[akpm-mm:mm-new 291/296] include/linux/mm.h:957:24: sparse: sparse: incorrect type in argument 2 (different modifiers)

[akpm-mm:mm-new 291/296] include/linux/mm.h:957:24: sparse: sparse: incorrect type in argument 2 (different modifiers)

[kernel test robot]

tree:   https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-new
head:   8017f1fc873a78f54df02521b4d150f7101916ae
commit: 85de428c7ca1b71457690b95e162662d0edb2792 [291/296] mm: set the VM_MAYBE_GUARD flag on guard region install
config: s390-randconfig-r131-20251110 (https://download.01.org/0day-ci/archive/20251110/202511102224.9qWchj6g-lkp@intel.com/config)
compiler: s390-linux-gcc (GCC) 12.5.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251110/202511102224.9qWchj6g-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/202511102224.9qWchj6g-lkp@intel.com/

sparse warnings: (new ones prefixed by >>)
   mm/khugepaged.c:830:25: sparse: sparse: symbol 'khugepaged_collapse_control' was not declared. Should it be static?
   mm/khugepaged.c: note: in included file (through include/linux/rbtree.h, include/linux/mm_types.h, include/linux/mmzone.h, ...):
   include/linux/rcupdate.h:897:25: sparse: sparse: context imbalance in '__collapse_huge_page_swapin' - unexpected unlock
   mm/khugepaged.c:1196:26: sparse: sparse: context imbalance in 'collapse_huge_page' - different lock contexts for basic block
   mm/khugepaged.c:1705:28: sparse: sparse: context imbalance in 'collapse_pte_mapped_thp' - unexpected unlock
   mm/khugepaged.c: note: in included file:
>> include/linux/mm.h:957:24: sparse: sparse: incorrect type in argument 2 (different modifiers) @@     expected unsigned long const volatile *ptr @@     got unsigned long [noderef] * @@
   include/linux/mm.h:957:24: sparse:     expected unsigned long const volatile *ptr
   include/linux/mm.h:957:24: sparse:     got unsigned long [noderef] *
>> include/linux/mm.h:957:24: sparse: sparse: incorrect type in argument 2 (different modifiers) @@     expected unsigned long const volatile *addr @@     got unsigned long [noderef] * @@
   include/linux/mm.h:957:24: sparse:     expected unsigned long const volatile *addr
   include/linux/mm.h:957:24: sparse:     got unsigned long [noderef] *
   mm/khugepaged.c:1822:28: sparse: sparse: context imbalance in 'retract_page_tables' - different lock contexts for basic block

vim +957 include/linux/mm.h

bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  946  
bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  947  /*
bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  948   * Test for VMA flag atomically. Requires no locks. Only specific valid flags
bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  949   * are allowed to do this.
bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  950   *
bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  951   * This is necessarily racey, so callers must ensure that serialisation is
bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  952   * achieved through some other means, or that races are permissible.
bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  953   */
bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  954  static inline bool vma_flag_test_atomic(struct vm_area_struct *vma, int bit)
bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  955  {
bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  956  	if (__vma_flag_atomic_valid(vma, bit))
bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07 @957  		return test_bit(bit, &vma->__vm_flags);
85de428c7ca1b71 Lorenzo Stoakes 2025-11-07  958  
85de428c7ca1b71 Lorenzo Stoakes 2025-11-07  959  	return false;
bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  960  }
bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  961  

:::::: The code at line 957 was first introduced by commit
:::::: bf91f4ed2a530f2e55e7b202c1692a4c682888d6 mm: add atomic VMA flags and set VM_MAYBE_GUARD as such

:::::: TO: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
:::::: CC: Andrew Morton <akpm@linux-foundation.org>

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

Re: [akpm-mm:mm-new 291/296] include/linux/mm.h:957:24: sparse: sparse: incorrect type in argument 2 (different modifiers)

[Lorenzo Stoakes]

On Mon, Nov 10, 2025 at 10:29:05PM +0800, kernel test robot wrote:
> tree:   https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-new
> head:   8017f1fc873a78f54df02521b4d150f7101916ae
> commit: 85de428c7ca1b71457690b95e162662d0edb2792 [291/296] mm: set the VM_MAYBE_GUARD flag on guard region install
> config: s390-randconfig-r131-20251110 (https://download.01.org/0day-ci/archive/20251110/202511102224.9qWchj6g-lkp@intel.com/config)
> compiler: s390-linux-gcc (GCC) 12.5.0
> reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251110/202511102224.9qWchj6g-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/202511102224.9qWchj6g-lkp@intel.com/
>
> sparse warnings: (new ones prefixed by >>)
> >> include/linux/mm.h:957:24: sparse: sparse: incorrect type in argument 2 (different modifiers) @@     expected unsigned long const volatile *ptr @@     got unsigned long [noderef] * @@

Yeah my mistake this is the ACCESS_PRIVATE() stuff.

I'm not sure why it's not highlighting the vma_flags_set_atomic() stuff but
there we are :)

I'll fix this up.

>
> vim +957 include/linux/mm.h
>
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  946
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  947  /*
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  948   * Test for VMA flag atomically. Requires no locks. Only specific valid flags
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  949   * are allowed to do this.
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  950   *
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  951   * This is necessarily racey, so callers must ensure that serialisation is
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  952   * achieved through some other means, or that races are permissible.
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  953   */
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  954  static inline bool vma_flag_test_atomic(struct vm_area_struct *vma, int bit)
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  955  {
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  956  	if (__vma_flag_atomic_valid(vma, bit))
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07 @957  		return test_bit(bit, &vma->__vm_flags);
> 85de428c7ca1b71 Lorenzo Stoakes 2025-11-07  958
> 85de428c7ca1b71 Lorenzo Stoakes 2025-11-07  959  	return false;
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  960  }
> bf91f4ed2a530f2 Lorenzo Stoakes 2025-11-07  961
>
> :::::: The code at line 957 was first introduced by commit
> :::::: bf91f4ed2a530f2e55e7b202c1692a4c682888d6 mm: add atomic VMA flags and set VM_MAYBE_GUARD as such
>
> :::::: TO: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> :::::: CC: Andrew Morton <akpm@linux-foundation.org>
>
> --
> 0-DAY CI Kernel Test Service
> https://github.com/intel/lkp-tests/wiki

Thanks, Lorenzo