qtIgQrZqItOt7ExDg9FTBKvBneIHLB3Dd7dD1xEHh+PyuoXh1u9vz2J6PbB/P/OWstUNqFQJLCen gYIJGqwq5aTy9mUZw46mclRQv2pXZw0Wfud3F/WaG6HZiUaR9McinNJypl+2BuorsKG06ubq3ZsJ FT3E1dk/5KzLyXmNgX636/PZvhj2xUxPDPPSnU2jSmhocEu371czBUeR83er+fJv4hbXrIRVNZMu KcCeF3HcjaIZrL1Cb9RByRPTCi7VN1WGivku5pq5uOoO9olXzRb/u+pn/v2Znw21EmCiJgJeWgXM 0UxfV2uSphboXV28feepaCMlgFQY7DKcWAZIrm0A7ADiifvsm6vzsZdmhZkmGyJZm+N4Ao1LJi2k mwtoYMOruQ0LZVpdJLDG4NoEJ36NmA7V0PdUJ63NPL7yJ/2EQRywoJBgTEuybW9ExQS4RXl1fj64 stLtcl1abx9LKZUQzzoD3lyDN9lAGMMycJAc8UPnxsNM897p46+fDl+h+wBGFk9fsZsvTjZeEpUH UFdBHJQ5gERY18vGcN3AqoE8okGgJMwFjZLpkGbKcmgQnBc6dR2wJpVY6lxNxm0a35ULsZo6O7Cs JovY6hwzGYG/vbxIwE+LLGvDbgBAtSKqZhThiOOMRdoUTCEuNYECwl4Hpyxt/t/ag6shYUML0ATY anQFqpx6IWyH7WxnEPfHArcaszbOwmfuEiEacQHlAHSXVKx//bh/OXxa/GUR6tfnp8939zanOOZ9 QazL00cd78lqQtz1A90bMliAXzA2cpfUhBGqxHDhLJhuLxdjSN0uw3AhanY7qaYKJUZ+dyCiwsXF tOBwWGJil7BWxePJoI6NSwqWPu4POhmEwZu25EoBqh2zGy0vEYLFizYVaF0KcLlMRBHLemjJy15q 5YdwLrXd5Fwb1O7gk16xTWK0gO3kJjaSLrE2fK4A6ioO2n/dMDdd2ecoErWMEgGdTukIkZeS690J VqvPz8AGTZIhaGJj62vycmVqgBKYei93gLxNoieEtrwOewCm2TOHZuQI8WviKQbS7TEhADwqdybF PwGc9f75eIcbYaG/fz04lhh6qLk2CpeuMdHi2iow0NUo4Tk8nwWetyQVicOQQJQxJbYxFxnIcapO tUjS8JBoRtC4Ls3oTwlLriiP9o5vvZnonZTKZiao5EsysuKNayL5D2RKQn8koVKhfiBTpGVcove9 Sx4fBvh46Y48VrapYjOzIrIkMQbLZtrCs9N370+25ewrp3znCEIdd3dOed2uOZQR/oYCMqalfaLB SfagVIzZei8tA+W4sCF3Cu4d+xSz9KPUapf4ye2ekWTX8aNOr+lBC1V17lrWbuOrGoAjOhzqw9S8 D6wV6JooAZfI0jnWHXPuZmzsn8Ptt+P+4/3B3LpYmIzS0bEVCa+yUiP48FKWHYxzcAVg5rQp6+E0 HeFKd/gSM5i2WkUlr/WI9joyOCo6ErFurNpd97l+m0GVh4en5++Lcv+4/3J4iILRLvvhzAvmAypM FWJSxMt2qLoAsFRrg3wAHKmrIeTsDuUT9K++YnfxWlylTcQnGa6Sl2VaKadH/USWiF5LjtYvlVdv zj4M4Y85xasBoyNiWzlFacHArHfh/NClTAqAaxDbxZPq0ZOam1oIz/XcJE0cAt1cZqKI+cYbA7IE df1pT8N8UDwvaIC0mSdE3Kvg4HcYEWa91ozaBJ+bQcUpMUfP0dqXeEAErjMviZyJ2W3SrtbMwvLO AXfqN69h49o4Hh+Pe2AEiNIctVolLdtqVvWw0OhudTj+9+n5L0C7U6UFhVq51drvNuVk6U4umIiY N9OF51vh89QxG7K1iFm4bSYdVcMvjGq6TLpLJcVSuP0yRMywRxs0XNVAqCwKTuNnUkYGnCymHE9U AssOaBOwxEzvW0C/o3kxFF7jhh2JuGIrtnO735Fi7Qd1A4yk3rUkVcaRyDatzSEk09FTRatD45ao 7dEUJdHbWMDu4VwrRaMD31O3GU8Qs7MT+6Jvoi66u2Xx005bfydK/MPjgQtBTSJUfJ1AqK7idsjs lZqfYi7R17CyiWm5lcCcKASY43qqHeBVCDM4C05heb3W3F/3JnXKe01nookPB5kkn1lDX916yrBp JpxefVxyqJ+GaPQtHKnhRInGTvjVQkM92R8MzgEy5oYkySZWH5JgaZSWwgmvsBX4c+mGG6Oh6ZkJ pzGr1bNpAwLuggycDbS3ESLulgapHP76gYT6scguKWJuchBYsyVR0V5W69NV48kl7rjTUsUPOrhm lTjVvx0j+agWA5kXgCkFVxFWSlEbp3SaLiMrnCRyimCs6NDZngyy0cH0fBkfS8/u27x6dfvpyyu3 J2X6VnFfp+v1u3ggVQdrPu52vHML0TdFiODbjFrXeMFXKZ7tPI4pUuc7k1kDK1vWHrwDiYwXgWEe iMPumATy9On5gMAAkO7x8Dx323msaAI1Rhb85Se2R1YG4WsBWFbydMlOCBBZn6gZ70455hTvNlSV gXAeFW9YhRfTOjJUBLjEM71OLf2KxNbMFcN7YG4qxWPCRmPFHDPT9WzjXMZslCeSSDDq6KLjs4DD SyAQxOs/cQHFJ+3rfnbjjVfEbwy+u1H4NFuzT4PQikvmXsTsGCVR1w2TEMB5LIsKPAXuid3KxR13 BmNoyiWLhUPIpNofNIwAU5adk5wrZG80hCVBi2YK2CvuDy4Jx+lTzJT4pGCGpw4WaSL5Q7LMnS3d XjfCu1dmqv8jmG/dpeX96iD8yMORocefGZrFkUEvYRdsd8FimXm1Bwfos+MqNQp52Kjb69tOCx6s fdqaQPxlcfv08PHu8fBp8fCEOYyXmG3a4ozLVVj0uH/+cjjOldBELpnuTEvEcm7NCsXN+KQWoGF4 Hx3WIJP5ZiwiMmyck1LOLjop1/uZk0Lgqko1mfWH/fH2zxOTjS8KMHWgd/VcJ6xQzFJPpWx4NrsO VggjjADJ9DeyT/kyJ8JQbqBrv/GQ0z397agJx/Vs3XcjIackTjrJZ5orKCEPt17rGkuf7iMCn3eq PpPvmXbT4U4jPq/ZmPNxZXCc0dqh3h9UH7cEgUTFZsbW1T/L5BmhbNI1c5NLBXWuPfhsCLOH1pYL xghXU12dX3Qp3HqtFsfn/ePL16fnI54fHp9un+4X90/7T4uP+/v94y3mV16+fUW+m+O1FeJRsWjn ggFXBsKk2W5ZCZKb8O97vDywfljem7aBrqiuh5Q1jvelzxyPhsDKSzmd0I2Us+0WNFiPdjMlZSKk iHUWkopkWhBpMiSmeUjxXZ+llfmJ9VDRcznLq657m2lmSuXzkwXqOqjTe6dMeaJMacvwKmVbXwf3 X7/e390a27b483D/dVq2yuj4RKv+v59A+RkGQJKYmOaNB80sArB0F+ZZFDCV71BbSO8RSoSRNnVf vQ/VeGLocSzRtWPjhjkkMlvaadOPE/w4xNImY/SH4uoMMHk9zXGNhzEnVmM6uC5qCubFUs2lovjg zEOTvmg8mvCZXY0gw5IBhPk8YGAQ0uhpMWThahUWNMeYHtJ1OO/PLtrLKIeUGIRHObKO0v0Ax2EY eB47LRpFfJTkMOqV9hGOw1M63pF1Qaq5nktWF7uZfqZxtBl0s9UzxXvoeLoKNbcUAQx2OAa2xmqt h7zDuAFSOskyIqlP6xmThIQFpTx9mb+22lXVotjFKVc9SF0GW3Fk/LC4ziRt7bWOYafOdnIcQndl Pt/f/hVcR+ornjTsVx9U4AIZ6ucL8Ntc+4NwkFYzl96NTJ8dM2npNsdT/zJ9GzttnxNXOTn3zhnm BGdfm5oSP9mDSMuu8tjGg7MoOfM+DHDwzOUMHQtwiwt3B+NX/2bYbczQ15ex/mv3elGf4Qq3EF+W sJ6VEHX8uLE35zJ8om72jiJBWg9J0TGi2UGDen4daSRl1Ds7tN/jiU4/1sJ7wwyf0UvLmhSOucQr TKSuC9aRnV2QpnG0u72I6UNB6mSsts6F12XOGMPhvfWAwkhtq6L7w7zB4iWroEOxXT8WCdMboITT Jqwazr1ZTGns3UpaKbwoLfAXCzyzCppIzA2paGWiZtVabThEu5FK192p3jgnPaXfISG5AK1LbOw9 6om5vTLIxNrxJaYp1S4pG27Lso7e6LPPEr3jvFzFk/Rmqs3wg5Sfwy8uMf+BiQFM6A6dupbaQR/4 hem0gFLmPFzZivqvvDtW94DTJOale9nHYdhsfepNTCu3bdKoXes/j0uu/SeDYDskI+V4rc89pl8c Dy/+o3nTjZVeMmdAxhZIUbewNry/qtB5l0lFAcO9CDDanFKS1Ay1u+h3+9fhuJD7T3dPQ7TrJIQI bGLnjhB8tSkpCT6pWvuZfim8++IyOL+171i2r8EmPHb9/nT4++72sPj0fPd3eF9qxWcupb7DiwtR TlJfM52zWHCakB1od4sva7N06wCjkZ6nzhvJjg7LNpFltReJ7EgZdfwnBzoomf+6Bj7xbDSmpsBJ aBkKL+dk/zj/cPkhlOZK6HqyIsBZpLZ76XQdsNwaReLtrLd2CA5JFZFRxbe55eC9VXuXxHsdF+nX sBI+NsaIhaXRZccfHfB8NRLS6J0ygN0qM7/048tHfl5kZCpWZNq/VTwSW0bTPM7xfkcHGBkjujEX E+xzSrMEyf23w/Hp6fjndJ+MJfHWcRFMR055ohXs8Xingd0QN9820tr8TTD8npFQFffxjgzR+WXs lMkRsd2NN0GW77axWxlWZA3/el0u5bqYEFoctk/Vq442qNbsxA5WDmL/ray9GwM9DYZgsgKFmDFQ g+BcOCK3K++BR9au3PdpocvoyJgmkd0Dg460gUCw8M5qe0pr9binwlfwdsaQ/J+7MCRV7yZC3HG/ NFsicjp3/JOBY+fmZ57wtfRUFnc3KwTeNdwQWQE8VhEhvJ8PvTcvzPESElumSUQMH7N0P9RkRPBW lf9IZWzVhjJ1bO86UjYSmLZEZUqcd6bTBnB6YvCWJ3aKvoeU1ty2h3L1LI/Scp6pVzzGnLzB7uDt eaRzPQsPWfq0/9Y+jTsblQjPSh68z85Om8dCV+8HTc5WvHA2of02JnbsZ0fkVd3oCXVZu9sVsc6H OvzuL2KH5OAHOSjhmbth8Xt2DxrmcFnAJTbKCU4oq3OTMZhQ8AcAtN6Ffei5qKZecOAlL+PRa60I QOvo70hhZjbzMoTFxp4ix4IScDDBc1wAs9A37+cUDAbHS8ml8m+Z4Hb1r3lkhBciGAQgLS1EEbtb YDMns5CippTI6T0V8yrr7rYrsRDD/dXxaqp9OJazoo6CPOi3Lmt3L/cUwM72p63G1IUmVUoKEb2G D0trWsq4LM17TvPLc71bzu6eH/67fz6YEyE3KZ9tzBsvz2j3JHMbOcXfBRqZYD4kGRpx3jeNpcxv wdgBu72PCsAyFUUyB47HIvEnX51rDAc3GAFi3nCu3fv0vf8yz8PivIDq3NIyqM/8rsbMNa4OFsow w+4JGDBlq4GtVoKSxu6voRBRu4r2ouaFpmO22NK7s/8/zp5sOXIcx/f9Cj9tzERMRevIQ/mwD0pK mam2LotKp7JeFNlV3inHuKoctnum+++XIHWAJJjZsQ8+BID3BYAAqL77LGAWjOO4KQPs5JtZ9UWB Zckxv+bBzk9MzARObRvDGNp2EpC4DmKayDm0w3MMULsUIt5I91LMQzvWlOIuf38f2B9d8Kq6lrzN 5RlsTuDqK+RepEo5ZBKAjx8FuqIVxWWjDbsSGxpY49NW9yWnDvKiRUaw4kOO9mTmMHv4vF7e3jXe GWjjZi09gzRFICCw25Cj1L7aTWkRVIyQjGx0BaX099L1QzrgfPL1wrUspAu6DEPiUEzZKcB/uCrz M9n1do/IjjqKf+8KZXkjI8G0cAP+om4f88ufVtdt83uxOo0WqvbYoL5Bp/eu1U2fjK++wXEkB/ys VNglkAE1P/ku0Rh2Xjgo5eBVtVF3GafTmAaT15hYdkoZZR1bTVz80lTFL7uXy/u3uy/fnl9tMU3O qF1m5v5rmqRMbkWOSu7Tctqq9Om5y0C3KANZVWRQMaCCXWQbl/e9jNHWI4aUwAZXsQsdC+VnPgEL CBhIfCCOUG0oBK9CujwPBOKYjvVhAuixzXJjacWFAagMQLzl4kjHUuCVkVOOYJfXV1CgDUDwElNU ly9i3zKHt4LdsYN+q3UBR86jw5kXxPRS4MGAz7m6R7Jqd5ME2Glpr+WmzGNxTNA6q1stVrEKIQLy l58/Pi7SWk/keU2NJ0qEWEm7XIgazjoV7FAH4X2wXDmmAudtsMzN7uO50RCtQ6wpIX5MGAQVaas2 zpUEhN3jBmzaSE9/wPpBZG1sgTp+FMf7/P6vT9WPTwx6zM3+yi6p2D4kh+B27yoNsuBezUzFLgVg 9yERn/qrBIK3sQiUnydjon7/FDXSzJ5Gb0oCO+mhoZ6SOK9hVv63+hvc1ay4+6784BzTRiWg+uh2 Vv9ltqpqjM1eAaWYtpBm4eI81pkYQXXcZs7OOpwFyy24HEoOaRHfWGlhAwXPcCyz1hESu9pJT1II a4Az6NO4yc806r7a/qoBYPFrKiEB0xjPaqe7F1Zg5yKkiEc449LCqC2IfUaQx1FIihs9oNgA6OMu itabFbrhGxBiAS1s8hL4FtRhQ7QEC9CXxzyHDyRtJ+rKYfj8rJb3VH347k9N1qZOqUiSDH64lKvo rKYcKgE3bZSKckAnzVbzT4LvfowbDmpD2qV3auI20VS5A5je5mTr4bqIJY+IB9bAg7AA5mgzm60R nKzoWqNY18Zy9EHSRzeR6s5QDYNd0a29eZSPRXrHp21j5PQEVKkkv2sg7BWI4bt420B8B516xwyA snUmgWLCiWPk0Bw1VQzCO4YWkzjKE3BITOOUA9S8I+LuUHzG8/sXJImN3GtacrElgZlrmD96AVYY J8tg2fVJXaGWIqAUXEkESK+zSHksirPcG2bzScY3YcAXHpJohWSZVxwuJ2CTGG5pxrzrhG8iL4hz NDIZz4ON54WIi5eQwJtpxta1ArNcopB0I2J78CFUnZVAlrjxkHryULBVuAxQu7i/itA3bJui2r3g tcMh4KymCaJXVwcxHLueJ7tUs5WoH+u4zChbahbgiIRpKjj3Ap2XY3dKuFhdAdoLZ+ASdbwC5uk+ ZmeLtoi7VbReWvBNyLoV3kQmeNctKP5qwAs2v482hzrlnVWFNPU9b4EVC0brVOj8pz8u73fZj/eP t9+/y4ii798ub4KJmY2pX4BJ+Crm+/Mr/IuP/BaYffK4/3/kSy0iffJrGE3RE4OtWwySR43u8lN2 qHAHaItW8cVgYDDwataoywhERaWdDU2cJfAiAxniAhKgVQHJtbAjEmL540ko3OqoQEBzvYYK3X38 +fp09zfRU//6x93H5fXpH3cs+SRG8u94KMbdnDuidx8ahSZNEMe0SEE1JdADxY5Q0gZGtmTaetC4 AVz8D/pbbCEj4Xm13xvvDEg4h4tmqQC0TifZO+04kd6NEQM2UY6QUdCODeDvRkmZ/H1tUHsOL7UQ eQI8z7biD4GA10X0S2OFampUkVGEMJpkdNFJ3eFoUUcB4/RYkFipXpKBrB2WgILmuOMHRknzoln4 5JSflTmo1i2WhCp+w11mcqDFKGIxohvaljatK6jKD8f4wKvM+xUTe6aUAak0AgkhB7PKTFJzZr42 gbCgzacs8Ub7rpm/0ZLtSNceNaomQySvvPRTvMgsaylGR/5UV2pSh45vTAHatpq9s4QdaJMrQE1X eqPo/PH2/Nvv8H4T/8/zx5dvevBKLCKOhrV/Mcm0gYNlkOHCBC19TMtECIJxHjOQFMjNaDgSWq67 wE1pi/izHr8HIx+OYqfK6AmH6WjnYERwbKoGhwCV30JkiCLPcxSu3Ikr0utzpmJxkpbYw6qI2daR I4sfs+Ot/GRkIlTVfVpkZTYPwcyquQck/cwOmcO5fqTZV9U+123QRtThGJ/SzJG3VEneGpAiFiJx TumPMZGgiMuq0zQHecdP1grC6N3pZuEZa/5CFQVVZXaTk5CLUbhFWMbtXyIDY7+yKugWYkJqAxDz oGLkSqoFVwZB3MkhFQspBxPVGfnA4rXneRZAv6V7kFZ7mndjU7gnXpOWqThyb7WsAaNZ6hYa0fC4 4Eccw4F3+206CNRUpjxNH24VzKs8bnbi52bn84rB1QVppILJWjk99NvxAuz0RUVvJD2XVc3P+Mb9 xPou32u9jRI84iDG4qMHwyCWybPDpj5lnzVVlfruT0shjmD7lAEaElB5DznGKJjah5BZabvTUHRx SUeVQtVVIiOlKjucdT5dApB1Ez/VB83TPk8TiLa038Nl9YFSv+2yLpV6f5TLbnLqFEf6HaRzGQ7G hUqL7duSrHSUNZ5hVhKl6duayWZJYjh/3ASsWC78hecoWKDXXdcN5c7AaBFFvt54gK4JUmVTrfob 8yeZOPViZ72GI8yJT8QxeK1dGatzMWnoVuVdq9ddnkd9d4rPvTkPQApqfc/3mSOz4Zgyh2YE+97e lVAeMVa68Ui5kkrhW2MApjNGH4JSPo8RWwXFbeSFnbMDH8acKPvJFBiye734YdvWC4dNemwMWicF ZwakTX2vw4GKBL8HnueM64RJHYVREJiNAXDLIt931FgmW0REXqs1lVe02jh75lFwqZynjpIGtdVe bABBA7/1ARIDfs+jzWaJdQlFklWDBbYBVIYdA6TaSaCdrtGECZkua7cxPvgUFCScMlPxFJDQJVBw b0nNNsAdMpC49QNcIjS1jYSIgWUgzuBbPklZPyw8f2Omrx8ib7WY9kwwmC5+f/l4fn15+kO/wB86 oy+Ond1FAFWGjDRqDIrXYb9ZnaKAQPP7yWqFcefOLXB9VzPNVJ6gn8hz/Y3TuqZ4ap7jeBU8PzD8 ldWTzRgO9iwRvIi1R2UAJt9Mgv9WY9cefr5/fHp//vp0d+TbSRkF9Xh6+gpv9v58k5jRnyX+enkF x2lLj3bKdd8C+J4EpKQQuwJ9jmOylr6I1mkKB/eNqa5IV5iMCUkcTQ2MGq3hHaiGZwgLOqBYG0wF mSwIb1aYEGloujTJYqMvKbLxLNBZq5z5kgdPHHqbmaKmpANcQhMPfBNZzeEcuJUHVqRiBG9dGbe3 6vX5nGA1HUZJViktS82fE7Qdot07LV+5Mk7PRdyJ329PL0/v73fbt5+Xr7/BU6PzhZC6TvghwxXj 5fPxU2TzNOQACEJLcjN71AExHQkcuVEOmh9aJgBrMLzxAt+FDIfH3HhS6l89i2tNUC8ASvPaj4XV gdmP198/nHp3w+hcfirz9O86bLeDa+DBmWJuu8SBi58rNJmi4NJZ476IyYAskqSIIS76vbIznUzu XmAsnuHJtf+9GIaYQ7IK3kK4Wviv1fk6QfpIu1+NWNC7fse9aRmvaAnu0/O2iht0EowQsQXXy2UU OTGbudtnTHu/pfJ6EGzv0iMSAGKtabwQKvBXHtHUiSIZPFqbVbQkCs3v77cJUaZ+umtgOT1SKlHL 4tXCXxHpBCZa+FQ/qWlCZJYXURiEmHHSUCH92CXKt1uHy821rikYJ7MXJ78f0G9eTTRlempJy4aJ AlyeQSdGlzFoSq5lwNvqFJ/wO3Ez6lje65YTEyp74KuAfshtrppYlFT8mHm4iqBvqyM7GEGdZ4JT vvBC73oxHUz0a8WIjdD3u45ooOH6OQ9NC48PkLfRaPtAbCl89jVHt+MTqI/zmhOk/facUOC82mfi b11TSH4u47pVD3K4kYJ71CSMmYSda92KaUZJMx35ViGFTXM4ftnhGs5dLJiIprn+htJUrpwAGVnq rmLAitDFkqXxtFFR4TUoO8e1xjgoMFQcpB1anyxJHnnXdTEVKEjhDfFE1WEaByVJ2ecKxEGn+CtF IEMIaY+ww/fQNMFPCNYY2TgMaaAXuRBEU8QGICDcXsPLulqobYyPE76OFisXch2t19r1mImltkCN qPEFs6Ob0Gh4EA36QrdnJgn6NlzfKuwojo+sYzhiKMZvj4Hv+SHdWIkMNnRKYMvhHa+MlVEoTxqy suwcsbaI/QW9d9mke9+nzledsG15bRh3EQSqix1lSQpDQLhCunDdWGLSJN54y4DuzATWgq4Ex+hD XNT8kDlkLEyZpu3tSqf7OI/JZxYsonmnoHPqWChkqRtZ7Y6/Zi0/0sOxr6ok65wtz5I0pdhaTJTl mZiLzjz4ip/XK8odVavHsfycumZEet/uAj+4taDSPHbsK2le0c2X+1R/ijxs9GYTaMomjBacle9H nu/AMr70PM/VrKLgvk/xHhpRmu9iLuSieuEoRH7QuKzoVse8b7ljP8vKtMNHg5bv/doPXFUX/Jz0 qrw1IgmEk1523oouX/7f6O/8WvhTVrqq0YIXSRguO2jizYWnNtubZKeklZcJxplLUQre2+8cE6fY rDvnogCsR4VHMol8x5YlcY6zoWB+uI7CK12eCVEpdNVNdKXcd8jXAHS6wPO6K3u9olhcQy6vIdeu KtaMFLYxCbxvwenMeZZrj2zqOH7tbOKtH4TBzSnE22JHug5qRMcGYtuOtol0Rl20Wi5ultfWfLX0 1rSogwk/p+0qCGhxUaOTr1beJGuqQzGwIrfzFNLYkgzxoRWclVkrZWBTF5I5VnhTZDYDoBTPl7ev 0p05+6W6My38Ui3gk/yE33osYgWu40ZTDChonm1BlsIX5xJOx/BRuMGWSJPBhjJ4UKgXcvUE8Miz TR3XWwU1ylYqCU5P0KNrUe/jItW9OUZIX/LlMppLn+C5ZhBMdfQUDJVS0SlV17fL2+ULaPot+/cW P7H5iCom/vAql87OJc/l7SKOlNaOBEhxfkKwWSfbIgS8HZfQcfvgQaxN1NetHmZEmWVLMNnVeSJm q2Dw28p8gmDwont7vrzY1zuD8CQ9fhgOxjYgokC3lp+AfZIKkVm6CY9uoTSdv1ouvbh/jAVImdBq M2gk28EFBCX7YSKrq7UKafbKCFHIA2qrz+gRWTb9UbpOLyhsIxiOrEivkchX0RLtogqXHZdnGbbd 2fCY1/Ay3iMUQdudImLpkg8+FDc6KknbVL2/7iq2cYR71HJxbStTJm0QRZ3VcvBUFytFbOjFqOgt f/74BElETnIiyhsF22xdpb+HODtlkVkZC9439D17Qip4Z40w9GmetSnRCSNqnFPulk6U01TxDQrz mfAJaE/Ycf3rYfyGJDzbZY9UZRXidlU5Y2VX2/lKsHP1cOavMg78J9mQCX0loaHQsfAuyXogHA6p X9t4by4DklAPJ2bjYEKoRWcuWUy0jY8JPI73P76/FKyfQZntulW3sufaYPFQc1WL62hnn4tTlugw OHtvDjIQiamo2udbeex43uf19W6UNFm5y9Nu6EozFwaGfDIGS7bPmDhTaCFmXH/ARvkhGYx3mAp1 Y++QANSOyskpWTuszMXO2iY3FIwDqlSOGol2W1T2hyTH5oH9Hr9cKt081ek/85AQ2GSIYktxjxLN 4ap4dgt7HKO9EOMqX/gmfXfrJiu1+OY5sU7rWnNQHXwiRzItMm6R9QfRATlZb4HeDoY36r5ff+hC sCcNWMNquv8JKJ8JE9xckRbkVJgJt/EipJQvM4UZYGvGMDG42JBnxnRZfUgbTVstmmLUZkbcaw/m yhcA1eigoLydgkNYEfRSS8vET+3qhJoqTybJuCmXKqgFsKSvGdyzZkkq1wYSsc8qha82WxFSLOqs TMmrMUxWHh+rVvchAPSjaF4vH2C4kp63Yfi5DhZUFUacS5NhkmmG02LrzM8q2M+U8QizglGMUTst jh6JacOQNUfeSociFZDKvtIXdbVv8rH+DbpM3oCJ7q108PAYlg4TXJrxKhyA6TdAATME1ALuXc/J uMuRw5/vq23W2sCaxdO1umjRJCFB8KO5eXOL/3z/ePp+9xuERhrib/zt+8/3j5c/756+//b0Feyk fhmoPgnuDQJz/F3rn57BwOgbMYCTFGIqykBm5svbBlrG9HX0CSKbmEmtmLRIHwMdNNRFK04KqNgl viK3RkF5nxZ1npijVslbZEcS0eeY1dUSNvchrR2RA5sVgkN3om3T7uEBdDHTf4ijUdD8IuaGGLvL YLpmSXayembMBAC2ccV7sXOOk6X6+CbSzzmiGYFNe5xzymhYe6QCl0vUEMFZpwfg4Pjr7i1JBFEs IJqFc8JAhDadhZ3h8n1nayYCxjiYtQYTbQypjU3bx2AfN+I0AkgFdDJg8pBSErpgkYvL+/B+y8fb z5cX8a9lEyNdRCUTixhggHXKfTQt9+IA0HGWeawEHlvgAPKzDp6ds2YmRrZnXI1063sQMICf5Lpx FaAcxraAyou11+e5/iaFgAOD6pQZBL4S0zgrqVMKsKMZtpmrkEeijK88WlklKaSg5ci26DKmd2wn /YN0kFy5ep9+PpcPRd3vH6xpEhdTOB85/rM1MCEcQxWkPfBEX4+PhamJY0wT8aPxjbLDJ8/OFD/6 Cag2T1dB55mj59qjueA1EQfH9Q/tuFSqUZ7dfZkm9mRoKMEvz+Bnj63RIAs4RimWudZ0KeKTivw3 