On Jul 30, 2012, at 10:34 AM, ext Thiago Macieira <[email protected]> wrote:
> On domingo, 29 de julho de 2012 10.12.50, [email protected] wrote: >>> After changed with _protected_ visibility, that kind of relocation is >>> reduced, but I still don't know why more R_ARM_RELATIVE relocation >>> introduced. >> Answer my own question, that is because the loading address of the module >> needs to be added to know actual address of each virtual functions. >> >> So for the qt(5), should we change all the exported symbol 's visibility to >> _protected_ ? Or is there still some exited use case to use _default_ >> visibility ? > > I have a pending patch that turns all Qt relocations to "protected". However, > it cannot be enabled by default since it often runs into bugs in the linker. > Protected visibility seems not to be tested properly, as the compiler and > linker people have different interpretations on how it should be used. Hope they can agree at some point, so we can start using it. Until then it would be great if we could have it as a configure option (off by default). Cheers, Lars _______________________________________________ Development mailing list [email protected] http://lists.qt-project.org/mailman/listinfo/development
