[GIT PULL] Please pull NFS client fixes for Linux 7.3

[GIT PULL] Please pull NFS client fixes for Linux 7.3

[Trond Myklebust]

Hi Linus,

The following changes since commit db2ddb87143519e20a95aa36c60b36107b736a58:

  Linux 7.2-rc7 (2026-08-09 14:54:50 -0700)

are available in the Git repository at:

  git://git.linux-nfs.org/projects/trondmy/linux-nfs.git tags/nfs-for-7.3-1

for you to fetch changes up to ee5a386cfe60f3f8286de16a9db8e1a08f0bc124:

  NFSv4.1: fix layout segment leak on the pnfs_layout_process() forget path (2026-08-20 15:18:20 -0700)

Thanks!
  Trond
----------------------------------------------------------------
NFS client updates for Linux 7.3

Highlights include:

Stable fixes:
- SunRPC: Use-after-free fixes for the sunrpc client code
- NFSv4: Delegation hash table leak
- lockd: NULL dereference on lockowner allocation failure
- SunRPC: Fix a handshake completion race in the TLS code
- NFSv4.1/pNFS: Fix an error sign checking issue when deciding whether
  the layout is still in use, or can be returned.
- NFSv4.1: Fix a layout segment leak in pnfs_layout_process()

Other bugfixes:
- SunRPC: Fix a missing NULL check in the rpcbind client
- SunRPC: annotate shared socket callbacks with READ_ONCE/WRITE_ONCE
- NFSv4: nfs_inode_set_delegation() error paths should return the delegation
- NFSv4: Use clear_and_wake_up_bit() in nfs_clear_invalid_mapping() and
  the pNFS code.
- NFSv4: Fix the nfs4_alloc_client() error paths to free the IDR
  allocation
- NFS: fix folio dereference before NULL check in nfs_inode_remove_request()
- NFS: Fix delayed delegation return
- NFSv4: Fix another state manager race with umount
- pNFS/blocklayout: Fix device leaks on parse failure
- pNFS: Avoid cancelling in-flight I/O during a layout recall if the
  server doesn't require it
- NFSv4/flexfiles: report cancelled I/O as a layout error
- NFSv4/flexfiles: fix NULL dereference for NFSv4.0 data servers
- NFSv4: Fix incorrect argument passed to nfs4_delete_lease()
- NFSv3: Fix several symlink issues resulting from nfs_atomic_open_v23()
- NFSv4.1: Fix an uninitialised variable issue in the callback code
- NFSv4.2: fix LAYOUTSTATS send buffer exhaustion

Features and cleanups:
- NFSv4.2: Allow the server to specify that file data may not be cached
- NFS/localio: optimise I/O submission when when not doing memory reclaim
- NFS/localio: Remove duplicate wait code in nfs_local_commit
- NFSv4/flexfiles: support loosely coupled NFSv4.x data servers
- NFSv4/pnfs: key the data server cache on the NFS version

----------------------------------------------------------------
Arnaud Bonnet (2):
      nfs: replace atomic bitops sequence with clear_and_wake_up_bit helper
      nfs: refactor pNFS functions using clear_and_wake_up_bit

Benjamin Coddington (3):
      pNFS: report clora_changed in the cb_layoutrecall_file tracepoint
      pNFS: honor clora_changed when recalling a layout
      NFSv4/flexfiles: report cancelled I/O as a layout error

Chuck Lever (1):
      NFS: Return a delegation the client fails to record

Guangshuo Li (1):
      NFS: Fix delayed delegation return list handling

Jeuk Kim (2):
      NFSv4/flexfiles: fix NULL dereference for NFSv4.0 data servers
      NFSv4/flexfiles: support loosely coupled data servers

Jia Zhu (1):
      NFSv4: pin the superblock for active state owners

Jiangshan Yi (1):
      NFS: fix folio dereference before NULL check in nfs_inode_remove_request()

Junrui Luo (3):
      NFSv4.2: fix LAYOUTSTATS send buffer exhaustion
      NFSv4/pnfs: key the data server cache on the NFS version
      NFSv4.1: fix layout segment leak on the pnfs_layout_process() forget path

Jérémy Jean (2):
      SUNRPC: wait for in-flight client TLS handshake callback
      NFSv4.1: zero referring call lists before decoding

Luxiao Xu (1):
      sunrpc: fix use-after-free in __rpc_clnt_handle_event and __rpc_clnt_remove_pipedir

Michael Nemanov (1):
      nfs: fix ENXIO on O_CREAT open of existing symlink over NFSv3

Mike Snitzer (5):
      nfs4.2: request UNCACHEABLE_FILE_DATA only for regular files
      nfs4.2: open UNCACHEABLE_FILE_DATA files with O_DIRECT
      NFS/localio: issue IO inline when not in a memory-reclaim context
      NFS/localio: remove dead FLUSH_SYNC handling from nfs_local_commit
      NFS/localio: issue commit inline when not in a memory-reclaim context

Nate Prodromou (1):
      NFS: fix delegation_hash_table leak when nfs4_server_common_setup() fails

Runyu Xiao (1):
      sunrpc: xprtsock: annotate shared socket callbacks with READ_ONCE/WRITE_ONCE

Ruoyu Wang (1):
      NFSv4: remove callback IDR entry on client allocation failure

Shuangpeng Bai (1):
      lockd: fix NULL dereference on lockowner allocation failure

Tim Menninger (1):
      pNFS: Fix EBUSY check in pnfs_layout_need_return

Tom Haynes (1):
      nfs4.2: add UNCACHEABLE_FILE_DATA attribute support

Weiming Shi (1):
      SUNRPC: check rpc_sockaddr2uaddr() return value in rpcb_register_inet4/6

ZhangGuoDong (2):
      pnfs/blocklayout: Fix device leaks on parse failure
      NFS: Verify symlink inode before caching target

Zhansong Gao (1):
      NFSv4: Fix incorrect argument passed to nfs4_delete_lease() in nfs4_add_lease()

 fs/lockd/clntproc.c                       |  3 ++
 fs/nfs/blocklayout/dev.c                  | 21 +++++++---
 fs/nfs/callback_proc.c                    |  5 ++-
 fs/nfs/callback_xdr.c                     |  2 +-
 fs/nfs/client.c                           | 14 ++++++-
 fs/nfs/delegation.c                       | 20 ++++++---
 fs/nfs/dir.c                              | 17 ++++++++
 fs/nfs/filelayout/filelayoutdev.c         |  5 ++-
 fs/nfs/flexfilelayout/flexfilelayout.c    | 26 +++++++++++-
 fs/nfs/flexfilelayout/flexfilelayoutdev.c |  6 ++-
 fs/nfs/inode.c                            | 28 ++++++++++---
 fs/nfs/internal.h                         |  8 ++--
 fs/nfs/localio.c                          | 56 +++++++++++++++++++------
 fs/nfs/nfs4client.c                       | 23 +++++++----
 fs/nfs/nfs4file.c                         |  2 +
 fs/nfs/nfs4proc.c                         | 69 ++++++++++++++++++++++++++++---
 fs/nfs/nfs4session.c                      | 21 ++++++----
 fs/nfs/nfs4session.h                      |  3 +-
 fs/nfs/nfs4state.c                        | 40 ++++++++++++++----
 fs/nfs/nfs4trace.h                        | 58 +++++++++++++++++++++++++-
 fs/nfs/nfs4xdr.c                          | 35 +++++++++++++++-
 fs/nfs/nfstrace.h                         |  3 +-
 fs/nfs/pnfs.c                             | 60 +++++++++++----------------
 fs/nfs/pnfs.h                             |  8 ++--
 fs/nfs/pnfs_nfs.c                         | 28 ++++++++-----
 fs/nfs/write.c                            | 13 +++---
 include/linux/nfs4.h                      |  9 ++++
 include/linux/nfs_fs.h                    |  4 ++
 include/linux/nfs_xdr.h                   | 10 ++++-
 include/linux/workqueue.h                 |  1 +
 kernel/workqueue.c                        | 24 +++++++++++
 net/sunrpc/clnt.c                         | 22 +++++++---
 net/sunrpc/rpcb_clnt.c                    |  4 ++
 net/sunrpc/xprtsock.c                     | 32 +++++++++-----
 34 files changed, 534 insertions(+), 146 deletions(-)

-- 
Trond Myklebust
Linux NFS client maintainer, Hammerspace
trondmy@kernel.org, trond.myklebust@hammerspace.com

Re: [GIT PULL] Please pull NFS client fixes for Linux 7.3

[Linus Torvalds]

On Wed, 26 Aug 2026 at 13:57, Trond Myklebust <trondmy@kernel.org> wrote:
>
>
> Highlights include:
>
> Stable fixes:
> - SunRPC: Use-after-free fixes for the sunrpc client code
> - NFSv4: Delegation hash table leak
> - lockd: NULL dereference on lockowner allocation failure
[..]

Please give a real summary of what has happened, not just a lightly
edited git shortlog list sorted by "is stable cc'd or not".

And even if the end result is that "lightly edited shortlog", then at
least make it more useful and legible to users.

For example, just to look at that first bullet point: why have that
"SunRPC:" thing at all? It adds no value. This is a NFS client merge
message, and the line goes on to say "fixes for the sunrpc client
code".

Same foes for pretty much every single bullet entry - that first one
was not the one that stood out - they all stood out as "

The merge message shouldn't be about some minor details - that's what
the individual commits messages are for.

I did some more light editing to at least make it slightly more
readable, but it's still not exactly great.  Please don't make me do
that. Give me that overview of what happened and what matters. And
it's ok to say "And misc fixes" for things that really doesn't matter
and people who care should just look at the individual commits.

                  Linus

Re: [GIT PULL] Please pull NFS client fixes for Linux 7.3

[pr-tracker-bot]

The pull request you sent on Wed, 26 Aug 2026 16:57:09 -0400:

> git://git.linux-nfs.org/projects/trondmy/linux-nfs.git tags/nfs-for-7.3-1

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/73e3f0710014fe6d4ed98cfc02292f6121db7558

Thank you!

-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/prtracker.html