cvdtDRTWTgqwoVjqaTXIlOUZBOW8l3wvXY+RRirfzEoNuGEPvp7BEJZ2qto/IZDi5ePnm3V0120t Kv7zy7/IaovW+ssoEtlWZgSa0XbbSj9VJytBskIq2bQUDD8bEb18oQBxIwKu2acjejAy3x1LZugF ISfxH12EhlBbs1WlsSpCvA+8jQ2HK/eVpqUeMQWrg5B7ETUOA4kQ6Pa6PDVhOn/pkZ5WI0Fb7JD6 aSo07tbrVeBReTb3EXkdN+IrlubVFN6jERPi/fJ+9/r848vH2wt1QLtIzIzF1DiU8T5u7O4rgD+O bTjji3XuLx2I0IXYIO0JzHDNB20ACNaMtxCAZ3judukHI0W1M5jLMUnWPAxm4NpsMbk/yflZMQQw kqn9UU8BwP6REmAlepiy48AUKhLY98vrq2CY5VFnsWIy3XrRdWOcVr1AdZyT25eqkDqSXRVKTtr7 NorfbeGP53tWYdOKG7hWZzMbsjsP+YmyKZQ4sNBjj8xKU2yjFV9Ty0eh0/KzH6yNFvC4iJdJICZU tT2auFFPqgOrzgSVsMOCo4sBP3PxiRelBDv9KdUoFUm/Ywesp7oy9pOYJaFPf7xefny158Rsv2zM CAV3KPoHkrI2mrU/9XWe2N0PprgebXA1EziMVtX1Hos3y9DZMxKNgycN0F20XHdWL7d1xoLI95w8 t9FjapXtkhs92WSfqzK2Stsma28ZRO62CQI/CqhjQS0uZcNlZivB1O4tsZMgqa2OOtwsQgsYrcOO AC5XS6tQtTtfH0c4bVzVklf2G99uTftQdBEVq0lhlbGxneqUrzyHIZ8kULYqrmwFdrNZ4EAyxCBP nO7VwRdbq79aEP3Vhf7GvzKx1dKgbc0VAQvDKHL2aJ3xijfG8HUN2DeGuGVEC/TZu9+LXSpWry1p 5Qs27ohc5U7+eO74n/7zPAjiM/M/Vf7kT49Z8WARUTFeMIl/QrqVGTFwpUSufJ+R65eoFa4tf7n8 +8ms6CDcH9KGVgBMJJxW9k54aKq3xPufjqIWuUaBrZv0pCs8BAgRhK7iaOZOSxx6juJC31FcGBrD gVHipKNUEzpVRBe59DoasY4clVxHvrPpqUfbD+lE/vraFBqmCuKbq5MMus/JaPEKC6+nYIUGhpoq mRrc8QE/g8R2EG2CpQlW+1gPYrO2FhVYEWN1PoT/l1CyE0CqhTALcMZ7pJHqNgbNzLmPWRttFkvt UBtx7BR4/pIs4P8Yu7LmyHHk/Ff05H1Zh0mQ4LER84AiWSWOyCKHYJVK/VIha9RuhdVSRx/2rn+9 kQAPHAlqHvqo/BIgkLgSQCJzZoFGQh8n6QyZtmwbdKN1DQQ/PZlZlBHixlf5TrsRmYVhEGfO3R8E vDF4AdtqzoZvS9y5ic1XjteT6A2i3ewXf3b1LYPumQ4mualYCr0IcesgERIa+tEsDaE6ia7heWI1 M8nOGmDxQ2cO0CNIqtkeTHQzBs6an9gU6vdjSzZjlNDQzQZqENMU+YCyPukmloQmaGKx/Gfu10Sj xCG9eIA8wAFCkWIAkEbGgqBBVHxlQ3jAkeWBL3GOKgVLd253UZxiTavMJvOtYXlgp0MlpF6QPA7d jjOMNIgiVwzDKOYKin3zVPAwCDAdYKmR0pKxxLsyz3OKmYwPRzomYbbMltptaYtfQMISbz0tUCQs qo/Dw0c21vBCCLVynZiqthLFPoKlHBSr20NIpoaJ8cbXgFMzs+kre6aCB2cZ5n0cajSW18w4B/w6 dOCcsuqv9zWvsBx1xj2rB2XGtVlZPYmMucN75rOOnZL4c0cY9fIiMFwXXM07Ax1eS6SNuv6ktbDT LCdlOulC8khEv5SXexiss6jDvRnBNQwGMV7RsOKc70Sf4LzemXaZgo4NxqJlOrtGNhZGYFPOMjv8 Nk9yTGGAwE3VtWhRj9w6m3XwojD7iFvti+HC4POvtycZN8Xnw0NsrKx7MaAs6oVJ5VEaGov/TCX4 wg/vQdWBBcH3gjI9G0mWBhte6oBJvmSDS6TCcxe4ct02RYlpu8ABTp7y4GIsrZJe5jQN23vsbbrM GQ6VL5Y4JM1y/bIvnZPSlWY+5dPohj9g2ST2qepCNBethZxhW4oFNRerlbzRbDCi0HOdBaXEzlRR Pc40NAZHZOrMxKWZ5/ULFdd+JjhErWQAFLvo6r4b7vj1wO02K8LocrmgRPMGVwfc5uxJQnKT+bZO YhJKsWnXYCPcwPO60LaVQBM5qnM6LQP1btz8kH2MBLQs69tM9zu3EqktRklO0HsL1S1t/W2iqisL q9kV3St2BWcJniz3N6ZkyOJNBqGHYQ/xFpQ4w0Wpl5uJ8sxJNCYRumGawTx1klTHPQl3LTYVVZ8u 8tWQNb4lyWqpoRpPnu/OewHjdnGiiWkX+/AC2+c3p2IXxoE7C+slmbVLs3wFHWmG7TUkepcFmVnL STc0ibwqHIMgSa/jNLlsrw28IRmMGk8ReEsDZ9GSRK+rL2C4e8jEEDCmOLa7UEREZsZj2/sEON0t GBU33gyKNjPR6WDYKjxs3jLsxGrKsGlPdpKeNS1DFVax6QgDfVOltiGh+U5060WX/KpkQA+KV9hd hSSdhL7BCHWxTsE1snUOruWHn+kvDFnyQUVy9GG5Bjsr30z3GDcaLJYZzISJmRu1jJ2O2NHRMWPs hAfYno7hEQXvvglJGllGqbJvtRF1B/lYRGLH61st1B2B1XPdm2SpM013MRjRXU5nwFmApfZimpnK WrU09FjwzLC3ZeWdgzOJS6pvqAkw1l+cTDR4hIvQ7IcQGuIzZ5pZaLDRr6bLEitn9RYSLrDQEzed Zdreo4mJPU+PoMeEJvts4WAUQDkSbcPgai2B653O1gZlOUmoDrA97IxThIXoj2u+cCjntOeuGdnB jLM6MYDB+Uk9a+CnVj/lWnlgUyv3tCvXv1wuoVQdxNyCZeDoZisEe60soViG2CmvhpY0QvumxjIN oKbsQk8mE4doVTgLR7uhxi13gB8xyX3RZrm0bZiL2T3SgGSXRKBZd3J7iLNTsTD8pNxkSrATMoOF mPYUFoZN7FoXZUcaUUqx4kssywK89TwKzMqgNh9Yxgo50wjtlDVv8ihASySghKShp1OK9SNBt40a i7s0aKDQeFK0wBIhOJKl5OIpj9QftssDugRa1VXLcCG1JOJtDmCSJh90rHmT9RfYKKpZGTyOMZmB ZkmMeRuyeBJPH552Uh9mkOs3IBaURl4oR0f0sm30YVmA9gaFETzdtHE3FSITN3xVmJAQgkdARR8K BXd7lmh7Ci74sLz7LKM5WiCBJBcc+SPNCTp4YT8ahmgigZj30iZGt5cTd6O7YmDpE1P8mM/gwlxl 6AzunlZD96dPEABsO4uzmC8TVDISyvyQbpWoQboJxEqWwRwmU2ekrBI+8d317LwicHgHxvtdNQwP YGttOPCyTdmxxLDH/ohHbbo35QZaHyaZYYwNn0I6Yp8I6Fh7Ru2NVhZO2p4F6GQPEA89OgunbZYm 2KZR41n37y7WHCBsA9oNVgUX+7DIM0hwMyuDKyPx9pojedIjVjqxF6ShGKb4CJj365u5AxOJEs9y oDbjHhcpNhtqm2kz5Z51Q6JhtD0tqj17fNnIQuyat7NYzNBc7d40El8B21uSgRjbOgNRWy18sDds V++M0A5D4Q1fNR946dsm8HUtEVDF8YcBimfCtW2YToZ4X6N57zrju3I4y7eEvGqsSCOTyfKfL4/z Xgwi9OlXRap4rJU3FlMJ/mWi7Mia7nAdzxqDVQh4NQ5hrlcebzUHBp7pPZ/i5eCDZgNmHy7NcPQC Lka7Tu3nhOe6rGSEAVvm4sc4gNvdYTa/O7/8+fweNy9vv/558/4NNreaEFU+57jRdJeVZt6MaHRo uUq0nHl0pBhYefbugxWH2gO39VEuNceD7tBSZi9jostIL4X4n2Zvo9D7Y1dWutEiVkWt/6xveTQB WFJGePQeuNxTqqBscxT4l9efzxBv8/GHqOXr8xOEXHv8efO3vQRuvuqJ/6YbE6quxErWj7jbcCmp 3WlPrEOxlY40mqRDzOaeY0jZqv5RH9D8WtY0evQp8YG1666h7azmRkz+7Q4h3cZ4ewM8zLH63vQy DaVeC16T4cKNt9cGw9j7u97Ech4NcyyoqRAbEX8wT+7rBbwY47ZE8IqLDOWoRpnM4a2b4SvS49vT y+vroxEA3pzWxpFJ+36ZiP368+VdTBNP72Cu+/ebb9/fn55//IAXWeBD/euLGbFDCWI8y1NaW8Zj