On Wed, 21 Apr 2010, Amerigo Wang wrote: > > CC arch/um/sys-i386/elfcore.o > >In file included from /data/linux-2.6/include/linux/elf.h:8, > > from /data/linux-2.6/arch/um/sys-i386/elfcore.c:2: > >/data/linux-2.6/arch/um/sys-i386/asm/elf.h:78: warning: ‘struct task_struct’ > >declared inside parameter list > >/data/linux-2.6/arch/um/sys-i386/asm/elf.h:78: warning: its scope is only > >this definition or declaration, which is probably not what you want > > > >I guess not many people build against i386 hosts anymore, so this > >remained widely unnoticed. > > > >> > >> If it built before, without having a task_struct declaration, I think > >> that means that the elf_core_copy_fpregs was never used. The > >> task_struct * in the declaration would become a private task_struct, > >> known only to the declaration. If the implementation or callers have > >> the regular task_struct, it will be a different one, and the > >> prototypes will conflict due to the different types of the first > >> parameter. > > > >This is just a forward declaration (that many arch elf header include), > >so no such problem exists. > > > >BTW, to answer the other question in this thread: We have a circular > >dependency that prevents including sched.h. > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > This is the right reason to do this. Ok then, thanks. > > But it looks like x86_64 needs this too. > > > BTW, I don't think compile warning fixes are trivial enough to go > to triv...@kernel.org.
Why? -- Jiri Kosina SUSE Labs, Novell Inc. ------------------------------------------------------------------------------ _______________________________________________ User-mode-linux-devel mailing list User-mode-linux-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
