您查看的文章来源于http://www.oklinux.cn
曾经试过很多次，都没有成功，连安装的界面都进不去。原因是在64位的机器上装了64位的系统，但是matlab的各种版本似乎都对64bit非常反感，只好作罢。毕竟那个时候还有Windows嘛，在Linux下装matlab只是兴趣使然。
昨天重新安装了Ubuntu系统，彻底的没有Windows了。今天尝试了很久，终于装上了。简单记录下过程，今后重装还有参考
1、一定装7.0及以上版本。6.5的能进入安装界面，但是在验证key文件的时候总是过不去。网上有人提出了修改系统时间的办法，试过了，未遂；尝试7.0，没有问题；
2、安装过程很顺利，运行时会出现比较大的问题：simulink无法启动，而且输入的命令会直接导致程序崩溃，ctrl c都无法响应，只能用kill －9的办法。网上提供了一套解决方案，copy如下：
simulink有问题，下面是各种错误信息的解决方法。
??? Can't load '/usr/local/matlab/bin/glnx86/libmwsimulink.so': libXft.so.1: cannot open shared object file: No such file or directory
Solution:
[root@localhost matlab]# updatedb && locate libXft.so
/usr/lib/libXft.so
/usr/lib/libXft.so.2
/usr/lib/libXft.so.2.1.2
[root@localhost matlab]# ln -s /usr/lib/libXft.so.2 /usr/lib/libXft.so.1
然后会出现
??? Can't load '/usr/local/matlab/bin/glnx86/libmwsimulink.so':
/usr/local/matlab/bin/glnx86/libqt-mt.so.3: undefined symbol: XftFreeTypeOpen
Solution :
# install qt 3.3
# mv /usr/local/matlab/bin/glnx86/libqt-mt.so.3 /usr/local/matlab/bin/glnx86/libqt-mt.so.3_back
# ln -s /usr/lib/libqt-mt.so.3.3 /usr/local/matlab/bin/glnx86/libqt-mt.so.3
然后又会出现
/usr/local/matlab/bin/glnx86/../../sys/os/glnx86/libgcc_s.so.1:
version `GCC_4.2.0' not found (required by /usr/lib/libstdc .so.6)
或者
/usr/local/matlab/bin/glnx86/../../sys/os/glnx86/libgcc_s.so.1: version `GCC_3.3' not found (required by /usr/lib/libstdc .so.6)
Solution :
# find libgcc_s.so.1 for 3.3 or 4.2
or a one for all way
Matlab was built with a different version of gcc (3.2.3 in the case of R14SP2), whereas Fedora Core 5 has gcc 4.1.0. The following trick is alluded to in the file ${MATLAB}/sys/os/glnx86/README.libstdc , but you need to be the root user. Type these commands ...
# cd $MATLAB
# cd sys/os/glnx86
# mkdir old
# mv libstdc .* libg2c.* libgcc_s* old
When matlab is restarted, it will then use the system libraries instead of those supplied by Mathworks. And then all is fine and well with the world. Note that if this sort of thing happens with an x86-64 machine, then the directory in question would be ${MATLAB}/sys/os/glnxa64 instead of ${MATLAB}/sys/os/glnx86.
还有一个搜索到的：
