- 新着としてマーク
- ブックマーク
- 購読
- ミュート
- RSS フィードを購読する
- ハイライト
- 印刷
- 不適切なコンテンツを報告
Hi, I'm student of Civil Engineering. I'm trying to compile a fortran 2003 in my linux system. I'm using Intel Fortran Composer XE 2011 for Linux, but the terminal says:
felipe@felipe-desktop:~/Documentos/LOSA PROYECTO FINAL DE CARRERA$ ifort Euclides.f03 -o a.out
ipo: warning #11010: file format not recognized for Euclides.f03
ld:Euclides.f03: file format not recognized; treating as linker script
ld:Euclides.f03:1: syntax error
How can I compile fortran 2003 file? With FORTRAN 90, I don't problems.
Thanks,
Regards
Felipe Rodriguez Fonte
Las Palmas, Spain
felipe@felipe-desktop:~/Documentos/LOSA PROYECTO FINAL DE CARRERA$ ifort Euclides.f03 -o a.out
ipo: warning #11010: file format not recognized for Euclides.f03
ld:Euclides.f03: file format not recognized; treating as linker script
ld:Euclides.f03:1: syntax error
How can I compile fortran 2003 file? With FORTRAN 90, I don't problems.
Thanks,
Regards
Felipe Rodriguez Fonte
Las Palmas, Spain
1 解決策
- 新着としてマーク
- ブックマーク
- 購読
- ミュート
- RSS フィードを購読する
- ハイライト
- 印刷
- 不適切なコンテンツを報告
ifortdoes not recognize the .f03 extension and does not offer an option on Linux to instruct it to treat that extension accordingly. You will need rename the file to something like, Euclides.f90 and compile with:
ifort Euclides.f90 -o a.out
ifort Euclides.f90 -o a.out
コピーされたリンク
12 返答(返信)
- 新着としてマーク
- ブックマーク
- 購読
- ミュート
- RSS フィードを購読する
- ハイライト
- 印刷
- 不適切なコンテンツを報告
ifortdoes not recognize the .f03 extension and does not offer an option on Linux to instruct it to treat that extension accordingly. You will need rename the file to something like, Euclides.f90 and compile with:
ifort Euclides.f90 -o a.out
ifort Euclides.f90 -o a.out
- 新着としてマーク
- ブックマーク
- 購読
- ミュート
- RSS フィードを購読する
- ハイライト
- 印刷
- 不適切なコンテンツを報告
But, Can I write the code in fortran 2003 and then compiling with extension .f90, or have I change the code at FORTRAN 90?
- 新着としてマーク
- ブックマーク
- 購読
- ミュート
- RSS フィードを購読する
- ハイライト
- 印刷
- 不適切なコンテンツを報告
Yes, you can use the .f90 extension with Fortran 2003 language features. The .f90 extension implies that Fortran free-form source rules apply not that the contents conform only to Fortran 90 language standards.
- 新着としてマーク
- ブックマーク
- 購読
- ミュート
- RSS フィードを購読する
- ハイライト
- 印刷
- 不適切なコンテンツを報告
Hi
Perhaps I should have mentioned this link: http://software.intel.com/en-us/forums/showthread.php?t=72203
where Steve mentions f90 => free form.
On a jovial note: "industry standard"? Absoft editor gives default extension .f95 to files of type f95/f90. Same is true with SilverFrost FTN95....Go IVF!
Abhi
Perhaps I should have mentioned this link: http://software.intel.com/en-us/forums/showthread.php?t=72203
where Steve mentions f90 => free form.
On a jovial note: "industry standard"? Absoft editor gives default extension .f95 to files of type f95/f90. Same is true with SilverFrost FTN95....Go IVF!
Abhi
- 新着としてマーク
- ブックマーク
- 購読
- ミュート
- RSS フィードを購読する
- ハイライト
- 印刷
- 不適切なコンテンツを報告
Well, my mistake, my apologies to everyone. -Tf works and assumes fixed-form by default so -free is needed if the file contains free-form source.
The 11.1 User Guidedoes not listLinux/Mac OS support under the /source equivalent that /Tf redirects one to, but the new Fortran Composesr XE 2011 User Guide does and /source now refers one back /Tf. Anyway, the option works, sorry for the oversight.
The 11.1 User Guidedoes not listLinux/Mac OS support under the /source equivalent that /Tf redirects one to, but the new Fortran Composesr XE 2011 User Guide does and /source now refers one back /Tf. Anyway, the option works, sorry for the oversight.
- 新着としてマーク
- ブックマーク
- 購読
- ミュート
- RSS フィードを購読する
- ハイライト
- 印刷
- 不適切なコンテンツを報告
Here is a minor complaint regarding options and how they are handled.
The Intel compiler has, over the years, accumulated a large number or options (about 700 options now!), almost on a par with GCC in that regard. Of those options, some have no potency anymore but have been retained to avoid breaking old makefiles. And, some of us imperfect users may look up one of those options in an Intel Fortran User Manual that is not the one that matches the compiler issue that we are using.
It is frustrating to try an option that the manual leads us to believe might solve a problem with our code, only to find after many perplexing runs that the option did nothing.
One of the good features of the Intel compiler is that we can type "ifort -help" and obtain a listing of the options and a brief description of the meaning of each option. This leads to my request: please, please (i) flag those options that are accepted but have no consequence, and (ii) if a do-nothing option is chosen for a compilation, have the compiler issue a warning that an inactive option has been chosen, with the additional feature that if warnings have been turned off even this requested warning is not displayed.
Thanks.
The Intel compiler has, over the years, accumulated a large number or options (about 700 options now!), almost on a par with GCC in that regard. Of those options, some have no potency anymore but have been retained to avoid breaking old makefiles. And, some of us imperfect users may look up one of those options in an Intel Fortran User Manual that is not the one that matches the compiler issue that we are using.
It is frustrating to try an option that the manual leads us to believe might solve a problem with our code, only to find after many perplexing runs that the option did nothing.
One of the good features of the Intel compiler is that we can type "ifort -help" and obtain a listing of the options and a brief description of the meaning of each option. This leads to my request: please, please (i) flag those options that are accepted but have no consequence, and (ii) if a do-nothing option is chosen for a compilation, have the compiler issue a warning that an inactive option has been chosen, with the additional feature that if warnings have been turned off even this requested warning is not displayed.
Thanks.
