linux-xfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Chandan Rajendra <chandan@linux.ibm.com>
To: "Darrick J. Wong" <darrick.wong@oracle.com>
Cc: linux-xfs@vger.kernel.org
Subject: Re: [PATCH 07/21] xfs: refactor log recovery icreate item dispatch for pass2 commit functions
Date: Fri, 01 May 2020 19:48:16 +0530	[thread overview]
Message-ID: <69061559.EVSPQyX58B@localhost.localdomain> (raw)
In-Reply-To: <158820770082.467894.7639027771832654915.stgit@magnolia>

On Thursday, April 30, 2020 6:18 AM Darrick J. Wong wrote: 
> From: Darrick J. Wong <darrick.wong@oracle.com>
> 
> Move the log icreate item pass2 commit code into the per-item source code
> files and use the dispatch function to call it.  We do these one at a
> time because there's a lot of code to move.  No functional changes.
>

The changes look good to me.

Reviewed-by: Chandan Rajendra <chandanrlinux@gmail.com>

> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> ---
>  fs/xfs/xfs_icreate_item.c |  132 +++++++++++++++++++++++++++++++++++++++++++++
>  fs/xfs/xfs_log_recover.c  |  126 -------------------------------------------
>  2 files changed, 132 insertions(+), 126 deletions(-)
> 
> 
> diff --git a/fs/xfs/xfs_icreate_item.c b/fs/xfs/xfs_icreate_item.c
> index 9f38a3c200a3..602a8c91371f 100644
> --- a/fs/xfs/xfs_icreate_item.c
> +++ b/fs/xfs/xfs_icreate_item.c
> @@ -6,13 +6,19 @@
>  #include "xfs.h"
>  #include "xfs_fs.h"
>  #include "xfs_shared.h"
> +#include "xfs_format.h"
>  #include "xfs_log_format.h"
> +#include "xfs_trans_resv.h"
> +#include "xfs_mount.h"
> +#include "xfs_inode.h"
>  #include "xfs_trans.h"
>  #include "xfs_trans_priv.h"
>  #include "xfs_icreate_item.h"
>  #include "xfs_log.h"
>  #include "xfs_log_priv.h"
>  #include "xfs_log_recover.h"
> +#include "xfs_ialloc.h"
> +#include "xfs_trace.h"
>  
>  kmem_zone_t	*xfs_icreate_zone;		/* inode create item zone */
>  
> @@ -117,6 +123,132 @@ xlog_icreate_reorder(
>  	return XLOG_REORDER_BUFFER_LIST;
>  }
>  
> +/*
> + * This routine is called when an inode create format structure is found in a
> + * committed transaction in the log.  It's purpose is to initialise the inodes
> + * being allocated on disk. This requires us to get inode cluster buffers that
> + * match the range to be initialised, stamped with inode templates and written
> + * by delayed write so that subsequent modifications will hit the cached buffer
> + * and only need writing out at the end of recovery.
> + */
> +STATIC int
> +xlog_recover_do_icreate_commit_pass2(
> +	struct xlog			*log,
> +	struct list_head		*buffer_list,
> +	struct xlog_recover_item	*item,
> +	xfs_lsn_t			lsn)
> +{
> +	struct xfs_mount		*mp = log->l_mp;
> +	struct xfs_icreate_log		*icl;
> +	struct xfs_ino_geometry		*igeo = M_IGEO(mp);
> +	xfs_agnumber_t			agno;
> +	xfs_agblock_t			agbno;
> +	unsigned int			count;
> +	unsigned int			isize;
> +	xfs_agblock_t			length;
> +	int				bb_per_cluster;
> +	int				cancel_count;
> +	int				nbufs;
> +	int				i;
> +
> +	icl = (struct xfs_icreate_log *)item->ri_buf[0].i_addr;
> +	if (icl->icl_type != XFS_LI_ICREATE) {
> +		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad type");
> +		return -EINVAL;
> +	}
> +
> +	if (icl->icl_size != 1) {
> +		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad icl size");
> +		return -EINVAL;
> +	}
> +
> +	agno = be32_to_cpu(icl->icl_ag);
> +	if (agno >= mp->m_sb.sb_agcount) {
> +		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad agno");
> +		return -EINVAL;
> +	}
> +	agbno = be32_to_cpu(icl->icl_agbno);
> +	if (!agbno || agbno == NULLAGBLOCK || agbno >= mp->m_sb.sb_agblocks) {
> +		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad agbno");
> +		return -EINVAL;
> +	}
> +	isize = be32_to_cpu(icl->icl_isize);
> +	if (isize != mp->m_sb.sb_inodesize) {
> +		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad isize");
> +		return -EINVAL;
> +	}
> +	count = be32_to_cpu(icl->icl_count);
> +	if (!count) {
> +		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad count");
> +		return -EINVAL;
> +	}
> +	length = be32_to_cpu(icl->icl_length);
> +	if (!length || length >= mp->m_sb.sb_agblocks) {
> +		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad length");
> +		return -EINVAL;
> +	}
> +
> +	/*
> +	 * The inode chunk is either full or sparse and we only support
> +	 * m_ino_geo.ialloc_min_blks sized sparse allocations at this time.