ydI4ciYDQc7FKoSQwzxPnZYaK3DvTQuUbm4Lpz7A+yj2GP5P3YxHkUetnBloFGN7whVuIsKcIjXn iASsLki0c8t1EhWMYvwIV3GINThN/Z8FOMrdjM89SXnbY9qMYuDd8eG6G/diY3XRp7q/1tjqIVbJ F0Z97pk+wFhCswztmUbKdRHZyE1M+mmIGprpeGRLH8hxdnHlA0ASYLZcK57FBE8oAFBbvIl3Yxbm dlEEkSZufoKcYIcdCr3jgfEeaOrNTZaI4iepm52Qeoof+uu4O6DgdCeNHfnNdKiuvVCP556GsZsV kKk7lM99GgTuwL8nme6jcqbmeRC5tZN0v7QADp0vn/tLROSkoPU06MuPRle3JzApKXfqKS6EqpnK 1A7Q/vz8tpG3266SnFGsE4epUy9FRrkjtyUlOUfJ1Nx4GsBmP2dlHmU5Mq2xuyxDn7tM7XTLMxIg Mlzkpcnw5auYgv7nGcKP34DjDUeYp75MxHYodKZeBWSR+x03z3Wh+w/F8vQueMTEB3c16Gdhfksp ueXO7OnNQT3hKYebn7/ehPY2Z2ss/y27kDCl6LRpJ1VL+suPp2exmr89v//6cfPl+fUblvUi+DRC DaWniYWSNHc6mnFPOVUeXIL2dRkQvfYbRVFlefz6/P1RfPZNrCeu38mp7/Qq/F/T2B+9rSlNnMK1 Ql7OBCKpyLIIdPTQb4VTNLMcUSoEPQqxY+YVNq2QFb07kyT2z9AAU2f1AGrmtIykop+g258QsDNx SCqypHTnJPEcdK4J0w8Z8HunlSH3KzndOSU0dMubpsSZnwU1cRVJoLozKOQQIw3bnTOxVm+WN9+W b564K2B3DqNM9+c6LU88SUjsFqId8zbwPMjTONDTphUPQ0dugtwHETLlC2C0vujgYYjoRAI4B+il n4a7Sj+QkfLxIYiCvogcAR677hiEKNTStmvsPel1KFnREod5+J3GR/ez9C5hzhoiqc6qKahxVRwQ vVIgdMf2G63G25qhkQAVXI1Zded0Ek6LNGqNlQyfTOU82wiau7Wb12yauSJhd2mUIhNJeZ+nqJP8 FU4yNFkWpNdz0aLrmFE+WeL96+OPL94VoYTbJkQdBOMd1EpxgZM40WVmfmZ5UWstmkYmBx4mibHK OSm0nTVgUwRF5AzIQK0jutNRmmKodfLXj5/vX1/+7/lmPKsF31jLtRTgPKtvPOZ5GhtspKXP2b/A mBH0TY7DpevG7rf0S20LzbPMWGYMuGLUilexwYca0mlcLa+N2BIGNhLDIsPG9Bs1B4t85RcoSfCl w2ILUfM3nQmirZl2iTp6KUiAPtY3mahx8WNisRdrL41ISPkWmo5eIRRxzDNUzzTYQNW1jBqdfoRe 4uls+0I0saeNJUY2sGirC/tSVn657QuhXvpkmmUDT0RS5x5j+uiJ5d7eymsSUu+oqcc8RK0+dKZB zPv+Jrs0URAO+4+6ZBuWoRBc7BGNxHeBCjuxLlXIfCYntPH9/fUHRNMUU/Lz6/u3m7fn/735/P39 7adIiUyg7sGl5Dl8f/z25eUJdV9VDm7gSiZo61qzbt40slqVvos19uY/f33+DO7n7MVpv7sWLQQ1 1EwrBO3YjfX+QSfpZ9pzwNirEBp2ZA2Zij/7ummGqhiNnAEouv5BJGcOULfsUO2a2kzCHzieFwBo XgDoea0lF6Xqhqo+HK/VUbQ49nRr/mKn+8bag6PdfTUMVXnVXRgJOrwjb2QAmq/Gh2QYY+UJFD// Fjxj3cgyjlbsALflvsyuGnVPUWtGS3hET4VEjzbfygBRGtl/1Sj1rr0eLmNM9atUQZ/MLQ1aW41D d+zaypKvu5hrGBfFCFJDm8G6p6zc7vHpv19f/uvLz5t/u2mK0hujVWDqBmyOQa41A2BNvA8CEpMR ncslR8tJFh325pMbiYzniAZ/YC/MAK6bOif69m0mRrp2CsSx7EjcmrTz4UDiiLDYJGv+RzUqa3mU 5PuD7rpiKrtoxbt9EJn020sW0dSkdWMbEUK1/cHSd20JOvjdWBIaYchife4gYINjvOefAWXfiQh1 ZZkfAiHZynv7eyNm6gouJgLIZ4UOnmUeFwoGTxrgGWAP+TF5JFHA8BwkiB23aCx9RqmnBj24QB6w MI0rz2JFg0hneg2K5m3bpWKlP1MSpA228VuZdmUSBin69aG4FMcj2omqUp8VPhj7c/rbUgaVUHPh u9hDvor58eXHt9fHeXF1p4vy1LYPrk92gyz+bU7tkf+WBTg+dPfgY1yb+AbWVrvTXiwRmMf/VYvY LuUyWruDtsjAL/Dvf7qIReWIA+cD00MFa0jRnIQ+b0TbcXSNORnvTkfdrz/8vHacOy9ZTOQKoWMa VmPONriR4bFUfudNUl+0JqFsmXJ860K8+mOepgz6wO7butSe5AIRvHlDINhrt9+Dt3Azye+i+7mU KVqU6WpVVRaevRtSOIKFxkW0uQDxy3BVOxu3UEQkEFASHu1KAxBuYuDpvmBDyX+LiPmp2Zioa0qw O/F9cuiK657bNTlXw66DoLMCRn0smkz1cbSkN3vkMXJVvhynZF4ZFWNzPbOmLqVXAd+3HZfLqj+c wLbCJasxa3aIhRtaxE0BPehanSs9OLCOmZmxIk9FFyurwqm0a9GgvOOW/y5vV3SNfaHpWd+CNyOh 1oKxiVCpPlW/JbHV1N7m5V1hlV7kIctqOJ+fkdmD5NaY64plKLnI2PWdmG8ekI+2IKTeaoIJKD4J PSAlYd5ectBVRLcubr2swwjH/QiP8h/AdZucOWlb3w2d7M+j8W4W8F3RJpF8Ds6v97c1HxuPfyfV hxff9YLfaVX+XkwmB5/fvwuN9vn5x9OjmOSL/rRYQxXvX7++v2mskwEVkuQfmtvnqSbgP5vxAamj 9KzNalf2ALR/IM0l8zqJlfOCY5zXtrAWqC9rbHet81T+0tSF2MbhVajw2tXtRZb1ZJh3bMpbzwIa 97ZOSAhv8RBR1O0BJcqEZsxQG+1OnuheGl/PBjGARe+0mBFWKdmNTyr8L31U9GUxSOpOhYQ5gvMZ hvq4n0fJeHfdjcWZl678ebeH4d2IKbFxJQVot8fpk8PnodtViOC5TKYGz9i+PH1/l4aB39/fQCvh sE+5AQ8syrpBPxGZu8BfT+WKabJvs4YyziQNwODQoJX+mtyqTHyygVwBXsZ9f2Bm7/t0EdvBFuns EPhDzeyLUguLCxLlT5/U5wXIxkp2up7GukHED1iY6pt/E7mEvjTJBuJEDbPxzSVLspnWNQYSmlHI bex6iwa3s7l8RbyLQ9SQSmfQ/R9o9Fi/DtXolMYof6K7pdTpMcGLRiP0JZnGABE2saRNQRPPm4GZ Z1eSzOKxOcSWsujcKjqeLxaAR7RBr3RNjgjJUwKxD6CIWiKBBANi0sRIb5IARfrxBJj23CbozS7B BSGgdFv+wBPh7jV1lgSNxaYxpAFe5tRT0XSjnmloOjfSscsFGQUT4BtdAo7CCL+303li7DLJYMhx MYPVKnass3BcSGCYWcyAVECRfqgUU+xjQm/aWkwrnoYRMu4FncTIfFvxLNLfF+p0gswrim57yrTQ 7Xn2MLZJgPSK+njsIMhREKGdeXkBfPVEUV70CSb0+QD1dWKwCJWfYR+SIN2cjSVLkrp1kEBOUm++ URrZbhn8jBwPmGqwmTZVZgG3umTL2ywPk+s9nO3KBzhoPhrX9AZns+hi5xYmGX7zrPOkWe7xTmFw 5ciQmQDToM4GrYh4Opwllw9bYObb7siCKzJuuC3AW0QJojMcgEKAzO1ZM4LPmgtqRgRaURqSf2JT 4wR93BiSyw7GNMFi0EboLfrC0IglHpl9hlFMyxn0c6y1YMsdbmkewBAhM4naq2Ofo0mGLKGKPhXD wdIAWdokWabAoBCtqyB7vxGixRVkPAU/jI1plLAg9aFlJe/dcs0I3i0XdKjEf3qsmdXN2pWJv+t9 jcYwWlmH/bR1UdsD93vTfsX9Cm9JhDo71zkSTE+fAN/on+GPRr/gi2mCu1hYeEZmhbhAGCg6N/Ox vnLUr9bMMTJOqP4M2QASVE0HKE22FTnJk26tCoIDPHK5ogUgDZHpWAIEraiAxJ7igyLBUybU/nfh 2LM8S1FdSHsW9GGj6rzb0/rCGYUX5IhqhckFUbQMGJ+sTRZPZ12ZtnrZxFUWlzDG24BHjJAUDbi2 