[PATCH V4 4/4] LoongArch: Enable ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP

[Arnd Bergmann]

On Tue, Jul 5, 2022 at 9:51 AM Muchun Song <songmuchun at bytedance.com> wrote:
>
> On Tue, Jul 5, 2022 at 2:22 PM Huacai Chen <chenhuacai at kernel.org> wrote:
> >
> > Hi, Arnd,
> >
> > On Mon, Jul 4, 2022 at 8:18 PM Arnd Bergmann <arnd at arndb.de> wrote:
> > >
> > > On Mon, Jul 4, 2022 at 1:25 PM Huacai Chen <chenhuacai at loongson.cn> wrote:
> > > > To avoid the following build error on LoongArch we should include linux/
> > > > static_key.h in page-flags.h.
> > > >
> > > > In file included from ./include/linux/mmzone.h:22,
> > > > from ./include/linux/gfp.h:6,
> > > > from ./include/linux/mm.h:7,
> > > > from arch/loongarch/kernel/asm-offsets.c:9:
> > > > ./include/linux/page-flags.h:208:1: warning: data definition has no
> > > > type or storage class
> > > > 208 | DECLARE_STATIC_KEY_MAYBE(CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON,
> > > > | ^~~~~~~~~~~~~~~~~~~~~~~~
> > > > ./include/linux/page-flags.h:208:1: error: type defaults to 'int' in
> > > > declaration of 'DECLARE_STATIC_KEY_MAYBE' [-Werror=implicit-int]
> > > > ./include/linux/page-flags.h:209:26: warning: parameter names (without
> > > > types) in function declaration
> > >
> > > I wonder if page_fixed_fake_head() should be moved out of line to avoid
> > > this, it's already nontrivial here, and that would avoid the static key
> > > in a central header.
> > I have some consideration here. I think both inline function and
> > static key are instruments to make things faster, in other words,
> > page_fixed_fake_head() is a performance critical function. If so, it
> > is not suitable to move it out of line.
>
> +1
>
> The static key is an optimization when HVO is disabled.

How about splitting up linux/page_flags.h so the static_key header is
only included
in those places that use one of the inline functions that depend on it?

       Arnd