It will be used outside hugetlb.c soon.

Signed-off-by: Peter Xu <peterx@redhat.com>
---
 include/linux/hugetlb.h | 3 +++
 mm/hugetlb.c            | 4 ++--
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 236ec7b63c54..bb07279b8991 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -174,6 +174,9 @@ u32 hugetlb_fault_mutex_hash(struct address_space *mapping, pgoff_t idx);
 
 pte_t *huge_pmd_share(struct mm_struct *mm, struct vm_area_struct *vma,
 		      unsigned long addr, pud_t *pud);
+bool hugetlbfs_pagecache_present(struct hstate *h,
+				 struct vm_area_struct *vma,
+				 unsigned long address);
 
 struct address_space *hugetlb_page_mapping_lock_write(struct page *hpage);
 
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 6feb3e0630d1..29705e5c6f40 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -6018,8 +6018,8 @@ static vm_fault_t hugetlb_wp(struct mm_struct *mm, struct vm_area_struct *vma,
 /*
  * Return whether there is a pagecache page to back given address within VMA.
  */
-static bool hugetlbfs_pagecache_present(struct hstate *h,
-			struct vm_area_struct *vma, unsigned long address)
+bool hugetlbfs_pagecache_present(struct hstate *h,
+				 struct vm_area_struct *vma, unsigned long address)
 {
 	struct address_space *mapping = vma->vm_file->f_mapping;
 	pgoff_t idx = linear_page_index(vma, address);
-- 
2.41.0

You're (thankfully) not actually exporting anything, so please
don't claim that in the subject.

>  pte_t *huge_pmd_share(struct mm_struct *mm, struct vm_area_struct *vma,
>  		      unsigned long addr, pud_t *pud);
> +bool hugetlbfs_pagecache_present(struct hstate *h,
> +				 struct vm_area_struct *vma,
> +				 unsigned long address);

Can you just follow the much more readable two tab indent of the
function above?

> -static bool hugetlbfs_pagecache_present(struct hstate *h,
> -			struct vm_area_struct *vma, unsigned long address)
> +bool hugetlbfs_pagecache_present(struct hstate *h,
> +				 struct vm_area_struct *vma, unsigned long address)

Same here.  The new indentation not only is less readable but also
creates a pointlessly overlong line.

On Wed, Nov 22, 2023 at 11:23:07PM -0800, Christoph Hellwig wrote:
> You're (thankfully) not actually exporting anything, so please
> don't claim that in the subject.

Ah, I'll rename the subject to "mm/hugetlb: Declare
hugetlbfs_pagecache_present() non-static".

> 
> >  pte_t *huge_pmd_share(struct mm_struct *mm, struct vm_area_struct *vma,
> >  		      unsigned long addr, pud_t *pud);
> > +bool hugetlbfs_pagecache_present(struct hstate *h,
> > +				 struct vm_area_struct *vma,
> > +				 unsigned long address);
> 
> Can you just follow the much more readable two tab indent of the
> function above?
> 
> > -static bool hugetlbfs_pagecache_present(struct hstate *h,
> > -			struct vm_area_struct *vma, unsigned long address)
> > +bool hugetlbfs_pagecache_present(struct hstate *h,
> > +				 struct vm_area_struct *vma, unsigned long address)
> 
> Same here.  The new indentation not only is less readable but also
> creates a pointlessly overlong line.

I can easily follow what you suggest for this single patch, but afaict the
kernel doesn't document that in the style guide.

https://www.kernel.org/doc/html/v4.10/process/coding-style.html#functions

And the reality is at least across mm codes it's used in a mixture of ways,
even more than these two major forms of indentations.

To be explicit, I think 2-tab is Vim's default, while this patch follows
Emacs's c-mode default.  It means if this patch indents wrongly, probably
99% of Emacs users are doomed. :(

Before we have a clear and thorough rule over this, shall we just allow
either sane indent to still be accepted?  Or maybe there's some rule that I
have missed?  Personally I actually prefer Emacs's indentations to align
with left bracket, but that's just subjective so doesn't count.

Thanks,

-- 
Peter Xu