> +	 */
> +	if (length != igeo->ialloc_blks &&
> +	    length != igeo->ialloc_min_blks) {
> +		xfs_warn(log->l_mp,
> +			 "%s: unsupported chunk length", __FUNCTION__);
> +		return -EINVAL;
> +	}
> +
> +	/* verify inode count is consistent with extent length */
> +	if ((count >> mp->m_sb.sb_inopblog) != length) {
> +		xfs_warn(log->l_mp,
> +			 "%s: inconsistent inode count and chunk length",
> +			 __FUNCTION__);
> +		return -EINVAL;
> +	}
> +
> +	/*
> +	 * The icreate transaction can cover multiple cluster buffers and these
> +	 * buffers could have been freed and reused. Check the individual
> +	 * buffers for cancellation so we don't overwrite anything written after
> +	 * a cancellation.
> +	 */
> +	bb_per_cluster = XFS_FSB_TO_BB(mp, igeo->blocks_per_cluster);
> +	nbufs = length / igeo->blocks_per_cluster;
> +	for (i = 0, cancel_count = 0; i < nbufs; i++) {
> +		xfs_daddr_t	daddr;
> +
> +		daddr = XFS_AGB_TO_DADDR(mp, agno,
> +				agbno + i * igeo->blocks_per_cluster);
> +		if (xlog_is_buffer_cancelled(log, daddr, bb_per_cluster))
> +			cancel_count++;
> +	}
> +
> +	/*
> +	 * We currently only use icreate for a single allocation at a time. This
> +	 * means we should expect either all or none of the buffers to be
> +	 * cancelled. Be conservative and skip replay if at least one buffer is
> +	 * cancelled, but warn the user that something is awry if the buffers
> +	 * are not consistent.
> +	 *
> +	 * XXX: This must be refined to only skip cancelled clusters once we use
> +	 * icreate for multiple chunk allocations.
> +	 */
> +	ASSERT(!cancel_count || cancel_count == nbufs);
> +	if (cancel_count) {
> +		if (cancel_count != nbufs)
> +			xfs_warn(mp,
> +	"WARNING: partial inode chunk cancellation, skipped icreate.");
> +		trace_xfs_log_recover_icreate_cancel(log, icl);
> +		return 0;
> +	}
> +
> +	trace_xfs_log_recover_icreate_recover(log, icl);
> +	return xfs_ialloc_inode_init(mp, NULL, buffer_list, count, agno, agbno,
> +				     length, be32_to_cpu(icl->icl_gen));
> +}
> +
>  const struct xlog_recover_item_type xlog_icreate_item_type = {
>  	.reorder_fn		= xlog_icreate_reorder,
> +	.commit_pass2_fn	= xlog_recover_do_icreate_commit_pass2,
>  };
> diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
> index 58a54d9e6847..6ba3d64d08de 100644
> --- a/fs/xfs/xfs_log_recover.c
> +++ b/fs/xfs/xfs_log_recover.c
> @@ -2489,130 +2489,6 @@ xlog_recover_bud_pass2(
>  	return 0;
>  }
>  
> -/*
> - * This routine is called when an inode create format structure is found in a
> - * committed transaction in the log.  It's purpose is to initialise the inodes
> - * being allocated on disk. This requires us to get inode cluster buffers that
> - * match the range to be initialised, stamped with inode templates and written
> - * by delayed write so that subsequent modifications will hit the cached buffer
> - * and only need writing out at the end of recovery.
