linux-xfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <darrick.wong@oracle.com>
To: Brian Foster <bfoster@redhat.com>
Cc: sandeen@sandeen.net, linux-xfs@vger.kernel.org
Subject: Re: [PATCH 3/9] xfs_repair: create a new class of btree rebuild cursors
Date: Thu, 14 May 2020 12:47:01 -0700	[thread overview]
Message-ID: <20200514194701.GG6714@magnolia> (raw)
In-Reply-To: <20200514151107.GC50849@bfoster>

On Thu, May 14, 2020 at 11:11:07AM -0400, Brian Foster wrote:
> On Sat, May 09, 2020 at 09:31:59AM -0700, Darrick J. Wong wrote:
> > From: Darrick J. Wong <darrick.wong@oracle.com>
> > 
> > Create some new support structures and functions to assist phase5 in
> > using the btree bulk loader to reconstruct metadata btrees.  This is the
> > first step in removing the open-coded rebuilding code.
> > 
> > Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> > ---
> 
> FYI, unused variable warnings:
> 
> phase5.c: In function ‘phase5_func’:
> phase5.c:2491:20: warning: unused variable ‘sc’ [-Wunused-variable]
>  2491 |  struct repair_ctx sc = { .mp = mp, };
>       |                    ^~
> At top level:
> phase5.c:509:1: warning: ‘finish_rebuild’ defined but not used [-Wunused-function]
>   509 | finish_rebuild(
>       | ^~~~~~~~~~~~~~
> phase5.c:468:1: warning: ‘rebuild_alloc_block’ defined but not used [-Wunused-function]
>   468 | rebuild_alloc_block(
>       | ^~~~~~~~~~~~~~~~~~~
> phase5.c:381:1: warning: ‘setup_rebuild’ defined but not used [-Wunused-function]
>   381 | setup_rebuild(
>       | ^~~~~~~~~~~~~
> phase5.c:366:1: warning: ‘init_rebuild’ defined but not used [-Wunused-function]
>   366 | init_rebuild(
>       | ^~~~~~~~~~~~

Yeah... these aren't used by anything until the next patch, and I was
trying to keep the "infrastructure you need for X" changes separate from
"X".  Though every time I do that, people complain about me adding code
that isn't used anywhere...

> >  repair/phase5.c |  240 ++++++++++++++++++++++++++++++++++++++++++++++++++-----
> >  1 file changed, 219 insertions(+), 21 deletions(-)
> > 
> > 
> > diff --git a/repair/phase5.c b/repair/phase5.c
> > index f3be15de..7eb24519 100644
> > --- a/repair/phase5.c
> > +++ b/repair/phase5.c
> ...
> > @@ -306,6 +324,157 @@ _("error - not enough free space in filesystem\n"));
> ...
> > +/* Reserve blocks for the new btree. */
> > +static void
> > +setup_rebuild(
> > +	struct xfs_mount	*mp,
> > +	xfs_agnumber_t		agno,
> > +	struct bt_rebuild	*btr,
> > +	uint32_t		nr_blocks)
> > +{
> > +	struct extent_tree_node	*ext_ptr;
> > +	struct extent_tree_node	*bno_ext_ptr;
> > +	uint32_t		blocks_allocated = 0;
> > +	int			error;
> > +
> > +	/*
> > +	 * grab the smallest extent and use it up, then get the
> > +	 * next smallest.  This mimics the init_*_cursor code.
> > +	 */
> > +	ext_ptr =  findfirst_bcnt_extent(agno);
> > +
> > +	/*
> > +	 * set up the free block array
> > +	 */
> > +	while (blocks_allocated < nr_blocks)  {
> > +		uint64_t	len;
> > +		xfs_agblock_t	new_start;
> > +		xfs_extlen_t	new_len;
> > +
> > +		if (!ext_ptr)
> > +			do_error(
> > +_("error - not enough free space in filesystem\n"));
> > +
> > +		/* Use up the extent we've got. */
> > +		len = min(ext_ptr->ex_blockcount,
> > +				btr->bload.nr_blocks - blocks_allocated);
> 
> What's the difference between the nr_blocks parameter and this one?

I think that's a bug, and should have been:

	len = min(ext_ptr->ex_blockcount, nr_blocks - blocks_allocated);

> > +		error = xrep_newbt_add_reservation(&btr->newbt,
> > +				XFS_AGB_TO_FSB(mp, agno,
> > +					       ext_ptr->ex_startblock),
> > +				len, NULL);
> > +		if (error)
> > +			do_error(_("could not set up btree reservation: %s\n"),
> > +				strerror(-error));
> > +		blocks_allocated += len;
> > +
> > +		error = rmap_add_ag_rec(mp, agno, ext_ptr->ex_startblock, len,
> > +				btr->newbt.oinfo.oi_owner);
> > +		if (error)
> > +			do_error(_("could not set up btree rmaps: %s\n"),
> > +				strerror(-error));
> > +
> > +		/* Figure out if we're putting anything back. */
> 
> The remaining extent replacement bits of this loop looks like it could
> warrant a little helper and a comment to explain exactly what's
> happening at a high level.

Ok, I'll figure something out.

--D

> Brian
> 
> > +		new_start = ext_ptr->ex_startblock + len;
> > +		new_len = ext_ptr->ex_blockcount - len;
> > +
> > +		/* Delete the used-up extent from both extent trees. */
> > +#ifdef XR_BLD_FREE_TRACE
> > +		fprintf(stderr, "releasing extent: %u [%u %u]\n",
> > +			agno, ext_ptr->ex_startblock, ext_ptr->ex_blockcount);
> > +#endif
> > +		bno_ext_ptr = find_bno_extent(agno, ext_ptr->ex_startblock);
> > +		ASSERT(bno_ext_ptr != NULL);
> > +		get_bno_extent(agno, bno_ext_ptr);
> > +		release_extent_tree_node(bno_ext_ptr);
> > +
> > +		ext_ptr = get_bcnt_extent(agno, ext_ptr->ex_startblock,
> > +				ext_ptr->ex_blockcount);
> > +		ASSERT(ext_ptr != NULL);
> > +		release_extent_tree_node(ext_ptr);
> > +
> > +		/*
> > +		 * If we only used part of this last extent, then we need only
> > +		 * to reinsert the extent in the extent trees and we're done.
> > +		 */
> > +		if (new_len > 0) {
> > +			add_bno_extent(agno, new_start, new_len);
> > +			add_bcnt_extent(agno, new_start, new_len);
> > +			break;
> > +		}
> > +
> > +		/* Otherwise, find the next biggest extent. */
> > +		ext_ptr = findfirst_bcnt_extent(agno);
> > +	}
> > +#ifdef XR_BLD_FREE_TRACE
> > +	fprintf(stderr, "blocks_allocated = %d\n",
> > +		blocks_allocated);
> > +#endif
> > +}
> > +
> > +/* Feed one of the new btree blocks to the bulk loader. */
> > +static int
> > +rebuild_alloc_block(
> > +	struct xfs_btree_cur	*cur,
> > +	union xfs_btree_ptr	*ptr,
> > +	void			*priv)
> > +{
> > +	struct bt_rebuild	*btr = priv;
> > +
> > +	return xrep_newbt_claim_block(cur, &btr->newbt, ptr);
> > +}
> > +
> >  static void
> >  write_cursor(bt_status_t *curs)
> >  {
> > @@ -336,6 +505,34 @@ finish_cursor(bt_status_t *curs)
> >  	free(curs->btree_blocks);
> >  }
> >  
> > +static void
> > +finish_rebuild(
> > +	struct xfs_mount	*mp,
> > +	struct bt_rebuild	*btr)
> > +{
> > +	struct xrep_newbt_resv	*resv, *n;
> > +
> > +	for_each_xrep_newbt_reservation(&btr->newbt, resv, n) {
> > +		xfs_agnumber_t	agno;
> > +		xfs_agblock_t	bno;
> > +		xfs_extlen_t	len;
> > +
> > +		if (resv->used >= resv->len)
> > +			continue;
> > +
> > +		/* XXX: Shouldn't this go on the AGFL? */
> > +		/* Put back everything we didn't use. */
> > +		bno = XFS_FSB_TO_AGBNO(mp, resv->fsbno + resv->used);
> > +		agno = XFS_FSB_TO_AGNO(mp, resv->fsbno + resv->used);
> > +		len = resv->len - resv->used;
> > +
> > +		add_bno_extent(agno, bno, len);
> > +		add_bcnt_extent(agno, bno, len);
> > +	}
> > +
> > +	xrep_newbt_destroy(&btr->newbt, 0);
> > +}
> > +
> >  /*
> >   * We need to leave some free records in the tree for the corner case of
> >   * setting up the AGFL. This may require allocation of blocks, and as
> > @@ -2290,28 +2487,29 @@ keep_fsinos(xfs_mount_t *mp)
> >  
> >  static void
> >  phase5_func(
> > -	xfs_mount_t	*mp,
> > -	xfs_agnumber_t	agno,
> > -	struct xfs_slab	*lost_fsb)
> > +	struct xfs_mount	*mp,
> > +	xfs_agnumber_t		agno,
> > +	struct xfs_slab		*lost_fsb)
> >  {
> > -	uint64_t	num_inos;
> > -	uint64_t	num_free_inos;
> > -	uint64_t	finobt_num_inos;
> > -	uint64_t	finobt_num_free_inos;
> > -	bt_status_t	bno_btree_curs;
> > -	bt_status_t	bcnt_btree_curs;
> > -	bt_status_t	ino_btree_curs;
> > -	bt_status_t	fino_btree_curs;
> > -	bt_status_t	rmap_btree_curs;
> > -	bt_status_t	refcnt_btree_curs;
> > -	int		extra_blocks = 0;
> > -	uint		num_freeblocks;
> > -	xfs_extlen_t	freeblks1;
> > +	struct repair_ctx	sc = { .mp = mp, };
> > +	struct agi_stat		agi_stat = {0,};
> > +	uint64_t		num_inos;
> > +	uint64_t		num_free_inos;
> > +	uint64_t		finobt_num_inos;
> > +	uint64_t		finobt_num_free_inos;
> > +	bt_status_t		bno_btree_curs;
> > +	bt_status_t		bcnt_btree_curs;
> > +	bt_status_t		ino_btree_curs;
> > +	bt_status_t		fino_btree_curs;
> > +	bt_status_t		rmap_btree_curs;
> > +	bt_status_t		refcnt_btree_curs;
> > +	int			extra_blocks = 0;
> > +	uint			num_freeblocks;
> > +	xfs_extlen_t		freeblks1;
> >  #ifdef DEBUG
> > -	xfs_extlen_t	freeblks2;
> > +	xfs_extlen_t		freeblks2;
> >  #endif
> > -	xfs_agblock_t	num_extents;
> > -	struct agi_stat	agi_stat = {0,};
> > +	xfs_agblock_t		num_extents;
> >  
> >  	if (verbose)
> >  		do_log(_("        - agno = %d\n"), agno);
> > @@ -2533,8 +2731,8 @@ inject_lost_blocks(
> >  		if (error)
> >  			goto out_cancel;
> >  
> > -		error = -libxfs_free_extent(tp, *fsb, 1, &XFS_RMAP_OINFO_AG,
> > -					    XFS_AG_RESV_NONE);
> > +		error = -libxfs_free_extent(tp, *fsb, 1,
> > +				&XFS_RMAP_OINFO_ANY_OWNER, XFS_AG_RESV_NONE);
> >  		if (error)
> >  			goto out_cancel;
> >  
> > 
> 

  reply	other threads:[~2020-05-14 19:49 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-09 16:31 [PATCH v4 0/9] xfs_repair: use btree bulk loading Darrick J. Wong
2020-05-09 16:31 ` [PATCH 1/9] xfs_repair: port the online repair newbt structure Darrick J. Wong
2020-05-14 15:09   ` Brian Foster
2020-05-14 19:20     ` Darrick J. Wong
2020-05-15 11:41       ` Brian Foster
2020-05-15 18:52         ` Darrick J. Wong
2020-05-15 19:43           ` Brian Foster
2020-05-09 16:31 ` [PATCH 2/9] xfs_repair: unindent phase 5 function Darrick J. Wong
2020-05-14 15:09   ` Brian Foster
2020-05-14 19:23     ` Darrick J. Wong
2020-05-09 16:31 ` [PATCH 3/9] xfs_repair: create a new class of btree rebuild cursors Darrick J. Wong
2020-05-14 15:11   ` Brian Foster
2020-05-14 19:47     ` Darrick J. Wong [this message]
2020-05-09 16:32 ` [PATCH 4/9] xfs_repair: rebuild free space btrees with bulk loader Darrick J. Wong
2020-05-14 15:12   ` Brian Foster
2020-05-14 19:53     ` Darrick J. Wong
2020-05-15 11:42       ` Brian Foster
2020-05-09 16:32 ` [PATCH 5/9] xfs_repair: rebuild inode " Darrick J. Wong
2020-05-09 16:32 ` [PATCH 6/9] xfs_repair: rebuild reverse mapping " Darrick J. Wong
2020-05-09 16:32 ` [PATCH 7/9] xfs_repair: rebuild refcount " Darrick J. Wong
2020-05-09 16:32 ` [PATCH 8/9] xfs_repair: remove old btree rebuild support code Darrick J. Wong
2020-05-09 16:32 ` [PATCH 9/9] xfs_repair: track blocks lost during btree construction via extents Darrick J. Wong
  -- strict thread matches above, loose matches on Subject: below --
2020-05-20  1:50 [PATCH v5 0/9] xfs_repair: use btree bulk loading Darrick J. Wong
2020-05-20  1:51 ` [PATCH 3/9] xfs_repair: create a new class of btree rebuild cursors Darrick J. Wong
2020-05-27 12:18   ` Brian Foster
2020-05-27 22:07     ` Darrick J. Wong
2020-05-28 15:09       ` Brian Foster
2020-05-29 21:08         ` Darrick J. Wong
2020-03-04  3:29 [PATCH v3 0/9] xfs_repair: use btree bulk loading Darrick J. Wong
2020-03-04  3:29 ` [PATCH 3/9] xfs_repair: create a new class of btree rebuild cursors Darrick J. Wong
2020-01-01  1:21 [PATCH v2 0/9] xfs_repair: use btree bulk loading Darrick J. Wong
2020-01-01  1:21 ` [PATCH 3/9] xfs_repair: create a new class of btree rebuild cursors Darrick J. Wong
2019-10-29 23:45 [PATCH RFC 0/9] xfs_repair: use btree bulk loading Darrick J. Wong
2019-10-29 23:45 ` [PATCH 3/9] xfs_repair: create a new class of btree rebuild cursors Darrick J. Wong

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=20200514194701.GG6714@magnolia \
    --to=darrick.wong@oracle.com \
    --cc=bfoster@redhat.com \
    --cc=linux-xfs@vger.kernel.org \
    --cc=sandeen@sandeen.net \
    /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).