Received-SPF: pass (google.com: domain of linux-kernel+bounces-94579-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Date: Thu, 7 Mar 2024 08:14:48 +1100
From: Dave Chinner <david@fromorbit.com>
To: John Garry <john.g.garry@oracle.com>
Cc: djwong@kernel.org, hch@lst.de, viro@zeniv.linux.org.uk,
	brauner@kernel.org, jack@suse.cz, chandan.babu@oracle.com,
	axboe@kernel.dk, martin.petersen@oracle.com,
	linux-kernel@vger.kernel.org, linux-xfs@vger.kernel.org,
	linux-fsdevel@vger.kernel.org, tytso@mit.edu, jbongio@google.com,
	ojaswin@linux.ibm.com, ritesh.list@gmail.com,
	linux-block@vger.kernel.org
Subject: Re: [PATCH v2 07/14] fs: iomap: Sub-extent zeroing
Message-ID: <ZejcyPPX3bVdvJyV@dread.disaster.area>
References: <20240304130428.13026-1-john.g.garry@oracle.com>
 <20240304130428.13026-8-john.g.garry@oracle.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20240304130428.13026-8-john.g.garry@oracle.com>

On Mon, Mar 04, 2024 at 01:04:21PM +0000, John Garry wrote:
> For FS_XFLAG_FORCEALIGN support, we want to treat any sub-extent IO like
> sub-fsblock DIO, in that we will zero the sub-extent when the mapping is
> unwritten.
> 
> This will be important for atomic writes support, in that atomically
> writing over a partially written extent would mean that we would need to
> do the unwritten extent conversion write separately, and the write could
> no longer be atomic.
> 
> It is the task of the FS to set iomap.extent_shift per iter to indicate
> sub-extent zeroing required.
> 
> Maybe a macro like i_blocksize() should be introduced for extent sizes,
> instead of using extent_shift. It would also eliminate excessive use
> of xfs_get_extss() for XFS in future.
> 
> Signed-off-by: John Garry <john.g.garry@oracle.com>
> ---
>  fs/iomap/direct-io.c  | 14 ++++++++------
>  include/linux/iomap.h |  1 +
>  2 files changed, 9 insertions(+), 6 deletions(-)
> 
> diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
> index bcd3f8cf5ea4..733f83f839b6 100644
> --- a/fs/iomap/direct-io.c
> +++ b/fs/iomap/direct-io.c
> @@ -277,7 +277,7 @@ static loff_t iomap_dio_bio_iter(const struct iomap_iter *iter,
>  {
>  	const struct iomap *iomap = &iter->iomap;
>  	struct inode *inode = iter->inode;
> -	unsigned int fs_block_size = i_blocksize(inode), pad;
> +	unsigned int zeroing_size, pad;
>  	loff_t length = iomap_length(iter);
>  	loff_t pos = iter->pos;
>  	blk_opf_t bio_opf;
> @@ -288,6 +288,8 @@ static loff_t iomap_dio_bio_iter(const struct iomap_iter *iter,
>  	size_t copied = 0;
>  	size_t orig_count;
>  
> +	zeroing_size = i_blocksize(inode) << iomap->extent_shift;

The iomap interfaces use units of bytes for offsets, sizes, ranges,
etc. Using shifts to define a granularity value seems like a
throwback to decades old XFS code and just a bit weird nowdays.  Can
we just pass this as a byte count? i.e.:

	zeroing_size = i_blocksize(inode);
	if (iomap->extent_size)
		zeroing_size = iomap->extent_size;

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com