Received: by 2002:a25:e74b:0:0:0:0:0 with SMTP id e72csp1848183ybh; Thu, 23 Jul 2020 20:36:45 -0700 (PDT) X-Google-Smtp-Source: ABdhPJyj09WEUt0//rnm1jIBS00nDSuiWyaWFBh5OMEhqwj6pY7Vs3drT5nFeP7rMvlFVYQutwMt X-Received: by 2002:a05:6402:1d0a:: with SMTP id dg10mr6803009edb.110.1595561805523; Thu, 23 Jul 2020 20:36:45 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1595561805; cv=none; d=google.com; s=arc-20160816; b=RIHU80Kh1jnhp+w/4OQlng38Paxfqx8wH+LntMxlPrSEVVE5oAuaAw2Tpa27B7UhTG AB0YcI2O/44cgo0Mv4X9SMxwYvp7Zd4rtFn//aeReZ3NWEMq821pc+3pOMpcq8k3gpU7 UXfBUFD78VjxOuQ9vdlfo2h9ZdLmugztl+QYYuqWIwiZRruIbu6nGYRpFXuqZNNk+02D UyVhYW6/1pFQPf9r6whl+aYJjMzWwYUHz7depeIrs+iWxo4z2LjkLH/BPBLRz+zsFGEQ iMRhIxOPlvQuyPJ+c2/slCf+RlZhKBFWA9YdVKd8vyic1EwWrfXicvmmHYeBQjjarn/v 1VFg== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:sender:user-agent:in-reply-to :content-disposition:mime-version:references:message-id:subject:cc :to:from:date:ironport-sdr:ironport-sdr; bh=HH4VRJBcoWVUwGW1TBSALtrwrlRUhGgiKrgOwyxYRKw=; b=FvrCANX1z/hruwmIY4g3zAR4FjfoAap6HFBiV06/6yG8lNlcDtfGdnJqgAlc6iq/OA 45rbliFiH50lQZ0h6hHhmLy01BMKA+oOKQef/3bfipF2tB2Ow6hU+t83cS0JQLowIKxs AJdUMO+8Xim7YEusrlGk1qFWcnawI+v0cV1jkoyLoBtlLMoou9kgxrfU6CpB8moMlAKJ I2GMoQBqT8rtEKBEkORVedEokvLyoR6wmdoqeDVat3+W+3uMXMgNK7i6OA+1HG4T3rf2 QBlpsRQle7xG7JbS3RO5zFNHa5CEmgb2++vmJZlohyZV1UZ3QOmW3a8DWx/atyUTe6eN nKag== ARC-Authentication-Results: i=1; mx.google.com; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=intel.com Return-Path: Received: from vger.kernel.org (vger.kernel.org. [23.128.96.18]) by mx.google.com with ESMTP id m1si3087953edr.207.2020.07.23.20.36.22; Thu, 23 Jul 2020 20:36:45 -0700 (PDT) Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18; Authentication-Results: mx.google.com; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=intel.com Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726719AbgGXDf4 (ORCPT + 99 others); Thu, 23 Jul 2020 23:35:56 -0400 Received: from mga17.intel.com ([192.55.52.151]:18292 "EHLO mga17.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726178AbgGXDf4 (ORCPT ); Thu, 23 Jul 2020 23:35:56 -0400 IronPort-SDR: DfP/E8ZEJ1QPKe6n/PoEyf2fXG9VFq4ZSYCi3upmc81AbXmPEOcEBfuw/twllspMoorhwP1igO c5RbB8MMfZ1A== X-IronPort-AV: E=McAfee;i="6000,8403,9691"; a="130728360" X-IronPort-AV: E=Sophos;i="5.75,389,1589266800"; d="gz'50?scan'50,208,50";a="130728360" X-Amp-Result: UNKNOWN X-Amp-Original-Verdict: FILE UNKNOWN X-Amp-File-Uploaded: False Received: from orsmga001.jf.intel.com ([10.7.209.18]) by fmsmga107.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 23 Jul 2020 20:15:49 -0700 IronPort-SDR: Iif1cxYVWhAPKZdYOgopKFrHbKGz+9GU1x7+e+LKXV/6yUtIUH/LC3Un87+jaMkPoTxaDscyBd eiFN+Lbu+rQQ== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.75,389,1589266800"; d="gz'50?scan'50,208,50";a="363245583" Received: from lkp-server01.sh.intel.com (HELO df0563f96c37) ([10.239.97.150]) by orsmga001.jf.intel.com with ESMTP; 23 Jul 2020 20:15:46 -0700 Received: from kbuild by df0563f96c37 with local (Exim 4.92) (envelope-from ) id 1jyoBW-000018-3b; Fri, 24 Jul 2020 03:15:46 +0000 Date: Fri, 24 Jul 2020 11:15:02 +0800 From: kernel test robot To: Nicholas Piggin , linux-kernel@vger.kernel.org Cc: kbuild-all@lists.01.org, Nicholas Piggin , linux-arch@vger.kernel.org, linuxppc-dev@lists.ozlabs.org, Peter Zijlstra , Ingo Molnar , Will Deacon , Alexey Kardashevskiy Subject: Re: [PATCH 1/2] lockdep: improve current->(hard|soft)irqs_enabled synchronisation with actual irq state Message-ID: <202007241111.k0s5ypfJ%lkp@intel.com> References: <20200723105615.1268126-1-npiggin@gmail.com> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="tThc/1wpZn/ma/RB" Content-Disposition: inline In-Reply-To: <20200723105615.1268126-1-npiggin@gmail.com> User-Agent: Mutt/1.10.1 (2018-07-13) Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --tThc/1wpZn/ma/RB Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Hi Nicholas, I love your patch! Perhaps something to improve: [auto build test WARNING on linux/master] [also build test WARNING on powerpc/next linus/master v5.8-rc6 next-20200723] [cannot apply to tip/locking/core] [If your patch is applied to the wrong git tree, kindly drop us a note. And when submitting patch, we suggest to use '--base' as documented in https://git-scm.com/docs/git-format-patch] url: https://github.com/0day-ci/linux/commits/Nicholas-Piggin/lockdep-improve-current-hard-soft-irqs_enabled-synchronisation-with-actual-irq-state/20200723-185938 base: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 9ebcfadb0610322ac537dd7aa5d9cbc2b2894c68 config: i386-randconfig-s001-20200723 (attached as .config) compiler: gcc-9 (Debian 9.3.0-14) 9.3.0 reproduce: # apt-get install sparse # sparse version: v0.6.2-93-g4c6cbe55-dirty # save the attached .config to linux build tree make W=1 C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' ARCH=i386 If you fix the issue, kindly add following tag as appropriate Reported-by: kernel test robot sparse warnings: (new ones prefixed by >>) kernel/locking/spinlock.c:149:17: sparse: sparse: context imbalance in '_raw_spin_lock' - wrong count at exit kernel/locking/spinlock.c: note: in included file (through include/linux/preempt.h): >> arch/x86/include/asm/preempt.h:79:9: sparse: sparse: context imbalance in '_raw_spin_lock_irqsave' - wrong count at exit >> arch/x86/include/asm/preempt.h:79:9: sparse: sparse: context imbalance in '_raw_spin_lock_irq' - wrong count at exit kernel/locking/spinlock.c:173:17: sparse: sparse: context imbalance in '_raw_spin_lock_bh' - wrong count at exit kernel/locking/spinlock.c:181:17: sparse: sparse: context imbalance in '_raw_spin_unlock' - unexpected unlock kernel/locking/spinlock.c:189:17: sparse: sparse: context imbalance in '_raw_spin_unlock_irqrestore' - unexpected unlock kernel/locking/spinlock.c:197:17: sparse: sparse: context imbalance in '_raw_spin_unlock_irq' - unexpected unlock kernel/locking/spinlock.c:205:17: sparse: sparse: context imbalance in '_raw_spin_unlock_bh' - unexpected unlock kernel/locking/spinlock.c:221:17: sparse: sparse: context imbalance in '_raw_read_lock' - wrong count at exit >> arch/x86/include/asm/preempt.h:79:9: sparse: sparse: context imbalance in '_raw_read_lock_irqsave' - wrong count at exit >> arch/x86/include/asm/preempt.h:79:9: sparse: sparse: context imbalance in '_raw_read_lock_irq' - wrong count at exit kernel/locking/spinlock.c:245:17: sparse: sparse: context imbalance in '_raw_read_lock_bh' - wrong count at exit kernel/locking/spinlock.c:253:17: sparse: sparse: context imbalance in '_raw_read_unlock' - unexpected unlock kernel/locking/spinlock.c:261:17: sparse: sparse: context imbalance in '_raw_read_unlock_irqrestore' - unexpected unlock kernel/locking/spinlock.c:269:17: sparse: sparse: context imbalance in '_raw_read_unlock_irq' - unexpected unlock kernel/locking/spinlock.c:277:17: sparse: sparse: context imbalance in '_raw_read_unlock_bh' - unexpected unlock kernel/locking/spinlock.c:293:17: sparse: sparse: context imbalance in '_raw_write_lock' - wrong count at exit >> arch/x86/include/asm/preempt.h:79:9: sparse: sparse: context imbalance in '_raw_write_lock_irqsave' - wrong count at exit >> arch/x86/include/asm/preempt.h:79:9: sparse: sparse: context imbalance in '_raw_write_lock_irq' - wrong count at exit kernel/locking/spinlock.c:317:17: sparse: sparse: context imbalance in '_raw_write_lock_bh' - wrong count at exit kernel/locking/spinlock.c:325:17: sparse: sparse: context imbalance in '_raw_write_unlock' - unexpected unlock kernel/locking/spinlock.c:333:17: sparse: sparse: context imbalance in '_raw_write_unlock_irqrestore' - unexpected unlock kernel/locking/spinlock.c:341:17: sparse: sparse: context imbalance in '_raw_write_unlock_irq' - unexpected unlock kernel/locking/spinlock.c:349:17: sparse: sparse: context imbalance in '_raw_write_unlock_bh' - unexpected unlock kernel/locking/spinlock.c:358:17: sparse: sparse: context imbalance in '_raw_spin_lock_nested' - wrong count at exit >> arch/x86/include/asm/preempt.h:79:9: sparse: sparse: context imbalance in '_raw_spin_lock_irqsave_nested' - wrong count at exit kernel/locking/spinlock.c:380:17: sparse: sparse: context imbalance in '_raw_spin_lock_nest_lock' - wrong count at exit -- kernel/trace/ring_buffer.c:699:32: sparse: sparse: incorrect type in return expression (different base types) @@ expected restricted __poll_t @@ got int @@ kernel/trace/ring_buffer.c:699:32: sparse: expected restricted __poll_t kernel/trace/ring_buffer.c:699:32: sparse: got int kernel/trace/ring_buffer.c: note: in included file (through include/linux/irqflags.h, arch/x86/include/asm/special_insns.h, arch/x86/include/asm/processor.h, ...): >> arch/x86/include/asm/irqflags.h:162:28: sparse: sparse: context imbalance in 'ring_buffer_peek' - different lock contexts for basic block >> arch/x86/include/asm/irqflags.h:162:28: sparse: sparse: context imbalance in 'ring_buffer_consume' - different lock contexts for basic block >> arch/x86/include/asm/irqflags.h:162:28: sparse: sparse: context imbalance in 'ring_buffer_empty' - different lock contexts for basic block >> arch/x86/include/asm/irqflags.h:162:28: sparse: sparse: context imbalance in 'ring_buffer_empty_cpu' - different lock contexts for basic block vim +/_raw_spin_lock_irqsave +79 arch/x86/include/asm/preempt.h c2daa3bed53a811 Peter Zijlstra 2013-08-14 72 c2daa3bed53a811 Peter Zijlstra 2013-08-14 73 /* c2daa3bed53a811 Peter Zijlstra 2013-08-14 74 * The various preempt_count add/sub methods c2daa3bed53a811 Peter Zijlstra 2013-08-14 75 */ c2daa3bed53a811 Peter Zijlstra 2013-08-14 76 c2daa3bed53a811 Peter Zijlstra 2013-08-14 77 static __always_inline void __preempt_count_add(int val) c2daa3bed53a811 Peter Zijlstra 2013-08-14 78 { b3ca1c10d7b32fd Christoph Lameter 2014-04-07 @79 raw_cpu_add_4(__preempt_count, val); c2daa3bed53a811 Peter Zijlstra 2013-08-14 80 } c2daa3bed53a811 Peter Zijlstra 2013-08-14 81 --- 0-DAY CI Kernel Test Service, Intel Corporation https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org --tThc/1wpZn/ma/RB Content-Type: application/gzip Content-Disposition: attachment; filename=".config.gz" Content-Transfer-Encoding: base64 H4sICHQ8Gl8AAy5jb25maWcAlFxLk9w2Dr7nV3Q5l+SQ7LzsOLU1B0qi1EyLokxSPd1zUU3G bWcq9ox3Hpv43y9AihJJUe3sHrIeAnwIBIEPINjff/f9irw8P3y+eb67vfn06evq4+H+8Hjz fHi/+nD36fDvVSFWjdArWjD9MzDXd/cvf//r7vztm9Xrn9/+fPLT4+35anN4vD98WuUP9x/u Pr5A77uH++++/y4XTcmqPs/7LZWKiabXdKcvX328vf3p19UPxeH3u5v71a8/n8Mwpxc/2n+9 8rox1Vd5fvnVNVXTUJe/npyfnDhCXYztZ+cXJ+Z/4zg1aaqRfOINvyaqJ4r3ldBimsQjsKZm DZ1ITL7rr4TcTC1Zx+pCM057TbKa9kpIPVH1WlJSwDClgP8Ai8KuIJnvV5UR86fV0+H55csk K9Yw3dNm2xMJX8U405fnZ8Du1iZ4y2AaTZVe3T2t7h+ecYRRDCIntfvSV69SzT3p/I816+8V qbXHvyZb2m+obGjdV9esndh9SgaUszSpvuYkTdldL/UQS4SLiRCuaZSKvyBfKjEDLusYfXd9 vLc4Tr5I7EhBS9LV2uyrJ2HXvBZKN4TTy1c/3D/cH358NQ2rrkibnE/t1Za1eWKyVii26/m7 jnae0vqt2DnXtS+8K6LzdW+oyelyKZTqOeVC7nuiNcnXSb5O0ZpliVWRDkxHtLlEwpyGgAsi dT3Ro1ZzWODcrZ5efn/6+vR8+Dwdloo2VLLcHMtWisz7aJ+k1uIqTaFlSXPNcEFl2XN7PCO+ ljYFa8zZTw/CWSWJxhPnfaMsgKRgE3tJFYyQ7pqv/cOFLYXghDVhm2I8xdSvGZUoyP3CuoiW sPEgRjj+Wsg0Fy5Pbs36ey6KyNiVQua0GOwYSGGiqpZIRQepjErgj1zQrKtKFSrL4f796uFD tKGTpRb5RokO5rRqWQhvRqMzPos5U19TnbekZgXRtK+J0n2+z+uEahirvZ3pnyOb8eiWNlod JfaZFKTIYaLjbBx2jBS/dUk+LlTftbhkp/L67vPh8Sml9Zrlm140FNTaP1bXoKmSiYLl/oY0 AimsqNOH25ATZ3bNqjWqhhGSVGbEYetmCxvtjKSUtxrGNC5znMO1b0XdNZrIfXIlA1fKqg39 cwHdnXjytvuXvnn6c/UMy1ndwNKenm+en1Y3t7cPL/fPd/cfI4FBh57kZoxAj1FTjSakiMZW qXwNR4Bsq1jZLUGvqeSkxmUq1cm0nDNVoIXKgQUn0kkmhAdKE61SQlAskKliow8pmELoUSQP 2j+Qk2fqQUZMidpYA384I3KZdyuVUEfYnh5o/vLgz57uQB9T+6kss989bMLeIIe6RrjDfbuK lIaC0BWt8qxm5iyN3xoucNzgjf2Ht+WbUbNEcFjYZg1mDvQ9Ca0QLJXgTFipL89O/HaUGyc7 j356Nmkva/QGEFZJozFOzwM16xo1QEijVsZ0OG1Xt38c3r98OjyuPhxunl8eD0+mefjuBDWw mVek0X2G9hTG7RpO2l7XWV/WnVp79rOSomuVLw9w+nmVkEVWbwb2uLtd/NRaEib7JCUvwXKS prhihV4HuqP9DonZh5laVqjZ9LLwgefQWIKKXVPpTzJQ1l1FQRRp3NMCotEqTbPdC7pl+QJq shwwSHzco6+gspwtOGvLxGKNP02dJ5FvRh6iSWD+AVuCpwa7k17lmuabVoCGorEHjEBT41v7 B2GD23IfisI2FhRsH0CM5G5JWhMPoKDqgNiM75aeOpi/CYfRrAv3sLIsoiAEGqLYA1qGkGNS o2IJzhvmNJQ3pIu0BRcCXRD+Oy3KvBctOBJ2TRE1mY0V4BiaPCXTmFvBPzxs5UB6YB9Ycfom 5gErm9PWgDcQX06jPm2u2g2sBSw6LsbbhtbTOmupp7+jmTg4GQZnIThACg4O4uV+wE9LkQpu 6JzDWYc1GIA6BAsmWLGoI4kF0Jp6sY21rg1nfkhb+SPSuoR9S2r2sngIQNuy82Fh2Wm6i/4E G+RJsRU+v2JVQ+rSU3HzUX6DAYZ+g1qDxfXXTlhaT5noO/jytCaSYssUdWJPSRFmyYiUjHox wQZ591zNW/oAHo+tRkZ4ujF+CjSrnzC1Hx1KA1fKlJEwbgqTLtPKYJAmNxvnHXNFvRjM2MOo DbrTovD9jD0HMHk/YvlJ2fLTk4sZzhmSWe3h8cPD4+eb+9vDiv73cA+giYC/zRE2AQCeANDC 4HZ5hggf32+5CcSSIO0fzugm3HI7nUXEDp47Haq7bO4qggQSATQgN+kTW5NUCI+DhpOINBvJ YAtlRR04DTsBFX0x4rZewtEXfGmQkQ0jaQiSgkPSlSXApJbANH5wG4oAMRmEqJqRlOUB5Fey OsD6xn4aXxiEO2GezjHv3r7pzz0HBH/7vkxp2eXGKhc0h6DaO2ei022ne+Mb9OWrw6cP52c/ YTrVz9ZtwKX2qmvbIJsI+DDfmGXOaZx7ANocJ444TzbgK5mNPi/fHqOT3eXpmzSD05lvjBOw BcONWQFF+gCgOYI12cGoZO9cWF8W+bwLGBiWSYzxixBhjLYENQDt0y5FIwBqekzuRq535AD1 gKPVtxWoSpy9AlxooZuNKiX1PsmEJ45kzBEMJTELse6azQKf0eUkm10Py6hsbI4GHKRiWR0v WXUKE1VLZBMCGNFBpDqg3tkIRqWUM1ywJHO6AiUHpe8Vb2dtNbne95VaGrIz2TmPXIKTp0TW +xzTTtRzvW1lo6AaTFytLsc4asjKK4JbhgcB94Xm9ugbu90+Ptwenp4eHlfPX7/YWNeLloZh rgX0D3Rw9jklJRrieIudQxJvTdbL00ZRFyXzwyhJNQABFqZAsK9VRwBtMg2WkCdjFSwnYa+Q SHcadhi1ZsIrQW+3mMXhwbZhorpV6WgAWQifxk/ENyP+UGXPsyAl4dqOOB+cQBb5+dnpbuEb QZUa0AjY4KYgfnyAxFGdhmQ0BJd15+MD6Ha2Oz2N5QKDMsnS32yDG8EZmGSIP8DeoAegMpUT 28NxBWgFcL3qqJ/GA6UgWyYDb+fajohjZFEta0z2MjHrBrx9NJ1NhLYdZvXgUNR6QJ7TwNu0 DowTRgmvlIN0rC4PMSUFLt6+Ubvk+EhKE14fIWiVL9I4X5jpzdKAYAUhWuGMfYN8nM6PUtMR It8sLGnzy0L723R7Ljsl0mkFTssSjmSYnZuoV6zBK4V8YSED+TwFwTl4yibY54oCcql2p+mx DLWvF7Yn30u2WxTylpH8vE9f1RnigsAwGljoBRgwvWfGMFrMsGBzzHlv8GssKrDZuQufpT5d pgEOqRqOcNwPmo23YA3jHTeWvySc1fuwp7EMEMRz5eFDRsBcoQvqgxQA8m/5bsk5DbllzCTQ Gryil1KCycFwWts/bzYbZaFskAZGGviCpcyYoa/3VaiI8dggFNLJ1NCAYhvFqSYw9ZEROp4H MNu1X6+J2PnXZOuWWqsmozbKuxpBotQekCn8dEFjIJjCgARAWEYrGPcsTcR7vLcxyUU6MWFq sD5G8TAaMo18SSnNlX1PWhbhL4j8542SSggrbCYpk2JDG5uuwivISCX9FNHQgFnpmlYk389I o9aEGKHJGeo7T2ID1xXvDNUaIEnc3Y77G02mRZHB3ahsQ3jnRcefH+7vnh8eg1seLwwfgE7X DOmDyQHOeCRpk/HhjDHHmxnqR4YLCwp20YgVDq4fHg5/BTI5fQNh05J9Em2N/6F+hkoLMEVZ kPFkbzcLI0iK6gCYO87vsxzsBli+pW30DdOAYZkHyhqB94RRzmpoukhjni1XbQ1g6/xbZMyP JpblGM6CHN/UGnebsZym0Q6cblGWENhdnvz99mQoIAq2qCVLYspbgtBTM6VZHsdAJZgfEAfY DpKI2kw4sUw2ttzBXbyd9yw4q1G7agdY8c67o1N5k1kYpvchqBAKE2mya4cqhTDmANVAtMfd PBOrHWBRqbBwAC+1ri7fjH4NHPF6MLpBRQTXUoZ/YSjHNLumi+2DUEYLe7LAhlLEVKQxvY75 9Ltg72L43CmqINZEG0HCiylDttmpKPzgJIoUAUdGLdZWaLUzu4UqFYs75kgjuQQnXsAkeWnJ ElukaI4JmeBkXvenJyfJMYB09nqRdB72CoY78Rzu9eWpV3dnPdxa4vW7lz+mO+q5r3a9Vwxd HRwgiYfvNCzek9Qk+MLDYXcH704wCR3ugMmrmF4qMYsBazDLWTDJGk5A3Rl44SWyx3Phkb2v