Re: [PATCH 6.6 CANDIDATE 00/24] xfs backports for 6.6.y (from 6.12)

["Darrick J. Wong" <djwong@kernel.org>]

On Wed, Feb 05, 2025 at 09:12:07AM +0100, Carlos Maiolino wrote:
> On Tue, Feb 04, 2025 at 07:07:08PM -0800, Catherine Hoang wrote:
> > Hi all,
> > 
> > This series contains backports for 6.6 from the 6.12 release. Tested on
> > 30 runs of kdevops with the following configurations:
> 
> Hi Catherine.
> 
> I believe the idea of xfs-stable creation was exactly to not flood linux-xfs
> with LTS kernel backports?
> 
> It's not a big deal, but IIRC, the whole idea of xfs-stable was to reduce
> linux-xfs traffic. If we keep Cc'ing both lists on those backports, xfs-stable
> kind of lose its meaning?

Yes, stable patches should be sent to xfs-stable and not linux-xfs.

(Says me who's guilty of doing exactly the same thing.)

--D

> Cheers,
> Carlos
> 
> > 
> > 1. CRC
> > 2. No CRC (512 and 4k block size)
> > 3. Reflink (1K and 4k block size)
> > 4. Reflink without rmapbt
> > 5. External log device
> > 
> > Andrew Kreimer (1):
> >   xfs: fix a typo
> > 
> > Brian Foster (2):
> >   xfs: skip background cowblock trims on inodes open for write
> >   xfs: don't free cowblocks from under dirty pagecache on unshare
> > 
> > Chi Zhiling (1):
> >   xfs: Reduce unnecessary searches when searching for the best extents
> > 
> > Christoph Hellwig (15):
> >   xfs: assert a valid limit in xfs_rtfind_forw
> >   xfs: merge xfs_attr_leaf_try_add into xfs_attr_leaf_addname
> >   xfs: return bool from xfs_attr3_leaf_add
> >   xfs: distinguish extra split from real ENOSPC from
> >     xfs_attr3_leaf_split
> >   xfs: distinguish extra split from real ENOSPC from
> >     xfs_attr_node_try_addname
> >   xfs: fold xfs_bmap_alloc_userdata into xfs_bmapi_allocate
> >   xfs: don't ifdef around the exact minlen allocations
> >   xfs: call xfs_bmap_exact_minlen_extent_alloc from xfs_bmap_btalloc
> >   xfs: support lowmode allocations in xfs_bmap_exact_minlen_extent_alloc
> >   xfs: pass the exact range to initialize to xfs_initialize_perag
> >   xfs: update the file system geometry after recoverying superblock
> >     buffers
> >   xfs: error out when a superblock buffer update reduces the agcount
> >   xfs: don't use __GFP_RETRY_MAYFAIL in xfs_initialize_perag
> >   xfs: update the pag for the last AG at recovery time
> >   xfs: streamline xfs_filestream_pick_ag
> > 
> > Darrick J. Wong (2):
> >   xfs: validate inumber in xfs_iget
> >   xfs: fix a sloppy memory handling bug in xfs_iroot_realloc
> > 
> > Ojaswin Mujoo (1):
> >   xfs: Check for delayed allocations before setting extsize
> > 
> > Uros Bizjak (1):
> >   xfs: Use try_cmpxchg() in xlog_cil_insert_pcp_aggregate()
> > 
> > Zhang Zekun (1):
> >   xfs: Remove empty declartion in header file
> > 
> >  fs/xfs/libxfs/xfs_ag.c         |  47 ++++----
> >  fs/xfs/libxfs/xfs_ag.h         |   6 +-
> >  fs/xfs/libxfs/xfs_alloc.c      |   9 +-
> >  fs/xfs/libxfs/xfs_alloc.h      |   4 +-
> >  fs/xfs/libxfs/xfs_attr.c       | 190 ++++++++++++++-------------------
> >  fs/xfs/libxfs/xfs_attr_leaf.c  |  40 +++----
> >  fs/xfs/libxfs/xfs_attr_leaf.h  |   2 +-
> >  fs/xfs/libxfs/xfs_bmap.c       | 140 ++++++++----------------
> >  fs/xfs/libxfs/xfs_da_btree.c   |   5 +-
> >  fs/xfs/libxfs/xfs_inode_fork.c |  10 +-
> >  fs/xfs/libxfs/xfs_rtbitmap.c   |   2 +
> >  fs/xfs/xfs_buf_item_recover.c  |  70 ++++++++++++
> >  fs/xfs/xfs_filestream.c        |  96 ++++++++---------
> >  fs/xfs/xfs_fsops.c             |  18 ++--
> >  fs/xfs/xfs_icache.c            |  39 ++++---
> >  fs/xfs/xfs_inode.c             |   2 +-
> >  fs/xfs/xfs_inode.h             |   5 +
> >  fs/xfs/xfs_ioctl.c             |   4 +-
> >  fs/xfs/xfs_log.h               |   1 -
> >  fs/xfs/xfs_log_cil.c           |  11 +-
> >  fs/xfs/xfs_log_recover.c       |   9 +-
> >  fs/xfs/xfs_mount.c             |   4 +-
> >  fs/xfs/xfs_reflink.c           |   3 +
> >  fs/xfs/xfs_reflink.h           |  19 ++++
> >  24 files changed, 375 insertions(+), 361 deletions(-)
> > 
> > --
> > 2.39.3
> > 
> > 
>