[maemo-developers] [maemo-developers] Re: gtkmm code size
From: Murray Cumming murrayc at murrayc.comDate: Thu May 11 12:36:48 EEST 2006
- Previous message: [maemo-developers] Platform Java on the 770 status
- Next message: [maemo-developers] Re: gtkmm code size
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Thu, 2006-05-11 at 11:16 +0200, dodji Seketeli wrote: > Hello, > > I wonder how the default symbols visibilty directives would affect the > total code size of gtkmm. > > In other words, if we say (like on win32) symbols are *not* exported by default, > and we explicitely flag the exported symbols, do you think we will > gain something ? > > The new __attribute__ ((visibility("default"))) and __attribute__ > ((visibility("hidden"))) > directives of gcc make this technique more useable than the older way > (maintain a > separate file with the list of the exported symbols) that was bug prone. > > Maybe we have to think about how to integrate this thing in the coming versions > of gtkmm. Feel free to try, but I am very afraid of missing even one symbol and causing big problems for existing applications. I'd much prefer to export everything by default and mark some things as not-exported. I did try to hack libtool to allow me to use regex to exclude anything with "Anonymous" in the name, to exclude anonymous namespaces, but it made no difference to code size, though the symbols no longer showed up in the nm output. Obviously there's more to it than that. > Checkout the documentation at http://gcc.gnu.org/wiki/Visibility . -- Murray Cumming murrayc at murrayc.com www.murrayc.com www.openismus.com
- Previous message: [maemo-developers] Platform Java on the 770 status
- Next message: [maemo-developers] Re: gtkmm code size
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]