tWmSNG1IhW0LJQL0wwuTcoKhU04HtpmV+74udJBBd+7mSHrDOsiHvw6PK/BHNx8Pnw/3z4aF 5C1bPXzBgkx7Z+w2x+aL0mFiKi2LELqaGaEwzYOTebTZX84fmu1UYDnExr9Ctok/cB96KBLD Lq2f/DMtIBsNFrMVV3j20SsrLx86XYMgr9mkKhnO27HaXPZOu8KuCKlLZadZ6i7pthdbKiUr qJ96C0eieaoWyucg8TdmRINN38etndahATfNW5hdLA1dknmHQiStuqGZYEXSd32rVDT9FGSM YChNZkW9SJwthrU8Zb8MLTyY8/2x05GqAuOPtwBL4wyIMpHmtWTjZrq2kqSIFx7TEvqVvoUz a8wZ3pWkgIwVtoAICSyCjCZ1383EAPbDYVWWTmbZvnElUDBhpyBQBk+v1+IIm6RFh3V/ePty hc5VNPU+5QLGY0pa6h32sH24ng2nQEJyAUWry/mp84wTw7tz2HC24DudZOHfyRNnAAEfo8XJ JIZ+1JWZrcrHw39eDve3X1dPtzefgpjDHZYwQjXHpxJbrJbFgFwvkMfqpiA8NmQ8XwtxsaG7 2i8cxqs0SI4V8KJcFezOYmA/64I3t6aE5J93EU1BYT3L2YNZD6ANZaj/z9IMwug0SznTQNKh iJIcTjBJEf4fclj6/rQCTF+9OG/yI0fl/BAr5+r9491/g/tpYLOSC/VwaDPZ/oJu0zi1NYZ+ EaK2ee6GWr5RGLxKzORwRU0aR/Yy+fnrs5N5s5kTd6oRV/3mzRLhl0VCBEpMcnBnwBAAmwju t4AkAWnYPJZkjfgWvY9gasjF8vUSSYX20az5wmbe+YKZHqJxs4mNqdJOJ0ZtLqmpZLccaCB9 DYdqkYFOx0POtPDpj5vHw3sPYyY/sWbZ0tebO1qseCStDTh91Js2wKP6s/efDqE5DrGHazEn qSZFQWfpyJHMadMtnu2RS9N0FBYwuUuZpMu0JHeBE3+s+SIvqWOO4bx+2wUF38T8RlTZy5Nr WP0AoGR1eL79+Uc/HkCkUglMFaS9riFzbv88wlIwmU7sWTJpPEyLTThj2GJHCNvcxEEuD9rz Jjs7AZm/69hCQQuWHGRdCgYMxQiYTvWSKipIqakcI8XkwKJu09dWEGKmL0Uaql+/Pklfp1R0 QfBaQZ/0Cswp2qsyS2rGwpZbdbi7v3n8uqKfXz7dRKd2iFyHCwE31ow/BHoAKbG6Q9gEiZmi vHv8/BcYhlUxuqMpYVGkbVrJJDdok1MsKknsWcGZn3+EP22NYdSUg0vhJF9jrI3XybTEOKiu MxJeZTKVKwhAsjKlsOVVn5dVPL7f6uL5iVoJUdV0/BB/roGkksHOQMR0tUncR45kIGM5NiAG UacGnohe5nl5KnPZnXVliRUow7RHJlzm2bZ+RSMI2pV1OE3Qh4+PN6sPTh8sPPHL1xcYHHmm SYHubbZB8hmvrDs4gdezJwzudELIt929PvUulbCqZE1O+4bFbWev38StuiWdGh/puOKwm8fb P+6eD7eYjfnp/eELLB2N8cwn5pKodVTUaBJiUZuL/+ydkbM4tmgt2HzXNhQEmlLdtqap8goj Lm+MeAQI0uYx0cYW7iSP628dB79OMpp2czCbuaw2V0VdYzJvWEeeY9Q+z8aal6iaNX2GDx+j UJKBdLDMLFGLtYlLi2wrluGkCKJNtw/DAHTvy1Qlddk1tqCPSokZDnN/Fb3529KwEHl66WhG XAuxiYjof+BvzapOdIlnZgokbECFfYCXyF8A9NOYNByq5ucMEH4Oif0FovW3PZ8J3a7cvmC2 BY391ZppOjy78cfC8jLVF/uGoCPQpi7c9Ij4zs8ypjEr3s8ehCqOucXhkXK8OxDtw7ltClv5 NehQ6LktX1AMHG4cPqhe7Li+6jP4UPsWIqJxhuh0IiuznIgJo0Os/epkAy4HtiQoro6LkBN6 gqkWBPzmoYctbIseh0yDJOZ3dcZyEFHR8eR+Tof2ODVR2c1511cEU2lDUgyLdJNkfKaVYhn0 zp4T+yQq5+0uX8dP7pyxGNQOr+cijqGfvSdfoBWiW6iEZG3e2/er7rF8QhjDDcpQCerFpgvt Xk/cghr0JSLO6hadWxhqGwPy7B1lSF7MxpmPZBrAz6AKxtXH+vLtZ49coFrxuJTe2bPG3LqB LLGaNNygSc5IwzHQmcp4C+G4u1tQmsOB8XLgQOrwhgD9Ar7fkLM7B5SXobiroNQygxLniIHu wBIlzWrY622oWqLdO5uoa2/MIc4ITQtE8HjbA5sAaNB/rSbwNxdYNVzCnM8IJPItIypH84nb lrLlGjyGdr9GIK+8sucjpLi7lXyye4o0ybqFPTo/c/d1oQ0ffTw4osCRj7qNls9/1JC8KfEe ifS0yeW+HZ8DV7nY/vT7zdPh/epP+5jiy+PDh7shWzthZWAbxHBsAsPmcFF0DXdspkAq+Fsl COJYk3xP8A3I6IaSCOo0QERPnObljcL3JN6VuD0xvkyH/TLJIRAwSQddA1fXHONwbvnYCErm 40+MLLwHc5wLGYSBjMqOb6mP8WBl+hV4ZqXwpxnGV4494+YmMtm1a0AF4XjteSbqhWhbMu74 NvjKKX1La2yTebgcX2FmQ7Xs+CcAIYwxJX0X1vBOT2jhQGGSNSThE8lMVcnGIJU2vafUtJJM J59aDqRen574CuIYsCY+vfWOAyye0LpOlwqbTxluto1TlfEkV1kqwPZEwPDNORzpfdxzpOdC pet67BLxKj552WN2ACu8W1LHg9tf93GmJPXkvb15fL7DA7nSX78cgiSGeVtkgWWxxYuFZFGt KoSaWMNA2W+eEnrRjP538HeY9Ao3GNow5mVi1jw8hba/nyKmR+JeOApcTNgyjQI8WpjL8Iib feYDHdeclR7Yhj96t1nuqfckbSAuvWKefhMkWOQUczen0yxdM2walswbo5XHj2WmagSblpLc +/EXY0ttZ9g5cRVcvcJBBM+yQDQeaoE2+jfzKzrFVM8/sSxT4s7yKt111j66K0xxYRVCTdoW zSEpCrSfvTGJKVfvXlb2GS3x/zBuCH/7xeM1hSj9lYTB/W+eHtUbDaN/H25fnm9+/3QwPxu2 MpWRz56uZawpuUbM5p2CugzzHgOTyiVr9awZzL2HEbHnEOWMKrS0CrNEfvj88Ph1xadk+Sw/ c7TgzlXycdJ0oTGZyvgsLZVNtJ3D0XpTdW77eU5kGs5UFXrfbCNa/E2byvc6w3rHn+7wh8Ka x1Yb7TVVyReT/ABT5nFFoKmElBRPUNraJ37tKDcZjD56LZYBjvPVzz48EYiSw6DRC5envJNK 1SC5u1+D0O3v6RTy8uLk1zdpCzB7J+S9o/ApiamORzgpKnz/FdkHCCzJxu377NTdMISMthzR k234qwbw57HHvY6adIVIhbUQdfmLa7puhQg0+TrrUn7s+ry0hdzub8Xdbk9dh7bxBR+35ig1 nGMNgY/LepkMssv5+VOYVJgRIibUNunB7TOubRRig+0yjwfw53X8ISv8uQ1AAGtOFm6QTKyH 9S+AglpTYp+u43IWs9XURrQkiBuWrY8bofELRdQmsw/+XDrMmLDm8PzXw+OfeM0/s11wmDc0 eJSGf/cFIx6MBP+5C//6H2dfthw5kiP4K7J5mO22ndriEWQwxqwfPEhGBFO8kmQcyheaKjOq S9ZKKU1SdVfv1y/gzsMPOKNmH7JKAYB+OxyAwwFgtoUGUT9R1Fz4MbwKVNY5QLuKdI/eyU7r +Autf6hqaFCW7xVXSQ7Ec4m++kIsd3ve2RwyOEl73Pb4GpO8CeEUgp2lRtWzP7ftS9A4tD5k 9eBxO08oKJuKTDuAxmopv91C4qHwY5yNuXlJzUO+pKSSnIlVNG+ZWlwKYHwy+oK0ngTYnr8i oSxLQFSXcqQ6/rtPDnGtVYZg7jlrqwwJGtZQt3s4OFmthnYUsH2DXuDFkbrXEBR9dyxLxT0e Oi5exZhuViAAAbS6z1JqCEWBpy5TZ/KYmJUgfFcd9QYDaG4SrWniPPXsYJlBdXGNEGnrqOWk KFZ2MTWmmeiNujA5kC9ZY9QQQwJVpiDo4tpYndk4VIiwtadhZ/pDBMJEt11T0VHisEr4c7+k dE008XErC0+j9DDi//YfX3//5enrf6ilF0mg2SemBXWSYzzAr2G/oOlrR2F43FcNIaL+IGPp E3MmQ21JaEiYZ8uQhuYcY21FVofqBgVgljNrKdZFEc5QtTTYJ5bB6tusM8gB1ocNNW0cXSYg 43OxuHuoU6035hYHoNh+SqeHS9DBzVzDwpGAlhsdLHiF0dyBhWhFWngb0NdZ0Rb9iXrTKGpP 92Gfny1jybEgiVCPAmcCLTqYWG91PhVLCe11F8v6Av4cF/QsE3MoVm8LLwzrAKMf4w0Iiksq H6y7GkMtt222e1Aw/JP68MAN13CuFrVy2QAU002KzEWHABzURhd2mNe3KwpEoOh9XN+M2NNE UVCtblc0aOCvPFPjdIwo8YQXBL2awg4fYow/CY2hp8qSS6sKlEcNNM6lAQFFgYxFNVMqjpgE GSteL2iFz2h+x0eOhEy162pL8VkTWzBENFQFD53jD9Lk+wG1X5lWabc0KeOw7/MjnPSd1uGS 9KIGxMQaZOrhWQqyO9qBayJZmCFo7bEATUYtG19zW8qcwjhZitMWVDdEBVdAJj9BaLX9BCeT tdrPx6qjTgLENemnVB/NbriGsnwCus9BbZUq/CNEyK1a22GpXh6oub1Ms8T384Ubdt7vvr5+ /+Xp5frt7vsr2golFUj+tB82h/Lpx+Pb368fti861uxTHuSvnDcXwUdmUpibRW4yU5Y7dR4J EoqrGkTAoovWGJLvjx9ff1sYCYzwjbYS9VAliGRuRfdc0JlK1RI14UQ+erwuMXFFb2hTauEB 4qRsYw7gR5tNATm11kiaAgurTlz9ut5gPa9P7d3H2+PL+4/Xtw+87/t4/fr6fPf8+vjt7pfH 58eXr6iRv//+A/Hy2SMKxMvOqtfkc5IGBGdrswQFO2i6gYSzItjBHCKBaWNVU5v7+z6a4s3+ NBYVEVDnpjGryml5htPnsd5kVWRGSHXa6aB8a36IsEYHJgcd0hqQwqSRHwQLUPl53HR8eNqD MkJaDfMaiqRvioVvCvFNVibpRV14jz9+PD995Tvi7rfr8w9qRsqderwMX//3gnw0n2mgFzWM y5ErmTMPzNqEC4Y9wqkDFDBLx64mQkELjjVVHMpCFkuBQBLfiIPL1gQYLaDJalMzQLgphgr4 wHpF0bY9bJU2BK5g5T5PzbJBFaYvwxYmb5jdf4ZL8ysPyjyTNFdUJtVOMs1uSE/JMI2hMrfz hITE9FoKA7XVNlGhGE08VvBjEbDbINB9swawMpPqZITzLNkbBLMl23SX54DcYvogDIO6bbJk bwnRr0rXVjqBQvJ0a6qpI1FNLvIkpo1HNT8gFNMZ/O6T7R6Fy7gkw7tzisHWIqxsQqksksAs iaBDZ2yyh9Yv0FPO1hKzBTYs1qtZuESNmqWqSaiR7bJaOpHwV1+k8Gkve2tIYDjqNTj3QlCM 4BysG9LmhyYddUeWe7LShr/GKyENevLlqjgooxYBx6SddEa2cg17hZkX8g+xIvXffbYvYF2V VaXaAQbsCd8DikUq0PoCL8gTQXiVogmxVYJx0gDYdvs+cjxXyecgI1mz8X2X4nMS0baJCyNH gU5gx6DhsFZyw8gU+/ac1TTK2qXUiim6expx336xjUGFsaEsdqCJ6HNsqRHmceM7Po1sPzHX dQIaCawuy2XuzdfEOF0GrN+f5FUnIQoFkaSxdjkiIPZrj1wWMuGH/GikY/m9XPapZ3Wdpyo4 72rljjSuavLSpk4SRXSHn+iWpN5+XryAaiOrldDM9aGy2S/CvDrXjHoVk6VpikMWSJLeDOvL fPiDR4rPMNid7HQtUeoKPvDVqVxljdmSOySx0pukRG/7tspP5ARtgQcy7oEl8ZEJNv6pvG2W 0aQhXCJIWGf5tKTf/kkUhfXGTa7AelmvE5H9G9+mUUWjvEBfkFfAc07AXLpYUQtP9kvF0cim 3gEVtXwzjJOKEOBbyhnGYbjD6MbgZ6Wskh3kmGN8pfCWihfqEjj3MV8XmhkU1Oema9RffSv7 sHNId9T4dRmr6XXwd1+lBfo19igx0jb5ppYGoNnx1D7yBcVFxg/+b9xU3sj+ehJC2M+1A6HB xDLtQ6/mF9h+Vv0Z613/iQwhxwPyd03Kitk7Uyp9h460o0lRdjK4+7i+f2j+07zx990+pZgI Z+FNVfdwHGZjSPZBUjbK1BCyR8N8pBcNS/hIDc6YX/9x/bhrHr89vU5GGMnmxYBBSgwafsEG LhgGlD+pVq+mUh4NNpXqjsNrY5f/A/z2ZWj3t+s/n75ezdgKxX0mP2YLa8XQv60/p/ggRhV7 qZW0ZQ8gR/T4fmeXXOTNPsEPBBwm1ICltXSUPLBCnobFTk3rkcm+XcBORq1HAm1jOtoq4vZn K+qTu/E35k0OnEmJaAvxWhi/O8XkscVRF6O9bR6rQY8QqNnrNRz6Dwu/FjpfHdFEae+R8X1A W7w0sk4wQgw1cEbwV4WgGJOXVBPZeCDMC/hyz8i0Grv+XhZALWxgl2375qjo2eesSXPxTnMe qd0ez3LXnL8R8XK9fnu/+3i9++UKA4em6W/ognk3SAGutG8GCJqG+c0Bz5/FQ/E7cxsKOXY/ /zlMFQ8OPz/NaXb3mcwbxW8YzvrYGdB9LTNfZFqbWv9t+FQPYGPkY5bRFyxxWh96Oh9muVPu zlDD22eKQIXAUvb0HgD9kTWdCj3oZO0h4VLrwMof3+52T9dnzCLy/fvvL6Mh8S9A+teBAcjX BjtUTzO8rVXtFDEmobSYJzBBA23aRlxdBqtVn3m0wDRTFCnlRzDgfV/tJAdhoSqYxzNVX3Qp 4IUvsH4N1ZxyEzJOgdIDjljqYtt5Lvyf6UQKiTnlAma2urzUxPoQwIFard7fnZsyMJs4HcJ/ aplISkbLQK6jbC7c3WgnHbeSs4MGUZNQJZj7QvV9BbkI9lEuS5k819mU3fOiXoSm4oKxL1r1 pg05u3p/zH1IB9/XkQmCzolO9DMEju4OSMzbZ/F+cJC4xq1mHGAKcSbr5akWZQR/g9q6Rbmw oMVkToIRT8ySxpgHIMbIj+M5qiRefipPSPQfQ2pZLUtSxv3B6TAqiGVtXSjFcAiVA2jCLQci U8nQoftPEdMp1xTCvu5oPsbDzLQUD0IMDzCjj8pSaocY33wJJ+chiq0ev1WibLvjVp0GLrPr QCY/YuWLIWbquPNXKnhuD/HvVGRWnbRaGm32a6ZoH7xE7a34EJlCWTgSUPgIfF7C9eWpkdst U2RbNZa0hOJRjaixVmrASDe3iNqDGrpHaBfw4dfXl4+312fMpGlI+vjhroP/unLIW4RiDm/D CjghBqd2fRP0F0w/dTGakVzfn/7+csZ4J9gifj/eShfLA8deIhOPXV5/gQ48PSP6ai1mgUpI do/frhjenKPn0Xmn7rp5r2KWpLD9eFA53n/7RlRI05oWu2/WP72bo6dvmtr05duP16eXD3VC 0zIZIz4oW3eEL4WR5HSwT7tUzUKr1DbV//6vp4+vv91cYe15sHd0aawXai9CbnzMGvotZ8Pq LFHzCc4xbJ6+DifXXWXGCj6Kh9mHNK9JKxwcr11Rq5aoEdYX6NxEXc7wFD+5mSua1zWFgsKo MKbv3xQTCF0w5Mv03Zm/dFa0mxHET/0EE/BKp+Sla9gcr2lOOjp/xYNqiL5ThUpoMsbUTEm/ a9ajHA09krRD/tQZDczjizdyegcdtslOljcTk5LbWBxYBQEu56EYOMEwCATtiI1kjL8kHIh5 AB1ioqeEbZgqDc5ATicJeRL6dMwxQ9kWWGOXyYJLk+6VF3LityoYD7Cza4CKQlblxm8bOZUk mokwRAVfHjvVYoPIHWdVPIwPOX2WTTQF35s1rdkwesj0yHRK8LZJ6h41ygqE5lhLdYhunkNM GXKa9iVpSig6xfUcfvIJbc1TcXqO/OPx7V01fXUYvmPNnzHLL/oALL0GV19bIbLaCTjdLAw9 zEOgEcWOKBGxCN/vief2P7nWAnjgKR6RQvboMclQBcQ3XgrPNfrOh+QIf8LByV0PeX7ODj3D RGC8u/zx38YgbfN72HVaX7RAAbtO996k3LizctcpZo6kVwBtq6RKbAu9XD7+2h2UgpzepGNu Gm5eN5YEyG4/N1Xx8+758R1Oo9+efhBmUVwEu0yv+lOapLGNTyABMINe4w9DUfxOo6rHwCJK sYguKwwbZu0XkmzhNHnAx3lnMrbgSJZLZFRN+7Qq0q6hXpQhCfKWLSvve57YvHfVnmhYbxG7 MkchcwmYZ0xxt9RBHlIUDj5ijAvQwhMTDmc1M6EYk1jbU7JYzwGqoZ2zjG2blh3J9RZWlpBp H3/8kCIcc9Mip3r8ipkRtOVXoYXiMr5C1bYfJoQozPkdwHZHZ5mo2tk+xzgoDEbHtsxHun2K KcLIlnEzpXjfrFXCQyPHFosbEpRpx2msBO027vcX6i0db0CstUiEoD1hQKpGayuI7o16w3Br jvhEttfnX39COfaR+3FDUeYFizqwRRwElDcG704+tkEZRgDatmiX6CsV03x0VYd5SNAGLT8p H7Agl7RDsljXi+TiOCv3xJEqdLin93/8VL38FGO/bXYh/DKp4r1k2dwKB2wQrIq/uSsTionn poG+PYZyTcDQS1ZqW3sA4nLFEILnJutSmmJWYglkJbsDyQjvgsx8b3AFjkzjGLWdAysKxQ3I QjC8i1XmGJ8RllqYdT4BeQ375u4/xf89UJ6Ku+/i+bVlfYkPKJ50uyi1pOPWkvMQcDyBKG1F U3kJiEjHMussr7AAe19tP81DBoAh9KICG8KAKDBF6oXfyptz+F2I+1a5JSKyCHXc6flaRFA9 NS+1DdDLN2MjDFSpTI2sMVODkrYj7WczBTcTqu0fsewSResN7do60sC2pnyGlYfP/NUz15MK GFrQJ6enGLV0IT27VrYMvqBrLWs92vaMUaPUD1GY5H6NgZnKY57jD9rRZCDaWdL8MExbs/gl GmraFjlmVvvehY5r/UXjtEYpxyJdJshBJl0kSJrtcoCq8ga+vdC5Vke8rQtx0uA1+30XJydL vpSO8S2C1wUkgbhKvDlJt3rYtBfTWlieilSy643aA0C1gMLTOJ0Kxd7CScXrVmZpPyc5nAsy nDJH7ti2URLfCWisAZQnqALC3bGN9kyPqyrS918iMesY4fixrWDjJc54DSaPppA4n96/mrek LAm84NIntXzfIgFVu0RyLIoHlfFm2wID3iqM6sBKLYXtgOmyXaHNJgetLxfZXTluN77Xrhwl +3VawkC0mC8W84yY3g3jKm+DwA/6Yre3JE881H2Wkymm6qTdRI7H1FiYubdxHMW/WMA8Ko8c 6ANt1bR9ByRBIBnXR8T24K7XSuS6EcOr3zhkuOsiDv1AUrCS1g0jRVU6DSZGtBCQq7vGl26H o+KR2NJypWI1nkzCA1LY+vs22aX0HXF9qlmZUVfDsacen+I3LCpoBmt6z+UDJkJwpSBtFZRp XmCAVXl05uoBb8aM1ykKdgmjNeWGOhBs/Pii+LYMcNBn+2hzqNOWmquBKE1dx1nJ2oTWpWkQ tmvX0faEgGm+khIQNlx7LCYDwhCa/o/H97vs5f3j7XcMyfM+JjOZH/s9g3R99w3YwNMP/HNm Ah2qxXJb/z8KMxd0nrW+xUmAoUs8TzRaK7EZUDkTzgvzZhuB8I80I43o7iIH6BLb4VRwBVA8 Mnr5uD7fgWAJEvDb9fnxA7pDrK9TVVtNmUtFTPMfHyQDLYZ5g97GGAA7Vm1IiGkwI6b1mo9t Wcl6lpFtUdj5xCh4OGQ5+J34IaS65+vj+xVKAQ3u9SufV27o+/np2xX//Z+39w+u4+LLvJ+f Xn59vXt9uUOxiqsH0qGBGQIvmG1BDbSH4I7f+rcqEOQKJTEJJlvXkjRPAT4B1wK9Sr1Xo4tw CJZKLYgJWSvjLVUQU/5lEh4+TS2f6he3cs8xUnxWKZnOeSpFNGvvpn2K44lGBPh6XDs///L7 3399+kM1qfN+mLneTbE2Zx16CiwSxUUSrmgjitQ5Tbw3CfjVAk8DO90FSt2Rb1KJwtXlLyC4 9jHSdNUklpuesYRqt9tWjAy2MpIYiv30LXDK0HOpKW2+YI7W273WWj9iWRqHNlVioskzN7j4 C3WwIlmvLhez4azLskttwvl0EvRdk+3ylECgUOSR8XK5uLQ0AEKeoov0QxN+qDs/JOCfuHta aSLa2PUcsm11llEn7DQ1XeSuPXKjdpHnLo04JyDGqWyj9colelsnsefATPdK6EADW6ZnooOn sxzqdAJnWaEEYpwRMLSuT3WszeONk4bUM9N5DRQgnJqlnjIWefGFWmZdHIWxw6Vtvq+rj9+u b7adLXS114/rf999f4UDA44iIIdz5fH5/fUOs5s9vcEh8+P69enxeQwr/ssrNPbH49vj9+uH Yjkcm7DiF6/EKOHuIfdG0sWet46oUTp0YRA6lLvqSPE5CQOq0GMBQ7H2TIRgA+MAYRjs0UJp 8DweI1vJwNewLOH5M2XnP/FKQ/5GVCBDtKODVzvUJxIn/wUEsH/8193H44/rf93FyU8gYP7V HNxWOT3jQyOgSzGoAU15R0zf7skSY1rr5n2J0ezL6Gj4nCCv9nvtlSSH8wxY/HbeMBnwAelG kfRdm4MWM7MOo64WuYsFwtYUkUuLmDEQTVorPM+28D/yA31iEco9qlo5zJlANbXU6tFArnVU G7iz8CGXlFaEKzYKAeK3xFoWMDE7l/3WF0TGaCFuJXC2AduWF8/8ept6tq/GVeafe2BJF75F tCYdaj1/HQCBfnMhr3tGtDkFDN2IjJLYgbmBZy2Jo1ce+dnaIkoJAhZjZ6zlZvFaYcIDAM/0 Fr0+x5CkvqdTNClPcZTm7KEv2r8FmPR9Nt0NRNzHaHIColSvgVAoksKf02zNYBcGkfZvRCVN yr2buu4B3S71G1C9uxv7fAF6s9JGAwHm4wTBVU8wu/bKitORzEgnuGuNtqxKXxsYka19MJZM Exfy2z0OTKFyT73CBCWGM3c49+l3ZBPFpO+YH2t9UnoEwpS5oAHqIQfhz0z2yh2e/NUS3iM4 VcGarv6sn0rHXXuI9b0pgOr16ojok3MMvIdG8q8GjYX4NMY3shJenV+5cELr0UmVnBMDS+my Sme4oH/A0aN6tYpTImftgRvBFpbcQ7NdxFJTOxgn6hPJblvaeDaICBff3bgmN9sJX3erLUEh ylpmJ9onFtv5eDJad9jo/VbGTeBHjjbMWa0vuKzMlHdJI5C5qjogRKF6oc1ZQau/Yjw7MoWg wD0UgR9HwHA8fTdMGJ5XVVw/4t0tN3m4NtoxLCvbt39zQwsV7ktOEa5sFIU5MLXOjwDSa+n9 JviUTEBGfObrHGPu28ZjoACGoU/f55xNB7zS5qxYuzpxEvub4A+dp2LnNuuVBi7b2tcH/5ys 3c3FaL/9iYJYIoVx8uoEkWPJWcvx4nbLuvcOei8PfZOw2IQe6r49m+C0MHkMgFl+1Ja2LPRp WsZ0TioiJd7boTg5g2oEGbY4BJ7SZlthKjLM/aiieHYlFaRe3vKK6mJKjBtLXuz/evr4Dfrw 8lO72929PH48/fN69/QC2t6vj1+vsmGLF8IONB8ZccQhwcFxemIa6HPVZJ+1NgIXid3QU5eQ 6A/IgkbtKk2b5Za7BY7d0Y8kCzJ88xiFUbnxios+03wmEYYZuuRtj7BaPyIQiG7aVFRevCdE l23iLnLQCDicGvdtPX80wHbHVsunISCoVxFFjEhZrhhghBwyYGLVpXOADpqgoe5hbI4719+s 7v6ye3q7nuHfX039G2TfFJ/8KgUPsL6i192Eh4GQmNEE1mKtzPCqfSD37mJTJ+EQo1p0VXsY 3M5VZ1AWA8c4FhXM9razxCgYnnnPLS4yNfbDsPhohxetVwICzN+hvNNGrBNIN7UDUDyqV2Gx 7OM+wqpi4/zxhw2uuruMZWewW+ztgU89x/EcqkyO6JVHhjqynu+Hkqf3j7enX37HS51WPE9h Ul470zF4G8j+boHPHY6GqdQQ6ARJIdqGbWlE2iTyvI4hhbaw/dudZyIGHwEdysou+zyFX1LW BeKLbk1bfyeCUxSloRM65Ndo0Y0PWY2Blzar9Zr2uaCoo/WGunlVqxV6slnQiOz3ebVlubdY 7c3gV59jFhHxpPDFY5feY5JzE9kWbWwPCiVjtQfPFIXupzYSnbIOZM60BzVh7UNv6Zue8RHd n1y9Y0t4LA/FY25ohtSIU1omVdP7cSUdw2muWKb9OHADcgJOVUNL3t1Dfai0YOFzbSxhdZeS L8slon0qn6hp5/qyOV+mzFmMvqCxLL3lWVyp8SCUL7pU9aCYWZK4tu7IjEdyEQX7oodRm1CK 8gY/I9d1rQ5Xuf4AeJTvkI3JcnOZhYFWcH/Zby1hGKX2fD4ij6CVK5muuTUluKAqZVBZZ9mb gCCPGAAr8mfuyiFHcssMH0GQVSyFAtKX2ygi7/akj0WEd3l9b1cr5Yd4r3uEY5rnUzNwPPfc Al4CxBhLXd5yaDadf8VlpkRd3FelstUExHRfm+2tUByp6j6ADlyovjdAq8YDg99wHGUVHVmG ozG8UtqMmVvsdElM20Q40up9p0wKPqiVG8vIqR+e3WpiE1258tkpO1KmI5nmkOatEtlKAPpO uUeeob1LxTqY8D5R0oqCDVHjDLgcN3qEiuypZBezpiGdsxWaNpY6mJaqDUym5Bn0aB4RX/Dp PqVUJMpal4pLUqOi7phnlrgr0ncYumK5TyAxqxfgqadJuQKytIsEAfyPurcYkb5eh2DVjQFu 7x8O7HxPDkT6BeUhaQr4776s0XZbwiGH0Qn71DaO+6ra5ymJOhzZOc1IVBZ5yt2njEJnKnmw 0ImN8nFDpwmdzqHPmmxP70eAn2hNOrvYPrEeaNnKWjvtPfOpuHkyFqw5pZbkzzIZ0LCyon1A ZDoQf9PbpY1C8p8ibNOCtmPIhA8NTbMDAbe82eySdXotBFGKEW6VXJ2ezDtPl73CpPH3+EQb DZjWJNtqHU1VVrfnrbw5JOUpS25LPNU9XRCIatUNWWhIV5iW+6xUM7McQDSE6SULfkjxdfwu u3E61mnZMvjLwquFCfdW50DqyzHMxy26JrnR1UFNUsU+yo0scv2NmmEPIV1F2VubyA03lv41 wA5tt4AyGQYApV3KJKqWFe3RGstzJEpTI7TyiKpy1uxy+qZVpstyNYpfG288x6qWTl+prgtZ u3EsN89Z625uiLqocyrFFfHG3VCvhNI6i/U7GPh447qkdImolWdYCKYhivEJ9sUWe3kk6zhH U+rsCm4Es73lkD4+3tgz7UNZ1co1L94jXvJ9IdvvZ5gquEsFdenh2KkGWQ651cLbFKfbHOmc fSlJ861EI9zz5QYODvvskoF0xGjOMNDkOXTHRrNLEuoYgLNKviJAfarBkGgNBQMhqcGczMO1 hNLJdmtxxRRmtTFRqwxU0y9zCJrHy0xMq2SbR1TWbRmdBnooqy+OF/MzAedvmG98y6/tmnRv tlJgh8R7FzUkB6c5ZHg/ax17TpPVn1eOu1kkiJyQ2tEcDfsUgwRmhVH7pSZTw9aHBy3UHAIk 7aQ9i+CK4n1Rlt3Bz4XXz+2OFmtYgjfXB/q1BhoyrLjBeKETjGjxUHI7hIAc5fK44E4sWlxI AEdrAabkflhXPGDwOALzd4MFwdpI+DRYuStniSBaRZFrqTrOQM1lenMHlczyTQIa7tAmib3V kR95ngns4sh1CdpVRADDtd4UAd5YmrLLLmmif5LFdQ67gv5CPOm4nNmDWn2OPjid67hurCEu nQoYRHca6Dp7vTlCNrc0Z7ZbE18JRGebu0lKV5tS8qTzLNdLxLCXHWYTMNehJF1Fjm9Hfx5r o8SqwaSttGWQpzQgSE9Sl6WTXoN0oPldFKkOLaSwU7LYNr2jWVvr+3AE7YGNeA3+l2JIuXqN UdeUpNlqVPh7CliVWvwDkIZ7QNnR6PnB/1LcrzmDO7y+f/z0/vTtendst5OLLlJdr9+GwMGI GUOHs2+PPzDJDvH+6Jyr4aE57vxUsMsd3iI+X9/f77Zvr4/ffnl8+Sa95xTP5nigYqURH693 +OJHlIAImTEPlwc3i5eadyPngnSVaeJ27D7NtyTqcG4zyQp7Ki5o4JancXf8lHXtsbe8+xS3 oVpkX0m8MGOiZm1SqvsZANTXJ+XEhJ99rb2AHh6Y/fj9w+oWPgZwnnkgAvo8JTPwCORu1xdp kSuBDwQG4+cruQIEuOXhsO+V+GICU7CuyS4DZor99IyTPHlpqO+PxGd4/UynBxUEn6oHoh3p iQQKn05psGwBRcQH9+kDf/Mjj9oIA6GBFpUkgjoIIvqpvEa0Ibo3k3T3W7oJn+EwCmiFTKFZ 36Tx3PAGTTIkrWjCiL5umyjz+3vL8/uJRJdlaQq+yCz8ciLsYhauXDoYhUwUrdwbUyFW6I2+ FZHv+bdp/Bs0wO7WfkCL0jNRTHOamaBuXI/2J5toyvTcWQzNEw0mMcFrpBvVEcYKYuKqPNll 7WEpkN5cYled2ZnRksRMdSxvrqgKGAvtMzUvgsLru+oYHwByg/Kcrxz/xoa4dDcbhbJXb3ml PhOxGuWtZSItRYTJIiUxE3/2dasGMhuBPcttEesmku0D3a+ZAk188P/aEjdromsfSlajKEaJ piYVCECqRj2RxA+1Gn5Haku2S7dVdU93l2f6M3yoDbIUHVWUO3oTN7WO6miKOoPF7Cm1hi8/ MrPMTLSrYhST1YRCM/pU8L+XR5RuqQgJtNBGkXULG7lAhPrkZk1q+RwfP7Ca6QOJg6i6oqhw PdK+huUdstYI+0C5Wx4602WX3CwT1+6WjGYmhi92XadmifnhqQWtnVFPvgV+yIWhj/m0uG05 DXQ6lJEXpJ0WiJTFPsJ6BhodmUN7pvAT+kvSsjah42rbSDM6wfc7j27JviGzHyr4XnUQn3HH DE79gsyVNxGhig47uyMa1WZJesbUvw1ZfFdY5La5bO6GvFT7mTVNJjsvTxh8UJvncg6duV01 i9Oq2dpQGDmFwmGuL1tfzlkCP5a78+WQlocjtWwnkmS7oWaXFWlcUV3pjs0WQ9HtLgSStYHj umR7UVa3hVyeiC61xbVDGv78HmYfxFpa9pkI60tzY653bcZCy2Uv33A8+aslT7EgQG7Zxk1K Pr8aTuislRNscxhL1q78hF6G6sxwwHVFmqO4YDBohWxbMFcO/DPoPv7F6bfHrpMndFDl4ra+ b3QoSKjRZrPuD5wrma0pYtdfR35fnxtRrL33BQjfZovgkCjVy34B55L/Nk3p8OgSTQKrU9vl EvaUAc+yFnDOWrzN7bedGgF3HOmctRy3MPFwvPCY3l1K+5hNKiTs7nKgtLbn/tJ92pgN4Rk7 QC+x+BZwmoeU6W4pGkVcuA6tagh8k+6POevQa4NP9iJpd/wTU97VbRh4bjSTEoN8qT3n0tcp 5VEzFCPkcKUUkoBPto48aimrhiFleYHWRnvD6ngXOKEPS7s4LgwFkEUBKQYN+HMxLGOiCsAt L1C+hpuqY80DBh6jF3rCNk7g9VVp02hmstA3yTTOc8l9iiFxMM2Rss+tF27snYgL5juOsfMH sCoODiUmKfAFDEMLf22ZwZWS5uSFsGrEQjWsUhwdBsvotYluimw1PpKZ+siBNqGNI0Eqpazc iNrJftkjhAvglQb3kiFglU7vugbE0yG+Y0BWRh92Aa0ZD0jFnCOsyY9v33j+hOzn6k6PbKB2 gQhbqlHwn30WOStPB8J/VcFdgOMu8mL1UR+H16wRljAVGmdC2VWgebYloMo7EQEaXKo1fXko uvXwzpQy8Ytvm7gnamH1lixO2LRa+rg4chqiKhTDhlGaiEdYX7ZBEC181OeSj+cETIuj69y7 BGZXREPQw8EyTy2FORoZYW8WJtzfHt8ev+IFgxHvseuUW5cTNbrHMrts4PDoHpTDWTxh52By CPOEBzs7dpXuHjwEyn7DiDDGMxqhFfcpa/IHRd4dEJEXOCQQJJC6SXk6BCmaPkEnot4qq2FE uWEQOKw/MQDZzGUy/Q41H+rElIkA1FayY6bSGOVZvIRILzK/lTFlwzP1tX9bUdjmWHZZkU4k ZLvTS5eCTkaGpZDIWFunMJonNTejMuZn7f5bRd4ov+m8KLrQJee1+iRDxhV0QA1BgUk45tei Q3yil5/wQ6Dmi47fhJmResT32NlcCReuIawTOhFMM+RqFOrLTwkolan3+FNLhjUQyDbbZSfq K4EYi11ayOJR2EIVcVxeqB0jEFQVOp0bZi06U5Ddn9B2jCqbDFhY5du0SRg5aMM58qlj+6P1 Flch1clUIgyXSu6CEWFdFMNFdt0a6TVJgtvDCcecUQkefbYGIA4WJB4o5oJsas/4AGDzCvY9 o8G7FhZNfWtYOVVWYgC6W6QxOgTCKuTZYmM4LmhHzXFV1HoWsCltgHKiaL0q4q7RM+QOqFLE okrE3eIojvGU8VqE2oc4Z8pDzPjhC1q/5OQR1YUJn5VcroqDuUeB0oCHMlZ9qkaIEghqgPV7 WX/S0rr3mByXkj3GGyBx1BNQcdZSHKjs963FJ7j6UtFu4Bi1XZMqhFN3a1OLh1nA+E9GpNVZ TBGhhaizliNUhSxfYkt1rUWcH8KY27/I6iIDgbpMcsVhAKEJ/uOmDw3Bc/8lIjDDrKhxDMZD FhdytrqEe6Owq+6UkB4cLYfdEABg9xrozLr4kFR7vVlow6h2O61VW6NKommHM4jsZaImvpmA yJBRaLZFxJ8J+e5YKn94FmmA96kyzjPilDEaPGw3ohEx8APS/TPpctV/va7xgaglpExVPpDx D4ozO0lNreNo7Yd/6Bm6QUDW7ylgHrQRnBH3gFH254mO2I3ZQ/Rn5Bh1gcMxX5cXhEqNyObo Sast5lbYDfv4kKL5GaedaEQXw7/atlZqMh48fpK1RoAMDjUA/CqKm2EVZiMh4fjJStvrXZmw PJ4q2oKGVKVsL0YAWenNyuKGsk8g5tRhit+mujyYnWw73/9Seys7RrcGweaKMZwIURnIGvmD cqk7QsY0LWN2UkNlnNeWmMDmiKlh66Ni35BxGFxQpIc0dD8045geUbKMh4Fm+MxUoNPtlbdE COXX8DDglQoWua002AFIFe8jAArPbuGn/Pvzx9OP5+sfGOMU2sUzG1GNA0FqK4wGUGSep6Uc rHUo1NjNMxz+SxsFB4q8i1e+Q3vOjDR1zDbBinomolL8YTaszkoUgEyE4qCOwCRdpC/yS1zn ibxYFodQ/n7IJYo2AbVgzdOAj3a+r7ZzDnIsd7J+YA5JLQRtHd9BIQD/DaPQkrloleHkYZh9 KujEhA19fSbtsZs5tkjWQWh8w6F9u4oi6gXpQIKv/okv+6K2fZQJ05AMaWWPCQEpOr1UjKJM msmRDfK7Vk//ZABDHzYWbzNOxZ/RwUo/WkrngYw3gTbzWRv6jl4jvm0KqddFiFQO+wFQ87dC fJJ5THbLrLexKrTO3Ojf7x/X73e/YHLSIb3bXzCe8fO/767ff7l+Q6/dnweqn15ffsJIyH9V 2UOMXFQ938V2arN9ySP8qceahmxzRWDQsFQQRI1kyx5A0M7I5JpaYbLNCXHp3nOMhZIW6cm2 +Cg2x1mjiHeXlZ94FlPL1/dpUctxszmf5x5vxmKN2VJsR05y0VYDAPT47Ahu7n07A26zgg4y gsjpDZXwqf4DTsUX0DMB9bNgOY+D+7Zl0Q1JuKyVdwz90k6FsTCHwNtTPdLqNOqImR7HUV6a wvOt7/DVlZxUNf7Dcxw4bLfqEO4G3XK0LNt4rzaI3ZF0lUGUubg5aEiPYq5qfAusR6oiSPCQ uEFiTd4hySBTu3xFkOLhRwFGpIudhefzLQotPpoEJxXog6zcHXiQ5lnWEfdBrZwPfnL25+Dn J0zPIq8OLAJlILINtep9KE7SroZyXr/+g3qOAMjeDaKoNyRMCwkayMjxN6uZdCVd9BgjiQ6I ft9URzl8KMCFQGfSo8SyO8Jn6nUAlgR/0VUoCLGIjCaNTWGtv/Y8Ao7X6YoLwYQpLCnrBnwR 157fOtQ10kiCgUdVQ82EubgBmTJqIugK2U1oBItreKrEKk7ziuZdIwl18BhEoCs2zcMpS8+L ZPlDebHH2J1qBFXJ5kY9VcjKsipzdk/zkIksTVgDx4tlLQ9USVqCMn2rSpHz9maVGQzpLZpP ePXS3CTL03PWbo8N7QU+TfqxbLI2vT2uXbY3K9XXJ2p0zFxCcbta53JmEAURBdTiSj8f4Wzd NnT4HRQolPelAwAj1naY06/PswI0hMCdYqNXO00CE6mGlailYylZ83kItyJZQnCzW59D8MJ4 uHxLa82cMxzKnxc4s8Ypkqx+f/zxA6RKXptxC8q/wywwWuxf0R9+YyE3XICLxJLjTmitIjqS nSA5s5o6wDlSvY4Usl6H/3NkzwB5EIiYrQLdEFN0yM+JBspUl2sO48EsTqSZmw/0Ngrb9UUr qGUFCxIPVl21PRpFiosyW4ltVhnFdcy7yN4g47KIZSuF8Hq7REGgwSZpUpu6fjf0d1Sq7atE HNNwZP40YNEbQFtHaid3azeKqHNBjHQXrc1xIZ0aR5Tvunofzlm5rUp9Fs+tG8arSO7ZYssn jYxDr3/8eHz5Zu6M4TGXPooCOiSk1DZHYklWK9bkGVYrdacs7V/HKJPDyRBrwrcEbTG+PkwD VM2aOWDQhU2n7+os9iLX0QVybXwEZ9klN8atyb5UcvQ0wRm0nE6CG3BvNg2oa25iU9b+ZuWb e7WO1r51cBAbhHql0+GiFiXEE/v0jQ6ONyiikNYAZ4qNS2m8Ai+cG/XlXUR+QAA3GyXZIzEx U8o0Y8IMro7GIVuztl100ZdMATJGZTLP2r6jMahhhqEf3NBYlFkqULIJWvhDJrEv0mZpnKPC eAO57nUw7n6zy3qP9/sm3TPafiA6CIrFUbogPbvy371gxHwk3Z/+9TQorcXj+4c2wEArFDf+ 2tAStGsmSlpvRRrxZBL3XCiNGRC6xWTGtHs6oyPRdLlL7fPjP696b4RSjaE36euqiaS13dJN FNhZh7b2qTT0rlNoyORraimhMmgzwvNpROQE2nDO35AhjFUK11Kdb6kOEH3cxPYqb48CrZjJ FGs5zYKKsLQ3Sp2VrU1R6q6XFtawgCSpnIfzZCea0QosTzpEaQcc2x7rOpfusGSonr9WwfHw ixIOA5wg3lTQWRKDwtnBxpHqEVx4+kS6CWsHr32iyWgU2WOHQWpwQmmAh+JBbeiizSpQjqMR F589x6VuD0YCnLXQMQudptkoUswzOfYKCRmHciDI0z2I6iefKr/dUmrLOAqAnRsrok2OQKOk 7WdvTWfdmhoKwoNP95KLFWQvpwnhTxEWShcEc3PHtwv69CMcxMHdMQXdnB33lJQ/lgmynLt2 VsSMDRjPglETRw6Y8SUEyDOSMjh2cHzSYH7XXORw+iM9f/yj5hwfUcTbaI0CBS1vTX1rCWE1 18oXgdmcvPPDwKVKxAFZBWrgd40kSTt+NSBoQ/W+TOnwho4fr9LQbHekgYW6cgP6WFdoyGB5 MoUXkEOIqDV5iyhRBJGcgHPacMXWX61NOJdDnQ2xEPkSxntib7Mi2NXo0kWsqi5w5NU2VtV0 wN4Cql/8wuDYbmvaVjmSHePWdZzl3bxNNptNQN04akyf/+xPWaKDhtsAYasR/rwinwzhOy5S jbNt1h33x+YoXXXoKJ/AJeuVq5ynCoZeazNJ4TqW8A8qDbVcVIqQahsiNhaE79KtLlyX3I0S xcZbOVSp3friWhArO8K1IELPgljbiloHZJdaf03t1Rkfr7XczhPqkvU7VqJ2ALI+dU06Ut5H GB+dLKNIMP5ys6debk1E+Gy/1bI7Te3b0jGUZwJ0uSfGpLvUZLe4nxe2d3HlJW3oWWKFThRu eGP5JhgvrbUlVxuJxCs4RkaKHYmy4B7Gcmt2Ew1VTrCjEZG321OYwF8HrYkY38QqZ/D0VRsf 1NwLI2afB25k8bmfKDynLcxC9yDyMRJMLP/hvryk2nDIDqFLHuvTCG4LlhJNAHidXgg4GllV hjtPRuAQuxCvX4d9oH+g2QtH+Kd4RQmnIxrks8b1PKKqPCtTJff1hOCHHckJBGqtPwmkqDZU lV0MogHBrhDhyTm/FYTnWdqy8la0sqzQhEtTKijITY7CUuiESycHJ3E31q9D6kZRptiQcwoY 311bou9IRKHGPCgK39a6MFxcOJwiICaRIzZrEgGtpua9iGvf8Yh5L/IL5rHdMWKHdHEYrIgV UYSEFJEXa59cJMX6xgopLCmTJIKlScyLiFrooBvTzbH4ckkEt5pDis0Smt4sxYYyB0nowPNJ KYyjSPdHlYLkF8IDe6nBSLHyiNVUdrGwqWVtVzUEPu5ge5GjjKj1jXkHmnVkkaInmpqHqV1s /S4KNtLCrlWfyolOD0YkS4vejaZuMSzqzvacbDqB+ni3s8S1mqjKtj42mAb2FmHjB94ibwEK jHNM9Slr6jZYkQnsJpI2DyMQFKid44GaTsji/FBZR9ZTaR3NQRyW+b0fUSfNwO3JHglebktb OhN5znpRfhAk1AEoeGdEt8tfrVaOpV1RaAmvOC29SwpH1PJRAurvyoFzdqHpQBL44ZrQhY5x snEoWQYRnkM2/JLUqest774vebgstWMECYso1x460lAo4anjCMD+HyQ4pqgn11pTHi9SOL+X lMAUBOWVQ5xkgPBcCyJEAyjRkKKNV+tiAUMfCgK79TdLDW27rl0HpIAEWkkYLvMuUANcL0qi m3p8u468pXOWQe8jWsfMSuY5VPhQmeBCiecl8z26zC4mw5pM6EMRU7JRV9SuQ441xywdwZwg IotcOXQbAXNDeQSSgLwHGglOGcP3JLTKAcgwCgnd6tS5HmV1OHWR5xPwc+Sv1z6hRiIicgnN GxEbK8KzIYhtw+HkJhUYZCHoF7UwSECYA2/uiINdoMKS7lvorQ87S9WASw908qWJit+wyCSL vvbTVsFnPMYFjEnW3TuuSzFYLnYx9a2VAKEfTpe1lnCWI1FapM0+LTFOxfDwEa0Y7KEv2r85 OrGmH4/gamfCzk3GY5H1XaNksh/xY+b3fXXCqOg1hrhKqV7IhDuWNcDhmcXtmfoE45SIgHV/ +pPhmi/Pq9gioIxfqW0yO3mzc0iASTZ6S6YNmW7uCV2R1mzFal0fR1LaFM0dkQmKAZ+kp12T fpYWm7GKUJjLqMUxuOcNQaw/rs936LT/XYlsMjVEpDLgPYlzVlBBGUGK6ut7vLosamr1iyLa Ku6TrqX6NO9LIPVXzuVGg5CEHr3hunixLKNv8WGxMHqIJAcQ6Zp3aUrH180Uq2y3MMRtm22V