fs/hugetlbfs/inode.c | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-)
When remove_inode_hugepages is introduced in
commit c86272287bc6 ("hugetlb: create remove_inode_single_folio to remove single file folio")
it used to return a boolean to indicate if it bailed out due to race with
page faults. However, since the race is already solved by [1],
remove_inode_hugepages() doesn't have any path to return false anymore.
Simplify remove_inode_hugepages() return type to void, remove the
unnecessary ret variable, and adjust the call site in
remove_inode_hugepages(). No functional change in this commit.
Changelog
=========
v2 -> v1 [2]
- Use two-tab alignment for parameters wrapped to new lines.
- Remove return type in remove_inode_hugepages()'s comment.
[1] https://lore.kernel.org/all/20220914221810.95771-10-mike.kravetz@oracle.com
[2] https://lore.kernel.org/linux-mm/20260202233617.2350734-1-jiaqiyan@google.com
Suggested-by: Jane Chu <jane.chu@oracle.com>
Reviewed-by: Jane Chu <jane.chu@oracle.com>
Reviewed-by: Muchun Song <muchun.song@linux.dev>
Acked-by: David Hildenbrand (arm) <david@kernel.org>
Signed-off-by: Jiaqi Yan <jiaqiyan@google.com>
---
fs/hugetlbfs/inode.c | 18 ++++++------------
1 file changed, 6 insertions(+), 12 deletions(-)
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 3b4c152c5c73a..09c88c09718b4 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -513,15 +513,11 @@ hugetlb_vmdelete_list(struct rb_root_cached *root, pgoff_t start, pgoff_t end,
/*
* Called with hugetlb fault mutex held.
- * Returns true if page was actually removed, false otherwise.
*/
-static bool remove_inode_single_folio(struct hstate *h, struct inode *inode,
- struct address_space *mapping,
- struct folio *folio, pgoff_t index,
- bool truncate_op)
+static void remove_inode_single_folio(struct hstate *h, struct inode *inode,
+ struct address_space *mapping, struct folio *folio,
+ pgoff_t index, bool truncate_op)
{
- bool ret = false;
-
/*
* If folio is mapped, it was faulted in after being
* unmapped in caller or hugetlb_vmdelete_list() skips
@@ -543,7 +539,6 @@ static bool remove_inode_single_folio(struct hstate *h, struct inode *inode,
*/
VM_BUG_ON_FOLIO(folio_test_hugetlb_restore_reserve(folio), folio);
hugetlb_delete_from_page_cache(folio);
- ret = true;
if (!truncate_op) {
if (unlikely(hugetlb_unreserve_pages(inode, index,
index + 1, 1)))
@@ -551,7 +546,6 @@ static bool remove_inode_single_folio(struct hstate *h, struct inode *inode,
}
folio_unlock(folio);
- return ret;
}
/*
@@ -599,9 +593,9 @@ static void remove_inode_hugepages(struct inode *inode, loff_t lstart,
/*
* Remove folio that was part of folio_batch.
*/
- if (remove_inode_single_folio(h, inode, mapping, folio,
- index, truncate_op))
- freed++;
+ remove_inode_single_folio(h, inode, mapping, folio,
+ index, truncate_op);
+ freed++;
mutex_unlock(&hugetlb_fault_mutex_table[hash]);
}
--
2.53.0.rc2.204.g2597b5adb4-googOn Wed, 4 Feb 2026 21:47:41 +0000 Jiaqi Yan <jiaqiyan@google.com> wrote:
> When remove_inode_hugepages is introduced in
> commit c86272287bc6 ("hugetlb: create remove_inode_single_folio to remove single file folio")
Why don't you wrap the above line for the 75 columns limit [1]?
> it used to return a boolean to indicate if it bailed out due to race with
> page faults. However, since the race is already solved by [1],
> remove_inode_hugepages() doesn't have any path to return false anymore.
>
> Simplify remove_inode_hugepages() return type to void, remove the
> unnecessary ret variable, and adjust the call site in
> remove_inode_hugepages(). No functional change in this commit.
>
> Changelog
> =========
>
> v2 -> v1 [2]
>
> - Use two-tab alignment for parameters wrapped to new lines.
>
> - Remove return type in remove_inode_hugepages()'s comment.
