A mixture of odd changes that didn't quite make it into the original
pull and fixes for things that did. Also the readpages changes had to
wait for the NFS tree to be pulled first.
The following changes since commit d888c83fcec75194a8a48ccd283953bdba7b2550:
fs: fix fd table size alignment properly (2022-03-29 23:29:18 -0700)
are available in the Git repository at:
git://git.infradead.org/users/willy/pagecache.git tags/folio-5.18d
for you to fetch changes up to 5a60542c61f3cce6e5dff2a38c8fb08a852a517b:
btrfs: Remove a use of PAGE_SIZE in btrfs_invalidate_folio() (2022-04-01 14:40:44 -0400)
----------------------------------------------------------------
Filesystem/VFS changes for 5.18, part two
- Remove ->readpages infrastructure
- Remove AOP_FLAG_CONT_EXPAND
- Move read_descriptor_t to networking code
- Pass the iocb to generic_perform_write
- Minor updates to iomap, btrfs, ext4, f2fs, ntfs
----------------------------------------------------------------
Christoph Hellwig (2):
mm: remove the pages argument to read_pages
mm: remove the skip_page argument to read_pages
Matthew Wilcox (Oracle) (13):
readahead: Remove read_cache_pages()
fs: Remove ->readpages address space operation
readahead: Update comments
iomap: Simplify is_partially_uptodate a little
fs: Remove read_actor_t
fs, net: Move read_descriptor_t to net.h
fs: Pass an iocb to generic_perform_write()
filemap: Remove AOP_FLAG_CONT_EXPAND
ext4: Correct ext4_journalled_dirty_folio() conversion
f2fs: Correct f2fs_dirty_data_folio() conversion
f2fs: Get the superblock from the mapping instead of the page
ntfs: Correct mark_ntfs_record_dirty() folio conversion
btrfs: Remove a use of PAGE_SIZE in btrfs_invalidate_folio()
Documentation/filesystems/fsverity.rst | 6 +-
Documentation/filesystems/locking.rst | 6 -
Documentation/filesystems/vfs.rst | 11 --
fs/btrfs/inode.c | 2 +-
fs/btrfs/reflink.c | 4 +-
fs/buffer.c | 3 +-
fs/ceph/file.c | 2 +-
fs/cifs/cifssmb.c | 2 +-
fs/cifs/inode.c | 2 +-
fs/crypto/crypto.c | 2 +-
fs/ext4/file.c | 2 +-
fs/ext4/inode.c | 2 +-
fs/ext4/readpage.c | 2 +-
fs/f2fs/checkpoint.c | 2 +-
fs/f2fs/data.c | 6 +-
fs/f2fs/file.c | 2 +-
fs/f2fs/node.c | 4 +-
fs/fuse/fuse_i.h | 2 +-
fs/iomap/buffered-io.c | 9 +-
fs/nfs/file.c | 2 +-
fs/ntfs/aops.c | 2 +-
fs/verity/verify.c | 4 +-
include/linux/fs.h | 31 +----
include/linux/fsverity.h | 2 +-
include/linux/net.h | 19 +++
include/linux/pagemap.h | 2 -
mm/filemap.c | 12 +-
mm/readahead.c | 204 +++++++++------------------------
28 files changed, 113 insertions(+), 236 deletions(-)
The pull request you sent on Fri, 1 Apr 2022 19:54:59 +0100:
> git://git.infradead.org/users/willy/pagecache.git tags/folio-5.18d
has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/cda4351252e710507d32178dfbb5a7f0f92488f8
Thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/prtracker.html
On Tue, Apr 05, 2022 at 02:08:48PM +0200, David Sterba wrote:
> Matthew, can you please always CC [email protected] for any
> patches that touch code under fs/btrfs? I've only noticed your folio
> updates in this pull request. Some of the changes are plain API switch,
> that's fine but I want to know about that, some changes seem to slightly
> modify logic that I'd really like to review and there are several missed
> opportunities to fix coding style. Thanks.
I'm sorry, that's an unreasonable request. There's ~50 filesystems
that use address_space_operations and cc'ing individual filesystems
on VFS-wide changes isn't feaasible.
On Tue, Apr 05, 2022 at 03:22:09PM +0100, Matthew Wilcox wrote:
> On Tue, Apr 05, 2022 at 02:08:48PM +0200, David Sterba wrote:
> > Matthew, can you please always CC [email protected] for any
> > patches that touch code under fs/btrfs? I've only noticed your folio
> > updates in this pull request. Some of the changes are plain API switch,
> > that's fine but I want to know about that, some changes seem to slightly
> > modify logic that I'd really like to review and there are several missed
> > opportunities to fix coding style. Thanks.
>
> I'm sorry, that's an unreasonable request. There's ~50 filesystems
> that use address_space_operations and cc'ing individual filesystems
> on VFS-wide changes isn't feaasible.
How many filesystems have you touched in the recent changes? I've
counted about 7 subsystems in commit 704528d895dd ("fs: Remove
->readpages address space operation"), the rest are VFS/MM changes or
individual filesystems in separate patches.
Examples of btrfs-only changes, there are more like that in the pull as
you probably know:
8e1dec8eb8b0 ("btrfs: Use folio_invalidate()").
895586eb6898 ("btrfs: Convert from invalidatepage to invalidate_folio")
...
You know you can slap a CC: [email protected] to the patch
tag and forget about it, is that unreasonable? No. If you're updating
the same filesystems repeatedly you can copy the CC list for all of
them.
On Fri, Apr 01, 2022 at 07:54:59PM +0100, Matthew Wilcox wrote:
> A mixture of odd changes that didn't quite make it into the original
> pull and fixes for things that did. Also the readpages changes had to
> wait for the NFS tree to be pulled first.
>
> The following changes since commit d888c83fcec75194a8a48ccd283953bdba7b2550:
>
> fs: fix fd table size alignment properly (2022-03-29 23:29:18 -0700)
>
> are available in the Git repository at:
>
> git://git.infradead.org/users/willy/pagecache.git tags/folio-5.18d
>
> for you to fetch changes up to 5a60542c61f3cce6e5dff2a38c8fb08a852a517b:
>
> btrfs: Remove a use of PAGE_SIZE in btrfs_invalidate_folio() (2022-04-01 14:40:44 -0400)
Matthew, can you please always CC [email protected] for any
patches that touch code under fs/btrfs? I've only noticed your folio
updates in this pull request. Some of the changes are plain API switch,
that's fine but I want to know about that, some changes seem to slightly
modify logic that I'd really like to review and there are several missed
opportunities to fix coding style. Thanks.
On Tue, Apr 05, 2022 at 03:22:09PM +0100, Matthew Wilcox wrote:
> On Tue, Apr 05, 2022 at 02:08:48PM +0200, David Sterba wrote:
> > Matthew, can you please always CC [email protected] for any
> > patches that touch code under fs/btrfs? I've only noticed your folio
> > updates in this pull request. Some of the changes are plain API switch,
> > that's fine but I want to know about that, some changes seem to slightly
> > modify logic that I'd really like to review and there are several missed
> > opportunities to fix coding style. Thanks.
>
> I'm sorry, that's an unreasonable request. There's ~50 filesystems
> that use address_space_operations and cc'ing individual filesystems
> on VFS-wide changes isn't feaasible.
FYI, for these kinds of global API changes I tend to add all the
mainling lists, but drop the multiple maintainers that would blow this
up even futher. But even that lead to occasional complaints.