YAjtVvmBYQTkPJ/8qzg7VPzGmPh6xGqlJFm18M2IVqHiae6ULZT+VCVSOPeMtVzhbeOCycXO t7CAMBYLf7736+8vXz+eXl/MfC3Dp8Uu0R4wcYjmOoYw6d5chrb+Wj6oR5hqUasLPvl1EFjM 1fwz1nnR2rE/EOFEPD4jhp/Q3m4bNIc8li3EiOBBdR013zSHJ5tg7RZnKgA+L1C7mZ5hWrTd 3RSOulcepyLCdIKbofawtXwyVuucVDYmrD5XhHP2BN7YZ0DgKSWNTyC//pefGYxA2aEQyxku C/RowyOGFuhHdEjVPyF9okTXEp4f0XvWpeequW/7fUtZlPkcxK5/uWjzOwCpThS1F3qUOoDI QxaCzMzHZi4Q1Lu+Zm0WK81HKBRPu6hiWYL5fj6y5l5+nzdQ5HU8eFNLgFZ1r57PFWwQOUoq SR8fuvOfJURuTr+CmluPAZ+4wPhn6NCL0Er2uQ09+tIf0Z9Y+aWPiyqhE3gBxeTlqnwXRXUR WRLuzXj7kuX4kHRsEftb+E6Y+/6yXtsu72aCgJKeZ3QU0uVu6MwIE0G0olS2AR1tHKq50Ya8 4p6wssF+BkZGSV3oW9JgjGiLiwhHp+XOc7eFbStj6FS1GaZXzhR6VLlBnKDqJuOFTh6xMnD0 vlDa18RBF0T20W/uI/INGMeVQRe6kVpNm8bE+dxmq3WoB97iiCJwXAJE9Kq9f4hgaXo6tZo/ km0vgXPjTG67oiZD5XOc9qyCA7VXFQjrsp4Vvh+A1NrGTD+5TWd0AY3WFuPkUGReUM/W+boY 38aNkn/dhq4TKLKBcNqxmDYFcm3nSIIgCi0NkDyCzM883clTI4hWpL/E2Gvuo2+Mr+GcL1UX EdAoNCSlwZ3eWvfgbU8UBlDqLAUccF8yR+robWeu8hHDjonqfg2I0FndWK7n3PXWvkEjr6vC D8y9PUems09N7AfRZmFJfC4u1gUxPitSl3gVH0q2Z5RCz8VQ/emHBKTGm4t4Hh2qlg9OEbiO TQJDpGssWP4ugjLwTkjjEADoauHMRbOUe7H4AIwEsnV0hpkC+fRsQ2a11aEAqX/tRqYyMOJA grWzlrmABaK2Q0GJWtoD19wZlZ/jZOOvtAUkx2OxKXRjydMNlVzyHHucO90RDZopRCbHU5V3 iuvGTIDRt44ivF97LFJLRWj34WafiW6xVpCi9oLdGCjUOyOZa6ko3ZVbwiaBb/EklYi4/von iCx3SRKRzb95JqH0Twkr1uONaoQStliNrpKpGNm5QcF4sv+fhnEpzI6VgR8E5NSo8oYU/J6r R3bMKfDJVmRtvvHVxyEKMvTWLhUnYyZCEWLt0gVwHMX1ZJJo7VlmznwTZyGyqL4aEekQJNGI U4YaJUSF65BCSYoIiQsi22dRuNrQ/eZI0vlCpREKBY2iFypHrcmVaigbOiqyNpZrTDdmQOhH Fr8NncyjznKJaLAhGEHvFYp1RCljKk2k3q/KyNoFMfJmc+vAlr9PJooiS7o8lcjy4FIm+rze WJRbiQp0vZu813wkapIYKp6E2x2/pK5D8pT6FEVOaEdFdtSGRsmPBGewodhJKEOLnHGtV9SM dHJRaVqaO7dBEa3DtaXsQfm7Mfhtvg8s+dolIijKCS2HMCAjb7V8XqHnhxv6lhU+Kma3igg9 P3TsRcAeWd5moz5HjaX5ekPDuUut198w00QW8cdUqSTc9HiZqPlkuXOeKXQpWsEoMnOjGyEa jI0kXejnmfqKclvvOIynU7assnhMLkR7inH8KYtTMiVdikHp4jTmTw21xA0IXsoKyT+Dqknk Ee+EjnmbRkhnJWlYVrYHllRnnUxpIdE6BQHidt6RsvFItk2aEw+t2KZ5Gk+BE4vrt6fHUQn4 +PcPOfvDMDiswJuFuQUKVqSx67uTjQCDdncYj91K0TB8P25BtkljQ40BRGx4/rxTHrgpdIbR ZWkovr6+EfmYT1mSVr0WGHoYn4o/GMnJ0U9O2/nUVupX6uH1n56+XV9X+dPL73/cvf5A5exd b8BplUsbeIbpGrrAsORkVdQEhVDSiqzEI4SVezl2OS93dy5FTOWp9VQrefOTp78/fTw+33Un qfVTi3AgCtjqljEqU2n6OC27QPtZ3WF2cDeUUclDyfDigDdb8VXlWB6fFLYlXsj3edW2vZZn USE/5qk5RlNniT7Ju8a4BBXrMs6kZScvrccfH78rq2u+tBXmTZgWaoTEZHVn4O0rffq7cxiR 1fz8+PL4/Pp3bDtRofg4O3UncmgE+pBesmMxBK66TVc1mSUEliArLnTyvmFzdb5LpDei+vTz b//+5e3p22LX4ovFk29Ee0Fk8UkbKSwmYYHedtFqCd8ytnZ92kAmUYQrS4+/zUsPb7OYCCJp 7Cl2WrsWwzKit8dkn3Z2Oyan8WJvuAmvrTdnSFjncHzQhzDfeQW0hFZM+dcdPdoCZ7nqwOQj 7XLzS3QKsTcq2TZZsrcTtEWGQUwW5imrj34fZxWp1/Oza2JUsgAlTrVstXYsWs5E4NKdFwTA CDP+1wJNl7JgHVryholqYK2tnZDOFDwWsgOt1iJjcQph8L5BENG9hdNkIMpaNvinUAIZH/Lt cedpguIMJ85ADi/SopJ97WZMUohTOtuT5RXca8z2YbtXjsRZ6hj6YB67MdulfRxnlDQ3HNmT WEV9bY/DNhzr/A3oEp+1xBIXWBFhzDpJMIwe/KOmSGMG6jiQhCi9LREKflfEP6Mn1B0UOwbL Vd1Fi5a7SmFKQGu7ubRnq2L39HY9Y1STv2Rpmt65/mb1VytL3WVNmujnoio7ytHIBOjx5evT 8/Pj279tUgHrOsZ9KoQvYMOjdAnau8ffP15/er8+X79+XL/d/fLvu//FACIAZsn/S5cKs2aQ /4Rz3+/fnl5BvP36itGT/uvux9vr1+v7++vbOxT17e770x9anwc54sRvwezSR8LWK9+QPgG8 ieQn4gM4ZeHKDfRdJeCeYy77oq39FWkgGLZF6/uy0/kIDXz1PdcMz32PMuQO7chPvuewLPb8 rV7oMWFwcBs9BV12vSbqQrhPefEMe7321m1RX8wPMTkIiBG7HrDkYvtzM8mnsknaiVBfHsD7 QxENbypZIZ+VD7kIU5fAt9fWbgq8Tyoh65XtUJgoQmdJUkKKaEUfTZM05trnALBBqM8nAEMD eN86rvzGb1iceRRCG0MDwUU411j9AkxMObe4r0mnmXEX1oGSaVoCB8S2AcTaIe9ZR+3Ai9QY UCN8syEfeUjokP6MvLQf1/rFFy+4pTWFS/VRWcnm6uIDtqZMc5K0vlKCHGoLVqrw+rJYjUdd MUt4+WmbtLbXxiwLMEntryw7wbd4VM0UgcWSPVJs/GizpEax+yhyl7Zbd2gjT780VwZ1GkBp UJ++A/f55/X79eXjDrMbEKN7rJNw5fjk9ZlMMXAJpUqz+Pks+1mQfH0FGmB/eGk9tsDgc+vA O7Ry8csliFSjSXP38fsLnMNGx1B+waeQrv7wdcwpp30qBIKn969XOLFfrq+YcOT6/IMqepqM tb+wEYvAU96pDwc+Ze1pMXV9nSX6Hc4oudhbJZr1+P369gjfvMABI2WC1VX9LAjo+5+haQWM 1hIv5wT05dBMECxp1kiwvlWFxTt5IvBvtcG33LEKgurkwUq2Thuig405RwiPllrGCW5UvF4t lxCEtwmWqwAC2m1sJAhDi6v0XML6JsGtNmyWCdZesMQpgWBNBt+d0KEptiJ0TRyzWNiNQY2i xX1RnTa3pmVza1BdPyLzRw+HbxuGHnHUF92mcCwvwiUKf0m8Qgp38WACilpzvzMpOoe8hpzx rmuI3QA+OWr2Iwnh20UfxLvylebAJhvHd+rYJ2a5rKrScTlyqR9BUeW0uisImk/BqlwaiTa4 Dxm7RbAkKADBKo33S8c8kARbRj/bHCiKjNV0XChBkHZReq9x4jEBHHla8OMiBxj1mGwUYILI cqE/yi9rf5E1JOfNevGEAYLIWfenuCCbrrRPWAeeH99/sx95LEHXiKXpQD9Piy/6RBCuQrI5 auVCIKkzU2wYJQ4dp915HUvuSSd68fv7x+v3p/97RSMyF1MMqwSnx/xGtfoSTMaCiu/y9Ni2 y8WJLPJkWcVAynfiZgVr14rdRNHaguSWT9uXHGn5sug8/QWXhrW9LtDJLKZrlcwL6dNBI3Mt LFQm+9y5Du05LRFdYs/xIrrrlzhQwmCouJWjRsJQWnjJ4dOA5n4m4ZqKBayQxatVG6mxSxU8 yt6WwA3mCrJc+MiEuxiOoNtDzMnoA9Eguz39Q+tul5euHJs3s1IryMd/Ym1GUdOGUCCd9ENp 4JFtbCKCyic8N7A8KJDIsm7jWtwYZbIGDoLbbYOV5DtuQx9kyrYo3MSFCbHYiQzSLQzNij7c CL4pM9T3KzdV795eXz7gkynFFnesfv94fPn2+Pbt7i/vjx+gcD19XP9696tEqt4+dVsn2tCq yIDXY8to+JOzcf5Yxi/cDwI+dN3lAkKb5Mfv6WGjW3yNOTqKktbX4olQg/X18Zfn693/vvu4 voGK/oGp2heGLWkudEYmRI7HSewl9C0J71dmZSy83WUUrdb2u06BN3sFuJ/aPzf18cVbLV3d crxHcxfehM63sBTEfslh2fj0mTPjFxZecHBXFkFtXFie5YJ8XLg2ZjZ9v7jw+cK8sfDteBQ4 HMsDunGROI7Fi3cswAvtC/+Utu7FYs3j3w+sMNH9HgkqsRQWGwttse8y4N+LXEKUb++rwNOM fV6KC5MBm2mBCXQtyCL2r4FBLA0RZm1iC40XM7l2yb3Y3f3lz3GUtgYZc6GHiLb3EAbIWy9P AODtu5XvNosGPvA7OyvLw5UtM8A8PvpbIImgvHSLWxUYjcXBd2QkvkU74k3Ptji9BW2zlilo J8mBYo0UtwhoZXYg2CzuQzFIdn7GdhubqIfoNL51Svvh0v5KPBCGLClER4KVa0syChRNl3uR xW4x4xdWIJ6H9u5/SVyQwtC5rkrIjRYPR/jCFkOOadP95zmw+GRJBPZZEIfK2mgg61poX/n6 9vHbHft+fXv6+vjy8/3r2/Xx5a6b2cPPMRdCku600AvYLZ5jcelBfNUEGHJsEe8uTMQ2Lvxg 4eDL90nn+wsNGAjsss1AENIWKEEBi2FhOSO3cuxnNztGgef1hg+FSXJa0fF7plpck61nbfI/ 4eubhQUFXCG6efR4julPwtugynn/+T9sWBfj47obEuZKVaMUP1upmrvXl+d/D5rKz3We63UB 6IYEAiMBZ+gtOYVTqTcrwmCVxmPK4dGSdffr65uQhgnZ3d9cHj7ZV1+5PXgLyxfR9sUH6Hph yjnaPur4AtCW12rCLxQv8HYOhSYyOzbft9E+X9q5gF8QtFi3BYVr4RQADhqGgV3byy5e4AT2 bcttCt7SlsFz0rf38FA1x9a3cx7WxlXn2b01D2mueXuK5fX6/fvrCw8f9vbr49fr3V/SMnA8 z/2rlAqbMkqPh6uzpIrU9I2qzRrAy+9eX5/f7z7QQ+Gf1+fXH3cv138t6LPHonjo9cjJis+Z 6WDGC9m/Pf747enrO+UIzfakM/me9ayRvZ4EgLvc7+sjd7efykBke846TNBc0T6oSWPmpmcA kw3q4228BBam97fH79e7X37/9VeYn2T6YCh5B5NTJBj+f24twMqqy3YPMki2H+6ypjizJu3T JKMCIu3QKThWCtxWVYfiCfHKBZuwQ5/APG/EsxkVEVf1A1TGDERWsH26zTP1k/ahpctCBFkW IuSy5n5Cq6omzfZln5ZJxkq6r7xGxTEWByDdpU2TJr38uBoHgsX3ebY/qG3Dt1e482rh7joj uiznzeoyHlbUnNHfHt++/evx7UptPRynrAF2QLe7LjyttwCBsdtVcAxh2JcShpD+NH7Ypo2n mLhlqDH9TH1zhpA2y2FEaeskn9y2syJhy7jUe15AHXGRKXUPAPn7ckXm8wTMYa9+XNVpiW7S rVZC6yb8wTBdSnnKYLnolXKgJVTGjNey9M0IeuU02UmvCEHWMHUjnniRo1FM9dHNzdYrdfbz NHKCdaSuCNbAVqzw5ZMc/gxXPM/3qbVcAPsCc5eWWk5yiu6h7bLPRyo2zEy0p+uwTkPDklSO eDSB1HglM9gyLwJpzibrHlwv0ncDB9JDrtHZUC0tEiCGnZjleQZiMwt7OBkr+MQfGyJj7eum isks7AMZvqQvatZlW9jm3YO6ntMKuG2mc4T7h4Y+/QDnJztaJsPqqiqpKlpgRHQXheR7YuSu TZakpc7zWUNbvjmDtJQEC70Qh6jCfQUUzm9W9OmJjH2r0MTHtuNBUeVS9imcDpajZwgipnDO bdHvL90qIH29sTwjTx2fNB4rRt2jKWyVsir0bqHS5tl4n3nTjcAWrRu0eYZ3Y60b2se7e0p6 4Wfc9vHrP56f/v7bB+iFeZyMr02Nd6WA6+Octe3wRnnuIWLGpxMzdNqDlq9m/BAql0KZYamk YmWeSI7ITFufqcipM16P86liAo9uAc+PdqPmz7B/+3OeUvLdTNWyA2sYVT9LMPqEY2lAYtiC za4TOTql4Q19h6yXozYkpo6CgByrmpVJRXfDDL0n9WIMqEv00BYXeG7NKfCcdV5TBW+T0JXj sEhVNvElLku6zlx/sjTsoRs7RVJSMKK7vEGqvRKzDX9jqrAjHL3AF8gVJNEYUhpFFOfHztOj nA0NNzSvsWFtdSzV1HGlaTM9gG5icINDpnwHP+f0vF2TlvuOStEOZA07zwNzJIoZuIFpuflx /YoGJGyO8XAJP2SrLpWFIw6Lm+NFr4ED+92ObqDY1VoxR9B1chW2TfP7rNTLRuWzoXIqCmQG vx6Mb6qjFu1OQRcsZnlOyyv8c+4vYKvyoQahu1WbDpOwr8pGi3U/Q7WxUWpLi9Y+dBivQY5F zmFf7tMHFbRPi23WmDO/syR758gc32yTKhiioY6uOqqRgDn8gTrxEXNmeVfVOv0pS89tVWaU UMub8dCM8fOV7zJ8DGn5Juu05fSJbWUuiaDunJUHZhR7n5YtaKyd5bE6kuSxLS84x8opOAWg rE6VBqv2mbl3Rij+qGuNfwkMuRAQ2xyLbZ7WLAE9eKcwwmy/WTkG8HxI07wVYG3xg4BbwLzb BreAaWzM+SjYww7EDgsT6ptUrHa1x0WGMVSqXWeUVpXA2lLbvi6OeZeNy0+Cl12mAqqmS/8f ZU+y3biu476/Iuet7l3cLluyPCzeghpsqyJaiig7Tm10clN+VTkviasznFPVX98AqYED6Hq9 8bEAcCZBgASBaxMEeyaGB4DZbawHDexfblXWsOJud7RyBCYD25Pdhg4MYqIvt45A18PIHHDv 883GgSZLfUu1J0ny2qp4wdA/CSw+i19Vdc6Zw8YFg0l47SlEMC72egQZCcTXyLBdWkMgmoxx BwTzEfYh88BCova7qvDyIdBBLF5XZ9mOiVxb7wPIWAYyb87q5nN5hwUY27IG90+HJrcXNjA4 kdkcoNkCR+E2rAaNiYPgYnq31OH+gve4q7eVCM1Mb/OclzbvO+Y7btXyS1aXXYs7aA8hGMKX uxT275I6QpQ9KAPxtNt97IybwijNsPvySQBFZbzMokQPdbkWJLR4hK7re9mmDxhi0/YIHTgI RiJuy22St3hkWWTdsakmOGFIDtfREoL3RZW38Z42fEUC+LvzydWIB9kYeD4T7dZkIoDzpKiS vD9TRSJsiSaiDfDq+6+3xwfox+L+F33HsSsrmeExyXL6agexMt7wwdfEhm0PpV3ZobMv1MMq hKE7ErqEu+qSm6wSxkvdQxDdxbnugP22FtkNiFRmiO0OrA4A6DzaGCMBjTkNoN7V0HIQ8dEb wZ4ZPqeAGH0Q9GOmvBooxwbb89v7VTJeRKWOiwCe2IdxCBLpNskJUIs+R5IERNBSP5Ef8Wru DG1HBEj/5Rb/kX08JvWHWBhzL5o1GTMFKG5jkdpFsyIhuYLssnwNfCM1G+GeASE0iRf6U2sE HaSHMGP4EbyHeuZzmDMWeXKzdTtmK2687W1Ksc1jZnvF0Sh4c21kCJJ8kyfUBrrLbuUGrslo 8KXOcShYK0UtQ0JEXFyjFLGDsW+3t6D8oNcuV71EbdvR52R6xpppYHqLV/BdOAmiFaXyKHy1 t2rJRDi3nCYrOMZ+pE4iVf0TPg/1lwkjNLKh8qBqQgEDChi6wPmMoJyvgiMBnUxtaJWwVWR6 ZNThPn4vaTrHxWZCGX6ACtc4YCOnvlUUSdev3Ig+N+B0J8gjMCSKjiKPd58Ov4zI13k9djm3 RyIpsgM6KMoLqt8isjejox19pkfNQzuB7VdSAu1DxQHo9Jzyomq28lLgXTUR02A5sXPqItaI WTBxZmMTRiu3t/3+eCW6SRj6CLXyaookWk3NR0kqN+Wq+MLgwfQ3zTp0bNm49b5u0mC+choq wum6CKcrtw4dyjpZt5iNtPr5++nx5d9/TP+UYkG9ia+6o7+Pl69AQYh9V3+MEvSfFruKUbNw x1FF/vA1mBfHRIXosaAw/E5W6GHe37OgNy2WMX29owZNBgfp1qd3vImgIKodGx5OZ65FFXZZ 8/r47Zsly6ncYAvY0O6zlFTg3Gvl8LuDjWyXUjAVXJGzC0iVr159jSI7VlCxDUvu2kNWx0Ju dXtWUXumU2rGyVLl7STHfxXbKPMCqmiWpjX0B9tRnaHR8WabME8eEnfhyhnmzUyjJ2n0mid1 yqltVKM5KAuV6oCkY/vxq62Ppo9v6YQsv72cYV6Veexpn8S19mtYH53PYWgGGmILnBM9b4qk 3msGRBLlWM7UTQKKVmwCMJb0fDlduhhLDkLQNgH5644G9jdf/3h9f5j8Q+suIAF0A3oe0QjE WnI2gnYHNQ2VK6UGUvaGZMbiQ1LYCtZYAHm9PBDgBbQxiD0Cau2rVn0w9AfUbLEqjhTXE7M4 jr5k+vHAiMnKLyu7fIU5LunYEj2BE3Kxx6TCe0eqk5DBqDWC+YLMfXvHl5HnJW1PA7vg3Pfk R6PBKAG/p1lRjhRMioXbs1q0NAtTiygJ6bblopgGZLQukyII3Gw7DFHgEeCRC66S9VKJfxTC DNqhY0IvxotYEgg+mzbLCdULCtPeppTFTk8U34TBNZW681B+IekQh8tN23lFvzgpMODa3OMT pqcRoGOsJrTxak+zhv3cY4A7FAVrkHy/rRFEyykxxyBhQIx5xkF1I2ZrfQiNN+AjfKneXNvt izgBTGHZDx6G0f2Alyvh+RxuJECk06MDsN9ys1SEQUgsAAUfYm87nQkzMaBdehm9sErIpalw KndHBque7t9Bnn2+XPGEl4Kq9jSg2ATAI/NWXsdElMKgM88lBoHneXHnycHyHUQRkPsCYBaB x++QTjP7D2iWv6vDYkaOczCbzAi4HYNIg1OcSTTX00XDiFnPZ8vGjHOpY8JLlUaCaEVkKfg8 mJEzK76ZeYO/9JOvihI6LEdHgJNz4pbqhFsYl4IdvqTDfLnb3fDKhXfBDfrFen75K6n2v5nu yj2tm9e6gX+TKVGvMSyu2wWLcOK+8EHVUCjPMhfroh0TjvftGAj5YEd7UA9nOIv3a8o5vbjb JWiWTZ0L71UyqwyAtLw8ZJ0tOjnOHZnIijUKdh5Xuopom7HKIuhd0Ju1HhS9/THNRVUw0/Ah nc0WpJdOdGyp+09V360U2Sc/w8XSQqQZZh300GTNNsjRZprUPsLaGi1zgsmgmPANvuTI81Zd mfa1a6bzayOCKqtl8IqK7bJCB+8wVrxC/nNigesSR+qfkXGtCwh1FAoKpRA+g1IgqeVlbNGW HmsMnYRS5zV8fz6r10LrHUU4AvbmeSB8tklOVwJxlZzh2S6vb6hZid6Q0U2zojBKaVmWmABQ tJNShHbxGKiAMAwyaHZZQ+kLMnm91zU2BPG1ckHWgdBsU/OT3UPj8rjZZ3rYeyQ0u0dBME7v nqzbIa1IP5PbEv0gQiojMwndeW6eFFYkwhPiQKKRVYruxrA77HAYDH98eD2/nf/1frX99eP0 +tfh6tvH6e2depqzvasy25n14Kjxci5jJps6u/Nd2IlGHpuQuONyrnnl9sbGqbhS+Mdh6hlu W+WVFUp13EqG+V+XPBvKETYGyAtWKWMhG1GB/m5GOhtQDR2zeSx+TKJAHlv6HltXXGyoZEV1 KRlwoaZ0kl3H0kbj4uOEPgfkDyqejF0wJoxZ7WIOceIC5XHGWrgIdRlv3dkPSM8JhsTvRVyl zisrnhUF25VHwipZnay227KpCuMOU8H14/4tg00zKbRbVviAfmiLsrzeVy4hdHUGW4E23dRR a5fJ2DIg3YqUunkbE5Dhzk30yifjamQij8IZbdNvUUWUfGfSTGdU0xAz82J0Z8QaJkmTbDGZ e3ErXXvUcfJNZ5tUnm7posNdboodCFtHmTcxGuaQUHK3RjAG13VxXfQibu4ciCk2vE029M6x vQWldIfX+g4DT57OD/++EueP14eTK3ZCxqIGfrIMotCYvtmhsaHyszXNCYAyLtKBcjxfRgO5 ZJtXwFYbkKbIXYGs2rDOWF7E5nulgb/zLd0NVUJv+gzjiLGWQ37UlbcqqT+pHI+RYVT2/Rmw 06/16fn8fkL39ZR9Sp2hQRMemZItJxKrTH88v30jlIOOoeufktka+oeEylcEG7xgQgCliEky TTjpq2QUrW26aPl9m9fuA2UBjftD/Hp7Pz1flS9XyffHH39eveFV2L8eHzR7EPV49vnp/A3A 4pwY/dU/oiXQKh1kePrqTeZi1VOV1/P914fzsy8diZcEu2P1af16Or093D+drm7Or/mNL5Pf kUrax//mR18GDk4ibz7un6Bq3rqTeH280ADFGazj49Pjy08rz152ktFTgG/t9QlBpRjs3P6j oR+FLpTI1nV202vl3efV5gyEL2e9Mh0K5LJDZ9gESou6NjP0VY0MxE5kC8wKMUNRogG0gF14 XEw6egjMTaMrJkQu0xqNcCyfxva22UG9dusw2bFJ5JmfzCD7+f5wfukeWbnZKOKWpUn7mSWG YNCjjpXPX1lHsRYMNn9Kee4IbJOODqysA+A3nK2od78dmRs4eUSEoR79eoSbtxAd3N5oe3Cz 62JEmPC6wQjIzIELHkW6pUMH7m0BiYYC6nJATAzBRL7TyHURED7aeL9eZzUFa5OYBKOx0xgp XsNfyzfh6ohGA3eX5CiNE2Wpv7rYrKVxSGWpAtfOQBJoex9q2LedIkU3HvFj5opZPzycnk6v 5+eTHYOIpcciXATeR9IxZzPPNRgoRzAF5G099ZozZYEeizhloXkWnXLQSSYeV2sSR8VYkRj9 4E8zDJY1acPU6tJOC1FYpU2bFKBxd0nZMRceHN43WPjro0hX1qf5NPr6mHy+nk6mpgiWhAF5 u8Q5W8z0tdkB7BAMCKYjmANmaTzRA8AqiqZWfLMOagPMWkrnvJTADJh5oFdTNNfLUHemjoCY RUYIFWsKqmmpgkai45LOzxCwXeC17iRdTFbTmqoMoIKVMa8AMp/M23wN2wWe67GiyGiPREC5 WtE3dSzNgYnnyOOpQoG9TzBkqDbUiuV3sFE/QFdtk6knn2x3yIqywqCgTZY0pf6a4rjQ3ckX TRLMFjZgaUSIkiCP4RbuB6HPeTTooXOfn+KkCmcB5fd+x/YLZbymHY3iXgEM3WrtKAmrmOKe 3hCp3FJ5mSqrubGxjRyIyXJq9K2EClhd1LxAJIedzhql7moX7Y9M6ByhsubaQeF6Pp3YI9rJ ZEenjf08vzSn9VkvffhcZZaDHuQ5dSYSZj9LNrPXEndS/I8nEPGsdbPlycx2bjXI9UMCleL7 6Vla1Ku7EE3aYU0Bg1ptOy5rLDWJyr6U/scYMc/m+jagvk0umSRiqU/2nN0k1pEc5p9jxNdW bCqSeYpKWGEWviztxd2rt3Zbqe1CtUhYnJOgMDRTIoMCn6PsNoWrp20fv/ZXT5Cw8ytlvojt djYlGphGuRZ63PHHZylk/npTMNZhV00tqJ8QVZ9uqNOoYThIa7s0M6RxXaeaLuYwepec3MYm MDDkaKLHI4bvUJ9X8D2bzc19IIpWIW1tCbj5au45rU2rEp0GGKs+FbNZQFkb8XkQ6qYEwEyj 6cLYSAFiBf8dmetsoZ+SAdOCcqNoYUYgkWwIEPSV3aXuU8ZdMPZfP56fe2de+mg6uM6t0ul/ Pk4vD7+uxK+X9++nt8f/RcvZNBWdwz/tAGtzejm93r+fXz+lj+gg8O8PvDvUy7hIp4wfvt+/ nf4qgOz09ao4n39c/QHloD/Dvh5vWj30vP+/KUdnGhdbaEzMb79ez28P5x+nqzebN8Z8M50b 3A2/Lfc4RyYCdCVKwmzZTlvVm7u6BGmWmjjVPpzo2lcHsDPrVp3KCGVXSmFoNmEwMUQ1f8MV 2zrdP71/1zaKHvr6flXfv5+u+Pnl8d3cQ9bZzLC4QDV0MjXiQihIYDAwKk8NqVdDVeLj+fHr 4/svd6QYD8KpIS6l24Y8Zd6mKLMdSUa73fM8NYydt40I9FcR6tsZ1WZPcgCRLyYTo1YIsd3W 9g22G6dWNyyrd7Rxfz7dv328qqh2H9BZhiQQ87ybmCQ/XB9LsYSqeAmu+XFONSDfHXDyzeXk M/RuHUHOykLweSroHfpCo5TFuvThMQ7yyNiTCgS/grRWTz/DOFpKKEv3x6nPXzjDYBBeFIZ0 pYqpUrEK9ZktISvLAcx2uiCD3CDCtGlMeBhMl+R9CDfN2OA71K0xE3xnFFl5zefkNZEuuXSu YCynZJsqYNXEY2akkNAjkwltZTBIBaIIVhNP8BSTKKCsWCVqqm+a+gFA4b71VhhsDJHbZ8Gm ga651lU9MV45DVKc/earqS1/U8UBJsssoaYe8LbZbGJxO4QY9nG7kk1DUuMuqwZmlFarCqod TEyYyKdTvYb4bYYrBrU8DEljUFiO+0MuAkOp70DmbtYkIpzpt4gSYNoi933WwFj57KwlzhMy AXELTzwMwM2ikJrBexFNl4FxFn1IdoUd5cZAhWaI5YxLPZAilygrWl8xp4MUf4HxguGZ6luZ ybGUBcf9t5fTuzoQITas6+VqoQu8+B3p35PVSteauuM4zjY7EuicIrFNOCVng7ZuMGHWlDxD 7wGh+XI2jIKZ0SEdY5eFOeKGMztAOY2Ws9C75fR0NYc5S+xMvQUL1Yv/NQR9+fF0+mkoElLt 6TwK9VnohN2m+vD0+OIbGl3z2iVFvtP7h+I+6rS2rctGep7xbHlEkcq7b/c06+qvKxXC5un8 cjIbtK2bnGsHxobehVcGdb2vGhrdILcvyrKi0dJ4g9Iu6Wp12/MLSG4qGPPLt48n+P/j/PaI Ar+xaQ9r4/fkhkT+4/wOAsEjeZwdBQvquCoVUzv0F+hlvjAYqJlZm5SGAf6j8b+qsMVWTzXJ JkDXmZJawauVG8XFk7NKrXQoDDYMohIpFcXVZD7h1EurmFeBeT6D35TI1ksHMfM4p0yLLfBI T+iQSoQeUcrYpTNBumWpzKHLkwqj0nnCklXFdBp5dHtAAtPTj9hFNNe5qPo29zyEhQuC0zn1 HbepaEYGWt5WwWRudO2XioEQR4drdMZ0FH5fHl++0WvJRnaz4/zz8RmVB1xlX2UsrAdyrkgB K/LEZyvylNXo0SRrD9Qq4/E0MJ+wVz67wHqdYphbKhdRr80w8uK48sgtR6iqNnsx5dKWAEJL vh+27ygsJsdhqg99frGnOhOTt/MTvkD231wM9iQXKRV/Pz3/wKMRc/G6q6PJdLN6XhxXk7ku iSmIrhA0HAT2ufWt3fE2wNxNIVZCAtp5IlVPTXptPIHieWb7sumnxq1m1QcfaqsxJs8t9z4R RRxreFa02yJJE9vsC9HE7bGGXYuiXTdOIuk1gZZLFVoI20MHQdAZRHmKlh4IzMsbBDe39D1V h7O9WipJpb6RsVkJB0b1Ddp5GdoutDgn2SJL0Q5LGXePIomdt8YlKpZce8YVmGLW4E12U5dF YXqjUri4Trho4u6ig2YPklBZO2yo98iKoMm71/n9aXK1vbsSH3+/SROYsTM6y/MW0OOc04Ay mj3sUjo6Tnh7Xe4YmiIEXcpxTCBN9zylbcq6znaU/a1OZWauY0QO8hnz5S5YcaAcfyANzuGc H5f8Bitp58DzI/Te0DJPHtWRtcFyx9ut0D34GijsAad+MIsr10uTXj6rqm25y1qe8vncs10j YZlkRYm3A3Wa0fspUqnJkJQ8puWPkSZzPA/1zN2YHUNT0fgIGmPs8DKjmlX09MzTIgOaz1ZY