> - */
> -STATIC int
> -xlog_recover_do_icreate_pass2(
> -	struct xlog		*log,
> -	struct list_head	*buffer_list,
> -	xlog_recover_item_t	*item)
> -{
> -	struct xfs_mount	*mp = log->l_mp;
> -	struct xfs_icreate_log	*icl;
> -	struct xfs_ino_geometry	*igeo = M_IGEO(mp);
> -	xfs_agnumber_t		agno;
> -	xfs_agblock_t		agbno;
> -	unsigned int		count;
> -	unsigned int		isize;
> -	xfs_agblock_t		length;
> -	int			bb_per_cluster;
> -	int			cancel_count;
> -	int			nbufs;
> -	int			i;
> -
> -	icl = (struct xfs_icreate_log *)item->ri_buf[0].i_addr;
> -	if (icl->icl_type != XFS_LI_ICREATE) {
> -		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad type");
> -		return -EINVAL;
> -	}
> -
> -	if (icl->icl_size != 1) {
> -		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad icl size");
> -		return -EINVAL;
> -	}
> -
> -	agno = be32_to_cpu(icl->icl_ag);
> -	if (agno >= mp->m_sb.sb_agcount) {
> -		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad agno");
> -		return -EINVAL;
> -	}
> -	agbno = be32_to_cpu(icl->icl_agbno);
> -	if (!agbno || agbno == NULLAGBLOCK || agbno >= mp->m_sb.sb_agblocks) {
> -		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad agbno");
> -		return -EINVAL;
> -	}
> -	isize = be32_to_cpu(icl->icl_isize);
> -	if (isize != mp->m_sb.sb_inodesize) {
> -		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad isize");
> -		return -EINVAL;
> -	}
> -	count = be32_to_cpu(icl->icl_count);
> -	if (!count) {
> -		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad count");
> -		return -EINVAL;
> -	}
> -	length = be32_to_cpu(icl->icl_length);
> -	if (!length || length >= mp->m_sb.sb_agblocks) {
> -		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad length");
> -		return -EINVAL;
> -	}
> -
> -	/*
> -	 * The inode chunk is either full or sparse and we only support
> -	 * m_ino_geo.ialloc_min_blks sized sparse allocations at this time.
> -	 */
> -	if (length != igeo->ialloc_blks &&
> -	    length != igeo->ialloc_min_blks) {
> -		xfs_warn(log->l_mp,
> -			 "%s: unsupported chunk length", __FUNCTION__);
> -		return -EINVAL;
> -	}
> -
> -	/* verify inode count is consistent with extent length */
> -	if ((count >> mp->m_sb.sb_inopblog) != length) {
> -		xfs_warn(log->l_mp,
> -			 "%s: inconsistent inode count and chunk length",
> -			 __FUNCTION__);
> -		return -EINVAL;
> -	}
> -
> -	/*
> -	 * The icreate transaction can cover multiple cluster buffers and these
> -	 * buffers could have been freed and reused. Check the individual
> -	 * buffers for cancellation so we don't overwrite anything written after
> -	 * a cancellation.
> -	 */
> -	bb_per_cluster = XFS_FSB_TO_BB(mp, igeo->blocks_per_cluster);
> -	nbufs = length / igeo->blocks_per_cluster;
> -	for (i = 0, cancel_count = 0; i < nbufs; i++) {
> -		xfs_daddr_t	daddr;
> -
> -		daddr = XFS_AGB_TO_DADDR(mp, agno,
> -				agbno + i * igeo->blocks_per_cluster);
> -		if (xlog_is_buffer_cancelled(log, daddr, bb_per_cluster))
> -			cancel_count++;
> -	}
> -
> -	/*
> -	 * We currently only use icreate for a single allocation at a time. This
> -	 * means we should expect either all or none of the buffers to be
> -	 * cancelled. Be conservative and skip replay if at least one buffer is
> -	 * cancelled, but warn the user that something is awry if the buffers
> -	 * are not consistent.
> -	 *
> -	 * XXX: This must be refined to only skip cancelled clusters once we use
> -	 * icreate for multiple chunk allocations.
