Re: [PATCH] cacheinfo: move cache_setup_acpi to header

[Greg Kroah-Hartman <gregkh@linuxfoundation.org>]

On Tue, May 05, 2026 at 01:08:04AM -0700, Rosen Penev wrote:
> On Tue, May 5, 2026 at 1:04 AM Greg Kroah-Hartman
> <gregkh@linuxfoundation.org> wrote:
> >
> > On Tue, May 05, 2026 at 12:59:36AM -0700, Rosen Penev wrote:
> > > cacheinfo.h already has handling for acpi_get_cache_info to be a static
> > > inline ENOTSUPP function. Move it there as both functions are in the
> > > same boat.
> > >
> > > Signed-off-by: Rosen Penev <rosenp@gmail.com>
> > > ---
> > >  drivers/base/cacheinfo.c  | 5 -----
> > >  include/linux/cacheinfo.h | 8 +++++++-
> > >  2 files changed, 7 insertions(+), 6 deletions(-)
> > >
> > > diff --git a/drivers/base/cacheinfo.c b/drivers/base/cacheinfo.c
> > > index 391ac5e3d2f5..ca7f266729d1 100644
> > > --- a/drivers/base/cacheinfo.c
> > > +++ b/drivers/base/cacheinfo.c
> > > @@ -366,11 +366,6 @@ static inline int cache_setup_of_node(unsigned int cpu) { return 0; }
> > >  int init_of_cache_level(unsigned int cpu) { return 0; }
> > >  #endif
> > >
> > > -int __weak cache_setup_acpi(unsigned int cpu)
> > > -{
> > > -     return -ENOTSUPP;
> > > -}
> > > -
> > >  unsigned int coherency_max_size;
> > >
> > >  static int cache_setup_properties(unsigned int cpu)
> > > diff --git a/include/linux/cacheinfo.h b/include/linux/cacheinfo.h
> > > index c8f4f0a0b874..73ab4a6e3551 100644
> > > --- a/include/linux/cacheinfo.h
> > > +++ b/include/linux/cacheinfo.h
> > > @@ -86,7 +86,6 @@ int early_cache_level(unsigned int cpu);
> > >  int init_cache_level(unsigned int cpu);
> > >  int init_of_cache_level(unsigned int cpu);
> > >  int populate_cache_leaves(unsigned int cpu);
> > > -int cache_setup_acpi(unsigned int cpu);
> > >  bool last_level_cache_is_valid(unsigned int cpu);
> > >  bool last_level_cache_is_shared(unsigned int cpu_x, unsigned int cpu_y);
> > >  int fetch_cache_info(unsigned int cpu);
> > > @@ -106,7 +105,14 @@ int acpi_get_cache_info(unsigned int cpu,
> > >  {
> > >       return -ENOENT;
> > >  }
> > > +
> > > +static inline
> > > +int cache_setup_acpi(unsigned int cpu) {
> > > +     return -ENOTSUPP;
> > > +};
> > > +
> > >  #else
> > > +int cache_setup_acpi(unsigned int cpu);
> > >  int acpi_get_cache_info(unsigned int cpu,
> > >                       unsigned int *levels, unsigned int *split_levels);
> > >  #endif
> > > --
> > > 2.54.0
> > >
> > >
> >
> > Didn't checkpatch complain about this change?
> Hrm looks like it. I just matched the style of surrounding code.

Nope, look at the style of the inline function just above the one you
added.  You used a totally different (i.e. not kernel) style.

> Not sure about this though:
> WARNING: ENOTSUPP is not a SUSV4 error code, prefer EOPNOTSUPP

That's not the issue here.

> > And why get rid of the weak symbol, this should break some arch builds,
> > right?
> The contrary. Fixes compilation with
> 
> make LLVM=1 ARCH=mips

So what commit caused this to break?  Is this a new issue or something
that has always been present?

thanks,

greg k-h