From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 32125 invoked by alias); 10 Feb 2002 14:36:02 -0000 Mailing-List: contact gcc-prs-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-prs-owner@gcc.gnu.org Received: (qmail 32073 invoked by uid 71); 10 Feb 2002 14:36:01 -0000 Resent-Date: 10 Feb 2002 14:36:01 -0000 Resent-Message-ID: <20020210143601.32072.qmail@sources.redhat.com> Resent-From: gcc-gnats@gcc.gnu.org (GNATS Filer) Resent-To: nobody@gcc.gnu.org Resent-Cc: gcc-prs@gcc.gnu.org, gcc-bugs@gcc.gnu.org Resent-Reply-To: gcc-gnats@gcc.gnu.org, David.Billinghurst@riotinto.com Received:(qmail 29558 invoked by uid 61); 10 Feb 2002 14:34:41 -0000 Message-Id:<20020210143441.29556.qmail@sources.redhat.com> Date: Sun, 10 Feb 2002 06:36:00 -0000 From: David.Billinghurst@riotinto.com Reply-To: David.Billinghurst@riotinto.com To: gcc-gnats@gcc.gnu.org X-Send-Pr-Version:gnatsweb-2.9.3 (1.1.1.1.2.31) Subject: fortran/5651: Optimization (-funroll-loops) error with LAPACK sstebz.f X-SW-Source: 2002-02/txt/msg00234.txt.bz2 List-Id: >Number: 5651 >Category: fortran >Synopsis: Optimization (-funroll-loops) error with LAPACK sstebz.f >Confidential: no >Severity: serious >Priority: medium >Responsible: unassigned >State: open >Class: sw-bug >Submitter-Id: net >Arrival-Date: Sun Feb 10 06:36:00 PST 2002 >Closed-Date: >Last-Modified: >Originator: David.Billinghurst@riotinto.com >Release: 3.1 20020208 >Organization: >Environment: mips-sgi-irix6.5 >Description: LAPACK test program xeigtsts core dumps on test ssep when compiled with "-O3 -funroll-loops". The problem does not happen with "-O3" or "-O2". The problem is in file (subroutine) sstebz.f. Recompiling this routine alone at lower optimization fixes the problem. I have attempted to reduce the size of the problem file. It is now 600 lines, and I cannot make it smaller. The correct output from the test problem is: Calling SSTEBZ In sstebz: RANGE = A ORDER = E N = 2 VL = 3.68575473E+19 VU = -5.49755519E+11 IL = 1 IU = 1 ABSTOL = 0. D( 1) = -835964416. D( 2) = -3.03700045E+09 E( 1) = -1.59336973E+09 In SLAEBZ2: NITMAX = 99 The wrong output is: Calling SSTEBZ In sstebz: RANGE = A ORDER = E N = 2 VL = 3.68575473E+19 VU = -5.49755519E+11 IL = 1 IU = 1 ABSTOL = 0. D( 1) = -835964416. D( 2) = -3.03700045E+09 E( 1) = -1.59336973E+09 In SLAEBZ2: NITMAX = -2147483647 >How-To-Repeat: Compile four attached files with "-O3 -funroll-loops" Link Run >Fix: >Release-Note: >Audit-Trail: >Unformatted: ----gnatsweb-attachment---- Content-Type: application/x-gzip-compressed; name="bug.tar.gz" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="bug.tar.gz" H4sIALiEZjwAA+w8aW/bSLLzOb+iEQSwZFOKDh/jYLWAIssZx7LsteVkscC8ASW1bNoUqSEpKx7s j9+q6pOXJPvtTOY9DJFYEtldXV13VTd7GnlPPPqlWZ/98LtdjWajcbjf+KEB135LfDbbTfrEr/uN gx8aR/vtxlFrv3XUgvato/bRD6zx+6FkrmWcuBFjP4w93/eCu2lZu4hPl5Mk+CNQ+iMvRtciCu8i d86mUhreiNvefOF7Ey9hQRhwea/3U/e62xv1r5m+Lq9P+tcOu+4OP/Vlq7PhqP/JbgO3Bg7cPr2E v7cOu3DYEP7dXA3ORrLPdb87YOmr+/FmdAn9vuD/2zWwPw4ue+cV1mJVAE9A1Y+vl9fw4JBVy0c5 kW37FdbEz6/yt+j6I3RVfXGGjHXYTnfnjTV3utVXt4bio8NYS975MlB3WLt++OPB0cH+Ubu/1zxW z2/V89pBff/46ODgoHkMzxUfznR/fec2e0eQisZo1N/oiTWrBPbH9sHx4f5+89B61BKP2vVG+wg1 86C/11AY9XXPZv3guN0+PEaE4bF8voq8hFcOnd0q2+m5pDns5mbU//ivHdVkArflvYqgnKMkZag4 6gipgE/F6RNggyMgvGP560LJDHDHkWxXHBcMkyxfAwOF0PCUB9M331sJv+MV+y4f//Y97f9Bo9k8 FPb/8LC9f9Ag+9/6y/7/IZdQgpvbj9eXt6OzYZ/dDLqksWefLz+Csp2NLrr/JI29oC8XZ8Mr+PkR PrXSlqvadX9AWn119oU6kHazfgsgfOkOEIDDegAUxl4DZYjNbN1WKrz7ZhcsVI0Nulfd3jlzl988 33OjZxaFy8QLOKuAN4u9MACz26hCS2wP123gPdVZOGMjHgQ8jjl3zD0wZ94sjALPZR959Mh9/gzo dj+xQTKtOxJCL1xGbpCwsyBOvGSZAIBudBcCNDZ0ExjR9dnAHTvMDabs2ptwgo/IJM8SxOclNG43 HNY8Pj4WU4GrXmc3YDlBJAHecs6DJIZ7azwfsUn4VcEbySckKblYwSrByM2eVnFKcE4jZbDrRpH7 vC1ywoPuojMFJlYkbuI3CAA9WodTqkdPQjqRn3312RJfygVIo2HNB79eLaNFGHP82hGXuC90gE3C IHG9IGbJPWfg7yLiLPPDcBGz1b03uYcm8wVwn9i8jDm2RACzZTChtsPKCjAUbT0BaBIuQXBA0Lh3 x4Mn11/yGH+6LH6ez3kSeRMEAZ9Tz70jSZq78OsbGzEfhBWAuAELI8Z/XcKzJATUYuZKhoBrrgPh EwSx4BEI8hwessl9iEIIwySrkCXPCzEmTeWDmDRKUqfpsFno++GKT9n4Wd9ufQCQLHEfoZsbMy+A SQNU34tpJl4AxIGZxESHiCfLKIjNc0l1uuLlbAYhJWDqP7N4jjGC6b26B26wZYCUSxE/dufchmKT zhVNRC+cIPwII+/OQ8pp2HW7+wj5SXOwMI/AXIDEPTjNqkOfrerPDnsAktTrdQd1KwcDrIwNJN+f rcCrqKmUTWAoe+2JkYayM4oNB6Sa7G8d9oB/UKfrFrPaxBWLy2OYM1i/mLsRylvAuEufNpI2Fnl8 ARLAWYQeSVLloYocAxggc4ndE+QaHlY7pMUPVUcpQAymkXpJrqBzTTA4FCDTskANCFcU2ZkYyRXD grLMQFGmDrYCvgaS2HZ/iX1HYt9ZCTBBKCBtDaeEbWOOOpkSUcVKnJFUdXWJCQqWodazh+V8gUQA X3R3zzShloFS4kT0sYH4HurlMom9KRdWJ/ASz5Jjyf5hmHABQjSyhRjYjihPXOTGvevPauECZq4b kQvz6rzuKG1B8QEquM74Z6aNVTDxl1Pk5xj+LTEFBIlwmURgFDL3KfSA58EU5A8MBhFYWar4Plz6 U6RgDM4MbEkcCnTRVIFzu+MxzRucKxkt4BJw7S7iLiApDBwLnwTg3d1K8z3kKrtmLLq1T9YcZzuO Qx+NMOlTXSAXx8tIkHAeghFyJxNw2AmXBpsMOXYHCggOWwq5fiZgb8fEeCTIHOWMACi0cZYOQAMQ kaTVDefsa52du/j8bVdh0s+a/xtl/tnI2H4EcEGovAWnzBdhlLDezT7Y6Z5wPRHEC2BOwbif8EUC TL1J3AA4OsWeJuhwIN4Ai9tqYrxxeGjE6APN1tX+HETIQaoCcuDwfPC9MNFdcCB88gjTRwOxDIBR MWA39okSYhJytjoysLyq8qtotUjhyCRVVchga8DNgk+8GarBiuQlFgRn0zDgH+yGHdb8wBQR0IYS akU6k7b9HQbOjF0JowkWM+bCUxv/DgPOPDBnGf88yoBpbwbjBUD9RAYBpHI2CGkE5WiutiCofvcu 2vNYEmPKguV8DKxO28+8K0E0VUTg81mSQvkSZVIySzomF2MWGe1AN+G54VlyTyFlB7J/KSkUPq7l 3IiU5ps3X84Nuuytz5+4H7/F74ZKWom09RSE5FNJJ/AnxuhC15U3BZy+CrRR88Y42DQFI6WHrf+p 1ATSaDm+Ks+QsFTEYXe/d5/QPCDL7oDi7gx1S85bczWWfoQSEBDOmKfsdxKSBJrpB2FQMzBtiSSq ym4bqDr1QKEojdEBTlF0SHEfmKSYqJP21z4obMIUNy+Il6/gppkA0XMeko0FchPEQrKiSxK2JOHK C0uJI2b+uvQSS+IQ0J7GE4KuLfBUCl+CJ9AEMhYpNLaPseHguIormDFtHlWIG1C1YFjh7IzvWETh BPPMqQ1kGWMY4bInUAowXRiLC7L+xqNQxytgtClOEkkhNrJhqEAF9HiqDLAoBGr0MbHKcRZIhpyt KN9ZlRqGjshoHqDzFbFIk0rrJXptN4ogWZBKp5LIUP4WSSRLQH7jbLxEUUCEWAJ77oCDEG1UsRpH ZlBO3yMPAAqEAVFke1915fMJZUG02mFEcA9gjGYIdbDBIMkl/STSW9Ev4j6YBbAb66lXGHy/mHoW yWw4vz/1VJiA9o5EOqXf/sp9jm26ssidet9255B7YA2GL2LPDwOVclN9YTvqpgM7ESS9XXhPYfIW IxQkyQ04rHlqRpAVUzQk1Yk4v4us303hiMaN/ZtDSL672/o3203BcGf8F0CCUVaju6jbKjHTzWCE IplKCowE2YUJMHfqPWGMD1ENGT+I+XXAKyl1IsHYlAJrGrkQyxmPUBlW8+5COgYZXcdrnYYYrP/a wcLZ7EXjsbTwLkIITtGvQpqrUyWIOnAFYkjZmxuNvSSCtFZh2nolpvGvS/BFGrktEU9rmna0/ZZE z7sLwAkq24tZnkHuvUg1tQcpQBM9XApTMP+iEINmoOWQ24r4DOQNhHpaL2raFgkLJGseKrkJBTHu rIvJ2zNEJIt8SMQha1B2YhJGQKxFCApAzjmdZ3porTLlkGXsqCzXpNk6YLNzChxMVXUgU0vVFQwA IqFDuaeXqEJKOVLrE3qNkHaSthRpRpUIExUhWzmJUuZWy40ddkgUUEaYS5k/mXAIylN1IAlCQXio JfcajKiqyDkgoLGCY4OIvLv7ZD2gemHJKysD6bQPA+xwStmHA4mI7yWOrO4pKRk/Z1GBGGWy9ClQ lnTuKY5sTedibXAsRcs/Bj1ZhdFjvHAnfI2KBFqsRJHKxGcygvR+E8KfmdbMi9D8i6qaoLJ0P6mC mwpaL29HNOWM9pfMK50v2MmcHKM4jdVTRzvRzoIxmqqlvpAohCupligmyyonmTKpItubsrySpKdp Vd+o5uk5D1VtfCCRQsoPK/JBtYzRKT46BpBVqOGJSYUy+XIMwhnPnmUhPpiS8/mQaSSRaFarWHUd yh/me0t8r8gf1WxWL1L4OXcDmQh4VjLriZjeYnR2dDmeVQ6mIRWlTDwHs7yNgW3+cxYDRZWOoafD