> -	 */
> -	ASSERT(!cancel_count || cancel_count == nbufs);
> -	if (cancel_count) {
> -		if (cancel_count != nbufs)
> -			xfs_warn(mp,
> -	"WARNING: partial inode chunk cancellation, skipped icreate.");
> -		trace_xfs_log_recover_icreate_cancel(log, icl);
> -		return 0;
> -	}
> -
> -	trace_xfs_log_recover_icreate_recover(log, icl);
> -	return xfs_ialloc_inode_init(mp, NULL, buffer_list, count, agno, agbno,
> -				     length, be32_to_cpu(icl->icl_gen));
> -}
> -
>  STATIC int
>  xlog_recover_commit_pass1(
>  	struct xlog			*log,
> @@ -2662,8 +2538,6 @@ xlog_recover_commit_pass2(
>  		return xlog_recover_bui_pass2(log, item, trans->r_lsn);
>  	case XFS_LI_BUD:
>  		return xlog_recover_bud_pass2(log, item);
> -	case XFS_LI_ICREATE:
> -		return xlog_recover_do_icreate_pass2(log, buffer_list, item);
>  	case XFS_LI_QUOTAOFF:
>  		/* nothing to do in pass2 */
>  		return 0;
> 
> 


-- 
chandan




  reply	other threads:[~2020-05-01 14:15 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-30  0:47 [PATCH v2 00/21] xfs: refactor log recovery Darrick J. Wong
2020-04-30  0:47 ` [PATCH 01/21] xfs: refactor log recovery item sorting into a generic dispatch structure Darrick J. Wong
2020-04-30  5:53   ` Christoph Hellwig
2020-04-30 15:08     ` Darrick J. Wong
2020-04-30 18:16       ` Darrick J. Wong
2020-05-01  8:08         ` Christoph Hellwig
2020-05-01 10:40   ` Chandan Rajendra
2020-04-30  0:47 ` [PATCH 02/21] xfs: refactor log recovery item dispatch for pass2 readhead functions Darrick J. Wong
2020-05-01 12:10   ` Chandan Rajendra
2020-04-30  0:47 ` [PATCH 03/21] xfs: refactor log recovery item dispatch for pass1 commit functions Darrick J. Wong
2020-04-30  0:48 ` [PATCH 04/21] xfs: refactor log recovery buffer item dispatch for pass2 " Darrick J. Wong
2020-05-01 13:43   ` Chandan Rajendra
2020-04-30  0:48 ` [PATCH 05/21] xfs: refactor log recovery inode " Darrick J. Wong
2020-05-01 14:03   ` Chandan Rajendra
2020-04-30  0:48 ` [PATCH 06/21] xfs: refactor log recovery dquot " Darrick J. Wong
2020-05-01 14:14   ` Chandan Rajendra
2020-04-30  0:48 ` [PATCH 07/21] xfs: refactor log recovery icreate " Darrick J. Wong
2020-05-01 14:18   ` Chandan Rajendra [this message]
2020-04-30  0:48 ` [PATCH 08/21] xfs: remove log recovery quotaoff " Darrick J. Wong
2020-05-01 15:09   ` Chandan Rajendra
2020-05-01 17:41     ` Darrick J. Wong
2020-04-30  0:48 ` [PATCH 09/21] xfs: refactor log recovery EFI " Darrick J. Wong
2020-05-01 10:28   ` Christoph Hellwig
2020-05-01 17:56     ` Darrick J. Wong
2020-04-30  0:48 ` [PATCH 10/21] xfs: refactor log recovery RUI " Darrick J. Wong
2020-04-30  0:48 ` [PATCH 11/21] xfs: refactor log recovery CUI " Darrick J. Wong
2020-04-30  0:48 ` [PATCH 12/21] xfs: refactor log recovery BUI " Darrick J. Wong
2020-04-30  0:48 ` [PATCH 13/21] xfs: refactor recovered EFI log item playback Darrick J. Wong
2020-05-01 10:19   ` Christoph Hellwig
2020-05-01 17:58     ` Darrick J. Wong
2020-04-30  0:49 ` [PATCH 14/21] xfs: refactor recovered RUI " Darrick J. Wong
2020-04-30  0:49 ` [PATCH 15/21] xfs: refactor recovered CUI " Darrick J. Wong
2020-04-30  0:49 ` [PATCH 16/21] xfs: refactor recovered BUI " Darrick J. Wong
2020-04-30  0:49 ` [PATCH 17/21] xfs: refactor releasing finished intents during log recovery Darrick J. Wong
2020-04-30  0:49 ` [PATCH 18/21] xfs: refactor adding recovered intent items to the log Darrick J. Wong
2020-04-30  0:49 ` [PATCH 19/21] xfs: refactor intent item RECOVERED flag into the log item Darrick J. Wong
2020-04-30  0:49 ` [PATCH 20/21] xfs: refactor intent item iop_recover calls Darrick J. Wong
2020-04-30  0:49 ` [PATCH 21/21] xfs: remove unnecessary includes from xfs_log_recover.c Darrick J. Wong
2020-05-01 10:15 ` [PATCH v2 00/21] xfs: refactor log recovery Christoph Hellwig
2020-05-01 16:53   ` Darrick J. Wong
2020-05-01 17:03     ` 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=69061559.EVSPQyX58B@localhost.localdomain \
    --to=chandan@linux.ibm.com \
    --cc=darrick.wong@oracle.com \
    --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).