Re: [PATCH 4/4] pcpcntr: remove percpu_counter_sum_all()

[Dave Chinner <david@fromorbit.com>]

On Thu, Mar 16, 2023 at 03:22:31AM +0800, kernel test robot wrote:
> Hi Dave,
> 
> Thank you for the patch! Yet something to improve:

No, ithere is nothing wrong with my patch series, this is something
for _you_ to improve.

> [auto build test ERROR on linus/master]
> [also build test ERROR on v6.3-rc2 next-20230315]
> [cannot apply to dennis-percpu/for-next]
> [If your patch is applied to the wrong git tree, kindly drop us a note.
> And when submitting patch, we suggest to use '--base' as documented in
> https://git-scm.com/docs/git-format-patch#_base_tree_information]
> 
> url:    https://github.com/intel-lab-lkp/linux/commits/Dave-Chinner/cpumask-introduce-for_each_cpu_or/20230315-165202
> patch link:    https://lore.kernel.org/r/20230315084938.2544737-5-david%40fromorbit.com
> patch subject: [PATCH 4/4] pcpcntr: remove percpu_counter_sum_all()
> config: i386-randconfig-a005 (https://download.01.org/0day-ci/archive/20230316/202303160333.XqIRz3JU-lkp@intel.com/config)
> compiler: gcc-11 (Debian 11.3.0-8) 11.3.0
> reproduce (this is a W=1 build):
>         # https://github.com/intel-lab-lkp/linux/commit/8360dcb55f1eb08fe7a1f457f3b99bef8e306c8b
>         git remote add linux-review https://github.com/intel-lab-lkp/linux
>         git fetch --no-tags linux-review Dave-Chinner/cpumask-introduce-for_each_cpu_or/20230315-165202
>         git checkout 8360dcb55f1eb08fe7a1f457f3b99bef8e306c8b
>         # save the config file
>         mkdir build_dir && cp config build_dir/.config
>         make W=1 O=build_dir ARCH=i386 olddefconfig
>         make W=1 O=build_dir ARCH=i386 SHELL=/bin/bash drivers/hwmon/ fs/xfs/
> 
> If you fix the issue, kindly add following tag where applicable
> | Reported-by: kernel test robot <lkp@intel.com>
> | Link: https://lore.kernel.org/oe-kbuild-all/202303160333.XqIRz3JU-lkp@intel.com/
> 
> All errors (new ones prefixed by >>):
> 
>    In file included from include/linux/string.h:5,
>                     from include/linux/uuid.h:11,
>                     from fs/xfs/xfs_linux.h:10,
>                     from fs/xfs/xfs.h:22,
>                     from fs/xfs/xfs_super.c:7:
>    fs/xfs/xfs_super.c: In function 'xfs_destroy_percpu_counters':
> >> fs/xfs/xfs_super.c:1079:16: error: implicit declaration of function 'percpu_counter_sum_all'; did you mean 'percpu_counter_sum'? [-Werror=implicit-function-declaration]
>     1079 |                percpu_counter_sum_all(&mp->m_delalloc_blks) == 0);
>          |                ^~~~~~~~~~~~~~~~~~~~~~
>    include/linux/compiler.h:77:45: note: in definition of macro 'likely'
>       77 | # define likely(x)      __builtin_expect(!!(x), 1)
>          |                                             ^
>    fs/xfs/xfs_super.c:1078:9: note: in expansion of macro 'ASSERT'
>     1078 |         ASSERT(xfs_is_shutdown(mp) ||
>          |         ^~~~~~
>    cc1: some warnings being treated as errors
> 
> 
> vim +1079 fs/xfs/xfs_super.c
> 
> 8757c38f2cf6e5 Ian Kent        2019-11-04  1070  
> 8757c38f2cf6e5 Ian Kent        2019-11-04  1071  static void
> 8757c38f2cf6e5 Ian Kent        2019-11-04  1072  xfs_destroy_percpu_counters(
> 8757c38f2cf6e5 Ian Kent        2019-11-04  1073  	struct xfs_mount	*mp)
> 8757c38f2cf6e5 Ian Kent        2019-11-04  1074  {
> 8757c38f2cf6e5 Ian Kent        2019-11-04  1075  	percpu_counter_destroy(&mp->m_icount);
> 8757c38f2cf6e5 Ian Kent        2019-11-04  1076  	percpu_counter_destroy(&mp->m_ifree);
> 8757c38f2cf6e5 Ian Kent        2019-11-04  1077  	percpu_counter_destroy(&mp->m_fdblocks);
> 75c8c50fa16a23 Dave Chinner    2021-08-18  1078  	ASSERT(xfs_is_shutdown(mp) ||
> c35278f526edf1 Ye Bin          2023-03-14 @1079  	       percpu_counter_sum_all(&mp->m_delalloc_blks) == 0);

This change has not been committed to any tree that I am aware of.
It was only posted to the XFS list yesterday, and I effectively
NACK'd it and wrote this patchset instead to fix the issue.

IOWs, if -anyone- has actually committed this change to add
percpu_counter_sum_all() to XFS, they've done the wrong thing.
Hence this build failure is a robot issue, not a problem with my
patch series.

-Dave.
-- 
Dave Chinner
david@fromorbit.com