sCj1GU8uMLrdhPL5VbSlLt23GQ2RuQroJMK+K5GPssyQeRmecmGzL9AJsnOQT7+QJUTSUzwfTI8G OvWUh/qquD0WJUOCjXVA0BgcGgM8CMJKI+i+bdKEbo8OeM4e+CqUb2rNwGB6izGQDyqUp8i2XNKR bQrQM2RJvecMHtVgZfjUROABZKMQn+RpXJ70BFl7QONO9ScHCwAuUpAeo1ynYCURSLJZkCM7ZTRG JQlQhj5dMzgIsrgoABHy2LNE6GmM6Fe45smgVRClKMAVoXfmUMrBYWD9rcM4sWlXRuryqi7d63RB 1OUgfq6RIcahOh7GW7QMgnFg9yh0ukWtyiDrOVjbHGPz2/MTRJaHBM4hKfCzeKz0CyNJKwb9MfVC uu61o2xJ7XvTtbgknuBWzlO8XdXc1Uczl+IWfMSY3ytua/HLJnYnw78W0FpWsKaxGPuhK+u76sF4 3C1zkFaMPnFfiwd5gecpu2iOQ3ccLJf3K1WIx5OyajnIzigjeKqwAs8D9ZMotrd8h6rd1WhPlOhe N4mSlKYb6u7kiOFcn1lTYpMooKIE0luPKei7h8ok3LNm7Ho36+qed0f0CEgW6WGw3OIDtYbn/7Ys 6hFTCgD5ne0GZrOP9/Xx1rRXNPC76gjxLkf08hwYmsIKgyOJuhGVIhy7c2fRukPtDqOZCj96Q1AL ssefkAA+nNpdU/WsJFtchzwOtvD726pquI9DDaBDXbSik/iE3Yp2HrqjWbeWPUjvGqawhkqNBmuI 1sXQ8W4/2i3WdnD/VmFBjSV8asZ67olGwuPo67vdMFZ3dp/s2RFCKIjB4Bs6fTWy5uF4cVKKWQis qzypGgZeUkRn52aN+6EWq61das5Eo9958uKs5Sc9+oQkgscLCBPiZDVWDDMlmjDR5GIRqLiT7HTs G4/bLtl+Le6xVQ5ccN3FeI2Z1sm8Wzb8P2PX0pxIrqz/CtGrOREzt6EAGxazUFUJ0LheLhUGvKlw 27SbaNs4bBxn+vz6q5TqoUcK96bd5Jd6lpRKSanM6p98BwVoK7RGdZaPitkTSQgULppssa3EtLUk W7Uq17xqrF06RKc6pa1hOa0LPrbEGGNpXlEz+y3L0tweDLe0zM923+0uFuukV75wIYDAReU6tHNu kEhUHpyvyV/eUkhS4I5AsLW+i8qJqiZwnwrqiWY54PB2RjgasU0PfuDyVcRqMNAW+pSyGNdHHXCc 8YeUmjGkNyWYEdEU9S7foDwWWwtt39OSpUan5yZyqUM7KnQ7LEG1WxPLkZNIAN5WHOsVAXzl8VdI NFgd3094uHojH5/nIsB4LLpMHwIdUUiUaoHNa8lhuIuGMjb2b5XeoYbJmi4YNSIrKIRud1nOnQQr Nr6cz6KbwPADrbCrsV31tag7uyjzBH1PJhiia9VgjbTi13Y2aYV9qVToThXTrRBbSmdMp0Ul5qfD /U/M3U2TZJ1xsqAQU29tmuWnXOiM3uEi1DcJtVq1Xph/QDSJM7oBLULrZfhlu6TqacptFYrIFVGs SHlpwWEJNsIZ2HGuNhDJKltKKareidIYexMhExJSjfCHgArOxsNgOjduoRTAxxeTKbYWKBiC9Y3t SoJBmRlzq6dPcac8qvG2f00LLofD0WSEPmCVDDQZTYPh2DiNloA0zB86FZJk3FtPj+OX1i1+McEO Czp0bjyJaKnDkU1tPJuaRBWcOMCpTiwlCXot11XZEI/B23uATu3SkmJqPKtsiVPphjZNzWDTHRpg 99c9arcUiBdu0TP1KsfO3vNyoe8d882ATv+kg4DL8v9sMrQe7ytSrbHto2TqIiubxGgUTPhQPwtS ZW5Si9I7jXdmUBzMUDe9qmeq8XRud63jE1hSq4iAd0ybmkTT+cj93GIaTP+1iPAKRgxvp4qMj0eL ZDyan+nFhicwfftbMkxaNH57Orz8/GP0n4HQLwblMpS4SPMBoYYxRWjwR69J/seRgiHo2Ni6q7oq 2UJEdbOd4DvfIokFug53uiqpek+GLvHOCpAG+E1OhweX3snpBjpRVVmmY3Xe3vVe9XZ4fDQWJpWB WDmWtLRXp4bc2N0/o1gu1ptVXnlSplXsQVZUKF4hJb6U/XNBW4y1HFGBBQ8yWEgkdhes2nnKMCOC GlAbiFBuY2X3HV5Pd9+e9u+Dk+rDfqRl+5Ny3giOH78fHgd/QFef7t4e9yd3mHWdWpKMM5rhpqpm S6Vf0s/5xBaX4ddLBltGq5iib/XMzOB4MvP0j3L413UriSIKQQNZYvQ2GY12Qi0hLEmo9paoPZS8 +/nxCn0mn/q8v+739z80482CEhV4uD+QVaSa78RWXpSZVRxTPiy2Ik+S/Ew267ioPDbtBmOYoZeh Bk9Moyq58hcmcLpFjZwNtuRsJnCK8mkWvLjK19rrcxOttkXpBdv3IPqeEvtU3QZS/JuxkGTa1qKn qaiZKYn1BtmwGj/oR9BYSRw38wZpvsYH73DhnUY/EEtwV8rZxlMHVuQMC4dZVpHY1YZaPoKgdHbd o7MgrqIq5zusXoAKpBJbZDOfhti+0vrydroffjFzdTaQBprdWJHhlafGSoiAFyGNvt/dm0HnIQ3L qgWUvMA7u2OBx02exkhcPfty04En3zWjMvKpJz24aYYddrsuwWEDVNrZsLXMKsCM7s+6AUgYTm+p fpbTIzS/NS7je2Q7G+LqR8sSc3h9/SkLuh5rDBeXgVux1S6dTU2XIi2Uku3FHPf/33M0Uf8wYD7z 5CpDTZzJtg1k56Qt+TQaX6L+yRsOxpNRgCdWUPB5aiOcRYNsBR1pZxEt7Kt4A7LCZ2Is4wtkuEjE CxhBM9punYwqI9iCQa83ceVmhoSb7qDrcYCHge5mjPIDf+4zRhAHYe6Wy8VGdT4kbmUXKdgbo19e TBL0elZjmM5GWFMgaeAJPNSw0HQ8DM5PsPJmjHu80Rn07W9Pn82GyJfk09Rl5rGY6LNOIynYeUEE H3c+xvpLIp/IA9Egt1qSjox0oE+QZkj6pdsSoJs2toZYQe0Tuy6bG29c+g85gU/sVKHcXoxGaFFS XkzOfTUl4wLPBA5G6AuLLnFUXM6tvkIeA8FnBJev7rqCiPFxgIdwMCp1iXQCjM95FLSFFk93J7En fT4/gMRXCmaIuBP0qRl9WkemnnAW2lIzm9YLkrIEu0XT+C4nyIyJeTAZTrBP4vU/YTBgo1dGMkbm W3U1uqzIDJObswrrGaCPsRVP0KeIrEt5ehFM0AEWXk/w85HumxbTaIjKNPjY56Sh601DG0COKwyH 6XaXXaeFo8kdX/6CPe4nY/icz/xO0Ffif+cFuv1KqZ900iPOuWmpIie58uNyjPemc5Lbmc1w5eAQ nUMxxBMHTdnQvXuqJ1S8YHDdOcFuQD3i7kcQ0Lp4iSuSZVR/EgiofIzZlUwgDgkRA24JRbjdE29q smWQ0HiDIl/x4ilUWEkmwAvNKK55OKnGSB0XsK/p+lr6eFlBkjpdpprO0QM9s6gR1MaKENZQNUMr of2rdF0HRk+H/cvJGH5E7BUjsY+sPa1PSaPjO11el4R11xKCHK4XRiiNthqQ/4KhPpLWKpnxecTv Os1vaO+USx8kgHKaLKBS+NanYVpR4rlZtaranXGstzHjRUK0c49VPJlcmoEN4c0x4RFjtceUoBpd XOlv/QpSSh8VBcn0R8XyZwv+PbTIZQ499vfUJKuroDoVW2wIFGrlFeZ51WFftB2oSFaCP6QwAb8g aKfpLNiNt4ZbliKqWc8WY4+v9aM58UOMSSE6ljRj5bUJxBBpAwOKcq1frgm8DneFvEIjmWisZuSj Tgy6sCQaVT/lUr8h+rkRhLoh38QFNhMaNAQfFboBTUOXXlTcIlKs3BTeKSg3dHUvC1smUbxRK/Eb LsSxKq1yXtUsrxLtaEMRS5bpAV2AZrM0zTdoGXXYeMSZlVN9w3P9Erch2hWXVLAW4o0tQZ3QJYl2 jmyXb9vfj99Pg9Wv1/3bXzeDx4/9+wl7CP8Za1ulZUl3oWG7UpEl001yhGClMbN/d/fQNlUdI0vx w25pfRX+HQwnszNsQkXWOYcWa8p4hAXQaeAwz3Bj1ga3TRtsvBEs51gYJ5jjFpNpFuhhRzVizYlD v1J/1Umbpi4kiedULpldBuPQ/UrKP2W7tpCXh7fj4cFw+9iQtGWmSRnmuHXvktfgmgCEpCacMsZ3 nBfECEQGjgrRk9lUDug8LfKMZpXx2dJm9vhStZ6mzATSXTH6jSQcsxTTcyVmPYm94pe4XthMhLqN 7NOlMICaFKF6OnUmC+i7Mte23y3QOsh0EWWPZBFb75E2OV9ixLyAKySs5tLA80x9S7JxM7xhYWlf u3YNLFm8pHFdrLD9V8Emcl1X7kvv3n/uT5hHUgvR1D/QI6GzFrpDTbDkgbKtg9hVCvYaUCteh+gt NET1FP3L2djwkg9xzVXc5DbuZzvb7LDKF0NDsVnmSbxgfIXpNESoY1GiSXzxA06IxdeBq51fNiP4 ciqIrp2qC1Mrk47mhKTXoC7GrgecT2amhGoxdRyKIZxNx5MRmiFA+vMJE9KDVpjIZOLL7nJoifcW i+KIXqKRcSwm46hYx3gwHA7FNgwvugtP2n8zOH28GE6GaAI95jVW32KD6SAaw0009SSVQdHRzafG 1EQATE2Ll9WGFyyzTbnUBuPpeP9zwI8fb/d7d58pL7sNfz2KIn30aFebZaQAU5yDlXO0YoWY9dXF JEQ3Emj5Wh6EJWGONZqJRq+14I5Kouxf9m+H+4EEB8Xd417eUw+4qwR9xmqWIzWaRfespdw/H097 CBGGnkZQsJ+1b436IGBuYpXp6/P7I5pfITbWjeKL52ik7IQbLJcbVva3zMePl4fN4W3vHgJ0vHKz 3ApooaEO/uC/3k/750H+Moh+HF7/Axee94fvoud6uz6lYTw/HR8FGTxt6Y1otQ0EVungBvXBm8xF lafft+Pdw/3x2ZcOxSVDti2+9v6/ro9v7NqXyWesyhbi/9KtLwMHk+D1x92TqJq37iiuaWnwvos5 M3l7eDq8/Gvl2S+cMjpjtNavsrEU3TX3b336vlJFCqvvoqTXyFyl2yqS2z1ZUfrv6f740oxB1zpU MQuZwm7zzLCzbJAFJ2K9wvS0hsE0ZWmIYpUbTaaXlxgwHusKekMvqsyM59bQywqC3BOHztPp1DzN bwCw+PZa00FQ0RJ/38TMJG1H60HKxQ91gW2SrP0XkOCobVFpeicQ2TW/CPTLMCBK20fjWhGo0uBv hj3xB1QsiWZxgtA8FFLX2uW1jBmCGNyX17BEmIeJYhnDdGg43iwJJNHZ6W1W8HrJcGHrlNwVXIDf VnN7S0smCmdFHlUk0bU+TiszALCBhGWU8iqEXxExnL0qXJ1lLjHHMIrBNdVXdPA6sOOReZqkrliE ds0/vr3LCdr3ZuuXU8B9JTWiDF5WxwYcRml9JWYajNOgSdl/epGmOVSvhdpbWiZaCJfM/BlDOKOl +UjNQElyg453wQODl6XbWXoNlbRzAG+6Sd8ydC4BX7EldTDL0nrF0eFl8EBnmM3II5rkFQySmBqx 78xv0SWBjVVENL0yjYxnLuKnmGuoYQnpFA19295Ojywucxajo93d0scEU56ktUxfM/nTFiQNEXZI PCadTrDaDE5vd/eHl0cs6gSvUlyWySlgv+ttH9K4WWoHCcUSt/ZbcOzJm3wQICbTto/TnX48nQ6v T/t/92+YdpWutzWJl5fzADuxbFA+mgwNYwWg20JdgzoNvD1yQ+rQiR2W6z4rxC8QS+0q1ndtwlJ8 Fyuddor/ZyrYhbZ5WAPiOf3QBqY8kmk3z+0Jv7k8q+gSBzB4k+PcDLKovDBT8U3g0IyjZQpM6NL6 hBDrY1DrQ64h1FtSVaVLLnLOxNeIEhfiNFqXytxSX37HNeqUWiATy5l1Q+rLQMdcy9UW58vbWoD/ CePA/GVziDzTMCLRiuqLCxM9KZCF8Vy5IwvmyGMn07JIr90sW2DDVMu+63A3B1+XIHzYN/hHQmgl tw7UrlcLHlhtBufiAc4dVm4XtbRPvmbHJntSTqOl56t2rOVaaHokE1zy+FgbvIrF8WeuyISLbsJP kvus6QJ8nrMFVoGMJU239GMmUA3XT12BBG8u8L5qUtjzqyXr88vM8vyQlyyqD5HqqNN8lv0jpJPl p90ppHUY7uXjngXNJx3gcMCUMIqiXq/VZrwellA4wLkyrjhSsd7Cid/OxvVKCQ2m3BUeN/QCh89q ToyO6H0Y2XOEayYWtQz8fWWkWpf6TdeC23GXYpvAFEE90NCrQBSAlH29zitDU5MEOJuHdyFqwVmQ CLuOlk7+G/4NKTOrsxTga7NCq5JqYvB6kVb1jXYGpwiaQJWpokr78GRd5Qs+MaaLopkzaA0+KrQd QCQIzq2kzpCLT5KQnS2eOio8+GcQdKoWf9AhjPGSZENk+KgkyXH/JloqlsUUt9nVmFIqeiQv3GvC 6O7+hxHMi6tV59kiSCliTuYGAMf3+bIkuKLXcvmttVuOPASJUCeMY1sKyQMzTv86Hc25YeyRrnq6 MtO0WvVA/FeZp1/jm1jqM4g6w3g+v7gY+haudbxwoLYcPG9l3ZHzrwtSfaVb+Ffso8zSu0lZGQMu 5SKdNdxuFl4BT6ru/hRuogowY5iML3UNxps4q5yFVJJ8k1WC5cZQGs+1UW1c3/cfD8fBd6ztfYwR nXBlmuhIGuyK9fkuidBY8KDB4GbKhKIVS2Kxee3JV7TM9KKszU+VFuZSJgmf6BOKRy6umM5N4Vop KqlQlPW7A/jTL+TtntLtpi4fuPSWY33HK5pqlc5LMG2xZByJ269qEtSXa2kLR5WgckXDR8rKylL8 Vu4sjBxC6tPwQquO1Bl4/yy8Cl8k5rYhs+VvtaBbF4ANlFYen1bXa8JXaCE3W6uJKcvEl9crnad2 LxRWs66z7cTiEaQLnOQojmVTALbGChFXGkHrFAXmfgL7sFaPws9CFG9ym/8m3wTlc7hWWsAXt2qz SfBbxd3yKv6N8tzQMmhFWmnosDkMX57+N/lx/8Vhs2J+NXS4kTHOoRTZtyo2cJigrjd2/MaaOmv/ tomWuR8UCtomL6908YAJbt2YU/zo++DwfpzNpvO/Rl90uF1JarGSmAk75NKPXGpH6wYy0w/XLcQ4 R7cw/D2FxYRFSDZZLryl6wErLCTwImNvbhMvYlz1WhgeS9limn/ONB//Rk7zKXadYuUTeNoxn8x9 vXI5MRGhWcH4qmeerEaBd0wIaGR3l7Qh9VS8LWqE18BqTEse2+OuBbD3LDo+xYu5wIu5xLnnOFkP /mHQJ3juI6syVzmb1aXdeZK69o4NMI4WS5DHfVzLEVHwv+LpG8Ug9orrMjdrKpEyJxU4R3KRXcmS hEX2twBsSWhiX/XYLGL7iMnZFmei0kSPMNgB2doM/Wv0Ax75t2UR2/IrpjupAWBdLYzngHGCLw/r jME4R60c6s21rhgaR7Dqkn5///F2OP3S7MY7HXdnrCrwW2w6r9cUznthy4QvVrTkYhMlPhykACNY fLEJmyyRalfgdYzGdeNxsl261ClJSzdrVscriEisfCHiBQKXPLJgkcvVLqbNMRVYQnN5j1aVzDwW x06yHBDXe8E0a0XKmGZUPdaGPXYtI+4RY9fhMBnmOE4OC5EFWOjhVRI7cRZJZogPrcJDI7VrN359 F+gvGRKeCh3neP/z4fjflz9/3T3f/fl0vHt4Pbz8+X73fS/yOTz8CW+lH2Ew/fnt9fsXNb6u9m8v +ycZTHr/Anc0/TjTXD8NDi+H0+Hu6fC/O0C1C16xH4MmRFfi22XGJaeE4KpfRizsvUygVxyKdSHm tumPQouUjtajhf3N6Ewe7InUFr7NS7W9MHYdYiTn7RVT9Pbr9XQc3B/f9oPj2+DH/ul1/9b3gWIW 7VwS3WWZQQ5cOtWDj2pEl5VfRaxY6TfTFuAmWRkvEDSiy1oaJt8dDWXsdMlnu+LemhBf5a+KwuUW RDcH0OZdVudlg0l3EzQnpCh3HTNOwoTax/4N13IxCmYq1KsJZOsEJ7rFyz+x23PraiVEp0M33xQ1 xOYlV3NZW3x8ezrc//Vz/2twL4foI0Tx/eWMzJIbp70NNcbsWNtyIrdCNIpXCLGMDTP3ZmCmSPvX 5Q0NplP5bltdf3+cfuxfTof7u9P+YUBfZCPE3Bz893D6MSDv78f7g4Tiu9Od06pIj0nafieEFq3E WkiCYZEnO/MFcjfplgxeq7qtoNfsBmnyigiBddN+hVDaOD4fH/SD17bsMHLrswhdmnk/11HxE5Km Gm42SWm4/Wio+QJ7WtCABVTRbuLWPBtu5ybdbUqPY7q2K+ENS7VGrWCbanPed93q7v2Hr+fUY0JL