Messages in this thread | | | Date | Tue, 5 Jun 2012 19:22:08 +0200 | From | Sam Ravnborg <> | Subject | Re: [PATCH] x86,nmi: Fix section mismatch warnings on 32-bit |
| |
On Tue, Jun 05, 2012 at 05:18:40PM +0100, Jan Beulich wrote: > >>> On 05.06.12 at 18:13, Sam Ravnborg <sam@ravnborg.org> wrote: > > On Tue, Jun 05, 2012 at 10:47:34AM -0400, Don Zickus wrote: > >> On Tue, Jun 05, 2012 at 09:12:13AM +0100, Jan Beulich wrote: > >> > >>> On 04.06.12 at 21:56, Don Zickus <dzickus@redhat.com> wrote: > >> > > It was reported that compiling for 32-bit caused a bunch of section > >> > > mismatch warnings: > >> > > > >> > > VDSOSYM arch/x86/vdso/vdso32-syms.lds > >> > > LD arch/x86/vdso/built-in.o > >> > > LD arch/x86/built-in.o > >> > > WARNING: arch/x86/built-in.o(.data+0x5af0): Section mismatch in reference > >> > > from the variable > >> > > test_nmi_ipi_callback_na.10451 to the function > >> > > >> > Did you check what this (compiler introduced) variable actually > >> > represents? The problem clearly is that the compiler has no way > >> > of knowing that data it generates referencing an __init function > >> > would actually need to go into .init.data or alike. > >> > >> I do not know enough about behind the scene compiler magic to track this > >> down correctly. So any insight/help is greatly appreciated. :-) > > > > We register a function annotated __init. And the normal use of this > > register() > > is for functions that survive the init phase. > > So drop the __init annotation on the function we register and the mismatch > > is fixed. > > But in the case here the function really is being used at init time > only. This is only for testing. Ando nly if CONFIG_DEBUG_NMI_SELFTEST is set.
> Plus it still escapes me why there would be differences in > behavior depending on bit-ness or compiler version. It is often different inline decisions that has influence.
Sam
| |