Linux filesystem development
 help / color / mirror / Atom feed
From: "Pankaj Raghav (Samsung)" <pankaj.raghav@linux.dev>
To: Christoph Hellwig <hch@lst.de>
Cc: Carlos Maiolino <cem@kernel.org>,
	linux-xfs@vger.kernel.org,  linux-fsdevel@vger.kernel.org
Subject: Re: [PATCH 02/25] iomap: add support for data checksumming
Date: Wed, 11 Nov 2026 12:21:07 +0100	[thread overview]
Message-ID: <avRO_O8jx3zrccGq@quentin> (raw)
In-Reply-To: <20261104085929.317544-3-hch@lst.de>

On Wed, Nov 04, 2026 at 09:57:06AM +0100, Christoph Hellwig wrote:
> Add a new refcounted structure for a data checksum attached to

I don't see any new refcounted structure. Probably the commit message is
stale?

> an iomap_ioend, and helpers to allocate, split and access it.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>
>  
>  /*
>   * Flags that if set on either ioend prevent the merge of two ioends.
> @@ -522,14 +529,31 @@ static inline u16 iomap_ioend_flags(const struct iomap *iomap)
>  struct iomap_ioend {
>  	struct list_head	io_list;	/* next ioend in chain */
>  	u16			io_flags;	/* IOMAP_IOEND_* */
> +	u8			io_csum_shift;	/* ilog(2) of csum size */
>  	u32			io_bvec_offset;	/* offset into first bvec */
>  	struct inode		*io_inode;	/* file being written to */
>  	size_t			io_size;	/* size of the extent */
>  	atomic_t		io_remaining;	/* completetion defer count */
>  	int			io_error;	/* stashed away status */
> -	struct iomap_ioend	*io_parent;	/* parent for completions */
> +
> +	/*
> +	 * There are two kinds of ioend, the original ones submitted by iomap,
> +	 * and thus split of for I/O size limitations.  For the latter,

s/thus split of/those split off

> +	 * io_parent points to the original ioend.  For the original ioends,
> +	 * the space used by the parent pointer is reused by the checksum
> +	 * pointer when data checksum are used.  Split ioends always point
> +	 * into the original allocation with their io_csum pointer.  As a
> +	 * special case tiny checksum use the pointer as actual checksum
> +	 * storage if that fits.
> +	 */
> +	union {
> +		struct iomap_ioend *io_parent;	/* parent for completions */
> +		void		*io_csum_alloc; /* original csum allocation. */
> +		u8		io_csum_inline[sizeof(void *)];
> +	};
>  	loff_t			io_offset;	/* offset in the file */
>  	sector_t		io_sector;	/* start sector of ioend */
> +	void			*io_csum;	/* data checksum */
>  	void			*io_private;	/* file system private data */
>  	struct fsverity_info	*io_vi;		/* fsverity info */
>  	struct bio		io_bio;		/* MUST BE LAST! */
> @@ -547,6 +571,22 @@ static inline struct iomap_ioend *iomap_ioend_from_bio(struct bio *bio)
>  	.bi_offset	= (_ioend)->io_bvec_offset,	\
>  }
>  

-- 
Pankaj

  reply	other threads:[~2026-11-11 11:21 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-04  8:57 support for RT data checksums v2 Christoph Hellwig
2026-11-04  8:57 ` [PATCH 01/25] block: export fs_bio_integrity_verify Christoph Hellwig
2026-11-04  8:57 ` [PATCH 02/25] iomap: add support for data checksumming Christoph Hellwig
2026-11-11 11:21   ` Pankaj Raghav (Samsung) [this message]
2026-11-11 17:38     ` Christoph Hellwig
2026-11-04  8:57 ` [PATCH 03/25] xfs: always call xfs_update_stable_writes in xfs_ioctl_setattr_xflags Christoph Hellwig
2026-11-11 11:11   ` Pankaj Raghav (Samsung)
2026-11-11 17:37     ` Christoph Hellwig
2026-11-04  8:57 ` [PATCH 04/25] xfs: use a separate lockdep class for all rmap inode locks Christoph Hellwig
2026-11-04  8:57 ` [PATCH 05/25] xfs: add a xfs_buf_read_async buffer cache API Christoph Hellwig
2026-11-04  8:57 ` [PATCH 06/25] xfs: add xfs_daddr_to_rgno and xfs_daddr_to_rgbno helpers Christoph Hellwig
2026-11-04  8:57 ` [PATCH 07/25] xfs: prepare xfs_rtfile_initialize_blocks for larger than FSB blocks Christoph Hellwig
2026-11-04  8:57 ` [PATCH 08/25] xfs: relase zi_open_zones_lock over xfs_open_zone_put on unmount Christoph Hellwig
2026-11-04  8:57 ` [PATCH 09/25] xfs: put the open zone outside RCU protection in xfs_get_cached_zone Christoph Hellwig
2026-11-04  8:57 ` [PATCH 10/25] xfs: define the RT data checksum on-disk format Christoph Hellwig
2026-11-04  8:57 ` [PATCH 11/25] xfs: add support for per-RTG csum files Christoph Hellwig
2026-11-04  8:57 ` [PATCH 12/25] xfs: calculate the log reservation for logging data checksum buffers Christoph Hellwig
2026-11-04  8:57 ` [PATCH 13/25] xfs: core RT data checksum support Christoph Hellwig
2026-11-04  8:57 ` [PATCH 14/25] xfs: data checksums require stable writes Christoph Hellwig
2026-11-04  8:57 ` [PATCH 15/25] xfs: require file system block size alignment when using data checksums Christoph Hellwig
2026-11-04  8:57 ` [PATCH 16/25] xfs: add support for reading with " Christoph Hellwig
2026-11-11 11:31   ` Pankaj Raghav (Samsung)
2026-11-11 17:38     ` Christoph Hellwig
2026-11-04  8:57 ` [PATCH 17/25] xfs: add support for writing " Christoph Hellwig
2026-11-04  8:57 ` [PATCH 18/25] xfs: add data checksum support to zoned garbage collection Christoph Hellwig
2026-11-04  8:57 ` [PATCH 19/25] xfs: simplify range tracking in xfs_verify_media Christoph Hellwig
2026-11-04  8:57 ` [PATCH 20/25] xfs: verify data checksums during media verification Christoph Hellwig
2026-11-11 11:57   ` Pankaj Raghav (Samsung)
2026-11-11 17:39     ` Christoph Hellwig
2026-11-04  8:57 ` [PATCH 21/25] xfs: don't try to verify checksums on empty zones Christoph Hellwig
2026-11-04  8:57 ` [PATCH 22/25] xfs: report RT data checksum information via XFS_FSOP_GEOM Christoph Hellwig
2026-11-04  8:57 ` [PATCH 23/25] xfs: add an experimental feature warning for RT data checksums Christoph Hellwig
2026-11-04  8:57 ` [PATCH 24/25] xfs: enable " Christoph Hellwig
2026-11-04  8:57 ` [PATCH 25/25] xfs: add a RT checksum log item Christoph Hellwig

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=avRO_O8jx3zrccGq@quentin \
    --to=pankaj.raghav@linux.dev \
    --cc=cem@kernel.org \
    --cc=hch@lst.de \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-xfs@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox