linux-xfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <darrick.wong@oracle.com>
To: Christoph Hellwig <hch@lst.de>
Cc: linux-xfs@vger.kernel.org
Subject: Re: [PATCH] xfs: only reclaim unwritten COW extents periodically
Date: Mon, 27 Feb 2017 21:26:51 -0800	[thread overview]
Message-ID: <20170228052651.GD5297@birch.djwong.org> (raw)
In-Reply-To: <20170228004910.9495-1-hch@lst.de>

On Mon, Feb 27, 2017 at 04:49:10PM -0800, Christoph Hellwig wrote:
> We only want to reclaim preallocations from our periodic work item.
> Currenrly this is archived by looking for a diry inode, but that check
> is rather fragile.  Instead add flags to the xfs_reflink_cancel_cow_* so
> that the caller can ask for just cancelling unwritten extents in the COw
> fork.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
>  fs/xfs/xfs_aops.c    |  2 +-
>  fs/xfs/xfs_icache.c  |  2 +-
>  fs/xfs/xfs_inode.c   |  2 +-
>  fs/xfs/xfs_reflink.c | 14 +++++++++-----
>  fs/xfs/xfs_reflink.h |  5 +++--
>  fs/xfs/xfs_super.c   |  2 +-
>  6 files changed, 16 insertions(+), 11 deletions(-)
> 
> diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
> index b480647e00e7..290c0fb306e7 100644
> --- a/fs/xfs/xfs_aops.c
> +++ b/fs/xfs/xfs_aops.c
> @@ -293,7 +293,7 @@ xfs_end_io(
>  			goto done;
>  		if (ioend->io_bio->bi_error) {
>  			error = xfs_reflink_cancel_cow_range(ip,
> -					ioend->io_offset, ioend->io_size);
> +					ioend->io_offset, ioend->io_size, 0);
>  			goto done;
>  		}
>  		error = xfs_reflink_end_cow(ip, ioend->io_offset,
> diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c
> index 7234b9748c36..1f7d158266c1 100644
> --- a/fs/xfs/xfs_icache.c
> +++ b/fs/xfs/xfs_icache.c
> @@ -1608,7 +1608,7 @@ xfs_inode_free_cowblocks(
>  	xfs_ilock(ip, XFS_IOLOCK_EXCL);
>  	xfs_ilock(ip, XFS_MMAPLOCK_EXCL);
>  
> -	ret = xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF);
> +	ret = xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, XFS_CC_UNWRITTEN);

Ok, so the periodic work item leaves real cow extents alone now, which
is good since real extents now represent pending cow operations.  Good.