ZlE4xzJDxJ+8cBmLBXqguc7GkvsMGMG8eubupeKik2FjQZ2bUaBVCdzKY9WBWKkSjPGbQauq69yP aoHEC6b8W77UkeQsmKmAZgrxmZ55dzBBz4P0EDaN0BPfxvQt4lkbkELAcrEA8hGFAzuLH61i0W1n dcCinl7rsNiVGVyWjamy/HfhJr2qKNcD5W3bBP5mxciWkxxm4GBZ2Ih0qpyirb3IAAxdMyBENK7E 0XLkWQqxCsBc+mGmv/QgkPzGuHgbcy7iJuHsqpQA2dP9m5puBs4Wk7enK1UoAyQzecp2ojnpFJnA lFlBbYnDeVYwpl6ceLjg9yRTlUpchR4zrI5I5fKEuulAEVckSTS139reKX61XB1VVSUwG9Z6M70e HJa22Hr/1WGND/JrN7WCZ9LGTK22VsOBP1Ac4mM2JSpouuPUmxIvFIBMf1EWgjEvSkpObMXtQpOc y+kyoqLnCQf98tMlabswjRdlimrjAEb2ABoskcyaJfuE8YbIJyS4YuMBQpAdqT64jI3jngFhERqJ 8zLmUW0WQb499Z+Fj7IGQjBUw16FbCFWptXabMWtUpUYC0aW50AQgUjMUdAdJW1SaKmiiRlmGKlJ UsUuVyAn2V4uRFgvTGIakp6/giMKdsI9SNIBjBiBDBUIAByK1W5BDxtxBefM8A8hBmEqAlrR4Evh 44SxicIVGVCrvknzJ+3HsBsGBf03td7MyLRgZLtjBQVtvkyhHGMgMVWn+HCCSx7I/8iYkymXd/V0 TnFZ/ps7X/jADCSqAtRRnWpNlKc7npiKjglSDJ51ZlkO8htS/gzhW0Rk1ALVHRc+02tRVumEQj1R 4NexuwgBbMevWdqdiSopDF8Rw1WptZNbuhaximGGTOIVkZCu7mIRhd88yJ858FHRtgLEUiFUJupO QpiEi2UcswQyflalTdpTQ8UxqeaVsVEPm7kVd8cZ7/xsr9qmxKsyq1KMtar4VUskbY2iBWxBdZd9 4lF8fxdGdjxH66aYZt+QDiko4Dooi6aBWcusyatVWDYOKUMjVQSqKDCXgaYCeUW1RCo1UUFkM5Ak ogT5abJG2Sba19Girs8hG8s0cQnC2GGYVtdN7GnpfU2zVoU4xFbTYLiH4X3IZr5L1OdRBIz1ZnZR QoHQVJuGPCYV9T1asdZELBX6lIS2iVCgapiVwlDAyRpG6IKlKtDiubzcbAFjDxh7UBBDZjKzl0BP 9kEpL+0wgFEkLfWMZo5a9TGiqsXXIcweqx2Qsg583YsAu0cI4Boil1ZQ/g4OVaJt42vhJ0JDxEah 00R0HhGOlpeKxNPGeC+S2b/YEICiTnTy90TUqi16xa9FClF/T0bBYEamtH7vzT0wVr6qwnX+G5e9 6e3KjSCIhGlbO8oKtoT9q3996bDRV/jzU3dwKhteda+7F/3UnvQKNcVApd7oN6gLLv3SD9GraKMY wsRoBZux99TJ3jOmkB2EIJhyn96GHXCjiysQGfyA3PUz/DvD/yP4c+Ww81P8P+x/hY912yjPB9Cm nChiDByiAFfwsGCvYm8CwZDYlJbG+PpseHPWM8DAxjpMbfUsgNf/xifLBJf+cadBwvUuQMXMHu4U oECmj5YAgi45GEo1rl9QSCDvxWLR1dyYuokrbrxvvJf3/PDOQ4JP+Rg8v7w5CedziGfZe3H3l/fq sd5/TvvsOvJzz2ydx8AT5EL9PBV7XeuDUb3J6pfXdfHz06jeZlU2+qk/fGPYKfrWmubWdX90e62a 9IcnAFCT4kyXldCWGELMWEUgW5Ub6XdpIz1tj+iwHUds6tzRU9hxJI0sAHK6gCKaDYOPDdB7CMei O34rbgMo4lIItRLfS9rJJsVP5xoGfntT0ghyQ9kIvpUMM8YVFTEUfi1u5Y5jsLHUSnwtbhZxH54J aOJ7cbuF96RGFV9lMwjLgdZZQen/AwRFSoa1bqS2omxR0MvvTzF4UZyG4tk7Obt+x4aXX/q90eW1 aXByydoNsCFop8SeX/NMPG7B4yvxuJV6xoStgEcn9KoLq9EeW7JJV3onrr5wxmAOKqJTFTVELt9V C2yVhFwTTbKwhvZAlvqlBkMQ9UG/Tqa7aIwslKZNOT3/JswfjT3E8dmnKQp8Jgrg9uHPNaTGe/Fs PVFeTJeNpNl28gXzT/+WNg6mr6/e5XB0Nryl17JajYKb+pJyRx97BnCLaKR/NgVm7UYxnC2N4SxU Oxnks/NTiofOByLYEZsScRsyhuxYSIVPq0prskZRIz0/Bf6l9+/Ui1qDv6Xm4EDjhEpfXOf3EYcM QhQOmMKpo53GOb5ZRbqmp2NK0mb1gPXqmcdtR+0LF9l/vFwsfIwR0ztjzbC2mWnlHBBI9/4a5e8h l0gyMJLZrTCLb3s2S4mL+0VczHItte1dY4laRlZoJDCRrxdY5nAAzWlJ215wsjwpTPN8UDs/3WvW P/XrYs9NvTs8qYv9N+h/GwVWFq6P/M6jjVAYM0J+4LMv8i0PGe6nMNX4Hkq6UdA1yMBk1lbCyqS6 blnG7MGf5K2PeNNA8kAbWcmWnMlLtS60iKYFmgZlZTJCUQKwWdAi1QBg2SM4ykDlMdVCkbkPZD1Y a2yz9MibXKtFKaFeRYwcPexfe4XUeT2BDI3g20GJAT7MGeA0QU0o2jeqn5NT9UJG7z7EtyXcbP21 oNJRz3OOUgrUhkEBT482qApiEcgN1q7Y34opO8TjYYJvCudGY1leEGVtJ7Mm+5EX2JdKyhE5KZhV 9T5Y5rpdTF0Zlf1jyZcc0353au3FxGiaajH3rv+UW43DS1VL1lOUSQZaUgomPOVIy1gqrqFd/FTB IrgLy4RmVyrl9YA725Sb8WlHmf3CQCE5bbw6WvGKGvcHN/21U2u+bmppRLeZWpoYG6fWTE2teGLF EIx24GepsRBODJqgxtJ6xHoy4JgpMsgaoKlOkWiKaSbZdTNz/YqCXDp9SQFCTPHX4nZpp2Gu13Cb blavTQTPj9PMWua1w+Tlde0YpvnGAcpFgemsnxhcKgvG/Jc+OCrxC6mxUM9gvPqwj/FPgVVMxdj6 ogDVrhNZ8yp2Ie0P7GNq+woA4XxhthwXvPRUJpEr+B/Xk7pwBR25xazAs/yY9ix5SFk7A4ol3hSV Rr7c2W+r9YylvMhG6ShlbBbVTy9GdQtZfpEgG1R/3CRq+YiuQFT6kJNtjrMLY/MbHmGNY3NkXhKI UBq/dcS+2jJiX+WFUubmJUzAoqodyWPrnMpKEGUljdZWobsEUmRdJBIUM4lC8uZAPUOqLih55Ik9 K/gKnIeLhFMv4rRrgV6ymvIZh2BOvI3g/aZfkbMv9YpVwL8lmv3vrgbvWO8CD4bp7LAdWbjS17D/ z9FNrzvoXr/pvYNH5om6m2+vH305eWdVu3UxrHfS770rug/tWcH93S+pga37lz0L/uDy8spRQ599 vLBAXV33T/vXCq/0nHd3Cm3d8aYyVGtNGapVImzsZWLFjGSJz1L/9QIhY7alOS6yNDlybJfasFdn N/+NBEXS6UWpSVFSIqqCf5Z0hMqK/3cSEZL59VlIbkZ/7vyjRIv1fGQCMdomgXhJ9rAhS9gyxXl5 JvGaNCKXEZQQjRVmD+WNC2SrHKgVZa3hWYmR3SZFKI7cWVmIWRy7s5eG76VZJLPtKYbwmyN4ZhuV 14bmW7BhPWPXR+SE22tj8f+9iBSzUt3FmFZchVG5lcUVRuLbRdX5INCsyasFkmCS8oK0C4FCbXMP Q/Bmeb4m3SUthFm0qaWscrYHhRjoM7Pd0HOm7jVlTc8GYJbJwFpKKAWnFu3SQFVazM/5blvGSEy2 8Ys9/Qo3GNmblbsQO8LU22tyE8caTWPyjUzcR0jbuahHcZz0+QwXPwRHSiVXUz+1Lpdv1NposlXc k13jK2jW2mzM0xpMtvpU2eo1zVvZ5sXQMwZiuBF+RmuHG0dIQyy3CGlAol1usV0hkAJYZjmGOZBn JTDtJcL8HpVUS8VbbQtLjJuxlaqxlL917ZWIvtoapi5jgooCEmNBmyVrz7Req9TKsnyUR1x48lU9 W+XQwLXKDdyLFlObrTRaFgriSCK581ofEBDKFWeHTkBIhWkUFZ6SERhkSoCfLtkIsN7HWgMuwObG 05aKAELDLKVUpILm83yw16zhzHWlUW0DODfOX0YtagDliWijpm4E7HkD1/c+zfCv66VXHCd4/uf+ 9zv/c/+o2T6S5z+3j1oNcf7nUeOv8z//iEuor33+55/3xF7xI30odeYI0P+nB39+z0O3FXoGzxcd ASpP5t61T+betRbv154Amj3q86v6fOmBniTU6rTO3OtImeM3s+c/WOc+jOQ+LjpSyI0f9UuIqfPj MjcQhqw55c/jYxVB55/lwS8ceFajsw3EizRntzXx8kO+4qkP4VMnh6ROrlPH3eRP4MO+8s2jdSfw 5Q7gw37ZM/jyB/AJ0Ue6CGji9SZxAN8ENMrT77Kody8LD9H7Ex+et8VJd/KkeH1KiVbg9GuNO92d D6zytuv7/2nv2rraOJLwe37FLC8IIWEkQRyTkHO4CO14xeC1PWhf8iCJwdYaJEVCsb3HP37r1j19 HYGTrLPnaI7BSH2Z6u7q6qruqq+3dpLE5RsdY0sIlk5E5PY1Fb3GnFDYLRrlt0ojVfFi9ZtTenOK YdpbOzGWtaip8U5F5buhJycLxb07HgVYVsD21/bqKRF4+jk5vZuNPwiNoY59BwTPNdRt1UMIGM3Z 7W0ywjqTGoYt26uZFU4TexR4BgXja3QgsAbUhMSNBS+G3nsIfAIJccemS03vmn255QZOB2szu0cd 81lDsq4S1aexFq4rLx3wNGg8etkaULyMwqukXrmKwcOfkisYorhUYMTRnD6Gedcwjg8I95Z23EfI qB4UjNPwkUK8FShNO5i3Gzy0VlDPZj3AMNcMFWaJbNTPzJh+Fe+5Rw3/CdItfskscCE8eWK5RYIJ K4eprqLCnX4zxiPNo0l+76XSexitOUbJgOvHclxw8CYN507iSwvNR9jfeu0yuks5cUt7zbIEoAz0 4++ckDsoZjDDMLwfMeEYsmC9kAqMwmnMMU/pqmvVVY8B4NNLZhnpqiedxRIn5jmfWQtGz47vYI3H DgvAvAF/jjESHkmFJRcBRUdFYQmWG4y3u0evdxJAHIlpxs8yHjdj28ErpGFlNDlNW+BThi2U5EmI d/EFBqxh3n9V//L2i1mPiWSoIEEgi4ZRKZJWc4pYqwLFapbtBkS76Ho3Nv4vrOIfkQBNq8VkjPDw 8BHVdjrvU8oOLm0FaC2gtLTrb/onl2d/r22/2d5hfC4DPPCrQdumfypsW7PlvbA2xYhZWNCaX/Ni 5RpRBRJBfD5+WFnInPYyg0s7zDjENqeJaU25Gup+BCIgkcTjxWT+IOceBBbabnT25JidjZr1BJWU 6KbROqo1JqOhjgiRVwidmDxwe+FRw60DlynMhbClLq2+H1hwOq4otOVDTUwbtEfUMfWiuJdTvaxp 1GwjrZWAWdKDrMz4Pbi2PScPjDO/mH18Ngapdj9N/i0TuCsQ7DhDUBYIxKTTohIb0BVjpHWhPJoh uj0pXnFgPicmXMV5qk0HxMtZanRpqNGDxxHNjhSXlo/ly3ziKpNgk9rINKPibjZ99wePTYS7Hwd8 iJ1Yhhqhfapgpt8mI1AfPiwZ/AM7fDWizhwX/o0JAoMmOT7xUC2ZZWHslzL4S+47sfVbNpIRxckX iLzlldIFdksoSPmubVdSPIz3Ghr7gKclWhwxgixNhGuUQqGXcdLOsSOMrtS5OneDCAM9gDRjWdox VAmvYyMbPWk5mY7LYDBEF8H148MUVhgE5prMFhPGPWfWkpe4sEUY6dYAHQXY5KaEtYUFClZt0Su5 LYp9noA2dFDPFM89CVaoU5Z7NH7QcjVGVOTb1R3NWjPDT5ShnMXNiaiNOM76rpP3wxvSV0A1fAeC 3NORfubXzO7RUCATWVY2axUaTu4U5iWfnTgajsxJlGeMUMSKRcBhiFE7yTNDIZ0HasfxCaBIGY9B 3o9I79KmGIEZEG3jHaOiDOMVTaFfCJaKRRvLMgUJensLNDLUBsOoPYogE0wbQerVjszQQ3i2H63x Moik4CZr9XepYIQmIZc3eQScEiYYIddT61fT4hNIdRgQ+HoyVfRUELIAC/u+eJiMXWA7eARP8ag0 V8ih5oj0/zADxd+U9o/ApCDG8fdR7KdLg3GEahCYIbutZtqPZz7Dxh8lBC4vLo2TsdGuBnVP4I6S 8EMYyiVwMlsPVt0VhJ+tFkgKjKgC/ywagmwuXXiyzbJ6Phl/WE8MQhqY3EZ9yDhXNI8JxaTa04Ie QWySTjAQQrYu8vNedwvZeH0t98MPKK4Jqg7FYIhjkgOCQjPsW7UyBeCC4q8skYCIsz/ifJoJoNOe 48FYwXPEbUpEVYzbq8VsRCAeY2amz7MVbmozZDcai8u72Xz+iC5SD5hJQ1QymgyQWjnL1MPMkwq4 Vnig1o41cB47tEvHL0AivgMdS21cYM+jml8izFj7tS4czetu/+LkjE8jGugUP1zdPTB0TLHvqkS+ HGOT3UCgwkt16Foff2Nvyy++hWteMmoOQQ4wJfXV3byOGJpfhl8aX0ZfbFWI1ewtyLOl4DPVIM6h Yya8OCNAIe8xKNXW1cnYoZ/x5dVICtStLBqyutOgJIHusSyCZOt2dYNL3RCDCbZI2watfhqbFrSD cFN4UKbDEr+spU4NcB4yLPRsqhor8HE+z5pCkVC9MOJf7RZxe2U5VE25tRf/+Wp0NxnDrISxYgQ4 OVlcGhBYVFFSTEmNhLoZNcntYyB8+aCHZ7hg69e6y2mypLmHMHi0IJhVQGvVIYpahWhFLcGWvy0o 01XW/SpkJiiXCOTS74JpipJHLNuQ6VRFF/O2er2SBPL5KUBQ/ateembTkZ1l16970MCrqwsdoxc8 L0V7tdtDb8b0PH1z1pf/8eS9iz/ZOfymM914B6WXeOwLP1cXF/ibz4pTuhMylZN9vtmzog6BrhrQ US3Wg/BVQN5LJAf+6+LlkvAzqIKuygZ5fGRO3l7100Zyml29BoJ70NRebjhzcuqbk4vuZVbxihL+ Cn5zVXn/VSMZ/CPtQ5UD+oF6B/QTutOSh5GOtdecZp/nl5cnFAwWqKb7SZaaANJWoPW8jyjp3X8B M2ZWHkmveM+b1UjhjpZvCtdEzlpeTV8FDgbfN5DFS5gwGKR/vn4bpDQKFsYVKkywSehiadT5yD3p aDuUXN7G3Uhg8XxXBPKU13M3+HAhkEff123AW5np+vZuSP/tLpQhNzKsQg0xapiEakiNGiahGvTt 3i7i1c0smTAEiweVxSXPa9uNSWN7h8ve4GYVZ0Jwq5uZX40JbmZW1LUqKiIVGSW+03wkgGtqZTkH je2mENcVE/eGR3Sv+HWPTlU89DV15KIotOMxdLIKxBSfUfUC5gVUVVzC+COLaPhi7+IEKlbTgCW2 8bV8f2kUZRlFS1Lr+eHBi4POYbfZ+UFSQRpBkhwbJNuvsGV/w6wv2vsv2i+6zf3nSkqgyIO8eD5i rVfZKb1NOpAuyYP5f4ZaM51nZMctox8zC6NsrAxrMtvxEtcp60+L4n72Gyp66JIxw2gzvKgKJut9 2amyBWUFpio3npYLM0OOyTQKjF7EeD615LqPnvYSREsQe9c5+vDLN8EADbN7vaGGZ6Dez5VYacrp yCNQ6jW+cX2RqxCsVLzxG72h+sr2KfY6i/2UMiIDtR6lfXBg5THqPlA4iPQWwzETj+6uRIvW6227 /ymIgqNK6+cqpHSnjhxVb+8yo6JrsUQpev2uBlfagpS/TJpOJ+qGyh+ut7ZEqTdbbsuDI2ll9jzJ dWeRy7JSDpj80Di2fJdnv0mZOxL8R537xxtxBSHGGPU4SW28Ls3znUAIid4ZOLIwAkN3yq4z8WEh KZ1srENQVXMIopgcDdiVo0aHATueS4e6vxokgcbxt1EKTQf1Xh6Yd71+4EvhVuIA2/O+zcyN2GEe e0moCSoVGn/JCwYiIognUGlkjt/F9+1y+I+dua8ie1HL5MxNzNwMZBaiVfSF59kfJyETErx+MV+d yautfkK91YqNgmIqJApes+PVCL+abLHUqTDN8Kz8bnBVd+EGiVz4tRsop76TMgHOKpHg5sZOiiF8 6KbMY1QPayDzQUGscbN2VRx5k77TIes7yTNfqediaNARSpcv3BD1W6Mk+uLrxFhB6e0Vgkfl5Vo9 WWK0TeT4LouoXt9PaUdTOpyS+ykH0ZTDaG3fe2XEj5ZJM5U3SWjHEpiwzBbeknZQkcbEpX172koi 05fmZu+BKdwXG6RT6yjLE91w2vQPrUgxBcXmU1K+yrfccB0nmhrmKDXMrhTLN1IPO59bzunib+6E 5JQtREmf5iH2G+DUNOhwEnMjteOkZlS2HE27qFnywEs03+pKs4zKluNatRJX1WNT79JgUe8VrWq3 Tb3bcIv6TmDFN00Bf9HVvlkUuFk8OIt3EpK8SivdpU9d+RRgH5U7s3JnTRW6ai94nVKb8xc8kwjZ wzAUOaN20ed2g8ysM5Wxs53o4vUt5alrKMRm0nXf56PrfB0H2/qmKudooabe7+j5F3gft+lThvEl CGd6hXCmp+wuxMeYYwVBNOBrKYBd1fYHfTUer+7pUGxZgU+0d1fs4TaVLpg/qWApa02bCTcKzY+O