Re: [PATCH v2 03/25] mm/percpu: Annotate static information into meminspect

[kernel test robot <lkp@intel.com>]

Hi Mukesh,

kernel test robot noticed the following build errors:

[auto build test ERROR on 343f51842f4ed7143872f3aa116a214a5619a4b9]

url:    https://github.com/intel-lab-lkp/linux/commits/Mukesh-Ojha/kernel-Introduce-meminspect/20260311-042334
base:   343f51842f4ed7143872f3aa116a214a5619a4b9
patch link:    https://lore.kernel.org/r/20260311-minidump-v2-v2-3-f91cedc6f99e%40oss.qualcomm.com
patch subject: [PATCH v2 03/25] mm/percpu: Annotate static information into meminspect
config: sparc64-allmodconfig (https://download.01.org/0day-ci/archive/20260313/202603130444.K3MmZnxY-lkp@intel.com/config)
compiler: clang version 23.0.0git (https://github.com/llvm/llvm-project 7d47b695929cc7f85eeb0f87d0189adc04c1c629)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20260313/202603130444.K3MmZnxY-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/202603130444.K3MmZnxY-lkp@intel.com/

All errors (new ones prefixed by >>):

>> mm/percpu.c:3363:25: error: use of undeclared identifier '__per_cpu_offset'; did you mean '__per_cpu_start'?
    3363 | MEMINSPECT_SIMPLE_ENTRY(__per_cpu_offset);
         |                         ^~~~~~~~~~~~~~~~
         |                         __per_cpu_start
   include/linux/meminspect.h:103:40: note: expanded from macro 'MEMINSPECT_SIMPLE_ENTRY'
     103 |         MEMINSPECT_ENTRY(MEMINSPECT_ID_##sym, sym, sizeof(sym))
         |                                               ^~~
   include/linux/meminspect.h:94:19: note: expanded from macro 'MEMINSPECT_ENTRY'
      94 |                 .va = (void *)&(sym),                                           \
         |                                 ^~~
   include/asm-generic/sections.h:42:13: note: '__per_cpu_start' declared here
      42 | extern char __per_cpu_start[], __per_cpu_end[];
         |             ^
>> mm/percpu.c:3363:25: error: use of undeclared identifier '__per_cpu_offset'; did you mean '__per_cpu_start'?
    3363 | MEMINSPECT_SIMPLE_ENTRY(__per_cpu_offset);
         |                         ^~~~~~~~~~~~~~~~
         |                         __per_cpu_start
   include/linux/meminspect.h:103:52: note: expanded from macro 'MEMINSPECT_SIMPLE_ENTRY'
     103 |         MEMINSPECT_ENTRY(MEMINSPECT_ID_##sym, sym, sizeof(sym))
         |                                                           ^~~
   include/linux/meminspect.h:95:12: note: expanded from macro 'MEMINSPECT_ENTRY'
      95 |                 .size = (sz),                                                   \
         |                          ^~
   include/asm-generic/sections.h:42:13: note: '__per_cpu_start' declared here
      42 | extern char __per_cpu_start[], __per_cpu_end[];
         |             ^
>> mm/percpu.c:3363:1: error: invalid application of 'sizeof' to an incomplete type 'char[]'
    3363 | MEMINSPECT_SIMPLE_ENTRY(__per_cpu_offset);
         | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   include/linux/meminspect.h:103:51: note: expanded from macro 'MEMINSPECT_SIMPLE_ENTRY'
     103 |         MEMINSPECT_ENTRY(MEMINSPECT_ID_##sym, sym, sizeof(sym))
         |         ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~
   include/linux/meminspect.h:95:12: note: expanded from macro 'MEMINSPECT_ENTRY'
      95 |                 .size = (sz),                                                   \
         |                          ^~
   3 errors generated.


vim +3363 mm/percpu.c

  3362	
> 3363	MEMINSPECT_SIMPLE_ENTRY(__per_cpu_offset);
  3364	/*
  3365	 * pcpu_nr_pages - calculate total number of populated backing pages
  3366	 *
  3367	 * This reflects the number of pages populated to back chunks.  Metadata is
  3368	 * excluded in the number exposed in meminfo as the number of backing pages
  3369	 * scales with the number of cpus and can quickly outweigh the memory used for
  3370	 * metadata.  It also keeps this calculation nice and simple.
  3371	 *
  3372	 * RETURNS:
  3373	 * Total number of populated backing pages in use by the allocator.
  3374	 */
  3375	unsigned long pcpu_nr_pages(void)
  3376	{
  3377		return data_race(READ_ONCE(pcpu_nr_populated)) * pcpu_nr_units;
  3378	}
  3379	

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki