Quote Originally Posted by schlett_tamas View Post
But theoretically, if i can tell for the armeabi compiler the locaton of the qwt headers, that should be work??
Of course - setting this path is done in the project files of the examples and is part of what you include with "CONFIG += qwt". As you have disabled the silent mode now you should be able to see what path gets added when looking at the commands used for compiling.

I ask that because i compiled the qwt with mingw and i got .dll-s in the qwt6.1.1\lib, so under unix based os's cant handle theese kind of files so maybe thats why armeabi cant see the headers.
No for compiling you don't need any dll's ( those are for linking ), so this can't be the issue. But of course there shouldn't be any dlls when cross compiling - if you have them your build of Qwt is not a cross compiled one.

Maybe i should compile qwt under linux and get the .so files and copy that to windows, and after that under windows i should link that .so files to armaebi compiler?
No all you need to do is to do a "qmake -spec xxx qwt.pro", where xxx is the spec file with all the settings for the target device ( the same specs you have used for cross compiling your standalone application ).

Uwe