On quinta-feira, 1 de agosto de 2013 19:38:12, Laszlo Papp wrote: > 3 This discussion is going off on a tangent but...
you do realise that there's an overhead associated with dynamic libraries, right? The code runs slightly slower due to indirect access to functons and data, and there's a start-time impact due to the relocation and dynamic linking. Even if you use prelink, the problems do not go away completely. I find it hard to believe that you have a device with "small NOR/NAND flash" (your words), a very quick processor and large amounts of RAM to cache everything. Without knowing the specifics, I find the situation inconsistent. You're probably better off statically linking everything into one huge binary, using the Qt feature system to drop features from the build. Remember to drop the - fPIC flag when building Qt too, to improve code generation. Please take a look at this email from almost exactly one year ago and the thread it belongs to for more information: http://lists.qt-project.org/pipermail/development/2012-July/005428.html > > On Thu, Aug 1, 2013 at 7:25 PM, Hausmann Simon <[email protected]>wrote: > > Interesting point. > > > > In your environment where you are already building your custom version > > > > of Qt, how many processes do you expect to be running simultaneously that > > are also using QtSceneGraph (without QtQml/QtQuick)? -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ Development mailing list [email protected] http://lists.qt-project.org/mailman/listinfo/development
