From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <pip@iszf.irk.ru>
X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on sa.int.altlinux.org
X-Spam-Level: 
X-Spam-Status: No, score=-1.5 required=5.0 tests=BAYES_00,DNS_FROM_OPENWHOIS
	autolearn=no version=3.2.5
From: Valery Pipin <pip@iszf.irk.ru>
Organization: ISTP
To: ALT Linux Team development discussions <devel@lists.altlinux.org>
Date: Mon, 4 Apr 2011 21:20:19 -0700
User-Agent: KMail/1.13.6 (Linux/2.6.38-std-def-alt1; KDE/4.6.1; x86_64; ; )
References: <201104042044.11351.pip@iszf.irk.ru>
	<4D9A93A3.4020305@mmedia2.kemsu.ru>
In-Reply-To: <4D9A93A3.4020305@mmedia2.kemsu.ru>
MIME-Version: 1.0
Content-Type: Text/Plain;
  charset="koi8-r"
Content-Transfer-Encoding: 8bit
Message-Id: <201104042120.19252.pip@iszf.irk.ru>
Subject: Re: [devel] atlas is broken again?
X-BeenThere: devel@lists.altlinux.org
X-Mailman-Version: 2.1.12
Precedence: list
Reply-To: ALT Linux Team development discussions <devel@lists.altlinux.org>
List-Id: ALT Linux Team development discussions <devel.lists.altlinux.org>
List-Unsubscribe: <https://lists.altlinux.org/mailman/options/devel>,
	<mailto:devel-request@lists.altlinux.org?subject=unsubscribe>
List-Archive: <http://lists.altlinux.org/pipermail/devel>
List-Post: <mailto:devel@lists.altlinux.org>
List-Help: <mailto:devel-request@lists.altlinux.org?subject=help>
List-Subscribe: <https://lists.altlinux.org/mailman/listinfo/devel>,
	<mailto:devel-request@lists.altlinux.org?subject=subscribe>
X-List-Received-Date: Tue, 05 Apr 2011 04:20:26 -0000
Archived-At: <http://lore.altlinux.org/devel/201104042120.19252.pip@iszf.irk.ru/>
List-Archive: <http://lore.altlinux.org/devel/>
List-Post: <mailto:devel@altlinux.org>

В сообщении от 4 апреля 2011 20:59:31 автор REAL написал:
> 05.04.2011 10:44, Valery Pipin пишет:
> > In [1]: from numpy.random import *
> > 
> > In [2]: from numpy.linalg import *
> > 
> > In [3]: m=randn(2000,2000)
> > 
> > In [4]: %time U,s,V=svd(m)
> > Ошибка сегментирования
> 
> NumPy нужно пересобрать с новым atlas. Сейчас сделаем.
Разве это поможет? Помоему numpy просто использует внешнюю библиотеку и все
Например по совету со scipy.org:
 ldd /usr/lib64/python2.6/site-packages/numpy/linalg/lapack_lite.so
        linux-vdso.so.1 =>  (0x00007fffc555e000)
        liblapack.so.3 => /usr/lib64/liblapack.so.3 (0x00007fe4b85a0000)
        libpython2.6.so.1.0 => /usr/lib64/libpython2.6.so.1.0 
(0x00007fe4b81f6000)
        libc.so.6 => /lib64/libc.so.6 (0x00007fe4b7e8d000)
        libblas.so.3 => /usr/lib64/libblas.so.3 (0x00007fe4b7c6e000)
        libgfortran.so.3 => /usr/lib64/libgfortran.so.3 (0x00007fe4b7987000)
        libm.so.6 => /lib64/libm.so.6 (0x00007fe4b7705000)
        libgcc_s.so.1 => /lib64/libgcc_s.so.1 (0x00007fe4b74ef000)
        libpthread.so.0 => /lib64/libpthread.so.0 (0x00007fe4b72d2000)
        libdl.so.2 => /lib64/libdl.so.2 (0x00007fe4b70ce000)
        libutil.so.1 => /lib64/libutil.so.1 (0x00007fe4b6ecb000)
        /lib64/ld-linux-x86-64.so.2 (0x00007fe4b9031000)
        libatlas.so.3 => /usr/lib64/libatlas.so.3 (0x00007fe4b68a6000)