linux-next: build failure after merge of the kvm tree

From: Stephen Rothwell
Date: Sun Nov 19 2023 - 23:22:45 EST


Hi all,

After merging the kvm tree, today's linux-next build (x86_64 allmodconfig)
failed like this:

arch/x86/kvm/../../../virt/kvm/guest_memfd.c:306:10: error: 'const struct address_space_operations' has no member named 'error_remove_page'; did you mean 'error_remove_folio'?
306 | .error_remove_page = kvm_gmem_error_page,
| ^~~~~~~~~~~~~~~~~
| error_remove_folio
arch/x86/kvm/../../../virt/kvm/guest_memfd.c:306:30: error: initialization of 'int (*)(struct folio *)' from incompatible pointer type 'int (*)(struct address_space *, struct page *)' [-Werror=incompatible-pointer-types]
306 | .error_remove_page = kvm_gmem_error_page,
| ^~~~~~~~~~~~~~~~~~~
arch/x86/kvm/../../../virt/kvm/guest_memfd.c:306:30: note: (near initialization for 'kvm_gmem_aops.launder_folio')

Caused by commit

640be5bc564f ("fs: convert error_remove_page to error_remove_folio")

from the mm tree intercting with commit

a7800aa80ea4 ("KVM: Add KVM_CREATE_GUEST_MEMFD ioctl() for guest-specific backing memory")

I have applied the following supplied merge fix patch (thanks Andrew).

From: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Date: Fri, 17 Nov 2023 09:28:33 -0800
Subject: [PATCH] fs: Convert error_remove_page to error_remove_folio

On Fri, 17 Nov 2023 16:14:47 +0000 "Matthew Wilcox (Oracle)" <willy@xxxxxxxxxxxxx> wrote:

> There were already assertions that we were not passing a tail page
> to error_remove_page(), so make the compiler enforce that by converting
> everything to pass and use a folio.
>
> Signed-off-by: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx>
> ---
> Documentation/filesystems/locking.rst | 4 ++--
> Documentation/filesystems/vfs.rst | 6 +++---
> block/fops.c | 2 +-
> fs/afs/write.c | 2 +-
> fs/bcachefs/fs.c | 2 +-
> fs/btrfs/inode.c | 2 +-
> fs/ceph/addr.c | 4 ++--
> fs/ext2/inode.c | 2 +-
> fs/ext4/inode.c | 6 +++---
> fs/f2fs/compress.c | 2 +-
> fs/f2fs/inode.c | 2 +-
> fs/gfs2/aops.c | 4 ++--
> fs/hugetlbfs/inode.c | 6 +++---
> fs/nfs/file.c | 2 +-
> fs/ntfs/aops.c | 6 +++---
> fs/ocfs2/aops.c | 2 +-
> fs/xfs/xfs_aops.c | 2 +-
> fs/zonefs/file.c | 2 +-
> include/linux/fs.h | 2 +-
> include/linux/mm.h | 3 ++-
> mm/memory-failure.c | 10 +++++-----
> mm/shmem.c | 6 +++---
> mm/truncate.c | 9 ++++-----
> virt/kvm/guest_memfd.c | 9 +++++----

virt/kvm/guest_memfd.c exists only in the KVM tree (and hence
linux-next). So I assume Stephen will use the change from this patch
when doing his resolution.

This:
---
virt/kvm/guest_memfd.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/virt/kvm/guest_memfd.c b/virt/kvm/guest_memfd.c
index b99272396119..451435123fe7 100644
--- a/virt/kvm/guest_memfd.c
+++ b/virt/kvm/guest_memfd.c
@@ -267,7 +267,8 @@ static int kvm_gmem_migrate_folio(struct address_space *mapping,
return -EINVAL;
}

-static int kvm_gmem_error_page(struct address_space *mapping, struct page *page)
+static int kvm_gmem_error_folio(struct address_space *mapping,
+ struct folio *folio)
{
struct list_head *gmem_list = &mapping->private_list;
struct kvm_gmem *gmem;
@@ -275,8 +276,8 @@ static int kvm_gmem_error_page(struct address_space *mapping, struct page *page)

filemap_invalidate_lock_shared(mapping);

- start = page->index;
- end = start + thp_nr_pages(page);
+ start = folio->index;
+ end = start + folio_nr_pages(folio);

list_for_each_entry(gmem, gmem_list, entry)
kvm_gmem_invalidate_begin(gmem, start, end);
@@ -303,7 +304,7 @@ static const struct address_space_operations kvm_gmem_aops = {
#ifdef CONFIG_MIGRATION
.migrate_folio = kvm_gmem_migrate_folio,
#endif
- .error_remove_page = kvm_gmem_error_page,
+ .error_remove_folio = kvm_gmem_error_folio,
};

static int kvm_gmem_getattr(struct mnt_idmap *idmap, const struct path *path,
--
2.40.1

--
Cheers,
Stephen Rothwell

Attachment: pgp1ZYGLVFeXc.pgp
Description: OpenPGP digital signature