gBmv99rKYp18ozjV0ezhcf9+cksoo3GAfB4gu4VsV4aTwoYcJuSKBqGH7soRkXk1GkKcFWfkohLb O2bTeILQED4mRqu0n4mQL1qm8Qh1PaDhF0MnU0EOzPh2PTBGXw+3E2pFRu4sE5NzeuEUIsjTUYBU WgDY0/lO8ozdrODKKdTfTrQydlmO5oFD3hRQcjNGo8PrD8OCsZMa2BwR1Bp11tzi2TpkHOk6UkZ4 UI9uROUb++2BIy9ISsXOCjN06TjU85xU91mtYVNMCgk6aoLfLCPU0ZvqhfyLiZAVuSWeV1vNtyIJ J+fGUCvOsZFB6fm8aVngJm7d4Jkg6zC2eFab3P4abVDn17f9+7uhJHc9JY9vnRKS29yhzSbusExu 3Tkvz0MdTjhlbaVfeffycHweZB/P3/Zvyi7e0uG7oclZHRWYOhiX4VI9hrRLkggqlhWCCTWJRJWr iwHglPAPA+dsFKwzi52DKpdaZhA0C5KV8H/zjs2rZXccpXnZjcBistwUv1GWVP59bRFanlRF8xAO pyuKjXS4vvWXAy2WhkHWDubp8O3tTuyi3o4fp8MLsuomLGwEG0IXEgoFmsWuC8XhVlfj8lcamNTk 1nLyseBQp2Gez0FXRF0YE2hAb9dioU6zW/r36BzLueK9a3rfujPKKjB5VszVBtPy6U1NKrEugE5/ Tjj2jJD9cHJmFQHW5sG4u5ABCE6ntxHF71E1vigSC/P5ckgKYR2ierlNXOFj4rbZAOG7NKVwQCTP lqpdQVGwWIdJw8PXYcPWG+T2jFWR6lxIxbfT4byOaNmcYtHGXlK7YLqK+P9XdjS9ceu4v9LjLrCv aIqi3UsPHtuTMca2HH/MNL0Y2XSQF3STFslk8X7+kpRkkxKV9l2CDEnJskxRJEVS/8YomgNisTNH 8cApPvkiBWv71X9HeLQO5716lQTGEWGGZmlDmijmzLnUFnFwejpjpgcYV89UX/b5/u7x5vzydHpz ++fp9vv94x1LHqKTJu7060UIVYwfRG0Fhy+/jH3G5ybloDNtkfXX4fN0ats1iBasoDqMOrGP3viN l/bvtKlaHANFOy330tdJ6YllPj7OHSvK4CHzBmx72B97ltSJRQ2yfqbjfKFDYs5CMPplPKDCYho6 42Af+N+WY3iru0dtq7bA1GOYmU0VBA/3herZx6s8yrmdmg3m+7H3Qf7J6vgZXV6Fkb9g5cDShn1b gC4+SorYEMrnapxm2UraYvBzqW0ihR1hYBmXm2vdoGEEH5SmWX9M8aSlgAlMYROXPOaBSr+CP3Fm 2CyGKG+pvYQzQlnMW1GN8V4DjFWYRs6UQ/Fj+vXjIrQoY/hX3GxAjZBa61e7SQZQHm8goVrPPOqA zUUYZcCoWS8LvQgmCMCCfo1K/jonEvgdNyvOfgqHPGR1ELiYDYPJK1gWhxJUo17UxMkoGJ7nYCBI lGmlKks8PrUF44WgsJRIvwsKNsG464xiFnakDrOR9HjnUoZBvOPUxT2veKwzhOit6X0t1l9Q5d2k kCAW89eVwfjaUnIICG1N68nnRsgLxC6ozphaovoyonaRlQoG9edADxDgeeBb12VtvzgjvuIyrjYi Hwl/L6tKPS6UodcLV42mqXJe6qrupzlICMjrr/OYMaceVvHBis4rpOkqUaO3qBrxG35seTlOQ3dU XcLO2F/ztw4mmc5virLjdT8GkHdicvGkrb2U0tftrNHGKE+WvD5B0J9P94/n71S18NvD6flOq4dF 2+6ecs5TIcj7Ga/BE8aajeXBGhk1bJX1clTxKUlxNVXl+PnDMrtO54p6+MBORbGClBtBUdaZfsxY XLcZfPBXwmQFBZUpU0OHmo1B1bTseyAv+Zwn53FxQ9z/9/TH+f7BaTfPRHpr4U/arNuhJNJJnEna TOg/wlyAdeK3PQyNgsE/X7x7/4HzSwdSEjPPeBhjD6Yv9QUovrZ2JeZ/YgA0sGOtZaO49WsTDTBQ sslGfmtfiKExzaatZYU06gVkWw6DLrM9HoiHBfZXnfF351HUJ3CcX5z+83J3h4eR1ePz+enlwRWW 8+yGF9GhCksJrzFwOQi1c//53V8XGpVNa9V7cCmvAwYFtDmVPJOzwEW0g5BAPOJfZdYGOicjggbT z3Telj3huXDqtJ12lP1lwWSY+7WGBcDveWdaM7lj2kRIOdG5F3YrnVunhE6XxCP0vtAOSqbNkDFv Kf3E0uwiq9lCN1hnQXNxWDTG9q4dkQlnO2fppr/FQ/KbYbBzqXwtfFzkoHTn6Eu/LD4bZR9YaXht JHcP284Q67fK4DkLyksJx7ia7MZnmGMrTGKyc02Ft2hKB5vEACO53Cw9rkUS4+V1r3AnUfelXurP ktgMBp1dhnraeDI9hIooUm464n338UBNrEEMxdPqMa8M0QZeTLhradsHCOrC0ZRtEcrtVZujvg7N 3F2OJGmCT39o4sEBNR6xhVFGIU2/UZt2l2A9JOKOwtEku7e1FSh2hOvZy8zSi2PSzBakUDwMgdbM tJwmZ5/hyo39eRaL4biWNYkzsZgd3tzgrQ4Zu7KuuWgsOyxkEC5Von9jfvx8/teb+sft95efdvPZ 3TzeSX0JL0PB+BljOjWomuMxnXcqRTnNytb3wUs03jEONtsR5S2aFOUIS8HoYUmImncTzMKYDXvO PHb7WlDLQy7es8egRtVloEMwwi683eVXtMtLLd0er2D7ByWgMLoCRtLXvp66878++TY0DxSCby90 B14sTe3iCywRC3SOeQ5b07F8tJLSd8g1OJ37suxS3jEn6cBqbbr4kkd8Kba9/OP55/0jRjLA+z68 nE9/neCf0/n27du3/2RXKBl/DyFVElzrMy26OZaljbIsLRivI6UOWpj7QNQTHGchudzBpGymsfxS RjqLr34VwhPkx6PFgIg2xy4bdyFBfxxE4o2F0ggDY9EmpXQaqQK2ViA8ttSb4IzSQZQv3htNECxB TBCNSlt6nl/ezBtpa/GTfCtac5Xj73DCorz0GejPILVIkAemJCFXGKnhMHN4G2VZFrACrIdK2fLs npuQhN+tSvTt5nzzBnWhW/Td8lLXdiYr7vBya8ABQ7H72rqxoahB5c9VatHOPxfZiK6Lvp+i9OBA kiQGL8eZ9zA97Qgq7FqLLJ9UZc2uqFxU512AqbRLwQE8GQmaYHGbJGchQdCYYTC1fW0ucWBZzWTQ LdL//YV8MHGLOsmILa+UFEU5cAoBni+JI8GYq4xePUXOZCATrpxF2HuXV8ANNqEctGP0m2kzhN7Q Nr/GG+JXl57p7Nsxbdf+xrJUc7BM7JLKpSQjn8tm2m55H+UBHWVIH5TrbUcc4XCs0PoNn8y6cpbZ cOQOHrdRoNeJmoIu3fJQyOh53rulvaK6O2z9G4vNH/nCt9Fig/sr0ES261MYu9OAHUbXwmnDfYVg d4QPpxDIrzW0WSev/gsQ3okQTOkGBB18CVerPIqQ9vCsBWGT4WGXbVAmUug8OSymVwk39Z7ORisT Ly7P1c7/RKwkEmXxWrQFuqpLxJyW8WyxB12rWhhnPfjSxAljRX5AFj0ObHl0hydq41t+Uo45PGLM evSIS+S6en6HgvRETEiGWR/0xaR3wpYTOQqDgF420biQQmyGJcPEF7Ag/iHUBAlOZf2UYa8Oac8s QpzbgZUH76kqaMIcdiRYuPc1AvtLTb5wFIctVn9H5m0KzJwQNbWcpAesdhvczdOttlNefNyTUiLU a0nL3dXj6fmMWhDq+vmP/52ebu5O3NbaT23iINArDOitNf2vSqKky6ZICxPsytwc/Aflp3xTayWn VaajUv31vhi1OFxr+FRUcJtfjEjwpmrpWojAS+Uo10mwvMBL5eiCyCuDtEaTysUGT9ZipQQPxAZT G6x+m2gqzuZWxdYvwbJHmZxo6w9lFG2ZXnpXfimmpgug7hjEpukMMXLIeZSaDdMA8GjEnTkEt9EF aevUnsCkPuE08cLhBPoSnEMScPGAhI/v8TA55WKxMyBCnglUFVkAqfdNAIFx4+GfBB4aazuFo6Cg t7CWieit24b9Y7jHzpCjkYXnU7ADPJvtKLIdq8Eux2DrTaQ/xBSd9MiPTglcLhlOtgQ5lYN+kfyI JJnIgxWsw7Jxfi3ZHyVIoe9Sl8PQLF6LMkNKl29RGpU9s/s/J7pmGBahAQA= --===============7686323094706785872==--