Re: [brauner-github:vfs-7.2.misc 28/28] fs/open.c:112:29: warning: converting the result of '<<' to a boolean always evaluates to true

[Jeff Layton <jlayton@kernel.org>]

On Sat, 2026-06-06 at 21:14 -0700, Nathan Chancellor wrote:
> On Sun, Jun 07, 2026 at 10:49:04AM +0800, kernel test robot wrote:
> > Hi Jori,
> > 
> > FYI, the error/warning was bisected to this commit, please ignore it if it's irrelevant.
> > 
> > tree:   https://github.com/brauner/linux.git vfs-7.2.misc
> > head:   0da79c259ad0554b36761a7135d4f92eb7c46263
> > commit: 0da79c259ad0554b36761a7135d4f92eb7c46263 [28/28] vfs: uapi: retire octal and hex numbers in favor of (1 << n) for O_ flags
> > config: powerpc64-randconfig-001-20260607 (https://download.01.org/0day-ci/archive/20260607/202606071029.DKCs8WOs-lkp@intel.com/config)
> > compiler: clang version 23.0.0git (https://github.com/llvm/llvm-project 7917772d7d61384696c61102c08c2ea158e610fa)
> > reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20260607/202606071029.DKCs8WOs-lkp@intel.com/reproduce)
> > 
> > If you fix the issue in a separate patch/commit (i.e. not just a new version of
> > the same patch/commit), kindly add following tags
> > > Reported-by: kernel test robot <lkp@intel.com>
> > > Closes: https://lore.kernel.org/oe-kbuild-all/202606071029.DKCs8WOs-lkp@intel.com/
> > 
> > All warnings (new ones prefixed by >>):
> > 
> > > > fs/open.c:112:29: warning: converting the result of '<<' to a boolean always evaluates to true [-Wtautological-constant-compare]
> >      112 |         error = break_lease(inode, O_WRONLY);
> >          |                                    ^
> >    include/uapi/asm-generic/fcntl.h:20:21: note: expanded from macro 'O_WRONLY'
> >       20 | #define O_WRONLY        (1 << 0)
> >          |                            ^
> 
> Hmmm, this appears to be pointing out an inconsistency between the
> different versions of break_lease(), depending on whether
> CONFIG_FILE_LOCKING is set or not.  The CONFIG_FILE_LOCKING=n stub has a
> second parameter of 'bool wait' whereas the CONFIG_FILE_LOCKING=y case
> has a second parameter of 'unsigned int mode'. 4be9f3cc582a ("filelock:
> rework the __break_lease API to use flags") changed this, which seems
> incorrect to me.

Good catch. I missed updating the stub when CONFIG_FILE_LOCKING=n. I'll
send Christian a follow-on patch to fix that.

Maybe this is another sign that we should just kill CONFIG_FILE_LOCKING
and just always enable it? It's not like it increases the size of the
binary much, and I imagine only the tiniest embedded systems disable
it.

Thanks,
-- 
Jeff Layton <jlayton@kernel.org>