AE2AjJ2lWp1e8YWf3E6e1XGmYa3f1/ECmbY+mNc92Mwpmb18fT0/fjVEwl1alznN6HvyQXZm 2mkomk8bgP49u/45sHJ1Knt79f56/yCFEptLCZMDw6cydwcxU5CsaqTAh/maCRciZGgTOz9R 7mtYRwARJfnqSiPaZqxu4ow17lxotmRnEY3TzlWrDXVGuBaaoy34kD4ssGN3ZWr6Cwdc57rP 4/5Ho7Cs7zUMk24T6ZNfoAL+Se1UEhVnaPFk1rZMdOEYPWdWRXYcrY00fdq1aQTdG7S0zWIV MD0T21MPQgZja1dPd6ySK96WlSYeibw8ml+4W1mFiCLnse61EAHKUjdpau19lVShExUGRDtf Lfe7RteXQZxob/YsTc3HIbwUDTl3LANDdff2+ATilWRQuvFlwpJt1t6iM1HXPcqBoWAOQjlo 6hWrBelLFnB5yZnWR9mxCdq1cADtkTWNsW33iKoUOQxfQtl79TQiS/a1cUcCmNAuJzSys8oK 9XzoomZ2hjM7QwtFVmtmO+j4HKeB+eXErRCg5cjRMOWaHHodcORDl88SoeXra/tnT7sNAr/z GJkcz5jQLRy9PxydOo5n+GsR0PXHABfdTBmoe1hbBgnlVXLAD7a76ODc9k46UGG16WopEtlo 5GfXRUmpAzqVWdG4ccdl1FrywtvqdWANmwRgRV2otmgsMDnOPfLiWEsimGcgaF+oHwP+20ki llPrvhAMP4IHFc7J4zijYUOnOQa5oPDZhbn6FKTzX2nGf8pBUEJwrrvyRaNxNPG68+DX+Pox qe+qrk0UGLSZjRmLSKBHKJpjrIX9wiy1AbkCyPmqFclsupt92RjCugTgg0/5uENuFWh7SInb NWA7+ltW74w2K7DFbRSwqTPNBulmzZv2YKjtCkSdpcgMksaYfWzflGsxo2eVQlpraL1Hl/Ik a4AuL9idsSJGGHrpVlFyUt1VNEXAilsmo4cVRXlLkua7NDNuJDXcDqfG0X4PRFHyDPqjrNwH tcn9w3fTBdNaSD5P25ApakWe/lWX/FN6SOXu7WzeuShXoP+YW0BZ5JkmS3wBIrPP9+na4Vp9 4XSB6oC5FJ/WrPmUHfEXtNL/q+zIluPGcb/iytNu1cxU7NiJvVV50NXd2tYVHd22X1Q9To/T NfFRPmqT/foFQFECSVDOPsw4TUC8CQIgDrNLE1PSAKaPLG4WROREGFUsb4aitSgmFegtzVRF uPBbmSWa674So573r18fjv6SZhr9uqytS0VrNDgTD2RJ7qnqgPDCKkD3/rJIW9PkkYDRKs1i EPl9NVZo8ocxmPGy4PzlOqkLPj9aTBp+tnlldp4KZrkuhWHdPatuCdQo5FUPRTQuRocT5U6d GMlO1J+JCGjB1J14tp3SRkW1wHhoSe65bZMWONi1D09jZXwHZbiB6Jb7/O7w/HB+fnbx+/E7 Do5AaKLlOv3wyfxwhHzyQ/ibqgE55wZWFuTEC/HXZjwgmDDRlt9COfZV/PFkpmLpDcJCOZ35 XHZOtpBkFw4LSQ52ZSBdfJAcmkwU75pcfPCtycXphW/qPjljByqMO6yXnv6Mb49Pzt57Jw6A koUC4lCkEl+r8qMLx5CfKzmGb8U13DtkyfCEwz/6PpQ9DjiGf/HHkcsadwPl9G0U3xjWZXre 1/YAqLTzfJIHUQ+XLA/Zr4ujJGu5qnQqB96vq0u7HYLVJchlgcyBj0hXmA9NVHtplGWQZFLb mJFi7Ran0FcjEOsIKDqeLdYYcSoNuu3qtRElHQFdu2BhfuLM0OXBzxlZtSvSyErmN1nnch2I Mknf37w+4aPcFFBpvFKvjCsTfwNX+QVjvvQC/6bvaRBKQUyGFcMvgC9fSldRi7kzklg3ojkg JYNM5bzxPl5hqlSVY0iqU8t8GFKnoXeMtk4jQ+E4o/rQIONux+gVFOGjSFSkW2RwgaMGucp0 pHGQeKtuDQuoAsOLSA/ZDjJSPpDzTREUGHKUqZR2VTbzQX1FRNVg6l+V+Vcy4FBMAJu+gB2D rMk/v/u5u9v99v1h9/XxcP/b8+6vPXx++Pobxle9xc3zTu2l9f7pfv+d0gXv6aV82lNKfbm/ e3j6eXS4P6AJ6uG/u8GgfZQT0xZ7DZJuUZp5LglEkinM/Nhjj9itkRdwcr24Y1ZusUsa7B/R 6LVhn5+R18OdXGrNbfT08/Hl4ejm4Wl/9PB09G3//ZH8BQxkFLyDimmwjeITtzzhwZ5ZoYva rKO0WnH52wK4n6yMuFis0EWtubg9lYmII9PpdNzbk8DX+XVVudhrrq3WNaCqxkUFkhwshXqH cvcDU4VhYmNC6yDMEoqq0ThYy8XxyXneZQ6g6DK50GA/h/KK/gqneIDTH2FTdO0KiKtQIXbW X12T5m5ly6zTicQxCJXe4tXrn98PN7//vf95dEO7/RZTK/50NnnNc2cPZbG70xL+FjKWiYh1 bGbf1nPR1Zvk5OzMjAer3u1eX76hndfN7mX/9Si5pw6jVdx/Di/fjoLn54ebA4Hi3cvOGUEU 5e6kCGXRCm7J4OR9VWZXaIAs9DFIlikGGZ1ZguRLuhEGvQqAzm305IfkCnT38HX/7HY3lBY+ EpOdaWDrbvNI2NQJd/MfyrJ6KzRXzjVXRaG72JdtI9QDjMC2Dip/XcWKTbc12Rj2re1yoVrU 4m6cbbLaPX8bJ9WZwDyQY4spQpkH0qxfwkjFS2uAb6xKtWnj/vnFXdc6+nAiLi0B1OvqHKmI PkgkhsphSTIgVnNdvbxcWUkEbYwwC9bJiWzGYKDIMcx0b9rj93G6cE+ceEF5Fz+PT4Uy6Uzm KRwusu+YXao6j62DK2GIGogJfnL20ekVFH/gEWM1IVgFx1KhVAUUnx1LiwsASY7V0PyDWxXq yMPSveDbZX184V6R20q1rE7M4fGbGX1Lkz3pbEOpFcjHhRep2tfS50UXip54Gl5H7hYIs3Jr JqmyAJN7sHNQAgymJ6aXHTFQVLLcixnM3aVY6i5nLE7XwuEFHGK5Cq4D2TxBr26QNYFoxWnd Ye6+MJInjoV1ZYT/Mcv7pklO+rNzYcPmp8II22RmctttKS7cUO6bdw0+mziX6OHuEQ2CDWlk nPpFZupwh1vuuhQ6fH4qPRWNn7jbD8pWEgm/blrXJLHe3X99uDsqXu/+3D9pv1/tE2yfhSbt o6oupCddPbQ6XOogqwJkuMEkiER4CSLxDQhwCv+dYkLsBC0ZqysHqpLVVKkwMRrUv3X5jIha wvFPxIhaF8u5JlG0mj1M9lO7KzthphZbKPx++PNpB0Lo08Pry+FeYOCyNBwoplCuiJrdFQS9 ebcikjrf2mLTU5NCmhs6YYn8tosXe4air26QENLr5PPxHMp8fzXamz22GPT5fnvu2pXE7qLN WRXEdnxKF2mZWJZpDLZKF0X/6eJMjqTDEJW5duKJBekg4kDen86QVkSNIleMxvIvpgmfCQFB 7vzi7MfbHUHcCOPJ/xLixxPJisLT9GYx0z9qcyMl6xXa3LhMJ80MyL4Nt3YbCvq0ajBScZGl ReJC22xMdyT3T0WofGtCmmCRXMqBwYzVU5YN4ibIMa141C8vpUqC5irPE9ScktIVE7pOY2HA qguzAafpQi9aW+UWzmQxdfb+oo8SVGSmERrZuRZ2k1Z5HTXnaOaxQUSscAZZt+miKLKLLvB/ kez/TKnung+398r74ubb/ubvw/0tM6Ol11Sur64N4xIX3nx+xzJZDfDksq0DPlhZA10WcVBf vdkaUHYMTd20v4BBtw7+S3VLGzr8whzoKsO0wE6Rjc1C312Z99LC3R/UfY1x/BmhR78Go6dh CpIFxpRnG0c7DoDQUUSoJ6/L3FK6cZQsKTzQImn7rk35g7cGLdIihv/VMDdhahid1rF5o2Bu 9aQvujyU0/Spp4ggc9uoKLuvYROqQVbxmFF6gbLCYPOb8iERBpoJwaECvq4YfF2NOzSCAw8c lVF0/NHEcMVo6Ezb9eZXtkoAdQE694WHNBEK0IIkvPKJwwxFfmwcUIJ6K58NBTfXq44+Ggy1 Kd1FPC9nGo7KkgmBPbAFXZy2jKkYF7iIy5zNwAS6Ro4AmDlTOLhW/I1VCrLCaCtolsaJVH4q YoOgIJebtUzk9RoB4oOPRu+X1ynbiwwQAuBEhGTXPNeh3tXCk1gYrYwf5CfQUuBAnkeAzPo2 QWbZ4l0GdR1cqQPBL5amjFLY/5ukJ4QJhGcIzhb3mFBFlJvEOHNYbmRsLECW7RuKgdkDTVm2 KwuGAKiCOHweKheKKTuGYZnZbNOyzUITLaL2lDpv/9fu9fsLuiu+HG5fH16fj+7UM9Tuab87 wohJ/2IyACWOvE76PLyCLTglUhkBDeqqFNCISMLAVVLjG7o3o4pRVSo/q5lIol0rogRZuixy VAKcsxdrBKC3lMfnsFlmagsZbFOEUfKXRdB2VhqJqkN7Zcz+Qs+FUk+qrq+NRY+/cDqdlaH5 SzjiRWYaf0XZdd8GPK5t/QVFDVZvXqVGSsw4zY3fZRpj8mK4jWtj68J21idpEzele76WSYuB CMpFHAh+dvhNz+10DEBLtxg3HS5RFWNn+qTS8x/82qAitFuFyTEcNBr07SrZwLVRXrTeBjwt BBXFSVXyj4GMG0uD7/TFkq8A87q22AzzoVkza1T6+HS4f/lbeRff7Z9vXZMGYmHWNCHWPY/F EQY0FdUEysGpB7Y5A44lG58wP3kxvnRohHo67guVzsit4XTqRViWre4K5W0ST2F8VQSYvNVv /2Fg9B77TOAlwhL596SuAd0I/42fwX+YBLhsEr4a3hke1WiH7/vfXw53Axf5TKg3qvzJXQ/V 1qASccrQhLmLEisI+gjVN5LH/4phNsBRycwLQ4q3Qb2QmZNlHGKy3LQStTtJQa++eYe6XiRa 7PzAVZeQQfrn4/cnp3y/V3CXoTcfvwrrJIipLgDxQa8S9CJGO204TpkYEprG0ShXBTQZzYOW X782hPrUl0V2ZXe2KlPT9Wrw/CjRiW+bBGuKnw3kle+LX1552iekuDzc6FMc7/98vb1F44r0 /vnl6RVDknG3tgBlVRBreH4sVjhaeKh1+Pz+x/E0cxwP5IQ08E+eY/dDJG0Na8+XAn9LQrPm 4ruwCQrgY4u0xasy4JcDwXhlCrmVnzMVMMRcFI1VB9kEuxXxVmXLMEIbL2hZdkbrLkIU7WV+ ae3MiUXTbJ5SbXA4UkPgdkFjZYxgI9EE2RmD1JpvMKoWhBPfIBmM47fltjD9oqgUNnlTFr4A GlPVcCLloGsKpS7jAJ1IZBOKcU8o5O2lPQW8ZBQc27jjwSjU796O2jsUD1mjvHu6DNF7qXHn bQCIkp2IuLDUSSaUYiR5MiMaiGik/mZbddQRyfO3h+wsMHiDm+ibFQ4UWt+5jD40WRd61fZE BYY9DHJFBsTP7ZKG+IkyGbR1jXIPmFqGqyIegEkRq5vj7V20yftq2RKhc7qykXPD2x/+QiMg 0nRBJrSgAN6xqswRZIBnb/bh1kApyia0bJ7QnWhhuCnNAqOIer0OkK450vsAxT2H7GVRTgQy jk0pWtVArdD2MK0DJ8JkT0izshI7DgIe4B+VD4/Pvx1hBN3XR3Udrnb3t5wbhe5EaKhYGuKj UYweyB17C1FAEgU6llQT3xm6Soh235SL1gtEjpPEcY5GLfwKjt01NFy1mqI8NXy1HQypIYbm 7YyNM3aGLRC20K86WPoWpEVh026/AB8E3FBcGs9+dAWqysU7cH55lVU1MEFfX5HzES41dfQd 5y0qFrx1taGpUKV5SHBXrJOkUlpWpaVFq7Dpkv7H8+PhHi3FoOd3ry/7H3v4x/7l5o8//vgn U+CirylVuSSxzBYUqxpT5k6upUyOQkAdbFUVBcyj744lBBysl5ag0qZrk0uuQx4O3ZD4zWEp ZPTtVkGA2JfbKuC6naGlbZPkzmfUQ0sngWUgzLqEcQB4B4MJOJERzRLf1zjT9Fo9m2eYOgXH BjUiPr5jGq8kT/8fG0JXCAwq8P5A6BZZsOR6LiSXBJzKSK6AWeu7Ag1EYJ8rHapwcarb2UM8 /1b85dfdy+4IGcsbfJowrPCGqUtnmJeKnj3s3bO0S8gPOU3MhAPEPxQ9MXjAfWGUR1+Aydke m01FINIC552qeLrKqCPqRM5XnaXIyICLnBGSVGfpDYw39gei4E1MsuR4lZwcW5XUgcevAKHJ FzFago7NZgzJXjKgukqarIkjmOFZlNc7sP/4bOl5fYCBDAlLlXJWh8eSjw8gFNFVW4oCF5p5 THvapXsFRfIEkOH3Aau06AolXs9DlyDorWQcrbNZWMdJAPbbtF2hEtFhowS0wQUcNVs2+oCW ExMN9eHzl4WC/ru0RxCTFAN2JdHwoaplAqq6I5NKk5LPTuNF6eII33gcxFXEhW9gGJE7aawq otFbQOQaxQrElRxOLIj9Yued9oYCtuiTq41zFNgRTWOQMVdRevzh4pS03APvO93qASb0E+MB TEw3RXZKGzUUU+GlfJgGHIdY/jj/KBEPk2K7OzkJ6uxK6xpVfLMBgkmlB8UfcVs8ly//ylNX HC6N+81uqL+MQ/lJihiYLFxkHbcto02G0YrsIzm9ckGH8XUqxsM7906JqdZRq9q/vzyXA6Yx DI9OccToHAWtjTF4MJn0jDS8+v2L0bDA+zCiPkQ7uSu7tiJPhUcLNR+k0KqYYV/VoVMUsiF2 IIyu2GL8h9rR/43k3NxkXAHf7p9fkI1A7jfCxJa72z2/o9fYqjiT+s5F/TNFex7CrIjIVigW SUOvRD0Q4qJyM2z/ij/kAnXCV5BWsaROavlsHbey9KyEATSkaHwxMAglTwtUWlR+DO/34XTl wHLO3O0hPpHOwOkVs8xKzKzsPwf8vdWPNqhYfLot4mc/npo8Jh/tKrlEFdXMdKhHH+VHJLqI DlhNZHpTUvkaAG0pW48RgrJW8VU7vkCZH0ExbMpMPvxKM9rZof04VD1a++Fag+HHqNFcx1HT WFPrs3wlaBrLhppqI69ndjmMvqw8UZcRvskdVas1OchB2SFgrDYqWaGqgGi0tCpJVbeRKQHa 7kA/+xD4uFUemNpEs7ZFWucgjsxMpIpPMjMe/zPcsDnJwxZtvmYIQ5JHAezH2WpQ/vNQSl2J jTCAAWLLeLPE2fFsVU+o/wM9hElzRzECAA== --tThc/1wpZn/ma/RB--