>
> [1] https://lore.kernel.org/all/20220914221810.95771-10-mike.kravetz@oracle.com
> [2] https://lore.kernel.org/linux-mm/20260202233617.2350734-1-jiaqiyan@google.com
I found David and Andrew replied the changelog will be removed in the final
one, so will not add comment.
>
> Suggested-by: Jane Chu <jane.chu@oracle.com>
> Reviewed-by: Jane Chu <jane.chu@oracle.com>
> Reviewed-by: Muchun Song <muchun.song@linux.dev>
> Acked-by: David Hildenbrand (arm) <david@kernel.org>
> Signed-off-by: Jiaqi Yan <jiaqiyan@google.com>
The above comment is only trivial, and the diff looks good to me. So,
Reviewed-by: SeongJae Park <sj@kernel.org>
Thanks,
SJ
[...]
On 2/4/26 22:47, Jiaqi Yan wrote:
> When remove_inode_hugepages is introduced in
> commit c86272287bc6 ("hugetlb: create remove_inode_single_folio to remove single file folio")
> it used to return a boolean to indicate if it bailed out due to race with
> page faults. However, since the race is already solved by [1],
> remove_inode_hugepages() doesn't have any path to return false anymore.
>
> Simplify remove_inode_hugepages() return type to void, remove the
> unnecessary ret variable, and adjust the call site in
> remove_inode_hugepages(). No functional change in this commit.
>
> Changelog
> =========
>
> v2 -> v1 [2]
>
> - Use two-tab alignment for parameters wrapped to new lines.
>
> - Remove return type in remove_inode_hugepages()'s comment.
>
> [1] https://lore.kernel.org/all/20220914221810.95771-10-mike.kravetz@oracle.com
> [2] https://lore.kernel.org/linux-mm/20260202233617.2350734-1-jiaqiyan@google.com
>
> Suggested-by: Jane Chu <jane.chu@oracle.com>
> Reviewed-by: Jane Chu <jane.chu@oracle.com>
> Reviewed-by: Muchun Song <muchun.song@linux.dev>
> Acked-by: David Hildenbrand (arm) <david@kernel.org>
> Signed-off-by: Jiaqi Yan <jiaqiyan@google.com>
> ---
For the future, the changelog always belongs down here, under the ---, like
Signed-off-by: Jiaqi Yan <jiaqiyan@google.com>
---
Changelog
...
fs/hugetlbfs/inode.c | 18 ++++++------------
1 file changed, 6 insertions(+), 12 deletions(-)
I think I always add another "---" above the file list, not sure if that
is required.
I'm sure Andrew can fix that up when applying, thanks!
--
Cheers,
David
On Wed, 4 Feb 2026 22:54:48 +0100 "David Hildenbrand (arm)" <david@kernel.org> wrote: > > Changelog > > ========= > > > > v2 -> v1 [2] > > > > - Use two-tab alignment for parameters wrapped to new lines. > > > > - Remove return type in remove_inode_hugepages()'s comment. > > > > [1] https://lore.kernel.org/all/20220914221810.95771-10-mike.kravetz@oracle.com > > [2] https://lore.kernel.org/linux-mm/20260202233617.2350734-1-jiaqiyan@google.com > > > > Suggested-by: Jane Chu <jane.chu@oracle.com> > > Reviewed-by: Jane Chu <jane.chu@oracle.com> > > Reviewed-by: Muchun Song <muchun.song@linux.dev> > > Acked-by: David Hildenbrand (arm) <david@kernel.org> > > Signed-off-by: Jiaqi Yan <jiaqiyan@google.com> > > --- > > For the future, the changelog always belongs down here, under the ---, like > > Signed-off-by: Jiaqi Yan <jiaqiyan@google.com> > --- > > Changelog > ... > > fs/hugetlbfs/inode.c | 18 ++++++------------ > 1 file changed, 6 insertions(+), 12 deletions(-) > > I think I always add another "---" above the file list, not sure if that > is required. Interpretations might vary but the "^---$" inventor sayeth that everything below the "^---$" is not part of the changelog and is to be removed. Also, my preference is that the permanent upstreamable changelog not contain short-term development-time information, as that's uninteresting to someone who is looking at the commit three years from now. Material such as "how this patch differes from the previous version" and "how I tested it" is interesting and useful, but below the "^---$", please. Beware that git interprets the separator more loosely: "^-----------" will match. So using a row of "-" characters to underline a heading is a pitfall.
© 2016 - 2026 Red Hat, Inc.