>  	xfs_iunlock(ip, XFS_MMAPLOCK_EXCL);
>  	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
> diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
> index edfa6a55b064..164e74a511b1 100644
> --- a/fs/xfs/xfs_inode.c
> +++ b/fs/xfs/xfs_inode.c
> @@ -1615,7 +1615,7 @@ xfs_itruncate_extents(
>  
>  	/* Remove all pending CoW reservations. */
>  	error = xfs_reflink_cancel_cow_blocks(ip, &tp, first_unmap_block,
> -			last_block);
> +			last_block, 0);
>  	if (error)
>  		goto out;
>  
> diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c
> index 08ca9f81b087..83605af3b135 100644
> --- a/fs/xfs/xfs_reflink.c
> +++ b/fs/xfs/xfs_reflink.c
> @@ -551,7 +551,8 @@ xfs_reflink_cancel_cow_blocks(
>  	struct xfs_inode		*ip,
>  	struct xfs_trans		**tpp,
>  	xfs_fileoff_t			offset_fsb,
> -	xfs_fileoff_t			end_fsb)
> +	xfs_fileoff_t			end_fsb,
> +	unsigned			flags)
>  {
>  	struct xfs_ifork		*ifp = XFS_IFORK_PTR(ip, XFS_COW_FORK);
>  	struct xfs_bmbt_irec		got, del;
> @@ -575,7 +576,8 @@ xfs_reflink_cancel_cow_blocks(
>  					&idx, &got, &del);
>  			if (error)
>  				break;
> -		} else {
> +		} else if (del.br_state == XFS_EXT_UNWRITTEN ||
> +			   !(flags & XFS_CC_UNWRITTEN)) {
>  			xfs_trans_ijoin(*tpp, ip, 0);
>  			xfs_defer_init(&dfops, &firstfsb);
>  
> @@ -623,7 +625,8 @@ int
>  xfs_reflink_cancel_cow_range(
>  	struct xfs_inode	*ip,
>  	xfs_off_t		offset,
> -	xfs_off_t		count)
> +	xfs_off_t		count,
> +	unsigned		flags)
>  {
>  	struct xfs_trans	*tp;
>  	xfs_fileoff_t		offset_fsb;
> @@ -649,7 +652,8 @@ xfs_reflink_cancel_cow_range(
>  	xfs_trans_ijoin(tp, ip, 0);
>  
>  	/* Scrape out the old CoW reservations */
> -	error = xfs_reflink_cancel_cow_blocks(ip, &tp, offset_fsb, end_fsb);
> +	error = xfs_reflink_cancel_cow_blocks(ip, &tp, offset_fsb, end_fsb,
> +			flags);
>  	if (error)
>  		goto out_cancel;
>  
> @@ -1437,7 +1441,7 @@ xfs_reflink_clear_inode_flag(
>  	 * We didn't find any shared blocks so turn off the reflink flag.
>  	 * First, get rid of any leftover CoW mappings.
>  	 */
> -	error = xfs_reflink_cancel_cow_blocks(ip, tpp, 0, NULLFILEOFF);
> +	error = xfs_reflink_cancel_cow_blocks(ip, tpp, 0, NULLFILEOFF, 0);
>  	if (error)
>  		return error;
>  
> diff --git a/fs/xfs/xfs_reflink.h b/fs/xfs/xfs_reflink.h
> index 33ac9b8db683..9416279b3c89 100644
> --- a/fs/xfs/xfs_reflink.h
> +++ b/fs/xfs/xfs_reflink.h
> @@ -37,11 +37,12 @@ extern bool xfs_reflink_find_cow_mapping(struct xfs_inode *ip, xfs_off_t offset,
>  extern void xfs_reflink_trim_irec_to_next_cow(struct xfs_inode *ip,
>  		xfs_fileoff_t offset_fsb, struct xfs_bmbt_irec *imap);
>  
> +#define XFS_CC_UNWRITTEN	0x01

Could we have a comment here about what CC_UNWRITTEN means?

I /think/ it means "don't remove real extents", since when it's specified
delalloc reservations still get removed by _reflink_cancel_cow_*, correct?

Somewhere we ought to record the fact that flags == 0 removes everything.

--D

>  extern int xfs_reflink_cancel_cow_blocks(struct xfs_inode *ip,
>  		struct xfs_trans **tpp, xfs_fileoff_t offset_fsb,
> -		xfs_fileoff_t end_fsb);
> +		xfs_fileoff_t end_fsb, unsigned flags);
>  extern int xfs_reflink_cancel_cow_range(struct xfs_inode *ip, xfs_off_t offset,
> -		xfs_off_t count);
> +		xfs_off_t count, unsigned flags);
>  extern int xfs_reflink_end_cow(struct xfs_inode *ip, xfs_off_t offset,
>  		xfs_off_t count);
>  extern int xfs_reflink_recover_cow(struct xfs_mount *mp);
> diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
> index 890862f2447c..9136854030d5 100644
> --- a/fs/xfs/xfs_super.c
> +++ b/fs/xfs/xfs_super.c
> @@ -953,7 +953,7 @@ xfs_fs_destroy_inode(
>  	XFS_STATS_INC(ip->i_mount, vn_remove);
>  
>  	if (xfs_is_reflink_inode(ip)) {
> -		error = xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF);
> +		error = xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, 0);
>  		if (error && !XFS_FORCED_SHUTDOWN(ip->i_mount))
>  			xfs_warn(ip->i_mount,
>  "Error %d while evicting CoW blocks for inode %llu.",
> -- 
> 2.11.0
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

  reply	other threads:[~2017-02-28  5:28 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-02-28  0:49 [PATCH] xfs: only reclaim unwritten COW extents periodically Christoph Hellwig
2017-02-28  5:26 ` Darrick J. Wong [this message]
2017-03-07  0:17   ` Darrick J. Wong
2017-03-07  5:00     ` 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=20170228052651.GD5297@birch.djwong.org \
    --to=darrick.wong@oracle.com \
    --cc=hch@lst.de \
    --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;
as well as URLs for NNTP newsgroup(s).