rUsdVX7KxUDnzwIBeSr6mRkvglVdXeDuK77A+JL2LPFrTLWXDaZEaeFQrTGLuQzmaFJRj0uyGN52 4tjBMEgpmsEBTE3NaQLwPlD2qFC904zhN3M/4W/XxAlXnD++4pwqzh9XcQ1pZovZrZqNbH6zJxVC pEQihkn+hMArwNK+VJa2qszNoyxuzvjy1Ep3bewAsEVPbgYtRzL5mdC9zUwBGycNnas+zZhJSnsm 2Dpl1wQTI/ZNopGakbvVdj5zude/OkrfEN5unicZO8kT7Z3EM3kO1iDVeuRQ23zbJ0QJZ/UsIHhO n2IEJb4ddKrtmbQ0hAKY664dZJWLG0JJyYG80imP7fFqsaArepRrXEAGrV9pE3uxfWwnhDBpYgtx EkFyNkRO7+1ecFtIcvVo3w7EYRR1wJSYQax7U/IFM3CM+3MPTi8MG8D8RaxIBz2xuUMcSCdA4YsO erQm9HJfVMdICmLtoY6ZzxXYvSGfbIUkak8aialOz7+zMhj2VA0X6H057IOfVoU1FbF/EmuVoFjL 8gPToj/Hq+CTsYgN1q6n8k16WVWHWGKwjEhutarIR2WXJYndIRbPhQ0oi+ukVb61QQedtA42zYrC QqBrbtbZ1QS2Qnp5s9d/0lZIEtgNifFBu9bWdvX/DS9UGOXJ47nBG535Z0t5T9GvecABPlSDlT1H fQNUUpTlppMkfbGa6vskqtD25HIB1paxVZHVVWBEZENzFzcL1d/QJ1kYM/ACr1VDbzwDuSkA6kbA QSi98f1N6ZuojNYnXW9f5/qgy6oOdf+mo8clUQQ0yh7IHZTYcmlAt7TrXxJ3t3hqXlxEmSItu4tK UO4AOcAzUH0Feo9iJcmVEuVfC9uvFOfUtI1N0GcPpMULh5smNbD/BrlxFbB1zeKgr26DHuRlHfQ1 Ooz26VpKkG0/U5j/zWQ5Hi6AZT89LIYu4yp7MHxsUfaQc/7Gvih63w/vhBn0nWQlYZvO/h+9i8rz rR58yRaVsK75MDtVsKK7PKkuQ+wuo4w6Fe5NsR1kvicOtqsbw38ElHINz6jeAh2FO1XGW+LEO30i f0TeEiqJnZ8GjbGEuDtlO0toq2BwyWjye3yS/lB9QRDRFNlBSR4/yq64vikYfXjIN5GK5xsFcYzQ v1978NkO92/QGxhD8/woUOH/4oY85fHQeyITrJ//YvuTwrx5P3n3/vGV5DhNf7HqWM4YJfTDhPzD xTVPBaxylzxTMavcIVZ5Czpi9BnvWpzffdYXmpq1PXMujHXK15Y7LhhuDeWEA39Ly4lcXgtvEM9I BcpQwCrjHT6+H87nxXTPvSDCHfPwXCBHLGTX3E1RKN9YhSyfVFuIowdJAGI70UjLMUlhEGpMAobX p8kc10AMEQM2DzWCmXqAMnN/DbCkovll/GIB3THReZzEoJaT+BUCiTH9GZ6N0ML2HwkJb9Wjxted wrGGhG6TIYw1Z4Q9TkiqR5iAKP/UIe79FYa41fpdQ9z+miG+9HpdXe5Vtfj6PR1fB7/Jatbq/FGr Wd9dzfoRUVd6ijlqddVhA8hnckw73j5F/7wZLjUYcIK7oG6oGt2lfQdi+wZWidnioTDgRmDhsABB ynq7UC/m9iPfIhgkjpZI1cheNF/QxsGFwa0i5J1Dg3c8rSftevwm5lFo/M8J7JBsK0jbrdQFWU7H /JqMt78MJBg0VDHVuo1R6rBu1YTQcPyKpwNMPOCv43PCLLxmQlTbQq4h5OUpm35YcVugfHKd5zaA kH/5B++jBNnwZ8I/rsF/3G897+wr/Mfvvz88JPzHw9YG//F/8fBcXaD1dSshAMnybng/fl8bKlky fj9cDMcIMjSUbzhHchx0Yi6mN0ogmLiSsk+Z8OXWWekEdKnCCF41+O5PjS0Z19sY6NtxFEq66Fxx jYFpJ/CKswah2VYFp2A2E4LSRppUASBMMcMoMplCMsXTZJpquQRVUcghGGGIcvcNtLFV32kQCjTV Th8QZrnuhjaZOc6o0Dn97vLvNv0nFUpRJ6pCxuJIKE78y7+XD7N5sg3DXIz+095+xMB2NgP7LQbW HKeOOU7fWq5sns2zeTbP5tk8m2fzbJ7Ns3k2z+bZPJtn82yezbN5Ns/m+bbPfwGVyVUHAMgAAA==