|
轉自 http://blog.sina.com.cn/s/blog_3ebc3c2901018v59.html 最近兩天被MatLab攪得是雞飛狗跳。先是在Mac下的2012a莫名其妙的啟動之后，CPU占有率飆升超過100%，完全沒法用，連用個logo命令都不得安生。無奈重裝。同樣問題。煩！暴走回Windows。今天安裝完Windows的2011a版之后發現有這樣一個配置說明：
1. The following products require a supported
compiler:
MATLAB Compiler 4.15
MATLAB Builder NE 4.0
MATLAB Builder JA 2.2.2
MATLAB Builder EX 2.0
Stateflow 7.7
Simulink Coder 8.0
xPC Target 5.0
MATLAB Coder 2.0
2. Simulink requires a C compiler for simulation acceleration,
model reference, and MATLAB Function Block capabilities. It is
recommended that you install a supported compiler on your
machine.
3. To accelerate computations with the following products, a
supported compiler is required:
SimBiology 3.4
Fixed-Point Toolbox 3.3
4. To configure Real-Time Windows Target you must type rtwintgt
-setup in a MATLAB command window.
5. After this installation is complete, you should continue with
your configuration of the MATLAB Distributed Computing Server as
outlined in the instructions obtained from www./distconfig.
6. You have installed xPC Target. Use the xPC Target Explorer
GUI (type xpcexplr in the MATLAB Command Window) to correctly
configure xPC Target with a valid compiler pathname. For further
help, type doc xpcexplr.
7. Run mbuild -setup from MATLAB before using MATLAB Compiler or
builder products.
乍一看還以為是編譯器出了問題。趕緊上網搜了一搜。MathWorks的網頁上有這樣一份說明。原文鏈接如下：
http://www./support/compilers/R2012a/win32.html
Supported and Compatible Compilers – Release
2012a
A number of MathWorks products or product features require that
you have a third-party compiler installed on your system. The
tables below outline the compilers that are supported by various
MathWorks products. These compilers are provided by a number of
vendors and are available under a variety of commercial, academic,
or open source terms; visit the providers' Web sites for further
information.
Windows
(32-bit)
On 32-bit Windows, the lcc C compiler is installed along with
MATLAB, providing out-of-the-box support for most MathWorks
products. Further options are available as outlined in this
table.
MATLAB Product Family – Release 2012a
Simulink Product Family – Release 2012a
Notes for the Windows (32-bit) Platform
- Support for this version of this
compiler will be discontinued in a future release, at which time a
new version will be supported. Consult the platform road map for more information.
- Support for C++ exception handling is
limited. You can find more information in
Solution 1-4OKNSV.
- Intel compilers depend on tools provided by Microsoft. The
following combinations are supported:
|
Microsoft
Visual Studio 2010
Professional (32-bit) |
Microsoft
Windows SDK 7.1 |
Microsoft
Visual Studio 2008 SP1
Professional Edition (32-bit) |
Microsoft
Visual Studio 2008
Shell
(32-bit) |
| Intel Visual C++ Composer XE 2011 |
√ |
√ |
√ |
|
| Intel Visual Fortran Composer XE 2011 |
√ |
√ |
√ |
√ |
- To build .NET components, a Microsoft
.NET Framework must be installed. The .NET Framework v3.0 does not
contain a framework-specific compiler; compatible components can be
built using the v2.0 compiler. The .NET Framework is automatically
installed by Visual Studio. It can also be downloaded from the
Microsoft Web site. To execute applications that use the resulting
.NET components, the target machine must have the matching .NET
Framework installed.
- MATLAB Builder NE supports building
.NET assemblies but not COM objects when using the Microsoft .NET
Framework SDK without Microsoft Visual Studio.
- Fortran compilers are supported with
Simulink only for creating Simulink S-Functions using the MATLAB
MEX command. The S-Functions can be used with normal and
accelerated simulations.
- xPC Target supports Fortran code in
Simulink models using C-MEX wrapper S-Functions.
- .sln project generation is not
supported when using the Microsoft Windows SDK.
MathWorks的意思其實就是說，如果你的Windows電腦裝了Visual C++
2008或者Visual C++ 2010，再加上Java Development
Kit（JDK）1.6的話，那編譯MatLab上的任何文件都沒有問題了。
同樣的情況也有Mac系統的說明，就是說Xcode +
JDK可以解決一切MatLab所需要的編譯活兒了。
所以就算出現了前面的說明也不用緊張，沒有VC++、JDK下載一個就是了。
|
